On Friday, 1 April 2016 10:01:08 UTC+2, Andreas wrote: > On Thursday, 31 March 2016 11:08:42 UTC+2, Ken Takata wrote: > > Hi Andreas, > > > > Thank you for trying the patches! > > > > 2016/3/30 Wed 21:22:05 UTC+9 Andreas wrote: > > > On Tuesday, 29 March 2016 03:30:40 UTC+2, Ken Takata wrote: > > > > Hi Bram, > > > > > > > > 2016/3/29 Tue 3:11:03 UTC+9 Bram Moolenaar wrote: > > > > > Ken Takata wrote: > > > > > > > > > > > 2016/2/1 Mon 22:34:43 UTC+9 Ken Takata wrote: > > > > > > > I think that all the style changes have completed. > > > > > > > Here are the updated large file patches for 7.4.1229. > > > > > > > > > > > > I have updated the large file patches for 7.4.1665. > > > > > > Please check the attached five patches. > > > > > > > > > > > > Bram, I'm afraid that you are mistaking some of my patches. > > > > > > Attached patches are related to the following items from the > > > > > > todo.txt: > > > > > > > > > > > > L170-173 > > > > > > > Win32: patch to use 64 bit stat() if possible. (Ken Takata, 2014 > > > > > > > May 12) > > > > > > > More tests May 14. Update May 29. Update Aug 10. > > > > > > > Now part of large file patches. (Ken Takata, 2016 Feb 1) > > > > > > > Two patches now? New update Feb 24. > > > > > > > > > > > > L1929-1931 > > > > > > > Win64: Seek error in swap file for a very big file (3 Gbyte). > > > > > > > Check storing > > > > > > > pointer in long and seek offset in 64 bit var. > > > > > > > Patches from Ken Takata might help (2014 Apr 17) > > > > > > > > > > > > > > > > > > The large file patches are also available from the following URLs: > > > > > > https://bitbucket.org/k_takata/vim-ktakata-mq/src/tip/0001_support-largefiles-on-windows.patch > > > > > > https://bitbucket.org/k_takata/vim-ktakata-mq/src/tip/0002_fix-off_T-display.patch > > > > > > https://bitbucket.org/k_takata/vim-ktakata-mq/src/tip/0003_use-stat_T.patch > > > > > > https://bitbucket.org/k_takata/vim-ktakata-mq/src/tip/0004_add-stat-test.patch > > > > > > https://bitbucket.org/k_takata/vim-ktakata-mq/src/tip/0005_test_largefile.patch > > > > > > (I will keep them up to date in the same URLs.) > > > > > > > > > > So those patches are supposed to be included in order and sent out as > > > > > separate patches? Or could I just include all of these and send out > > > > > one > > > > > big patch? > > > > > > > > I separate them mainly for readability. They can be merged as one > > > > patch. > > > > > > > > More detail: > > > > > > > > * 0001: This is a core part of them. This should fix the reported two > > > > problems: > > > > > > https://groups.google.com/d/topic/vim_dev/8epHlRuIAHc/discussion > > > > > > https://groups.google.com/d/topic/vim_use/RLM8Vqa_aME/discussion > > > > * 0002: Only fixes one display problem. This can be considered as a > > > > part of > > > > 0001. > > > > * 0003: This fixes another problem: 32-bit stat() returns an error for > > > > a file > > > > larger than 4 GiB. :checktime, getfperm() or other stat > > > > functions > > > > might fail for large files. > > > > * 0004: Adds basic tests for stat functions. This can be considered as > > > > a part > > > > of 0003. > > > > * 0005: Adds tests for 0001 and 0003. This is not added as a normal test > > > > target. A special target would be needed as you said: > > > > > For the test, we can add a "torture" target. > > > > > > > > > > > > > > I think these patches are rather important, as some people reported > > > > > > that very > > > > > > large file can be truncated or Vim can crash: > > > > > > https://groups.google.com/d/topic/vim_dev/8epHlRuIAHc/discussion > > > > > > https://groups.google.com/d/topic/vim_use/RLM8Vqa_aME/discussion > > > > > > > > > > > > Note: these patches doesn't fix some display problems (e.g. g<C-G>) > > > > > > because > > > > > > 64-bit integers are needed. To fix them, the num64 patch is needed. > > > > > > https://groups.google.com/d/topic/vim_dev/p8Fl_vJDGy8/discussion > > > > > > (Additionally, the 0002_fix-off_T-display.patch can be simpler when > > > > > > the num64 > > > > > > patch is included.) > > > > > > > > > > I hope to get back to including patches soon. Just a bit more cleanup > > > > > for new features to do. > > > > > > > > Regards, > > > > Ken Takata > > > > > > I tried these pathces on version 1685 on windows 10 using Visual Studio > > > 2015. > > > It didn't manage to patch src/proto/ fileio.pro, memline.pro, > > > os_mswin.pro and src/testdir/Make_all.mak. > > > Got some warnings of the type "warning C4133: 'function': incompatible > > > types - from 'stat_T *' to 'stat *'" from the compiler. > > > > I don't know why patching failed, but it might be the cause of the warnings. > > The patches apply cleanly in my environment (7.4.1689), and I don't see such > > warnings (with VC2010, VC2015 and MinGW-w64). > > > > > > > Tried the resulting gvim.exe on 8GB file and it worked nicely. The only > > > problem I found was that the number of charatcers read is 32-bit and also > > > the byte offset if you use G + CTRL-G. > > > > Yes, this is a known problem. If you apply the following patch, the number > > should be shown properly: > > https://groups.google.com/d/msg/vim_dev/p8Fl_vJDGy8/CKSZ1ytQHAAJ > > > > Regards, > > Ken Takata > > Is is probably because I'm not used to downloading patches. I apply them by > just typing "patch.exe < <patch file>" in the src directory. I guess I should > use some Git function instead. > > Regards > Andreas
I remember now why. When I try to do the patch using TortoiseGit I get the message: git.exe am --3way --ignore-space-change --keep-cr "C:\files\0001_support-largefiles-on-windows(1).patch" Patch does not have a valid e-mail address. Fail Regards Andreas -- -- You received this message from the "vim_dev" maillist. Do not top-post! Type your reply below the text you are replying to. For more information, visit http://www.vim.org/maillist.php --- You received this message because you are subscribed to the Google Groups "vim_dev" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
