contrib/hook-scripts/case-insensitive.py

   1 #!/usr/bin/env python
   2
   3 # Licensed under the same terms as Subversion.
   4
   5 # A pre-commit hook to detect case-insensitive filename clashes.
   6 #
   7 # What this script does:
   8 #  - Detects new paths that 'clash' with existing, or other new, paths.
   9 #  - Ignores existings paths that already 'clash'
  10 #  - Exits with an error code, and a diagnostic on stderr, if 'clashes'
  11 #    are detected.
  12 #
  13 # How it does it:
  14 #  - Get a list of changed paths.
  15 #  - From that list extract the new paths that represent adds or replaces.
  16 #  - For each new path:
  17 #    - Split the path into a directory and a name.
  18 #    - Get the names of all the entries in the version of the directory
  19 #      within the txn.
  20 #    - Compare the canonical new name with each canonical entry name.
  21 #    - If the canonical names match and the pristine names do not match
  22 #      then we have a 'clash'.
  23 #
  24 # Notes:
  25 #  - All the paths from the Subversion filesystem bindings are encoded
  26 #    in UTF-8 and the separator is '/' on all OS's.
  27 #  - The canonical form determines what constitutes a 'clash', at present
  28 #    a simple 'lower case' is used.  That's probably not identical to the
  29 #    behaviour of Windows or OSX, but it might be good enough.
  30 #  - Hooks get invoked with an empty environment so this script explicitly
  31 #    sets a locale; make sure it is a sensible value.
  32 #  - If used with Apache the 'clash' diagnostic must be ASCII irrespective
  33 #    of the locale, see the 'Force' comment near the end of the script for
  34 #    one way to achieve this.
  35 #
  36 # How to call it:
  37 #
  38 #   On a Unix system put this script in the hooks directory and add this to
  39 #   the pre-commit script:
  40 #
  41 #     $REPOS/hooks/case-insensitive.py "$REPOS" "$TXN" || exit 1
  42 #
  43 #   On a windows machine add this to pre-commit.bat:
  44 #
  45 #     python <path-to-script>\case-insensitive.py %1 %2
  46 #     if errorlevel 1 goto :ERROR
  47 #     exit 0
  48 #     :ERROR
  49 #     echo Error found in commit 1>&2
  50 #     exit 1
  51 #
  52 #   Make sure the python bindings are installed and working on Windows.  The
  53 #   zip file can be downloaded from the Subversion site.  The bindings depend
  54 #   on dll's shipped as part of the Subversion binaries, if the script cannot
  55 #   load the _fs dll it is because it cannot find the other Subversion dll's.
  56 #
  57 # $HeadURL$
  58 # $LastChangedRevision$
  59 # $LastChangedDate$
  60 # $LastChangedBy$
  61
  62 import sys, locale
  63 sys.path.append('/usr/local/subversion/lib/svn-python')
  64 from svn import repos, fs
  65 locale.setlocale(locale.LC_ALL, 'en_GB')
  66
  67 def canonicalize(path):
  68   return path.decode('utf-8').lower().encode('utf-8')
  69
  70 def get_new_paths(txn_root):
  71   new_paths = []
  72   for path, change in fs.paths_changed(txn_root).iteritems():
  73     if (change.change_kind == fs.path_change_add
  74         or change.change_kind == fs.path_change_replace):
  75       new_paths.append(path)
  76   return new_paths
  77
  78 def split_path(path):
  79   slash = path.rindex('/')
  80   if (slash == 0):
  81     return '/', path[1:]
  82   return path[:slash], path[slash+1:]
  83
  84 def join_path(dir, name):
  85   if (dir == '/'):
  86     return '/' + name
  87   return dir + '/' + name
  88
  89 def ensure_names(path, names, txn_root):
  90   if (not names.has_key(path)):
  91      names[path] = []
  92      for name, dirent in fs.dir_entries(txn_root, path).iteritems():
  93        names[path].append([canonicalize(name), name])
  94
  95 names = {}   # map of: key - path, value - list of two element lists of names
  96 clashes = {} # map of: key - path, value - map of: key - path, value - dummy
  97
  98 native = locale.getlocale()[1]
  99 if not native: native = 'ascii'
 100 repos_handle = repos.open(sys.argv[1].decode(native).encode('utf-8'))
 101 fs_handle = repos.fs(repos_handle)
 102 txn_handle = fs.open_txn(fs_handle, sys.argv[2].decode(native).encode('utf-8'))
 103 txn_root = fs.txn_root(txn_handle)
 104
 105 new_paths = get_new_paths(txn_root)
 106 for path in new_paths:
 107   dir, name = split_path(path)
 108   canonical = canonicalize(name)
 109   ensure_names(dir, names, txn_root)
 110   for name_pair in names[dir]:
 111     if (name_pair[0] == canonical and name_pair[1] != name):
 112       canonical_path = join_path(dir, canonical)
 113       if (not clashes.has_key(canonical_path)):
 114         clashes[canonical_path] = {}
 115       clashes[canonical_path][join_path(dir, name)] = True
 116       clashes[canonical_path][join_path(dir, name_pair[1])] = True
 117
 118 if (clashes):
 119   # native = 'ascii' # Force ASCII output for Apache
 120   for canonical_path in clashes.iterkeys():
 121     sys.stderr.write(u'Clash:'.encode(native))
 122     for path in clashes[canonical_path].iterkeys():
 123       sys.stderr.write(u' \''.encode(native) +
 124                        str(path).decode('utf-8').encode(native, 'replace') +
 125                        u'\''.encode(native))
 126     sys.stderr.write(u'\n'.encode(native))
 127   sys.exit(1)