From: "Torsten Bögershausen" <tboegi@web.de>
To: "Toralf Förster" <toralf.foerster@gmx.de>
Cc: Andrew Ardill <andrew.ardill@gmail.com>,
"git@vger.kernel.org" <git@vger.kernel.org>
Subject: Re: problem with BOINC repository and CR/LF
Date: Tue, 18 Dec 2012 13:15:30 +0100 [thread overview]
Message-ID: <50D05E62.7090605@web.de> (raw)
In-Reply-To: <50D03D80.3090005@gmx.de>
On 18.12.12 10:55, Toralf Förster wrote:
> On 12/18/2012 02:56 AM, Andrew Ardill wrote:
>> On 18 December 2012 03:01, Toralf Förster <toralf.foerster@gmx.de> wrote:
>>> On 12/17/2012 12:38 PM, Andrew Ardill wrote:
>>>> On 17 December 2012 21:23, Toralf Förster <toralf.foerster@gmx.de> wrote:
>>>>> Hello,
>>>>>
>>>>> I'm faced with this situation :
>>>>> http://lists.ssl.berkeley.edu/mailman/private/boinc_alpha/2012-December/017371.html
>>>>> and even a "git stash" doesn't help.
>>>>
>>>> Hi Toralf,
>>>>
>>>> That list is private and not visible without an account. Can you
>>>> transcribe the relevant parts?
>>>>
>>>> Regards,
>>>>
>>>> Andrew Ardill
>>>>
>>> Oh of course :
>>>
>>>
>>> On 12/17/2012 12:03 AM, Gianfranco Costamagna wrote:
>>>> So if you have further issues with boinc feel free to look in our debian
>>>> git and feel free to download appropriate patches :-)
>>>>
>>>> Gianfranco
>>> thx
>>>
>>> Currently I'm struggling with a git problem of the boinc repository
>>> itself and b/c I'm using git for the linux kernel tree w/o any problems
>>> since eons /me wonders whether this is a BOINC-repository specific problem :
>>>
>>>
>>> After doing the following sequence with git 1.8.0.2 :
>>>
>>> $> git clone git://boinc.berkeley.edu/boinc.git
>>> $> cd boinc
>>> $> git checkout client_release_7.0.39
>>> $> git checkout master
>>> (sometimes I've to repeat this :
>>> $> git checkout client_release_7.0.39
>>> $> git checkout master
>>> )
>>> I'm faced with this situation :
>>>
>>> $ git status
>>> # On branch master
>>> # Changes not staged for commit:
>>> # (use "git add <file>..." to update what will be committed)
>>> # (use "git checkout -- <file>..." to discard changes in working
>>> directory)
>>> #
>>> # modified: clientgui/AsyncRPC.cpp
>>> # modified: clientgui/sg_BoincSimpleFrame.cpp
>>> #
>>> no changes added to commit (use "git add" and/or "git commit -a")
>>>
>>> (sometimes only clientgui/sg_BoincSimpleFrame.cpp is mentioned)
>>>
>>> Now these commands
>>>
>>> $ git checkout -- clientgui/AsyncRPC.cpp
>>> $ git checkout -- clientgui/sg_BoincSimpleFrame.cpp
>>>
>>> doesn't help - the status is still the same (and ofc now I'm no longer
>>> allowed to make a "git checkout" - due to un-commited changes).
>>>
>>> Now I'm wondering where to start to investigate this issue ...
>>
>> Hi Toralf,
>>
>> That does look like a weird issue. What operating system are you on?
>
> I'm running a stable Gentoo Linux x86, 32bit with gcc 4.6.3 and current
> stable kernel 3.6.1x and 3.7.1, file system is ext4 at an external USB
> 2.0 drive.
>
> FWIW from the boinc maintainer I know that all tags till 7.0.3X are
> imported from svn.
>
> I upgraded to git 1.8.0.2 from 1.7.8.6 - situation is the same. The
> emerge gave :
>
> failed test(s): t3600 t7508
>
> fixed 0
> success 8342
> failed 8
> broken 56
> total 8528
>
> Ok, now answering your other questions:
>
>
> $> git stash
> warning: CRLF will be replaced by LF in clientgui/AsyncRPC.cpp.
> The file will have its original line endings in your working directory.
> warning: CRLF will be replaced by LF in clientgui/sg_BoincSimpleFrame.cpp.
> The file will have its original line endings in your working directory.
> warning: CRLF will be replaced by LF in clientgui/AsyncRPC.cpp.
> The file will have its original line endings in your working directory.
> warning: CRLF will be replaced by LF in clientgui/sg_BoincSimpleFrame.cpp.
> The file will have its original line endings in your working directory.
> Saved working directory and index state WIP on master: 4a296dc - client
> simulator: fix build errors
> HEAD is now at 4a296dc - client simulator: fix build errors
>
> After that the situation is unchanged.
