SQL: extract types in "create type t" by masatake · Pull Request #4383 · universal-ctags/ctags

masatake · 2026-02-10T03:51:28Z

https://www.postgresql.jp/document/17/html/sql-createtype.html

codecov · 2026-02-10T03:59:01Z

Codecov Report

❌ Patch coverage is 89.47368% with 4 lines in your changes missing coverage. Please review.
✅ Project coverage is 85.93%. Comparing base (1bccbf2) to head (39d97ca).
⚠️ Report is 7 commits behind head on master.

Files with missing lines	Patch %	Lines
parsers/sql.c	89.47%	4 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #4383      +/-   ##
==========================================
+ Coverage   85.91%   85.93%   +0.02%     
==========================================
  Files         252      252              
  Lines       62683    62715      +32     
==========================================
+ Hits        53854    53895      +41     
+ Misses       8829     8820       -9

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

Copilot

Pull request overview

Adds first-class tagging support for SQL CREATE TYPE statements so type names (e.g., t in CREATE TYPE t ...) can be extracted by the SQL parser, with accompanying unit tests and release notes.

Changes:

Introduce a new SQL kind type (k) and emit it from parseType.
Suppress the type tag when a more specific kind (record, table, cursor) is emitted for the same name.
Add/adjust unit tests covering PostgreSQL CREATE TYPE variants and update news.

Reviewed changes

Copilot reviewed 7 out of 7 changed files in this pull request and generated 3 comments.

Show a summary per file

File	Description
parsers/sql.c	Adds `SQLTAG_TYPE` kind and updates `parseType` to emit/suppress type tags appropriately.
docs/news/HEAD.rst	Documents the new SQL `type` kind extraction behavior.
Units/parser-sql.r/types.d/input.sql	New fixture with PostgreSQL `CREATE TYPE` examples.
Units/parser-sql.r/types.d/expected.tags	Expected tags for the new `types.d` test case, including `k` kind outputs.
Units/parser-sql.r/types.d/args.ctags	Test args for the new `types.d` suite.
Units/parser-sql.r/funcions.d/expected.tags	Updates expected tags to reflect record tagging behavior for `create type ... as (...)`.
Units/parser-sql.r/funcions.d/args.ctags	Enables SQL `record` kind (`+r`) for the `funcions.d` test suite to match updated expectations.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

parsers/sql.c

Copilot

Pull request overview

Copilot reviewed 7 out of 7 changed files in this pull request and generated 2 comments.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

parsers/sql.c

Copilot

Pull request overview

Copilot reviewed 7 out of 7 changed files in this pull request and generated 2 comments.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot · 2026-02-10T14:16:47Z

parsers/sql.c

+					markCorkEntryAsPlaceholder(r, true);
 					makeSqlTag (name, SQLTAG_RECORD);


In the KEYWORD_record/KEYWORD_object branch, the code always marks the newly-created SQLTAG_TYPE entry as a placeholder before attempting to emit SQLTAG_RECORD. If the record kind is disabled (SqlKinds['r'] is false by default), makeSqlTag(name, SQLTAG_RECORD) returns CORK_NIL and the placeholder call suppresses the only tag for this CREATE TYPE statement. Consider only calling markCorkEntryAsPlaceholder(r, true) when SQLTAG_RECORD is enabled / when makeSqlTag(...) actually created a tag, so a type tag is still emitted when record tags are disabled.

