pestctrl/guess-filename.py
1
0
Fork 0
forked from Github/guess-filename.py
Code Issues Pull requests Projects Releases Packages Wiki Activity

added MEDIATHEKVIEW_SHORT_REGEX and manual/interactive fall-back handling for truncated ORF file names

Browse source
This commit is contained in:
Karl Voit 2018-06-09 12:38:52 +02:00
parent 83f62df07d
commit 9a9fac31d5
1 changed files with 121 additions and 12 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

133
guessfilename/__init__.py
View file

@ -1,6 +1,6 @@
#!/usr/bin/env python3
# -*- coding: utf-8 -*-
PROG_VERSION = u"Time-stamp: <2018-05-31 12:11:00 vk>"
PROG_VERSION = u"Time-stamp: <2018-06-09 14:56:05 vk>"
# TODO:
@ -165,11 +165,35 @@ class GuessFilename(object):
# 20180521T193000 ORF - ZIB 1 - Parlament bereitet sich auf EU-Vorsitz vor -ORIGINAL- 2018-05-21_1930_tl_02_ZIB-1_Parlament-berei__13977453__o__277886215b__s14303762_2__WEB03HD_19350304P_19371319P_Q4A.mp4
# 20180522T220000 ORF - Willkommen Österreich mit Stermann & Grissemann - Beachvolleyball-Duo Clemens Doppler und Alexander Horst -ORIGINAL- 2018-05-22_2200_l__13977514__o__1745__s14352_2__BCK1HD_22394018P_22.mp4
# -ORIGINAL- 2018-05-22_2200_l__13977514__o__1745__s14352_2__BCK1HD_22394018P_22.mp4
MEDIATHEKVIEW_LONG_REGEX = re.compile(DATESTAMP_REGEX + 'T?' + TIMESTAMP_REGEX + # e.g., "20180510T090000"
' (.+) - (.+) - (.+) -ORIGINAL- ' + # e.g., " ORF - ZIB - Signation -ORIGINAL- "
'.+__o__(\d+b?)__s(\d+)_' + # e.g., "2018-05-10_0900_tl_02_ZIB-9-00_Signation__13976423__o__1368225677__s14297692"
# SHORT_REGEX: if MediathekView is NOT able to generate the full length file name because
# of file name length restrictions, this RegEx is a fall-back in order to
# recognize the situation.
MEDIATHEKVIEW_SHORT_REGEX_STRING = DATESTAMP_REGEX + 'T?' + TIMESTAMP_REGEX + ' (.+) - (.+) - (.+) -ORIGINAL- ' # e.g., "20180510T090000 ORF - ZIB - Signation -ORIGINAL- "
MEDIATHEKVIEW_SHORT_REGEX = re.compile(MEDIATHEKVIEW_SHORT_REGEX_STRING + '(.+).mp4')
# URL has format like: http://apasfpd.apa.at/cms-worldwide/online/7db1010b02753288e65ff61d5e1dff58/1528531468/2018-06-08_2140_tl_01_Was-gibt-es-Neu_Promifrage-gest__13979244__o__1391278651__s14313058_8__BCK1HD_22050122P_22091314P_Q4A.mp4
# but with varying quality indicator: Q4A (low), Q6A (high), Q8C (HD)
# which gets parsed like:
# http://apasfpd.apa.at/cms-worldwide/online/ → required
# 7db1010b02753288e65ff61d5e1dff58/1528531468
# /2018-06-08_2140_tl_ → required
# 01_Was-gibt-es-Neu_Promifrage-gest__13979244__o__1391278651__s14313058_8__BCK1HD
# _22050122P_22091314P_Q4A.mp4 → required
FILM_URL_REGEX = re.compile('http://apasfpd.apa.at/cms-worldwide/online/' +
'.+' + # e.g., "7db1010b02753288e65ff61d5e1dff58/1528531468"
'/' + DATESTAMP_REGEX + '_' + TIMESTAMP_REGEX + '_tl_' + # e.g., "/2018-06-08_2140_tl_"
'.+' + # e.g., "01_Was-gibt-es-Neu_Promifrage-gest__13979244__o__1391278651__s14313058_8__BCK1HD"
'_' + TIMESTAMP_REGEX + '\d\dP_' + TIMESTAMP_REGEX + '\d\dP_' + # e.g., "_22050122P_22091314P_"
'(Q4A|Q6A|Q8C).mp4') # "Q4A.mp4" or "Q6A.mp4" or "Q8C.mp4"
