OLD | NEW |
---|---|
1 # Copyright 2013 The Chromium Authors. All rights reserved. | 1 # Copyright 2013 The Chromium Authors. All rights reserved. |
2 # Use of this source code is governed by a BSD-style license that can be | 2 # Use of this source code is governed by a BSD-style license that can be |
3 # found in the LICENSE file. | 3 # found in the LICENSE file. |
4 | 4 |
5 import json | 5 import json |
6 import logging | 6 import logging |
7 import os | 7 import os |
8 import re | 8 import re |
9 import shutil | 9 import shutil |
10 import tempfile | 10 import tempfile |
11 | 11 |
12 from telemetry import page as page_module | 12 from telemetry import page as page_module |
13 from telemetry.util import cloud_storage | 13 from telemetry.util import cloud_storage |
14 | 14 |
15 | 15 |
16 def AssertValidCloudStorageBucket(bucket): | 16 def AssertValidCloudStorageBucket(bucket): |
17 is_valid = bucket in (None, | 17 is_valid = bucket in (None, |
18 cloud_storage.PUBLIC_BUCKET, | 18 cloud_storage.PUBLIC_BUCKET, |
19 cloud_storage.PARTNER_BUCKET, | 19 cloud_storage.PARTNER_BUCKET, |
20 cloud_storage.INTERNAL_BUCKET) | 20 cloud_storage.INTERNAL_BUCKET) |
21 if not is_valid: | 21 if not is_valid: |
22 raise ValueError("Cloud storage privacy bucket %s is invalid" % bucket) | 22 raise ValueError("Cloud storage privacy bucket %s is invalid" % bucket) |
23 | 23 |
24 | 24 |
25 # TODO(chrishenry): Rename this (and module) to wpr_archive_info.WprArchiveInfo | 25 class WprArchiveInfo(object): |
26 # and move to telemetry.user_story or telemetry.wpr or telemetry.core. | |
27 class PageSetArchiveInfo(object): | |
28 def __init__(self, file_path, data, bucket, ignore_archive=False): | 26 def __init__(self, file_path, data, bucket, ignore_archive=False): |
29 AssertValidCloudStorageBucket(bucket) | 27 AssertValidCloudStorageBucket(bucket) |
30 self._file_path = file_path | 28 self._file_path = file_path |
31 self._base_dir = os.path.dirname(file_path) | 29 self._base_dir = os.path.dirname(file_path) |
32 self._bucket = bucket | 30 self._bucket = bucket |
33 | 31 |
34 # Ensure directory exists. | 32 # Ensure directory exists. |
35 if not os.path.exists(self._base_dir): | 33 if not os.path.exists(self._base_dir): |
36 os.makedirs(self._base_dir) | 34 os.makedirs(self._base_dir) |
37 | 35 |
38 # Download all .wpr files. | 36 # Download all .wpr files. |
39 if not ignore_archive: | 37 if not ignore_archive: |
40 if not self._bucket: | 38 if not self._bucket: |
41 logging.warning('page_set in %s has no bucket specified, and cannot be' | 39 logging.warning('User story set in %s has no bucket specified, and ' |
42 'downloaded from cloud_storage.', file_path) | 40 'cannot be downloaded from cloud_storage.', file_path) |
43 else: | 41 else: |
44 for archive_path in data['archives']: | 42 for archive_path in data['archives']: |
45 archive_path = self._WprFileNameToPath(archive_path) | 43 archive_path = self._WprFileNameToPath(archive_path) |
46 try: | 44 try: |
47 cloud_storage.GetIfChanged(archive_path, bucket) | 45 cloud_storage.GetIfChanged(archive_path, bucket) |
48 except (cloud_storage.CredentialsError, | 46 except (cloud_storage.CredentialsError, |
49 cloud_storage.PermissionError): | 47 cloud_storage.PermissionError): |
50 if os.path.exists(archive_path): | 48 if os.path.exists(archive_path): |
51 # If the archive exists, assume the user recorded their own and | 49 # If the archive exists, assume the user recorded their own and |
52 # simply warn. | 50 # simply warn. |
53 logging.warning('Need credentials to update WPR archive: %s', | 51 logging.warning('Need credentials to update WPR archive: %s', |
54 archive_path) | 52 archive_path) |
55 | 53 |
56 # Map from the relative path (as it appears in the metadata file) of the | 54 # Map from the relative path (as it appears in the metadata file) of the |
57 # .wpr file to a list of page names it supports. | 55 # .wpr file to a list of user story names it supports. |
