Fix third-party license information for a number of projects

tools/licenses.py

 #!/usr/bin/env python
 # Copyright (c) 2012 The Chromium Authors. All rights reserved.
 # Use of this source code is governed by a BSD-style license that can be
 # found in the LICENSE file.
 
 """Utility for checking and processing licensing information in third_party
 directories.
 
 Usage: licenses.py <command>
 
@@ skipping 75 matching lines @@
     # Harfbuzz-ng is not currently shipping in any product:
     os.path.join('third_party','harfbuzz-ng'),
 ])
 
 # Directories we don't scan through.
 PRUNE_DIRS = ('.svn', '.git',             # VCS metadata
               'out', 'Debug', 'Release',  # build files
               'layout_tests')             # lots of subdirs
 
 ADDITIONAL_PATHS = (
+    os.path.join('googleurl'),
+    os.path.join('native_client_sdk'),
     # The directory with the word list for Chinese and Japanese segmentation
     # with different license terms than ICU.
     os.path.join('third_party','icu','source','data','brkitr'),
     # Fake directory so we can include the strongtalk license.
     os.path.join('v8', 'strongtalk'),
-    # Fake directory so we can include the google-url license.
-    'googleurl',
 )
 
 
 # Directories where we check out directly from upstream, and therefore
 # can't provide a README.chromium.  Please prefer a README.chromium
 # wherever possible.
 SPECIAL_CASES = {
     'googleurl': {
         "Name": "google-url",
         "URL": "http://code.google.com/p/google-url/",
         "License File": "LICENSE.txt",
     },
+    os.path.join('third_party', 'GTM'): {
+        "Name": "Google Toolbox for Mac",
+        "URL": "http://code.google.com/p/google-toolbox-for-mac/",
+        "License": "Apache 2.0",
+        "License File": "COPYING",
+    },
+    os.path.join('third_party', 'WebKit'): {
+        "Name": "WebKit",
+        "URL": "http://webkit.org/",
+        "License": "BSD and GPL v2",
+        # Absolute path here is resolved as relative to the source root.
+        "License File": "/webkit/LICENSE",

[Nico, 2012/07/26 19:45:04]

nit: It looks like someone tried to keep these ordered alphabetically by path
(case-insensitively). These two entries should be further down, I think.

[Steve Block, 2012/07/26 22:11:00]

Hmm, they were still out of order, even if considered case-insensitive, but that
might create less churn - done.

+    },
     os.path.join('third_party', 'angle'): {
         "Name": "Almost Native Graphics Layer Engine",
         "URL": "http://code.google.com/p/angleproject/",
     },
+    os.path.join('third_party', 'cros_system_api'): {
+        "Name": "Chromium OS system API",
+        "URL": "http://code.google.com/apis/protocolbuffers",

[Nico, 2012/07/26 19:45:04]

This looks wrong.

[Steve Block, 2012/07/26 22:11:00]

This is from the README, but I guess it's specific to proto-buffers, so I've
replaced it with a generic Chromium OS URL.

+        "License": "BSD",
+        # Absolute path here is resolved as relative to the source root.
+        "License File": "/LICENSE.chromium_os",
+    },
     os.path.join('third_party', 'lss'): {
         "Name": "linux-syscall-support",
         "URL": "http://code.google.com/p/lss/",
     },
     os.path.join('third_party', 'mozrunner'): {
         "Name": "mozrunner",
         "URL": "http://pypi.python.org/packages/source/m/mozrunner",
         "License": "MPL 1.1/GPL 2.0/LGPL 2.1",
     },
     os.path.join('third_party', 'ots'): {
         "Name": "OTS (OpenType Sanitizer)",
         "URL": "http://code.google.com/p/ots/",
     },
+    os.path.join('third_party', 'pdfsqueeze'): {
+        "Name": "pdfsqueeze",
+        "URL": "http://code.google.com/p/pdfsqueeze/",
+        "License": "Apache 2.0",
+        "License File": "COPYING",
+    },
     os.path.join('third_party', 'ppapi'): {
         "Name": "ppapi",
         "URL": "http://code.google.com/p/ppapi/",
     },
     os.path.join('third_party', 'pylib_simplejson'): {
         "Name": "simplejson",
         "URL": "http://pypi.python.org/packages/source/s/simplejson",
         "License": "MIT",
     },
-    os.path.join('third_party', 'WebKit'): {
-        "Name": "WebKit",
-        "URL": "http://webkit.org/",
-        # Absolute path here is resolved as relative to the source root.
-        "License File": "/webkit/LICENSE",
+    os.path.join('third_party', 'scons-2.0.1'): {
+        "Name": "scons-2.0.1",
+        "URL": " http://www.scons.org",
+        "License": "MIT",
     },
-    os.path.join('third_party', 'GTM'): {
-        "Name": "Google Toolbox for Mac",
-        "URL": "http://code.google.com/p/google-toolbox-for-mac/",
-        "License File": "COPYING",
-    },
-    os.path.join('third_party', 'pdfsqueeze'): {
-        "Name": "pdfsqueeze",
-        "URL": "http://code.google.com/p/pdfsqueeze/",
-        "License File": "COPYING",
+    os.path.join('third_party', 'trace-viewer'): {
+        "Name": "trace-viewer",
+        "URL": "http://code.google.com/p/trace-viewer",
+        "License": "BSD",
     },
     os.path.join('third_party', 'v8-i18n'): {
         "Name": "Internationalization Library for v8",
         "URL": "http://code.google.com/p/v8-i18n/",
     },
+    os.path.join('third_party', 'webpagereplay'): {
+        "Name": "webpagereplay",
+        "URL": "http://code.google.com/p/web-page-replay",
+        "License": "Apache 2.0",
+    },
     os.path.join('v8', 'strongtalk'): {
         "Name": "Strongtalk",
         "URL": "http://www.strongtalk.org/",
+        # Absolute path here is resolved as relative to the source root.
         "License File": "/v8/LICENSE.strongtalk",
     },
 }
 
