Added tests for issues #224/#225 and #229. Since
these are known bugs, the tests are currently
marked as expected failures. Once the solution
is implemented, that designation can be removed
and we'll still keep the test to make sure there
aren't any regressions.
Added a test which demonstrates a current bug where
if multiple TODOs with the same title appear in
the diff of the same file, only one of those issues
will have its URL succesfully inserted.
(Specifically, the first from the top.) As such,
the test is configured as an expected failure until
such time as the bug can be fixed.
The diff file has a couple of hunks which for
some unknown reason have parsing issues which
result in the wrong line being detected for where
the TODO is. This causes the issue URL insertion
to fail. That's its own issue that should be
investigated, but in order to allow the process_diff
unit test to pass, the diff file is being split
into two to isolate the trouble-causing hunks.
If regex characters are part of the title (for
example, parentheses), then this breaks the ability
to find the comment line when trying to insert
the issue URL. This commit escapes the title
to fix this problem.
Closes github.com/alstr/todo-to-issue-action#226
When searching for the comment line after which
the issue URL will be inserted, use a case-insensitive
search for the identifier portion of the line.
This matches how TodoParser identifies TODOs using
a case-insensitive search.
Closes github.com/alstr/todo-to-issue-action#224
There are issues currently where the wrong text
may be inserted in place of the issue URL. (See
github.com/alstr/todo-to-issue-action issues #224
, #225, and #226, for example.) When such a
situation is detected, the offending line is now
NOT inserted and instead a warning message is
printed out. This doesn't completely resolve
those issues, but mitigates against them and
other similar issues which may arise in the future.
Insertion of the URL is different from simply
creating/updating the issue, so we want to have
this level of output to ensure that as many
file updates as expected are actually occurring.
Note: The unit test is currently failing!
Reads the diff file to generate a simulated
filesystem, inserts the URL comments into these
new files, and then parses the output to confirms
that all of the issues that were expected to be
created were created.
test_new.diff had some issues which would cause
errors if trying to actually use the diff file
with 'patch' to create the files it references.
1) had some line numbers which were invalid
2) had path specifications and, unfortunately,
'patch' does not support creating absent directories
The file has been cleaned up to have the create
line numbers and to assume that all files are at
the root.
Additionally, for the few cases where the diff
file referred to an edit rather than creation of a
whole new file, those diff hunks were moved into a
new, separate, file.
There is no functional change with this commit,
but it sets up future test features.
Fix bug where if there were multiple issue URLs
to insert into a file, after the first insertion
all further insertions would copy a "random"
line (based on *original* line numbering) and
insert that copy rather than the issue URL.
Now processes files in order by file name
and then in reverse order by line number,
thereby ensuring that any line insertions/
deletions at bottom of file don't affect
line numbering above them.
Closes github.com/alstr/todo-to-issue-action#225
New unittest file consists of two tests. Both
run mypy, but one does so with arguments that are
expected to allow the test to pass. The other does
not provide those arguments and is expected to fail.
The expected failure is intentional, to serve as
an ongoing reminder to try to 1) move towards having
strict typing in the codebase and to 2) not
add any additional errors/warnings in the meantime.
Separated requirements into runtime (requirements.txt)
and additional development (requirements-dev.txt).
Modified devcontainer and workflow to reflect the
change.
Add additional, non-default, Dockerfile target
which installs app into an ubuntu image. This is
useful for local development and enables other
operations (such as committing back to the repo)
to occur within the container.
There's no reason to have main directly construct
the URL. A different VCS (such as GitLab) could
have a different URL construction. Further, in the
case of LocalClient, there is no URL.
CustomLanguageTest was modifying a few environment
variables and not restoring them when complete. This
led to problems for later tests that expected those
variables to be at their default values.
Add a new method, get_issue_url(), which returns
the VCS-specific web URL for viewing an issue.
This results in moving GitHub-specific code from
main into GitHubClient, but with no change in behavior.
This is just a safe fallback for local testing.
If environment variables which activate the
creation of the GitHub client aren't set, then
the Local client is created. It acts on the most
recent commit of the repo in the working directory.
Minor edit to GitHubClient so that it raises an
EnvironmentError exception if INPUT_GITHUB_URL
environment variable is not defined. This allows
main to detect the error and fall back to trying
to use the Local client
Split main.py into modules to (mostly) isolate the
GitHub-specific code from the general TODO
detection and program logic.
This is both for readability/maintainability and
to prepare for potentially supporting other
version control systems (e.g. GitLab, BitBucket).