58 self._wpr_file_to_page_names = data['archives'] | 56 self._wpr_file_to_user_story_names = data['archives'] |
59 | 57 |
60 # Map from the page name to a relative path (as it appears in the metadata | 58 # Map from the user_story name to a relative path (as it appears |
61 # file) of the .wpr file. | 59 # in the metadata file) of the .wpr file. |
62 self._page_name_to_wpr_file = dict() | 60 self._user_story_name_to_wpr_file = dict() |
63 # Find out the wpr file names for each page. | 61 # Find out the wpr file names for each user_story. |
64 for wpr_file in data['archives']: | 62 for wpr_file in data['archives']: |
65 page_names = data['archives'][wpr_file] | 63 user_story_names = data['archives'][wpr_file] |
66 for page_name in page_names: | 64 for user_story_name in user_story_names: |
67 self._page_name_to_wpr_file[page_name] = wpr_file | 65 self._user_story_name_to_wpr_file[user_story_name] = wpr_file |
68 self.temp_target_wpr_file_path = None | 66 self.temp_target_wpr_file_path = None |
69 | 67 |
70 @classmethod | 68 @classmethod |
71 def FromFile(cls, file_path, bucket, ignore_archive=False): | 69 def FromFile(cls, file_path, bucket, ignore_archive=False): |
72 if os.path.exists(file_path): | 70 if os.path.exists(file_path): |
73 with open(file_path, 'r') as f: | 71 with open(file_path, 'r') as f: |
74 data = json.load(f) | 72 data = json.load(f) |
75 return cls(file_path, data, bucket, ignore_archive=ignore_archive) | 73 return cls(file_path, data, bucket, ignore_archive=ignore_archive) |
76 return cls(file_path, {'archives': {}}, bucket, | 74 return cls(file_path, {'archives': {}}, bucket, |
77 ignore_archive=ignore_archive) | 75 ignore_archive=ignore_archive) |
78 | 76 |
79 def WprFilePathForUserStory(self, story): | 77 def WprFilePathForUserStory(self, story): |
80 if self.temp_target_wpr_file_path: | 78 if self.temp_target_wpr_file_path: |
81 return self.temp_target_wpr_file_path | 79 return self.temp_target_wpr_file_path |
82 wpr_file = self._page_name_to_wpr_file.get(story.display_name, None) | 80 wpr_file = self._user_story_name_to_wpr_file.get(story.display_name, None) |
83 if wpr_file is None and isinstance(story, page_module.Page): | 81 if wpr_file is None and isinstance(story, page_module.Page): |
84 # Some old page sets always use the URL to identify a page rather than the | 82 # Some old pages always use the URL to identify a page rather than the |
85 # display_name, so try to look for that. | 83 # display_name, so try to look for that. |
86 wpr_file = self._page_name_to_wpr_file.get(story.url, None) | 84 wpr_file = self._user_story_name_to_wpr_file.get(story.url, None) |
87 if wpr_file: | 85 if wpr_file: |
88 return self._WprFileNameToPath(wpr_file) | 86 return self._WprFileNameToPath(wpr_file) |
89 return None | 87 return None |
90 | 88 |
91 def AddNewTemporaryRecording(self, temp_wpr_file_path=None): | 89 def AddNewTemporaryRecording(self, temp_wpr_file_path=None): |
92 if temp_wpr_file_path is None: | 90 if temp_wpr_file_path is None: |
93 temp_wpr_file_handle, temp_wpr_file_path = tempfile.mkstemp() | 91 temp_wpr_file_handle, temp_wpr_file_path = tempfile.mkstemp() |
94 os.close(temp_wpr_file_handle) | 92 os.close(temp_wpr_file_handle) |
95 self.temp_target_wpr_file_path = temp_wpr_file_path | 93 self.temp_target_wpr_file_path = temp_wpr_file_path |
96 | 94 |
97 def AddRecordedPages(self, pages, upload_to_cloud_storage=False): | 95 def AddRecordedUserStories(self, user_stories, upload_to_cloud_storage=False): |
98 if not pages: | 96 if not user_stories: |
99 os.remove(self.temp_target_wpr_file_path) | 97 os.remove(self.temp_target_wpr_file_path) |
100 return | 98 return |
101 | 99 |
102 (target_wpr_file, target_wpr_file_path) = self._NextWprFileName() | 100 (target_wpr_file, target_wpr_file_path) = self._NextWprFileName() |
103 for page in pages: | 101 for user_story in user_stories: |
104 self._SetWprFileForPage(page.display_name, target_wpr_file) | 102 self._SetWprFileForUserStory(user_story.display_name, target_wpr_file) |
nednguyen
2014/12/08 00:53:16
