Release v1.28.0 #528
Merged
Release v1.28.0 #528
+1,260
−386
Conversation
This file contains 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
Go back to development after releasing v1.27.0.
Add a FOSSA check to the build steps for ThriftRW Go. Refs GO-468
Currently, the FOSSA analysis is set to run as part of CI. This makes it impossible to run builds for incoming pull requests because PRs don't have access to the secrets needed for the FOSSA analysis. Resolve this by running the FOSSA analysis only when we push to a branch of the repository.
Thrift allows implicit assignment of field identifiers. This is legacy syntax, but it's useful to support it for compatibility with old files. Auto-assigned fields start at -1 and get increasingly negative. If a negative field identifier is explicitly given, it will be used as the new basis for any additional auto-assigned fields. This behavior is consistent with the Apache Thrift implementation.
Previously, parse errors (`idl.internal.parseError`) weren't visible to `idl.Parse()`'s callers. This was inconvenient because it's useful to have access to (a) the full list of parse errors and (b) the line number that caused each parse error. This change introduces `idl.ParseError` which provides a list of errors and their lines. Fixes #489
The templating language used for code generation uses `<` and `>` as its delimiters. This results in syntax errors when trying to use the beginning delimiter as an actual character, not as invoking a template. Add an explicit command (`<lessthan>`) to inject `<` as needed.
Line-based data types are too limiting should we add support for other positional information, such as offsets/columns.
This change does two things: First, it records document positions for parsed nodes that cannot store their own position data. This primarily applies to primitive constants. Second, it makes this information available through a new Config-based parsing interface using an optional idl.Info structure. Node positions are available though idl.Info.Pos(), which first attempts to return a node's "native" position (line) value before falling back to its internal nodePositions map. Closes #493
With idl.Position moved to ast.Position, there's no need for a separate internal.Position type.
This improves stack-level reports for test failures more accurate.
This doesn't change anything in current practice aside from moving to the latest stable version of Ragel. From the changelog: Ragel 6.10 - Mar 24, 2017 ========================= -C codegen: test P vs PE in goto/call/ret statements in EOF actions, just before re-entering. If at the end of the input block then the EOF check is jumped to. This change prevents overrunning the buffer if control flow is issued in an EOF action without fixing the input pointer first. If a program properly issues an fhold before the control flow the program won't be affected. -Updated action label generation. The previous set of conditions for generating the label didn't cover actions coming from the eofAction pointer (eof trans covered since it points into the set of transitions). -Use separate signed/unsigned values for host type min/max. Using separate values avoids the need to type cast before the data goes into FsmCtx structs. Keep it in native types until it is used. -Optionally do not generate entry point variables. Adds noentry write option for data. -Various warning elimination and build updates.
Add a new hidden flag `--enum-text-marshal-strict` that will cause the MarshalText and MarshalJSON methods methods for enums to fail with an error if the enum being serialized is unknown. This change is necessary because some clients crash during serialization of JSON or YAML if they receive an integer instead of a string (for unknown enums). Refs EDGE-577 Co-authored-by: Abhinav Gupta <abg@uber.com>
Codecov Report
@@ Coverage Diff @@
## master #528 +/- ##
==========================================
+ Coverage 78.98% 79.02% +0.04%
==========================================
Files 125 129 +4
Lines 16055 16199 +144
==========================================
+ Hits 12681 12802 +121
- Misses 2064 2086 +22
- Partials 1310 1311 +1
Continue to review full report at Codecov.
|
Update the changelog to reflect the commits going in this release.
r-hang
approved these changes
Jul 26, 2021
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
This release v1.28.0, which includes a series of changes cherry-picked from
dev. It skips the following changes that are currently on dev:progress in the streamdev branch and should not be released without the rest
of those changes.
AST identities. These cannot be released without ast: add column values to the AST nodes #522, which is not yet
ready.
Besides that, all changes from dev are included in this change.
Here's an API comparison between the last release (v1.27.0) and this release,
generated with the help of apidiff.
Of the two incompatible changes: