Bugzilla

:bgrins and I have a video call yesterday and the plan is for me to get up an MVP prototype of this in the next week.  Specifically, integration of code coverage data in to the HTML renderings of source files.  This will not initially include anything at the directory listing level[1] or anything in the search results[2].

My current UX plan is to expose a persistent (like hotkeys) <select> combo-box in the navigation panel for styling choices, which includes disabling the coverage, in order to let us land this sooner and iterate and get feedback on which presentations are least disruptive for normal code browsing and most helpful for reading the code while directly processing (which are different use cases).  I'll provide a few options and then ideally other (possibly new!) contributors can help provide additional options and/or refinements of the initial options.

1: It should be reasonably easy to make a subsequent directory level enhancement, although, as with the test info bug, the harder issue is likely the UX of the directory listing.
2: The cross-referencing process would need to be involved in processing coverage for these purposes, and for accuracy this might want some additional help from the indexers to most appropriately attribute the coverage to the right symbol on a line, either in the data emitted by the indexer or as policy information taught to crossref so that it can perform proper attribution.

:bgrins and I had a video call yesterday and the plan is for me to get up an MVP prototype of this in the next week.  Specifically, integration of code coverage data in to the HTML renderings of source files.  This will not initially include anything at the directory listing level[1] or anything in the search results[2].

My current UX plan is to expose a persistent (like hotkeys) <select> combo-box in the navigation panel for styling choices, which includes disabling the coverage, in order to let us land this sooner and iterate and get feedback on which presentations are least disruptive for normal code browsing and most helpful for reading the code while directly processing (which are different use cases).  I'll provide a few options and then ideally other (possibly new!) contributors can help provide additional options and/or refinements of the initial options.

1: It should be reasonably easy to make a subsequent directory level enhancement, although, as with the test info bug, the harder issue is likely the UX of the directory listing.
2: The cross-referencing process would need to be involved in processing coverage for these purposes, and for accuracy this might want some additional help from the indexers to most appropriately attribute the coverage to the right symbol on a line, either in the data emitted by the indexer or as policy information taught to crossref so that it can perform proper attribution.