https://bugs.kde.org/show_bug.cgi?id=397563
--- Comment #2 from Mark <mpic...@gmail.com> --- (In reply to Nate Graham from comment #1) > Would the need for this go away if Kate was better about choosing a syntax > highlighting mode in the first place? What are the ones that it's not so > good at? I do not think it's possible for it to get good enough on this. Some of the file types I work with don't have a mime type and/or lack an extension, so I don't think detection is feasible. Even if 100% were detected, would we really want mime type X to be mapped to something that's _kinda_ like it, say python? That might not bother me or you short term, but I think it'd become a problem sooner or later. Types it isn't good at include: * Starlark (formerly skylark), file names BUILD, BUILD.bazel, *.BUILD, *.bzl. Quite similar to python; for this one I simply added the extensions to kate's python highlighter. * Jinja2 templates. The underlying content can be practically anything text-based: C++, bash, python, html, ansible, other yaml, etc etc. In some cases I prefer to use Jinja2 highlighting (making template items more visible) and in others I prefer to switch to highlighting of the underlying content. I've installed a set of highlighters for Jinja2 with common underlying formats, which is helpful - but a) this doesn't cover every underlying format, and b) even when the underlying format _is_ detected, I still desire to switch highlight modes at times. * ERB templates, golang's templates, etc. Similar issues as with Jinja2. * Various outdated (or proprietary) formats that I use from time to time. * Plus there's at least one extension that's used for two drastically different things; *.bas springs to mind though I don't recall what would use that other than some dialect of BASIC. -- You are receiving this mail because: You are watching all bug changes.
