Cutlery/calibre
1
0
Fork 0
mirror of https://github.com/kovidgoyal/calibre.git synced 2025-11-19 04:53:05 -05:00
Code Issues Packages Projects Releases Wiki Activity

Only check if name is present in container once

Browse Source 
Apparently on some windows systems the check fails when run in a thread
for mysterious reasons. I cant replicate the failure, but no real harm
in this, maybe a very tiny performance decrease at worst.
This commit is contained in:
Kovid Goyal 2025-09-22 10:24:42 +05:30
parent 051c25f1fa
commit 9d363b30a8
No known key found for this signature in database
GPG Key ID: 06BC317B515ACE7C
1 changed files with 28 additions and 22 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

50
src/calibre/srv/render_book.py
View File

@ -71,7 +71,7 @@ def decode_url(x):
return decode_component(parts[0]), (parts[1] if len(parts) > 1 else '') return decode_component(parts[0]), (parts[1] if len(parts) > 1 else '')
def create_link_replacer(container, link_uid, changed): def create_link_replacer(container, link_uid, changed, present_names):
resource_template = link_uid + '|{}|' resource_template = link_uid + '|{}|'
def link_replacer(base, url): def link_replacer(base, url):
@ -94,7 +94,7 @@ def create_link_replacer(container, link_uid, changed):
url, frag = purl.path, purl.fragment url, frag = purl.path, purl.fragment
name = container.href_to_name(url, base) name = container.href_to_name(url, base)
if name: if name:
if container.has_name_and_is_not_empty(name): if name in present_names:
frag = urlunquote(frag) frag = urlunquote(frag)
url = resource_template.format(encode_url(name, frag)) url = resource_template.format(encode_url(name, frag))
else: else:
@ -258,12 +258,12 @@ def create_cover_page(container, input_fmt, is_comic, book_metadata=None):
return raster_cover_name, titlepage_name return raster_cover_name, titlepage_name
def transform_style_sheet(container, name, link_uid, virtualize_resources, virtualized_names): def transform_style_sheet(container, name, link_uid, virtualize_resources, virtualized_names, present_names):
changed = False changed = False
link_replacer = None link_replacer = None
if virtualize_resources: if virtualize_resources:
changed_names = set() changed_names = set()
link_replacer = partial(create_link_replacer(container, link_uid, changed_names), name) link_replacer = partial(create_link_replacer(container, link_uid, changed_names, present_names), name)
raw = container.raw_data(name, decode=True) raw = container.raw_data(name, decode=True)
nraw = transform_properties(raw, is_declaration=False, url_callback=link_replacer) nraw = transform_properties(raw, is_declaration=False, url_callback=link_replacer)
if virtualize_resources: if virtualize_resources:
@ -282,10 +282,10 @@ def transform_style_sheet(container, name, link_uid, virtualize_resources, virtu
f.write(raw.encode('utf-8')) f.write(raw.encode('utf-8'))
def transform_svg_image(container, name, link_uid, virtualize_resources, virtualized_names): def transform_svg_image(container, name, link_uid, virtualize_resources, virtualized_names, present_names):
if not virtualize_resources: if not virtualize_resources:
return return
link_replacer = create_link_replacer(container, link_uid, set()) link_replacer = create_link_replacer(container, link_uid, set(), present_names)
xlink = XLINK('href') xlink = XLINK('href')
altered = False altered = False
xlink_xpath = XPath('//*[@xl:href]') xlink_xpath = XPath('//*[@xl:href]')
@ -326,7 +326,7 @@ def parse_smil_time(x):
return seconds return seconds
def transform_smil(container, name, link_uid, virtualize_resources, virtualized_names, smil_map): def transform_smil(container, name, link_uid, virtualize_resources, virtualized_names, smil_map, present_names):
root = container.parsed(name) root = container.parsed(name)
text_tag, audio_tag = SMIL('text'), SMIL('audio') text_tag, audio_tag = SMIL('text'), SMIL('audio')
body_tag, seq_tag, par_tag = SMIL('body'), SMIL('seq'), SMIL('par') body_tag, seq_tag, par_tag = SMIL('body'), SMIL('seq'), SMIL('par')
@ -419,7 +419,7 @@ def transform_inline_styles(container, name, transform_sheet, transform_style):
return changed return changed
def transform_html(container, name, virtualize_resources, link_uid, link_to_map, virtualized_names): def transform_html(container, name, virtualize_resources, link_uid, link_to_map, virtualized_names, present_names):
link_xpath = XPath('//h:*[@href and (self::h:a or self::h:area)]') link_xpath = XPath('//h:*[@href and (self::h:a or self::h:area)]')
svg_link_xpath = XPath('//svg:a') svg_link_xpath = XPath('//svg:a')
img_xpath = XPath('//h:img[@src]') img_xpath = XPath('//h:img[@src]')
@ -427,7 +427,7 @@ def transform_html(container, name, virtualize_resources, link_uid, link_to_map,
res_link_xpath = XPath('//h:link[@href]') res_link_xpath = XPath('//h:link[@href]')
root = container.parsed(name) root = container.parsed(name)
changed_names = set() changed_names = set()
link_replacer = create_link_replacer(container, link_uid, changed_names) link_replacer = create_link_replacer(container, link_uid, changed_names, present_names)
# Used for viewing images # Used for viewing images
for img in img_xpath(root): for img in img_xpath(root):
@ -464,7 +464,7 @@ def transform_html(container, name, virtualize_resources, link_uid, link_to_map,
