Opened 11 years ago

Closed 10 years ago

Last modified 10 years ago

#8898 closed defect (fixed)

some files in sage-4.4.{0,1} have dos line ending instead of a unix line ending

Reported by: fbissey Owned by: jason
Priority: trivial Milestone: sage-4.6.1
Component: misc Keywords:
Cc: fredrik.johansson Merged in:
Authors: Minh Van Nguyen, Jeroen Demeyer Reviewers: François Bissey
Report Upstream: N/A Work issues:
Branch: Commit:
Dependencies: Stopgaps:

Description (last modified by jdemeyer)

Various files in the sage spkg have dos line ending rather than unix ones.

I found about this while trying to build sage with python-2.6.5 which absolutely refused to parse these files as is. Not sure how to submit a patch for line endings.

Execute from SAGE_ROOT/devel/sage:

  1. 8898.sh

Attachments (3)

trac_8898-unix-endlines.2.patch (105.2 KB) - added by fbissey 10 years ago.
update of the patch based on sage-4.6
trac-8898_unix-endlines.patch (107.0 KB) - added by mvngu 10 years ago.
8898.sh (460 bytes) - added by jdemeyer 10 years ago.
Script to do the changes (to be executed in SAGE_ROOT/devel/sage)

Download all attachments as: .zip

Change History (21)

comment:1 Changed 11 years ago by mvngu

Some files under sage/logic/ also have DOS line ending.

comment:2 Changed 11 years ago by fbissey

just checked sage/logic/booleval.py is actually in "mac format" in sage-4.4 I will check sage-4.4.1 later. But those are pure python files, they may be ok but cython with python-2.6.5 refused to deal with the other 4.

comment:3 Changed 10 years ago by AlexGhitza

  • Component changed from algebra to misc
  • Owner changed from AlexGhitza to jason

comment:4 Changed 10 years ago by mvngu

  • Authors set to Minh Van Nguyen
  • Cc fredrik.johansson added
  • Description modified (diff)
  • Priority changed from minor to trivial
  • Status changed from new to needs_review

The attached patch converts the following files to use Unix line endings:

  • sage/libs/mpmath/ext_impl.pxd
  • sage/libs/mpmath/ext_main.pyx
  • sage/libs/mpmath/ext_main.pxd
  • sage/libs/mpmath/ext_libmp.pyx
  • sage/logic/booleval.py

I used the Perl script at

http://www.obviously.com/tech_tips/dos2unix.html

to convert to Unix end lines. Fredrik Johansson is a main developer of mpmath. I have CC'd him so he is aware of this Unix line endings issue.

comment:5 Changed 10 years ago by fbissey

  • Status changed from needs_review to needs_work

sage/libs/mpmath/ext_impl.pxd at least as been changed since this patch has been posted. It may need rebasing for all 4 files.

Changed 10 years ago by fbissey

update of the patch based on sage-4.6

comment:6 Changed 10 years ago by fbissey

  • Status changed from needs_work to needs_review

OK - so I updated the patch (but could not delete the old one, I don't have the right to do it).

This fairly trivial patch now needs a review.

comment:7 Changed 10 years ago by cschwan

I have tested the patch with python-2.6.6 and sage-on-gentoo - everything fine here.

Changed 10 years ago by mvngu

comment:8 Changed 10 years ago by mvngu

  • Description modified (diff)

Here are some problems with trac_8898-unix-endlines.2.patch:

  • It fails to apply on Sage 4.6.1.alpha0; I got the following failure:
    [mvngu@sage sage-main]$ pwd
    /dev/shm/mvngu/sage-4.6.1.alpha0/devel/sage-main
    [mvngu@sage sage-main]$ hg qimport http://trac.sagemath.org/sage_trac/raw-attachment/ticket/8898/trac_8898-unix-endlines.2.patch && hg qpush 
    adding trac_8898-unix-endlines.2.patch to series file
    applying trac_8898-unix-endlines.2.patch
    patching file sage/libs/mpmath/ext_main.pyx
    Hunk #1 FAILED at 0
    1 out of 1 hunks FAILED -- saving rejects to file sage/libs/mpmath/ext_main.pyx.rej
    patch failed, unable to continue (try -v)
    patch failed, rejects left in working dir
    errors during apply, please fix and refresh trac_8898-unix-endlines.2.patch
    

My rebased patch should take care of the above issues for Sage 4.6.1.alpha0. See the ticket description for which patch to apply.

comment:9 Changed 10 years ago by fbissey

Sorry I missed sage/logic/booleval.py somehow. In any case this one is no bother and is only included for consistency. It doesn't prevent building the way the other do and it is usable. I also didn't see there was a change affecting this in 4.6.1.alpha0. I will test it shortly and report.

comment:10 Changed 10 years ago by fbissey

  • Status changed from needs_review to positive_review

4.6.1.alpha0 took me for a little ride. Anyway, Minh's new patch applies cleanly to 4.6.1.alpha0 and it compiles cleanly with a python-2.6.5 install as expected. sage starts without problems. For safety I also did a run of sage -t --long in the sage/libs/mpmath/ and sage/logic/ and everything ran ok.

I am putting this back to positive review, hopefully no one messed up with those files in alpha1 so it will apply cleanly there as well.

Changed 10 years ago by jdemeyer

Script to do the changes (to be executed in SAGE_ROOT/devel/sage)

comment:11 Changed 10 years ago by jdemeyer

  • Status changed from positive_review to needs_work

comment:12 follow-up: Changed 10 years ago by jdemeyer

  • Description modified (diff)
  • Status changed from needs_work to needs_review

Any complaints if I execute the script 8898.sh instead of applying the patch? The script also fixes some more files.

comment:13 in reply to: ↑ 12 Changed 10 years ago by fbissey

Replying to jdemeyer:

Any complaints if I execute the script 8898.sh instead of applying the patch? The script also fixes some more files.

No complaints here. As I initially said my main concern are the cython files. I'll be happy if a fix goes in script or patch.

comment:14 Changed 10 years ago by fbissey

I am building 4.6.1.alpha2 which includes the fix. cython parsed everything without trouble using python-2.6.5 and compilation is now underway. So it looks good to me.

comment:15 follow-up: Changed 10 years ago by jdemeyer

  • Authors changed from Minh Van Nguyen to Minh Van Nguyen, Jeroen Demeyer
  • Reviewers set to François Bissey
  • Status changed from needs_review to positive_review

François, I am interpreting your post as a positive review, okay?

comment:16 in reply to: ↑ 15 Changed 10 years ago by fbissey

Replying to jdemeyer:

François, I am interpreting your post as a positive review, okay?

OK, the build finished successfully, so yes positive review.

comment:17 Changed 10 years ago by jdemeyer

  • Merged in set to sage-4.6.1.alpha2
  • Resolution set to fixed
  • Status changed from positive_review to closed

comment:18 Changed 10 years ago by jdemeyer

  • Merged in sage-4.6.1.alpha2 deleted
Note: See TracTickets for help on using tickets.