1#!/usr/bin/env python
2#
3# Copyright (C) 2012 The Android Open Source Project
4#
5# Licensed under the Apache License, Version 2.0 (the "License");
6# you may not use this file except in compliance with the License.
7# You may obtain a copy of the License at
8#
9#      http://www.apache.org/licenses/LICENSE-2.0
10#
11# Unless required by applicable law or agreed to in writing, software
12# distributed under the License is distributed on an "AS IS" BASIS,
13# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
14# See the License for the specific language governing permissions and
15# limitations under the License.
16"""
17Usage: generate-notice-files --text-output [plain text output file] \
18               --html-output [html output file] \
19               --xml-output [xml output file] \
20               -t [file title] -s [directory of notices]
21
22Generate the Android notice files, including both text and html files.
23
24-h to display this usage message and exit.
25"""
26from collections import defaultdict
27import argparse
28import hashlib
29import itertools
30import os
31import os.path
32import re
33import sys
34
35MD5_BLOCKSIZE = 1024 * 1024
36HTML_ESCAPE_TABLE = {
37    "&": "&",
38    '"': """,
39    "'": "'",
40    ">": ">",
41    "<": "&lt;",
42    }
43
44def hexify(s):
45    return ("%02x"*len(s)) % tuple(map(ord, s))
46
47def md5sum(filename):
48    """Calculate an MD5 of the file given by FILENAME,
49    and return hex digest as a string.
50    Output should be compatible with md5sum command"""
51
52    f = open(filename, "rb")
53    sum = hashlib.md5()
54    while 1:
55        block = f.read(MD5_BLOCKSIZE)
56        if not block:
57            break
58        sum.update(block)
59    f.close()
60    return hexify(sum.digest())
61
62
63def html_escape(text):
64    """Produce entities within text."""
65    return "".join(HTML_ESCAPE_TABLE.get(c,c) for c in text)
66
67HTML_OUTPUT_CSS="""
68<style type="text/css">
69body { padding: 0; font-family: sans-serif; }
70.same-license { background-color: #eeeeee; border-top: 20px solid white; padding: 10px; }
71.label { font-weight: bold; }
72.file-list { margin-left: 1em; color: blue; }
73</style>
74"""
75
76def combine_notice_files_html(file_hash, input_dir, output_filename):
77    """Combine notice files in FILE_HASH and output a HTML version to OUTPUT_FILENAME."""
78
79    SRC_DIR_STRIP_RE = re.compile(input_dir + "(/.*).txt")
80
81    # Set up a filename to row id table (anchors inside tables don't work in
82    # most browsers, but href's to table row ids do)
83    id_table = {}
84    id_count = 0
85    for value in file_hash:
86        for filename in value:
87             id_table[filename] = id_count
88        id_count += 1
89
90    # Open the output file, and output the header pieces
91    output_file = open(output_filename, "wb")
92
93    print >> output_file, "<html><head>"
94    print >> output_file, HTML_OUTPUT_CSS
95    print >> output_file, '</head><body topmargin="0" leftmargin="0" rightmargin="0" bottommargin="0">'
96
97    # Output our table of contents
98    print >> output_file, '<div class="toc">'
99    print >> output_file, "<ul>"
100
101    # Flatten the list of lists into a single list of filenames
102    sorted_filenames = sorted(itertools.chain.from_iterable(file_hash))
103
104    # Print out a nice table of contents
105    for filename in sorted_filenames:
106        stripped_filename = SRC_DIR_STRIP_RE.sub(r"\1", filename)
107        print >> output_file, '<li><a href="#id%d">%s</a></li>' % (id_table.get(filename), stripped_filename)
108
109    print >> output_file, "</ul>"
110    print >> output_file, "</div><!-- table of contents -->"
111    # Output the individual notice file lists
112    print >>output_file, '<table cellpadding="0" cellspacing="0" border="0">'
113    for value in file_hash:
114        print >> output_file, '<tr id="id%d"><td class="same-license">' % id_table.get(value[0])
115        print >> output_file, '<div class="label">Notices for file(s):</div>'
116        print >> output_file, '<div class="file-list">'
117        for filename in value:
118            print >> output_file, "%s <br/>" % (SRC_DIR_STRIP_RE.sub(r"\1", filename))
119        print >> output_file, "</div><!-- file-list -->"
120        print >> output_file
121        print >> output_file, '<pre class="license-text">'
122        print >> output_file, html_escape(open(value[0]).read())
123        print >> output_file, "</pre><!-- license-text -->"
124        print >> output_file, "</td></tr><!-- same-license -->"
125        print >> output_file
126        print >> output_file
127        print >> output_file
128
129    # Finish off the file output
130    print >> output_file, "</table>"
131    print >> output_file, "</body></html>"
132    output_file.close()
133
134def combine_notice_files_text(file_hash, input_dir, output_filename, file_title):
135    """Combine notice files in FILE_HASH and output a text version to OUTPUT_FILENAME."""