transform_inline_styles(container, name, transform_sheet=transform_sheet, transform_style=transform_declaration) transform_inline_styles(container, name, transform_sheet=transform_sheet, transform_style=transform_declaration)
if virtualize_resources: if virtualize_resources:
virtualize_html(container, name, link_uid, link_to_map, virtualized_names) virtualize_html(container, name, link_uid, link_to_map, virtualized_names, present_names)
else: else:
def handle_link(a, attr='href'): def handle_link(a, attr='href'):
@ -497,12 +497,12 @@ def transform_html(container, name, virtualize_resources, link_uid, link_to_map,
f.write(shtml) f.write(shtml)
def virtualize_html(container, name, link_uid, link_to_map, virtualized_names): def virtualize_html(container, name, link_uid, link_to_map, virtualized_names, present_names):
changed = set() changed = set()
link_xpath = XPath('//h:*[@href and (self::h:a or self::h:area)]') link_xpath = XPath('//h:*[@href and (self::h:a or self::h:area)]')
svg_link_xpath = XPath('//svg:a') svg_link_xpath = XPath('//svg:a')
link_replacer = create_link_replacer(container, link_uid, changed) link_replacer = create_link_replacer(container, link_uid, changed, present_names)
virtualized_names.add(name) virtualized_names.add(name)
root = container.parsed(name) root = container.parsed(name)
@ -542,7 +542,7 @@ def virtualize_html(container, name, link_uid, link_to_map, virtualized_names):
__smil_file_names__ = '' __smil_file_names__ = ''
def process_book_file(virtualize_resources, link_uid, container, name): def process_book_file(virtualize_resources, link_uid, container, present_names, name):
link_to_map = {} link_to_map = {}
html_data = {} html_data = {}
smil_map = {__smil_file_names__: []} smil_map = {__smil_file_names__: []}
@ -556,14 +556,14 @@ def process_book_file(virtualize_resources, link_uid, container, name):
'has_maths': check_for_maths(root), 'has_maths': check_for_maths(root),
'anchor_map': anchor_map(root) 'anchor_map': anchor_map(root)
} }
transform_html(container, name, virtualize_resources, link_uid, link_to_map, virtualized_names) transform_html(container, name, virtualize_resources, link_uid, link_to_map, virtualized_names, present_names)
elif mt in OEB_STYLES: elif mt in OEB_STYLES:
transform_style_sheet(container, name, link_uid, virtualize_resources, virtualized_names) transform_style_sheet(container, name, link_uid, virtualize_resources, virtualized_names, present_names)
elif mt == 'image/svg+xml': elif mt == 'image/svg+xml':
transform_svg_image(container, name, link_uid, virtualize_resources, virtualized_names) transform_svg_image(container, name, link_uid, virtualize_resources, virtualized_names, present_names)
elif mt in ('application/smil', 'application/smil+xml'): elif mt in ('application/smil', 'application/smil+xml'):
smil_map[__smil_file_names__].append(name) smil_map[__smil_file_names__].append(name)
transform_smil(container, name, link_uid, virtualize_resources, virtualized_names, smil_map) transform_smil(container, name, link_uid, virtualize_resources, virtualized_names, smil_map, present_names)
return link_to_map, html_data, virtualized_names, smil_map return link_to_map, html_data, virtualized_names, smil_map
@ -599,10 +599,16 @@ def process_exploded_book(
# We do not add zero byte sized files as the IndexedDB API in the # We do not add zero byte sized files as the IndexedDB API in the
# browser has no good way to distinguish between zero byte files and # browser has no good way to distinguish between zero byte files and
# load failures. # load failures.
excluded_names = { excluded_names = set()
name for name, mt in container.mime_map.items() if present_names = set()
name == container.opf_name or mt == guess_type('a.ncx') or name.startswith('META-INF/') or for name, mt in container.mime_map.items():
name == 'mimetype' or not container.has_name_and_is_not_empty(name)} if container.has_name_and_is_not_empty(name):
present_names.add(name)
if name == container.opf_name or mt == guess_type('a.ncx') or name.startswith(
'META-INF/') or name == 'mimetype':
excluded_names.add(name)
else:
excluded_names.add(name)
raster_cover_name, titlepage_name = create_cover_page(container, input_fmt.lower(), is_comic, book_metadata) raster_cover_name, titlepage_name = create_cover_page(container, input_fmt.lower(), is_comic, book_metadata)
tocobj = get_toc(container, verify_destinations=False) tocobj = get_toc(container, verify_destinations=False)
@ -647,7 +653,7 @@ def process_exploded_book(
names_that_need_work = tuple(n for n, mt in container.mime_map.items() if needs_work(mt)) names_that_need_work = tuple(n for n, mt in container.mime_map.items() if needs_work(mt))
num_workers = calculate_number_of_workers(names_that_need_work, container, max_workers) num_workers = calculate_number_of_workers(names_that_need_work, container, max_workers)
results = [] results = []
f = partial(process_book_file, virtualize_resources, book_render_data['link_uid'], container) f = partial(process_book_file, virtualize_resources, book_render_data['link_uid'], container, present_names)
if num_workers < 2: if num_workers < 2:
results.extend(map(f, names_that_need_work)) results.extend(map(f, names_that_need_work))
else: else: