Fix /doc/build/ -> ./doc/build/ in update_copyright.py
[update-copyright.git] / update_copyright.py
index 10fed81c017efcaaaa43c1fbf06635ae88c4c530..37f4ba5fa841cf234a3b35d30dcd71d6149503bd 100755 (executable)
@@ -1,6 +1,22 @@
 #!/usr/bin/python
 #
-# COPYRIGHT
+# Copyright (C) 2010 W. Trevor King <wking@drexel.edu>
+#
+# This file is part of Hooke.
+#
+# Hooke is free software: you can redistribute it and/or
+# modify it under the terms of the GNU Lesser General Public
+# License as published by the Free Software Foundation, either
+# version 3 of the License, or (at your option) any later version.
+#
+# Hooke is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU Lesser General Public License for more details.
+#
+# You should have received a copy of the GNU Lesser General Public
+# License along with Hooke.  If not, see
+# <http://www.gnu.org/licenses/>.
 
 """Automatically update copyright boilerplate.
 
@@ -47,24 +63,81 @@ License along with %(project)s.  If not, see
 <http://www.gnu.org/licenses/>.
 """.strip()
 
-COPY_RIGHT_TAG='-xyz-COPY-RIGHT-zyx-' # unlikely to occur in the wild :p
+COPY_RIGHT_TAG='-xyz-COPY' + '-RIGHT-zyx-' # unlikely to occur in the wild :p
 
 ALIASES = {
+    'A. Seeholzer':
+        ['A. Seeholzer'],
     'Alberto Gomez-Casado':
         ['albertogomcas'],
     'Massimo Sandal <devicerandom@gmail.com>':
-        ['devicerandom',
+        ['Massimo Sandal',
+         'devicerandom',
          'unknown'],
-    'Fabrizio Benedetti':['fabrizio.benedetti'],
-    'il':['illysam'],
-    'Marco Brucale':['marcobrucale'],
-    'pp':['pancaldi.paolo'],
+    'Fabrizio Benedetti':
+        ['fabrizio.benedetti.82'],
+    'Richard Naud <richard.naud@epfl.ch>':
+        ['Richard Naud'],
+    'Rolf Schmidt <rschmidt@alcor.concordia.ca>':
+        ['Rolf Schmidt',
+         'illysam'],
+    'Marco Brucale':
+        ['marcobrucale'],
+    'Pancaldi Paolo':
+        ['pancaldi.paolo'],
     }
 
-IGNORED_PATHS = ['./.hg/', './doc/img', './test/data/',
-                 './build/', '/doc/build/']
+IGNORED_PATHS = ['./.hg/', './doc/img/', './test/data/',
+                 './build/', './doc/build/']
 IGNORED_FILES = ['COPYING', 'COPYING.LESSER']
 
+# Work around missing author holes in the VCS history
+AUTHOR_HACKS = {
+    ('hooke','driver','hdf5.py'):['Massimo Sandal'],
+    ('hooke','driver','mcs.py'):['Allen Chen'],
+    ('hooke','driver','mfp3d.py'):['A. Seeholzer','Richard Naud','Rolf Schmidt',
+                                   'Alberto Gomez-Casado'],
+    ('hooke','util','peak.py'):['Fabrizio Benedetti'],
+    ('hooke','plugin','showconvoluted.py'):['Rolf Schmidt'],
+    ('hooke','ui','gui','formatter.py'):['Francesco Musiani','Massimo Sandal'],
+    ('hooke','ui','gui','prettyformat.py'):['Rolf Schmidt'],
+    }
+
+# Work around missing year holes in the VCS history
+YEAR_HACKS = {
+    ('hooke','driver','hdf5.py'):2009,
+    ('hooke','driver','mfp3d.py'):2008,
+    ('hooke','driver','picoforce.py'):2006,
+    ('hooke','driver','picoforcealt.py'):2006,
+    ('hooke','util','peak.py'):2007,
+    ('hooke','plugin','showconvoluted.py'):2009,
+    ('hooke','plugin','tutorial.py'):2007,
+    ('hooke','ui','gui','formatter.py'):2006,
+    ('hooke','ui','gui','prettyformat.py'):2009,
+    }
+
+# Helpers for VCS-specific commands
+
+def splitpath(path):
+    """Recursively split a path into elements.
+
+    Examples
+    --------
+
+    >>> splitpath(os.path.join('a', 'b', 'c'))
+    ('a', 'b', 'c')
+    >>> splitpath(os.path.join('.', 'a', 'b', 'c'))
+    ('a', 'b', 'c')
+    """
+    path = os.path.normpath(path)
+    elements = []
+    while True:
+        dirname,basename = os.path.split(path)
+        elements.insert(0,basename)
+        if dirname in ['', '.']:
+            break
+        path = dirname
+    return tuple(elements)
 
