From 42bd98aeb90b0cd58f8d367f99d5c8f288546eca Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Thu, 18 Aug 2022 17:21:00 -0400 Subject: [PATCH 01/62] Introduce @defer and @stream. Update Section 3 -- Type System.md Update Section 3 -- Type System.md Update Section 3 -- Type System.md Update Section 6 -- Execution.md Update Section 6 -- Execution.md Update Section 6 -- Execution.md Update Section 6 -- Execution.md Update Section 6 -- Execution.md Update Section 6 -- Execution.md Update Section 6 -- Execution.md Update Section 6 -- Execution.md Update Section 6 -- Execution.md Amend changes change initial_count to initialCount add payload fields to Response section add stream validation for overlapping fields spelling updates add note about re-execution add note about final payloads label is optional fix build Update ExecuteQuery with hasNext logic fix spelling fix spaces Update execution to add defer/stream to mutations and subscriptions clarify stream records Apply suggestions from code review Co-authored-by: Benjie Gillam missing bracket Update spec/Section 7 -- Response.md Co-authored-by: Benjie Gillam clarify line about stream record iterator update visitedFragments with defer Updates to consolidate subsequent payload logic for queries, mutations, and subscriptions Apply suggestions from code review Co-authored-by: Benjie Gillam address review feedback Add handling of termination signal more formatting fix spelling Add assertion for record type add "Stream Directives Are Used On List Fields" validation rule Add defaultValue to @stream initialCount Update spec/Section 5 -- Validation.md Co-authored-by: Benjie Gillam # Conflicts: # spec/Section 3 -- Type System.md # spec/Section 5 -- Validation.md # spec/Section 6 -- Execution.md # spec/Section 7 -- Response.md --- cspell.yml | 2 + spec/Section 3 -- Type System.md | 63 +++++- spec/Section 5 -- Validation.md | 37 ++++ spec/Section 6 -- Execution.md | 334 +++++++++++++++++++++++++++---- spec/Section 7 -- Response.md | 67 +++++-- 5 files changed, 451 insertions(+), 52 deletions(-) diff --git a/cspell.yml b/cspell.yml index e8aa73355..66902e41a 100644 --- a/cspell.yml +++ b/cspell.yml @@ -4,12 +4,14 @@ ignoreRegExpList: - /[a-z]{2,}'s/ words: # Terms of art + - deprioritization - endianness - interoperation - monospace - openwebfoundation - parallelization - structs + - sublist - subselection # Fictional characters / examples - alderaan diff --git a/spec/Section 3 -- Type System.md b/spec/Section 3 -- Type System.md index d32b08566..d1c45e1a5 100644 --- a/spec/Section 3 -- Type System.md +++ b/spec/Section 3 -- Type System.md @@ -794,8 +794,9 @@ And will yield the subset of each object type queried: When querying an Object, the resulting mapping of fields are conceptually ordered in the same order in which they were encountered during execution, excluding fragments for which the type does not apply and fields or fragments -that are skipped via `@skip` or `@include` directives. This ordering is -correctly produced when using the {CollectFields()} algorithm. +that are skipped via `@skip` or `@include` directives or temporarily skipped via +`@defer`. This ordering is correctly produced when using the {CollectFields()} +algorithm. Response serialization formats capable of representing ordered maps should maintain this ordering. Serialization formats which can only represent unordered @@ -1942,6 +1943,11 @@ by a validator, executor, or client tool such as a code generator. GraphQL implementations should provide the `@skip` and `@include` directives. +GraphQL implementations are not required to implement the `@defer` and `@stream` +directives. If they are implemented, they must be implemented according to this +specification. GraphQL implementations that do not support these directives must +not make them available via introspection. + GraphQL implementations that support the type system definition language must provide the `@deprecated` directive if representing deprecated portions of the schema. @@ -2162,3 +2168,56 @@ to the relevant IETF specification. ```graphql example scalar UUID @specifiedBy(url: "https://tools.ietf.org/html/rfc4122") ``` + +### @defer + +```graphql +directive @defer( + label: String + if: Boolean +) on FRAGMENT_SPREAD | INLINE_FRAGMENT +``` + +The `@defer` directive may be provided for fragment spreads and inline fragments +to inform the executor to delay the execution of the current fragment to +indicate deprioritization of the current fragment. A query with `@defer` +directive will cause the request to potentially return multiple responses, where +non-deferred data is delivered in the initial response and data deferred is +delivered in a subsequent response. `@include` and `@skip` take precedence over +`@defer`. + +```graphql example +query myQuery($shouldDefer: Boolean) { + user { + name + ...someFragment @defer(label: 'someLabel', if: $shouldDefer) + } +} +fragment someFragment on User { + id + profile_picture { + uri + } +} +``` + +### @stream + +```graphql +directive @stream(label: String, initialCount: Int = 0, if: Boolean) on FIELD +``` + +The `@stream` directive may be provided for a field of `List` type so that the +backend can leverage technology such as asynchronous iterators to provide a +partial list in the initial response, and additional list items in subsequent +responses. `@include` and `@skip` take precedence over `@stream`. + +```graphql example +query myQuery($shouldStream: Boolean) { + user { + friends(first: 10) { + nodes @stream(label: "friendsStream", initialCount: 5, if: $shouldStream) + } + } +} +``` diff --git a/spec/Section 5 -- Validation.md b/spec/Section 5 -- Validation.md index dceec126b..0c1a06c80 100644 --- a/spec/Section 5 -- Validation.md +++ b/spec/Section 5 -- Validation.md @@ -422,6 +422,7 @@ FieldsInSetCanMerge(set): {set} including visiting fragments and inline fragments. - Given each pair of members {fieldA} and {fieldB} in {fieldsForName}: - {SameResponseShape(fieldA, fieldB)} must be true. + - {SameStreamDirective(fieldA, fieldB)} must be true. - If the parent types of {fieldA} and {fieldB} are equal or if either is not an Object Type: - {fieldA} and {fieldB} must have identical field names. @@ -455,6 +456,16 @@ SameResponseShape(fieldA, fieldB): - If {SameResponseShape(subfieldA, subfieldB)} is false, return false. - Return true. +SameStreamDirective(fieldA, fieldB): + +- If neither {fieldA} nor {fieldB} has a directive named `stream`. + - Return true. +- If both {fieldA} and {fieldB} have a directive named `stream`. + - Let {streamA} be the directive named `stream` on {fieldA}. + - Let {streamB} be the directive named `stream` on {fieldB}. + - If {streamA} and {streamB} have identical sets of arguments, return true. +- Return false. + **Explanatory Text** If multiple field selections with the same response names are encountered during @@ -1517,6 +1528,32 @@ query ($foo: Boolean = true, $bar: Boolean = false) { } ``` +### Stream Directives Are Used On List Fields + +**Formal Specification** + +- For every {directive} in a document. +- Let {directiveName} be the name of {directive}. +- If {directiveName} is "stream": + - Let {adjacent} be the AST node the directive affects. + - {adjacent} must be a List type. + +**Explanatory Text** + +GraphQL directive locations do not provide enough granularity to distinguish the +type of fields used in a GraphQL document. Since the stream directive is only +valid on list fields, an additional validation rule must be used to ensure it is +used correctly. + +For example, the following document will only pass validation if `field` is +defined as a List type in the associated schema. + +```graphql counter-example +query { + field @stream(initialCount: 0) +} +``` + ## Variables ### Variable Uniqueness diff --git a/spec/Section 6 -- Execution.md b/spec/Section 6 -- Execution.md index f357069f9..c415e4882 100644 --- a/spec/Section 6 -- Execution.md +++ b/spec/Section 6 -- Execution.md @@ -30,13 +30,21 @@ request is determined by the result of executing this operation according to the "Executing Operations” section below. ExecuteRequest(schema, document, operationName, variableValues, initialValue): +Note: the execution assumes implementing language supports coroutines. +Alternatively, the socket can provide a write buffer pointer to allow +{ExecuteRequest()} to directly write payloads into the buffer. - Let {operation} be the result of {GetOperation(document, operationName)}. - Let {coercedVariableValues} be the result of {CoerceVariableValues(schema, operation, variableValues)}. - If {operation} is a query operation: - - Return {ExecuteQuery(operation, schema, coercedVariableValues, - initialValue)}. + - Let {executionResult} be the result of calling {ExecuteQuery(operation, + schema, coercedVariableValues, initialValue, subsequentPayloads)}. + - If {executionResult} is an iterator: + - For each {payload} in {executionResult}: + - Yield {payload}. + - Otherwise: + - Return {executionResult}. - Otherwise if {operation} is a mutation operation: - Return {ExecuteMutation(operation, schema, coercedVariableValues, initialValue)}. @@ -128,15 +136,28 @@ An initial value may be provided when executing a query operation. ExecuteQuery(query, schema, variableValues, initialValue): +- Let {subsequentPayloads} be an empty list. - Let {queryType} be the root Query type in {schema}. - Assert: {queryType} is an Object type. - Let {selectionSet} be the top level Selection Set in {query}. - Let {data} be the result of running {ExecuteSelectionSet(selectionSet, - queryType, initialValue, variableValues)} _normally_ (allowing - parallelization). + queryType, initialValue, variableValues, subsequentPayloads)} _normally_ + (allowing parallelization). - Let {errors} be the list of all _field error_ raised while executing the selection set. -- Return an unordered map containing {data} and {errors}. +- If {subsequentPayloads} is empty: + - Return an unordered map containing {data} and {errors}. +- If {subsequentPayloads} is not empty: + - Yield an unordered map containing {data}, {errors}, and an entry named + {hasNext} with the value {true}. + - Let {iterator} be the result of running + {YieldSubsequentPayloads(subsequentPayloads)}. + - For each {payload} yielded by {iterator}: + - If a termination signal is received: + - Send a termination signal to {iterator}. + - Return. + - Otherwise: + - Yield {payload}. ### Mutation @@ -150,14 +171,27 @@ mutations ensures against race conditions during these side-effects. ExecuteMutation(mutation, schema, variableValues, initialValue): +- Let {subsequentPayloads} be an empty list. - Let {mutationType} be the root Mutation type in {schema}. - Assert: {mutationType} is an Object type. - Let {selectionSet} be the top level Selection Set in {mutation}. - Let {data} be the result of running {ExecuteSelectionSet(selectionSet, - mutationType, initialValue, variableValues)} _serially_. + mutationType, initialValue, variableValues, subsequentPayloads)} _serially_. - Let {errors} be the list of all _field error_ raised while executing the selection set. -- Return an unordered map containing {data} and {errors}. +- If {subsequentPayloads} is empty: + - Return an unordered map containing {data} and {errors}. +- If {subsequentPayloads} is not empty: + - Yield an unordered map containing {data}, {errors}, and an entry named + {hasNext} with the value {true}. + - Let {iterator} be the result of running + {YieldSubsequentPayloads(subsequentPayloads)}. + - For each {payload} yielded by {iterator}: + - If a termination signal is received: + - Send a termination signal to {iterator}. + - Return. + - Otherwise: + - Yield {payload}. ### Subscription @@ -290,22 +324,36 @@ MapSourceToResponseEvent(sourceStream, subscription, schema, variableValues): - Return a new event stream {responseStream} which yields events as follows: - For each {event} on {sourceStream}: - - Let {response} be the result of running + - Let {executionResult} be the result of running {ExecuteSubscriptionEvent(subscription, schema, variableValues, event)}. - - Yield an event containing {response}. + - For each {response} yielded by {executionResult}: + - Yield an event containing {response}. - When {responseStream} completes: complete this event stream. ExecuteSubscriptionEvent(subscription, schema, variableValues, initialValue): +- Let {subsequentPayloads} be an empty list. - Let {subscriptionType} be the root Subscription type in {schema}. - Assert: {subscriptionType} is an Object type. - Let {selectionSet} be the top level Selection Set in {subscription}. - Let {data} be the result of running {ExecuteSelectionSet(selectionSet, - subscriptionType, initialValue, variableValues)} _normally_ (allowing - parallelization). + subscriptionType, initialValue, variableValues, subsequentPayloads)} + _normally_ (allowing parallelization). - Let {errors} be the list of all _field error_ raised while executing the selection set. -- Return an unordered map containing {data} and {errors}. +- If {subsequentPayloads} is empty: + - Return an unordered map containing {data} and {errors}. +- If {subsequentPayloads} is not empty: + - Yield an unordered map containing {data}, {errors}, and an entry named + {hasNext} with the value {true}. + - Let {iterator} be the result of running + {YieldSubsequentPayloads(subsequentPayloads)}. + - For each {payload} yielded by {iterator}: + - If a termination signal is received: + - Send a termination signal to {iterator}. + - Return. + - Otherwise: + - Yield {payload}. Note: The {ExecuteSubscriptionEvent()} algorithm is intentionally similar to {ExecuteQuery()} since this is how each event result is produced. @@ -321,6 +369,44 @@ Unsubscribe(responseStream): - Cancel {responseStream} +## Yield Subsequent Payloads + +If an operation contains subsequent payload records resulting from `@stream` or +`@defer` directives, the {YieldSubsequentPayloads} algorithm defines how the +payloads should be processed. + +YieldSubsequentPayloads(subsequentPayloads): + +- While {subsequentPayloads} is not empty: +- If a termination signal is received: + - For each {record} in {subsequentPayloads}: + - If {record} is a Stream Record: + - Let {iterator} be the correspondent fields on the Stream Record + structure. + - Send a termination signal to {iterator}. + - Return. +- Let {record} be the first complete item in {subsequentPayloads}. + - Remove {record} from {subsequentPayloads}. + - Assert: {record} must be a Deferred Fragment Record or a Stream Record. + - If {record} is a Deferred Fragment Record: + - Let {payload} be the result of running + {ResolveDeferredFragmentRecord(record, variableValues, + subsequentPayloads)}. + - If {record} is a Stream Record: + - Let {payload} be the result of running {ResolveStreamRecord(record, + variableValues, subsequentPayloads)}. + - If {payload} is {null}: + - If {subsequentPayloads} is empty: + - Yield a map containing a field `hasNext` with the value {false}. + - Return. + - If {subsequentPayloads} is not empty: + - Continue to the next record in {subsequentPayloads}. + - If {record} is not the final element in {subsequentPayloads} + - Add an entry to {payload} named `hasNext` with the value {true}. + - If {record} is the final element in {subsequentPayloads} + - Add an entry to {payload} named `hasNext` with the value {false}. + - Yield {payload} + ## Executing Selection Sets To execute a selection set, the object value being evaluated and the object type @@ -331,10 +417,13 @@ First, the selection set is turned into a grouped field set; then, each represented field in the grouped field set produces an entry into a response map. -ExecuteSelectionSet(selectionSet, objectType, objectValue, variableValues): +ExecuteSelectionSet(selectionSet, objectType, objectValue, variableValues, +subsequentPayloads, parentPath): -- Let {groupedFieldSet} be the result of {CollectFields(objectType, - selectionSet, variableValues)}. +- If {subsequentPayloads} is not provided, initialize it to the empty set. +- If {parentPath} is not provided, initialize it to an empty list. +- Let {groupedFieldSet} be the result of {CollectFields(objectType, objectValue, + selectionSet, variableValues, subsequentPayloads, parentPath)}. - Initialize {resultMap} to an empty ordered map. - For each {groupedFieldSet} as {responseKey} and {fields}: - Let {fieldName} be the name of the first entry in {fields}. Note: This value @@ -343,7 +432,7 @@ ExecuteSelectionSet(selectionSet, objectType, objectValue, variableValues): {objectType}. - If {fieldType} is defined: - Let {responseValue} be {ExecuteField(objectType, objectValue, fieldType, - fields, variableValues)}. + fields, variableValues, subsequentPayloads, parentPath)}. - Set {responseValue} as the value for {responseKey} in {resultMap}. - Return {resultMap}. @@ -464,7 +553,13 @@ Before execution, the selection set is converted to a grouped field set by calling {CollectFields()}. Each entry in the grouped field set is a list of fields that share a response key (the alias if defined, otherwise the field name). This ensures all fields with the same response key (including those in -referenced fragments) are executed at the same time. +referenced fragments) are executed at the same time. A deferred selection set's +fields will not be included in the grouped field set. Rather, a record +representing the deferred fragment and additional context will be stored in a +list. The executor revisits and resumes execution for the list of deferred +fragment records after the initial execution is initiated. This deferred +execution would ‘re-execute’ fields with the same response key that were present +in the grouped field set. As an example, collecting the fields of this selection set would collect two instances of the field `a` and one of field `b`: @@ -489,7 +584,8 @@ The depth-first-search order of the field groups produced by {CollectFields()} is maintained through execution, ensuring that fields appear in the executed response in a stable and predictable order. -CollectFields(objectType, selectionSet, variableValues, visitedFragments): +CollectFields(objectType, objectValue, selectionSet, variableValues, +deferredFragments, parentPath, visitedFragments): - If {visitedFragments} is not provided, initialize it to the empty set. - Initialize {groupedFields} to an empty ordered map of lists. @@ -512,8 +608,12 @@ CollectFields(objectType, selectionSet, variableValues, visitedFragments): - Append {selection} to the {groupForResponseKey}. - If {selection} is a {FragmentSpread}: - Let {fragmentSpreadName} be the name of {selection}. - - If {fragmentSpreadName} is in {visitedFragments}, continue with the next - {selection} in {selectionSet}. + - If {fragmentSpreadName} provides the directive `@defer` and it's {if} + argument is {true} or is a variable in {variableValues} with the value + {true}: + - Let {deferDirective} be that directive. + - If {fragmentSpreadName} is in {visitedFragments} and {deferDirective} is + not defined, continue with the next {selection} in {selectionSet}. - Add {fragmentSpreadName} to {visitedFragments}. - Let {fragment} be the Fragment in the current Document whose name is {fragmentSpreadName}. @@ -523,6 +623,12 @@ CollectFields(objectType, selectionSet, variableValues, visitedFragments): - If {DoesFragmentTypeApply(objectType, fragmentType)} is false, continue with the next {selection} in {selectionSet}. - Let {fragmentSelectionSet} be the top-level selection set of {fragment}. + - If {deferDirective} is defined: + - Let {deferredFragment} be the result of calling + {DeferFragment(objectType, objectValue, fragmentSelectionSet, + parentPath)}. + - Append {deferredFragment} to {deferredFragments}. + - Continue with the next {selection} in {selectionSet}. - Let {fragmentGroupedFieldSet} be the result of calling {CollectFields(objectType, fragmentSelectionSet, variableValues, visitedFragments)}. @@ -538,9 +644,18 @@ CollectFields(objectType, selectionSet, variableValues, visitedFragments): fragmentType)} is false, continue with the next {selection} in {selectionSet}. - Let {fragmentSelectionSet} be the top-level selection set of {selection}. + - If {InlineFragment} provides the directive `@defer`, let {deferDirective} + be that directive. + - If {deferDirective}'s {if} argument is {true} or is a variable in + {variableValues} with the value {true}: + - Let {deferredFragment} be the result of calling + {DeferFragment(objectType, objectValue, fragmentSelectionSet, + parentPath)}. + - Append {deferredFragment} to {deferredFragments}. + - Continue with the next {selection} in {selectionSet}. - Let {fragmentGroupedFieldSet} be the result of calling {CollectFields(objectType, fragmentSelectionSet, variableValues, - visitedFragments)}. + visitedFragments, parentPath)}. - For each {fragmentGroup} in {fragmentGroupedFieldSet}: - Let {responseKey} be the response key shared by all fields in {fragmentGroup}. @@ -549,6 +664,9 @@ CollectFields(objectType, selectionSet, variableValues, visitedFragments): - Append all items in {fragmentGroup} to {groupForResponseKey}. - Return {groupedFields}. +Note: The steps in {CollectFields()} evaluating the `@skip` and `@include` +directives may be applied in either order since they apply commutatively. + DoesFragmentTypeApply(objectType, fragmentType): - If {fragmentType} is an Object Type: @@ -561,8 +679,47 @@ DoesFragmentTypeApply(objectType, fragmentType): - if {objectType} is a possible type of {fragmentType}, return {true} otherwise return {false}. -Note: The steps in {CollectFields()} evaluating the `@skip` and `@include` -directives may be applied in either order since they apply commutatively. +DeferFragment(objectType, objectValue, fragmentSelectionSet, parentPath): + +- Let {label} be the value or the variable to {deferDirective}'s {label} + argument. +- Let {deferredFragmentRecord} be the result of calling + {CreateDeferredFragmentRecord(label, objectType, objectValue, + fragmentSelectionSet, parentPath)}. +- return {deferredFragmentRecord}. + +#### Deferred Fragment Record + +Let {deferredFragmentRecord} be an inline fragment or fragment spread with +`@defer` provided. + +Deferred Fragment Record is a structure containing: + +- {label}: value derived from the `@defer` directive. +- {objectType}: of the {deferredFragmentRecord}. +- {objectValue}: of the {deferredFragmentRecord}. +- {fragmentSelectionSet}: the top level selection set of + {deferredFragmentRecord}. +- {path}: a list of field names and indices from root to + {deferredFragmentRecord}. + +CreateDeferredFragmentRecord(label, objectType, objectValue, +fragmentSelectionSet, path): + +- If {path} is not provided, initialize it to an empty list. +- Construct a deferred fragment record based on the parameters passed in. + +ResolveDeferredFragmentRecord(deferredFragmentRecord, variableValues, +subsequentPayloads): + +- Let {label}, {objectType}, {objectValue}, {fragmentSelectionSet}, {path} be + the corresponding fields in the deferred fragment record structure. +- Let {payload} be the result of calling + {ExecuteSelectionSet(fragmentSelectionSet, objectType, objectValue, + variableValues, subsequentPayloads, path)}. +- Add an entry to {payload} named `label` with the value {label}. +- Add an entry to {payload} named `path` with the value {path}. +- Return {payload}. ## Executing Fields @@ -572,16 +729,29 @@ coerces any provided argument values, then resolves a value for the field, and finally completes that value either by recursively executing another selection set or coercing a scalar value. -ExecuteField(objectType, objectValue, fieldType, fields, variableValues): +ExecuteField(objectType, objectValue, fieldType, fields, variableValues, +subsequentPayloads, parentPath): - Let {field} be the first entry in {fields}. - Let {fieldName} be the field name of {field}. - Let {argumentValues} be the result of {CoerceArgumentValues(objectType, field, variableValues)} +- If {field} provides the directive `@stream`, let {streamDirective} be that + directive. + - Let {initialCount} be the value or variable provided to {streamDirective}'s + {initialCount} argument. + - Let {resolvedValue} be {ResolveFieldGenerator(objectType, objectValue, + fieldName, argumentValues, initialCount)}. + - Let {result} be the result of calling {CompleteValue(fieldType, fields, + resolvedValue, variableValues, subsequentPayloads, parentPath)}. + - Append {fieldName} to the {path} field of every {subsequentPayloads}. + - Return {result}. - Let {resolvedValue} be {ResolveFieldValue(objectType, objectValue, fieldName, argumentValues)}. -- Return the result of {CompleteValue(fieldType, fields, resolvedValue, - variableValues)}. +- Let {result} be the result of calling {CompleteValue(fieldType, fields, + resolvedValue, variableValues, subsequentPayloads)}. +- Append {fieldName} to the {path} for every {subsequentPayloads}. +- Return {result}. ### Coercing Field Arguments @@ -644,11 +814,20 @@ must only allow usage of variables of appropriate types. While nearly all of GraphQL execution can be described generically, ultimately the internal system exposing the GraphQL interface must provide values. This is exposed via {ResolveFieldValue}, which produces a value for a given field on a -type for a real value. +type for a real value. In addition, {ResolveFieldGenerator} will be exposed to +produce an iterator for a field with `List` return type. The internal system may +optionally define a generator function. In the case where the generator is not +defined, the GraphQL executor provides a default generator. For example, a +trivial generator that yields the entire list upon the first iteration. + +As an example, a {ResolveFieldValue} might accept the {objectType} `Person`, the +{field} {"soulMate"}, and the {objectValue} representing John Lennon. It would +be expected to yield the value representing Yoko Ono. -As an example, this might accept the {objectType} `Person`, the {field} -{"soulMate"}, and the {objectValue} representing John Lennon. It would be -expected to yield the value representing Yoko Ono. +A {ResolveFieldGenerator} might accept the {objectType} `MusicBand`, the {field} +{"members"}, and the {objectValue} representing Beatles. It would be expected to +yield a iterator of values representing, John Lennon, Paul McCartney, Ringo +Starr and George Harrison. ResolveFieldValue(objectType, objectValue, fieldName, argumentValues): @@ -657,18 +836,74 @@ ResolveFieldValue(objectType, objectValue, fieldName, argumentValues): - Return the result of calling {resolver}, providing {objectValue} and {argumentValues}. +ResolveFieldGenerator(objectType, objectValue, fieldName, argumentValues, +initialCount): + +- If {objectType} provide an internal function {generatorResolver} for + generating partially resolved value of a list field named {fieldName}: + - Let {generatorResolver} be the internal function. + - Return the iterator from calling {generatorResolver}, providing + {objectValue}, {argumentValues} and {initialCount}. +- Create {generator} from {ResolveFieldValue(objectType, objectValue, fieldName, + argumentValues)}. +- Return {generator}. + Note: It is common for {resolver} to be asynchronous due to relying on reading an underlying database or networked service to produce a value. This necessitates the rest of a GraphQL executor to handle an asynchronous execution -flow. +flow. In addition, a common implementation of {generator} is to leverage +asynchronous iterators or asynchronous generators provided by many programming +languages. ### Value Completion After resolving the value for a field, it is completed