def reorder(self):
if len(self.blocks) > 1:
pos = -1
for i in range(len(self.blocks)):
if self.blocks[i].type_ == 'dic':
pos = i
break
if pos >= 0:
self.blocks.append(self.blocks[pos])
del self.blocks[pos]
pos = -1
for i in range(len(self.blocks)):
if self.blocks[i].type_ == 'comment':
pos = i
break
if pos >= 0:
self.blocks.append(self.blocks[pos])
del self.blocks[pos]
if sh.Text(self.pattern).has_cyrillic():
for i in range(len(self.blocks)):
if self.blocks[i-1].type_ == 'term' \
and self.blocks[i].type_ == 'term' \
and self.blocks[i-1].lang != 2 \
and self.blocks[i].lang == 2:
self.blocks[
i -
1], self.blocks[i] = self.blocks[i], self.blocks[i
-
1]
def swap_dics(self):
for i in range(len(self.blocks)):
if self.is_dic(self.blocks[i]):
Condition = False
if i > 0:
if sh.Text(text=self.blocks[i]).has_cyrillic() \
and sh.Text(text=self.blocks[i-1]).has_latin():
Condition = True
elif sh.Text(text=self.blocks[i]).has_latin() \
and sh.Text(text=self.blocks[i-1]).has_cyrillic():
Condition = True
if Condition:
self.blocks[
i -
1], self.blocks[i] = self.blocks[i], self.blocks[i - 1]
return self.blocks
def get_priorities(self):
f = '[MClient] logic.Lists.get_priorities'
if self.Success:
text = sh.ReadTextFile(self.priorlst,True).get()
text = sh.Text(text,True).text
return text.splitlines()
else:
sh.com.cancel(f)
def run(self):
f = '[MClient] plugins.dsl.cleanup.CleanUp.run'
if self.text:
self.delete_trash()
self.text = sh.Text(self.text).delete_unsupported()
return self.text
else:
sh.com.rep_empty(f)
return ''
def run(self):
f = '[MClient] plugins.stardict.cleanup.Common.run'
if self.text:
self.decode_entities()
self.delete_trash()
self.text = sh.Text(self.text).delete_unsupported()
return self.text
else:
sh.com.rep_empty(f)
return ''
def pretty(self):
f = '[Samurai] samurai.Parse.pretty'
timer = sh.Timer(f) #TODO: del
timer.start() #TODO: del
if self.text:
text = sh.Text(text=self.text)
text.convert_line_breaks()
text.strip_lines()
text.tabs2spaces()
text.replace_x()
text.delete_duplicate_spaces()
self.text = text.text
# Allow only 2 consequent line breaks
while '\n\n\n' in self.text:
self.text = self.text.replace('\n\n\n','\n\n')
else:
sh.com.rep_empty(f)
timer.end() #TODO: del
def gen_poses(self):
''' We generate positions here according to the text produced by
TkinterHtml's 'widget.text()' command.
Peculiarities of the retrieved text:
- TkinterHtml adds some empty lines from the top and
the bottom;
- The number of these lines varies each time (and we don't
know the rule according to which they are generated)
- Each cell occupies a single line
- Blocks within a cell are spaced with a single space
(except for the blocks having such text as '(')
- Each line is stripped
- Pos2 of the previous cell and pos1 of the next cell are
sometimes equal; this corresponds to the position system
used by Tkinter
- Duplicate spaces are removed
'''
f = '[MClient] cells.Pos.gen_poses'
last = 0
not_found = []
for block in self.blocks:
text = sh.Text(block.text.strip()).delete_duplicate_spaces()
if text:
search = sh.Search(text=self.rawtext, pattern=text)
search.i = last
result = sh.Input(f, search.get_next()).get_integer()
if result >= last:
block.first = result
else:
block.first = last
not_found.append(text)
else:
block.first = last
block.last = block.first + len(text)
last = block.last
''' This may be caused by the following:
a) Missing types in db.DB.types
b) Defective HTML generation algorithm
'''
if not_found:
not_found = ['"' + item + '"' for item in not_found]
not_found = '\n' + '\n'.join(not_found)
mes = _('Unable to find: {}').format(not_found)
sh.objs.get_mes(f, mes).show_error()
def run(self):
f = '[MClient] plugins.multitrancom.cleanup.CleanUp.run'
if self.text:
self.delete_trash_tags()
self.fix_href()
self.fix_tags()
self.run_common()
''' #TODO: do we really need this heaviest operation (takes
~0.53s for 'set' (EN-RU) on AMD E-300, whereas the
entire module takes ~0.58s, since we have already
deleted unicode control codes?
'''
# Delete a non-breaking space before a user name
self.text = self.text.replace(' ',' ')
self.text = sh.Text(self.text).delete_unsupported()
return self.text
else:
sh.com.rep_empty(f)
return ''
def check(self):
f = '[shared] spelling.Spelling.check'
''' We do not warn about empty input/output since this happens
quite often. Moreover, we return True in such cases since
this is actually not a spelling error.
'''
if self.word:
self.word = self.word.strip()
itext = sh.Text(self.word)
self.word = itext.delete_punctuation()
if self.word and not itext.has_digits():
if len(self.word) == 1:
return True
elif itext.has_cyrillic():
return self.ru.check(self.word)
elif itext.has_latin():
return self.gb.check(self.word) \
or self.us.check(self.word)
else:
# Ignore an empty input and words with digits
return True
else:
return True
def _break_word(self, word):
lst = sh.Text(word).split_by_len(self.max_word_len)
return ' '.join(lst)