 # VCS-specific commands
 
@@ -85,25 +158,33 @@ def mercurial_cmd(*args):
     return (tmp_stdout.getvalue().rstrip('\n'),
             tmp_stderr.getvalue().rstrip('\n'))
 
-def original_year(filename):
+def original_year(filename, year_hacks=YEAR_HACKS):
     # shortdate filter: YEAR-MONTH-DAY
     output,error = mercurial_cmd('log', '--follow',
                                  '--template', '{date|shortdate}\n',
                                  filename)
     years = [int(line.split('-', 1)[0]) for line in output.splitlines()]
+    if splitpath(filename) in year_hacks:
+        years.append(year_hacks[splitpath(filename)])
     years.sort()
     return years[0]
 
-def authors(filename):
+def authors(filename, author_hacks=AUTHOR_HACKS):
     output,error = mercurial_cmd('log', '--follow',
                                  '--template', '{author}\n',
                                  filename)
-    return list(set(output.splitlines()))
+    ret = list(set(output.splitlines()))
+    if splitpath(filename) in author_hacks:
+        ret.extend(author_hacks[splitpath(filename)])
+    return ret
 
-def authors_list():
+def authors_list(author_hacks=AUTHOR_HACKS):
     output,error = mercurial_cmd('log', '--follow',
                                  '--template', '{author}\n')
-    return list(set(output.splitlines()))
+    ret = list(set(output.splitlines()))
+    for path,authors in author_hacks.items():
+        ret.extend(authors)
+    return ret
 
 def is_versioned(filename):
     output,error = mercurial_cmd('log', '--follow',
@@ -244,12 +325,13 @@ def _tag_copyright(contents):
     ... (copyright ends)
     ... bla bla bla
     ... '''
-    >>> print _tag_copyright(contents),
+    >>> print _tag_copyright(contents).replace('COPY-RIGHT', 'CR')
     Some file
     bla bla
-    -xyz-COPY-RIGHT-zyx-
+    -xyz-CR-zyx-
     (copyright ends)
     bla bla bla
+    <BLANKLINE>
     """
     lines = []
     incopy = False
@@ -291,33 +373,34 @@ def _update_copyright(contents, original_year, authors):
     contents = _tag_copyright(contents)
     return contents.replace(COPY_RIGHT_TAG, copyright_string)
 
-def ignored_file(filename, ignored_paths=None, ignored_files=None):
+def ignored_file(filename, ignored_paths=None, ignored_files=None,
+                 check_disk=True, check_vcs=True):
     """
     >>> ignored_paths = ['./a/', './b/']
     >>> ignored_files = ['x', 'y']
-    >>> ignored_file('./a/z', ignored_paths, ignored_files)
+    >>> ignored_file('./a/z', ignored_paths, ignored_files, False, False)
     True
-    >>> ignored_file('./ab/z', ignored_paths, ignored_files)
+    >>> ignored_file('./ab/z', ignored_paths, ignored_files, False, False)
     False
-    >>> ignored_file('./ab/x', ignored_paths, ignored_files)
+    >>> ignored_file('./ab/x', ignored_paths, ignored_files, False, False)
     True
-    >>> ignored_file('./ab/xy', ignored_paths, ignored_files)
+    >>> ignored_file('./ab/xy', ignored_paths, ignored_files, False, False)
     False
-    >>> ignored_file('./z', ignored_paths, ignored_files)
+    >>> ignored_file('./z', ignored_paths, ignored_files, False, False)
     False
     """
     if ignored_paths == None:
         ignored_paths = IGNORED_PATHS
     if ignored_files == None:
         ignored_files = IGNORED_FILES
-    if os.path.isfile(filename) == False:
+    if check_disk == True and os.path.isfile(filename) == False:
         return True
     for path in ignored_paths:
         if filename.startswith(path):
             return True
     if os.path.basename(filename) in ignored_files:
         return True
-    if is_versioned(filename) == False:
+    if check_vcs == True and is_versioned(filename) == False:
         return True
     return False
 
@@ -352,9 +435,11 @@ def _set_contents(filename, contents, original_contents=None, dry_run=False,
 # Update commands
 
 def update_authors(authors_fn=authors_list, dry_run=False, verbose=0):
+    authors = authors_fn()
+    authors = _replace_aliases(authors, with_email=True, aliases=ALIASES)
     new_contents = '%s was written by:\n%s\n' % (
         PROJECT_INFO['project'],
-        '\n'.join(authors_fn())
+        '\n'.join(authors)
         )
     _set_contents('AUTHORS', new_contents, dry_run=dry_run, verbose=verbose)