by ensuring it adheres to the expected return type. If the return type is another Object type, then the -field execution process continues recursively. - -CompleteValue(fieldType, fields, result, variableValues): +field execution process continues recursively. In the case where a value +returned for a list type field is an iterator due to `@stream` specified on the +field, value completion iterates over the iterator until the number of items +yield by the iterator satisfies `initialCount` specified on the `@stream` +directive. Unresolved items in the iterator will be stored in a stream record +which the executor resumes to execute after the initial execution finishes. + +#### Stream Record + +Let {streamField} be a list field with a `@stream` directive provided. + +A Stream Record is a structure containing: + +- {label}: value derived from the `@stream` directive's `label` argument. +- {iterator}: created by {ResolveFieldGenerator}. +- {resolvedItems}: items resolved from the {iterator} but not yet delivered. +- {index}: indicating the position of the item in the complete list. +- {path}: a list of field names and indices from root to {streamField}. +- {fields}: the group of fields grouped by CollectFields() for {streamField}. +- {innerType}: inner type of {streamField}'s type. + +CreateStreamRecord(label, initialCount, iterator, resolvedItems, index, fields, +innerType): + +- Construct a stream record based on the parameters passed in. + +ResolveStreamRecord(streamRecord, variableValues, subsequentPayloads): + +- Let {label}, {iterator}, {resolvedItems}, {index}, {path}, {fields}, + {innerType} be the correspondent fields on the Stream Record structure. +- Wait for the next item from {iterator}. +- If an item is not retrieved because {iterator} has completed: + - Return {null} +- Let {item} be the item retrieved from {iterator}. +- Append {index} to {path}. +- Increment {index}. +- Let {payload} be the result of calling CompleteValue(innerType, fields, item, + variableValues, subsequentPayloads, path)}. +- Add an entry to {payload} named `label` with the value {label}. +- Add an entry to {payload} named `path` with the value {path}. +- Append {streamRecord} to {subsequentPayloads}. +- Return {payload}. + +CompleteValue(fieldType, fields, result, variableValues, subsequentPayloads, +parentPath): - If the {fieldType} is a Non-Null type: - Let {innerType} be the inner type of {fieldType}. @@ -679,11 +914,34 @@ CompleteValue(fieldType, fields, result, variableValues): - If {result} is {null} (or another internal value similar to {null} such as {undefined}), return {null}. - If {fieldType} is a List type: + - If {result} is an iterator: + - Let {field} be the first entry in {fields}. + - Let {innerType} be the inner type of {fieldType}. + - Let {streamDirective} be the `@stream` directive provided on {field}. + - Let {initialCount} be the value or variable provided to + {streamDirective}'s {initialCount} argument. + - Let {label} be the value or variable provided to {streamDirective}'s + {label} argument. + - Let {resolvedItems} be an empty list + - For each {members} in {result}: + - Append all items from {members} to {resolvedItems}. + - If the length of {resolvedItems} is greater or equal to {initialCount}: + - Let {initialItems} be the sublist of the first {initialCount} items + from {resolvedItems}. + - Let {remainingItems} be the sublist of the items in {resolvedItems} + after the first {initialCount} items. + - Let {streamRecord} be the result of calling {CreateStreamRecord(label, + initialCount, result, remainingItems, initialCount, fields, innerType, + parentPath)} + - Append {streamRecord} to {subsequentPayloads}. + - Let {result} be {initialItems}. + - Exit for each loop. - If {result} is not a collection of values, raise a _field error_. - Let {innerType} be the inner type of {fieldType}. - Return a list where each list item is the result of calling - {CompleteValue(innerType, fields, resultItem, variableValues)}, where - {resultItem} is each item in {result}. + {CompleteValue(innerType, fields, resultItem, variableValues, + subsequentPayloads, parentPath)}, where {resultItem} is each item in + {result}. - If {fieldType} is a Scalar or Enum type: - Return the result of {CoerceResult(fieldType, result)}. - If {fieldType} is an Object, Interface, or Union type: @@ -693,8 +951,8 @@ CompleteValue(fieldType, fields, result, variableValues): - Let {objectType} be {ResolveAbstractType(fieldType, result)}. - Let {subSelectionSet} be the result of calling {MergeSelectionSets(fields)}. - Return the result of evaluating {ExecuteSelectionSet(subSelectionSet, - objectType, result, variableValues)} _normally_ (allowing for - parallelization). + objectType, result, variableValues, subsequentPayloads, parentPath)} + _normally_ (allowing for parallelization). **Coercing Results** diff --git a/spec/Section 7 -- Response.md b/spec/Section 7 -- Response.md index 8dcd9234c..db28604eb 100644 --- a/spec/Section 7 -- Response.md +++ b/spec/Section 7 -- Response.md @@ -10,11 +10,11 @@ the case that any _field error_ was raised on a field and was replaced with ## Response Format -A response to a GraphQL request must be a map. +A response to a GraphQL request must be a map or an event stream of maps. -If the request raised any errors, the response map must contain an entry with -key `errors`. The value of this entry is described in the "Errors" section. If -the request completed without raising any errors, this entry must not be +If the operation encountered any errors, the response map must contain an entry +with key `errors`. The value of this entry is described in the "Errors" section. +If the request completed without raising any errors, this entry must not be present. If the request included execution, the response map must contain an entry with @@ -22,6 +22,24 @@ key `data`. The value of this entry is described in the "Data" section. If the request failed before execution, due to a syntax error, missing information, or validation error, this entry must not be present. +When the response of the GraphQL operation is an event stream, the first value +will be the initial response. All subsequent values may contain `label` and +`path` entries. These two entries are used by clients to identify the the +`@defer` or `@stream` directive from the GraphQL operation that triggered this +value to be returned by the event stream. The combination of these two entries +must be unique across all values returned by the event stream. + +If the response of the GraphQL operation is an event stream, each response map +must contain an entry with key `hasNext`. The value of this entry is `true` for +all but the last response in the stream. The value of this entry is `false` for +the last response of the stream. This entry is not required for GraphQL +operations that return a single response map. + +The GraphQL server may determine there are no more values in the event stream +after a previous value with `hasNext` equal to `true` has been emitted. In this +case the last value in the event stream should be a map without `data`, `label`, +and `path` entries, and a `hasNext` entry with a value of `false`. + The response map may also contain an entry with key `extensions`. This entry, if set, must have a map as its value. This entry is reserved for implementors to extend the protocol however they see fit, and hence there are no additional @@ -42,6 +60,11 @@ requested operation. If the operation was a query, this output will be an object of the query root operation type; if the operation was a mutation, this output will be an object of the mutation root operation type. +If the result of the operation is an event stream, the `data` entry in +subsequent values will be an object of the type of a particular field in the +GraphQL result. The adjacent `path` field will contain the path segments of the +field this data is associated with. + If an error was raised before execution begins, the `data` entry should not be present in the result. @@ -107,14 +130,8 @@ syntax element. If an error can be associated to a particular field in the GraphQL result, it must contain an entry with the key `path` that details the path of the response field which experienced the error. This allows clients to identify whether a -`null` result is intentional or caused by a runtime error. - -This field should be a list of path segments starting at the root of the -response and ending with the field associated with the error. Path segments that -represent fields should be strings, and path segments that represent list -indices should be 0-indexed integers. If the error happens in an aliased field, -the path to the error should use the aliased name, since it represents a path in -the response, not in the request. +`null` result is intentional or caused by a runtime error. The value of this +field is described in the "Path" section. For example, if fetching one of the friends' names fails in the following operation: @@ -244,6 +261,32 @@ discouraged. } ``` +## Path + +A `path` field allows for the association to a particular field in a GraphQL +result. This field should be a list of path segments starting at the root of the +response and ending with the field to be associated with. Path segments that +represent fields should be strings, and path segments that represent list +indices should be 0-indexed integers. If the path is associated to an aliased +field, the path should use the aliased name, since it represents a path in the +response, not in the request. + +When the `path` field is present on a GraphQL response, it indicates that the +`data` field is not the root query or mutation result, but is rather associated +to a particular field in the root result. + +When the `path` field is present on an "Error result", it indicates the response +field which experienced the error. + +## Label + +If the response of the GraphQL operation is an event stream, subsequent values +may contain a string field `label`. This `label` is the same label passed to the +`@defer` or `@stream` directive that triggered this value. This allows clients +to identify which `@defer` or `@stream` directive is associated with this value. +`label` will not be present if the corresponding `@defer` or `@stream` directive +is not passed a `label` argument. + ## Serialization Format GraphQL does not require a specific serialization format. However, clients From 583acb616b83f1e9275b0fe688059982295a8484 Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Wed, 17 Feb 2021 13:03:38 -0500 Subject: [PATCH 02/62] fix typos # Conflicts: # spec/Section 6 -- Execution.md --- spec/Section 6 -- Execution.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/spec/Section 6 -- Execution.md b/spec/Section 6 -- Execution.md index c415e4882..39f2dfd40 100644 --- a/spec/Section 6 -- Execution.md +++ b/spec/Section 6 -- Execution.md @@ -826,8 +826,8 @@ be expected to yield the value representing Yoko Ono. A {ResolveFieldGenerator} might accept the {objectType} `MusicBand`, the {field} {"members"}, and the {objectValue} representing Beatles. It would be expected to -yield a iterator of values representing, John Lennon, Paul McCartney, Ringo -Starr and George Harrison. +yield a iterator of values representing John Lennon, Paul McCartney, Ringo Starr +and George Harrison. ResolveFieldValue(objectType, objectValue, fieldName, argumentValues): From a4bbad68adf9c769ce071e1ad00be8722840e51c Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Wed, 17 Feb 2021 13:06:20 -0500 Subject: [PATCH 03/62] clear up that it is legal to support either defer or stream individually # Conflicts: # spec/Section 3 -- Type System.md --- spec/Section 3 -- Type System.md | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/spec/Section 3 -- Type System.md b/spec/Section 3 -- Type System.md index d1c45e1a5..cb231a086 100644 --- a/spec/Section 3 -- Type System.md +++ b/spec/Section 3 -- Type System.md @@ -1944,9 +1944,9 @@ by a validator, executor, or client tool such as a code generator. GraphQL implementations should provide the `@skip` and `@include` directives. GraphQL implementations are not required to implement the `@defer` and `@stream` -directives. If they are implemented, they must be implemented according to this -specification. GraphQL implementations that do not support these directives must -not make them available via introspection. +directives. If either or both of these directives are implemented, they must be +implemented according to this specification. GraphQL implementations that do not +support these directives must not make them available via introspection. GraphQL implementations that support the type system definition language must provide the `@deprecated` directive if representing deprecated portions of the From b4fb65c378ab8691c3e88566dcb707519661a255 Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Wed, 17 Feb 2021 13:12:04 -0500 Subject: [PATCH 04/62] Add sumary of arguments to Type System --- spec/Section 3 -- Type System.md | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+) diff --git a/spec/Section 3 -- Type System.md b/spec/Section 3 -- Type System.md index cb231a086..3fff10d6d 100644 --- a/spec/Section 3 -- Type System.md +++ b/spec/Section 3 -- Type System.md @@ -2201,6 +2201,15 @@ fragment someFragment on User { } ``` +#### @defer Arguments + +- `if: Boolean` - When true, fragment may be deferred. If omitted, defaults to + `true`. +- `label: String` - A unique label across all `@defer` and `@stream` directives + in an operation. This label should be used by GraphQL clients to identify the + data from patch responses and associate it with the correct fragments. If + provided, the GraphQL Server must add it to the payload. + ### @stream ```graphql @@ -2221,3 +2230,14 @@ query myQuery($shouldStream: Boolean) { } } ``` + +#### @stream Arguments + +- `if: Boolean` - When true, field may be streamed. If omitted, defaults to + `true`. +- `label: String` - A unique label across all `@defer` and `@stream` directives + in an operation. This label should be used by GraphQL clients to identify the + data from patch responses and associate it with the correct fragments. If + provided, the GraphQL Server must add it to the payload. +- `initialCount: Int` - The number of list items the server should return as + part of the initial response. If omitted, defaults to `0`. From 3257f192bfb8c4a72043f880d96b742f144280b6 Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Sat, 15 May 2021 11:24:10 -0400 Subject: [PATCH 05/62] Update Section 3 -- Type System.md # Conflicts: # spec/Section 3 -- Type System.md --- spec/Section 3 -- Type System.md | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/spec/Section 3 -- Type System.md b/spec/Section 3 -- Type System.md index 3fff10d6d..f6091d8fb 100644 --- a/spec/Section 3 -- Type System.md +++ b/spec/Section 3 -- Type System.md @@ -2203,8 +2203,9 @@ fragment someFragment on User { #### @defer Arguments -- `if: Boolean` - When true, fragment may be deferred. If omitted, defaults to - `true`. +- `if: Boolean` - When `true`, fragment may be deferred. When `false`, fragment + will not be deferred and data will be included in the initial response. If + omitted, defaults to `true`. - `label: String` - A unique label across all `@defer` and `@stream` directives in an operation. This label should be used by GraphQL clients to identify the data from patch responses and associate it with the correct fragments. If @@ -2233,8 +2234,9 @@ query myQuery($shouldStream: Boolean) { #### @stream Arguments -- `if: Boolean` - When true, field may be streamed. If omitted, defaults to - `true`. +- `if: Boolean` - When `true`, field may be streamed. When `false`, the field + will not be streamed and all list items will be included in the initial + response. If omitted, defaults to `true`. - `label: String` - A unique label across all `@defer` and `@stream` directives in an operation. This label should be used by GraphQL clients to identify the data from patch responses and associate it with the correct fragments. If From 3cce3fa41c5dcb4793fe6df15145650fe8e5a2ea Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Fri, 19 Nov 2021 15:41:51 -0500 Subject: [PATCH 06/62] clarification on defer/stream requirement # Conflicts: # spec/Section 3 -- Type System.md --- spec/Section 3 -- Type System.md | 22 +++++++++++++++++----- 1 file changed, 17 insertions(+), 5 deletions(-) diff --git a/spec/Section 3 -- Type System.md b/spec/Section 3 -- Type System.md index f6091d8fb..47af2af1e 100644 --- a/spec/Section 3 -- Type System.md +++ b/spec/Section 3 -- Type System.md @@ -2203,9 +2203,9 @@ fragment someFragment on User { #### @defer Arguments -- `if: Boolean` - When `true`, fragment may be deferred. When `false`, fragment - will not be deferred and data will be included in the initial response. If - omitted, defaults to `true`. +- `if: Boolean` - When `true`, fragment _should_ be deferred. When `false`, + fragment will not be deferred and data will be included in the initial + response. If omitted, defaults to `true`. - `label: String` - A unique label across all `@defer` and `@stream` directives in an operation. This label should be used by GraphQL clients to identify the data from patch responses and associate it with the correct fragments. If @@ -2234,8 +2234,8 @@ query myQuery($shouldStream: Boolean) { #### @stream Arguments -- `if: Boolean` - When `true`, field may be streamed. When `false`, the field - will not be streamed and all list items will be included in the initial +- `if: Boolean` - When `true`, field _should_ be streamed. When `false`, the + field will not be streamed and all list items will be included in the initial response. If omitted, defaults to `true`. - `label: String` - A unique label across all `@defer` and `@stream` directives in an operation. This label should be used by GraphQL clients to identify the @@ -2243,3 +2243,15 @@ query myQuery($shouldStream: Boolean) { provided, the GraphQL Server must add it to the payload. - `initialCount: Int` - The number of list items the server should return as part of the initial response. If omitted, defaults to `0`. + +Note: The ability to defer and/or stream parts of a response can have a +potentially significant impact on application performance. Developers generally +need clear, predictable control over their application's performance. It is +highly recommended that GraphQL servers honor the `@defer` and `@stream` +directives on each execution. However, the specification allows advanced +use-cases where the server can determine that it is more performant to not defer +and/or stream. Therefore, GraphQL clients _must_ be able to process a response +that ignores the `@defer` and/or `@stream` directives. This also applies to the +`initialCount` argument on the `@stream` directive. Clients _must_ be able to +process a streamed response that contains a different number of initial list +items than what was specified in the `initialCount` argument. From f81d74546960d8aa11c783b5fb26492316da7b5c Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Sat, 20 Nov 2021 11:58:41 -0500 Subject: [PATCH 07/62] clarify negative values of initialCount # Conflicts: # spec/Section 3 -- Type System.md --- spec/Section 3 -- Type System.md | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/spec/Section 3 -- Type System.md b/spec/Section 3 -- Type System.md index 47af2af1e..d0fc4c372 100644 --- a/spec/Section 3 -- Type System.md +++ b/spec/Section 3 -- Type System.md @@ -2242,7 +2242,8 @@ query myQuery($shouldStream: Boolean) { data from patch responses and associate it with the correct fragments. If provided, the GraphQL Server must add it to the payload. - `initialCount: Int` - The number of list items the server should return as - part of the initial response. If omitted, defaults to `0`. + part of the initial response. If omitted, defaults to `0`. If the value of + this argument is less than `0`, it is treated the same as `0`. Note: The ability to defer and/or stream parts of a response can have a potentially significant impact on application performance. Developers generally From b0cae7f4368f84e8affa3fc90a7ace670e47605d Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Thu, 25 Nov 2021 11:06:48 -0500 Subject: [PATCH 08/62] allow extensions only subsequent payloads # Conflicts: # spec/Section 7 -- Response.md --- spec/Section 7 -- Response.md | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/spec/Section 7 -- Response.md b/spec/Section 7 -- Response.md index db28604eb..19b073161 100644 --- a/spec/Section 7 -- Response.md +++ b/spec/Section 7 -- Response.md @@ -43,7 +43,9 @@ and `path` entries, and a `hasNext` entry with a value of `false`. The response map may also contain an entry with key `extensions`. This entry, if set, must have a map as its value. This entry is reserved for implementors to extend the protocol however they see fit, and hence there are no additional -restrictions on its contents. +restrictions on its contents. When the response of the GraphQL operation is an +event stream, implementors may send subsequent payloads containing only +`hasNext` and `extensions` entries. To ensure future changes to the protocol do not break existing services and clients, the top level response map must not contain any entries other than the From 9449853523deb86ca8e0fc1e60a6255df6dd82d2 Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Fri, 26 Nov 2021 09:25:24 -0500 Subject: [PATCH 09/62] fix typo # Conflicts: # spec/Section 7 -- Response.md --- spec/Section 7 -- Response.md | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/spec/Section 7 -- Response.md b/spec/Section 7 -- Response.md index 19b073161..353f40d2c 100644 --- a/spec/Section 7 -- Response.md +++ b/spec/Section 7 -- Response.md @@ -24,10 +24,10 @@ validation error, this entry must not be present. When the response of the GraphQL operation is an event stream, the first value will be the initial response. All subsequent values may contain `label` and -`path` entries. These two entries are used by clients to identify the the -`@defer` or `@stream` directive from the GraphQL operation that triggered this -value to be returned by the event stream. The combination of these two entries -must be unique across all values returned by the event stream. +`path` entries. These two entries are used by clients to identify the `@defer` +or `@stream` directive from the GraphQL operation that triggered this value to +be returned by the event stream. The combination of these two entries must be +unique across all values returned by the event stream. If the response of the GraphQL operation is an event stream, each response map must contain an entry with key `hasNext`. The value of this entry is `true` for From c81289029f6e57541721d9fa3c848c10bb44ee11 Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Thu, 18 Aug 2022 17:27:07 -0400 Subject: [PATCH 10/62] Raise a field error if initialCount is less than zero # Conflicts: # spec/Section 3 -- Type System.md # spec/Section 6 -- Execution.md --- spec/Section 3 -- Type System.md | 4 ++-- spec/Section 6 -- Execution.md | 1 + 2 files changed, 3 insertions(+), 2 deletions(-) diff --git a/spec/Section 3 -- Type System.md b/spec/Section 3 -- Type System.md index d0fc4c372..4a25974b0 100644 --- a/spec/Section 3 -- Type System.md +++ b/spec/Section 3 -- Type System.md @@ -2242,8 +2242,8 @@ query myQuery($shouldStream: Boolean) { data from patch responses and associate it with the correct fragments. If provided, the GraphQL Server must add it to the payload. - `initialCount: Int` - The number of list items the server should return as - part of the initial response. If omitted, defaults to `0`. If the value of - this argument is less than `0`, it is treated the same as `0`. + part of the initial response. If omitted, defaults to `0`. A field error will + be raised if the value of this argument is less than `0`. Note: The ability to defer and/or stream parts of a response can have a potentially significant impact on application performance. Developers generally diff --git a/spec/Section 6 -- Execution.md b/spec/Section 6 -- Execution.md index 39f2dfd40..1c2b8d5b3 100644 --- a/spec/Section 6 -- Execution.md +++ b/spec/Section 6 -- Execution.md @@ -920,6 +920,7 @@ parentPath): - Let {streamDirective} be the `@stream` directive provided on {field}. - Let {initialCount} be the value or variable provided to {streamDirective}'s {initialCount} argument. + - If {initialCount} is less than zero, raise a _field error_. - Let {label} be the value or variable provided to {streamDirective}'s {label} argument. - Let {resolvedItems} be an empty list From c1fefe4870b0a08528d31ee04cf96527f3605ba4 Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Mon, 6 Dec 2021 13:15:13 -0500 Subject: [PATCH 11/62] data is not necessarily an object in subsequent payloads # Conflicts: # spec/Section 7 -- Response.md --- spec/Section 7 -- Response.md | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/spec/Section 7 -- Response.md b/spec/Section 7 -- Response.md index 353f40d2c..47ee7e893 100644 --- a/spec/Section 7 -- Response.md +++ b/spec/Section 7 -- Response.md @@ -63,9 +63,9 @@ of the query root operation type; if the operation was a mutation, this output will be an object of the mutation root operation type. If the result of the operation is an event stream, the `data` entry in -subsequent values will be an object of the type of a particular field in the -GraphQL result. The adjacent `path` field will contain the path segments of the -field this data is associated with. +subsequent values will be of the type of a particular field in the GraphQL +result. The adjacent `path` field will contain the path segments of the field +this data is associated with. If an error was raised before execution begins, the `data` entry should not be present in the result. From 6dd9b79e9ff300b31fe36982652ed961ba23fa20 Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Mon, 6 Dec 2021 17:20:37 -0500 Subject: [PATCH 12/62] add Defer And Stream Directives Are Used On Valid Root Field rule --- spec/Section 5 -- Validation.md | 28 ++++++++++++++++++++++++++++ 1 file changed, 28 insertions(+) diff --git a/spec/Section 5 -- Validation.md b/spec/Section 5 -- Validation.md index 0c1a06c80..8303ce776 100644 --- a/spec/Section 5 -- Validation.md +++ b/spec/Section 5 -- Validation.md @@ -1528,6 +1528,34 @@ query ($foo: Boolean = true, $bar: Boolean = false) { } ``` +### Defer And Stream Directives Are Used On Valid Root Field + +** Formal Specification ** + +- For every {directive} in a document. +- Let {directiveName} be the name of {directive}. +- Let {mutationType} be the root Mutation type in {schema}. +- Let {subscriptionType} be the root Subscription type in {schema}. +- If {directiveName} is "defer" or "stream": + - The parent type of {directive} must not be {mutationType} or + {subscriptionType}. + +**Explanatory Text** + +The defer and stream directives are not allowed to be used on root fields of the +mutation or subscription type. + +For example, the following document will not pass validation because `@defer` +has been used on a root mutation field: + +```raw graphql counter-example +mutation { + ... @defer { + mutationField + } +} +``` + ### Stream Directives Are Used On List Fields **Formal Specification** From ebfefb6e00cbfd044dfb9b82b0300b0c6fd3b941 Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Thu, 18 Aug 2022 17:21:50 -0400 Subject: [PATCH 13/62] wait for parent async record to ensure correct order of payloads # Conflicts: # spec/Section 