>
>> What happens if you do a hard reset to the branch?
>
> $> git reset --hard HEAD~1
>
> not better.
>
>
>> What is the ouptut of git diff --cached ?
>
> The output is empty but "git status" shows still modified files.
>
>
>
> FWIW there's a related issue I'm wondering about which might help:
>
> $> git clone git://boinc.berkeley.edu/boinc.git
> $> tar -cpf boinc.tar boinc/
> $> rm -rf boinc/
> $> tar -xpf boinc.tar
> $> cd boinc/
> $> git status
> # On branch master
> # Changes not staged for commit:
> # (use "git add <file>..." to update what will be committed)
> # (use "git checkout -- <file>..." to discard changes in working
> directory)
> #
> # modified: client/win/boinc_log.h
> # modified: client/win/boinc_log.rc
> # modified: clientctrl/boincsvcctrl.cpp
> # modified: clientctrl/boincsvcctrl.h
> # modified: clientctrl/boincsvcctrl.rc
> # modified: clientgui/AsyncRPC.cpp
> # modified: clientgui/DlgEventLog.cpp
> # modified: clientgui/DlgEventLog.h
> # modified: clientgui/DlgEventLogListCtrl.cpp
> # modified: clientgui/DlgEventLogListCtrl.h
> # modified: clientgui/DlgExitMessage.h
> # modified: clientgui/DlgItemProperties.h
> # modified: clientgui/TermsOfUsePage.cpp
> # modified: clientgui/TermsOfUsePage.h
> # modified: clientgui/ViewNotices.cpp
> # modified: clientgui/ViewNotices.h
> # modified: clientgui/sg_BoincSimpleFrame.cpp
> # modified: clientscr/boinc_ss_opengl.h
> # modified: clientscr/boinc_ss_opengl.rc
> # modified: clientscr/screensaver.cpp
> # modified: clienttray/boinc_tray.h
> # modified: clienttray/boinc_tray.rc
> # modified: clienttray/tray_win.cpp
> # modified: clienttray/tray_win.h
> # modified: coprocs/NVIDIA/include/nvapi.h
> #
> no changes added to commit (use "git add" and/or "git commit -a")
> $> git diff --cached
> $>
>
>
>
> Meaning, w/o any other interaction a tar'ed archive has modified files -
> and the diff is empty...
>
Hej,
I could re-produce the problem here:
git version 1.8.0.197.g5a90748
Mac OS X (that what I had at hands fastest)
After doing
git checkout 5db4a05b5c8f9c420fc418727cafbb58e6051f1e
(same as master ?)
We see that
clientgui/AsyncRPC.cpp
is full of CRLF and as it seems only CRLF, no single LF.
The file is classified as text:
$git check-attr text clientgui/AsyncRPC.cpp
$clientgui/AsyncRPC.cpp: text: set
(And we can see this in .gitattributes as well)
(And there are more files affected, but I will only look at one of them)
If we remove the text attribute like this:
$mv .gitattributes .gitattributes.sav
we see
$git status
deleted: .gitattributes
# modified: clientgui/sg_BoincSimpleFrame.cpp
If we dig into the file:
$git ls-files -s clientgui/AsyncRPC.cpp
100644 6832333ad133181986ada54fe0229b45a30c614a 0 clientgui/AsyncRPC.cpp
We see that it is recorded under 6832333ad1
And if we look into it:
$git show 6832333 | od -c
[snip]
we can see that the file has CRLF in the repo.
So my conclusion is:
The file has CRLF in the repo, but should have LF.
This is not a good thing, and the files need to be normalized.
A very good instruction how to do this, is found here:
http://kernel.org/pub/software/scm/git/docs/gitattributes.html
(You may want to search for "End-of-line conversion" or "core.autocrlf")
HTH
/Torsten
next prev parent reply other threads:[~2012-12-18 12:15 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-12-17 10:23 problem with BOINC repository and CR/LF Toralf Förster
2012-12-17 11:38 ` Andrew Ardill
2012-12-17 16:01 ` Toralf Förster
2012-12-18 1:56 ` Andrew Ardill
2012-12-18 9:55 ` Toralf Förster
2012-12-18 9:58 ` Toralf Förster
2012-12-18 12:15 ` Torsten Bögershausen [this message]
2012-12-18 16:41 ` Jeff King
2012-12-19 10:44 ` Toralf Förster
2012-12-22 12:31 ` Toralf Förster
2012-12-22 18:54 ` Dmitry Potapov
2012-12-22 19:17 ` Junio C Hamano
2012-12-19 10:43 ` Toralf Förster
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=50D05E62.7090605@web.de \
--to=tboegi@web.de \
--cc=andrew.ardill@gmail.com \
--cc=git@vger.kernel.org \
--cc=toralf.foerster@gmx.de \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.