Coq now uses .mlg rather than .ml4 (since https://github.com/coq/coq/pull/8763), so we have to ignore its generated dependency files. The `native_compute_profile_*.data` files are generated by `Set NativeCompute Profiling` (see https://github.com/coq/coq/pull/950). Finally `.coq-native` is a directory that may be generated in any subdirectory, not only at top level, so we should not use absolute paths for it.
I added this .gitignore to a project that included a file named CoverageSearchModel.cs, and the file was wrongly ignored. This change fixes the incorrect use of the range operator on the Coverlet rules.
The problem here was two fold:
1. the folder "/target/" would be top-level of the repo only, it should be "target/" to properly exclude target folders anywhere in the repo
2. the default Rust/Cargo folder when compiling code is "debug/", which gets used perhaps more often that "target/", added that
After running `yarn set version berry` and `yarn install`, the file `.yarn/install-state.gz` is created.
The documentation at https://yarnpkg.com/advanced/qa#which-files-should-be-gitignored mentions that this file should be ignored:
> .yarn/install-state.tgz is an optimization file that you shouldn't have to ever commit. It simply stores the exact state of your project so that the next commands can boot without having to resolve your workspaces again.
The documentation has a minor error; the generated file is `.gz` instead of `.tgz` (source: https://github.com/yarnpkg/berry/pull/998/files#diff-23dd4c2e823c25186f1107e88e962032R201)
pip generated this folder for a few versions, as part of it's initial
implementation of PEP 517.
pip has not generated this folder for a few versions now, so it should
be OK to remove this from the standard gitignore file.
Visual Studio .Net used Win32/ as one of the default output directories for C and C++ projects. Later, when 64-bit support was added to the toolchain (circa 2005), x64/ was used. The Gitignore files include x64/, but not Win32/. The commit adds support for both Win32/ and x64/.
* Added Addressables.
Prevent automatically generated addressable files to end up in Git.
* Update .gitignore to exclude packed Addressables and Android auto-generated files.
After years of use I've come up with some improvements to the
`JENKINS_HOME.gitignore` example.
- Major performance improvement: On very large Jenkins installations that
have been running for more than one year, there tends to be many builds
(hundreds of thousands of builds). The `builds` directory of these
jobs contain millions of files which would cause Git to hang for
several minutes on simple commands like `git status` and longer for
committing changes. `strace` was used on Git to figure out the
performance impact and this proposed change includes the optimization.
I also added a clear comment explaining the line's purpose.
- There's an example for how to include Jenkins encryption keys, and
there's a disclaimer informing the user why they shouldn't but still
giving an example.
- Comments have been reworded and slightly reformatted to be a little
more clear.
Cython extension modules built with `gdb_debug=True` spit out debug symbols in the `cython_debug` directory at the top level of the project. The files in this directory contain hardcoded paths and are not shareable/meaningful across environments, so I think it makes sense to include them in a default Python .gitignore.
Since October 2019, Raku is the name of the language formerly known as
Perl 6. This reflects the change. It's the same language, so changes
are mostly cosmetic.
