.gitattributes 2.6 KB

12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758596061626364656667
  1. # ensure LF endings on all checkouts
  2. configure.ac crlf=input
  3. # ensure native line endings on checkout
  4. *.c crlf
  5. *.h crlf
  6. *.cs crlf
  7. *.sh crlf
  8. *.il crlf
  9. .gitattributes crlf
  10. *akefile* crlf
  11. *.sources crlf
  12. # don't do anything to line-endings. Let CRLFs go into the repo, and CRLF on checkout
  13. *.bat -crlf
  14. *.sln -crlf
  15. *.*proj* -crlf
  16. *.xml -crlf
  17. # CRLF Handling
  18. # -------------
  19. #
  20. # The ideal situation would be to do no EOL normalization. Each file
  21. # would have a default EOL, and tools on Windows and Linux would handle
  22. # both EOL formats.
  23. #
  24. # We're not in the ideal world. A popular editor on Windows (possibly
  25. # Visual Studio) silently introduces EOL corruption -- it displays an
  26. # LF-file normally, but any newly added lines have CRLF. On Linux,
  27. # Emacs and versions of VI handle LF-files and CRLF-files properly.
  28. # However, emacs doesn't like files with both LF and CRLF EOLs. Editing
  29. # the file without additional action will increase the EOL corruption
  30. # in the file.
  31. #
  32. # Another vector for mixed EOLs is scripts. We mostly don't have scripts
  33. # that add new lines -- so we rarely see this. However, one major event
  34. # in the tree was the addition of copyright headers using a script. That
  35. # script introduced EOL corruption.
  36. #
  37. # Any automated EOL normalization of files already in the repository will
  38. # cause difficulties in traversing histories, assigning blame, etc. So, we
  39. # don't want to change what's in the repository significantly, even if it
  40. # causes trouble.
  41. #
  42. # What we do now:
  43. #
  44. # a) we ensure that there's no further corruption of LF-files. So, we use
  45. # git's 'crlf' attribute on those files to ensure that things are fine
  46. # when we work on Windows. We could use 'crlf=input', but it doesn't buy
  47. # us much -- we might as well be working with consistent EOLs for files in
  48. # working directories as well as in the repository
  49. #
  50. # b) if the file already of CRLFs, we don't do any normalization. We use '-crlf'
  51. # so that git doesn't do any EOL-conversion of the file. As I said, this
  52. # is mostly harmless on Linux. We can't mark these files as 'crlf' or use
  53. # the new (git 1.7.2) 'eol=crlf' attribute, since it changes the contents
  54. # _inside_ the repository [1], and hence makes history traversal annoying.
  55. # So, we live with occasional EOL corruption.
  56. #
  57. # c) We can handle mixed-EOL files on a case-by-case basis, converting them to
  58. # LF- or CRLF-files based on which causes fewer lines to change
  59. #
  60. # d) We try to ensure no further headaches, by declaring EOL normalization on
  61. # code files, and Unix-flavoured files, like shell-scripts, makefiles, etc.
  62. #
  63. # [1] GIT use LFs as the normalized internal representation.