Environment configuration #326

THardy98 · 2025-08-23T06:43:11Z

What was changed

Added environment configuration

Closes [Feature Request] Environment Configuration #287
How was this tested:
Added test suite
Any docs updates needed?
Yes

temporalio/lib/temporalio/envconfig.rb

cretz · 2025-08-26T13:24:42Z

temporalio/lib/temporalio/envconfig.rb

+    # TLS configuration as specified as part of client configuration
+    #
+    # @!attribute [r] disabled
+    #   @return [Boolean] If true, TLS is explicitly disabled


Isn't this Boolean, nil?

Folding this into the comment below

Unresolving and mentioned later, remember this is a tri-state bool if we want this to properly represent what's in TOML

oop sorry - I knew I had this remaining, was updating the submodule, didn't realize i resolved this

updated

cretz · 2025-08-26T13:27:47Z

temporalio/lib/temporalio/envconfig.rb

+      # Create a ClientConfigTLS from a hash
+      # @param hash [Hash, nil] Hash representation
+      # @return [ClientConfigTLS, nil] The TLS configuration or nil if hash is nil/empty
+      def self.from_hash(hash)


Hrmm, help me remember, why do we have this (and the to_hash) functionality exposed to users? We don't in Go but I saw we do in Python. If it's for TOML serialization ability, I figure that's at the profile level, but even if we do want it here, shouldn't that mean disabled can be a boolean or nil? There's a difference between setting disabled as false and not having it set at all I think.

If we are going to have a to and from hash, they should be symmetrical and exactly represent what is in TOML without defaults

temporalio/sdk-python#895 (comment)

That's a good point. I've mimicked Python here which exposes disabled only as a bool, this is because it's only exposed as a bool from core:

struct TomlClientConfigTLS { #[serde(default, skip_serializing_if = "std::ops::Not::not")] disabled: bool, #[serde(skip_serializing_if = "Option::is_none")] client_cert_path: Option<String>, #[serde(skip_serializing_if = "Option::is_none")] client_cert_data: Option<String>, // String in TOML, not Vec<u8> #[serde(skip_serializing_if = "Option::is_none")] client_key_path: Option<String>, #[serde(skip_serializing_if = "Option::is_none")] client_key_data: Option<String>, // String in TOML, not Vec<u8> #[serde(skip_serializing_if = "Option::is_none")] server_ca_cert_path: Option<String>, #[serde(skip_serializing_if = "Option::is_none")] server_ca_cert_data: Option<String>, // String in TOML, not Vec<u8> #[serde(skip_serializing_if = "Option::is_none")] server_name: Option<String>, #[serde(default, skip_serializing_if = "std::ops::Not::not")] disable_host_verification: bool, } impl TomlClientConfigTLS { fn new() -> Self { Self { disabled: false, client_cert_path: None, client_cert_data: None, client_key_path: None, client_key_data: None, server_ca_cert_path: None, server_ca_cert_data: None, server_name: None, disable_host_verification: false, } }

which uses that default and populates it with the given config. disabled is the only field that is non optional.

If we think this is blocking, I can open a PR. Might be best to do this now given that only Python is released so far.

Yeah, if we want users to be able to convert this to and from TOML, it should represent exactly what the TOML is I figured, right? It's only when we convert to the options a client uses for connect that we should apply defaults IMO.

Up to you on the scheduling of the fix, but yeah prior to marking GA. But if done in separate issue, we might as well remove the to_h/from_h from this PR if we want to get this one merged before that. Up to you.

temporalio/sdk-core#1002
opened core PR

temporalio/lib/temporalio/envconfig.rb

cretz · 2025-08-26T13:45:27Z

temporalio/lib/temporalio/envconfig.rb

+          # If it's a string path (from TOML), read the file
+          # Otherwise return as content
+          if File.exist?(source)


You should know whether this is a path or not from TOML, should not have to use File.exists to switch between behaviors

TOML fields passed in as file paths are now read as Pathname, only data fields are string

temporalio/lib/temporalio/envconfig.rb

cretz · 2025-08-26T13:50:56Z

temporalio/lib/temporalio.rb

@@ -1,5 +1,6 @@
 # frozen_string_literal: true

+require 'temporalio/envconfig'


Doesn't matter much, but I think this should only be required where it is used, though I could also see it being required from client.rb as a convenience, but I don't think it needs to be required from top-level for convenience. I should have said the same about versioning_override when it was added here.

