[LucaCappelletti94]

ALTER VIEW v OWNER TO o and ALTER VIEW v SET (security_invoker = true) fail with Expected: AS, found: OWNER, because parse_alter_view only knows the MySQL form ALTER VIEW v AS SELECT .... The other PostgreSQL forms of the statement fail the same way.

Statement::AlterView now wraps an AlterView struct holding IF EXISTS and an AlterViewOperation, the shape AlterCollation already uses. The MySQL form becomes one variant beside the PostgreSQL ones, which changes the public AST of Statement::AlterView. PostgreSQL lets a SET (...) option omit its value, so such an option parses to SqlOption::Ident.

The syntax follows https://www.postgresql.org/docs/current/sql-alterview.html.

Closes #886.

[codecov-commenter]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 82.81%. Comparing base (2d9629f) to head (4e8582d).
⚠️ Report is 3 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #2730      +/-   ##
==========================================
+ Coverage   82.67%   82.81%   +0.13%     
==========================================
  Files          42       42              
  Lines       35737    35865     +128     
  Branches    35737    35865     +128     
==========================================
+ Hits        29547    29702     +155     
+ Misses       2671     2653      -18     
+ Partials     3519     3510       -9     

☔ View full report in Codecov by Harness.
📢 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.

[sabir-akhadov-localstack]

Pre-existing, but the parser reads (columns) before WITH (...) while this writes them the other way round, so ALTER VIEW v (a) WITH (x = 1) AS SELECT 1 displays as ALTER VIEW v WITH (x = 1) (a) AS SELECT 1, which doesn't re-parse. Swapping the two blocks here fixes the round-trip.

[LucaCappelletti94]

Done!

[sabir-akhadov-localstack]

parse_alter_view has no dialect check, so these could use all_dialects() next to the existing parse_alter_view tests in sqlparser_common.rs.

[LucaCappelletti94]

Moved, and adjusted for the DuckDB corner case