RE: [PATCH] D39737: [MC] Fix regression tests on Windows when git “core.autocrlf” is set to true

Zhen Cao via llvm-commits llvm-commits at lists.llvm.org
Thu Nov 9 13:59:23 PST 2017


Thanks. But do you think it is better to require the user to run “git config --local core.autocrlf false” rather than to use .gitattributes to automatically set the repository?


From: Zachary Turner [mailto:zturner at google.com]
Sent: Thursday, November 9, 2017 4:52 PM
To: Zhen Cao <zhen.cao at autodesk.com>
Cc: reviews+D39737+public+a17c03644213de66 at reviews.llvm.org; Eli Friedman via Phabricator <reviews at reviews.llvm.org>; rnk at google.com; grosbach at apple.com; bigcheesegs at gmail.com; javed.absar at arm.com; llvm-commits at lists.llvm.org
Subject: Re: [PATCH] D39737: [MC] Fix regression tests on Windows when git “core.autocrlf” is set to true

You can set it locally for a specific repo.

git config --local core.autocrlf false

This won't affect other repositories.

On Thu, Nov 9, 2017 at 1:51 PM Zhen Cao <zhen.cao at autodesk.com<mailto:zhen.cao at autodesk.com>> wrote:
Some people use git for other projects as well. They may not want to set “core.autocrlf=false” for all the repos. On the other hand, if some version control system settings have to be required to pass all the regression tests, I think it is better to have documentation on it.

From: Zachary Turner [mailto:zturner at google.com<mailto:zturner at google.com>]
Sent: Thursday, November 9, 2017 4:31 PM
To: Zhen Cao <zhen.cao at autodesk.com<mailto:zhen.cao at autodesk.com>>
Cc: reviews+D39737+public+a17c03644213de66 at reviews.llvm.org<mailto:reviews%2BD39737%2Bpublic%2Ba17c03644213de66 at reviews.llvm.org>; Eli Friedman via Phabricator <reviews at reviews.llvm.org<mailto:reviews at reviews.llvm.org>>; rnk at google.com<mailto:rnk at google.com>; grosbach at apple.com<mailto:grosbach at apple.com>; bigcheesegs at gmail.com<mailto:bigcheesegs at gmail.com>; javed.absar at arm.com<mailto:javed.absar at arm.com>; llvm-commits at lists.llvm.org<mailto:llvm-commits at lists.llvm.org>

Subject: Re: [PATCH] D39737: [MC] Fix regression tests on Windows when git “core.autocrlf” is set to true

I guess, going back to my original question, can we just require core.autocrlf=false and call it a day?

I’m not sure why you’re having trouble checking out with svn, but what I’ve always done is use a monorepo and then you’ll have an svn checkout locally, and you can use Tortoise Svn to modify the props in a gui
On Thu, Nov 9, 2017 at 1:25 PM Zhen Cao <zhen.cao at autodesk.com<mailto:zhen.cao at autodesk.com>> wrote:
What about I add the following to the “Getting Started” section of the page (http://llvm.org/docs/GettingStartedVS.html<https://urldefense.proofpoint.com/v2/url?u=http-3A__llvm.org_docs_GettingStartedVS.html&d=DwMFaQ&c=76Q6Tcqc-t2x0ciWn7KFdCiqt6IQ7a_IF9uzNzd_2pA&r=5bO7zfHLitDTGEV53jsIoKY9YZM-dEXQXjvwmt4K1Xs&m=UW85ZemgrJlItSvSy9cLehlsQ2aLniu3SryzULBLG5U&s=kgkrRbqp-j4XQ1u19DS9ak2mlKFctKTu4d1ySfRv3HU&e=>)?

Note: some regression tests require Linux style line ending (\n). To pass all regression tests, please add the following two lines to “C:\Users\<username>\AppData\Roaming\Subversion\config”:

enable-auto-props = yes
* = svn:mime-type=application/octet-stream

From: Zachary Turner [mailto:zturner at google.com<mailto:zturner at google.com>]
Sent: Wednesday, November 8, 2017 7:39 PM
To: reviews+D39737+public+a17c03644213de66 at reviews.llvm.org<mailto:reviews%2BD39737%2Bpublic%2Ba17c03644213de66 at reviews.llvm.org>; Eli Friedman via Phabricator <reviews at reviews.llvm.org<mailto:reviews at reviews.llvm.org>>
Cc: Zhen Cao <zhen.cao at autodesk.com<mailto:zhen.cao at autodesk.com>>; rnk at google.com<mailto:rnk at google.com>; grosbach at apple.com<mailto:grosbach at apple.com>; bigcheesegs at gmail.com<mailto:bigcheesegs at gmail.com>; javed.absar at arm.com<mailto:javed.absar at arm.com>; llvm-commits at lists.llvm.org<mailto:llvm-commits at lists.llvm.org>
Subject: Re: [PATCH] D39737: [MC] Fix regression tests on Windows when git “core.autocrlf” is set to true

By “your git or svn client”, you mean everyone right (ie what I said earlier)? Because it’s not sufficient for Zhen just to fix this locally before pushing a patch
On Wed, Nov 8, 2017 at 4:09 PM Eli Friedman via Phabricator via llvm-commits <llvm-commits at lists.llvm.org<mailto:llvm-commits at lists.llvm.org>> wrote:
efriedma added a comment.

With svn, this is controlled by svn:eol-style and svn:mime-type.

That said, we don't have to worry about any of this if you just configure your git or svn client not to mess with line endings.


https://reviews.llvm.org/D39737<https://urldefense.proofpoint.com/v2/url?u=https-3A__reviews.llvm.org_D39737&d=DwMFaQ&c=76Q6Tcqc-t2x0ciWn7KFdCiqt6IQ7a_IF9uzNzd_2pA&r=5bO7zfHLitDTGEV53jsIoKY9YZM-dEXQXjvwmt4K1Xs&m=QiVMrF__xcFR41pB7Dx8zjHdggqK6rIFVvSD2-LR9I0&s=vHN0zUMQzdfO93b_VDKUuj20gOQSoAJjnZrQmzgP7kw&e=>



_______________________________________________
llvm-commits mailing list
llvm-commits at lists.llvm.org<mailto:llvm-commits at lists.llvm.org>
http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-commits<https://urldefense.proofpoint.com/v2/url?u=http-3A__lists.llvm.org_cgi-2Dbin_mailman_listinfo_llvm-2Dcommits&d=DwMFaQ&c=76Q6Tcqc-t2x0ciWn7KFdCiqt6IQ7a_IF9uzNzd_2pA&r=5bO7zfHLitDTGEV53jsIoKY9YZM-dEXQXjvwmt4K1Xs&m=QiVMrF__xcFR41pB7Dx8zjHdggqK6rIFVvSD2-LR9I0&s=dcj06dTL4nPn9WulSgRnOkVnF4IlLkhZLJPAlwnm49k&e=>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20171109/dc7c0a92/attachment-0001.html>


More information about the llvm-commits mailing list