Bump here, we should remove this IMO

missed this, removed

cretz · 2025-08-26T14:07:24Z

temporalio/ext/src/envconfig.rs

+    let hash = RHash::new();
+    match ds {
+        DataSource::Path(p) => {
+            hash.aset("path", ruby.str_new(p))?;


All of the hashes in this Rust library should use symbols for keys IMO. But now that I think about it, maybe we should be returning a Rust struct that is accessible in Ruby. Maybe something like https://docs.rs/magnus/latest/magnus/derive.TypedData.html. Or if we did want struct/data, https://docs.rs/magnus/latest/magnus/struct.Ruby.html#method.define_struct or https://docs.rs/magnus/latest/magnus/struct.Ruby.html#method.define_data.

Not sure, can discuss, but I think hashes of string keys is probably the least safe/efficient for complex objects like these. While efficiency doesn't strictly matter for non-user-facing code and for how rare this is called, I think we should set a reasonable example for complex objects returned to Ruby from calls in Rust.

EDIT: Revisiting the Core envconfig layer, I wonder if it'd be worth having the Rust side serialize to a TOML-like-but-still-typesafe structure, and then we can convert that to Ruby hash. It probably would not be hard to have a serde-to-Ruby-hash. Sure it would still be string hash keys, but it would be cleaner I think. There's not that many data types to convert from TOML/Rust to Ruby.

I opted to just use symbols. I'd like to avoid more extensive changes here to get this in.

cretz · 2025-08-26T14:19:18Z

temporalio/ext/src/envconfig.rs

+    )
+}
+
+// load_client_connect_config(profile: String|nil, path: String|nil, data: String|nil, disable_file: bool, disable_env: bool, config_file_strict: bool, env_vars: Hash|nil)


Feel free to just accept a defined-in-Ruby struct options set as we do in other Rust-from-Ruby calls if that makes it cleaner once it gets to this many parameters

- Refactor ClientConfigTLS, ClientConfigProfile, and ClientConfig to use Data.define

- Rename from_hash → from_h for all envconfig classes - Rename to_hash → to_h for all envconfig classes

- Rename `to_connect_tls_config` → `to_tls_options` - Return `Connection::TLSOptions` object instead of plain Hash

Changed `to_client_connect_options` to return `[positional_args, keyword_args]` tuple instead of a hash, enabling clean one-liner client connections: ```ruby args, kwargs = profile.to_client_connect_options client = Temporalio::Client.connect(*args, **kwargs) ```

…ction - TOML paths (*_path fields) become Pathname objects - TOML data (*_data fields) remain String objects

Changed all hash keys from strings to symbols in the Rust-to-Ruby bridge

cretz · 2025-09-09T13:59:57Z

temporalio/lib/temporalio/envconfig.rb

Just realized, I think this file should be called env_config.rb to match convention if we're capitalizing the C in EnvConfig (which is proper IMO)

renamed the file

cretz · 2025-09-09T14:00:52Z

temporalio/lib/temporalio/envconfig.rb

+    # - String: TOML configuration content
+    # - nil: No configuration source
+
+    # Convert a data source to path and data parameters for the bridge


This does not seem to be prefixed with underscore

temporalio/lib/temporalio/envconfig.rb

cretz · 2025-09-09T14:08:22Z

temporalio/lib/temporalio.rb

@@ -1,5 +1,6 @@
 # frozen_string_literal: true

+require 'temporalio/envconfig'


Bump here, we should remove this IMO

cretz · 2025-09-09T14:12:33Z

temporalio/lib/temporalio/envconfig.rb

+      when Pathname
+        [source.to_s, nil]
+      when String
+        [nil, source.encode('UTF-8').bytes]


Suggested change

[nil, source.encode('UTF-8').bytes]

[nil, source.b]

Shouldn't need to build an array of bytes to pass to Rust, a Ruby string is just fine. Can just accept a https://docs.rs/magnus/latest/magnus/r_string/struct.RString.html on the Rust side instead of a String, and on the Rust side, call unsafe { whatever_your_var_is.as_slice().to_vec() } to get a Vec<u8>.

cretz · 2025-09-09T14:17:32Z

temporalio/lib/temporalio/envconfig.rb

+    # TLS configuration as specified as part of client configuration
+    #
+    # @!attribute [r] disabled
+    #   @return [Boolean] If true, TLS is explicitly disabled


Unresolving and mentioned later, remember this is a tri-state bool if we want this to properly represent what's in TOML

temporalio/lib/temporalio/envconfig.rb

…332)

…ibility

Renamed to_tls_options -> to_client_tls_options

cretz

LGTM, only blocking concern is the other stuff showing up in the PR changeset, but updating branch with main may fix that, then I'll mark approved,

temporalio/lib/temporalio/env_config.rb

cretz · 2025-09-09T21:04:22Z

temporalio/lib/temporalio/internal/worker/workflow_instance.rb

I wonder why all of this other stuff is showing up in the PR, may just need to merge from main

Yup - mostly resolved via merge from main. Had to revert the changes in cancellation.rb explicitly

cretz

LGTM, nothing blocking, thanks!

