Fixes for sass_interface (C Bindings) #181
Conversation
|
Hello First I want to thank you all, specially akhleung for working hard on the @extend implementation, for the ongoing work. I see you already fixed the "alpha opacity" with a much shorter patch. I just would like to ask about some feedback about this pull request. As I see it was not yet closed, but did not get any comments at all. There are a few bugs addressed in this pull request that havent yet been reported IMO. If you'd like I could also open issues for each one of these (like the "expression" bug, which is another IE specific thing)? We really would like to use libsass via our perl deployment! A quick feedback would be greatly appreciated! Kind regards |
|
Sorry, I think I overlooked this because I confused it with a different pull request (also regarding sass_interface) that was ultimately withdrawn. I'll go through this tomorrow morning and get it merged in. Thanks! |
|
+1, I want to have c function support from sass_interface |
|
I rebased my patches to the latest libsass version here: https://github.com/mgreter/libsass/tree/rebase-02. Added another patch that enables the use of sourcemaps with data not directly loaded by libsass. I also updated my branch for the perl Module CSS::Sass to handle source maps. |
|
@mgreter Failed? |
|
I have pushed all my changes which I use for CSS::Sass. The latests commits should solve #16. Just be aware that this has not been tested in the wild at all! I've included a small testcase in CSS::Sass to test the imports of indented sass files: |
Has obviously been renamed from C_Function_Data
Libsass crashes if implementer does not set a default option for image_path. Protect the sass_interface the same way as SassC does.
Re-introduced cookie for c functions (pointer to anything). Moved actual function registration right after the built in functions are registered.
They are still usefull so why disable it? Added clause to only produce it when enabled!
Convert all files with sass extension via sass2scss.
Also searches for files with sass extension.
|
Thanks, I'll take a look and get this merged in! |
|
Always believe that something wonderful is about to happen. |
|
I AM SO STOKED RIGHT NOW |
|
Would be really cool if this would be finally merged in! The pull request is now open for half a year and I tried to keep up with the master branch. I also added the official spec tests, which give the same results before and after adding my changes. It would allow us to release a new version for perl and I guess all others trying to use the c_interface will profit from it. I really would love to see it (at least partially) be merged in! Don't get me wrong, but It starts to feel a bit like working for the bin. |
|
I realize it's been a long time, but now that the sass2scss stuff is part of this pull request, I just want to review it carefully before I merge it. Please hold on for a few more days! |
|
Thanks for the feedback! Otherwise I'll bump it up again in a few weeks :) |
|
I could really use this merged so I can implement it as a Hexo (node static site generator) plugin ASAP. It's tricky using ruby in this environment so indented sass support would be fantastic! |
Fixes for sass_interface (C Bindings)
|
Credit goes to @mgreter; I just clicked the merge button! |
|
Woohoo! What great news and I really hope I didn't break anything for anyone else! If I did I already appologise for that. And please excuse my last "rude" comment about working for the bin, but yeah, I myself need some kick in the ass sometimes to get things done :) Anyway, IMO the only commit that I'm not too sure about is in 691e101. I've just now added a comment for the specific change which I'm not sure if it will behave correctly (it does not seem to break with any existing test). Thanks @akhleung for the merge! |
|
@mgreter don't forget to claim the bounty on this! https://www.bountysource.com/issues/1057273-sass-files-not-supported/claims |
Hi
I'm currently trying to get the latest Version of libsass to work with Perl CSS::Sass. Unfortunately I've hit a few obstacles on the way:
A) The current sass_interface does not seem to be in sync with the rest of the library
B) The sass_interface crashes if there is no image_path set. This could be solved by the perl implementation, but I guess it doesn't hurt to include this protection for all implementations. By the way, it was crashing with this message:
terminate called after throwing an instance of 'std::logic_error'
what(): basic_string::_S_construct null not valid
I hope this pull request is useful!
Best wishes
Marcel Greter