def handle_companion_file(fname, dstfname=None, book=None):
src = os.path.join(path(download_cache).abspath(), fname)
if dstfname is None:
dstfname = fname
dst = os.path.join(path(static_folder).abspath(), dstfname)
# optimization based on mime/extension
if path(fname).ext in ('.png', '.jpg', '.jpeg', '.gif'):
copy_from_cache(src, dst)
optimize_image(path_for_cmd(dst))
elif path(fname).ext == '.epub':
tmp_epub = tempfile.NamedTemporaryFile(suffix='.epub',
dir=TMP_FOLDER)
tmp_epub.close()
optimize_epub(src, tmp_epub.name)
path(tmp_epub.name).move(dst)
else:
# excludes files created by Windows Explorer
if src.endswith('_Thumbs.db'):
return
# copy otherwise (PDF mostly)
logger.debug("\t\tshitty ext: {}".format(dst))
copy_from_cache(src, dst)
def handle_companion_file(fname, dstfname=None, book=None):
src = os.path.join(path(download_cache).abspath(), fname)
if dstfname is None:
dstfname = fname
dst = os.path.join(path(static_folder).abspath(), dstfname)
# optimization based on mime/extension
if path(fname).ext in ('.png', '.jpg', '.jpeg', '.gif'):
copy_from_cache(src, dst)
optimize_image(path_for_cmd(dst))
elif path(fname).ext == '.epub':
tmp_epub = tempfile.NamedTemporaryFile(suffix='.epub',
dir=TMP_FOLDER)
tmp_epub.close()
optimize_epub(src, tmp_epub.name)
path(tmp_epub.name).move(dst)
else:
# excludes files created by Windows Explorer
if src.endswith('_Thumbs.db'):
return
# copy otherwise (PDF mostly)
logger.debug("\t\tshitty ext: {}".format(dst))
copy_from_cache(src, dst)
def optimize_epub(src, dst):
logger.info("\t\tCreating ePUB at {}".format(dst))
zipped_files = []
# create temp directory to extract to
tmpd = tempfile.mkdtemp(dir=TMP_FOLDER)
with zipfile.ZipFile(src, 'r') as zf:
zipped_files = zf.namelist()
zf.extractall(tmpd)
remove_cover = False
for fname in zipped_files:
fnp = os.path.join(tmpd, fname)
if path(fname).ext in ('.png', '.jpeg', '.jpg', '.gif'):
# special case to remove ugly cover
if fname.endswith('cover.jpg') and is_bad_cover(fnp):
zipped_files.remove(fname)
remove_cover = True
else:
optimize_image(path_for_cmd(fnp))
if path(fname).ext in ('.htm', '.html'):
f = open(fnp, 'r')
html = update_html_for_static(book=book,
html_content=f.read(),
epub=True)
f.close()
with open(fnp, 'w') as f:
f.write(html)
if path(fname).ext == '.ncx':
pattern = "*** START: FULL LICENSE ***"
f = open(fnp, 'r')
ncx = f.read()
f.close()
soup = BeautifulSoup(ncx, ["lxml", "xml"])
for tag in soup.findAll('text'):
if pattern in tag.text:
s = tag.parent.parent
s.decompose()
for s in s.next_siblings:
s.decompose()
s.next_sibling
with open(fnp, 'w') as f:
f.write(soup.encode())
# delete {id}/cover.jpg if exist and update {id}/content.opf
if remove_cover:
# remove cover
path(os.path.join(tmpd, str(book.id), 'cover.jpg')).unlink_p()
soup = None
opff = os.path.join(tmpd, str(book.id), 'content.opf')
if os.path.exists(opff):
with open(opff, 'r') as fd:
soup = BeautifulSoup(fd.read(), ["lxml", "xml"])
for elem in soup.findAll():
if getattr(elem, 'attrs', {}).get('href') == 'cover.jpg':
elem.decompose()
with (open(opff, 'w')) as fd:
fd.write(soup.encode())
with cd(tmpd):
exec_cmd('zip -q0X "{dst}" mimetype'.format(dst=path_for_cmd(dst)))
exec_cmd('zip -qXr9D "{dst}" {files}'.format(
dst=path_for_cmd(dst),
files=" ".join(
[f for f in zipped_files if not f == 'mimetype'])))
path(tmpd).rmtree_p()
def optimize_epub(src, dst):
logger.info("\t\tCreating ePUB at {}".format(dst))
zipped_files = []
# create temp directory to extract to
tmpd = tempfile.mkdtemp(dir=TMP_FOLDER)
with zipfile.ZipFile(src, 'r') as zf:
zipped_files = zf.namelist()
zf.extractall(tmpd)
remove_cover = False
for fname in zipped_files:
fnp = os.path.join(tmpd, fname)
if path(fname).ext in ('.png', '.jpeg', '.jpg', '.gif'):
# special case to remove ugly cover
if fname.endswith('cover.jpg') and is_bad_cover(fnp):
zipped_files.remove(fname)
remove_cover = True
else:
optimize_image(path_for_cmd(fnp))
if path(fname).ext in ('.htm', '.html'):
f = open(fnp, 'r')
html = update_html_for_static(book=book,
html_content=f.read(),
epub=True)
f.close()
with open(fnp, 'w') as f:
f.write(html)
if path(fname).ext == '.ncx':
pattern = "*** START: FULL LICENSE ***"
f = open(fnp, 'r')
ncx = f.read()
f.close()
soup = BeautifulSoup(ncx, ["lxml", "xml"])
for tag in soup.findAll('text'):
if pattern in tag.text:
s = tag.parent.parent
s.decompose()
for s in s.next_siblings:
s.decompose()
s.next_sibling
with open(fnp, 'w') as f:
f.write(soup.encode())
# delete {id}/cover.jpg if exist and update {id}/content.opf
if remove_cover:
# remove cover
path(os.path.join(tmpd, str(book.id), 'cover.jpg')).unlink_p()
soup = None
opff = os.path.join(tmpd, str(book.id), 'content.opf')
if os.path.exists(opff):
with open(opff, 'r') as fd:
soup = BeautifulSoup(fd.read(), ["lxml", "xml"])
for elem in soup.findAll():
if getattr(elem, 'attrs', {}).get('href') == 'cover.jpg':
elem.decompose()
with(open(opff, 'w')) as fd:
fd.write(soup.encode())
with cd(tmpd):
exec_cmd('zip -q0X "{dst}" mimetype'.format(dst=path_for_cmd(dst)))
exec_cmd('zip -qXr9D "{dst}" {files}'
.format(dst=path_for_cmd(dst),
files=" ".join([f for f in zipped_files
if not f == 'mimetype'])))
path(tmpd).rmtree_p()