cretz · 2025-09-10T18:27:17Z

temporalio/lib/temporalio/env_config.rb

I think in theory all of this is "experimental" and we usually should mark it as such (same for Python IMO)

marked experimental

envconfig impl

dd91bde

THardy98 force-pushed the env_config branch from 954b9e2 to dd91bde Compare August 25, 2025 20:00

THardy98 added 2 commits August 25, 2025 16:12

linting

3504fad

fix steep type checks

4d79036

THardy98 force-pushed the env_config branch from aba9891 to 4d79036 Compare August 25, 2025 21:13

THardy98 and others added 2 commits August 25, 2025 14:13

Merge branch 'main' into env_config

e89c4ab

formatting

a79438c

THardy98 changed the title ~~envconfig impl~~ Environment configuration Aug 25, 2025

THardy98 marked this pull request as ready for review August 25, 2025 21:33

THardy98 requested a review from a team as a code owner August 25, 2025 21:33

cretz reviewed Aug 26, 2025

View reviewed changes

THardy98 added 8 commits September 8, 2025 14:49

Refactor envconfig classes to use Data.define for immutability

6dfaca3

- Refactor ClientConfigTLS, ClientConfigProfile, and ClientConfig to use Data.define

Rename hash methods to use idiomatic Ruby naming

3fb3b51

- Rename from_hash → from_h for all envconfig classes - Rename to_hash → to_h for all envconfig classes

Replace TLS config hash with Connection::TLSOptions object

12d3824

- Rename `to_connect_tls_config` → `to_tls_options` - Return `Connection::TLSOptions` object instead of plain Hash

Refactor to use inline pattern instead of temporary variables

a3696a0

Make hash methods private

902dbfe

Remove File.exist? check by using type system for path vs data distin…

4979a3d

…ction - TOML paths (*_path fields) become Pathname objects - TOML data (*_data fields) remain String objects

Use symbol keys instead of string keys in Rust FFI bridge

33eccab

Changed all hash keys from strings to symbols in the Rust-to-Ruby bridge

cretz reviewed Sep 9, 2025

View reviewed changes

dependabot bot and others added 9 commits September 9, 2025 08:02

Bump golang.org/x/net in /temporalio/test/golangworker (#303)

1ded95f

Bump tracing-subscriber from 0.3.19 to 0.3.20 in /temporalio (#331)

0d16b46

Reduce the scope of the illegal call tracer and other tracing fixes (#…

5c49d24

…332)

Remove unnecessary fully qualified import names

b0b337e

Remove envconfig import in temporalio.rb

928070c

_source_to_path_and_data prefixed with underscore and private doc vis…

fae411e

…ibility

TLS disabled field now optional

35c2f47

Rubocop fixes

5af7ceb

Steep fixes

7f2f777

THardy98 force-pushed the env_config branch from dbeb122 to 7f2f777 Compare September 9, 2025 16:25

THardy98 added 4 commits September 9, 2025 09:27

Renamed envconfig.rb -> env_config.rb

4a276af

Renamed to_tls_options -> to_client_tls_options

Pass data string as RString, Rust bridge converts to Vec

6fe479e

Rename load_client_connect_config -> load_client_connect_options

20f28ff

Format envconfig.rs

dee3b0a

THardy98 requested a review from cretz September 9, 2025 18:11

cretz reviewed Sep 9, 2025

View reviewed changes

THardy98 and others added 4 commits September 9, 2025 14:43

Merge branch 'main' into env_config

f3b478f

Nits

656a511

Update core submodule

cde7c6f

Revert irrelevant cancellation changes

89e03de

THardy98 requested a review from cretz September 9, 2025 22:30

Update Cargo.lock

aeaf17a

THardy98 force-pushed the env_config branch from b529547 to b4280b8 Compare September 10, 2025 00:37

Propagate set_headers failure in client bridge

6db2dcc

THardy98 force-pushed the env_config branch from b4280b8 to 6db2dcc Compare September 10, 2025 00:46

cretz approved these changes Sep 10, 2025

View reviewed changes

Add experimental warning

edfed08

THardy98 merged commit 51b5f95 into main Sep 10, 2025
8 checks passed

THardy98 deleted the env_config branch September 10, 2025 19:29

		@@ -1,5 +1,6 @@
		# frozen_string_literal: true

		require 'temporalio/envconfig'

Environment configuration #326

Environment configuration #326

Uh oh!

Conversation

THardy98 commented Aug 23, 2025

What was changed

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

THardy98 Sep 9, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

THardy98 Sep 5, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

cretz Sep 5, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

cretz left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

cretz left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

THardy98 Sep 9, 2025 •

edited

Loading

THardy98 Sep 5, 2025 •

edited

Loading

cretz Sep 5, 2025 •

edited

Loading