6 -- Execution.md --- cspell.yml | 1 - spec/Section 6 -- Execution.md | 192 ++++++++++++++++++--------------- 2 files changed, 108 insertions(+), 85 deletions(-) diff --git a/cspell.yml b/cspell.yml index 66902e41a..8bc4a231c 100644 --- a/cspell.yml +++ b/cspell.yml @@ -11,7 +11,6 @@ words: - openwebfoundation - parallelization - structs - - sublist - subselection # Fictional characters / examples - alderaan diff --git a/spec/Section 6 -- Execution.md b/spec/Section 6 -- Execution.md index 1c2b8d5b3..900f6b230 100644 --- a/spec/Section 6 -- Execution.md +++ b/spec/Section 6 -- Execution.md @@ -417,13 +417,13 @@ First, the selection set is turned into a grouped field set; then, each represented field in the grouped field set produces an entry into a response map. -ExecuteSelectionSet(selectionSet, objectType, objectValue, variableValues, -subsequentPayloads, parentPath): +ExecuteSelectionSet(selectionSet, objectType, objectValue, variableValues, path, +subsequentPayloads, asyncRecord): +- If {path} is not provided, initialize it to an empty list. - If {subsequentPayloads} is not provided, initialize it to the empty set. -- If {parentPath} is not provided, initialize it to an empty list. - Let {groupedFieldSet} be the result of {CollectFields(objectType, objectValue, - selectionSet, variableValues, subsequentPayloads, parentPath)}. + selectionSet, variableValues, path subsequentPayloads, asyncRecord)}. - Initialize {resultMap} to an empty ordered map. - For each {groupedFieldSet} as {responseKey} and {fields}: - Let {fieldName} be the name of the first entry in {fields}. Note: This value @@ -432,7 +432,7 @@ subsequentPayloads, parentPath): {objectType}. - If {fieldType} is defined: - Let {responseValue} be {ExecuteField(objectType, objectValue, fieldType, - fields, variableValues, subsequentPayloads, parentPath)}. + fields, variableValues, path, subsequentPayloads, asyncRecord)}. - Set {responseValue} as the value for {responseKey} in {resultMap}. - Return {resultMap}. @@ -584,8 +584,8 @@ The depth-first-search order of the field groups produced by {CollectFields()} is maintained through execution, ensuring that fields appear in the executed response in a stable and predictable order. -CollectFields(objectType, objectValue, selectionSet, variableValues, -deferredFragments, parentPath, visitedFragments): +CollectFields(objectType, objectValue, selectionSet, variableValues, path, +subsequentPayloads, asyncRecord, visitedFragments): - If {visitedFragments} is not provided, initialize it to the empty set. - Initialize {groupedFields} to an empty ordered map of lists. @@ -624,14 +624,16 @@ deferredFragments, parentPath, visitedFragments): with the next {selection} in {selectionSet}. - Let {fragmentSelectionSet} be the top-level selection set of {fragment}. - If {deferDirective} is defined: - - Let {deferredFragment} be the result of calling - {DeferFragment(objectType, objectValue, fragmentSelectionSet, - parentPath)}. - - Append {deferredFragment} to {deferredFragments}. + - Let {label} be the value or the variable to {deferDirective}'s {label} + argument. + - Let {deferredFragmentRecord} be the result of calling + {CreateDeferredFragmentRecord(label, objectType, objectValue, + fragmentSelectionSet, path, asyncRecord)}. + - Append {deferredFragmentRecord} to {subsequentPayloads}. - Continue with the next {selection} in {selectionSet}. - Let {fragmentGroupedFieldSet} be the result of calling - {CollectFields(objectType, fragmentSelectionSet, variableValues, - visitedFragments)}. + {CollectFields(objectType, objectValue, fragmentSelectionSet, + variableValues, path, subsequentPayloads, asyncRecord, visitedFragments)}. - For each {fragmentGroup} in {fragmentGroupedFieldSet}: - Let {responseKey} be the response key shared by all fields in {fragmentGroup}. @@ -648,14 +650,16 @@ deferredFragments, parentPath, visitedFragments): be that directive. - If {deferDirective}'s {if} argument is {true} or is a variable in {variableValues} with the value {true}: - - Let {deferredFragment} be the result of calling - {DeferFragment(objectType, objectValue, fragmentSelectionSet, - parentPath)}. - - Append {deferredFragment} to {deferredFragments}. + - Let {label} be the value or the variable to {deferDirective}'s {label} + argument. + - Let {deferredFragmentRecord} be the result of calling + {CreateDeferredFragmentRecord(label, objectType, objectValue, + fragmentSelectionSet, path, asyncRecord)}. + - Append {deferredFragmentRecord} to {subsequentPayloads}. - Continue with the next {selection} in {selectionSet}. - Let {fragmentGroupedFieldSet} be the result of calling - {CollectFields(objectType, fragmentSelectionSet, variableValues, - visitedFragments, parentPath)}. + {CollectFields(objectType, objectValue, fragmentSelectionSet, + variableValues, path, subsequentPayloads, asyncRecord, visitedFragments)}. - For each {fragmentGroup} in {fragmentGroupedFieldSet}: - Let {responseKey} be the response key shared by all fields in {fragmentGroup}. @@ -679,44 +683,53 @@ DoesFragmentTypeApply(objectType, fragmentType): - if {objectType} is a possible type of {fragmentType}, return {true} otherwise return {false}. -DeferFragment(objectType, objectValue, fragmentSelectionSet, parentPath): +#### Async Payload Record + +An Async Payload Record is either a Deferred Fragment Record or a Stream Record. +All Async Payload Records are structures containing: -- Let {label} be the value or the variable to {deferDirective}'s {label} - argument. -- Let {deferredFragmentRecord} be the result of calling - {CreateDeferredFragmentRecord(label, objectType, objectValue, - fragmentSelectionSet, parentPath)}. -- return {deferredFragmentRecord}. +- {label}: value derived from the corresponding `@defer` or `@stream` directive. +- {parentRecord}: optionally an Async Payload Record. +- {errors}: a list of field errors encountered during execution. +- {dataExecution}: A result that can notify when the corresponding execution has + completed. +- {path}: a list of field names and indices from root to the location of the + corresponding `@defer` or `@stream` directive. #### Deferred Fragment Record Let {deferredFragmentRecord} be an inline fragment or fragment spread with `@defer` provided. -Deferred Fragment Record is a structure containing: +Deferred Fragment Record is a structure containing all the entries of Async +Payload Record, and additionally: -- {label}: value derived from the `@defer` directive. - {objectType}: of the {deferredFragmentRecord}. - {objectValue}: of the {deferredFragmentRecord}. - {fragmentSelectionSet}: the top level selection set of {deferredFragmentRecord}. -- {path}: a list of field names and indices from root to - {deferredFragmentRecord}. CreateDeferredFragmentRecord(label, objectType, objectValue, -fragmentSelectionSet, path): +fragmentSelectionSet, path, parentRecord): -- If {path} is not provided, initialize it to an empty list. - Construct a deferred fragment record based on the parameters passed in. +- Initialize {errors} to an empty list. ResolveDeferredFragmentRecord(deferredFragmentRecord, variableValues, subsequentPayloads): -- Let {label}, {objectType}, {objectValue}, {fragmentSelectionSet}, {path} be - the corresponding fields in the deferred fragment record structure. -- Let {payload} be the result of calling - {ExecuteSelectionSet(fragmentSelectionSet, objectType, objectValue, - variableValues, subsequentPayloads, path)}. +- Let {label}, {objectType}, {objectValue}, {fragmentSelectionSet}, {path}, + {parentRecord} be the corresponding fields in the deferred fragment record + structure. +- Let {dataExecution} be the asynchronous future value of: + - Let {payload} be the result of {ExecuteSelectionSet(fragmentSelectionSet, + objectType, objectValue, variableValues, path, subsequentPayloads, + deferredFragmentRecord)}. + - If {parentRecord} is defined: + - Wait for the result of {dataExecution} on {parentRecord}. + - Return {payload}. +- Set {dataExecution} on {deferredFragmentRecord}. +- Let {payload} be the result of waiting for {dataExecution}. - Add an entry to {payload} named `label` with the value {label}. - Add an entry to {payload} named `path` with the value {path}. - Return {payload}. @@ -729,11 +742,12 @@ coerces any provided argument values, then resolves a value for the field, and finally completes that value either by recursively executing another selection set or coercing a scalar value. -ExecuteField(objectType, objectValue, fieldType, fields, variableValues, -subsequentPayloads, parentPath): +ExecuteField(objectType, objectValue, fieldType, fields, variableValues, path, +subsequentPayloads, asyncRecord): - Let {field} be the first entry in {fields}. - Let {fieldName} be the field name of {field}. +- Append {fieldName} to {path}. - Let {argumentValues} be the result of {CoerceArgumentValues(objectType, field, variableValues)} - If {field} provides the directive `@stream`, let {streamDirective} be that @@ -741,16 +755,14 @@ subsequentPayloads, parentPath): - Let {initialCount} be the value or variable provided to {streamDirective}'s {initialCount} argument. - Let {resolvedValue} be {ResolveFieldGenerator(objectType, objectValue, - fieldName, argumentValues, initialCount)}. + fieldName, argumentValues)}. - Let {result} be the result of calling {CompleteValue(fieldType, fields, - resolvedValue, variableValues, subsequentPayloads, parentPath)}. - - Append {fieldName} to the {path} field of every {subsequentPayloads}. + resolvedValue, variableValues, path, subsequentPayloads, asyncRecord)}. - Return {result}. - Let {resolvedValue} be {ResolveFieldValue(objectType, objectValue, fieldName, argumentValues)}. - Let {result} be the result of calling {CompleteValue(fieldType, fields, - resolvedValue, variableValues, subsequentPayloads)}. -- Append {fieldName} to the {path} for every {subsequentPayloads}. + resolvedValue, variableValues, path, subsequentPayloads, asyncRecord)}. - Return {result}. ### Coercing Field Arguments @@ -836,14 +848,13 @@ ResolveFieldValue(objectType, objectValue, fieldName, argumentValues): - Return the result of calling {resolver}, providing {objectValue} and {argumentValues}. -ResolveFieldGenerator(objectType, objectValue, fieldName, argumentValues, -initialCount): +ResolveFieldGenerator(objectType, objectValue, fieldName, argumentValues): - If {objectType} provide an internal function {generatorResolver} for generating partially resolved value of a list field named {fieldName}: - Let {generatorResolver} be the internal function. - Return the iterator from calling {generatorResolver}, providing - {objectValue}, {argumentValues} and {initialCount}. + {objectValue} and {argumentValues}. - Create {generator} from {ResolveFieldValue(objectType, objectValue, fieldName, argumentValues)}. - Return {generator}. @@ -863,52 +874,58 @@ field execution process continues recursively. In the case where a value returned for a list type field is an iterator due to `@stream` specified on the field, value completion iterates over the iterator until the number of items yield by the iterator satisfies `initialCount` specified on the `@stream` -directive. Unresolved items in the iterator will be stored in a stream record -which the executor resumes to execute after the initial execution finishes. +directive. #### Stream Record Let {streamField} be a list field with a `@stream` directive provided. -A Stream Record is a structure containing: +A Stream Record is a structure containing all the entries of Async Payload +Record, and additionally: -- {label}: value derived from the `@stream` directive's `label` argument. - {iterator}: created by {ResolveFieldGenerator}. -- {resolvedItems}: items resolved from the {iterator} but not yet delivered. - {index}: indicating the position of the item in the complete list. -- {path}: a list of field names and indices from root to {streamField}. - {fields}: the group of fields grouped by CollectFields() for {streamField}. - {innerType}: inner type of {streamField}'s type. -CreateStreamRecord(label, initialCount, iterator, resolvedItems, index, fields, -innerType): +CreateStreamRecord(label, iterator, index, fields, innerType, path, +parentRecord): - Construct a stream record based on the parameters passed in. +- Initialize {errors} to an empty list. ResolveStreamRecord(streamRecord, variableValues, subsequentPayloads): -- Let {label}, {iterator}, {resolvedItems}, {index}, {path}, {fields}, +- Let {label}, {parentRecord}, {iterator}, {index}, {path}, {fields}, {innerType} be the correspondent fields on the Stream Record structure. -- Wait for the next item from {iterator}. -- If an item is not retrieved because {iterator} has completed: - - Return {null} -- Let {item} be the item retrieved from {iterator}. -- Append {index} to {path}. -- Increment {index}. -- Let {payload} be the result of calling CompleteValue(innerType, fields, item, - variableValues, subsequentPayloads, path)}. +- Let {indexPath} be {path} with {index} appended. +- Let {dataExecution} be the asynchronous future value of: + - Wait for the next item from {iterator}. + - If an item is not retrieved because {iterator} has completed: + - Return {null}. + - Let {item} be the item retrieved from {iterator}. + - Let {payload} be the result of calling {CompleteValue(innerType, fields, + item, variableValues, indexPath, subsequentPayloads, parentRecord)}. + - Increment {index}. + - Let {nextStreamRecord} be the result of calling {CreateStreamRecord(label, + iterator, index, fields, innerType, path, streamRecord)}. + - Append {nextStreamRecord} to {subsequentPayloads}. + - If {parentRecord} is defined: + - Wait for the result of {dataExecution} on {parentRecord}. + - Return {payload}. +- Set {dataExecution} on {streamRecord}. +- Let {payload} be the result of waiting for {dataExecution}. - Add an entry to {payload} named `label` with the value {label}. -- Add an entry to {payload} named `path` with the value {path}. -- Append {streamRecord} to {subsequentPayloads}. +- Add an entry to {payload} named `path` with the value {indexPath}. - Return {payload}. -CompleteValue(fieldType, fields, result, variableValues, subsequentPayloads, -parentPath): +CompleteValue(fieldType, fields, result, variableValues, path, +subsequentPayloads, asyncRecord): - If the {fieldType} is a Non-Null type: - Let {innerType} be the inner type of {fieldType}. - Let {completedResult} be the result of calling {CompleteValue(innerType, - fields, result, variableValues)}. + fields, result, variableValues, path)}. - If {completedResult} is {null}, raise a _field error_. - Return {completedResult}. - If {result} is {null} (or another internal value similar to {null} such as @@ -923,26 +940,33 @@ parentPath): - If {initialCount} is less than zero, raise a _field error_. - Let {label} be the value or variable provided to {streamDirective}'s {label} argument. - - Let {resolvedItems} be an empty list - - For each {members} in {result}: - - Append all items from {members} to {resolvedItems}. - - If the length of {resolvedItems} is greater or equal to {initialCount}: - - Let {initialItems} be the sublist of the first {initialCount} items - from {resolvedItems}. - - Let {remainingItems} be the sublist of the items in {resolvedItems} - after the first {initialCount} items. + - Let {initialItems} be an empty list + - Let {index} be zero. + - While {result} is not closed: + - If {streamDirective} was not provided or {index} is not greater than or + equal to {initialCount}: + - Wait for the next item from {result}. + - Let {resultItem} be the item retrieved from {result}. + - Let {indexPath} be {path} with {index} appended. + - Let {resolvedItem} be the result of calling {CompleteValue(innerType, + fields, resultItem, variableValues, indexPath, subsequentPayloads, + asyncRecord)}. + - Append {resolvedItem} to {initialItems}. + - Increment {index}. + - If {streamDirective} was provided and {index} is greater than or equal + to {initialCount}: - Let {streamRecord} be the result of calling {CreateStreamRecord(label, - initialCount, result, remainingItems, initialCount, fields, innerType, - parentPath)} + result, index, fields, innerType, path, asyncRecord)}. - Append {streamRecord} to {subsequentPayloads}. - Let {result} be {initialItems}. - - Exit for each loop. + - Exit while loop. + - Return {initialItems}. - If {result} is not a collection of values, raise a _field error_. - Let {innerType} be the inner type of {fieldType}. - Return a list where each list item is the result of calling - {CompleteValue(innerType, fields, resultItem, variableValues, - subsequentPayloads, parentPath)}, where {resultItem} is each item in - {result}. + {CompleteValue(innerType, fields, resultItem, variableValues, indexPath, + subsequentPayloads, asyncRecord)}, where {resultItem} is each item in + {result} and {indexPath} is {path} with the index of the item appended. - If {fieldType} is a Scalar or Enum type: - Return the result of {CoerceResult(fieldType, result)}. - If {fieldType} is an Object, Interface, or Union type: @@ -952,7 +976,7 @@ parentPath): - Let {objectType} be {ResolveAbstractType(fieldType, result)}. - Let {subSelectionSet} be the result of calling {MergeSelectionSets(fields)}. - Return the result of evaluating {ExecuteSelectionSet(subSelectionSet, - objectType, result, variableValues, subsequentPayloads, parentPath)} + objectType, result, variableValues, path, subsequentPayloads, asyncRecord)} _normally_ (allowing for parallelization). **Coercing Results** From 30674c8434963ea5dab295f8d31921ea3c6a80b0 Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Mon, 20 Dec 2021 16:00:43 -0500 Subject: [PATCH 14/62] Simplify execution, payloads should begin execution immediately # Conflicts: # spec/Section 6 -- Execution.md --- spec/Section 6 -- Execution.md | 115 ++++++++++----------------------- 1 file changed, 34 insertions(+), 81 deletions(-) diff --git a/spec/Section 6 -- Execution.md b/spec/Section 6 -- Execution.md index 900f6b230..f7920e622 100644 --- a/spec/Section 6 -- Execution.md +++ b/spec/Section 6 -- Execution.md @@ -380,21 +380,13 @@ YieldSubsequentPayloads(subsequentPayloads): - While {subsequentPayloads} is not empty: - If a termination signal is received: - For each {record} in {subsequentPayloads}: - - If {record} is a Stream Record: - - Let {iterator} be the correspondent fields on the Stream Record - structure. + - If {record} contains {iterator}: - Send a termination signal to {iterator}. - Return. -- Let {record} be the first complete item in {subsequentPayloads}. +- Let {record} be the first item in {subsequentPayloads} with a completed + {dataExecution}. - Remove {record} from {subsequentPayloads}. - - Assert: {record} must be a Deferred Fragment Record or a Stream Record. - - If {record} is a Deferred Fragment Record: - - Let {payload} be the result of running - {ResolveDeferredFragmentRecord(record, variableValues, - subsequentPayloads)}. - - If {record} is a Stream Record: - - Let {payload} be the result of running {ResolveStreamRecord(record, - variableValues, subsequentPayloads)}. + - Let {payload} be the completed result returned by {dataExecution}. - If {payload} is {null}: - If {subsequentPayloads} is empty: - Yield a map containing a field `hasNext` with the value {false}. @@ -626,10 +618,9 @@ subsequentPayloads, asyncRecord, visitedFragments): - If {deferDirective} is defined: - Let {label} be the value or the variable to {deferDirective}'s {label} argument. - - Let {deferredFragmentRecord} be the result of calling - {CreateDeferredFragmentRecord(label, objectType, objectValue, - fragmentSelectionSet, path, asyncRecord)}. - - Append {deferredFragmentRecord} to {subsequentPayloads}. + - Call {ExecuteDeferredFragment(label, objectType, objectValue, + fragmentSelectionSet, path, variableValues, asyncRecord, + subsequentPayloads)}. - Continue with the next {selection} in {selectionSet}. - Let {fragmentGroupedFieldSet} be the result of calling {CollectFields(objectType, objectValue, fragmentSelectionSet, @@ -652,10 +643,8 @@ subsequentPayloads, asyncRecord, visitedFragments): {variableValues} with the value {true}: - Let {label} be the value or the variable to {deferDirective}'s {label} argument. - - Let {deferredFragmentRecord} be the result of calling - {CreateDeferredFragmentRecord(label, objectType, objectValue, - fragmentSelectionSet, path, asyncRecord)}. - - Append {deferredFragmentRecord} to {subsequentPayloads}. + - Call {ExecuteDeferredFragment(label, objectType, objectValue, + fragmentSelectionSet, path, asyncRecord, subsequentPayloads)}. - Continue with the next {selection} in {selectionSet}. - Let {fragmentGroupedFieldSet} be the result of calling {CollectFields(objectType, objectValue, fragmentSelectionSet, @@ -689,50 +678,31 @@ An Async Payload Record is either a Deferred Fragment Record or a Stream Record. All Async Payload Records are structures containing: - {label}: value derived from the corresponding `@defer` or `@stream` directive. -- {parentRecord}: optionally an Async Payload Record. +- {path}: a list of field names and indices from root to the location of the + corresponding `@defer` or `@stream` directive. +- {iterator}: The underlying iterator if created from a `@stream` directive. - {errors}: a list of field errors encountered during execution. - {dataExecution}: A result that can notify when the corresponding execution has completed. -- {path}: a list of field names and indices from root to the location of the - corresponding `@defer` or `@stream` directive. - -#### Deferred Fragment Record - -Let {deferredFragmentRecord} be an inline fragment or fragment spread with -`@defer` provided. - -Deferred Fragment Record is a structure containing all the entries of Async -Payload Record, and additionally: -- {objectType}: of the {deferredFragmentRecord}. -- {objectValue}: of the {deferredFragmentRecord}. -- {fragmentSelectionSet}: the top level selection set of - {deferredFragmentRecord}. +#### Execute Deferred Fragment -CreateDeferredFragmentRecord(label, objectType, objectValue, -fragmentSelectionSet, path, parentRecord): +ExecuteDeferredFragment(label, objectType, objectValue, fragmentSelectionSet, +path, variableValues, parentRecord, subsequentPayloads): -- Construct a deferred fragment record based on the parameters passed in. -- Initialize {errors} to an empty list. - -ResolveDeferredFragmentRecord(deferredFragmentRecord, variableValues, -subsequentPayloads): - -- Let {label}, {objectType}, {objectValue}, {fragmentSelectionSet}, {path}, - {parentRecord} be the corresponding fields in the deferred fragment record - structure. +- Let {deferRecord} be an async payload record created from {label} and {path}. +- Initialize {errors} on {deferRecord} to an empty list. - Let {dataExecution} be the asynchronous future value of: - Let {payload} be the result of {ExecuteSelectionSet(fragmentSelectionSet, objectType, objectValue, variableValues, path, subsequentPayloads, - deferredFragmentRecord)}. + deferRecord)}. - If {parentRecord} is defined: - Wait for the result of {dataExecution} on {parentRecord}. + - Add an entry to {payload} named `label` with the value {label}. + - Add an entry to {payload} named `path` with the value {path}. - Return {payload}. - Set {dataExecution} on {deferredFragmentRecord}. -- Let {payload} be the result of waiting for {dataExecution}. -- Add an entry to {payload} named `label` with the value {label}. -- Add an entry to {payload} named `path` with the value {path}. -- Return {payload}. +- Append {deferRecord} to {subsequentPayloads}. ## Executing Fields @@ -876,28 +846,14 @@ field, value completion iterates over the iterator until the number of items yield by the iterator satisfies `initialCount` specified on the `@stream` directive. -#### Stream Record - -Let {streamField} be a list field with a `@stream` directive provided. - -A Stream Record is a structure containing all the entries of Async Payload -Record, and additionally: - -- {iterator}: created by {ResolveFieldGenerator}. -- {index}: indicating the position of the item in the complete list. -- {fields}: the group of fields grouped by CollectFields() for {streamField}. -- {innerType}: inner type of {streamField}'s type. - -CreateStreamRecord(label, iterator, index, fields, innerType, path, -parentRecord): - -- Construct a stream record based on the parameters passed in. -- Initialize {errors} to an empty list. +#### Execute Stream Field -ResolveStreamRecord(streamRecord, variableValues, subsequentPayloads): +ExecuteStreamRecord(label, iterator, index, fields, innerType, path +streamRecord, variableValues, subsequentPayloads): -- Let {label}, {parentRecord}, {iterator}, {index}, {path}, {fields}, - {innerType} be the correspondent fields on the Stream Record structure. +- Let {streamRecord} be an async payload record created from {label}, {path}, + and {iterator}. +- Initialize {errors} on {streamRecord} to an empty list. - Let {indexPath} be {path} with {index} appended. - Let {dataExecution} be the asynchronous future value of: - Wait for the next item from {iterator}. @@ -907,17 +863,15 @@ ResolveStreamRecord(streamRecord, variableValues, subsequentPayloads): - Let {payload} be the result of calling {CompleteValue(innerType, fields, item, variableValues, indexPath, subsequentPayloads, parentRecord)}. - Increment {index}. - - Let {nextStreamRecord} be the result of calling {CreateStreamRecord(label, - iterator, index, fields, innerType, path, streamRecord)}. - - Append {nextStreamRecord} to {subsequentPayloads}. + - Call {ExecuteStreamRecord(label, iterator, index, fields, innerType, path, + streamRecord, variableValues, subsequentPayloads)}. - If {parentRecord} is defined: - Wait for the result of {dataExecution} on {parentRecord}. + - Add an entry to {payload} named `label` with the value {label}. + - Add an entry to {payload} named `path` with the value {indexPath}. - Return {payload}. - Set {dataExecution} on {streamRecord}. -- Let {payload} be the result of waiting for {dataExecution}. -- Add an entry to {payload} named `label` with the value {label}. -- Add an entry to {payload} named `path` with the value {indexPath}. -- Return {payload}. +- Append {streamRecord} to {subsequentPayloads}. CompleteValue(fieldType, fields, result, variableValues, path, subsequentPayloads, asyncRecord): @@ -955,9 +909,8 @@ subsequentPayloads, asyncRecord): - Increment {index}. - If {streamDirective} was provided and {index} is greater than or equal to {initialCount}: - - Let {streamRecord} be the result of calling {CreateStreamRecord(label, - result, index, fields, innerType, path, asyncRecord)}. - - Append {streamRecord} to {subsequentPayloads}. + - Call {ExecuteStreamRecord(label, result, index, fields, innerType, + path, asyncRecord, subsequentPayloads)}. - Let {result} be {initialItems}. - Exit while loop. - Return {initialItems}. From 101516b7d7b53900f785f4dcf339543b2b771c88 Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Mon, 20 Dec 2021 16:08:58 -0500 Subject: [PATCH 15/62] Clarify error handling # Conflicts: # spec/Section 6 -- Execution.md --- spec/Section 6 -- Execution.md | 28 +++++++++++++++++++--------- 1 file changed, 19 insertions(+), 9 deletions(-) diff --git a/spec/Section 6 -- Execution.md b/spec/Section 6 -- Execution.md index f7920e622..ca848af82 100644 --- a/spec/Section 6 -- Execution.md +++ b/spec/Section 6 -- Execution.md @@ -393,9 +393,9 @@ YieldSubsequentPayloads(subsequentPayloads): - Return. - If {subsequentPayloads} is not empty: - Continue to the next record in {subsequentPayloads}. - - If {record} is not the final element in {subsequentPayloads} + - If {record} is not the final element in {subsequentPayloads}: - Add an entry to {payload} named `hasNext` with the value {true}. - - If {record} is the final element in {subsequentPayloads} + - If {record} is the final element in {subsequentPayloads}: - Add an entry to {payload} named `hasNext` with the value {false}. - Yield {payload} @@ -693,11 +693,16 @@ path, variableValues, parentRecord, subsequentPayloads): - Let {deferRecord} be an async payload record created from {label} and {path}. - Initialize {errors} on {deferRecord} to an empty list. - Let {dataExecution} be the asynchronous future value of: - - Let {payload} be the result of {ExecuteSelectionSet(fragmentSelectionSet, + - Let {payload} be an unordered map. + - Let {data} be the result of {ExecuteSelectionSet(fragmentSelectionSet, objectType, objectValue, variableValues, path, subsequentPayloads, deferRecord)}. + - Append any encountered field errors to {errors}. - If {parentRecord} is defined: - Wait for the result of {dataExecution} on {parentRecord}. + - If {errors} is not empty: + - Add an entry to {payload} named `errors` with the value {errors}. + - Add an entry to {payload} named `data` with the value {data}. - Add an entry to {payload} named `label` with the value {label}. - Add an entry to {payload} named `path` with the value {path}. - Return {payload}. @@ -848,8 +853,8 @@ directive. #### Execute Stream Field -ExecuteStreamRecord(label, iterator, index, fields, innerType, path -streamRecord, variableValues, subsequentPayloads): +ExecuteStreamField(label, iterator, index, fields, innerType, path streamRecord, +variableValues, subsequentPayloads): - Let {streamRecord} be an async payload record created from {label}, {path}, and {iterator}. @@ -859,14 +864,19 @@ streamRecord, variableValues, subsequentPayloads): - Wait for the next item from {iterator}. - If an item is not retrieved because {iterator} has completed: - Return {null}. + - Let {payload} be an unordered map. - Let {item} be the item retrieved from {iterator}. - - Let {payload} be the result of calling {CompleteValue(innerType, fields, - item, variableValues, indexPath, subsequentPayloads, parentRecord)}. + - Let {data} be the result of calling {CompleteValue(innerType, fields, item, + variableValues, indexPath, subsequentPayloads, parentRecord)}. + - Append any encountered field errors to {errors}. - Increment {index}. - - Call {ExecuteStreamRecord(label, iterator, index, fields, innerType, path, + - Call {ExecuteStreamField(label, iterator, index, fields, innerType, path, streamRecord, variableValues, subsequentPayloads)}. - If {parentRecord} is defined: - Wait for the result of {dataExecution} on {parentRecord}. + - If {errors} is not empty: + - Add an entry to {payload} named `errors` with the value {errors}. + - Add an entry to {payload} named `data` with the value {data}. - Add an entry to {payload} named `label` with the value {label}. - Add an entry to {payload} named `path` with the value {indexPath}. - Return {payload}. @@ -909,7 +919,7 @@ subsequentPayloads, asyncRecord): - Increment {index}. - If {streamDirective} was provided and {index} is greater than or equal to {initialCount}: - - Call {ExecuteStreamRecord(label, result, index, fields, innerType, + - Call {ExecuteStreamField(label, result, index, fields, innerType, path, asyncRecord, subsequentPayloads)}. - Let {result} be {initialItems}. - Exit while loop. From 80ff450997e75751c3f37da8c624d0cda814a170 Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Thu, 30 Dec 2021 12:57:17 -0500 Subject: [PATCH 16/62] add isCompletedIterator to AsyncPayloadRecord to track completed iterators # Conflicts: # spec/Section 6 -- Execution.md --- spec/Section 6 -- Execution.md | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/spec/Section 6 -- Execution.md b/spec/Section 6 -- Execution.md index ca848af82..94c33485e 100644 --- a/spec/Section 6 -- Execution.md +++ b/spec/Section 6 -- Execution.md @@ -386,13 +386,13 @@ YieldSubsequentPayloads(subsequentPayloads): - Let {record} be the first item in {subsequentPayloads} with a completed {dataExecution}. - Remove {record} from {subsequentPayloads}. - - Let {payload} be the completed result returned by {dataExecution}. - - If {payload} is {null}: + - If {isCompletedIterator} on {record} is {true}: - If {subsequentPayloads} is empty: - Yield a map containing a field `hasNext` with the value {false}. - Return. - If {subsequentPayloads} is not empty: - Continue to the next record in {subsequentPayloads}. + - Let {payload} be the completed result returned by {dataExecution}. - If {record} is not the final element in {subsequentPayloads}: - Add an entry to {payload} named `hasNext` with the value {true}. - If {record} is the final element in {subsequentPayloads}: @@ -681,6 +681,8 @@ All Async Payload Records are structures containing: - {path}: a list of field names and indices from root to the location of the corresponding `@defer` or `@stream` directive. - {iterator}: The underlying iterator if created from a `@stream` directive. +- {isCompletedIterator}: a boolean indicating the payload record was generated + from an iterator that has completed. - {errors}: a list of field errors encountered during execution. - {dataExecution}: A result that can notify when the corresponding execution has completed. @@ -863,6 +865,7 @@ variableValues, subsequentPayloads): - Let {dataExecution} be the asynchronous future value of: - Wait for the next item from {iterator}. - If an item is not retrieved because {iterator} has completed: + - Set {isCompletedIterator} to {true} on {streamRecord}. - Return {null}. - Let {payload} be an unordered map. - Let {item} be the item retrieved from {iterator}. From b73af22d3c54da1c88e4c6d6890b99b37dd3da48 Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Fri, 21 Jan 2022 07:42:28 -0500 Subject: [PATCH 17/62] fix typo --- spec/Section 6 -- Execution.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/spec/Section 6 -- Execution.md b/spec/Section 6 -- Execution.md index 94c33485e..77f6110bd 100644 --- a/spec/Section 6 -- Execution.md +++ b/spec/Section 6 -- Execution.md @@ -600,7 +600,7 @@ subsequentPayloads, asyncRecord, visitedFragments): - Append {selection} to the {groupForResponseKey}. - If {selection} is a {FragmentSpread}: - Let {fragmentSpreadName} be the name of {selection}. - - If {fragmentSpreadName} provides the directive `@defer` and it's {if} + - If {fragmentSpreadName} provides the directive `@defer` and its {if} argument is {true} or is a variable in {variableValues} with the value {true}: - Let {deferDirective} be that directive. From f634192c098589e6dc955b1529fe337538762d0d Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Wed, 2 Feb 2022 14:45:30 -0500 Subject: [PATCH 18/62] deferDirective and visitedFragments --- spec/Section 6 -- Execution.md | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/spec/Section 6 -- Execution.md b/spec/Section 6 -- Execution.md index 77f6110bd..874ffbb0b 100644 --- a/spec/Section 6 -- Execution.md +++ b/spec/Section 6 -- Execution.md @@ -604,9 +604,10 @@ subsequentPayloads, asyncRecord, visitedFragments): argument is {true} or is a variable in {variableValues} with the value {true}: - Let {deferDirective} be that directive. - - If {fragmentSpreadName} is in {visitedFragments} and {deferDirective} is - not defined, continue with the next {selection} in {selectionSet}. - - Add {fragmentSpreadName} to {visitedFragments}. + - If {deferDirective} is not defined: + - If {fragmentSpreadName} is in {visitedFragments}, continue with the next + {selection} in {selectionSet}. + - Add {fragmentSpreadName} to {visitedFragments}. - Let {fragment} be the Fragment in the current Document whose name is {fragmentSpreadName}. - If no such {fragment} exists, continue with the next {selection} in From 7eafac873417eecaf0a7cf2d34c93bf1f94a8dce Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Mon, 7 Feb 2022 15:33:51 -0500 Subject: [PATCH 19/62] stream if argument, indexPath -> itemPath --- spec/Section 6 -- Execution.md | 24 +++++++++++++----------- 1 file changed, 13 insertions(+), 11 deletions(-) diff --git a/spec/Section 6 -- Execution.md b/spec/Section 6 -- Execution.md index 874ffbb0b..59b57ee80 100644 --- a/spec/Section 6 -- Execution.md +++ b/spec/Section 6 -- Execution.md @@ -862,7 +862,7 @@ variableValues, subsequentPayloads): - Let {streamRecord} be an async payload record created from {label}, {path}, and {iterator}. - Initialize {errors} on {streamRecord} to an empty list. -- Let {indexPath} be {path} with {index} appended. +- Let {itemPath} be {path} with {index} appended. - Let {dataExecution} be the asynchronous future value of: - Wait for the next item from {iterator}. - If an item is not retrieved because {iterator} has completed: @@ -871,7 +871,7 @@ variableValues, subsequentPayloads): - Let {payload} be an unordered map. - Let {item} be the item retrieved from {iterator}. - Let {data} be the result of calling {CompleteValue(innerType, fields, item, - variableValues, indexPath, subsequentPayloads, parentRecord)}. + variableValues, itemPath, subsequentPayloads, parentRecord)}. - Append any encountered field errors to {errors}. - Increment {index}. - Call {ExecuteStreamField(label, iterator, index, fields, innerType, path, @@ -882,7 +882,7 @@ variableValues, subsequentPayloads): - Add an entry to {payload} named `errors` with the value {errors}. - Add an entry to {payload} named `data` with the value {data}. - Add an entry to {payload} named `label` with the value {label}. - - Add an entry to {payload} named `path` with the value {indexPath}. + - Add an entry to {payload} named `path` with the value {itemPath}. - Return {payload}. - Set {dataExecution} on {streamRecord}. - Append {streamRecord} to {subsequentPayloads}. @@ -902,7 +902,9 @@ subsequentPayloads, asyncRecord): - If {result} is an iterator: - Let {field} be the first entry in {fields}. - Let {innerType} be the inner type of {fieldType}. - - Let {streamDirective} be the `@stream` directive provided on {field}. + - If {field} provides the directive `@stream` and its {if} argument is + {true} or is a variable in {variableValues} with the value {true} and : + - Let {streamDirective} be that directive. - Let {initialCount} be the value or variable provided to {streamDirective}'s {initialCount} argument. - If {initialCount} is less than zero, raise a _field error_. @@ -911,18 +913,18 @@ subsequentPayloads, asyncRecord): - Let {initialItems} be an empty list - Let {index} be zero. - While {result} is not closed: - - If {streamDirective} was not provided or {index} is not greater than or + - If {streamDirective} is not defined or {index} is not greater than or equal to {initialCount}: - Wait for the next item from {result}. - Let {resultItem} be the item retrieved from {result}. - - Let {indexPath} be {path} with {index} appended. + - Let {itemPath} be {path} with {index} appended. - Let {resolvedItem} be the result of calling {CompleteValue(innerType, - fields, resultItem, variableValues, indexPath, subsequentPayloads, + fields, resultItem, variableValues, itemPath, subsequentPayloads, asyncRecord)}. - Append {resolvedItem} to {initialItems}. - Increment {index}. - - If {streamDirective} was provided and {index} is greater than or equal - to {initialCount}: + - If {streamDirective} is defined and {index} is greater than or equal to + {initialCount}: - Call {ExecuteStreamField(label, result, index, fields, innerType, path, asyncRecord, subsequentPayloads)}. - Let {result} be {initialItems}. @@ -931,9 +933,9 @@ subsequentPayloads, asyncRecord): - If {result} is not a collection of values, raise a _field error_. - Let {innerType} be the inner type of {fieldType}. - Return a list where each list item is the result of calling - {CompleteValue(innerType, fields, resultItem, variableValues, indexPath, + {CompleteValue(innerType, fields, resultItem, variableValues, itemPath, subsequentPayloads, asyncRecord)}, where {resultItem} is each item in - {result} and {indexPath} is {path} with the index of the item appended. + {result} and {itemPath} is {path} with the index of the item appended. - If {fieldType} is a Scalar or Enum type: - Return the result of {CoerceResult(fieldType, result)}. - If {fieldType} is an Object, Interface, or Union type: From 83e058d945a96d619efe35b4db1cefced6d06f5f Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Mon, 7 Feb 2022 15:43:48 -0500 Subject: [PATCH 20/62] Clarify stream only applies to outermost list of multi-dimensional arrays --- spec/Section 6 -- Execution.md | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/spec/Section 6 -- Execution.md b/spec/Section 6 -- Execution.md index 59b57ee80..35fc37f0e 100644 --- a/spec/Section 6 -- Execution.md +++ b/spec/Section 6 -- Execution.md @@ -903,7 +903,9 @@ subsequentPayloads, asyncRecord): - Let {field} be the first entry in {fields}. - Let {innerType} be the inner type of {fieldType}. - If {field} provides the directive `@stream` and its {if} argument is - {true} or is a variable in {variableValues} with the value {true} and : + {true} or is a variable in {variableValues} with the value {true} and + {innerType} is the outermost return type of the list type defined for + {field}: - Let {streamDirective} be that directive. - Let {initialCount} be the value or variable provided to {streamDirective}'s {initialCount} argument. From 5aa915b7c910ad41e44e01096d38b13dbb573442 Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Mon, 7 Mar 2022 16:30:36 -0500 Subject: [PATCH 21/62] =?UTF-8?q?add=20validation=20=E2=80=9CDefer=20And?= =?UTF-8?q?=20Stream=20Directive=20Labels=20Are=20Unique=E2=80=9D?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- spec/Section 5 -- Validation.md | 65 +++++++++++++++++++++++++++++++++ 1 file changed, 65 insertions(+) diff --git a/spec/Section 5 -- Validation.md b/spec/Section 5 -- Validation.md index 8303ce776..018d7b91a 100644 --- a/spec/Section 5 -- Validation.md +++ b/spec/Section 5 -- Validation.md @@ -1556,6 +1556,71 @@ mutation { } ``` +### Defer And Stream Directive Labels Are Unique + +** Formal Specification ** + +- For every {directive} in a document. +- Let {directiveName} be the name of {directive}. +- If {directiveName} is "defer" or "stream": + - For every {argument} in {directive}: + - Let {argumentName} be the name of {argument}. + - Let {argumentValue} be the value passed to {argument}. + - If {argumentName} is "label": + - {argumentValue} must not be a variable. + - Let {labels} be all label values passed to defer or stream directive label + arguments. + - {labels} must be a set of one. + +**Explanatory Text** + +The `@defer` and `@stream` directives each accept an argument "label". This +label may be used by GraphQL clients to uniquely identify response payloads. If +a label is passed, it must not be a variable and it must be unique within all +other `@defer` and `@stream` directives in the document. + +For example the following document is valid: + +```graphql example +{ + dog { + ...fragmentOne + ...fragmentTwo @defer(label: "dogDefer") + } + pets @stream(label: "petStream") { + name + } +} + +fragment fragmentOne on Dog { + name +} + +fragment fragmentTwo on Dog { + owner { + name + } +} +``` + +For example, the following document will not pass validation because the same +label is used in different `@defer` and `@stream` directives.: + +```raw graphql counter-example +{ + dog { + ...fragmentOne @defer(label: "MyLabel") + } + pets @stream(label: "MyLabel") { + name + } +} + +fragment fragmentOne on Dog { + name +} +``` + ### Stream Directives Are Used On List Fields **Formal Specification** From d7fe43aa0a754aea52a17cf681d31df97df1ad5f Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Tue, 8 Mar 2022 10:41:32 -0500 Subject: [PATCH 22/62] Clarification on labels --- spec/Section 3 -- Type System.md | 19 +++++++++++-------- 1 file changed, 11 insertions(+), 8 deletions(-) diff --git a/spec/Section 3 -- Type System.md b/spec/Section 3 -- Type System.md index 4a25974b0..49ecd38e0 100644 --- a/spec/Section 3 -- Type System.md +++ b/spec/Section 3 -- Type System.md @@ -2206,10 +2206,11 @@ fragment someFragment on User { - `if: Boolean` - When `true`, fragment _should_ be deferred. When `false`, fragment will not be deferred and data will be included in the initial response. If omitted, defaults to `true`. -- `label: String` - A unique label across all `@defer` and `@stream` directives - in an operation. This label should be used by GraphQL clients to identify the - data from patch responses and associate it with the correct fragments. If - provided, the GraphQL Server must add it to the payload. +- `label: String` - May be used by GraphQL clients to identify the data from + responses and associate it with the corresponding defer directive. If + provided, the GraphQL Server must add it to the corresponding payload. `label` + must be unique label across all `@defer` and `@stream` directives in a + document. `label` must not be provided as a variable. ### @stream @@ -2237,10 +2238,12 @@ query myQuery($shouldStream: Boolean) { - `if: Boolean` - When `true`, field _should_ be streamed. When `false`, the field will not be streamed and all list items will be included in the initial response. If omitted, defaults to `true`. -- `label: String` - A unique label across all `@defer` and `@stream` directives - in an operation. This label should be used by GraphQL clients to identify the - data from patch responses and associate it with the correct fragments. If - provided, the GraphQL Server must add it to the payload. +- `label: String` - May be used by GraphQL clients to identify the data from + responses and associate it with the corresponding stream directive. If + provided, the GraphQL Server must add it to the corresponding payload. `label` + must be unique label across all `@defer` and `@stream` directives in a + document. `label` must not be provided as a variable. + - `initialCount: Int` - The number of list items the server should return as part of the initial response. If omitted, defaults to `0`. A field error will be raised if the value of this argument is less than `0`. From 379f10caf01ac8734a0326ff257613a0121703fb Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Wed, 23 Mar 2022 14:42:50 -0400 Subject: [PATCH 23/62] fix wrong quotes --- spec/Section 3 -- Type System.md | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/spec/Section 3 -- Type System.md b/spec/Section 3 -- Type System.md index 49ecd38e0..771c5c0b0 100644 --- a/spec/Section 3 -- Type System.md +++ b/spec/Section 3 -- Type System.md @@ -2188,10 +2188,10 @@ delivered in a subsequent response. `@include` and `@skip` take precedence over ```graphql example query myQuery($shouldDefer: Boolean) { - user { - name - ...someFragment @defer(label: 'someLabel', if: $shouldDefer) - } + user { + name + ...someFragment @defer(label: "someLabel", if: $shouldDefer) + } } fragment someFragment on User { id From 1ffad84f9594a6b350caad061855fc643aeff146 Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Wed, 23 Mar 2022 14:45:42 -0400 Subject: [PATCH 24/62] remove label/path requirement --- spec/Section 7 -- Response.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/spec/Section 7 -- Response.md b/spec/Section 7 -- Response.md index 47ee7e893..690c8a016 100644 --- a/spec/Section 7 -- Response.md +++ b/spec/Section 7 -- Response.md @@ -26,8 +26,8 @@ When the response of the GraphQL operation is an event stream, the first value will be the initial response. All subsequent values may contain `label` and `path` entries. These two entries are used by clients to identify the `@defer` or `@stream` directive from the GraphQL operation that triggered this value to -be returned by the event stream. The combination of these two entries must be -unique across all values returned by the event stream. +be returned by the event stream. When a label is provided, the combination of +these two entries will be unique across all values returned by the event stream. If the response of the GraphQL operation is an event stream, each response map must contain an entry with key `hasNext`. The value of this entry is `true` for From 21dd0bc62b2d57ce48173d440e7675f17cd2baa1 Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Thu, 9 Jun 2022 17:26:33 -0500 Subject: [PATCH 25/62] add missing line --- spec/Section 6 -- Execution.md | 1 + 1 file changed, 1 insertion(+) diff --git a/spec/Section 6 -- Execution.md b/spec/Section 6 -- Execution.md index 35fc37f0e..4b9ba73c1 100644 --- a/spec/Section 6 -- Execution.md +++ b/spec/Section 6 -- Execution.md @@ -30,6 +30,7 @@ request is determined by the result of executing this operation according to the "Executing Operations” section below. ExecuteRequest(schema, document, operationName, variableValues, initialValue): + Note: the execution assumes implementing language supports coroutines. Alternatively, the socket can provide a write buffer pointer to allow {ExecuteRequest()} to directly write payloads into the buffer. From 3a569b6d79dfb57ccd251f9ea98cb953b844dbee Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Thu, 9 Jun 2022 17:28:59 -0500 Subject: [PATCH 26/62] fix ExecuteRequest --- spec/Section 6 -- Execution.md | 9 ++------- 1 file changed, 2 insertions(+), 7 deletions(-) diff --git a/spec/Section 6 -- Execution.md b/spec/Section 6 -- Execution.md index 4b9ba73c1..18eb21a45 100644 --- a/spec/Section 6 -- Execution.md +++ b/spec/Section 6 -- Execution.md @@ -39,13 +39,8 @@ Alternatively, the socket can provide a write buffer pointer to allow - Let {coercedVariableValues} be the result of {CoerceVariableValues(schema, operation, variableValues)}. - If {operation} is a query operation: - - Let {executionResult} be the result of calling {ExecuteQuery(operation, - schema, coercedVariableValues, initialValue, subsequentPayloads)}. - - If {executionResult} is an iterator: - - For each {payload} in {executionResult}: - - Yield {payload}. - - Otherwise: - - Return {executionResult}. + - Return {ExecuteQuery(operation, schema, coercedVariableValues, + initialValue)}. - Otherwise if {operation} is a mutation operation: - Return {ExecuteMutation(operation, schema, coercedVariableValues, initialValue)}. From a596eadb76b1521bf53d7c6f1e8d4fbd4ce24229 Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Thu, 9 Jun 2022 17:31:06 -0500 Subject: [PATCH 27/62] fix response --- spec/Section 7 -- Response.md | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/spec/Section 7 -- Response.md b/spec/Section 7 -- Response.md index 690c8a016..3743d4f8d 100644 --- a/spec/Section 7 -- Response.md +++ b/spec/Section 7 -- Response.md @@ -12,9 +12,9 @@ the case that any _field error_ was raised on a field and was replaced with A response to a GraphQL request must be a map or an event stream of maps. -If the operation encountered any errors, the response map must contain an entry -with key `errors`. The value of this entry is described in the "Errors" section. -If the request completed without raising any errors, this entry must not be +If the request raised any errors, the response map must contain an entry with +key `errors`. The value of this entry is described in the "Errors" section. If +the request completed without raising any errors, this entry must not be present. If the request included execution, the response map must contain an entry with From 4543a479c3e7b1414f63985701878b7eb90a3b54 Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Wed, 3 Aug 2022 14:41:10 -0400 Subject: [PATCH 28/62] Align deferred fragment field collection with reference implementation --- spec/Section 6 -- Execution.md | 57 ++++++++++++++++++++++++---------- 1 file changed, 40 insertions(+), 17 deletions(-) diff --git a/spec/Section 6 -- Execution.md b/spec/Section 6 -- Execution.md index 18eb21a45..2ae221126 100644 --- a/spec/Section 6 -- Execution.md +++ b/spec/Section 6 -- Execution.md @@ -410,8 +410,9 @@ subsequentPayloads, asyncRecord): - If {path} is not provided, initialize it to an empty list. - If {subsequentPayloads} is not provided, initialize it to the empty set. -- Let {groupedFieldSet} be the result of {CollectFields(objectType, objectValue, - selectionSet, variableValues, path subsequentPayloads, asyncRecord)}. +- Let {groupedFieldSet} and {deferredGroupedFieldsList} be the result of + {CollectFields(objectType, objectValue, selectionSet, variableValues, path, + asyncRecord)}. - Initialize {resultMap} to an empty ordered map. - For each {groupedFieldSet} as {responseKey} and {fields}: - Let {fieldName} be the name of the first entry in {fields}. Note: This value @@ -422,6 +423,10 @@ subsequentPayloads, asyncRecord): - Let {responseValue} be {ExecuteField(objectType, objectValue, fieldType, fields, variableValues, path, subsequentPayloads, asyncRecord)}. - Set {responseValue} as the value for {responseKey} in {resultMap}. +- For each {deferredGroupFieldSet} and {label} in {deferredGroupedFieldsList} + - Call {ExecuteDeferredFragment(label, objectType, objectValue, + deferredGroupFieldSet, path, variableValues, asyncRecord, + subsequentPayloads)} - Return {resultMap}. Note: {resultMap} is ordered by which fields appear first in the operation. This @@ -573,10 +578,12 @@ is maintained through execution, ensuring that fields appear in the executed response in a stable and predictable order. CollectFields(objectType, objectValue, selectionSet, variableValues, path, -subsequentPayloads, asyncRecord, visitedFragments): +asyncRecord, visitedFragments, deferredGroupedFieldsList): - If {visitedFragments} is not provided, initialize it to the empty set. - Initialize {groupedFields} to an empty ordered map of lists. +- If {deferredGroupedFieldsList} is not provided, initialize it to an empty + list. - For each {selection} in {selectionSet}: - If {selection} provides the directive `@skip`, let {skipDirective} be that directive. @@ -615,13 +622,17 @@ subsequentPayloads, asyncRecord, visitedFragments): - If {deferDirective} is defined: - Let {label} be the value or the variable to {deferDirective}'s {label} argument. - - Call {ExecuteDeferredFragment(label, objectType, objectValue, - fragmentSelectionSet, path, variableValues, asyncRecord, - subsequentPayloads)}. + - Let {deferredGroupedFields} be the result of calling + {CollectFields(objectType, objectValue, fragmentSelectionSet, + variableValues, path, asyncRecord, visitedFragments, + deferredGroupedFieldsList)}. + - Append a record containing {label} and {deferredGroupedFields} to + {deferredGroupedFieldsList}. - Continue with the next {selection} in {selectionSet}. - Let {fragmentGroupedFieldSet} be the result of calling {CollectFields(objectType, objectValue, fragmentSelectionSet, - variableValues, path, subsequentPayloads, asyncRecord, visitedFragments)}. + variableValues, path, asyncRecord, visitedFragments, + deferredGroupedFieldsList)}. - For each {fragmentGroup} in {fragmentGroupedFieldSet}: - Let {responseKey} be the response key shared by all fields in {fragmentGroup}. @@ -640,19 +651,24 @@ subsequentPayloads, asyncRecord, visitedFragments): {variableValues} with the value {true}: - Let {label} be the value or the variable to {deferDirective}'s {label} argument. - - Call {ExecuteDeferredFragment(label, objectType, objectValue, - fragmentSelectionSet, path, asyncRecord, subsequentPayloads)}. + - Let {deferredGroupedFields} be the result of calling + {CollectFields(objectType, objectValue, fragmentSelectionSet, + variableValues, path, asyncRecord, visitedFragments, + deferredGroupedFieldsList)}. + - Append a record containing {label} and {deferredGroupedFields} to + {deferredGroupedFieldsList}. - Continue with the next {selection} in {selectionSet}. - Let {fragmentGroupedFieldSet} be the result of calling {CollectFields(objectType, objectValue, fragmentSelectionSet, - variableValues, path, subsequentPayloads, asyncRecord, visitedFragments)}. + variableValues, path, asyncRecord, visitedFragments, + deferredGroupedFieldsList)}. - For each {fragmentGroup} in {fragmentGroupedFieldSet}: - Let {responseKey} be the response key shared by all fields in {fragmentGroup}. - Let {groupForResponseKey} be the list in {groupedFields} for {responseKey}; if no such list exists, create it as an empty list. - Append all items in {fragmentGroup} to {groupForResponseKey}. -- Return {groupedFields}. +- Return {groupedFields} and {deferredGroupedFieldsList}. Note: The steps in {CollectFields()} evaluating the `@skip` and `@include` directives may be applied in either order since they apply commutatively. @@ -686,22 +702,29 @@ All Async Payload Records are structures containing: #### Execute Deferred Fragment -ExecuteDeferredFragment(label, objectType, objectValue, fragmentSelectionSet, -path, variableValues, parentRecord, subsequentPayloads): +ExecuteDeferredFragment(label, objectType, objectValue, groupedFieldSet, path, +variableValues, parentRecord, subsequentPayloads): - Let {deferRecord} be an async payload record created from {label} and {path}. - Initialize {errors} on {deferRecord} to an empty list. - Let {dataExecution} be the asynchronous future value of: - Let {payload} be an unordered map. - - Let {data} be the result of {ExecuteSelectionSet(fragmentSelectionSet, - objectType, objectValue, variableValues, path, subsequentPayloads, - deferRecord)}. + - Initialize {resultMap} to an empty ordered map. + - For each {groupedFieldSet} as {responseKey} and {fields}: + - Let {fieldName} be the name of the first entry in {fields}. Note: This + value is unaffected if an alias is used. + - Let {fieldType} be the return type defined for the field {fieldName} of + {objectType}. + - If {fieldType} is defined: + - Let {responseValue} be {ExecuteField(objectType, objectValue, fieldType, + fields, variableValues, path, subsequentPayloads, asyncRecord)}. + - Set {responseValue} as the value for {responseKey} in {resultMap}. - Append any encountered field errors to {errors}. - If {parentRecord} is defined: - Wait for the result of {dataExecution} on {parentRecord}. - If {errors} is not empty: - Add an entry to {payload} named `errors` with the value {errors}. - - Add an entry to {payload} named `data` with the value {data}. + - Add an entry to {payload} named `data` with the value {resultMap}. - Add an entry to {payload} named `label` with the value {label}. - Add an entry to {payload} named `path` with the value {path}. - Return {payload}. From 5a093cda76c550fb1167dea0d6117c5f6e45b373 Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Thu, 18 Aug 2022 14:20:15 -0400 Subject: [PATCH 29/62] spec updates to reflect latest discussions --- spec/Section 6 -- Execution.md | 64 ++++++++++++++---------- spec/Section 7 -- Response.md | 91 +++++++++++++++++++++++++--------- 2 files changed, 107 insertions(+), 48 deletions(-) diff --git a/spec/Section 6 -- Execution.md b/spec/Section 6 -- Execution.md index 2ae221126..83d631f43 100644 --- a/spec/Section 6 -- Execution.md +++ b/spec/Section 6 -- Execution.md @@ -144,10 +144,10 @@ ExecuteQuery(query, schema, variableValues, initialValue): - If {subsequentPayloads} is empty: - Return an unordered map containing {data} and {errors}. - If {subsequentPayloads} is not empty: - - Yield an unordered map containing {data}, {errors}, and an entry named - {hasNext} with the value {true}. + - Let {intialResponse} be an unordered map containing {data}, {errors}, and an + entry named {hasNext} with the value {true}. - Let {iterator} be the result of running - {YieldSubsequentPayloads(subsequentPayloads)}. + {YieldSubsequentPayloads(intialResponse, subsequentPayloads)}. - For each {payload} yielded by {iterator}: - If a termination signal is received: - Send a termination signal to {iterator}. @@ -178,10 +178,10 @@ ExecuteMutation(mutation, schema, variableValues, initialValue): - If {subsequentPayloads} is empty: - Return an unordered map containing {data} and {errors}. - If {subsequentPayloads} is not empty: - - Yield an unordered map containing {data}, {errors}, and an entry named - {hasNext} with the value {true}. + - Let {intialResponse} be an unordered map containing {data}, {errors}, and an + entry named {hasNext} with the value {true}. - Let {iterator} be the result of running - {YieldSubsequentPayloads(subsequentPayloads)}. + {YieldSubsequentPayloads(intialResponse, subsequentPayloads)}. - For each {payload} yielded by {iterator}: - If a termination signal is received: - Send a termination signal to {iterator}. @@ -340,10 +340,10 @@ ExecuteSubscriptionEvent(subscription, schema, variableValues, initialValue): - If {subsequentPayloads} is empty: - Return an unordered map containing {data} and {errors}. - If {subsequentPayloads} is not empty: - - Yield an unordered map containing {data}, {errors}, and an entry named - {hasNext} with the value {true}. + - Let {intialResponse} be an unordered map containing {data}, {errors}, and an + entry named {hasNext} with the value {true}. - Let {iterator} be the result of running - {YieldSubsequentPayloads(subsequentPayloads)}. + {YieldSubsequentPayloads(intialResponse, subsequentPayloads)}. - For each {payload} yielded by {iterator}: - If a termination signal is received: - Send a termination signal to {iterator}. @@ -371,29 +371,42 @@ If an operation contains subsequent payload records resulting from `@stream` or `@defer` directives, the {YieldSubsequentPayloads} algorithm defines how the payloads should be processed. -YieldSubsequentPayloads(subsequentPayloads): +YieldSubsequentPayloads(intialResponse, subsequentPayloads): +- Let {initialRecords} be any items in {subsequentPayloads} with a completed + {dataExecution}. +- Initialize {initialIncremental} to an empty list. +- For each {record} in {initialRecords}: + - Remove {record} from {subsequentPayloads}. + - If {isCompletedIterator} on {record} is {true}: + - Continue to the next record in {records}. + - Let {payload} be the completed result returned by {dataExecution}. + - Append {payload} to {initialIncremental}. +- If {initialIncremental} is not empty: + - Add an entry to {intialResponse} named `incremental` containing the value + {incremental}. +- Yield {intialResponse}. - While {subsequentPayloads} is not empty: - If a termination signal is received: - For each {record} in {subsequentPayloads}: - If {record} contains {iterator}: - Send a termination signal to {iterator}. - Return. -- Let {record} be the first item in {subsequentPayloads} with a completed +- Wait for at least one record in {subsequentPayloads} to have a completed {dataExecution}. - - Remove {record} from {subsequentPayloads}. - - If {isCompletedIterator} on {record} is {true}: - - If {subsequentPayloads} is empty: - - Yield a map containing a field `hasNext` with the value {false}. - - Return. - - If {subsequentPayloads} is not empty: - - Continue to the next record in {subsequentPayloads}. - - Let {payload} be the completed result returned by {dataExecution}. - - If {record} is not the final element in {subsequentPayloads}: - - Add an entry to {payload} named `hasNext` with the value {true}. - - If {record} is the final element in {subsequentPayloads}: - - Add an entry to {payload} named `hasNext` with the value {false}. - - Yield {payload} +- Let {subsequentResponse} be an unordered map with an entry {incremental} + initialized to an empty list. +- Let {records} be the items in {subsequentPayloads} with a completed + {dataExecution}. + - For each {record} in {records}: + - Remove {record} from {subsequentPayloads}. + - If {isCompletedIterator} on {record} is {true}: + - Continue to the next record in {records}. + - Let {payload} be the completed result returned by {dataExecution}. + - Append {payload} to {incremental}. + - If {subsequentPayloads} is empty: + - Add an entry to {response} named `hasNext` with the value {false}. + - Yield {response} ## Executing Selection Sets @@ -899,7 +912,8 @@ variableValues, subsequentPayloads): - Wait for the result of {dataExecution} on {parentRecord}. - If {errors} is not empty: - Add an entry to {payload} named `errors` with the value {errors}. - - Add an entry to {payload} named `data` with the value {data}. + - Add an entry to {payload} named `items` with a list containing the value + {data}. - Add an entry to {payload} named `label` with the value {label}. - Add an entry to {payload} named `path` with the value {itemPath}. - Return {payload}. diff --git a/spec/Section 7 -- Response.md b/spec/Section 7 -- Response.md index 3743d4f8d..b2fe30a9f 100644 --- a/spec/Section 7 -- Response.md +++ b/spec/Section 7 -- Response.md @@ -23,11 +23,14 @@ request failed before execution, due to a syntax error, missing information, or validation error, this entry must not be present. When the response of the GraphQL operation is an event stream, the first value -will be the initial response. All subsequent values may contain `label` and -`path` entries. These two entries are used by clients to identify the `@defer` -or `@stream` directive from the GraphQL operation that triggered this value to -be returned by the event stream. When a label is provided, the combination of -these two entries will be unique across all values returned by the event stream. +will be the initial response. All subsequent values may contain an `incremental` +entry, containing a list of Defer or Stream responses. + +The `label` and `path` entries on Defer and Stream responses are used by clients +to identify the `@defer` or `@stream` directive from the GraphQL operation that +triggered this response to be included in an `incremental` entry on a value +returned by the event stream. When a label is provided, the combination of these +two entries will be unique across all values returned by the event stream. If the response of the GraphQL operation is an event stream, each response map must contain an entry with key `hasNext`. The value of this entry is `true` for @@ -45,11 +48,13 @@ set, must have a map as its value. This entry is reserved for implementors to extend the protocol however they see fit, and hence there are no additional restrictions on its contents. When the response of the GraphQL operation is an event stream, implementors may send subsequent payloads containing only -`hasNext` and `extensions` entries. +`hasNext` and `extensions` entries. Defer and Stream responses may also contain +an entry with the key `extensions`, also reserved for implementors to extend the +protocol however they see fit. To ensure future changes to the protocol do not break existing services and clients, the top level response map must not contain any entries other than the -three described above. +five described above. Note: When `errors` is present in the response, it may be helpful for it to appear first when serialized to make it more clear when errors are present in a @@ -62,11 +67,6 @@ requested operation. If the operation was a query, this output will be an object of the query root operation type; if the operation was a mutation, this output will be an object of the mutation root operation type. -If the result of the operation is an event stream, the `data` entry in -subsequent values will be of the type of a particular field in the GraphQL -result. The adjacent `path` field will contain the path segments of the field -this data is associated with. - If an error was raised before execution begins, the `data` entry should not be present in the result. @@ -263,7 +263,43 @@ discouraged. } ``` -## Path +### Incremental + +The `incremental` entry in the response is a non-empty list of Defer or Stream +responses. If the response of the GraphQL operation is an event stream, this +field may appear on both the initial and subsequent values. + +#### Stream response + +A stream response is a map that may appear as an item in the `incremental` entry +of a response. A stream response is the result of an associated `@stream` +directive in the operation. A stream response must contain `items` and `path` +entries and may contain `label`, `errors`, and `extensions` entries. + +##### Items + +The `items` entry in a stream response is a list of results from the execution +of the associated @stream directive. This output will be a list of the same type +of the field with the associated `@stream` directive. If `items` is set to +`null`, it indicates that an error has caused a `null` to bubble up to a field +higher than the list field with the associated `@stream` directive. + +#### Defer response + +A defer response is a map that may appear as an item in the `incremental` entry +of a response. A defer response is the result of an associated `@defer` +directive in the operation. A defer response must contain `data` and `path` +entries and may contain `label`, `errors`, and `extensions` entries. + +##### Data + +The `data` entry in a Defer response will be of the type of a particular field +in the GraphQL result. The adjacent `path` field will contain the path segments +of the field this data is associated with. If `data` is set to `null`, it +indicates that an error has caused a `null` to bubble up to a field higher than +the field that contains the fragment with the associated `@defer` directive. + +#### Path A `path` field allows for the association to a particular field in a GraphQL result. This field should be a list of path segments starting at the root of the @@ -273,21 +309,30 @@ indices should be 0-indexed integers. If the path is associated to an aliased field, the path should use the aliased name, since it represents a path in the response, not in the request. -When the `path` field is present on a GraphQL response, it indicates that the -`data` field is not the root query or mutation result, but is rather associated -to a particular field in the root result. +When the `path` field is present on a Stream response, it indicates that the +`items` field represents the partial result of the list field containing the +corresponding `@stream` directive. All but the non-final path segments must +refer to the location of the list field containing the corresponding `@stream` +directive. The final segment of the path list must be a 0-indexed integer. This +integer indicates that this result is set at a range, where the beginning of the +range is at the index of this integer, and the length of the range is the length +of the data. + +When the `path` field is present on a Defer response, it indicates that the +`data` field represents the result of the fragment containing the corresponding +`@defer` directive. The path segments must point to the location of the result +of the field containing the associated `@defer` directive. When the `path` field is present on an "Error result", it indicates the response field which experienced the error. -## Label +#### Label -If the response of the GraphQL operation is an event stream, subsequent values -may contain a string field `label`. This `label` is the same label passed to the -`@defer` or `@stream` directive that triggered this value. This allows clients -to identify which `@defer` or `@stream` directive is associated with this value. -`label` will not be present if the corresponding `@defer` or `@stream` directive -is not passed a `label` argument. +Stream and Defer responses may contain a string field `label`. This `label` is +the same label passed to the `@defer` or `@stream` directive associated with the +response. This allows clients to identify which `@defer` or `@stream` directive +is associated with this value. `label` will not be present if the corresponding +`@defer` or `@stream` directive is not passed a `label` argument. ## Serialization Format From 8fa426eda1f2bf3d7a16d1534311ad1dc0154b0b Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Thu, 18 Aug 2022 14:26:33 -0400 Subject: [PATCH 30/62] Note about mutation execution order --- spec/Section 6 -- Execution.md | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/spec/Section 6 -- Execution.md b/spec/Section 6 -- Execution.md index 83d631f43..f4756002c 100644 --- a/spec/Section 6 -- Execution.md +++ b/spec/Section 6 -- Execution.md @@ -553,6 +553,11 @@ A correct executor must generate the following result for that selection set: } ``` +When subsections contain a `@stream` or `@defer` directive, these subsections +are no longer required to execute serially. Exeuction of the deferred or +streamed sections of the subsection may be executed in parallel, as defined in +{ExecuteStreamField} and {ExecuteDeferredFragment}. + ### Field Collection Before execution, the selection set is converted to a grouped field set by From fcc63936ee1603a06e7bb033a78681a65c3fb884 Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Thu, 18 Aug 2022 14:38:06 -0400 Subject: [PATCH 31/62] minor change for uniqueness --- spec/Section 7 -- Response.md | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/spec/Section 7 -- Response.md b/spec/Section 7 -- Response.md index b2fe30a9f..32dedbe14 100644 --- a/spec/Section 7 -- Response.md +++ b/spec/Section 7 -- Response.md @@ -30,7 +30,8 @@ The `label` and `path` entries on Defer and Stream responses are used by clients to identify the `@defer` or `@stream` directive from the GraphQL operation that triggered this response to be included in an `incremental` entry on a value returned by the event stream. When a label is provided, the combination of these -two entries will be unique across all values returned by the event stream. +two entries will be unique across all Defer and Stream responses returned in the +event stream. If the response of the GraphQL operation is an event stream, each response map must contain an entry with key `hasNext`. The value of this entry is `true` for From aa22c5ff0b78eb3edf9384b41c59aebf01815267 Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Thu, 18 Aug 2022 16:57:09 -0400 Subject: [PATCH 32/62] fix typos --- spec/Section 6 -- Execution.md | 26 +++++++++++++------------- 1 file changed, 13 insertions(+), 13 deletions(-) diff --git a/spec/Section 6 -- Execution.md b/spec/Section 6 -- Execution.md index f4756002c..1945225c7 100644 --- a/spec/Section 6 -- Execution.md +++ b/spec/Section 6 -- Execution.md @@ -144,10 +144,10 @@ ExecuteQuery(query, schema, variableValues, initialValue): - If {subsequentPayloads} is empty: - Return an unordered map containing {data} and {errors}. - If {subsequentPayloads} is not empty: - - Let {intialResponse} be an unordered map containing {data}, {errors}, and an - entry named {hasNext} with the value {true}. + - Let {initialResponse} be an unordered map containing {data}, {errors}, and + an entry named {hasNext} with the value {true}. - Let {iterator} be the result of running - {YieldSubsequentPayloads(intialResponse, subsequentPayloads)}. + {YieldSubsequentPayloads(initialResponse, subsequentPayloads)}. - For each {payload} yielded by {iterator}: - If a termination signal is received: - Send a termination signal to {iterator}. @@ -178,10 +178,10 @@ ExecuteMutation(mutation, schema, variableValues, initialValue): - If {subsequentPayloads} is empty: - Return an unordered map containing {data} and {errors}. - If {subsequentPayloads} is not empty: - - Let {intialResponse} be an unordered map containing {data}, {errors}, and an - entry named {hasNext} with the value {true}. + - Let {initialResponse} be an unordered map containing {data}, {errors}, and + an entry named {hasNext} with the value {true}. - Let {iterator} be the result of running - {YieldSubsequentPayloads(intialResponse, subsequentPayloads)}. + {YieldSubsequentPayloads(initialResponse, subsequentPayloads)}. - For each {payload} yielded by {iterator}: - If a termination signal is received: - Send a termination signal to {iterator}. @@ -340,10 +340,10 @@ ExecuteSubscriptionEvent(subscription, schema, variableValues, initialValue): - If {subsequentPayloads} is empty: - Return an unordered map containing {data} and {errors}. - If {subsequentPayloads} is not empty: - - Let {intialResponse} be an unordered map containing {data}, {errors}, and an - entry named {hasNext} with the value {true}. + - Let {initialResponse} be an unordered map containing {data}, {errors}, and + an entry named {hasNext} with the value {true}. - Let {iterator} be the result of running - {YieldSubsequentPayloads(intialResponse, subsequentPayloads)}. + {YieldSubsequentPayloads(initialResponse, subsequentPayloads)}. - For each {payload} yielded by {iterator}: - If a termination signal is received: - Send a termination signal to {iterator}. @@ -371,7 +371,7 @@ If an operation contains subsequent payload records resulting from `@stream` or `@defer` directives, the {YieldSubsequentPayloads} algorithm defines how the payloads should be processed. -YieldSubsequentPayloads(intialResponse, subsequentPayloads): +YieldSubsequentPayloads(initialResponse, subsequentPayloads): - Let {initialRecords} be any items in {subsequentPayloads} with a completed {dataExecution}. @@ -383,9 +383,9 @@ YieldSubsequentPayloads(intialResponse, subsequentPayloads): - Let {payload} be the completed result returned by {dataExecution}. - Append {payload} to {initialIncremental}. - If {initialIncremental} is not empty: - - Add an entry to {intialResponse} named `incremental` containing the value + - Add an entry to {initialResponse} named `incremental` containing the value {incremental}. -- Yield {intialResponse}. +- Yield {initialResponse}. - While {subsequentPayloads} is not empty: - If a termination signal is received: - For each {record} in {subsequentPayloads}: @@ -554,7 +554,7 @@ A correct executor must generate the following result for that selection set: ``` When subsections contain a `@stream` or `@defer` directive, these subsections -are no longer required to execute serially. Exeuction of the deferred or +are no longer required to execute serially. Execution of the deferred or streamed sections of the subsection may be executed in parallel, as defined in {ExecuteStreamField} and {ExecuteDeferredFragment}. From 089d06c75fd8b498cd49fbf8fd7667ea5a9d6fb1 Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Tue, 23 Aug 2022 12:25:56 -0400 Subject: [PATCH 33/62] if: Boolean! = true --- spec/Section 3 -- Type System.md | 21 ++++++++++++--------- 1 file changed, 12 insertions(+), 9 deletions(-) diff --git a/spec/Section 3 -- Type System.md b/spec/Section 3 -- Type System.md index 771c5c0b0..e0ff53adc 100644 --- a/spec/Section 3 -- Type System.md +++ b/spec/Section 3 -- Type System.md @@ -2174,7 +2174,7 @@ scalar UUID @specifiedBy(url: "https://tools.ietf.org/html/rfc4122") ```graphql directive @defer( label: String - if: Boolean + if: Boolean! = true ) on FRAGMENT_SPREAD | INLINE_FRAGMENT ``` @@ -2203,9 +2203,9 @@ fragment someFragment on User { #### @defer Arguments -- `if: Boolean` - When `true`, fragment _should_ be deferred. When `false`, - fragment will not be deferred and data will be included in the initial - response. If omitted, defaults to `true`. +- `if: Boolean! = true` - When `true`, fragment _should_ be deferred. When + `false`, fragment will not be deferred and data will be included in the + initial response. If omitted, defaults to `true`. - `label: String` - May be used by GraphQL clients to identify the data from responses and associate it with the corresponding defer directive. If provided, the GraphQL Server must add it to the corresponding payload. `label` @@ -2215,7 +2215,11 @@ fragment someFragment on User { ### @stream ```graphql -directive @stream(label: String, initialCount: Int = 0, if: Boolean) on FIELD +directive @stream( + label: String + if: Boolean! = true + initialCount: Int = 0 +) on FIELD ``` The `@stream` directive may be provided for a field of `List` type so that the @@ -2235,15 +2239,14 @@ query myQuery($shouldStream: Boolean) { #### @stream Arguments -- `if: Boolean` - When `true`, field _should_ be streamed. When `false`, the - field will not be streamed and all list items will be included in the initial - response. If omitted, defaults to `true`. +- `if: Boolean! = true` - When `true`, field _should_ be streamed. When `false`, + the field will not be streamed and all list items will be included in the + initial response. If omitted, defaults to `true`. - `label: String` - May be used by GraphQL clients to identify the data from responses and associate it with the corresponding stream directive. If provided, the GraphQL Server must add it to the corresponding payload. `label` must be unique label across all `@defer` and `@stream` directives in a document. `label` must not be provided as a variable. - - `initialCount: Int` - The number of list items the server should return as part of the initial response. If omitted, defaults to `0`. A field error will be raised if the value of this argument is less than `0`. From a154bda6aa3942750fd2bc83598792eb47f61659 Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Tue, 23 Aug 2022 19:59:38 -0400 Subject: [PATCH 34/62] address pr feedback --- spec/Section 6 -- Execution.md | 32 ++++++++++++++++++-------------- 1 file changed, 18 insertions(+), 14 deletions(-) diff --git a/spec/Section 6 -- Execution.md b/spec/Section 6 -- Execution.md index 1945225c7..b472b3dfc 100644 --- a/spec/Section 6 -- Execution.md +++ b/spec/Section 6 -- Execution.md @@ -387,26 +387,30 @@ YieldSubsequentPayloads(initialResponse, subsequentPayloads): {incremental}. - Yield {initialResponse}. - While {subsequentPayloads} is not empty: -- If a termination signal is received: - - For each {record} in {subsequentPayloads}: - - If {record} contains {iterator}: - - Send a termination signal to {iterator}. - - Return. -- Wait for at least one record in {subsequentPayloads} to have a completed - {dataExecution}. -- Let {subsequentResponse} be an unordered map with an entry {incremental} - initialized to an empty list. -- Let {records} be the items in {subsequentPayloads} with a completed - {dataExecution}. + - If a termination signal is received: + - For each {record} in {subsequentPayloads}: + - If {record} contains {iterator}: + - Send a termination signal to {iterator}. + - Return. + - Wait for at least one record in {subsequentPayloads} to have a completed + {dataExecution}. + - Let {subsequentResponse} be an unordered map with an entry {incremental} + initialized to an empty list. + - Let {records} be the items in {subsequentPayloads} with a completed + {dataExecution}. - For each {record} in {records}: - Remove {record} from {subsequentPayloads}. - If {isCompletedIterator} on {record} is {true}: - Continue to the next record in {records}. - Let {payload} be the completed result returned by {dataExecution}. - - Append {payload} to {incremental}. + - Append {payload} to the {incremental} entry on {subsequentResponse}. - If {subsequentPayloads} is empty: - - Add an entry to {response} named `hasNext` with the value {false}. - - Yield {response} + - Add an entry to {subsequentResponse} named `hasNext` with the value + {false}. + - Otherwise, if {subsequentPayloads} is not empty: + - Add an entry to {subsequentResponse} named `hasNext` with the value + {true}. + - Yield {subsequentResponse} ## Executing Selection Sets From 67689c5a8bb1864c47c8f461a4537869f05fa202 Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Tue, 23 Aug 2022 20:02:18 -0400 Subject: [PATCH 35/62] clarify null behavior of if --- spec/Section 3 -- Type System.md | 4 ++-- spec/Section 6 -- Execution.md | 37 ++++++++++++++++---------------- 2 files changed, 21 insertions(+), 20 deletions(-) diff --git a/spec/Section 3 -- Type System.md b/spec/Section 3 -- Type System.md index e0ff53adc..1852c4bbf 100644 --- a/spec/Section 3 -- Type System.md +++ b/spec/Section 3 -- Type System.md @@ -2205,7 +2205,7 @@ fragment someFragment on User { - `if: Boolean! = true` - When `true`, fragment _should_ be deferred. When `false`, fragment will not be deferred and data will be included in the - initial response. If omitted, defaults to `true`. + initial response. Defaults to `true` when omitted or null. - `label: String` - May be used by GraphQL clients to identify the data from responses and associate it with the corresponding defer directive. If provided, the GraphQL Server must add it to the corresponding payload. `label` @@ -2241,7 +2241,7 @@ query myQuery($shouldStream: Boolean) { - `if: Boolean! = true` - When `true`, field _should_ be streamed. When `false`, the field will not be streamed and all list items will be included in the - initial response. If omitted, defaults to `true`. + initial response. Defaults to `true` when omitted or null. - `label: String` - May be used by GraphQL clients to identify the data from responses and associate it with the corresponding stream directive. If provided, the GraphQL Server must add it to the corresponding payload. `label` diff --git a/spec/Section 6 -- Execution.md b/spec/Section 6 -- Execution.md index b472b3dfc..f93a3c86e 100644 --- a/spec/Section 6 -- Execution.md +++ b/spec/Section 6 -- Execution.md @@ -626,8 +626,8 @@ asyncRecord, visitedFragments, deferredGroupedFieldsList): - If {selection} is a {FragmentSpread}: - Let {fragmentSpreadName} be the name of {selection}. - If {fragmentSpreadName} provides the directive `@defer` and its {if} - argument is {true} or is a variable in {variableValues} with the value - {true}: + argument is not {false} and is not a variable in {variableValues} with the + value {false}: - Let {deferDirective} be that directive. - If {deferDirective} is not defined: - If {fragmentSpreadName} is in {visitedFragments}, continue with the next @@ -667,19 +667,20 @@ asyncRecord, visitedFragments, deferredGroupedFieldsList): fragmentType)} is false, continue with the next {selection} in {selectionSet}. - Let {fragmentSelectionSet} be the top-level selection set of {selection}. - - If {InlineFragment} provides the directive `@defer`, let {deferDirective} - be that directive. - - If {deferDirective}'s {if} argument is {true} or is a variable in - {variableValues} with the value {true}: - - Let {label} be the value or the variable to {deferDirective}'s {label} - argument. - - Let {deferredGroupedFields} be the result of calling - {CollectFields(objectType, objectValue, fragmentSelectionSet, - variableValues, path, asyncRecord, visitedFragments, - deferredGroupedFieldsList)}. - - Append a record containing {label} and {deferredGroupedFields} to - {deferredGroupedFieldsList}. - - Continue with the next {selection} in {selectionSet}. + - If {InlineFragment} provides the directive `@defer` and its {if} argument + is not {false} and is not a variable in {variableValues} with the value + {false}: + - Let {deferDirective} be that directive. + - If {deferDirective} is defined: + - Let {label} be the value or the variable to {deferDirective}'s {label} + argument. + - Let {deferredGroupedFields} be the result of calling + {CollectFields(objectType, objectValue, fragmentSelectionSet, + variableValues, path, asyncRecord, visitedFragments, + deferredGroupedFieldsList)}. + - Append a record containing {label} and {deferredGroupedFields} to + {deferredGroupedFieldsList}. + - Continue with the next {selection} in {selectionSet}. - Let {fragmentGroupedFieldSet} be the result of calling {CollectFields(objectType, objectValue, fragmentSelectionSet, variableValues, path, asyncRecord, visitedFragments, @@ -944,9 +945,9 @@ subsequentPayloads, asyncRecord): - If {result} is an iterator: - Let {field} be the first entry in {fields}. - Let {innerType} be the inner type of {fieldType}. - - If {field} provides the directive `@stream` and its {if} argument is - {true} or is a variable in {variableValues} with the value {true} and - {innerType} is the outermost return type of the list type defined for + - If {field} provides the directive `@stream` and its {if} argument is not + {false} and is not a variable in {variableValues} with the value {false} + and {innerType} is the outermost return type of the list type defined for {field}: - Let {streamDirective} be that directive. - Let {initialCount} be the value or variable provided to From 9230085f180891c43b8298c0e506c6862657bdce Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Thu, 8 Sep 2022 11:03:48 -0400 Subject: [PATCH 36/62] Add error boundary behavior --- spec/Section 6 -- Execution.md | 36 +++++++++++++++++++++++++++++----- 1 file changed, 31 insertions(+), 5 deletions(-) diff --git a/spec/Section 6 -- Execution.md b/spec/Section 6 -- Execution.md index f93a3c86e..f0f933b58 100644 --- a/spec/Section 6 -- Execution.md +++ b/spec/Section 6 -- Execution.md @@ -455,8 +455,11 @@ If during {ExecuteSelectionSet()} a field with a non-null {fieldType} raises a _field error_ then that error must propagate to this entire selection set, either resolving to {null} if allowed or further propagated to a parent field. -If this occurs, any sibling fields which have not yet executed or have not yet -yielded a value may be cancelled to avoid unnecessary work. +If this occurs, any defer or stream executions with a path that starts with the +same path as the resolved {null} must not return their results to the client. +These defer or stream executions or any sibling fields which have not yet +executed or have not yet yielded a value may be cancelled to avoid unnecessary +work. Note: See [Handling Field Errors](#sec-Handling-Field-Errors) for more about this behavior. @@ -747,7 +750,11 @@ variableValues, parentRecord, subsequentPayloads): - Wait for the result of {dataExecution} on {parentRecord}. - If {errors} is not empty: - Add an entry to {payload} named `errors` with the value {errors}. - - Add an entry to {payload} named `data` with the value {resultMap}. + - If a field error was raised, causing a {null} to be propagated to + {responseValue}, and {objectType} is a Non-Nullable type: + - Add an entry to {payload} named `data` with the value {null}. + - Otherwise: + - Add an entry to {payload} named `data` with the value {resultMap}. - Add an entry to {payload} named `label` with the value {label}. - Add an entry to {payload} named `path` with the value {path}. - Return {payload}. @@ -922,8 +929,12 @@ variableValues, subsequentPayloads): - Wait for the result of {dataExecution} on {parentRecord}. - If {errors} is not empty: - Add an entry to {payload} named `errors` with the value {errors}. - - Add an entry to {payload} named `items` with a list containing the value - {data}. + - If a field error was raised, causing a {null} to be propagated to {data}, + and {innerType} is a Non-Nullable type: + - Add an entry to {payload} named `items` with the value {null}. + - Otherwise: + - Add an entry to {payload} named `items` with a list containing the value + {data}. - Add an entry to {payload} named `label` with the value {label}. - Add an entry to {payload} named `path` with the value {itemPath}. - Return {payload}. @@ -1099,6 +1110,21 @@ resolves to {null}, then the entire list must resolve to {null}. If the `List` type is also wrapped in a `Non-Null`, the field error continues to propagate upwards. +When a field error is raised inside `ExecuteDeferredFragment` or +`ExecuteStreamField`, the defer and stream payloads act as error boundaries. +That is, the null resulting from a `Non-Null` type cannot propagate outside of +the boundary of the defer or stream payload. + +If a fragment with the `defer` directive is spread on a Non-Nullable object +type, and a field error has caused a {null} to propagate to the associated +object, the {null} should not propagate any further, and the associated Defer +Payload's `data` field must be set to {null}. + +If the `stream` directive is present on a list field with a Non-Nullable inner +type, and a field error has caused a {null} to propagate to the list item, the +{null} should not propagate any further, and the associated Stream Payload's +`item` field must be set to {null}. + If all fields from the root of the request to the source of the field error return `Non-Null` types, then the {"data"} entry in the response should be {null}. From 4bf09d0306496029cfddc1f47e1c56602ecb4327 Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Thu, 8 Sep 2022 11:22:29 -0400 Subject: [PATCH 37/62] defer/stream response => payload --- spec/Section 7 -- Response.md | 54 +++++++++++++++++------------------ 1 file changed, 27 insertions(+), 27 deletions(-) diff --git a/spec/Section 7 -- Response.md b/spec/Section 7 -- Response.md index 32dedbe14..b8db8364e 100644 --- a/spec/Section 7 -- Response.md +++ b/spec/Section 7 -- Response.md @@ -24,13 +24,13 @@ validation error, this entry must not be present. When the response of the GraphQL operation is an event stream, the first value will be the initial response. All subsequent values may contain an `incremental` -entry, containing a list of Defer or Stream responses. +entry, containing a list of Defer or Stream payloads. -The `label` and `path` entries on Defer and Stream responses are used by clients +The `label` and `path` entries on Defer and Stream payloads are used by clients to identify the `@defer` or `@stream` directive from the GraphQL operation that triggered this response to be included in an `incremental` entry on a value returned by the event stream. When a label is provided, the combination of these -two entries will be unique across all Defer and Stream responses returned in the +two entries will be unique across all Defer and Stream payloads returned in the event stream. If the response of the GraphQL operation is an event stream, each response map @@ -49,7 +49,7 @@ set, must have a map as its value. This entry is reserved for implementors to extend the protocol however they see fit, and hence there are no additional restrictions on its contents. When the response of the GraphQL operation is an event stream, implementors may send subsequent payloads containing only -`hasNext` and `extensions` entries. Defer and Stream responses may also contain +`hasNext` and `extensions` entries. Defer and Stream payloads may also contain an entry with the key `extensions`, also reserved for implementors to extend the protocol however they see fit. @@ -267,38 +267,38 @@ discouraged. ### Incremental The `incremental` entry in the response is a non-empty list of Defer or Stream -responses. If the response of the GraphQL operation is an event stream, this +payloads. If the response of the GraphQL operation is an event stream, this field may appear on both the initial and subsequent values. -#### Stream response +#### Stream payload -A stream response is a map that may appear as an item in the `incremental` entry -of a response. A stream response is the result of an associated `@stream` -directive in the operation. A stream response must contain `items` and `path` +A stream payload is a map that may appear as an item in the `incremental` entry +of a response. A stream payload is the result of an associated `@stream` +directive in the operation. A stream payload must contain `items` and `path` entries and may contain `label`, `errors`, and `extensions` entries. ##### Items -The `items` entry in a stream response is a list of results from the execution -of the associated @stream directive. This output will be a list of the same type -of the field with the associated `@stream` directive. If `items` is set to -`null`, it indicates that an error has caused a `null` to bubble up to a field -higher than the list field with the associated `@stream` directive. +The `items` entry in a stream payload is a list of results from the execution of +the associated @stream directive. This output will be a list of the same type of +the field with the associated `@stream` directive. If `items` is set to `null`, +it indicates that an error has caused a `null` to bubble up to a field higher +than the list field with the associated `@stream` directive. -#### Defer response +#### Defer payload -A defer response is a map that may appear as an item in the `incremental` entry -of a response. A defer response is the result of an associated `@defer` -directive in the operation. A defer response must contain `data` and `path` -entries and may contain `label`, `errors`, and `extensions` entries. +A defer payload is a map that may appear as an item in the `incremental` entry +of a response. A defer payload is the result of an associated `@defer` directive +in the operation. A defer payload must contain `data` and `path` entries and may +contain `label`, `errors`, and `extensions` entries. ##### Data -The `data` entry in a Defer response will be of the type of a particular field -in the GraphQL result. The adjacent `path` field will contain the path segments -of the field this data is associated with. If `data` is set to `null`, it -indicates that an error has caused a `null` to bubble up to a field higher than -the field that contains the fragment with the associated `@defer` directive. +The `data` entry in a Defer payload will be of the type of a particular field in +the GraphQL result. The adjacent `path` field will contain the path segments of +the field this data is associated with. If `data` is set to `null`, it indicates +that an error has caused a `null` to bubble up to a field higher than the field +that contains the fragment with the associated `@defer` directive. #### Path @@ -310,7 +310,7 @@ indices should be 0-indexed integers. If the path is associated to an aliased field, the path should use the aliased name, since it represents a path in the response, not in the request. -When the `path` field is present on a Stream response, it indicates that the +When the `path` field is present on a Stream payload, it indicates that the `items` field represents the partial result of the list field containing the corresponding `@stream` directive. All but the non-final path segments must refer to the location of the list field containing the corresponding `@stream` @@ -319,7 +319,7 @@ integer indicates that this result is set at a range, where the beginning of the range is at the index of this integer, and the length of the range is the length of the data. -When the `path` field is present on a Defer response, it indicates that the +When the `path` field is present on a Defer payload, it indicates that the `data` field represents the result of the fragment containing the corresponding `@defer` directive. The path segments must point to the location of the result of the field containing the associated `@defer` directive. @@ -329,7 +329,7 @@ field which experienced the error. #### Label -Stream and Defer responses may contain a string field `label`. This `label` is +Stream and Defer payloads may contain a string field `label`. This `label` is the same label passed to the `@defer` or `@stream` directive associated with the response. This allows clients to identify which `@defer` or `@stream` directive is associated with this value. `label` will not be present if the corresponding From 1499e640582dd8e5cfdfe43d24b7a993d784514d Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Thu, 8 Sep 2022 11:12:12 -0400 Subject: [PATCH 38/62] event stream => response stream --- spec/Section 7 -- Response.md | 24 ++++++++++++------------ 1 file changed, 12 insertions(+), 12 deletions(-) diff --git a/spec/Section 7 -- Response.md b/spec/Section 7 -- Response.md index b8db8364e..bc656d353 100644 --- a/spec/Section 7 -- Response.md +++ b/spec/Section 7 -- Response.md @@ -10,7 +10,7 @@ the case that any _field error_ was raised on a field and was replaced with ## Response Format -A response to a GraphQL request must be a map or an event stream of maps. +A response to a GraphQL request must be a map or a response stream of maps. If the request raised any errors, the response map must contain an entry with key `errors`. The value of this entry is described in the "Errors" section. If @@ -22,33 +22,33 @@ key `data`. The value of this entry is described in the "Data" section. If the request failed before execution, due to a syntax error, missing information, or validation error, this entry must not be present. -When the response of the GraphQL operation is an event stream, the first value +When the response of the GraphQL operation is a response stream, the first value will be the initial response. All subsequent values may contain an `incremental` entry, containing a list of Defer or Stream payloads. The `label` and `path` entries on Defer and Stream payloads are used by clients to identify the `@defer` or `@stream` directive from the GraphQL operation that triggered this response to be included in an `incremental` entry on a value -returned by the event stream. When a label is provided, the combination of these -two entries will be unique across all Defer and Stream payloads returned in the -event stream. +returned by the response stream. When a label is provided, the combination of +these two entries will be unique across all Defer and Stream payloads returned +in the response stream. -If the response of the GraphQL operation is an event stream, each response map +If the response of the GraphQL operation is a response stream, each response map must contain an entry with key `hasNext`. The value of this entry is `true` for all but the last response in the stream. The value of this entry is `false` for the last response of the stream. This entry is not required for GraphQL operations that return a single response map. -The GraphQL server may determine there are no more values in the event stream +The GraphQL server may determine there are no more values in the response stream after a previous value with `hasNext` equal to `true` has been emitted. In this -case the last value in the event stream should be a map without `data`, `label`, -and `path` entries, and a `hasNext` entry with a value of `false`. +case the last value in the response stream should be a map without `data`, +`label`, and `path` entries, and a `hasNext` entry with a value of `false`. The response map may also contain an entry with key `extensions`. This entry, if set, must have a map as its value. This entry is reserved for implementors to extend the protocol however they see fit, and hence there are no additional -restrictions on its contents. When the response of the GraphQL operation is an -event stream, implementors may send subsequent payloads containing only +restrictions on its contents. When the response of the GraphQL operation is a +response stream, implementors may send subsequent response maps containing only `hasNext` and `extensions` entries. Defer and Stream payloads may also contain an entry with the key `extensions`, also reserved for implementors to extend the protocol however they see fit. @@ -267,7 +267,7 @@ discouraged. ### Incremental The `incremental` entry in the response is a non-empty list of Defer or Stream -payloads. If the response of the GraphQL operation is an event stream, this +payloads. If the response of the GraphQL operation is a response stream, this field may appear on both the initial and subsequent values. #### Stream payload From 722de48877321ea65e524774284687b4ce48661c Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Thu, 8 Sep 2022 11:14:27 -0400 Subject: [PATCH 39/62] link to path section --- spec/Section 7 -- Response.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/spec/Section 7 -- Response.md b/spec/Section 7 -- Response.md index bc656d353..0be0b5e03 100644 --- a/spec/Section 7 -- Response.md +++ b/spec/Section 7 -- Response.md @@ -134,7 +134,7 @@ If an error can be associated to a particular field in the GraphQL result, it must contain an entry with the key `path` that details the path of the response field which experienced the error. This allows clients to identify whether a `null` result is intentional or caused by a runtime error. The value of this -field is described in the "Path" section. +field is described in the [Path](#sec-Path) section. For example, if fetching one of the friends' names fails in the following operation: From c4bdfaff92a16fdde62931d4521d0578463fad89 Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Thu, 8 Sep 2022 11:15:16 -0400 Subject: [PATCH 40/62] use case no dash --- spec/Section 3 -- Type System.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/spec/Section 3 -- Type System.md b/spec/Section 3 -- Type System.md index 1852c4bbf..ccbe4b01c 100644 --- a/spec/Section 3 -- Type System.md +++ b/spec/Section 3 -- Type System.md @@ -2255,8 +2255,8 @@ Note: The ability to defer and/or stream parts of a response can have a potentially significant impact on application performance. Developers generally need clear, predictable control over their application's performance. It is highly recommended that GraphQL servers honor the `@defer` and `@stream` -directives on each execution. However, the specification allows advanced -use-cases where the server can determine that it is more performant to not defer +directives on each execution. However, the specification allows advanced use +cases where the server can determine that it is more performant to not defer and/or stream. Therefore, GraphQL clients _must_ be able to process a response that ignores the `@defer` and/or `@stream` directives. This also applies to the `initialCount` argument on the `@stream` directive. Clients _must_ be able to From d84dd0b1b6e9e369a3b987b80fa44a6034f2104e Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Thu, 8 Sep 2022 11:15:56 -0400 Subject: [PATCH 41/62] remove "or null" --- spec/Section 3 -- Type System.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/spec/Section 3 -- Type System.md b/spec/Section 3 -- Type System.md index ccbe4b01c..0d4b3aab0 100644 --- a/spec/Section 3 -- Type System.md +++ b/spec/Section 3 -- Type System.md @@ -2205,7 +2205,7 @@ fragment someFragment on User { - `if: Boolean! = true` - When `true`, fragment _should_ be deferred. When `false`, fragment will not be deferred and data will be included in the - initial response. Defaults to `true` when omitted or null. + initial response. Defaults to `true` when omitted. - `label: String` - May be used by GraphQL clients to identify the data from responses and associate it with the corresponding defer directive. If provided, the GraphQL Server must add it to the corresponding payload. `label` @@ -2241,7 +2241,7 @@ query myQuery($shouldStream: Boolean) { - `if: Boolean! = true` - When `true`, field _should_ be streamed. When `false`, the field will not be streamed and all list items will be included in the - initial response. Defaults to `true` when omitted or null. + initial response. Defaults to `true` when omitted. - `label: String` - May be used by GraphQL clients to identify the data from responses and associate it with the corresponding stream directive. If provided, the GraphQL Server must add it to the corresponding payload. `label` From 2248f35bfe78fe028383435ebba4d63f7782c316 Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Thu, 8 Sep 2022 11:31:04 -0400 Subject: [PATCH 42/62] add detailed incremental example --- spec/Section 7 -- Response.md | 80 +++++++++++++++++++++++++++++++++++ 1 file changed, 80 insertions(+) diff --git a/spec/Section 7 -- Response.md b/spec/Section 7 -- Response.md index 0be0b5e03..477c964ca 100644 --- a/spec/Section 7 -- Response.md +++ b/spec/Section 7 -- Response.md @@ -270,6 +270,86 @@ The `incremental` entry in the response is a non-empty list of Defer or Stream payloads. If the response of the GraphQL operation is a response stream, this field may appear on both the initial and subsequent values. +For example, a query containing both defer and stream: + +```graphql example +query { + person(id: "cGVvcGxlOjE=") { + ...HomeWorldFragment @defer(label: "homeWorldDefer") + name + films @stream(initialCount: 1, label: "filmsStream") { + title + } + } +} +fragment HomeWorldFragment on Person { + homeWorld { + name + } +} +``` + +The response stream might look like: + +Response 1, the initial response does not contain any deferred or streamed +results. + +```json example +{ + "data": { + "person": { + "name": "Luke Skywalker", + "films": [{ "title": "A New Hope" }] + } + }, + "hasNext": true +} +``` + +Response 2, contains the defer payload and the first stream payload. + +```json example +{ + "incremental": [ + { + "label": "homeWorldDefer", + "path": ["person"], + "data": { "homeWorld": { "name": "Tatooine" } } + }, + { + "label": "filmsStream", + "path": ["person", "films", 1], + "items": [{ "title": "The Empire Strikes Back" }] + } + ], + "hasNext": true +} +``` + +Response 3, contains an additional stream payload. + +```json example +{ + "incremental": [ + { + "label": "filmsStream", + "path": ["person", "films", 2], + "items": [{ "title": "Return of the Jedi" }] + } + ], + "hasNext": true +} +``` + +Response 4, contains no incremental payloads, {hasNext} set to {false} indicates +the end of the stream. + +```json example +{ + "hasNext": false +} +``` + #### Stream payload A stream payload is a map that may appear as an item in the `incremental` entry From 2fb409c0e1ea28be6ecd67b05ab5af62355f20cf Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Thu, 8 Sep 2022 12:22:32 -0400 Subject: [PATCH 43/62] update label validation rule --- spec/Section 5 -- Validation.md | 22 +++++++++++----------- 1 file changed, 11 insertions(+), 11 deletions(-) diff --git a/spec/Section 5 -- Validation.md b/spec/Section 5 -- Validation.md index 018d7b91a..78a910cbf 100644 --- a/spec/Section 5 -- Validation.md +++ b/spec/Section 5 -- Validation.md @@ -1560,17 +1560,17 @@ mutation { ** Formal Specification ** -- For every {directive} in a document. -- Let {directiveName} be the name of {directive}. -- If {directiveName} is "defer" or "stream": - - For every {argument} in {directive}: - - Let {argumentName} be the name of {argument}. - - Let {argumentValue} be the value passed to {argument}. - - If {argumentName} is "label": - - {argumentValue} must not be a variable. - - Let {labels} be all label values passed to defer or stream directive label - arguments. - - {labels} must be a set of one. +- Let {labelValues} be an empty set. +- For every {directive} in the document: + - Let {directiveName} be the name of {directive}. + - If {directiveName} is "defer" or "stream": + - For every {argument} in {directive}: + - Let {argumentName} be the name of {argument}. + - Let {argumentValue} be the value passed to {argument}. + - If {argumentName} is "label": + - {argumentValue} must not be a variable. + - {argumentValue} must not be present in {labelValues}. + - Append {argumentValue} to {labelValues}. **Explanatory Text** From 66c3f7bc48b267cc1fe38807cf2799347a553eae Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Thu, 8 Sep 2022 17:27:57 -0400 Subject: [PATCH 44/62] clarify hasNext on incremental example --- spec/Section 7 -- Response.md | 14 +++++++++----- 1 file changed, 9 insertions(+), 5 deletions(-) diff --git a/spec/Section 7 -- Response.md b/spec/Section 7 -- Response.md index 477c964ca..5cf6fc567 100644 --- a/spec/Section 7 -- Response.md +++ b/spec/Section 7 -- Response.md @@ -41,8 +41,8 @@ operations that return a single response map. The GraphQL server may determine there are no more values in the response stream after a previous value with `hasNext` equal to `true` has been emitted. In this -case the last value in the response stream should be a map without `data`, -`label`, and `path` entries, and a `hasNext` entry with a value of `false`. +case the last value in the response stream should be a map without `data` and +`incremental` entries, and a `hasNext` entry with a value of `false`. The response map may also contain an entry with key `extensions`. This entry, if set, must have a map as its value. This entry is reserved for implementors to @@ -326,7 +326,10 @@ Response 2, contains the defer payload and the first stream payload. } ``` -Response 3, contains an additional stream payload. +Response 3, contains the final stream payload. In this example, the underlying +iterator does not close synchronously so {hasNext} is set to {true}. If this +iterator did close synchronously, {hasNext} would be set to {true} and this +would be the final response. ```json example { @@ -341,8 +344,9 @@ Response 3, contains an additional stream payload. } ``` -Response 4, contains no incremental payloads, {hasNext} set to {false} indicates -the end of the stream. +Response 4, contains no incremental payloads. {hasNext} set to {false} indicates +the end of the response stream. This response is sent when the underlying +iterator of the `films` field closes. ```json example { From 9287b6291bd436bb4ea4ad0fd3b034a62c0c4180 Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Thu, 8 Sep 2022 17:45:10 -0400 Subject: [PATCH 45/62] clarify canceling of subsequent payloads --- spec/Section 6 -- Execution.md | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-) diff --git a/spec/Section 6 -- Execution.md b/spec/Section 6 -- Execution.md index f0f933b58..cdd5ef5c7 100644 --- a/spec/Section 6 -- Execution.md +++ b/spec/Section 6 -- Execution.md @@ -455,11 +455,16 @@ If during {ExecuteSelectionSet()} a field with a non-null {fieldType} raises a _field error_ then that error must propagate to this entire selection set, either resolving to {null} if allowed or further propagated to a parent field. -If this occurs, any defer or stream executions with a path that starts with the -same path as the resolved {null} must not return their results to the client. -These defer or stream executions or any sibling fields which have not yet -executed or have not yet yielded a value may be cancelled to avoid unnecessary -work. +If this occurs, any sibling fields which have not yet executed or have not yet +yielded a value may be cancelled to avoid unnecessary work. + +Additionally, the path of each {asyncRecord} in {subsequentPayloads} must be +compared with the path of the field that ultimately resolved to {null}. If the +path of any {asyncRecord} starts with, but is not equal to, the path of the +resolved {null}, the {asyncRecord} must be removed from {subsequentPayloads} and +its result must not be sent to clients. If these async records have not yet +executed or have not yet yielded a value they may also be cancelled to avoid +unnecessary work. Note: See [Handling Field Errors](#sec-Handling-Field-Errors) for more about this behavior. From e0c0ad9be267eebf94c6df8609f174332d2d6b05 Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Thu, 8 Sep 2022 18:20:12 -0400 Subject: [PATCH 46/62] Add examples for non-null cases --- spec/Section 6 -- Execution.md | 149 ++++++++++++++++++++++++++++++--- 1 file changed, 139 insertions(+), 10 deletions(-) diff --git a/spec/Section 6 -- Execution.md b/spec/Section 6 -- Execution.md index cdd5ef5c7..0e743fe35 100644 --- a/spec/Section 6 -- Execution.md +++ b/spec/Section 6 -- Execution.md @@ -460,11 +460,35 @@ yielded a value may be cancelled to avoid unnecessary work. Additionally, the path of each {asyncRecord} in {subsequentPayloads} must be compared with the path of the field that ultimately resolved to {null}. If the -path of any {asyncRecord} starts with, but is not equal to, the path of the -resolved {null}, the {asyncRecord} must be removed from {subsequentPayloads} and -its result must not be sent to clients. If these async records have not yet -executed or have not yet yielded a value they may also be cancelled to avoid -unnecessary work. +path of any {asyncRecord} starts with the path of the resolved {null}, the +{asyncRecord} must be removed from {subsequentPayloads} and its result must not +be sent to clients. If these async records have not yet executed or have not yet +yielded a value they may also be cancelled to avoid unnecessary work. + +For example, assume the field `alwaysThrows` is a `Non-Null` type that always +raises a field error: + +```graphql example +{ + myObject { + ... @defer { + name + } + alwaysThrows + } +} +``` + +In this case, only one response should be sent. The async payload record +associated with the `@defer` directive should be removed and it's execution may +be cancelled. + +```json example +{ + "data": { "myObject": null }, + "hasNext": false +} +``` Note: See [Handling Field Errors](#sec-Handling-Field-Errors) for more about this behavior. @@ -756,7 +780,7 @@ variableValues, parentRecord, subsequentPayloads): - If {errors} is not empty: - Add an entry to {payload} named `errors` with the value {errors}. - If a field error was raised, causing a {null} to be propagated to - {responseValue}, and {objectType} is a Non-Nullable type: + {responseValue}: - Add an entry to {payload} named `data` with the value {null}. - Otherwise: - Add an entry to {payload} named `data` with the value {resultMap}. @@ -1120,16 +1144,121 @@ When a field error is raised inside `ExecuteDeferredFragment` or That is, the null resulting from a `Non-Null` type cannot propagate outside of the boundary of the defer or stream payload. -If a fragment with the `defer` directive is spread on a Non-Nullable object -type, and a field error has caused a {null} to propagate to the associated -object, the {null} should not propagate any further, and the associated Defer -Payload's `data` field must be set to {null}. +If a field error is raised while executing the selection set of a fragment with +the `defer` directive, causing a {null} to propagate to the object containing +this fragment, the {null} should not propagate any further. In this case, the +associated Defer Payload's `data` field must be set to {null}. + +For example, assume the `month` field is a `Non-Null` type that raises a field +error: + +```graphql example +{ + birthday { + ... @defer { + month + } + } +} +``` + +Response 1, the initial response is sent: + +```json example +{ + "data": { "birthday": {} }, + "hasNext": true +} +``` + +Response 2, the defer payload is sent. The {data} entry has been set to {null}, +as this {null} as propagated as high as the error boundary will allow. + +```json example +{ + "incremental": [ + { + "path": ["birthday"], + "data": null + } + ], + "hasNext": false +} +``` If the `stream` directive is present on a list field with a Non-Nullable inner type, and a field error has caused a {null} to propagate to the list item, the {null} should not propagate any further, and the associated Stream Payload's `item` field must be set to {null}. +For example, assume the `films` field is a `List` type with an `Non-Null` inner +type. In this case, the second list item raises a field error: + +```graphql example +{ + films @stream(initialCount: 1) +} +``` + +Response 1, the initial response is sent: + +```json example +{ + "data": { "films": ["A New Hope"] }, + "hasNext": true +} +``` + +Response 2, the first stream payload is sent. The {items} entry has been set to +{null}, as this {null} as propagated as high as the error boundary will allow. + +```json example +{ + "incremental": [ + { + "path": ["films", 1], + "items": null + } + ], + "hasNext": false +} +``` + +In this alternative example, assume the `films` field is a `List` type without a +`Non-Null` inner type. In this case, the second list item also raises a field +error: + +```graphql example +{ + films @stream(initialCount: 1) +} +``` + +Response 1, the initial response is sent: + +```json example +{ + "data": { "films": ["A New Hope"] }, + "hasNext": true +} +``` + +Response 2, the first stream payload is sent. The {items} entry has been set to +a list containing {null}, as this {null} has only propagated as high as the list +item. + +```json example +{ + "incremental": [ + { + "path": ["films", 1], + "items": [null] + } + ], + "hasNext": false +} +``` + If all fields from the root of the request to the source of the field error return `Non-Null` types, then the {"data"} entry in the response should be {null}. From 99baf54e52238f7da9d661d1cceca22f5a956765 Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Fri, 9 Sep 2022 06:58:54 -0400 Subject: [PATCH 47/62] typo --- spec/Section 6 -- Execution.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/spec/Section 6 -- Execution.md b/spec/Section 6 -- Execution.md index 0e743fe35..3078bfe09 100644 --- a/spec/Section 6 -- Execution.md +++ b/spec/Section 6 -- Execution.md @@ -480,7 +480,7 @@ raises a field error: ``` In this case, only one response should be sent. The async payload record -associated with the `@defer` directive should be removed and it's execution may +associated with the `@defer` directive should be removed and its execution may be cancelled. ```json example From 4931f7a0785d90c3fc5786a9b5174d63cd9115bf Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Fri, 9 Sep 2022 06:59:06 -0400 Subject: [PATCH 48/62] improve non-null example --- spec/Section 6 -- Execution.md | 27 ++++++++++++++++++++++++--- 1 file changed, 24 insertions(+), 3 deletions(-) diff --git a/spec/Section 6 -- Execution.md b/spec/Section 6 -- Execution.md index 3078bfe09..1b50f37bd 100644 --- a/spec/Section 6 -- Execution.md +++ b/spec/Section 6 -- Execution.md @@ -1155,9 +1155,12 @@ error: ```graphql example { birthday { - ... @defer { + ... @defer(label: "monthDefer") { month } + ... @defer(label: "yearDefer") { + year + } } } ``` @@ -1171,14 +1174,16 @@ Response 1, the initial response is sent: } ``` -Response 2, the defer payload is sent. The {data} entry has been set to {null}, -as this {null} as propagated as high as the error boundary will allow. +Response 2, the defer payload for label "monthDefer" is sent. The {data} entry +has been set to {null}, as this {null} as propagated as high as the error +boundary will allow. ```json example { "incremental": [ { "path": ["birthday"], + "label": "monthDefer", "data": null } ], @@ -1186,6 +1191,22 @@ as this {null} as propagated as high as the error boundary will allow. } ``` +Response 3, the defer payload for label "yearDefer" is sent. The data in this +payload is unaffected by the previous null error. + +```json example +{ + "incremental": [ + { + "path": ["birthday"], + "label": "yearDefer", + "data": { "year": "2022" } + } + ], + "hasNext": false +} +``` + If the `stream` directive is present on a list field with a Non-Nullable inner type, and a field error has caused a {null} to propagate to the list item, the {null} should not propagate any further, and the associated Stream Payload's From da96e9830321e6be4e5c8d5774d0432061358d19 Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Fri, 9 Sep 2022 07:26:51 -0400 Subject: [PATCH 49/62] Add FilterSubsequentPayloads algorithm --- spec/Section 6 -- Execution.md | 48 +++++++++++++++++++++++++++------- 1 file changed, 39 insertions(+), 9 deletions(-) diff --git a/spec/Section 6 -- Execution.md b/spec/Section 6 -- Execution.md index 1b50f37bd..3d0ccfd1d 100644 --- a/spec/Section 6 -- Execution.md +++ b/spec/Section 6 -- Execution.md @@ -458,12 +458,45 @@ either resolving to {null} if allowed or further propagated to a parent field. If this occurs, any sibling fields which have not yet executed or have not yet yielded a value may be cancelled to avoid unnecessary work. -Additionally, the path of each {asyncRecord} in {subsequentPayloads} must be -compared with the path of the field that ultimately resolved to {null}. If the -path of any {asyncRecord} starts with the path of the resolved {null}, the -{asyncRecord} must be removed from {subsequentPayloads} and its result must not -be sent to clients. If these async records have not yet executed or have not yet -yielded a value they may also be cancelled to avoid unnecessary work. +Additionally, async payload records in {subsequentPayloads} must be filtered if +their path points to a location that has resolved to {null} due to propagation +of a field error. This is described in +[Filter Subsequent Payloads](#sec-Filter-Subsequent-Payloads). These async +payload records must be removed from {subsequentPayloads} and their result must +not be sent to clients. If these async records have not yet executed or have not +yet yielded a value they may also be cancelled to avoid unnecessary work. + +Note: See [Handling Field Errors](#sec-Handling-Field-Errors) for more about +this behavior. + +### Filter Subsequent Payloads + +When a field error is raised, there may be async payload records in +{subsequentPayloads} with a path that points to a location that has been removed +or set to null due to null propagation. These async payload records must be +removed from subsequent payloads and their results must not be sent to clients. + +In {FilterSubsequentPayloads}, {nullPath} is the path which has resolved to null +after propagation as a result of a field error. {currentAsyncRecord} is the +async payload record where the field error was raised. {currentAsyncRecord} will +not be set for field errors that were raised during the initial execution +outside of {ExecuteDeferredFragment} or {ExecuteStreamField}. + +FilterSubsequentPayloads(subsequentPayloads, nullPath, currentAsyncRecord): + +- For each {asyncRecord} in {subsequentPayloads}: + - If {asyncRecord} is the same record as {currentAsyncRecord}: + - Continue to the next record in {subsequentPayloads}. + - Initialize {index} to zero. + - While {index} is less then the length of {nullPath}: + - Initialize {nullPathItem} to the element at {index} in {nullPath}. + - Initialize {asyncRecordPathItem} to the element at {index} in the {path} + of {asyncRecord}. + - If {nullPathItem} is not equivalent to {asyncRecordPathItem}: + - Continue to the next record in {subsequentPayloads}. + - Increment {index} by one. + - Remove {asyncRecord} from {subsequentPayloads}. Optionally, cancel any + incomplete work in the execution of {asyncRecord}. For example, assume the field `alwaysThrows` is a `Non-Null` type that always raises a field error: @@ -490,9 +523,6 @@ be cancelled. } ``` -Note: See [Handling Field Errors](#sec-Handling-Field-Errors) for more about -this behavior. - ### Normal and Serial Execution Normally the executor can execute the entries in a grouped field set in whatever From aff1113ebf758c5874c8195d6ef605de937838c1 Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Wed, 12 Oct 2022 17:22:54 -0400 Subject: [PATCH 50/62] link to note on should --- spec/Section 3 -- Type System.md | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-) diff --git a/spec/Section 3 -- Type System.md b/spec/Section 3 -- Type System.md index 0d4b3aab0..97b17e99c 100644 --- a/spec/Section 3 -- Type System.md +++ b/spec/Section 3 -- Type System.md @@ -2203,9 +2203,10 @@ fragment someFragment on User { #### @defer Arguments -- `if: Boolean! = true` - When `true`, fragment _should_ be deferred. When - `false`, fragment will not be deferred and data will be included in the - initial response. Defaults to `true` when omitted. +- `if: Boolean! = true` - When `true`, fragment _should_ be deferred (See + [related note](#note-088b7)). When `false`, fragment will not be deferred and + data will be included in the initial response. Defaults to `true` when + omitted. - `label: String` - May be used by GraphQL clients to identify the data from responses and associate it with the corresponding defer directive. If provided, the GraphQL Server must add it to the corresponding payload. `label` @@ -2239,9 +2240,10 @@ query myQuery($shouldStream: Boolean) { #### @stream Arguments -- `if: Boolean! = true` - When `true`, field _should_ be streamed. When `false`, - the field will not be streamed and all list items will be included in the - initial response. Defaults to `true` when omitted. +- `if: Boolean! = true` - When `true`, field _should_ be streamed (See + [related note](#note-088b7)). When `false`, the field will not be streamed and + all list items will be included in the initial response. Defaults to `true` + when omitted. - `label: String` - May be used by GraphQL clients to identify the data from responses and associate it with the corresponding stream directive. If provided, the GraphQL Server must add it to the corresponding payload. `label` From fb3e05c03e9c5e76168dcf39e5857cd62e41484c Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Tue, 1 Nov 2022 15:35:52 -0400 Subject: [PATCH 51/62] update on hasNext --- spec/Section 7 -- Response.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/spec/Section 7 -- Response.md b/spec/Section 7 -- Response.md index 5cf6fc567..e00f5a254 100644 --- a/spec/Section 7 -- Response.md +++ b/spec/Section 7 -- Response.md @@ -36,7 +36,7 @@ in the response stream. If the response of the GraphQL operation is a response stream, each response map must contain an entry with key `hasNext`. The value of this entry is `true` for all but the last response in the stream. The value of this entry is `false` for -the last response of the stream. This entry is not required for GraphQL +the last response of the stream. This entry must not be present for GraphQL operations that return a single response map. The GraphQL server may determine there are no more values in the response stream From 039e4ebd70321a6d8eda55fb1420491ec7d5972e Mon Sep 17 00:00:00 2001 From: Yaacov Rydzinski Date: Mon, 7 Nov 2022 16:07:31 -0500 Subject: [PATCH 52/62] small fixes (#3) * add comma * remove unused parameter --- spec/Section 6 -- Execution.md | 31 +++++++++++++------------------ 1 file changed, 13 insertions(+), 18 deletions(-) diff --git a/spec/Section 6 -- Execution.md b/spec/Section 6 -- Execution.md index 3d0ccfd1d..1cf078077 100644 --- a/spec/Section 6 -- Execution.md +++ b/spec/Section 6 -- Execution.md @@ -428,8 +428,7 @@ subsequentPayloads, asyncRecord): - If {path} is not provided, initialize it to an empty list. - If {subsequentPayloads} is not provided, initialize it to the empty set. - Let {groupedFieldSet} and {deferredGroupedFieldsList} be the result of - {CollectFields(objectType, objectValue, selectionSet, variableValues, path, - asyncRecord)}. + {CollectFields(objectType, selectionSet, variableValues, path, asyncRecord)}. - Initialize {resultMap} to an empty ordered map. - For each {groupedFieldSet} as {responseKey} and {fields}: - Let {fieldName} be the name of the first entry in {fields}. Note: This value @@ -661,8 +660,8 @@ The depth-first-search order of the field groups produced by {CollectFields()} is maintained through execution, ensuring that fields appear in the executed response in a stable and predictable order. -CollectFields(objectType, objectValue, selectionSet, variableValues, path, -asyncRecord, visitedFragments, deferredGroupedFieldsList): +CollectFields(objectType, selectionSet, variableValues, path, asyncRecord, +visitedFragments, deferredGroupedFieldsList): - If {visitedFragments} is not provided, initialize it to the empty set. - Initialize {groupedFields} to an empty ordered map of lists. @@ -707,16 +706,14 @@ asyncRecord, visitedFragments, deferredGroupedFieldsList): - Let {label} be the value or the variable to {deferDirective}'s {label} argument. - Let {deferredGroupedFields} be the result of calling - {CollectFields(objectType, objectValue, fragmentSelectionSet, - variableValues, path, asyncRecord, visitedFragments, - deferredGroupedFieldsList)}. + {CollectFields(objectType, fragmentSelectionSet, variableValues, path, + asyncRecord, visitedFragments, deferredGroupedFieldsList)}. - Append a record containing {label} and {deferredGroupedFields} to {deferredGroupedFieldsList}. - Continue with the next {selection} in {selectionSet}. - Let {fragmentGroupedFieldSet} be the result of calling - {CollectFields(objectType, objectValue, fragmentSelectionSet, - variableValues, path, asyncRecord, visitedFragments, - deferredGroupedFieldsList)}. + {CollectFields(objectType, fragmentSelectionSet, variableValues, path, + asyncRecord, visitedFragments, deferredGroupedFieldsList)}. - For each {fragmentGroup} in {fragmentGroupedFieldSet}: - Let {responseKey} be the response key shared by all fields in {fragmentGroup}. @@ -737,16 +734,14 @@ asyncRecord, visitedFragments, deferredGroupedFieldsList): - Let {label} be the value or the variable to {deferDirective}'s {label} argument. - Let {deferredGroupedFields} be the result of calling - {CollectFields(objectType, objectValue, fragmentSelectionSet, - variableValues, path, asyncRecord, visitedFragments, - deferredGroupedFieldsList)}. + {CollectFields(objectType, fragmentSelectionSet, variableValues, path, + asyncRecord, visitedFragments, deferredGroupedFieldsList)}. - Append a record containing {label} and {deferredGroupedFields} to {deferredGroupedFieldsList}. - Continue with the next {selection} in {selectionSet}. - Let {fragmentGroupedFieldSet} be the result of calling - {CollectFields(objectType, objectValue, fragmentSelectionSet, - variableValues, path, asyncRecord, visitedFragments, - deferredGroupedFieldsList)}. + {CollectFields(objectType, fragmentSelectionSet, variableValues, path, + asyncRecord, visitedFragments, deferredGroupedFieldsList)}. - For each {fragmentGroup} in {fragmentGroupedFieldSet}: - Let {responseKey} be the response key shared by all fields in {fragmentGroup}. @@ -964,8 +959,8 @@ directive. #### Execute Stream Field -ExecuteStreamField(label, iterator, index, fields, innerType, path streamRecord, -variableValues, subsequentPayloads): +ExecuteStreamField(label, iterator, index, fields, innerType, path, +streamRecord, variableValues, subsequentPayloads): - Let {streamRecord} be an async payload record created from {label}, {path}, and {iterator}. From 777c21b031262f6fbeef36d2383744c97d3f1c53 Mon Sep 17 00:00:00 2001 From: Yaacov Rydzinski Date: Wed, 16 Nov 2022 17:15:20 +0200 Subject: [PATCH 53/62] remove ResolveFIeldGenerator (#4) * streamline stream execution Currently, these spec changes introduce a new internal function named `ResolveFieldGenerator` that is suggested parallels `ResolveFieldValue`. This function is used during field execution such that if the stream directive is specified, it is called instead of `ResolveFieldValue`. The reference implementation, however, does not require any such function, simply utilizing the result of `ResolveFieldValue`. With incremental delivery, collections completed by `CompleteValue` should be explicitly iterated using a well-defined iterator, such that the iterator can be passed to `ExecuteStreamField`. But this does not require a new internal function to be specified/exposed. Moreover, introducing this function causes a mixing of concerns between the `ExecuteField` and `CompleteValue` algorithms; Currently, if stream is specified for a field, `ExecuteField` extracts the iterator and passes it to `CompleteValue`, while if stream is not specified, the `ExecuteField` passes the collection, i.e. the iterable, not the iterator. In the stream case, this shunts some of the logic checking the validity of resolution results into field execution. In fact, it exposes a specification "bug" => in the stream case, no checking is actually done that `ResolveFieldGenerator` returns an iterator! This change removes `ResolveFieldGenerator` and with it some complexity, and brings it in line with the reference implementation. The reference implementation contains some simplification of the algorithm for the synchronous iterator case (we don't have to preserve the iterator on the StreamRecord, because there will be no early close required and we don't have to set isCompletedIterator, beacuse we don't have to create a dummy payload for termination of the asynchronous stream), We could consider also removing these bits as well, as they are an implementation detail in terms of how our dispatcher is managing its iterators, but that should be left for another change. * run prettier --- spec/Section 6 -- Execution.md | 117 +++++++++++++-------------------- 1 file changed, 44 insertions(+), 73 deletions(-) diff --git a/spec/Section 6 -- Execution.md b/spec/Section 6 -- Execution.md index 1cf078077..ac2520336 100644 --- a/spec/Section 6 -- Execution.md +++ b/spec/Section 6 -- Execution.md @@ -831,15 +831,6 @@ subsequentPayloads, asyncRecord): - Append {fieldName} to {path}. - Let {argumentValues} be the result of {CoerceArgumentValues(objectType, field, variableValues)} -- If {field} provides the directive `@stream`, let {streamDirective} be that - directive. - - Let {initialCount} be the value or variable provided to {streamDirective}'s - {initialCount} argument. - - Let {resolvedValue} be {ResolveFieldGenerator(objectType, objectValue, - fieldName, argumentValues)}. - - Let {result} be the result of calling {CompleteValue(fieldType, fields, - resolvedValue, variableValues, path, subsequentPayloads, asyncRecord)}. - - Return {result}. - Let {resolvedValue} be {ResolveFieldValue(objectType, objectValue, fieldName, argumentValues)}. - Let {result} be the result of calling {CompleteValue(fieldType, fields, @@ -907,20 +898,17 @@ must only allow usage of variables of appropriate types. While nearly all of GraphQL execution can be described generically, ultimately the internal system exposing the GraphQL interface must provide values. This is exposed via {ResolveFieldValue}, which produces a value for a given field on a -type for a real value. In addition, {ResolveFieldGenerator} will be exposed to -produce an iterator for a field with `List` return type. The internal system may -optionally define a generator function. In the case where the generator is not -defined, the GraphQL executor provides a default generator. For example, a -trivial generator that yields the entire list upon the first iteration. +type for a real value. -As an example, a {ResolveFieldValue} might accept the {objectType} `Person`, the -{field} {"soulMate"}, and the {objectValue} representing John Lennon. It would -be expected to yield the value representing Yoko Ono. +As an example, this might accept the {objectType} `Person`, the {field} +{"soulMate"}, and the {objectValue} representing John Lennon. It would be +expected to yield the value representing Yoko Ono. -A {ResolveFieldGenerator} might accept the {objectType} `MusicBand`, the {field} -{"members"}, and the {objectValue} representing Beatles. It would be expected to -yield a iterator of values representing John Lennon, Paul McCartney, Ringo Starr -and George Harrison. +List values are resolved similarly. For example, {ResolveFieldValue} might also +accept the {objectType} `MusicBand`, the {field} {"members"}, and the +{objectValue} representing the Beatles. It would be expected to yield a +collection of values representing John Lennon, Paul McCartney, Ringo Starr and +George Harrison. ResolveFieldValue(objectType, objectValue, fieldName, argumentValues): @@ -929,33 +917,23 @@ ResolveFieldValue(objectType, objectValue, fieldName, argumentValues): - Return the result of calling {resolver}, providing {objectValue} and {argumentValues}. -ResolveFieldGenerator(objectType, objectValue, fieldName, argumentValues): - -- If {objectType} provide an internal function {generatorResolver} for - generating partially resolved value of a list field named {fieldName}: - - Let {generatorResolver} be the internal function. - - Return the iterator from calling {generatorResolver}, providing - {objectValue} and {argumentValues}. -- Create {generator} from {ResolveFieldValue(objectType, objectValue, fieldName, - argumentValues)}. -- Return {generator}. - Note: It is common for {resolver} to be asynchronous due to relying on reading an underlying database or networked service to produce a value. This necessitates the rest of a GraphQL executor to handle an asynchronous execution -flow. In addition, a common implementation of {generator} is to leverage -asynchronous iterators or asynchronous generators provided by many programming -languages. +flow. In addition, an implementation for collections may leverage asynchronous +iterators or asynchronous generators provided by many programming languages. +This may be particularly helpful when used in conjunction with the `@stream` +directive. ### Value Completion After resolving the value for a field, it is completed by ensuring it adheres to the expected return type. If the return type is another Object type, then the -field execution process continues recursively. In the case where a value -returned for a list type field is an iterator due to `@stream` specified on the -field, value completion iterates over the iterator until the number of items -yield by the iterator satisfies `initialCount` specified on the `@stream` -directive. +field execution process continues recursively. If the return type is a List +type, each member of the resolved collection is completed using the same value +completion process. In the case where `@stream` is specified on a field of list +type, value completion iterates over the collection until the number of items +yielded items satisfies `initialCount` specified on the `@stream` directive. #### Execute Stream Field @@ -1007,45 +985,38 @@ subsequentPayloads, asyncRecord): - If {result} is {null} (or another internal value similar to {null} such as {undefined}), return {null}. - If {fieldType} is a List type: - - If {result} is an iterator: - - Let {field} be the first entry in {fields}. - - Let {innerType} be the inner type of {fieldType}. - - If {field} provides the directive `@stream` and its {if} argument is not - {false} and is not a variable in {variableValues} with the value {false} - and {innerType} is the outermost return type of the list type defined for - {field}: - - Let {streamDirective} be that directive. + - If {result} is not a collection of values, raise a _field error_. + - Let {field} be the first entry in {fields}. + - Let {innerType} be the inner type of {fieldType}. + - If {field} provides the directive `@stream` and its {if} argument is not + {false} and is not a variable in {variableValues} with the value {false} and + {innerType} is the outermost return type of the list type defined for + {field}: + - Let {streamDirective} be that directive. - Let {initialCount} be the value or variable provided to {streamDirective}'s {initialCount} argument. - If {initialCount} is less than zero, raise a _field error_. - Let {label} be the value or variable provided to {streamDirective}'s {label} argument. - - Let {initialItems} be an empty list - - Let {index} be zero. + - Let {iterator} be an iterator for {result}. + - Let {items} be an empty list. + - Let {index} be zero. - While {result} is not closed: - - If {streamDirective} is not defined or {index} is not greater than or - equal to {initialCount}: - - Wait for the next item from {result}. - - Let {resultItem} be the item retrieved from {result}. - - Let {itemPath} be {path} with {index} appended. - - Let {resolvedItem} be the result of calling {CompleteValue(innerType, - fields, resultItem, variableValues, itemPath, subsequentPayloads, - asyncRecord)}. - - Append {resolvedItem} to {initialItems}. - - Increment {index}. - - If {streamDirective} is defined and {index} is greater than or equal to - {initialCount}: - - Call {ExecuteStreamField(label, result, index, fields, innerType, - path, asyncRecord, subsequentPayloads)}. - - Let {result} be {initialItems}. - - Exit while loop. - - Return {initialItems}. - - If {result} is not a collection of values, raise a _field error_. - - Let {innerType} be the inner type of {fieldType}. - - Return a list where each list item is the result of calling - {CompleteValue(innerType, fields, resultItem, variableValues, itemPath, - subsequentPayloads, asyncRecord)}, where {resultItem} is each item in - {result} and {itemPath} is {path} with the index of the item appended. + - If {streamDirective} is defined and {index} is greater than or equal to + {initialCount}: + - Call {ExecuteStreamField(label, iterator, index, fields, innerType, + path, asyncRecord, subsequentPayloads)}. + - Return {items}. + - Otherwise: + - Retrieve the next item from {result} via the {iterator}. + - Let {resultItem} be the item retrieved from {result}. + - Let {itemPath} be {path} with {index} appended. + - Let {resolvedItem} be the result of calling {CompleteValue(innerType, + fields, resultItem, variableValues, itemPath, subsequentPayloads, + asyncRecord)}. + - Append {resolvedItem} to {initialItems}. + - Increment {index}. + - Return {items}. - If {fieldType} is a Scalar or Enum type: - Return the result of {CoerceResult(fieldType, result)}. - If {fieldType} is an Object, Interface, or Union type: From 7626800dd3e94094ca2c6fc56f4694ea96a2b3ca Mon Sep 17 00:00:00 2001 From: Yaacov Rydzinski Date: Fri, 18 Nov 2022 16:31:34 +0200 Subject: [PATCH 54/62] fix typos (#6) * fix whitespace * complete renaming of initialItems --- spec/Section 6 -- Execution.md | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/spec/Section 6 -- Execution.md b/spec/Section 6 -- Execution.md index ac2520336..35b4f2c1a 100644 --- a/spec/Section 6 -- Execution.md +++ b/spec/Section 6 -- Execution.md @@ -1001,7 +1001,7 @@ subsequentPayloads, asyncRecord): - Let {iterator} be an iterator for {result}. - Let {items} be an empty list. - Let {index} be zero. - - While {result} is not closed: + - While {result} is not closed: - If {streamDirective} is defined and {index} is greater than or equal to {initialCount}: - Call {ExecuteStreamField(label, iterator, index, fields, innerType, @@ -1014,9 +1014,9 @@ subsequentPayloads, asyncRecord): - Let {resolvedItem} be the result of calling {CompleteValue(innerType, fields, resultItem, variableValues, itemPath, subsequentPayloads, asyncRecord)}. - - Append {resolvedItem} to {initialItems}. + - Append {resolvedItem} to {items}. - Increment {index}. - - Return {items}. + - Return {items}. - If {fieldType} is a Scalar or Enum type: - Return the result of {CoerceResult(fieldType, result)}. - If {fieldType} is an Object, Interface, or Union type: From d16a4322c082ca8e37cb2ced6a37c7a89c7b9d1b Mon Sep 17 00:00:00 2001 From: Yaacov Rydzinski Date: Mon, 21 Nov 2022 21:33:58 +0200 Subject: [PATCH 55/62] Add error handling for stream iterators (#5) * Add error handling for stream iterators * also add iterator error handling within CompleteValue * incorporate feedback --- spec/Section 6 -- Execution.md | 37 +++++++++++++++++++--------------- 1 file changed, 21 insertions(+), 16 deletions(-) diff --git a/spec/Section 6 -- Execution.md b/spec/Section 6 -- Execution.md index 35b4f2c1a..93b974a47 100644 --- a/spec/Section 6 -- Execution.md +++ b/spec/Section 6 -- Execution.md @@ -950,25 +950,29 @@ streamRecord, variableValues, subsequentPayloads): - Set {isCompletedIterator} to {true} on {streamRecord}. - Return {null}. - Let {payload} be an unordered map. - - Let {item} be the item retrieved from {iterator}. - - Let {data} be the result of calling {CompleteValue(innerType, fields, item, - variableValues, itemPath, subsequentPayloads, parentRecord)}. - - Append any encountered field errors to {errors}. - - Increment {index}. - - Call {ExecuteStreamField(label, iterator, index, fields, innerType, path, - streamRecord, variableValues, subsequentPayloads)}. - - If {parentRecord} is defined: - - Wait for the result of {dataExecution} on {parentRecord}. - - If {errors} is not empty: - - Add an entry to {payload} named `errors` with the value {errors}. - - If a field error was raised, causing a {null} to be propagated to {data}, - and {innerType} is a Non-Nullable type: + - If an item is not retrieved because of an error: + - Append the encountered error to {errors}. - Add an entry to {payload} named `items` with the value {null}. - Otherwise: - - Add an entry to {payload} named `items` with a list containing the value - {data}. + - Let {item} be the item retrieved from {iterator}. + - Let {data} be the result of calling {CompleteValue(innerType, fields, + item, variableValues, itemPath, subsequentPayloads, parentRecord)}. + - Append any encountered field errors to {errors}. + - Increment {index}. + - Call {ExecuteStreamField(label, iterator, index, fields, innerType, path, + streamRecord, variableValues, subsequentPayloads)}. + - If a field error was raised, causing a {null} to be propagated to {data}, + and {innerType} is a Non-Nullable type: + - Add an entry to {payload} named `items` with the value {null}. + - Otherwise: + - Add an entry to {payload} named `items` with a list containing the value + {data}. + - If {errors} is not empty: + - Add an entry to {payload} named `errors` with the value {errors}. - Add an entry to {payload} named `label` with the value {label}. - Add an entry to {payload} named `path` with the value {itemPath}. + - If {parentRecord} is defined: + - Wait for the result of {dataExecution} on {parentRecord}. - Return {payload}. - Set {dataExecution} on {streamRecord}. - Append {streamRecord} to {subsequentPayloads}. @@ -1008,7 +1012,8 @@ subsequentPayloads, asyncRecord): path, asyncRecord, subsequentPayloads)}. - Return {items}. - Otherwise: - - Retrieve the next item from {result} via the {iterator}. + - Wait for the next item from {result} via the {iterator}. + - If an item is not retrieved because of an error, raise a _field error_. - Let {resultItem} be the item retrieved from {result}. - Let {itemPath} be {path} with {index} appended. - Let {resolvedItem} be the result of calling {CompleteValue(innerType, From 979e581f3a1e18800ace355cf12d2eb6e00b2c9a Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Tue, 22 Nov 2022 13:25:12 -0500 Subject: [PATCH 56/62] Raise a field error if defer/stream encountered during subscription execution --- spec/Section 6 -- Execution.md | 24 ++++++++---------------- 1 file changed, 8 insertions(+), 16 deletions(-) diff --git a/spec/Section 6 -- Execution.md b/spec/Section 6 -- Execution.md index 93b974a47..2a15f5304 100644 --- a/spec/Section 6 -- Execution.md +++ b/spec/Section 6 -- Execution.md @@ -328,28 +328,15 @@ MapSourceToResponseEvent(sourceStream, subscription, schema, variableValues): ExecuteSubscriptionEvent(subscription, schema, variableValues, initialValue): -- Let {subsequentPayloads} be an empty list. - Let {subscriptionType} be the root Subscription type in {schema}. - Assert: {subscriptionType} is an Object type. - Let {selectionSet} be the top level Selection Set in {subscription}. - Let {data} be the result of running {ExecuteSelectionSet(selectionSet, - subscriptionType, initialValue, variableValues, subsequentPayloads)} - _normally_ (allowing parallelization). + subscriptionType, initialValue, variableValues)} _normally_ (allowing + parallelization). - Let {errors} be the list of all _field error_ raised while executing the selection set. -- If {subsequentPayloads} is empty: - - Return an unordered map containing {data} and {errors}. -- If {subsequentPayloads} is not empty: - - Let {initialResponse} be an unordered map containing {data}, {errors}, and - an entry named {hasNext} with the value {true}. - - Let {iterator} be the result of running - {YieldSubsequentPayloads(initialResponse, subsequentPayloads)}. - - For each {payload} yielded by {iterator}: - - If a termination signal is received: - - Send a termination signal to {iterator}. - - Return. - - Otherwise: - - Yield {payload}. +- Return an unordered map containing {data} and {errors}. Note: The {ExecuteSubscriptionEvent()} algorithm is intentionally similar to {ExecuteQuery()} since this is how each event result is produced. @@ -690,6 +677,8 @@ visitedFragments, deferredGroupedFieldsList): argument is not {false} and is not a variable in {variableValues} with the value {false}: - Let {deferDirective} be that directive. + - If this execution is for a subscription operation, raise a _field + error_. - If {deferDirective} is not defined: - If {fragmentSpreadName} is in {visitedFragments}, continue with the next {selection} in {selectionSet}. @@ -730,6 +719,8 @@ visitedFragments, deferredGroupedFieldsList): is not {false} and is not a variable in {variableValues} with the value {false}: - Let {deferDirective} be that directive. + - If this execution is for a subscription operation, raise a _field + error_. - If {deferDirective} is defined: - Let {label} be the value or the variable to {deferDirective}'s {label} argument. @@ -997,6 +988,7 @@ subsequentPayloads, asyncRecord): {innerType} is the outermost return type of the list type defined for {field}: - Let {streamDirective} be that directive. + - If this execution is for a subscription operation, raise a _field error_. - Let {initialCount} be the value or variable provided to {streamDirective}'s {initialCount} argument. - If {initialCount} is less than zero, raise a _field error_. From 120712104102e49c847b7f798df3ddea79f74d22 Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Tue, 22 Nov 2022 13:25:32 -0500 Subject: [PATCH 57/62] Add validation rule for defer/stream on subscriptions --- spec/Section 5 -- Validation.md | 49 +++++++++++++++++++++++++++++++++ 1 file changed, 49 insertions(+) diff --git a/spec/Section 5 -- Validation.md b/spec/Section 5 -- Validation.md index 78a910cbf..1e8fbd6d7 100644 --- a/spec/Section 5 -- Validation.md +++ b/spec/Section 5 -- Validation.md @@ -1556,6 +1556,55 @@ mutation { } ``` +### Defer And Stream Directives Are Used On Valid Operations + +** Formal Specification ** + +- Let {subscriptionFragments} be the empty set. +- For each {operation} in a document: + - If {operation} is a subscription operation: + - Let {fragments} be every fragment referenced by that {operation} + transitively. + - For each {fragment} in {fragments}: + - Let {fragmentName} be the name of {fragment}. + - Add {fragmentName} to {subscriptionFragments}. +- For every {directive} in a document: + - If {directiveName} is not "defer" or "stream": + - Continue to the next {directive}. + - Let {ancestor} be the ancestor operation or fragment definition of + {directive}. + - If {ancestor} is a fragment definition: + - If the fragment name of {ancestor} is not present in + {subscriptionFragments}: + - Continue to the next {directive}. + - If {ancestor} is not a subscription operation: + - Continue to the next {directive}. + - Let {if} be the argument named "if" on {directive}. + - {if} must be defined. + - Let {argumentValue} be the value passed to {if}. + - {argumentValue} must be a variable, or the boolean value "false". + +**Explanatory Text** + +The defer and stream directives can not be used to defer or stream data in +subscription operations. If these directives appear in a subscription operation +they must be disabled using the "if" argument. This rule will not permit any +defer or stream directives on a subscription operation that cannot be disabled +using the "if" argument. + +For example, the following document will not pass validation because `@defer` +has been used in a subscription operation with no "if" argument defined: + +```raw graphql counter-example +subscription sub { + newMessage { + ... @defer { + body + } + } +} +``` + ### Defer And Stream Directive Labels Are Unique ** Formal Specification ** From 6545a5587bb52ce84d191f37598895fe3fe87bb0 Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Wed, 23 Nov 2022 15:48:58 -0500 Subject: [PATCH 58/62] clarify label is not required --- spec/Section 6 -- Execution.md | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/spec/Section 6 -- Execution.md b/spec/Section 6 -- Execution.md index 2a15f5304..26fa4efb9 100644 --- a/spec/Section 6 -- Execution.md +++ b/spec/Section 6 -- Execution.md @@ -800,7 +800,8 @@ variableValues, parentRecord, subsequentPayloads): - Add an entry to {payload} named `data` with the value {null}. - Otherwise: - Add an entry to {payload} named `data` with the value {resultMap}. - - Add an entry to {payload} named `label` with the value {label}. + - If {label} is defined: + - Add an entry to {payload} named `label` with the value {label}. - Add an entry to {payload} named `path` with the value {path}. - Return {payload}. - Set {dataExecution} on {deferredFragmentRecord}. @@ -960,7 +961,8 @@ streamRecord, variableValues, subsequentPayloads): {data}. - If {errors} is not empty: - Add an entry to {payload} named `errors` with the value {errors}. - - Add an entry to {payload} named `label` with the value {label}. + - If {label} is defined: + - Add an entry to {payload} named `label` with the value {label}. - Add an entry to {payload} named `path` with the value {itemPath}. - If {parentRecord} is defined: - Wait for the result of {dataExecution} on {parentRecord}. From 4c2e3f9cfe630588b326f47c13ab2a029d88cacc Mon Sep 17 00:00:00 2001 From: Yaacov Rydzinski Date: Tue, 29 Nov 2022 15:50:09 +0200 Subject: [PATCH 59/62] fix parentRecord argument in ExecuteStreamField (#7) --- spec/Section 6 -- Execution.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/spec/Section 6 -- Execution.md b/spec/Section 6 -- Execution.md index 26fa4efb9..ef4c199c6 100644 --- a/spec/Section 6 -- Execution.md +++ b/spec/Section 6 -- Execution.md @@ -930,7 +930,7 @@ yielded items satisfies `initialCount` specified on the `@stream` directive. #### Execute Stream Field ExecuteStreamField(label, iterator, index, fields, innerType, path, -streamRecord, variableValues, subsequentPayloads): +parentRecord, variableValues, subsequentPayloads): - Let {streamRecord} be an async payload record created from {label}, {path}, and {iterator}. From e4ee3eb9a9a545e0418274245ead20ff7b1bd9da Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Mon, 5 Dec 2022 14:03:10 -0500 Subject: [PATCH 60/62] fix typo Co-authored-by: Simon Gellis <82392336+simongellis-attentive@users.noreply.github.com> --- spec/Section 7 -- Response.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/spec/Section 7 -- Response.md b/spec/Section 7 -- Response.md index e00f5a254..4eade1be1 100644 --- a/spec/Section 7 -- Response.md +++ b/spec/Section 7 -- Response.md @@ -328,7 +328,7 @@ Response 2, contains the defer payload and the first stream payload. Response 3, contains the final stream payload. In this example, the underlying iterator does not close synchronously so {hasNext} is set to {true}. If this -iterator did close synchronously, {hasNext} would be set to {true} and this +iterator did close synchronously, {hasNext} would be set to {false} and this would be the final response. ```json example From 22de5de5b892eebe8bbf4a201a2cb35301d96e4a Mon Sep 17 00:00:00 2001 From: Rob Richard Date: Sun, 15 Jan 2023 12:18:50 -0500 Subject: [PATCH 61/62] replace server with service --- spec/Section 3 -- Type System.md | 14 +++++++------- spec/Section 7 -- Response.md | 8 ++++---- 2 files changed, 11 insertions(+), 11 deletions(-) diff --git a/spec/Section 3 -- Type System.md b/spec/Section 3 -- Type System.md index 97b17e99c..e5a955a9b 100644 --- a/spec/Section 3 -- Type System.md +++ b/spec/Section 3 -- Type System.md @@ -2209,8 +2209,8 @@ fragment someFragment on User { omitted. - `label: String` - May be used by GraphQL clients to identify the data from responses and associate it with the corresponding defer directive. If - provided, the GraphQL Server must add it to the corresponding payload. `label` - must be unique label across all `@defer` and `@stream` directives in a + provided, the GraphQL service must add it to the corresponding payload. + `label` must be unique label across all `@defer` and `@stream` directives in a document. `label` must not be provided as a variable. ### @stream @@ -2246,19 +2246,19 @@ query myQuery($shouldStream: Boolean) { when omitted. - `label: String` - May be used by GraphQL clients to identify the data from responses and associate it with the corresponding stream directive. If - provided, the GraphQL Server must add it to the corresponding payload. `label` - must be unique label across all `@defer` and `@stream` directives in a + provided, the GraphQL service must add it to the corresponding payload. + `label` must be unique label across all `@defer` and `@stream` directives in a document. `label` must not be provided as a variable. -- `initialCount: Int` - The number of list items the server should return as +- `initialCount: Int` - The number of list items the service should return as part of the initial response. If omitted, defaults to `0`. A field error will be raised if the value of this argument is less than `0`. Note: The ability to defer and/or stream parts of a response can have a potentially significant impact on application performance. Developers generally need clear, predictable control over their application's performance. It is -highly recommended that GraphQL servers honor the `@defer` and `@stream` +highly recommended that GraphQL services honor the `@defer` and `@stream` directives on each execution. However, the specification allows advanced use -cases where the server can determine that it is more performant to not defer +cases where the service can determine that it is more performant to not defer and/or stream. Therefore, GraphQL clients _must_ be able to process a response that ignores the `@defer` and/or `@stream` directives. This also applies to the `initialCount` argument on the `@stream` directive. Clients _must_ be able to diff --git a/spec/Section 7 -- Response.md b/spec/Section 7 -- Response.md index 4eade1be1..db50408fa 100644 --- a/spec/Section 7 -- Response.md +++ b/spec/Section 7 -- Response.md @@ -39,10 +39,10 @@ all but the last response in the stream. The value of this entry is `false` for the last response of the stream. This entry must not be present for GraphQL operations that return a single response map. -The GraphQL server may determine there are no more values in the response stream -after a previous value with `hasNext` equal to `true` has been emitted. In this -case the last value in the response stream should be a map without `data` and -`incremental` entries, and a `hasNext` entry with a value of `false`. +The GraphQL service may determine there are no more values in the response +stream after a previous value with `hasNext` equal to `true` has been emitted. +In this case the last value in the response stream should be a map without +`data` and `incremental` entries, and a `hasNext` entry with a value of `false`. The response map may also contain an entry with key `extensions`. This entry, if set, must have a map as its value. This entry is reserved for implementors to From c630301560d9819d33255d3ba00f548e8abbcdc6 Mon Sep 17 00:00:00 2001 From: Yaacov Rydzinski Date: Mon, 16 Jan 2023 18:28:31 +0200 Subject: [PATCH 62/62] CollectFields does not require path or asyncRecord (#11) --- spec/Section 6 -- Execution.md | 24 ++++++++++++------------ 1 file changed, 12 insertions(+), 12 deletions(-) diff --git a/spec/Section 6 -- Execution.md b/spec/Section 6 -- Execution.md index ef4c199c6..5f690f311 100644 --- a/spec/Section 6 -- Execution.md +++ b/spec/Section 6 -- Execution.md @@ -415,7 +415,7 @@ subsequentPayloads, asyncRecord): - If {path} is not provided, initialize it to an empty list. - If {subsequentPayloads} is not provided, initialize it to the empty set. - Let {groupedFieldSet} and {deferredGroupedFieldsList} be the result of - {CollectFields(objectType, selectionSet, variableValues, path, asyncRecord)}. + {CollectFields(objectType, selectionSet, variableValues)}. - Initialize {resultMap} to an empty ordered map. - For each {groupedFieldSet} as {responseKey} and {fields}: - Let {fieldName} be the name of the first entry in {fields}. Note: This value @@ -647,8 +647,8 @@ The depth-first-search order of the field groups produced by {CollectFields()} is maintained through execution, ensuring that fields appear in the executed response in a stable and predictable order. -CollectFields(objectType, selectionSet, variableValues, path, asyncRecord, -visitedFragments, deferredGroupedFieldsList): +CollectFields(objectType, selectionSet, variableValues, visitedFragments, +deferredGroupedFieldsList): - If {visitedFragments} is not provided, initialize it to the empty set. - Initialize {groupedFields} to an empty ordered map of lists. @@ -695,14 +695,14 @@ visitedFragments, deferredGroupedFieldsList): - Let {label} be the value or the variable to {deferDirective}'s {label} argument. - Let {deferredGroupedFields} be the result of calling - {CollectFields(objectType, fragmentSelectionSet, variableValues, path, - asyncRecord, visitedFragments, deferredGroupedFieldsList)}. + {CollectFields(objectType, fragmentSelectionSet, variableValues, + visitedFragments, deferredGroupedFieldsList)}. - Append a record containing {label} and {deferredGroupedFields} to {deferredGroupedFieldsList}. - Continue with the next {selection} in {selectionSet}. - Let {fragmentGroupedFieldSet} be the result of calling - {CollectFields(objectType, fragmentSelectionSet, variableValues, path, - asyncRecord, visitedFragments, deferredGroupedFieldsList)}. + {CollectFields(objectType, fragmentSelectionSet, variableValues, + visitedFragments, deferredGroupedFieldsList)}. - For each {fragmentGroup} in {fragmentGroupedFieldSet}: - Let {responseKey} be the response key shared by all fields in {fragmentGroup}. @@ -725,21 +725,21 @@ visitedFragments, deferredGroupedFieldsList): - Let {label} be the value or the variable to {deferDirective}'s {label} argument. - Let {deferredGroupedFields} be the result of calling - {CollectFields(objectType, fragmentSelectionSet, variableValues, path, - asyncRecord, visitedFragments, deferredGroupedFieldsList)}. + {CollectFields(objectType, fragmentSelectionSet, variableValues, + visitedFragments, deferredGroupedFieldsList)}. - Append a record containing {label} and {deferredGroupedFields} to {deferredGroupedFieldsList}. - Continue with the next {selection} in {selectionSet}. - Let {fragmentGroupedFieldSet} be the result of calling - {CollectFields(objectType, fragmentSelectionSet, variableValues, path, - asyncRecord, visitedFragments, deferredGroupedFieldsList)}. + {CollectFields(objectType, fragmentSelectionSet, variableValues, + visitedFragments, deferredGroupedFieldsList)}. - For each {fragmentGroup} in {fragmentGroupedFieldSet}: - Let {responseKey} be the response key shared by all fields in {fragmentGroup}. - Let {groupForResponseKey} be the list in {groupedFields} for {responseKey}; if no such list exists, create it as an empty list. - Append all items in {fragmentGroup} to {groupForResponseKey}. -- Return {groupedFields} and {deferredGroupedFieldsList}. +- Return {groupedFields}, {deferredGroupedFieldsList} and {visitedFragments}. Note: The steps in {CollectFields()} evaluating the `@skip` and `@include` directives may be applied in either order since they apply commutatively.