Cutlery/calibre
1
0
Fork 0
mirror of https://github.com/kovidgoyal/calibre.git synced 2025-07-31 14:33:54 -04:00
Code Issues Packages Projects Releases Wiki Activity

PDF metadata: When both Info and XMP metadata present resolve which one ot use based on modification timestamps

Browse Source
This commit is contained in:
Kovid Goyal 2014-02-18 10:52:52 +05:30
parent 0ff835c4e6
commit 9d7a1f8417
2 changed files with 29 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
src/calibre/ebooks/metadata/pdf.py
View File

@ -134,7 +134,7 @@ def get_metadata(stream, cover=True):
if 'xmp_metadata' in info: if 'xmp_metadata' in info:
from calibre.ebooks.metadata.xmp import consolidate_metadata from calibre.ebooks.metadata.xmp import consolidate_metadata
mi = consolidate_metadata(mi, info['xmp_metadata']) mi = consolidate_metadata(mi, info)
# Look for recognizable identifiers in the info dict, if they were not # Look for recognizable identifiers in the info dict, if they were not
# found in the XMP metadata # found in the XMP metadata

32
src/calibre/ebooks/metadata/xmp.py
View File

@ -226,6 +226,12 @@ def metadata_from_xmp_packet(raw_bytes):
bkp = first_simple('//xmp:CreatorTool', root) bkp = first_simple('//xmp:CreatorTool', root)
if bkp: if bkp:
mi.book_producer = bkp mi.book_producer = bkp
md = first_simple('//xmp:MetadataDate', root)
if md:
try:
mi.metadata_date = parse_date(md)
except:
pass
rating = first_simple('//calibre:rating', root) rating = first_simple('//calibre:rating', root)
if rating is not None: if rating is not None:
try: try:
@ -289,15 +295,33 @@ def metadata_from_xmp_packet(raw_bytes):
return mi return mi
def consolidate_metadata(info_mi, xmp_packet): def consolidate_metadata(info_mi, info):
' When both the PDF Info dict and XMP metadata are present, prefer the xmp metadata ' ''' When both the PDF Info dict and XMP metadata are present, prefer the xmp
metadata unless the Info ModDate is never than the XMP MetadataDate. This
is the algorithm recommended by the PDF spec. '''
try: try:
xmp_mi = metadata_from_xmp_packet(xmp_packet) xmp_mi = metadata_from_xmp_packet(info['xmp_metadata'])
except: except:
import traceback import traceback
traceback.print_exc() traceback.print_exc()
return info_mi
info_title, info_authors, info_tags = info_mi.title or _('Unknown'), list(info_mi.authors or ()), list(info_mi.tags or ())
info_mi.smart_update(xmp_mi, replace_metadata=True)
prefer_info = False
if 'ModDate' in info and hasattr(xmp_mi, 'metadata_date'):
try:
info_date = parse_date(info['ModDate'])
except:
pass
else:
prefer_info = info_date > xmp_mi.metadata_date
if prefer_info:
info_mi.title, info_mi.authors, info_mi.tags = info_title, info_authors, info_tags
else: else:
info_mi.smart_update(xmp_mi, replace_metadata=True) # We'll use the xmp tags/authors but fallback to the info ones if the
# xmp does not have tags/authors. smart_update() should have taken care of
# the rest
info_mi.authors, info_mi.tags = xmp_mi.authors or info_mi.authors, xmp_mi.tags or info_mi.tags
return info_mi return info_mi
def nsmap(*args): def nsmap(*args):