136
137    SRC_DIR_STRIP_RE = re.compile(input_dir + "(/.*).txt")
138    output_file = open(output_filename, "wb")
139    print >> output_file, file_title
140    for value in file_hash:
141      print >> output_file, "============================================================"
142      print >> output_file, "Notices for file(s):"
143      for filename in value:
144        print >> output_file, SRC_DIR_STRIP_RE.sub(r"\1", filename)
145      print >> output_file, "------------------------------------------------------------"
146      print >> output_file, open(value[0]).read()
147    output_file.close()
148
149def combine_notice_files_xml(files_with_same_hash, input_dir, output_filename):
150    """Combine notice files in FILE_HASH and output a XML version to OUTPUT_FILENAME."""
151
152    SRC_DIR_STRIP_RE = re.compile(input_dir + "(/.*).txt")
153
154    # Set up a filename to row id table (anchors inside tables don't work in
155    # most browsers, but href's to table row ids do)
156    id_table = {}
157    for file_key in files_with_same_hash.keys():
158        for filename in files_with_same_hash[file_key]:
159             id_table[filename] = file_key
160
161    # Open the output file, and output the header pieces
162    output_file = open(output_filename, "wb")
163
164    print >> output_file, '<?xml version="1.0" encoding="utf-8"?>'
165    print >> output_file, "<licenses>"
166
167    # Flatten the list of lists into a single list of filenames
168    sorted_filenames = sorted(id_table.keys())
169
170    # Print out a nice table of contents
171    for filename in sorted_filenames:
172        stripped_filename = SRC_DIR_STRIP_RE.sub(r"\1", filename)
173        print >> output_file, '<file-name contentId="%s">%s</file-name>' % (id_table.get(filename), stripped_filename)
174
175    print >> output_file
176    print >> output_file
177
178    processed_file_keys = []
179    # Output the individual notice file lists
180    for filename in sorted_filenames:
181        file_key = id_table.get(filename)
182        if file_key in processed_file_keys:
183            continue
184        processed_file_keys.append(file_key)
185
186        print >> output_file, '<file-content contentId="%s"><![CDATA[%s]]></file-content>' % (file_key, html_escape(open(filename).read()))
187        print >> output_file
188
189    # Finish off the file output
190    print >> output_file, "</licenses>"
191    output_file.close()
192
193def get_args():
194    parser = argparse.ArgumentParser()
195    parser.add_argument(
196        '--text-output', required=True,
197        help='The text output file path.')
198    parser.add_argument(
199        '--html-output',
200        help='The html output file path.')
201    parser.add_argument(
202        '--xml-output',
203        help='The xml output file path.')
204    parser.add_argument(
205        '-t', '--title', required=True,
206        help='The file title.')
207    parser.add_argument(
208        '-s', '--source-dir', required=True,
209        help='The directory containing notices.')
210    parser.add_argument(
211        '-i', '--included-subdirs', action='append',
212        help='The sub directories which should be included.')
213    parser.add_argument(
214        '-e', '--excluded-subdirs', action='append',
215        help='The sub directories which should be excluded.')
216    return parser.parse_args()
217
218def main(argv):
219    args = get_args()
220
221    txt_output_file = args.text_output
222    html_output_file = args.html_output
223    xml_output_file = args.xml_output
224    file_title = args.title
225    included_subdirs = []
226    excluded_subdirs = []
227    if args.included_subdirs is not None:
228        included_subdirs = args.included_subdirs
229    if args.excluded_subdirs is not None:
230        excluded_subdirs = args.excluded_subdirs
231
232    # Find all the notice files and md5 them
233    input_dir = os.path.normpath(args.source_dir)
234    files_with_same_hash = defaultdict(list)
235    for root, dir, files in os.walk(input_dir):
236        for file in files:
237            matched = True
238            if len(included_subdirs) > 0:
239                matched = False
240                for subdir in included_subdirs:
241                    if (root == (input_dir + '/' + subdir) or
242                        root.startswith(input_dir + '/' + subdir + '/')):
243                        matched = True
244                        break
245            elif len(excluded_subdirs) > 0:
246                for subdir in excluded_subdirs:
247                    if (root == (input_dir + '/' + subdir) or
248                        root.startswith(input_dir + '/' + subdir + '/')):
249                        matched = False
250                        break
251            if matched and file.endswith(".txt"):
252                filename = os.path.join(root, file)
253                file_md5sum = md5sum(filename)
254                files_with_same_hash[file_md5sum].append(filename)
255
256    filesets = [sorted(files_with_same_hash[md5]) for md5 in sorted(files_with_same_hash.keys())]
257
258    combine_notice_files_text(filesets, input_dir, txt_output_file, file_title)
259
260    if html_output_file is not None:
261        combine_notice_files_html(filesets, input_dir, html_output_file)
262
263    if xml_output_file is not None:
264        combine_notice_files_xml(files_with_same_hash, input_dir, xml_output_file)
265
266if __name__ == "__main__":
267    main(sys.argv)
268