build: run tests pull requests for all branches #426
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
olavloite
added a commit
that referenced
this pull request
Jun 4, 2025
* chore: find parameters with comments in place Find query parameters in the SQL string while keeping comments in place. This both improves performance, as we need to copy the string fewer times, but more imporantly; it keeps the comments in the SQL string that is sent to Spanner. The latter is important for the PostgreSQL dialect, as query hints in that dialect are embedded into comments. * build: run tests pull requests for all branches (#426) * chore: ignore lint errors * perf: add a cache for parsed statements (#425) * perf: add a cache for parsed statements The functions for determining the type of statement and which query parameters are in a statement are deterministic and do not change over time. These can therefore safely be cached. This improves the execution speed of frequently executed SQL strings, as the same parsing does not need to happen for each execution. * chore: go mod tidy * chore: create statementParser struct and make it dialect-aware (#427) * chore: create statementParser struct and make it dialect-aware Refactors the statement parser into a struct that contains a field for the database dialect that is being used. This dialect field will be used in follow-up pull request to implement dialect-specific parsing. * chore: remove duplicate quoted string parsing (#428) * chore: remove duplicate quoted string parsing The calculateFindParams function contained its own logic for skipping over quoted string literals. This code was redundant, as it is also implemented in the skip function. This change removes this duplication and simplifies the calculateFindParams implementation significantly. This also ensures that we only have one place where string literals are found and skipped in the parser, which will make it easier to make this code dialect-aware. * chore: add support for PostgreSQL-style comments (#429) * chore: add support for PostgreSQL-style comments Adds support for PostgreSQL-style comments and sets some of the tests up for multi-dialect testing. * chore: support PostgreSQL-style parameters (#431) * chore: support PostgreSQL-style parameters Adds support for both recognizing PostgreSQL-style query parameters, and converting positional parameters to PostgreSQL-style query parameters. When using the PostgreSQL-dialect, the driver accepts both GoogleSQL and PostgreSQL-style named parameters (so both @param and $1). This to be consistent with other PostgreSQL drivers in Go (e.g. pgx). * chore: add PostgreSQL quoting rules (#432) * chore: add PostgreSQL quoting rules Add PostgreSQL quoting rules to the parser. This means: 1. Backslashes do not start an escape sequence. 2. Repeating the same quote twice inside a string literal is an escaped quote. 3. Backtick is not a valid quote character. 4. Triple-quoted strings are not supported. However, if the start of a string consists of three consequtive quotes, then that means 'start of a string' and then 'an escaped quote'. 5. Linefeeds are allowed in string literals. This change does not add support for dollar-quoted strings. That will be added in a follow-up pull request. * chore: add support for dollar-quoted strings (#433) * chore: add support for dollar-quoted strings Adds support for dollar-quoted strings. Also fixes an issue in the skipWhitespaces function, where multi-byte characters would be skipped as-if they were spaces. * test: add some more tests for PostgreSQL parsing (#434) * test: add some more tests for PostgreSQL parsing * feat: detect database dialect and support PostgreSQL databases (#435) * feat: detect database dialect and support PostgreSQL databases Automatically detect the dialect of the database that the driver connects to, and modify the way that query parameters are recognized based on the dialect of the database. This makes it possible to use this driver with Spanner databases that use the PostgreSQL dialect. * perf: cache client-side SQL parsing (#436) Cache the outcome of parsing a potential client-side SQL statement to prevent the same regular expressions to be tested repeatedly. Also, the initial check whether a statement is a client-side statement has been replaced with a keyword based check: Instead of trying to match the SQL statement against the list of all possible client-side statements, the parser first checks whether the first keyword is a keyword that could potentially be a client-side statement. If it is not, then the function returns early. Benchmarks: ``` BenchmarkDetectStatementTypeWithCache-8 12530068 942 ns/op BenchmarkDetectStatementTypeWithoutCache-8 2861143 4182 ns/op ```
olavloite
added a commit
that referenced
this pull request
Jun 4, 2025
