Yuke, there is one more thing I'd like to improve. Take a look at this screenshot: https://screenshot.googleplex.com/vXZjR6U9Kr0.png

Alignment of percentages doesn't look good. I'd prefer seeing it in the following manner:


file1     |  89.13% (123 / 2323)    | 100.00% (55 / 55)    |   9.33% (14545 / 234234)
file_asda |   1.78% ( 23 / 4323)    |  32.31% (555 / 6555) |  67.89% (1545 / 2234)
fileXXX   | 100.00% (12323 / 12323) | 100.00% (555 / 555)  | 100.00% (12345 / 12345)

i.e. I want to align numbers to the left and pad them with spaces, so the percentage values would be easy to scan while looking at them. WDYT? I guess it should be done on the LLVM side, since those .html reports are generated by llvm-cov?

Yes, I completely agree, this would make the reports much easier to read.

https://reviews.llvm.org/D42093

Landed: https://reviews.llvm.org/rL323892

Should be available in Chromium after next clang roll ( issue 803661 ).

Do we have any other ideas regarding HTML report improvements?

Below are notes from Code Coverage BoF session:

------------------------------------------------------------------
2. HTML report quality

There seems to be widespread interest in improving the quality of coverage reports. We need volunteers to work on this and would love your help! Here are some desired features:

* Search and filtering for coverage summaries
* Collapsing parts of a coverage summary by subdirectory
* Automatically generating a top 10 list of code regions which need better coverage
* Searching via complex queries (e.g: 'give me uncovered regions in covered lines', or 'give me uncovered regions after a call')
* Generating coverage deltas between two profiles, and identifying coverage regressions in a patch/commit
* Simplified tracking of coverage trends over time

There is some consensus that this functionality should not be built on top of the existing llvm-cov C++ codebase. It might be better to develop these features in a language more amenable to rapid prototyping and interoperation with popular web application frameworks (perhaps Python). To facilitate this, llvm-cov gained support for exporting all of its data to JSON (see CoverageExporterJson.cpp). If you are interested in working on these features, I would be happy to work with you on design issues and on code review.
------------------------------------------------------------------

That all sounds interesting, but as said above, that should not be part of the report generated by C++ code, but rather something separate. I guess we'd better implement things like search, tracking trends, etc in a way which we could use as-is on the coverage dashboard web-site.

The only potential improvement I see is to generate report by directory, but we're doing that already in our script together with grouping by component. Thus, having by-directory grouping landed upstream is not a priority for us at the moment, though might be useful for community. Filed a separate issue 820543 to track that.

Closing this one. Feel free to re-open or file a new one if you have any other suggestions.

Thanks Max,

This is not high priority, but would it be interesting for us to create a Github open source project to develop those tools so that any project (even non-Chrome) can just import our tools and generate code coverage report that supports all those features such as Search, filtering?

Yes, I'm thinking the same way, Yuke. We might consider LLVM repo instead of GitHub, though GitHub seems more convenient to me. Having a unified Coverage Dashboard app providing all those features would be really awesome and useful for many projects, not only Chrome.