.gitattributes 2.7 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869
  1. # ensure LF endings on all checkouts
  2. configure.ac crlf=input
  3. config.rpath crlf=input
  4. configure.host crlf=input
  5. mkinstalldirs crlf=input
  6. *.sh crlf=input
  7. *.sources crlf=input
  8. .gitattributes crlf=input
  9. *akefile* crlf=input
  10. # ensure native line endings on checkout
  11. *.c crlf
  12. *.h crlf
  13. *.cs crlf
  14. *.il crlf
  15. # don't do anything to line-endings. Let CRLFs go into the repo, and CRLF on checkout
  16. *.bat -crlf
  17. *.sln -crlf
  18. *.*proj* -crlf
  19. *.xml -crlf
  20. # CRLF Handling
  21. # -------------
  22. #
  23. # The ideal situation would be to do no EOL normalization. Each file
  24. # would have a default EOL, and tools on Windows and Linux would handle
  25. # both EOL formats.
  26. #
  27. # We're not in the ideal world. A popular editor on Windows (possibly
  28. # Visual Studio) silently introduces EOL corruption -- it displays an
  29. # LF-file normally, but any newly added lines have CRLF. On Linux,
  30. # Emacs and versions of VI handle LF-files and CRLF-files properly.
  31. # However, emacs doesn't like files with both LF and CRLF EOLs. Editing
  32. # the file without additional action will increase the EOL corruption
  33. # in the file.
  34. #
  35. # Another vector for mixed EOLs is scripts. We mostly don't have scripts
  36. # that add new lines -- so we rarely see this. However, one major event
  37. # in the tree was the addition of copyright headers using a script. That
  38. # script introduced EOL corruption.
  39. #
  40. # Any automated EOL normalization of files already in the repository will
  41. # cause difficulties in traversing histories, assigning blame, etc. So, we
  42. # don't want to change what's in the repository significantly, even if it
  43. # causes trouble.
  44. #
  45. # What we do now:
  46. #
  47. # a) we ensure that there's no further corruption of LF-files. So, we use
  48. # git's 'crlf' attribute on those files to ensure that things are fine
  49. # when we work on Windows. We could use 'crlf=input', but it doesn't buy
  50. # us much -- we might as well be working with consistent EOLs for files in
  51. # working directories as well as in the repository
  52. #
  53. # b) if the file already of CRLFs, we don't do any normalization. We use '-crlf'
  54. # so that git doesn't do any EOL-conversion of the file. As I said, this
  55. # is mostly harmless on Linux. We can't mark these files as 'crlf' or use
  56. # the new (git 1.7.2) 'eol=crlf' attribute, since it changes the contents
  57. # _inside_ the repository [1], and hence makes history traversal annoying.
  58. # So, we live with occasional EOL corruption.
  59. #
  60. # c) We can handle mixed-EOL files on a case-by-case basis, converting them to
  61. # LF- or CRLF-files based on which causes fewer lines to change
  62. #
  63. # d) We try to ensure no further headaches, by declaring EOL normalization on
  64. # code files, and Unix-flavoured files, like shell-scripts, makefiles, etc.
  65. #
  66. # [1] GIT use LFs as the normalized internal representation.