* chore: add simple token parser * chore: add multi-byte support * chore: add benchmark * chore: find parameters with comments in place (#424) * chore: find parameters with comments in place Find query parameters in the SQL string while keeping comments in place. This both improves performance, as we need to copy the string fewer times, but more imporantly; it keeps the comments in the SQL string that is sent to Spanner. The latter is important for the PostgreSQL dialect, as query hints in that dialect are embedded into comments. * build: run tests pull requests for all branches (#426) * chore: ignore lint errors * perf: add a cache for parsed statements (#425) * perf: add a cache for parsed statements The functions for determining the type of statement and which query parameters are in a statement are deterministic and do not change over time. These can therefore safely be cached. This improves the execution speed of frequently executed SQL strings, as the same parsing does not need to happen for each execution. * chore: go mod tidy * chore: create statementParser struct and make it dialect-aware (#427) * chore: create statementParser struct and make it dialect-aware Refactors the statement parser into a struct that contains a field for the database dialect that is being used. This dialect field will be used in follow-up pull request to implement dialect-specific parsing. * chore: remove duplicate quoted string parsing (#428) * chore: remove duplicate quoted string parsing The calculateFindParams function contained its own logic for skipping over quoted string literals. This code was redundant, as it is also implemented in the skip function. This change removes this duplication and simplifies the calculateFindParams implementation significantly. This also ensures that we only have one place where string literals are found and skipped in the parser, which will make it easier to make this code dialect-aware. * chore: add support for PostgreSQL-style comments (#429) * chore: add support for PostgreSQL-style comments Adds support for PostgreSQL-style comments and sets some of the tests up for multi-dialect testing. * chore: support PostgreSQL-style parameters (#431) * chore: support PostgreSQL-style parameters Adds support for both recognizing PostgreSQL-style query parameters, and converting positional parameters to PostgreSQL-style query parameters. When using the PostgreSQL-dialect, the driver accepts both GoogleSQL and PostgreSQL-style named parameters (so both @param and $1). This to be consistent with other PostgreSQL drivers in Go (e.g. pgx). * chore: add PostgreSQL quoting rules (#432) * chore: add PostgreSQL quoting rules Add PostgreSQL quoting rules to the parser. This means: 1. Backslashes do not start an escape sequence. 2. Repeating the same quote twice inside a string literal is an escaped quote. 3. Backtick is not a valid quote character. 4. Triple-quoted strings are not supported. However, if the start of a string consists of three consequtive quotes, then that means 'start of a string' and then 'an escaped quote'. 5. Linefeeds are allowed in string literals. This change does not add support for dollar-quoted strings. That will be added in a follow-up pull request. * chore: add support for dollar-quoted strings (#433) * chore: add support for dollar-quoted strings Adds support for dollar-quoted strings. Also fixes an issue in the skipWhitespaces function, where multi-byte characters would be skipped as-if they were spaces. * test: add some more tests for PostgreSQL parsing (#434) * test: add some more tests for PostgreSQL parsing * feat: detect database dialect and support PostgreSQL databases (#435) * feat: detect database dialect and support PostgreSQL databases Automatically detect the dialect of the database that the driver connects to, and modify the way that query parameters are recognized based on the dialect of the database. This makes it possible to use this driver with Spanner databases that use the PostgreSQL dialect. * perf: cache client-side SQL parsing (#436) Cache the outcome of parsing a potential client-side SQL statement to prevent the same regular expressions to be tested repeatedly. Also, the initial check whether a statement is a client-side statement has been replaced with a keyword based check: Instead of trying to match the SQL statement against the list of all possible client-side statements, the parser first checks whether the first keyword is a keyword that could potentially be a client-side statement. If it is not, then the function returns early. Benchmarks: ``` BenchmarkDetectStatementTypeWithCache-8 12530068 942 ns/op BenchmarkDetectStatementTypeWithoutCache-8 2861143 4182 ns/op ``` * chore: rename function to skipWhitespacesAndComments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Run tests for pull requests on any branch.