def encode_media(process_order_function=PROCESS_ORDER_FUNCS[DEFAULT_ORDER_FUNC], **kwargs):
meta_manager = MetaManagerExtended(**kwargs) # path_meta=kwargs['path_meta'], path_source=kwargs['path_source']
meta_manager.load_all()
encoder = Encoder(meta_manager, **kwargs)
# In the full system, encode will probably be driven from a rabitmq endpoint.
# For testing locally we are monitoring the 'pendings_actions' list
for name in progress_bar(
process_order_function(
m.name
for m in meta_manager.meta.values()
if PENDING_ACTION["encode"] in m.pending_actions or not m.source_hashs
# (
#'AKB0048 Next Stage - ED1 - Kono Namida wo Kimi ni Sasagu',
#'Cuticle Tantei Inaba - OP - Haruka Nichijou no Naka de',
#'Gosick - ED2 - Unity (full length)',
#'Ikimonogakari - Sakura', # Takes 2 hours to encode
#'Frozen Japanise (find real name)' # took too long to process
# 'Parasite Eve - Somnia Memorias', # Non unicode characterset
# 'Akira Yamaoka - Día de los Muertos', # Non unicode characterset
# 'Higurashi no Naku koro ni - ED - why or why not (full length)', # When subs import from SSA they have styling information still attached
# 'Gatekeepers - OP - For the Smiles of Tomorrow.avi', # It's buggered. Looks like it's trying to containerize subs in a txt file?
# 'Get Backers - ED2 - Namida no Hurricane', # It's just f****d
# 'Nana (anime) - OP - Rose', # SSA's have malformed unicode characters
# 'Lunar Silver Star Story - OP - Wings (Japanese Version)',
# 'Evangleion ED - Fly Me To The Moon', # Odd dimensions and needs to be normalised
# 'Ranma Half OP1 - Jajauma ni Sasenaide',
# 'Tamako Market - OP - Dramatic Market Ride',
# 'Fullmetal Alchemist - OP1 - Melissa', # Exhibits high bitrate pausing at end
# 'Samurai Champloo - OP - Battlecry', # Missing title sub with newline
# 'KAT-TUN Your side [Instrumental]',
)
):
encoder.encode(name)
def encode_media(process_order_function=PROCESS_ORDER_FUNCS[DEFAULT_ORDER_FUNC] , **kwargs):
meta = MetaManager(kwargs['path_meta'])
meta.load_all()
encoder = Encoder(meta, **kwargs)
# In the full system, encode will probably be driven from a rabitmq endpoint.
# For testing locally we are monitoring the 'pendings_actions' list
for name in progress_bar(process_order_function(
m.name for m in meta.meta.values()
if PENDING_ACTION['encode'] in m.pending_actions or not m.source_hash
#(
#'Lunar Silver Star Story - OP - Wings (Japanese Version)',
#'Evangleion ED - Fly Me To The Moon',
# 'AKB0048 Next Stage - ED1 - Kono Namida wo Kimi ni Sasagu',
#'Cuticle Tantei Inaba - OP - Haruka Nichijou no Naka de',
#'Gosick - ED2 - Unity (full length)',
#'Ranma Half OP1 - Jajauma ni Sasenaide',
#'Tamako Market - OP - Dramatic Market Ride',
#'Fullmetal Alchemist - OP1 - Melissa', # Exhibits high bitrate pausing at end
#'Samurai Champloo - OP - Battlecry', # Missing title sub with newline
#'KAT-TUN Your side [Instrumental]',
#)
)):
encoder.encode(name)
def import_media(**kwargs):
"""
"""
stats = dict(meta_set=set(), meta_imported=set(), meta_unprocessed=set(), db_removed=list(), missing_processed_deleted=set(), missing_processed_aborted=set(), db_start=set(), meta_hash_matched_db_hash=set())
def get_db_track_names():
return set(t.source_filename for t in DBSession.query(Track.source_filename))
meta_manager = MetaManagerExtended(**kwargs)
importer = TrackImporter(meta_manager=meta_manager)
stats['db_start'] = get_db_track_names()
meta_manager.load_all() # mtime=epoc(last_update())
meta_processed_track_ids = set(meta_manager.source_hashs)
stats['meta_set'] = set(m.name for m in meta_manager.meta_items if m.source_hash)
for name in progress_bar(meta_manager.meta.keys()):
try:
if importer.import_track(name):
stats['meta_imported'].add(name)
else:
stats['meta_hash_matched_db_hash'].add(name)
except TrackNotProcesedException:
log.debug('Unprocessed (no source_hash): %s', name)
stats['meta_unprocessed'].add(name)
except TrackMissingProcessedFiles as ex:
if ex.id:
log.warning('Missing (processed files) delete existing: %s', name)
delete_track(ex.id)
commit()
stats['missing_processed_deleted'].add(name)
else:
log.warning('Missing (processed files) abort import: %s', name)
stats['missing_processed_aborted'].add(name)
for unneeded_track_id in importer.exisiting_track_ids - meta_processed_track_ids:
log.warning('Remove: %s', unneeded_track_id)
stats['db_removed'].append(DBSession.query(Track).get(unneeded_track_id).source_filename or unneeded_track_id)
delete_track(unneeded_track_id)
commit()
stats['db_end'] = get_db_track_names()
#assert stats['db_end'] == stats['meta_hash_matched_db_hash'] | stats['meta_imported'] # TODO! Reinstate this
return stats
def encode_media(
process_order_function=PROCESS_ORDER_FUNCS[DEFAULT_ORDER_FUNC],
**kwargs):
meta_manager = MetaManagerExtended(
**kwargs
) #path_meta=kwargs['path_meta'], path_source=kwargs['path_source']
meta_manager.load_all()
encoder = Encoder(meta_manager, **kwargs)