Suggested change

markCorkEntryAsPlaceholder(r, true);

makeSqlTag (name, SQLTAG_RECORD);

if (isLanguageKindEnabled (Lang_sql, SQLTAG_RECORD))

{

markCorkEntryAsPlaceholder(r, true);

makeSqlTag (name, SQLTAG_RECORD);

}

Copilot · 2026-02-10T14:16:48Z

parsers/sql.c

+					markCorkEntryAsPlaceholder(r, true);
 					makeSqlTag (name, SQLTAG_TABLE);
 					break;

 				case KEYWORD_ref:
 					readToken (token);
 					if (isKeyword (token, KEYWORD_cursor))
+					{
+						markCorkEntryAsPlaceholder(r, true);
 						makeSqlTag (name, SQLTAG_CURSOR);


Similar to the record/object branch, the KEYWORD_table and KEYWORD_ref cursor branches unconditionally call markCorkEntryAsPlaceholder(r, true). If the alternative kind (table/cursor) is disabled via --kinds-SQL, this will suppress the type tag and leave the CREATE TYPE statement without any tag. Consider making the placeholder suppression conditional on successfully creating the table/cursor tag (or on isLanguageKindEnabled for that kind).

Suggested change

markCorkEntryAsPlaceholder(r, true);

makeSqlTag (name, SQLTAG_TABLE);

break;

case KEYWORD_ref:

readToken (token);

if (isKeyword (token, KEYWORD_cursor))

{

markCorkEntryAsPlaceholder(r, true);

makeSqlTag (name, SQLTAG_CURSOR);

if (isLanguageKindEnabled (Lang_sql, SQLTAG_TABLE))

{

/* When the table kind is enabled, emit a

* table tag and suppress the type tag. */

markCorkEntryAsPlaceholder(r, true);

makeSqlTag (name, SQLTAG_TABLE);

}

break;

case KEYWORD_ref:

readToken (token);

if (isKeyword (token, KEYWORD_cursor))

{

if (isLanguageKindEnabled (Lang_sql, SQLTAG_CURSOR))

{

/* When the cursor kind is enabled, emit a

* cursor tag and suppress the type tag. */

markCorkEntryAsPlaceholder(r, true);

makeSqlTag (name, SQLTAG_CURSOR);

}

Copilot

Pull request overview

Copilot reviewed 10 out of 10 changed files in this pull request and generated 3 comments.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

parsers/sql.c

misc/units.py

Units/parser-sql.r/types.d/input.sql

So units.py runs the unit test case and can report "FAILED (broken args.ctags?)" failure. Signed-off-by: Masatake YAMATO <yamato@redhat.com>

https://www.postgresql.jp/document/17/html/sql-createtype.html Signed-off-by: Masatake YAMATO <yamato@redhat.com>

Signed-off-by: Masatake YAMATO <yamato@redhat.com>

Copilot

Pull request overview

Copilot reviewed 10 out of 10 changed files in this pull request and generated 1 comment.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

parsers/sql.c

Signed-off-by: Masatake YAMATO <yamato@redhat.com>

Copilot

Pull request overview

Copilot reviewed 11 out of 11 changed files in this pull request and generated 1 comment.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot · 2026-02-11T08:58:52Z

parsers/sql.c

+/* Make the tag that OLDTAGINDEX specifies a placeholder, then make a new
+ * tag for TOKEN with KIND kind. */
+static int makeSqlTagInstead (tokenInfo *const token, const sqlKind kind, int oldTagIndex)
+{
+	markCorkEntryAsPlaceholder(oldTagIndex, true);
+	return makeSqlTag (token, kind);
+}


makeSqlTagInstead() only marks the original (leaf) cork entry as a placeholder. If XTAG_QUALIFIED_TAGS is enabled and the original tag had a non-empty scope, makeSqlTagFull() will also have emitted a separate qualified-tag entry; that qualified entry won’t be marked placeholder and can leak into the output even though the leaf tag was suppressed. Consider capturing the qualified-tag cork index when creating the original tag (use makeSqlTagFull(...,&fq)), and when replacing, mark both the leaf and qualified entries as placeholders (and likewise create the replacement tag via makeSqlTagFull so the replacement gets a qualified entry when applicable).

