I've traced it into app/models/merge_request_diff.rb, in the unmerged_diffs function. It calls Gitlab::Git::Diff.between(repository, source_branch, target_branch), which is the one that returns a diff for the PDF. I haven't gone beyond that since I'm unsure of the state of using Grit versus Rugged, and I'm also unsure of each respective project's support for .gitattributes. I've been told Rugged supports them to the extent of libgit2, but that didn't mean a lot to me. I'm not at all familiar with Grit, Rugged, or libgit2. If there was support for .gitattributes, I'm still not sure what to expect back from Gitlab::Git::Diff.between()-- no diff? A special binary "this file changed" kind of diff?
If you can point me where to go next, I shall! Kyle On Wednesday, May 28, 2014 10:18:10 AM UTC-4, Sytse Sijbrandij wrote: > > Consider starting to digg into app/models/commit.rb > > On Tue, May 6, 2014 at 7:28 PM, Kyle Fazzari > <gitlab_ma...@status.e4ward.com <javascript:>> wrote: > > Hello all. > > > > In a few of my C++ projects being hosted on GitLab, I include a PDF > > generated by Doxygen to serve as a source code reference. These PDFs can > get > > quite lengthy, as anyone familiar with Doxygen will know. After a little > > while, I noticed that GitLab wouldn't show me my diffs on merge requests > > anymore. They were always "too big," and required me to click the > "download" > > link to see them. When I looked at them, a diff was also generated for > the > > PDF, and it was huge. So I thought, okay, that makes sense-- I'll just > add a > > .gitattributes file that removes the diff ability from PDF files for > this > > repo. I did that, and now when I look at the diff it no longer includes > the > > PDF's diff, but GitLab STILL says it's too large (even though the diff > is > > only a few lines). I narrowed it down to the reload_diffs function of > the > > MergeRequestDiff model, which still seems to count the lines from the > PDF > > diff thus putting it in the :overflow_diff_lines_limit state, regardless > of > > the .gitattributes file. > > > > Does anyone else have this problem? Is there a workaround? I'd like to > > continue including the generated PDF if I can. > > > > Thank you for any help. > > > > Kyle Fazzari > > > > -- > > You received this message because you are subscribed to the Google > Groups > > "GitLab" group. > > To unsubscribe from this group and stop receiving emails from it, send > an > > email to gitlabhq+u...@googlegroups.com <javascript:>. > > For more options, visit https://groups.google.com/d/optout. > -- You received this message because you are subscribed to the Google Groups "GitLab" group. To unsubscribe from this group and stop receiving emails from it, send an email to gitlabhq+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/gitlabhq/90b0401c-1eb1-47b5-b596-198360421388%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.