# LONG_REGEX: MediathekView was able to generate the full length file name including
# the full length original file name at the end of the file name which ends
# with the quality indicator Q4A or Q8C when used with the ORF sender file format.
MEDIATHEKVIEW_LONG_REGEX = re.compile(MEDIATHEKVIEW_SHORT_REGEX_STRING +
'.+__o__(\d+b?)__s(\d+)_' + # e.g., "2018-05-10_0900_tl_02_ZIB-9-00_Signation__13976423__o__1368225677__s14297692"
'(.+_(' + TIMESTAMP_REGEX + ').+P_(' + TIMESTAMP_REGEX + ').+P_)?' + # OPTIONAL: time-stamps of chunks: "_2__WEB03HD_09000305P_09001400P"
'(Q4A|Q8C).mp4', re.UNICODE) # "Q4A.mp4" for lowquality or "Q8C.mp4" for highquality
'(Q4A|Q8C).mp4', re.UNICODE) # "Q4A.mp4" for lowquality or "Q8C.mp4" for highquality
# C112345678901EUR20150930001.pdf -> 2015-09-30 Bank Austria Kontoauszug 2017-001 12345678901.pdf
BANKAUSTRIA_BANK_STATEMENT_REGEX = re.compile('^C1(\d{11})EUR(\d{4})(\d{2})(\d{2})(\d{3}).pdf$', re.UNICODE)
@ -494,6 +518,20 @@ class GuessFilename(object):
return months[month]
def translate_ORF_quality_string_to_tag(self, quality_string):
"""
Returns a filetag which is derived from a key string. The key strings are defined
by the file names of the ORF company offering its download file names.
"""
if quality_string == 'Q4A':
return 'lowquality'
elif quality_string == 'Q6A' or quality_string == 'Q8C':
return 'highquality'
else:
return 'UNKNOWNQUALITY'
def derive_new_filename_from_old_filename(self, oldfilename):
"""
Analyses the old filename and returns a new one if feasible.
@ -534,6 +572,11 @@ class GuessFilename(object):
# 20180520T201500 ORF - Tatort - Tatort_ Aus der Tiefe der Zeit -ORIGINAL- 2018-05-20_2015_in_02_Tatort--Aus-der_____13977411__o__1151703583__s14303062_Q8C.mp4
# ('2018', '05', '20', '20', '15', '00', '00', 'ORF', 'Tatort', 'Tatort_ Aus der Tiefe der Zeit', '1151703583', '14303062', None, None, None, None, None, None, None, None, None, None, None, 'Q8C')
# -> 2018-05-20T20.15.00 ORF - Tatort - Tatort Aus der Tiefe der Zeit -- highquality.mp4
#
# LONG_REGEX: MediathekView was able to generate the full length file name including
# the full length original file name at the end of the file name which ends
# with the quality indicator Q4A or Q8C when used with the ORF sender file format.
#
regex_match = re.match(self.MEDIATHEKVIEW_LONG_REGEX, oldfilename)
if regex_match:
@ -546,13 +589,7 @@ class GuessFilename(object):
# where the files do not exist
pass
quality_string = regex_match.group(len(regex_match.groups())).upper()
if quality_string == 'Q4A':
qualitytag = 'lowquality'
elif quality_string == 'Q8C':
qualitytag = 'highquality'
else:
qualitytag = 'UNKNOWNQUALITY'
qualitytag = self.translate_ORF_quality_string_to_tag(regex_match.group(len(regex_match.groups())).upper())
if regex_match.group(13):
# the file name contained the optional chunk time-stamp(s)
@ -562,6 +599,78 @@ class GuessFilename(object):
MEDIATHEKVIEW_LONG_INDEXGROUPS = [1, '-', 2, '-', 3, 'T', 4, '.', 5, '.', 6, ' ', 8, ' - ', 9, ' - ', 10, ' -- ', qualitytag, '.mp4']
return self.build_string_via_indexgroups(regex_match, MEDIATHEKVIEW_LONG_INDEXGROUPS).replace('_', ' ')
