1 | import os |
---|
2 | import logging |
---|
3 | from shutil import unpack_archive, rmtree |
---|
4 | from uuid import uuid1 |
---|
5 | from datetime import datetime, timezone |
---|
6 | |
---|
7 | import pytz |
---|
8 | from repoze.folder import Folder |
---|
9 | from pyramid.security import Allow, Deny, Everyone, ALL_PERMISSIONS |
---|
10 | from fitparse.utils import FitHeaderError, FitEOFError |
---|
11 | from gpxpy.gpx import GPXXMLSyntaxException |
---|
12 | |
---|
13 | from ow.utilities import create_blob |
---|
14 | from ow.models.workout import Workout |
---|
15 | |
---|
16 | log = logging.getLogger(__name__) |
---|
17 | |
---|
18 | |
---|
19 | class BulkFile(Folder): |
---|
20 | |
---|
21 | """ |
---|
22 | Object that maps to a compressed file uploaded by a user to upload several |
---|
23 | workout tracking files at once. |
---|
24 | """ |
---|
25 | |
---|
26 | __parent__ = __name__ = None |
---|
27 | |
---|
28 | def __acl__(self): |
---|
29 | """ |
---|
30 | Owner of the compressed file has full permissions to the file, as well |
---|
31 | as system admins. Everybody else has no permissions. |
---|
32 | """ |
---|
33 | permissions = [ |
---|
34 | (Allow, str(self.uid), 'view'), |
---|
35 | (Allow, str(self.uid), 'edit'), |
---|
36 | (Allow, str(self.uid), 'delete'), |
---|
37 | (Deny, Everyone, ALL_PERMISSIONS) |
---|
38 | ] |
---|
39 | return permissions |
---|
40 | |
---|
41 | def __init__(self, **kw): |
---|
42 | super(BulkFile, self).__init__() |
---|
43 | self.bfid = uuid1() |
---|
44 | self.uid = kw['uid'] # required, so let it blow if none is given |
---|
45 | self.uploaded = kw.get('uploaded', datetime.now(timezone.utc)) |
---|
46 | self.compressed_file = kw.get('compressed_file', None) # Blob |
---|
47 | self.file_name = '' # unicode string |
---|
48 | self.file_type = '' # unicode string |
---|
49 | self.loaded = False # datetime when workouts have been loaded |
---|
50 | self.workout_ids = [] # ids of the workouts loaded from this file |
---|
51 | self.loaded_info = {} # per-file information (loaded or not, errors..) |
---|
52 | |
---|
53 | def _in_timezone(self, timezone, value): |
---|
54 | """ |
---|
55 | Return a string representation of the given value date and time, |
---|
56 | localized into the given timezone |
---|
57 | """ |
---|
58 | _value = value.astimezone(pytz.timezone(timezone)) |
---|
59 | return _value.strftime('%d/%m/%Y %H:%M (%Z)') |
---|
60 | |
---|
61 | def uploaded_in_timezone(self, timezone): |
---|
62 | return self._in_timezone(timezone, self.uploaded) |
---|
63 | |
---|
64 | def loaded_in_timezone(self, timezone): |
---|
65 | if self.loaded: |
---|
66 | return self._in_timezone(timezone, self.loaded) |
---|
67 | return '' |
---|
68 | |
---|
69 | def extract(self, tmp_path, path): |
---|
70 | """ |
---|
71 | Extract the files contained in this bulk/compressed file into the |
---|
72 | given path. |
---|
73 | """ |
---|
74 | if self.compressed_file is None: |
---|
75 | return [] |
---|
76 | |
---|
77 | # save the blob into a temporal file |
---|
78 | tmp_file_path = os.path.join(tmp_path, self.file_name) |
---|
79 | with open(tmp_file_path, 'wb') as tmp_file: |
---|
80 | with self.compressed_file.open() as blob: |
---|
81 | tmp_file.write(blob.read()) |
---|
82 | # extract |
---|
83 | unpack_archive(tmp_file_path, path) |
---|
84 | # remove temporary file |
---|
85 | os.remove(tmp_file_path) |
---|
86 | # analyze the extracted contents, return some data |
---|
87 | extracted = [] |
---|
88 | if os.path.exists(path): |
---|
89 | for extracted_file in os.listdir(path): |
---|
90 | extracted.append(os.path.join(path, extracted_file)) |
---|
91 | return extracted |
---|
92 | |
---|
93 | def load(self, root, tmp_path): |
---|
94 | user = root.get_user_by_uid(self.uid) |
---|
95 | # extract |
---|
96 | tmp_extract_path = os.path.join(tmp_path, str(self.bfid)) |
---|
97 | log.info(self.file_name + ' extracting to ' + tmp_extract_path) |
---|