Do we have a check to make sure that all user stor
| |
105 shutil.move(self.temp_target_wpr_file_path, target_wpr_file_path) | 103 shutil.move(self.temp_target_wpr_file_path, target_wpr_file_path) |
106 | 104 |
107 # Update the hash file. | 105 # Update the hash file. |
108 with open(target_wpr_file_path + '.sha1', 'wb') as f: | 106 with open(target_wpr_file_path + '.sha1', 'wb') as f: |
109 f.write(cloud_storage.CalculateHash(target_wpr_file_path)) | 107 f.write(cloud_storage.CalculateHash(target_wpr_file_path)) |
110 f.flush() | 108 f.flush() |
111 | 109 |
112 self._WriteToFile() | 110 self._WriteToFile() |
113 self._DeleteAbandonedWprFiles() | 111 self._DeleteAbandonedWprFiles() |
114 | 112 |
115 # Upload to cloud storage | 113 # Upload to cloud storage |
116 if upload_to_cloud_storage: | 114 if upload_to_cloud_storage: |
117 if not self._bucket: | 115 if not self._bucket: |
118 logging.warning('PageSet must have bucket specified to upload pages to' | 116 logging.warning('UserStorySet must have bucket specified to upload ' |
119 ' cloud storage.') | 117 'user stories to cloud storage.') |
120 return | 118 return |
121 try: | 119 try: |
122 cloud_storage.Insert(self._bucket, target_wpr_file, | 120 cloud_storage.Insert(self._bucket, target_wpr_file, |
123 target_wpr_file_path) | 121 target_wpr_file_path) |
124 except cloud_storage.CloudStorageError, e: | 122 except cloud_storage.CloudStorageError, e: |
125 logging.warning('Failed to upload wpr file %s to cloud storage. ' | 123 logging.warning('Failed to upload wpr file %s to cloud storage. ' |
126 'Error:%s' % target_wpr_file_path, e) | 124 'Error:%s' % target_wpr_file_path, e) |
127 | 125 |
128 def _DeleteAbandonedWprFiles(self): | 126 def _DeleteAbandonedWprFiles(self): |
129 # Update the metadata so that the abandoned wpr files don't have empty page | 127 # Update the metadata so that the abandoned wpr files don't have |
130 # name arrays. | 128 # empty user story name arrays. |
131 abandoned_wpr_files = self._AbandonedWprFiles() | 129 abandoned_wpr_files = self._AbandonedWprFiles() |
132 for wpr_file in abandoned_wpr_files: | 130 for wpr_file in abandoned_wpr_files: |
133 del self._wpr_file_to_page_names[wpr_file] | 131 del self._wpr_file_to_user_story_names[wpr_file] |
134 # Don't fail if we're unable to delete some of the files. | 132 # Don't fail if we're unable to delete some of the files. |
135 wpr_file_path = self._WprFileNameToPath(wpr_file) | 133 wpr_file_path = self._WprFileNameToPath(wpr_file) |
136 try: | 134 try: |
137 os.remove(wpr_file_path) | 135 os.remove(wpr_file_path) |
138 except Exception: | 136 except Exception: |
139 logging.warning('Failed to delete file: %s' % wpr_file_path) | 137 logging.warning('Failed to delete file: %s' % wpr_file_path) |
140 | 138 |
141 def _AbandonedWprFiles(self): | 139 def _AbandonedWprFiles(self): |
142 abandoned_wpr_files = [] | 140 abandoned_wpr_files = [] |
143 for wpr_file, page_names in self._wpr_file_to_page_names.iteritems(): | 141 for wpr_file, user_story_names in ( |
144 if not page_names: | 142 self._wpr_file_to_user_story_names.iteritems()): |
143 if not user_story_names: | |
145 abandoned_wpr_files.append(wpr_file) | 144 abandoned_wpr_files.append(wpr_file) |
146 return abandoned_wpr_files | 145 return abandoned_wpr_files |
147 | 146 |
148 def _WriteToFile(self): | 147 def _WriteToFile(self): |
149 """Writes the metadata into the file passed as constructor parameter.""" | 148 """Writes the metadata into the file passed as constructor parameter.""" |
150 metadata = dict() | 149 metadata = dict() |
151 metadata['description'] = ( | 150 metadata['description'] = ( |
152 'Describes the Web Page Replay archives for a page set. Don\'t edit by ' | 151 'Describes the Web Page Replay archives for a user story set. ' |
153 'hand! Use record_wpr for updating.') | 152 'Don\'t edit by hand! Use record_wpr for updating.') |
154 metadata['archives'] = self._wpr_file_to_page_names.copy() | 153 metadata['archives'] = self._wpr_file_to_user_story_names.copy() |
