Make --dbname and --dbuser optional when SQLite is active

[abhi3315]

wp config create requires --dbname and --dbuser even when the SQLite integration plugin is active. SQLite doesn't use these parameters, so users are forced to pass dummy MySQL credentials to create a wp-config.php for a SQLite-backed site.

Addresses the enhancement suggested by @mrsdizzie in #167 (comment).

Also incorporates all review feedback from #219 (batched error messages, is_readable() guard, isset() instead of empty()).

Changes

• Synopsis: --dbname and --dbuser are now optional ([--dbname=<dbname>], [--dbuser=<dbuser>])
• SQLite detection: New is_sqlite_integration_active() method scans wp-content/db.php for SQLITE_DB_DROPIN_VERSION, matching the approach used by DB_Command_SQLite::is_sqlite()
• Validation: When SQLite is not detected, missing --dbname/--dbuser still produces an error with both parameters reported in a single message
• MySQLi check: Skipped entirely when SQLite is detected

Test coverage

• New @require-sqlite scenario: verifies wp config create --skip-salts succeeds without --dbname/--dbuser when the SQLite drop-in is present
• New @skip-sqlite scenario: verifies the error message when both parameters are missing on MySQL/MariaDB

Fixes #167

--

Related wp-cli/wp-cli#6295

[codecov]

Codecov Report

❌ Patch coverage is 88.88889% with 2 lines in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
src/Config_Command.php	88.88%	2 Missing ⚠️

📢 Thoughts on this report? Let us know!

[gemini-code-assist]

Code Review

This pull request introduces support for SQLite integration in the wp config create command, allowing users to omit the --dbname and --dbuser parameters when an SQLite drop-in is detected. The changes include a new helper method to check for the SQLite drop-in, updated parameter validation logic, and corresponding feature tests. I have provided feedback regarding the robust construction of the wp-content directory path to avoid potential path concatenation errors and suggested simplifying the error messages for better clarity.

[swissspidy]

Please note that we already have an active PR for this at #219...

[abhi3315]

Hey @swissspidy. I missed that #219 had been updated recently. When I checked it earlier there were unaddressed review comments, but I see it's been iterating since.

If any of the specific fixes in this PR are useful (e.g. the rtrim(ABSPATH, '/\') for consistency with existing code in the same method, or the @skip-sqlite scenario for MySQL validation coverage), happy to move those as suggestions to #219.

[swissspidy]

Yes please move this to suggestions on the other PR, thank you!

[swissspidy]

Closing in favor of #219