tris/recoll
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

rclaudio: more fixes: complicated because the different file type handlers (e.g. flac, mp3) return data in different types.

Browse Source
This commit is contained in:
Jean-Francois Dockes 2018-06-21 12:30:43 +02:00
parent a012b831fa
commit f0bedb2201
1 changed files with 21 additions and 18 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

39
src/filters/rclaudio
View File

@ -164,6 +164,13 @@ tagdict = {
'\xa9wrt' : 'COMPOSER', '\xa9wrt' : 'COMPOSER',
} }
def tobytes(s):
if type(s) == type(b''):
return s
if type(s) != type(u''):
s = str(s)
return s.encode('utf-8', errors='replace')
# mp3: album, title, artist, genre, date, tracknumber # mp3: album, title, artist, genre, date, tracknumber
# flac: album, title, artist, genre, xxx, tracknumber # flac: album, title, artist, genre, xxx, tracknumber
# oggvorbis:album, title, artist, genre, date, tracknumber # oggvorbis:album, title, artist, genre, date, tracknumber
@ -276,30 +283,29 @@ class AudioTagExtractor:
minf['bits_per_sample'] = 16 minf['bits_per_sample'] = 16
for tag,val in minf.items(): for tag,val in minf.items():
val = str(val) minf[tag] = tobytes(val)
if type(val) == type(u''):
val = val.encode('utf-8', errors='replace')
minf[tag] = val
#################### ####################
# Metadata tags. The names vary depending on the file type. We # Metadata tags. The names vary depending on the file type. We
# just have a big translation dictionary for all # just have a big translation dictionary for all
for tag,val in mutf.items(): for tag,val in mutf.items():
#self.em.rclog("Original tag: <%s>, val <%s>" % (tag, val))
if tag.upper() in tagdict: if tag.upper() in tagdict:
tag = tag.upper() tag = tag.upper()
if tag in tagdict: if tag in tagdict:
#self.em.rclog("Original tag: <%s>, type0 %s val <%s>" %
# (tag, type(val), val))
# Some file types return lists of value (e.g. FLAC)
try:
val = " ".join(val)
#self.em.rclog("Joined tag: <%s>, type0 %s val <%s>" %
# (tag, type(val), val))
except:
pass
ntag = tagdict[tag].lower() ntag = tagdict[tag].lower()
#self.em.rclog("New tag: %s" % ntag) #self.em.rclog("New tag: %s" % ntag)
try: try:
if type(val) == type(u''): minf[ntag] = tobytes(val)
val = val.encode('utf-8', errors='replace') #self.em.rclog("Tag %s -> %s" % (ntag, val))
elif type(val) != type(b''):
val = str(val)
if type(val) == type(u''):
val = val.encode('utf-8', errors='replace')
minf[ntag] = val
#self.em.rclog("Tag %s -> %s" % (ntag, val0))
except Exception as err: except Exception as err:
self.em.rclog("Error while extracting tag: %s"%err) self.em.rclog("Error while extracting tag: %s"%err)
else: else:
@ -340,10 +346,7 @@ class AudioTagExtractor:
embdimg = self._embeddedImageFormat(mutf) embdimg = self._embeddedImageFormat(mutf)
if embdimg: if embdimg:
#self.em.rclog("Embedded image format: %s" % embdimg) #self.em.rclog("Embedded image format: %s" % embdimg)
if type(embdimg) == type(u''): embdimg = tobytes(embdimg)
minf["embdimg"] = embdimg.encode('ascii', errors='replace')
else:
minf["embdimg"] = embdimg
self.em.setmimetype("text/plain") self.em.setmimetype("text/plain")
self.em.setfield("charset", 'utf-8') self.em.setfield("charset", 'utf-8')
@ -356,7 +359,7 @@ class AudioTagExtractor:
self.em.setfield('author', val) self.em.setfield('author', val)
try: try:
docdata = mutf.pprint().encode('utf-8', errors='replace') docdata = tobytes(mutf.pprint())
except Exception as err: except Exception as err:
self.em.rclog("Doc pprint error: %s" % err) self.em.rclog("Doc pprint error: %s" % err)