155 # Don't write data for abandoned archives. | 154 # Don't write data for abandoned archives. |
156 abandoned_wpr_files = self._AbandonedWprFiles() | 155 abandoned_wpr_files = self._AbandonedWprFiles() |
157 for wpr_file in abandoned_wpr_files: | 156 for wpr_file in abandoned_wpr_files: |
158 del metadata['archives'][wpr_file] | 157 del metadata['archives'][wpr_file] |
159 | 158 |
160 with open(self._file_path, 'w') as f: | 159 with open(self._file_path, 'w') as f: |
161 json.dump(metadata, f, indent=4) | 160 json.dump(metadata, f, indent=4) |
162 f.flush() | 161 f.flush() |
163 | 162 |
164 def _WprFileNameToPath(self, wpr_file): | 163 def _WprFileNameToPath(self, wpr_file): |
165 return os.path.abspath(os.path.join(self._base_dir, wpr_file)) | 164 return os.path.abspath(os.path.join(self._base_dir, wpr_file)) |
166 | 165 |
167 def _NextWprFileName(self): | 166 def _NextWprFileName(self): |
168 """Creates a new file name for a wpr archive file.""" | 167 """Creates a new file name for a wpr archive file.""" |
169 # The names are of the format "some_thing_number.wpr". Read the numbers. | 168 # The names are of the format "some_thing_number.wpr". Read the numbers. |
170 highest_number = -1 | 169 highest_number = -1 |
171 base = None | 170 base = None |
172 for wpr_file in self._wpr_file_to_page_names: | 171 for wpr_file in self._wpr_file_to_user_story_names: |
173 match = re.match(r'(?P<BASE>.*)_(?P<NUMBER>[0-9]+)\.wpr', wpr_file) | 172 match = re.match(r'(?P<BASE>.*)_(?P<NUMBER>[0-9]+)\.wpr', wpr_file) |
174 if not match: | 173 if not match: |
175 raise Exception('Illegal wpr file name ' + wpr_file) | 174 raise Exception('Illegal wpr file name ' + wpr_file) |
176 highest_number = max(int(match.groupdict()['NUMBER']), highest_number) | 175 highest_number = max(int(match.groupdict()['NUMBER']), highest_number) |
177 if base and match.groupdict()['BASE'] != base: | 176 if base and match.groupdict()['BASE'] != base: |
178 raise Exception('Illegal wpr file name ' + wpr_file + | 177 raise Exception('Illegal wpr file name ' + wpr_file + |
179 ', doesn\'t begin with ' + base) | 178 ', doesn\'t begin with ' + base) |
180 base = match.groupdict()['BASE'] | 179 base = match.groupdict()['BASE'] |
181 if not base: | 180 if not base: |
182 # If we're creating a completely new info file, use the base name of the | 181 # If we're creating a completely new info file, use the base name of the |
183 # page set file. | 182 # user story set file. |
184 base = os.path.splitext(os.path.basename(self._file_path))[0] | 183 base = os.path.splitext(os.path.basename(self._file_path))[0] |
185 new_filename = '%s_%03d.wpr' % (base, highest_number + 1) | 184 new_filename = '%s_%03d.wpr' % (base, highest_number + 1) |
186 return new_filename, self._WprFileNameToPath(new_filename) | 185 return new_filename, self._WprFileNameToPath(new_filename) |
187 | 186 |
188 def _SetWprFileForPage(self, page_name, wpr_file): | 187 def _SetWprFileForUserStory(self, user_story_name, wpr_file): |
189 """For modifying the metadata when we're going to record a new archive.""" | 188 """For modifying the metadata when we're going to record a new archive.""" |
190 old_wpr_file = self._page_name_to_wpr_file.get(page_name, None) | 189 old_wpr_file = self._user_story_name_to_wpr_file.get(user_story_name, None) |
191 if old_wpr_file: | 190 if old_wpr_file: |
192 self._wpr_file_to_page_names[old_wpr_file].remove(page_name) | 191 self._wpr_file_to_user_story_names[old_wpr_file].remove(user_story_name) |
193 self._page_name_to_wpr_file[page_name] = wpr_file | 192 self._user_story_name_to_wpr_file[user_story_name] = wpr_file |
194 if wpr_file not in self._wpr_file_to_page_names: | 193 if wpr_file not in self._wpr_file_to_user_story_names: |
195 self._wpr_file_to_page_names[wpr_file] = [] | 194 self._wpr_file_to_user_story_names[wpr_file] = [] |
196 self._wpr_file_to_page_names[wpr_file].append(page_name) | 195 self._wpr_file_to_user_story_names[wpr_file].append(user_story_name) |
OLD | NEW |