# In the full system, encode will probably be driven from a rabitmq endpoint.
# For testing locally we are monitoring the 'pendings_actions' list
for name in progress_bar(
process_order_function(
m.name for m in meta_manager.meta.values()
if PENDING_ACTION['encode'] in m.pending_actions
or not m.source_hashs
#(
#'AKB0048 Next Stage - ED1 - Kono Namida wo Kimi ni Sasagu',
#'Cuticle Tantei Inaba - OP - Haruka Nichijou no Naka de',
#'Gosick - ED2 - Unity (full length)',
#'Ikimonogakari - Sakura', # Takes 2 hours to encode
#'Frozen Japanise (find real name)' # took too long to process
# 'Parasite Eve - Somnia Memorias', # Non unicode characterset
# 'Akira Yamaoka - Día de los Muertos', # Non unicode characterset
# 'Higurashi no Naku koro ni - ED - why or why not (full length)', # When subs import from SSA they have styling information still attached
# 'Gatekeepers - OP - For the Smiles of Tomorrow.avi', # It's buggered. Looks like it's trying to containerize subs in a txt file?
# 'Get Backers - ED2 - Namida no Hurricane', # It's just f****d
# 'Nana (anime) - OP - Rose', # SSA's have malformed unicode characters
# 'Lunar Silver Star Story - OP - Wings (Japanese Version)',
# 'Evangleion ED - Fly Me To The Moon', # Odd dimensions and needs to be normalised
# 'Ranma Half OP1 - Jajauma ni Sasenaide',
# 'Tamako Market - OP - Dramatic Market Ride',
# 'Fullmetal Alchemist - OP1 - Melissa', # Exhibits high bitrate pausing at end
# 'Samurai Champloo - OP - Battlecry', # Missing title sub with newline
# 'KAT-TUN Your side [Instrumental]',
)):
encoder.encode(name)
def import_media(**kwargs):
"""
"""
stats = dict(meta_set=set(),
meta_imported=set(),
meta_unprocessed=set(),
db_removed=list(),
missing_processed_deleted=set(),
missing_processed_aborted=set(),
db_start=set(),
meta_hash_matched_db_hash=set())
track_api = partial(_track_api, kwargs['api_host'])
meta_manager = MetaManagerExtended(**kwargs)
meta_manager.load_all() # mtime=epoc(last_update())
processed_track_ids = set(meta_manager.source_hashs)
processed_files_lookup = set(
f.relative
for f in fast_scan(meta_manager.processed_files_manager.path))
existing_tracks = track_api()['data']['tracks']
existing_track_ids = existing_tracks.keys()
generate_track_dict = partial(
_generate_track_dict,
meta_manager=meta_manager,
existing_track_ids=existing_track_ids,
processed_files_lookup=processed_files_lookup)
stats['db_start'] = set(existing_tracks.values())
stats['meta_set'] = set(m.name for m in meta_manager.meta_items
if m.source_hash)
tracks_to_add = []
track_ids_to_delete = []
log.info(
f'Importing tracks - Existing:{len(existing_track_ids)} Processed{len(processed_track_ids)}'
)
for name in progress_bar(meta_manager.meta.keys()):
try:
track = generate_track_dict(name)
if track:
stats['meta_imported'].add(name)
#tracks_to_add.append(track)
track_api([track], method='POST')
else:
stats['meta_hash_matched_db_hash'].add(name)
except TrackNotProcesedException:
log.debug('Unprocessed (no source_hash): %s', name)
stats['meta_unprocessed'].add(name)
except TrackMissingProcessedFiles as ex:
if ex.id:
log.warning('Missing (processed files) delete existing: %s',
name)
track_ids_to_delete.append(ex.id)
stats['missing_processed_deleted'].add(name)
else:
log.warning('Missing (processed files) abort import: %s', name)
stats['missing_processed_aborted'].add(name)
for unneeded_track_id in existing_track_ids - processed_track_ids:
log.warning('Remove: %s', unneeded_track_id)
stats['db_removed'].append(existing_tracks[unneeded_track_id])
track_ids_to_delete.append(unneeded_track_id)
log.info(
f"""{kwargs['api_host']} -> Add:{len(tracks_to_add)} Delete:{len(track_ids_to_delete)}"""
)
#track_api(tracks_to_add, method='POST')
track_api(track_ids_to_delete, method='DELETE')
stats['db_end'] = track_api()['data']['tracks'].values()
#assert stats['db_end'] == stats['meta_hash_matched_db_hash'] | stats['meta_imported'] # TODO! Reinstate this
return stats
def scan_media(**kwargs):
"""
"""
meta = MetaManager(kwargs['path_meta'])
meta.load_all()
log.info('1.) Read file structure into memory')