Marking fq tags as placeholder is not needed in this pull request.

masatake requested a review from Copilot February 10, 2026 04:29

Copilot started reviewing on behalf of masatake February 10, 2026 04:29 View session

Copilot AI reviewed Feb 10, 2026

View reviewed changes

parsers/sql.c Outdated Show resolved Hide resolved

parsers/sql.c Outdated Show resolved Hide resolved

parsers/sql.c Outdated Show resolved Hide resolved

masatake force-pushed the sql--type-for-postgresql branch from b180345 to 3046a12 Compare February 10, 2026 05:43

masatake requested a review from Copilot February 10, 2026 13:15

Copilot started reviewing on behalf of masatake February 10, 2026 13:15 View session

masatake force-pushed the sql--type-for-postgresql branch from 3046a12 to 41a70dd Compare February 10, 2026 13:18

Copilot AI reviewed Feb 10, 2026

View reviewed changes

parsers/sql.c Show resolved Hide resolved

parsers/sql.c Outdated Show resolved Hide resolved

masatake force-pushed the sql--type-for-postgresql branch from 41a70dd to d39ebed Compare February 10, 2026 13:53

masatake requested a review from Copilot February 10, 2026 14:11

Copilot started reviewing on behalf of masatake February 10, 2026 14:12 View session

Copilot AI reviewed Feb 10, 2026

View reviewed changes

masatake force-pushed the sql--type-for-postgresql branch from d39ebed to 0f95654 Compare February 10, 2026 23:35

masatake requested a review from Copilot February 11, 2026 00:03

Copilot started reviewing on behalf of masatake February 11, 2026 00:03 View session

Copilot AI reviewed Feb 11, 2026

View reviewed changes

parsers/sql.c Show resolved Hide resolved

misc/units.py Outdated Show resolved Hide resolved

Units/parser-sql.r/types.d/input.sql Outdated Show resolved Hide resolved

masatake added 4 commits February 11, 2026 10:53

units.py: don't consider "Filtered by LANGUAGES" if args.ctags is broken

6eae027

So units.py runs the unit test case and can report "FAILED (broken args.ctags?)" failure. Signed-off-by: Masatake YAMATO <yamato@redhat.com>

SQL: extract types in "create type t"

661e091

https://www.postgresql.jp/document/17/html/sql-createtype.html Signed-off-by: Masatake YAMATO <yamato@redhat.com>

SQL: extract type declaration with "type" kind and "decl" role

e69e056

Signed-off-by: Masatake YAMATO <yamato@redhat.com>

SQL: consdier name including '.' when extracting type kind tags

4b0e801

Signed-off-by: Masatake YAMATO <yamato@redhat.com>

masatake force-pushed the sql--type-for-postgresql branch from 0f95654 to 4b0e801 Compare February 11, 2026 01:55

masatake requested a review from Copilot February 11, 2026 01:57

Copilot started reviewing on behalf of masatake February 11, 2026 02:03 View session

Copilot AI reviewed Feb 11, 2026

View reviewed changes

parsers/sql.c Show resolved Hide resolved

masatake added 2 commits February 11, 2026 17:21

SQL: accept non-reserved keywords as types

009fa42

Signed-off-by: Masatake YAMATO <yamato@redhat.com>

SQL: skip "BODY", a reserved word when parsing "CREATE TYPE BODY t ..."

39d97ca

Signed-off-by: Masatake YAMATO <yamato@redhat.com>

masatake requested a review from Copilot February 11, 2026 08:48

Copilot started reviewing on behalf of masatake February 11, 2026 08:52 View session

Copilot AI reviewed Feb 11, 2026

View reviewed changes

masatake merged commit 381d57e into universal-ctags:master Feb 11, 2026
91 of 92 checks passed

		markCorkEntryAsPlaceholder(r, true);
		makeSqlTag (name, SQLTAG_RECORD);

Conversation

masatake commented Feb 10, 2026

Uh oh!

codecov bot commented Feb 10, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Copilot AI Feb 10, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 10, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Copilot AI Feb 11, 2026

Choose a reason for hiding this comment

Uh oh!

masatake Feb 11, 2026

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

codecov bot commented Feb 10, 2026 •

edited

Loading