Modularization #225
Modularization #225
Conversation
This required the options.show_branch flag to be extracted. It was previously used as a global variable to control the mode of `CoverageData.coverage()`.
the `CoverageData.summary()` method was removed and put into `print_text_report()` as it is mostly presentational.
Codecov Report
@@ Coverage Diff @@
## master #225 +/- ##
==========================================
+ Coverage 83.87% 84.17% +0.29%
==========================================
Files 3 11 +8
Lines 1228 1251 +23
Branches 248 248
==========================================
+ Hits 1030 1053 +23
Misses 138 138
Partials 60 60
Continue to review full report at Codecov.
|
gcovr/__main__.py
Outdated
| xrange = range | ||
| from .gcov import get_datafiles, process_existing_gcov_file, process_datafile | ||
| from .utils import get_global_stats, build_filter | ||
| from .version import __version__, version_str # noqa: F401 unused __version__ |
There was a problem hiding this comment.
Why import __version__ that requires # noqa: F401 unused __version__ ?
There was a problem hiding this comment.
🤔 hmm, good question. I just cargo-culted the existing code here because I'm not 100% sure about the Python conventions in this case. I see two alternatives:
- import the version in the
__init__so that we have agcovr.__version__attribute. - leave the version in the
gcovr.versionmodule until we offer an API that would profit from a version attribute.
What do you think would be best?
There was a problem hiding this comment.
On second thought, I'll keep __version__ but remove the version_str() function which no longer adds any value over accessing the __version__ directly.
This function previously included the SVN revision number in the output, but that is no longer useful since gcovr is maintained in a Git repository. So instead, we can use `__version__` directly.
Modularization was implemented in gcovr#225
This PR splits the gcovr script into multiple modules, as discussed in #215.