folder_structure = FolderStructure.factory(
path=kwargs['path_source'],
search_filter=fast_scan_regex_filter(
file_regex=file_extension_regex(ALL_EXTS),
ignore_regex=DEFAULT_IGNORE_FILE_REGEX,
)
)
log.info('2.) Locate primary files')
# Note: Duplicate media is completely ignored/removed in this list
primary_files = locate_primary_files(folder_structure, file_regex=file_extension_regex(PRIMARY_FILE_RANKED_EXTS))
log.info("3.) Find associated files as a 'file collection' (based on the name of the primary file)")
file_collections = {
f.file_no_ext: get_file_collection(folder_structure, f)
for f in primary_files
}
log.info('4.) Associate file_collections with existing metadata objects')
for name, file_collection in progress_bar(file_collections.items()):
meta.load(name)
m = meta.get(name)
for f in file_collection:
m.associate_file(f)
meta.save(name)
log.info('5.) Attempt to find associate unassociated files but finding them on the folder_structure in memory')
# TODO: The step in 5b (mtime and hash check) is not sufficent
# We need to identify all unmatched files and generate a hash lookup and match them that way.
# It may take longer but it's the only true way of matching a file.
# It might in some cases be quicker because we don't need to craw the entire file structure in memory.
# These are meta items that have a filecollection matched,
# but that file collection is incomplete, so we have some child files missing
has_unassociated_files = operator.attrgetter('unassociated_files')
for m in filter(has_unassociated_files, meta.meta.values()):
for filename, scan_data in m.unassociated_files.items():
# 5a.) The unassociated file may not have been found in the inital collection scan,
# check it's original location and associate if it exists
f = folder_structure.get(scan_data.get('relative')) if scan_data.get('relative') else None
if f:
m.associate_file(f)
log.warning('Associating found missing file %s to %s - this should not be a regular occurance, move/rename this so it is grouped effectivly', f.relative, m.name)
continue
# 5b.) Search the whole folder_structure in memory for a matching hash
mtime = scan_data['mtime']
for f in folder_structure.scan(
lambda f:
not IGNORE_SEARCH_EXTS_REGEX.search(f.file)
and
(f.file == filename or f.stats.st_mtime == mtime) # TODO: Depricate this!
and
str(f.hash) == scan_data['hash']
):
log.warning('Associating found missing file %s to %s - this should not be a regular occurance, move/rename this so it is grouped effectivly', f.relative, m.name)
m.associate_file(f)
break
# 5c.)
# We have done our best at locating missing files
# Remove them from the tracked list of files.
m.unlink_unassociated_files()
log.info('6.) Remove unmatched meta entrys')
for m in [m for m in meta.meta.values() if not m.file_collection]:
log.info('Removing meta %s', m.name)
meta.delete(m.name)
# (If processed data already exisits, it will be relinked at the encode level)
meta.save_all()
def scan_media(**kwargs):
"""
"""
meta = MetaManager(kwargs['path_meta'])
meta.load_all()
log.info('1.) Read file structure into memory')
folder_structure = FolderStructure.factory(
path=kwargs['path_source'],
search_filter=fast_scan_regex_filter(
file_regex=file_extension_regex(ALL_EXTS),
ignore_regex=DEFAULT_IGNORE_FILE_REGEX,
))
log.info('2.) Locate primary files')
# Note: Duplicate media is completely ignored/removed in this list
primary_files = locate_primary_files(
folder_structure,
file_regex=file_extension_regex(PRIMARY_FILE_RANKED_EXTS))
log.info(
"3.) Find associated files as a 'file collection' (based on the name of the primary file)"
)
file_collections = {
f.file_no_ext: get_file_collection(folder_structure, f)
for f in primary_files
}
log.info('4.) Associate file_collections with existing metadata objects')
for name, file_collection in progress_bar(file_collections.items()):
meta.load(name)
m = meta.get(name)
for f in file_collection:
m.associate_file(f)
meta.save(name)
log.info(
'5.) Attempt to find associate unassociated files but finding them on the folder_structure in memory'
)
