The --type filter was concatenated into ImmutableKeywords in
SearchQueryBuild, which bypassed search.Query's quoting and let values
that contained quotes (or extra qualifiers) corrupt the final query.
Per babakks's suggestion in the review thread, model it as a regular
qualifier instead.
Add an IssueType field to Qualifiers tagged `qualifier:"type"` so it
shares a key with the existing Type field, and rework Qualifiers.Map
to honour the tag and concatenate values when multiple fields share
the same key. SearchQueryBuild now drops its bespoke type:X
formatting and just sets Qualifiers.IssueType, leaving the
keyword/qualifier escaping to pkg/search.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Add greppable TODO identifiers above all if-statements that reference
featuredetection struct fields, as required by the featuredetection
linter. This ensures every feature detection branch is tagged for
future cleanup when GHES gains support.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
This commit moves the remaining searcher tests from using JSON marshaled types to using JSON responses for consistency. There appears to be a weird JSON marshaling error with search.Repository that does not map `Name` field in the process.
Additionally, the test scenarios around pulling multiple pages beneath the total results have been updated to demonstrate that the REST API returns full pages in both of these cases, which is below the total number of results.
This commit is focused on fixing the `searcher` tests for a few reasons:
1. Correcting the `.Total` logic in the previous commit caused changes to tests to fail
2. Tests involving results that exceed the max per page have been improved with new initialize helper, allowing testing table scenarios to be self contained
3. Tests that stub JSON response payloads have been standardized on maps rather than GitHub type primitives (Repository, Issue, Commit, Code, etc)
4. Tests had some minor formatting changes to make them easier to understand and maintain
This change restores the original logic of passing the search total count logic as is to the result.
Additionally, this undoes some of the contributor's formatting changes that increase the changed lines to review.