 class LicenseError(Exception):
     """We raise this exception when a directory's licensing info isn't
     fully filled out."""
     pass
 
 def AbsolutePath(path, filename):
@@ skipping 23 matching lines @@
     # Relative path to a file containing some html we're required to place in
     # about:credits.
     optional_keys = ["Required Text"]
 
     if path in SPECIAL_CASES:
         metadata.update(SPECIAL_CASES[path])
     else:
         # Try to find README.chromium.
         readme_path = os.path.join(path, 'README.chromium')
         if not os.path.exists(readme_path):
-            raise LicenseError("missing README.chromium")
+            raise LicenseError("missing README.chromium or licenses.py "
+                               "SPECIAL_CASES entry")
 
         for line in open(readme_path):
             line = line.strip()
             if not line:
                 break
             for key in metadata.keys() + optional_keys:
                 field = key + ": "
                 if line.startswith(field):
                     metadata[key] = line[len(field):]
 
@@ skipping 21 matching lines @@
     if "Required Text" in metadata:
         required_path = AbsolutePath(path, metadata["Required Text"])
         if required_path is not None:
             metadata["Required Text"] = required_path
         else:
             raise LicenseError("Required text file listed but not found.")
 
     return metadata
 
 
+def ContainsFiles(path):
+    """Determines whether a directory contains any files."""

[Nico, 2012/07/26 19:45:04]

"any files in any subdirectory"

[Steve Block, 2012/07/26 22:11:00]

Done.

+    for _, _, files in os.walk(path):
+        if files:
+            return True
+    return False
+
+
 def FindThirdPartyDirs():
     """Find all third_party directories underneath the current directory."""
     third_party_dirs = []
     for path, dirs, files in os.walk('.'):
         path = path[len('./'):]  # Pretty up the path.
 
         if path in PRUNE_PATHS:
             dirs[:] = []
             continue
 
         # Prune out directories we want to skip.
         # (Note that we loop over PRUNE_DIRS so we're not iterating over a
         # list that we're simultaneously mutating.)
         for skip in PRUNE_DIRS:
             if skip in dirs:
                 dirs.remove(skip)
 
         if os.path.basename(path) == 'third_party':
             # Add all subdirectories that are not marked for skipping.
             for dir in dirs:
                 dirpath = os.path.join(path, dir)
                 if dirpath not in PRUNE_PATHS:
                     third_party_dirs.append(dirpath)
 
             # Don't recurse into any subdirs from here.
             dirs[:] = []
             continue
 
+        # Don't recurse into paths in ADDITIONAL_PATHS, like we do with regular
+        # third_party/foo paths.
+        if path in ADDITIONAL_PATHS:
+            dirs[:] = []
+
     for dir in ADDITIONAL_PATHS:
         third_party_dirs.append(dir)
 
-    return third_party_dirs
+    # If a directory contains no files, assume it's a DEPS directory for a
+    # project not used by our current configuration and skip it.
+    return [x for x in third_party_dirs if ContainsFiles(x)]
 
 
 def ScanThirdPartyDirs():
     """Scan a list of directories and report on any problems we find."""
     third_party_dirs = FindThirdPartyDirs()
 
     errors = []
     for path in sorted(third_party_dirs):
         try:
             metadata = ParseDir(path)
@@ skipping 60 matching lines @@
     elif command == 'credits':
         if not GenerateCredits():
             return 1
     else:
         print __doc__
         return 1
 
 
 if __name__ == '__main__':
   sys.exit(main())