98 | extracted = self.extract(tmp_path, tmp_extract_path) |
---|
99 | log.info(self.file_name + ' ' + str(len(extracted)) + |
---|
100 | ' files extracted') |
---|
101 | |
---|
102 | # loop over extracted files and create the workouts, taking |
---|
103 | # care of duplicates. Store some stats/info in a dict, so we can |
---|
104 | # keep that somewhere, to show to the user later on |
---|
105 | for extracted_file in extracted: |
---|
106 | base_file_name = os.path.basename(extracted_file) |
---|
107 | log_header = self.file_name + '/' + base_file_name |
---|
108 | log.info(log_header + ' loading file') |
---|
109 | |
---|
110 | file_extension = os.path.splitext(base_file_name)[1].strip('.') |
---|
111 | |
---|
112 | # gpx files are text, but fit files are binary files |
---|
113 | open_mode = 'r' |
---|
114 | binary = False |
---|
115 | if file_extension == 'fit': |
---|
116 | open_mode = 'rb' |
---|
117 | binary = True |
---|
118 | |
---|
119 | with open(extracted_file, open_mode) as f_obj: |
---|
120 | blob = create_blob( |
---|
121 | f_obj.read(), file_extension=file_extension, binary=binary) |
---|
122 | |
---|
123 | workout = Workout() |
---|
124 | workout.tracking_file = blob |
---|
125 | workout.tracking_filetype = file_extension |
---|
126 | |
---|
127 | try: |
---|
128 | workout.load_from_file() |
---|
129 | except (FitHeaderError, FitEOFError, GPXXMLSyntaxException) as e: |
---|
130 | log.error(log_header + ' error loading tracking file ') |
---|
131 | log.error(e) |
---|
132 | self.loaded_info[base_file_name] = { |
---|
133 | 'loaded': False, |
---|
134 | 'error': 'tracking file load error', |
---|
135 | 'workout': None, |
---|
136 | } |
---|
137 | else: |
---|
138 | # check for duplicates |
---|
139 | # hashed is not "complete" for a workout that has not been |
---|
140 | # added yet, as it does not have the owner set, so we have to |
---|
141 | # "build it" |
---|
142 | hashed = str(self.uid) + workout.hashed |
---|
143 | duplicate = root.get_workout_by_hash(hashed) |
---|
144 | if duplicate: |
---|
145 | log.warning( |
---|
146 | log_header + |
---|
147 | ' cannot create workout, possible duplicate') |
---|
148 | self.loaded_info[base_file_name] = { |
---|
149 | 'loaded': False, |
---|
150 | 'error': 'Possible duplicate workout', |
---|
151 | 'workout': None, |
---|
152 | } |
---|
153 | else: |
---|
154 | # add the workout only if no errors happened |
---|
155 | user.add_workout(workout) |
---|
156 | log.info(log_header + ' workout added') |
---|
157 | self.loaded_info[base_file_name] = { |
---|
158 | 'loaded': True, |
---|
159 | 'error': None, |
---|
160 | 'workout': workout.workout_id, |
---|
161 | } |
---|
162 | self.workout_ids.append(workout.workout_id) |
---|
163 | |
---|
164 | # clean-up, we have to check if the temporary directory exists, |
---|
165 | # as extract() won't create such directory if the compressed file |
---|
166 | # is empty |
---|
167 | if os.path.exists(tmp_extract_path): |
---|
168 | rmtree(tmp_extract_path) |
---|
169 | |
---|
170 | # mark this bulk file as loaded |
---|
171 | self.loaded = datetime.now(timezone.utc) |
---|
172 | |
---|
173 | |
---|
174 | class BulkFiles(Folder): |
---|
175 | |
---|
176 | """ |
---|
177 | Container for bulk upload compressed files |
---|
178 | """ |
---|
179 | |
---|
180 | __parent__ = __name__ = None |
---|
181 | |
---|
182 | def __acl__(self): |
---|
183 | """ |
---|
184 | Everybody can view, super users can edit |
---|
185 | """ |
---|
186 | permissions = [ |
---|
187 | (Allow, Everyone, 'view'), |
---|
188 | (Allow, 'admins', 'edit'), |
---|
189 | (Deny, Everyone, ALL_PERMISSIONS) |
---|
190 | ] |
---|
191 | return permissions |
---|
192 | |
---|
193 | def add_bulk_file(self, bulk_file): |
---|
194 | self[str(bulk_file.bfid)] = bulk_file |
---|
195 | |
---|
196 | def get_by_uid(self, uid): |
---|
197 | """ |
---|
198 | Return bulk files owned by the given uid |
---|
199 | """ |
---|
200 | return [self[bf] for bf in self if self[bf].uid == str(uid)] |
---|