# TODO: The step in 5b (mtime and hash check) is not sufficent
# We need to identify all unmatched files and generate a hash lookup and match them that way.
# It may take longer but it's the only true way of matching a file.
# It might in some cases be quicker because we don't need to craw the entire file structure in memory.
# These are meta items that have a filecollection matched,
# but that file collection is incomplete, so we have some child files missing
has_unassociated_files = operator.attrgetter('unassociated_files')
for m in filter(has_unassociated_files, meta.meta.values()):
for filename, scan_data in m.unassociated_files.items():
# 5a.) The unassociated file may not have been found in the inital collection scan,
# check it's original location and associate if it exists
f = folder_structure.get(scan_data.get(
'relative')) if scan_data.get('relative') else None
if f:
m.associate_file(f)
log.warning(
'Associating found missing file %s to %s - this should not be a regular occurance, move/rename this so it is grouped effectivly',
f.relative, m.name)
continue
# 5b.) Search the whole folder_structure in memory for a matching hash
mtime = scan_data['mtime']
for f in folder_structure.scan(
lambda f: not IGNORE_SEARCH_EXTS_REGEX.search(f.file) and
(f.file == filename or f.stats.st_mtime == mtime
) # TODO: Depricate this!
and str(f.hash) == scan_data['hash']):
log.warning(
'Associating found missing file %s to %s - this should not be a regular occurance, move/rename this so it is grouped effectivly',
f.relative, m.name)
m.associate_file(f)
break
# 5c.)
# We have done our best at locating missing files
# Remove them from the tracked list of files.
m.unlink_unassociated_files()
log.info('6.) Remove unmatched meta entrys')
for m in [m for m in meta.meta.values() if not m.file_collection]:
log.info('Removing meta %s', m.name)
meta.delete(m.name)
# (If processed data already exisits, it will be relinked at the encode level)
meta.save_all()
def import_media(**kwargs):
"""
"""
stats = dict(meta_set=set(), meta_imported=set(), meta_unprocessed=set(), db_removed=list(), missing_processed_deleted=set(), missing_processed_aborted=set(), db_start=set(), meta_hash_matched_db_hash=set())
track_api = partial(_track_api, kwargs['api_host'])
meta_manager = MetaManagerExtended(**kwargs)
meta_manager.load_all() # mtime=epoc(last_update())
processed_track_ids = set(meta_manager.source_hashs)
processed_files_lookup = set(f.relative for f in fast_scan(meta_manager.processed_files_manager.path))
existing_tracks = track_api()['data']['tracks']
existing_track_ids = existing_tracks.keys()
generate_track_dict = partial(_generate_track_dict, meta_manager=meta_manager, existing_track_ids=existing_track_ids, processed_files_lookup=processed_files_lookup)
stats['db_start'] = set(existing_tracks.values())
stats['meta_set'] = set(m.name for m in meta_manager.meta_items if m.source_hash)
tracks_to_add = []
track_ids_to_delete = []
log.info('Importing tracks - Existing:{} Processed:{}'.format(len(existing_track_ids), len(processed_track_ids))) # TODO: replace with formatstring
for name in progress_bar(meta_manager.meta.keys()):
try:
track = generate_track_dict(name)
if track:
stats['meta_imported'].add(name)
#tracks_to_add.append(track)
track_api([track], method='POST')
else:
stats['meta_hash_matched_db_hash'].add(name)
except TrackNotProcesedException:
log.debug('Unprocessed (no source_hash): %s', name)
stats['meta_unprocessed'].add(name)
except TrackMissingProcessedFiles as ex:
if ex.id:
log.warning('Missing (processed files) delete existing: %s', name)
track_ids_to_delete.append(ex.id)
stats['missing_processed_deleted'].add(name)
else:
log.warning('Missing (processed files) abort import: %s', name)
stats['missing_processed_aborted'].add(name)
for unneeded_track_id in existing_track_ids - processed_track_ids:
log.warning('Remove: %s', unneeded_track_id)
stats['db_removed'].append(existing_tracks[unneeded_track_id])
track_ids_to_delete.append(unneeded_track_id)
log.info("""{api_host} -> Add:{add_count} Delete:{delete_count}""".format(
api_host=kwargs['api_host'],
add_count=len(tracks_to_add),
delete_count=len(track_ids_to_delete),
)) # TODO: replace with formatstring
#track_api(tracks_to_add, method='POST')
track_api(track_ids_to_delete, method='DELETE')
stats['db_end'] = track_api()['data']['tracks'].values()
#assert stats['db_end'] == stats['meta_hash_matched_db_hash'] | stats['meta_imported'] # TODO! Reinstate this
return stats