Commit Graph

16 Commits

Author SHA1 Message Date
8a1fd5438e chore: Add tests for LLMClientInterface (#131) 2025-03-10 21:54:17 +01:00
3aae923d43 feat: Improve describe recent changes in system prompt (#113)
Co-authored-by: Petr Mironychev <9195189+Palm1r@users.noreply.github.com>
2025-03-08 08:58:28 +01:00
f94c79a5ff fix: Improve support for code blocks without language (#108)
This makes it possible to represent code blocks in models that emit
their suggestion immediately after the ``` characters.
2025-03-07 15:30:22 +01:00
9a5047618d chore: Silence Qt warnings during tests (#110) 2025-03-07 01:58:09 +01:00
90beebf2ee Revert "refactor: Move all processing logic to CodeHandler::processText()" (#109) 2025-03-07 01:57:13 +01:00
521261e0a3 refactor: Move all processing logic to CodeHandler::processText() (#107)
This will become useful once more processing modes are available
2025-03-06 18:49:28 +01:00
81ac3c71fb chore: Add tests for CodeHandler (#105)
* chore: Extract test utils to separate file

This makes it possible to reuse the utils in other test files.

* chore: Add tests for CodeHandler
2025-03-06 13:45:12 +01:00
61ca5c9a1b fix: Properly omit copyright information (#103)
This commit ensures that copyright information is always excluded and
that context is always split into prefix and suffix at correct position.
2025-03-06 13:00:15 +01:00
8a167bf248 chore: Expand DocumentContextReaderTest to cover more conditions (#102)
Tests should have "before" and "after" cases side by side, so that it's
possible to easily verify that full context is extracted correctly.

Also tests should consistently cover the same conditions in all the
different scenarios.
2025-03-06 12:08:57 +01:00
ab97f39ea4 chore: Fix pretty printing for QString (#101)
Previously QString was printed symbol by symbol. E.g.:

reader.readWholeFileAfter(3, 1)
    Which is: { "i", "n", "e", " ", "2", "
", "L", "i", "n", "e", " ", "3" }
2025-03-06 12:08:16 +01:00
1d062e1fe4 chore: Use zero-based line numbers in tests (#100)
This makes the line index arguments passed to functions match the actual
test data, thus tests are less confusing.
2025-03-06 11:53:27 +01:00
5dceb5cd19 fix: Bring back old behavior of readStrings{After,Before}Cursor setting (#97) 2025-03-05 20:27:16 +01:00
69a8aa80d9 refactor: Make DocumentContextReader::prepareContext() testable (#96) 2025-03-05 20:18:59 +01:00
3dc0d910bf fix: Fix off by one errors in getContext{Before,After}() (#94)
This also specifies what exactly getContext*() functions do. Before this
commit linesCount was sometimes interpreted as exclusive of current
line, which was confusing as linesCount + 1 lines were being returned.
2025-03-05 19:32:53 +01:00
f9f2a86cea fix: Correctly pick whole file context (#85)
Currently the current line is duplicated in both "before" and "after"
context. This is due to DocumentContextReader::readWholeFileAfter()
picking the current line part of which has been already included into
the "before" context.
2025-03-05 19:17:51 +01:00
247256d4a4 chore: Add unit tests for DocumentContextReader (#90)
chore: Add unit tests for DocumentContextReader

The tests are based on GTest like some tests in Qt Creator itself, which
makes it easy to run as full Qt Creator does not need to be started.
2025-03-05 15:01:52 +01:00