Re: [Rd] Making headers self-contained for static analysis

2023-03-16 Thread Marc Schwartz via R-devel
Hi, There are a limited number of MIME file types that are accepted through the list server, with plain text being one. Even though a patch file should be plain text, it is possible that your mail client may not have set the correct MIME type for your patch file attachment. If so, that would

Re: [Rd] Making headers self-contained for static analysis

2023-03-16 Thread Lionel Henry via R-devel
People have let me know that the attachment didn't make it through. Do patches get filtered out? Please find it there: https://github.com/lionel-/r-svn/commit/e3de56798b1321a3fa8688a42bbb73d763b78024.patch I'm also happy to post it on the bugzilla if that makes sense. Best, Lionel On 3/16/23,

Re: [Rd] Making headers self-contained for static analysis

2023-03-16 Thread Ivan Krylov
Hello Lionel, Just letting you know off-list that the patch didn't make it through. Unfortunately, I don't remember the exact rules regarding attachments (does text/plain work?), but an external link is always an option, especially for large patches. -- Best regards, Ivan

[Rd] Making headers self-contained for static analysis

2023-03-16 Thread Lionel Henry via R-devel
Hello, I started using clangd to get better static analysis and code refactoring tooling with the R sources (using eglot-mode in Emacs, it just works once you've generated a `compile_commands.json` file with `bear make all`). I noticed that the static analyser can't understand several header