github-mirrors/bazarr
1
0
Fork 0
mirror of https://github.com/morpheus65535/bazarr.git synced 2025-04-24 06:37:16 -04:00
Code Issues Projects Releases Packages Wiki Activity

Improved uppercase detection / execution in Fix Uppercase mod

Browse source
This commit is contained in:
JayZed 2025-03-20 20:40:55 -04:00 • committed by GitHub
parent cefd1638dd
commit 2a330f63dc
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
1 changed files with 30 additions and 28 deletions
Download patch file Download diff file Expand all files Collapse all files

58
custom_libs/subzero/modification/main.py
View file

@ -22,7 +22,7 @@ class SubtitleModifications(object):
language = None
initialized_mods = {}
mods_used = []
only_uppercase = False
mostly_uppercase = False
f = None
font_style_tag_start = u"{\\"
@ -118,7 +118,7 @@ class SubtitleModifications(object):
identifier, self.language)
continue
if mod_cls.only_uppercase and not self.only_uppercase:
if mod_cls.only_uppercase and not self.mostly_uppercase:
if self.debug:
logger.debug("Skipping %s, because the subtitle isn't all uppercase", identifier)
continue
@ -188,41 +188,43 @@ class SubtitleModifications(object):
return line_mods, non_line_mods, used_mods
def detect_uppercase(self):
entries_used = 0
for entry in self.f:
entry_used = False
sub = entry.text
# skip HI bracket entries, those might actually be lowercase
sub = sub.strip()
for processor in registry.mods["remove_HI"].processors[:4]:
sub = processor.process(sub)
MAXIMUM_ENTRIES = 50
MINIMUM_UPPERCASE_PERCENTAGE = 90
MINIMUM_UPPERCASE_COUNT = 100
entry_count = 0
uppercase_count = 0
lowercase_count = 0
if sub.strip():
# only consider alphabetic characters to determine if uppercase
alpha_sub = ''.join([i for i in sub if i.isalpha()])
if alpha_sub and not alpha_sub.isupper():
return False
for entry in self.f:
sub = entry.text
# skip HI bracket entries, those might actually be lowercase
sub = sub.strip()
for processor in registry.mods["remove_HI"].processors[:4]:
sub = processor.process(sub)
entry_used = True
else:
# skip full entry
continue
if sub.strip():
uppercase_count += sum(1 for char in sub if char.isupper())
lowercase_count += sum(1 for char in sub if char.islower())
entry_count += 1
if entry_used:
entries_used += 1
if entry_count >= MAXIMUM_ENTRIES:
break
if entries_used == 40:
break
return True
total_character_count = lowercase_count + uppercase_count
if total_character_count > 0 and uppercase_count > MINIMUM_UPPERCASE_COUNT:
uppercase_percentage = uppercase_count * 100 / total_character_count
logger.debug(f"Uppercase mod percentage is {uppercase_percentage:.2f}% vs minimum of {MINIMUM_UPPERCASE_PERCENTAGE}%")
return uppercase_percentage >= MINIMUM_UPPERCASE_PERCENTAGE
return False
def modify(self, *mods):
new_entries = []
start = time.time()
self.only_uppercase = self.detect_uppercase()
self.mostly_uppercase = self.detect_uppercase()
if self.only_uppercase and self.debug:
logger.debug("Full-uppercase subtitle found")
if self.mostly_uppercase and self.debug:
logger.debug("Mostly-uppercase subtitle found")
line_mods, non_line_mods, mods_used = self.prepare_mods(*mods)
self.mods_used = mods_used