takino
/
ordigi
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Fix get_path, get_part and change get_date_from_string functions

This commit is contained in:
Cédric Leporcq 2021-08-08 13:09:13 +02:00
parent 38794cd41d
commit 9dfb870704
2 changed files with 69 additions and 70 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

135
dozo/filesystem.py
View File

@ -197,17 +197,20 @@ class FileSystem(object):
# Each item has its own custom logic and we evaluate a single item and return # Each item has its own custom logic and we evaluate a single item and return
# the evaluated string. # the evaluated string.
if item in ('basename'): part = ''
return os.path.basename(metadata['base_name']) if item == 'basename':
elif item is 'date': part = os.path.basename(metadata['base_name'])
elif item == 'name':
# Remove date prefix added to the name.
part = metadata['base_name']
for i, rx in self.match_date_from_string(metadata['base_name']):
part = re.sub(rx, '', part)
elif item == 'date':
date = self.get_date_taken(metadata) date = self.get_date_taken(metadata)
# early morning photos can be grouped with previous day # early morning photos can be grouped with previous day
date = self.check_for_early_morning_photos(date) date = self.check_for_early_morning_photos(date)
if date is not None: if date is not None:
return date.strftime(mask) part = date.strftime(mask)
else:
return ''
elif item in ('location', 'city', 'state', 'country'): elif item in ('location', 'city', 'state', 'country'):
place_name = geolocation.place_name( place_name = geolocation.place_name(
metadata['latitude'], metadata['latitude'],
@ -219,51 +222,33 @@ class FileSystem(object):
if item == 'location': if item == 'location':
mask = 'default' mask = 'default'
return self.get_location_part(mask, item, place_name) part = self.get_location_part(mask, item, place_name)
elif item in ('folder'): elif item == 'folder':
return os.path.basename(subdirs) part = os.path.basename(subdirs)
elif item in ('folders'): elif item == 'folders':
folders = pathlib.Path(subdirs).parts folders = pathlib.Path(subdirs).parts
folders = eval(mask) folders = eval(mask)
return os.path.join(*folders) part = os.path.join(*folders)
elif item in ('album','camera_make', 'camera_model', 'ext', elif item in ('album','camera_make', 'camera_model', 'ext',
'title'): 'title'):
if metadata[item]: if metadata[item]:
# return metadata[item] part = metadata[item]
return re.sub(self.whitespace_regex, '_', metadata[item].strip())
elif item in ('original_name'): elif item in ('original_name'):
# First we check if we have metadata['original_name']. # First we check if we have metadata['original_name'].
# We have to do this for backwards compatibility because # We have to do this for backwards compatibility because
# we original did not store this back into EXIF. # we original did not store this back into EXIF.
if metadata[item]: if metadata[item]:
part = os.path.splitext(metadata['original_name'])[0] part = os.path.splitext(metadata['original_name'])[0]
else:
# We didn't always store original_name so this is
# for backwards compatability.
# We want to remove the hardcoded date prefix we used
# to add to the name.
# This helps when re-running the program on file
# which were already processed.
part = re.sub(
'^\d{4}-\d{2}-\d{2}_\d{2}-\d{2}-\d{2}-',
'',
metadata['base_name']
)
if(len(part) == 0):
part = metadata['base_name']
# Lastly we want to sanitize the name
return re.sub(self.whitespace_regex, '_', part.strip())
elif item in 'custom': elif item in 'custom':
# Fallback string # Fallback string
return mask[1:-1] part = mask[1:-1]
return '' return part
def get_path(self, metadata, db, subdirs='', whitespace_sub='_'):
def get_path(self, metadata, db, subdirs=''):
"""path_format: {%Y-%d-%m}/%u{city}/{album} """path_format: {%Y-%d-%m}/%u{city}/{album}
Returns file path. Returns file path.
@ -278,23 +263,26 @@ class FileSystem(object):
# p = [] # p = []
for this_part in this_parts: for this_part in this_parts:
# parts = '' # parts = ''
for item, mask in self.items.items(): for item, regex in self.items.items():
matched = re.search(mask, this_part) matched = re.search(regex, this_part)
if matched: if matched:
# parts = re.split(mask, this_part) # parts = re.split(mask, this_part)
# parts = this_part.split('%')[1:] # parts = this_part.split('%')[1:]
part = self.get_part(item, matched.group()[1:-1], metadata, db, part = self.get_part(item, matched.group()[1:-1], metadata, db,
subdirs) subdirs)
part = part.strip()
# Capitalization # Capitalization
umask = '%u' + mask u_regex = '%u' + regex
lmask = '%l' + mask l_regex = '%l' + regex
if re.search(umask, this_part): if re.search(u_regex, this_part):
this_part = re.sub(umask, part.upper(), this_part) this_part = re.sub(u_regex, part.upper(), this_part)
elif re.search(lmask, this_part): elif re.search(l_regex, this_part):
this_part = re.sub(lmask, part.lower(), this_part) this_part = re.sub(l_regex, part.lower(), this_part)
else: else:
this_part = re.sub(mask, part, this_part) this_part = re.sub(regex, part, this_part)
if this_part: if this_part:
# Check if all masks are substituted # Check if all masks are substituted
@ -308,14 +296,18 @@ class FileSystem(object):
break break
# Else we continue for fallbacks # Else we continue for fallbacks
return os.path.join(*path) if(len(path[-1]) == 0):
path[-1] = metadata['base_name']
path_string = os.path.join(*path)
def get_date_from_string(self, string, user_regex=None): if whitespace_sub != ' ':
# If missing datetime from EXIF data check if filename is in datetime format. # Lastly we want to sanitize the name
# For this use a user provided regex if possible. path_string = re.sub(self.whitespace_regex, whitespace_sub, path_string)
# Otherwise assume a filename such as IMG_20160915_123456.jpg as default.
return path_string
def match_date_from_string(self, string, user_regex=None):
if user_regex is not None: if user_regex is not None:
matches = re.findall(user_regex, string) matches = re.findall(user_regex, string)
else: else:
@ -325,30 +317,38 @@ class FileSystem(object):
'a': re.compile( 'a': re.compile(
r'.*[_-]?(?P<year>\d{4})[_-]?(?P<month>\d{2})[_-]?(?P<day>\d{2})[_-]?(?P<hour>\d{2})[_-]?(?P<minute>\d{2})[_-]?(?P<second>\d{2})'), r'.*[_-]?(?P<year>\d{4})[_-]?(?P<month>\d{2})[_-]?(?P<day>\d{2})[_-]?(?P<hour>\d{2})[_-]?(?P<minute>\d{2})[_-]?(?P<second>\d{2})'),
'b': re.compile ( 'b': re.compile (
'[-_./](?P<year>\d{4})[-_.]?(?P<month>\d{2})[-_.]?(?P<day>\d{2})[-_./]'), r'[-_./](?P<year>\d{4})[-_.]?(?P<month>\d{2})[-_.]?(?P<day>\d{2})[-_./]'),
# not very accurate # not very accurate
'c': re.compile ( 'c': re.compile (
'[-_./](?P<year>\d{2})[-_.]?(?P<month>\d{2})[-_.]?(?P<day>\d{2})[-_./]'), r'[-_./](?P<year>\d{2})[-_.]?(?P<month>\d{2})[-_.]?(?P<day>\d{2})[-_./]'),
'd': re.compile ( 'd': re.compile (
'[-_./](?P<day>\d{2})[-_.](?P<month>\d{2})[-_.](?P<year>\d{4})[-_./]') r'[-_./](?P<day>\d{2})[-_.](?P<month>\d{2})[-_.](?P<year>\d{4})[-_./]')
} }
matches = []
for i, rx in regex.items(): for i, rx in regex.items():
match = re.findall(rx, string) yield i, rx
if match != []:
if i == 'c': def get_date_from_string(self, string, user_regex=None):
match = [('20'+match[0][0],match[0][1],match[0][2])] # If missing datetime from EXIF data check if filename is in datetime format.
elif i == 'd': # For this use a user provided regex if possible.
# reorder items # Otherwise assume a filename such as IMG_20160915_123456.jpg as default.
match = [(match[0][2],match[0][1],match[0][0])]
# matches = match + matches matches = []
if len(match) != 1: for i, rx in self.match_date_from_string(string, user_regex):
# The time string is not uniq match = re.findall(rx, string)
continue if match != []:
matches.append((match[0], rx)) if i == 'c':
# We want only the first match for the moment match = [('20' + match[0][0], match[0][1], match[0][2])]
break elif i == 'd':
# reorder items
match = [(match[0][2], match[0][1], match[0][0])]
# matches = match + matches
if len(match) != 1:
# The time string is not uniq
continue
matches.append((match[0], rx))
# We want only the first match for the moment
break
# check if there is only one result # check if there is only one result
if len(set(matches)) == 1: if len(set(matches)) == 1:
@ -368,7 +368,6 @@ class FileSystem(object):
return None return None
def get_date_taken(self, metadata): def get_date_taken(self, metadata):
''' '''
Get the date taken from metadata or filename Get the date taken from metadata or filename
@ -397,8 +396,8 @@ class FileSystem(object):
self.logger.warn(f"{basename} time mark is more recent than {date_created}") self.logger.warn(f"{basename} time mark is more recent than {date_created}")
return date_filename return date_filename
if True: if True:
# TODO warm and ask for confirmation
if date_created is not None: if date_created is not None:
# TODO warm and ask for confirmation
return date_created return date_created
elif metadata['date_modified'] is not None: elif metadata['date_modified'] is not None:
return metadata['date_modified'] return metadata['date_modified']

4
dozo/media/media.py
View File

@ -352,9 +352,9 @@ class Media():
try: try:
if(key in exif): if(key in exif):
# correct nasty formated date # correct nasty formated date
regex = re.compile('(\d{4}):(\d{2}):(\d{2})') regex = re.compile(r'(\d{4}):(\d{2}):(\d{2})')
if(re.match(regex , exif[key]) is not None): # noqa if(re.match(regex , exif[key]) is not None): # noqa
exif[key] = re.sub(regex ,'\g<1>-\g<2>-\g<3>',exif[key]) exif[key] = re.sub(regex , r'\g<1>-\g<2>-\g<3>', exif[key])
return parse(exif[key]) return parse(exif[key])
# if(re.match('\d{4}(-|:)\d{2}(-|:)\d{2}', exif[key]) is not None): # noqa # if(re.match('\d{4}(-|:)\d{2}(-|:)\d{2}', exif[key]) is not None): # noqa
# dt, tm = exif[key].split(' ') # dt, tm = exif[key].split(' ')