# SHORT_REGEX: if MediathekView is NOT able to generate the full length file name because
# of file name length restrictions, this RegEx is a fall-back in order to
# recognize the situation.
# This is a fall-back mechanism which requires INTERACTIVE correction: user gets asked to
# enter the original file URL: MediathekView > context menu of a chunk > "Film-URL kopieren"
# With this URL, guessfilename is able to extract the original time-stamps that were missing
# in the SHORT_REGEX.
#
# test it manually with following data: (no unit test because of interactive input)
# filename "20180608T214000 ORF - Was gibt es Neues? - Promifrage gestellt von Helmut Bohatsch_ Wie vergewisserte sich der Bischof von New York 1877, dass das erste Tonaufnahmegerät kein Teufelswerk ist? -ORIGINAL- 2018-06-08_2140_tl_01_Was-gibt-es-Neu_Promifr.mp4"
# Low quality URL:
# http://apasfpd.apa.at/cms-worldwide/online/7db1010b02753288e65ff61d5e1dff58/1528531468/2018-06-08_2140_tl_01_Was-gibt-es-Neu_Promifrage-gest__13979244__o__1391278651__s14313058_8__BCK1HD_22050122P_22091314P_Q4A.mp4
# High quality URL:
# http://apasfpd.apa.at/cms-worldwide/online/549c11b7cf10c9a232361003d78e5335/1528531468/2018-06-08_2140_tl_01_Was-gibt-es-Neu_Promifrage-gest__13979244__o__1391278651__s14313058_8__BCK1HD_22050122P_22091314P_Q6A.mp4
# HD URL:
# http://apasfpd.apa.at/cms-worldwide/online/6ade5772382b0833525870b4a290692c/1528531468/2018-06-08_2140_tl_01_Was-gibt-es-Neu_Promifrage-gest__13979244__o__1391278651__s14313058_8__BCK1HD_22050122P_22091314P_Q8C.mp4
regex_match = re.match(self.MEDIATHEKVIEW_SHORT_REGEX, oldfilename)
if regex_match:
try:
if 'Tatort' in oldfilename and os.stat(oldfilename).st_size < 2000000000 and not options.quiet:
print(' → ' + colorama.Style.BRIGHT + colorama.Fore.RED +
'WARNING: Tatort file seems to be too small (download aborted?): ' +
oldfilename + colorama.Style.RESET_ALL)
except OSError:
# ignore this error because this only(?) happens when
# the function is called within its unittest module
# where the files do not exist
pass
logging.warn('I recognized a MediathekView file which has a cut-off time-stamp because ' +
'of file name length restrictions.\nYou can fix it manually:')
url_valid = False
while not url_valid:
film_url = input("\nPlease enter: MediathekView > context menu of the " +
"corresponding chunk > \"Film-URL kopieren\": (C-k = lowquality; C-h = HD)\n")
# URL has format like: http://apasfpd.apa.at/cms-worldwide/online/7db1010b02753288e65ff61d5e1dff58/1528531468/2018-06-08_2140_tl_01_Was-gibt-es-Neu_Promifrage-gest__13979244__o__1391278651__s14313058_8__BCK1HD_22050122P_22091314P_Q4A.mp4
# but with varying quality indicator: Q4A (low), Q6A (high), Q8C (HD)
film_regex_match = re.match(self.FILM_URL_REGEX, film_url)
if regex_match.groups()[:5] != film_regex_match.groups()[:5]:
# plausibility check fails: date and time of the chunks differ
logging.warn('Sorry, there is a mismatch of the date and time contained bewteen the filename (' +
self.build_string_via_indexgroups(regex_match, [1, '-', 2, '-', 3, 'T', 4, '.', 5]) +
') and the URL pasted (' +
self.build_string_via_indexgroups(film_regex_match, [1, '-', 2, '-', 3, 'T', 4, '.', 5]) +
'). Please try again with the correct URL ...')
elif not film_regex_match:
logging.warn('You did not enter a valid Film-URL which looks like: ' +
'http://apasfpd.apa.at/cms-worldwide/online/.../.../2018-06-08_2140_tl_01_Description__' +
'13979244__o__1391278651__s14313058_8__BCK1HD_22050122P_22091314P_Q4A.mp4')
else:
url_valid = True
# "lowquality" or "highquality" or "UNKNOWNQUALITY"
qualitytag = self.translate_ORF_quality_string_to_tag(film_regex_match.group(len(film_regex_match.groups())).upper())
# e.g., "2018-06-08T"
datestamp = self.build_string_via_indexgroups(regex_match, [1, '-', 2, '-', 3, 'T'])
# e.g., "22.05.01 "
timestamp = self.build_string_via_indexgroups(film_regex_match, [8, '.', 9, '.', 10, ' '])
# e.g., "ORF - Was gibt es Neues? - Promifrage gestellt von Helmut Bohatsch_ Wie vergewisserte sich der Bischof von New York 1877, dass das erste Tonaufnahmegerät kein Teufelswerk ist? -- lowquality.mp4"
description = self.build_string_via_indexgroups(regex_match, [8, ' - ', 9, ' - ', 10, ' -- ', qualitytag, '.mp4'])
# combining them all to one final filename:
return datestamp + timestamp + description
# OLD # # MediathekView: Settings > modify Set > Targetfilename: "%DT%d h%i %s %t - %T - %N.mp4"
# OLD # # results in files like:
# OLD # # 20161227T201500 h115421 ORF Das Sacher. In bester Gesellschaft 1.mp4