def doSpellcheck(self):
canspell = IMC.spellCheck.isUp()
nwords = IMC.wordCensus.size()
if 0 >= nwords : # could be zero in a null document
return
pqMsgs.startBar(nwords,"Checking spelling...")
for i in range(IMC.wordCensus.size()):
(qword, cnt, wflags) = IMC.wordCensus.get(i)
wflags = wflags & (0xff - IMC.WordMisspelt) # turn off flag if on
# some words have /dict-tag, split that out as string or ""
(w,x,d) = unicode(qword).partition("/")
if IMC.goodWordList.check(w):
pass
elif IMC.badWordList.check(w) :
wflags |= IMC.WordMisspelt
elif canspell : # check word in its optional dictionary
if not ( IMC.spellCheck.check(w,d) ) :
wflags |= IMC.WordMisspelt
IMC.wordCensus.setflags(i,wflags)
if 0 == i & 0x1f :
pqMsgs.rollBar(i)
pqMsgs.endBar()
IMC.needMetadataSave |= IMC.wordlistsChanged
IMC.needSpellCheck = False
if IMC.spellingHiliteSwitch :
self.setHighlight(True) # force refresh of spell underlines
def doSpellcheck(self):
canspell = IMC.spellCheck.isUp()
nwords = IMC.wordCensus.size()
if 0 >= nwords: # could be zero in a null document
return
pqMsgs.startBar(nwords, "Checking spelling...")
for i in range(IMC.wordCensus.size()):
(qword, cnt, wflags) = IMC.wordCensus.get(i)
wflags = wflags & (0xff - IMC.WordMisspelt) # turn off flag if on
# some words have /dict-tag, split that out as string or ""
(w, x, d) = unicode(qword).partition("/")
if IMC.goodWordList.check(w):
pass
elif IMC.badWordList.check(w):
wflags |= IMC.WordMisspelt
elif canspell: # check word in its optional dictionary
if not (IMC.spellCheck.check(w, d)):
wflags |= IMC.WordMisspelt
IMC.wordCensus.setflags(i, wflags)
if 0 == i & 0x1f:
pqMsgs.rollBar(i)
pqMsgs.endBar()
IMC.needMetadataSave |= IMC.wordlistsChanged
IMC.needSpellCheck = False
if IMC.spellingHiliteSwitch:
self.setHighlight(True) # force refresh of spell underlines
def doCensus(self, page=False) :
global reLineSep, reTokens, reLang, qcLess
# Clear the current census values
IMC.wordCensus.clear()
IMC.charCensus.clear()
# Count chars locally for speed
local_char_census = defaultdict(int)
# Name of current alternate dictionary
alt_dict = QString() # isEmpty when none
# Tag from which we set an alternate dict
alt_dict_tag = QString()
# Start the progress bar based on the number of lines in the document
pqMsgs.startBar(self.document().blockCount(),"Counting words and chars...")
# Find the first text block of interest, skipping an HTML header file
qtb = self.document().begin() # first text block
if IMC.bookType.startsWith(QString(u"htm")) \
and qtb.text().startsWith(QString(u"<!DOCTYPE")) :
while (qtb != self.document().end()) \
and (not qtb.text().startsWith(QString(u"<body"))) :
qtb = qtb.next()
# Scan all lines of the document to the end.
while qtb != self.document().end() :
qsLine = qtb.text() # text of line as qstring
dbg = qsLine.size()
dbg2 = qtb.length()
if reLineSep.exactMatch(qsLine): # this is a page separator line
if page :
# We are doing page seps, it's for Open with no .meta seen,
# the page table has been cleared. Store the page sep
# data in the page table, with a textCursor to its start.
qsfilenum = reLineSep.cap(1) # xxx from "File: xxx.png"
qsproofers = reLineSep.cap(2) # \who\x\blah\etc
# proofer names can contain spaces, replace with en-space char
qsproofers.replace(QChar(" "),QChar(0x2002))
# create a new TextCursor instance
tcursor = QTextCursor(self.document())
# point it to this text block
tcursor.setPosition(qtb.position())
# dump all that in the page table
IMC.pageTable.loadPsep(tcursor, qsfilenum, qsproofers)
# else not doing pages, just ignore this psep line
else: # not psep, ordinary text line, count chars and words
pyLine = unicode(qsLine) # move into Python space to count
for c in pyLine :
local_char_census[c] += 1
j = 0
while True:
j = reTokens.indexIn(qsLine,j)
if j < 0 : # no more word-like units
break
qsWord = reTokens.cap(0)
j += qsWord.size()
if qsWord.startsWith(qcLess) :
# Examine a captured HTML production.
if not reTokens.cap(2).isEmpty() :
# HTML open tag, look for lang='dict'
if 0 <= reLang.indexIn(reTokens.cap(3)) :
# found it: save tag and dict name
alt_dict_tag = QString(reTokens.cap(2))
alt_dict = QString(reLang.cap(1))
alt_dict.prepend(u'/') # make "/en_GB"
# else no lang= attribute
else:
# HTML close tag, see if it closes alt dict use
if reTokens.cap(5) == alt_dict_tag :
# yes, matches open-tag for dict, clear it
alt_dict_tag = QString()
alt_dict = QString()
# else no alt dict in use, or didn't match
else : # did not start with "<", process as a word
# Set the property flags, which is harder now we don't
# look at every character. Use the QString facilities
# rather than python because python .isalnum fails
# for a hyphenated number "1850-1910".
flag = 0
if 0 != qsWord.compare(qsWord.toLower()) :
flag |= IMC.WordHasUpper
if 0 != qsWord.compare(qsWord.toUpper()) :
flag |= IMC.WordHasLower
if qsWord.contains(qcHyphen) :
flag |= IMC.WordHasHyphen
if qsWord.contains(qcApostrophe) or qsWord.contains(qcCurlyApostrophe) :
flag |= IMC.WordHasApostrophe
if qsWord.contains(reDigit) :
flag |= IMC.WordHasDigit
IMC.wordCensus.count(qsWord.append(alt_dict),flag)
# end "while any more words in this line"
# end of not-a-psep-line processing
qtb = qtb.next() # move on to next block
if (0 == (qtb.blockNumber() & 255)) : #every 256th block
pqMsgs.rollBar(qtb.blockNumber()) # roll the bar
QApplication.processEvents()
# end of scanning all text blocks in the doc
pqMsgs.endBar()
# we accumulated the char counts in localCharCensus. Now read it out
# in sorted order and stick it in the IMC.charCensus list.
for one_char in sorted(local_char_census.keys()):
qc = QChar(ord(one_char)) # get to QChar for category() method
IMC.charCensus.append(QString(qc),local_char_census[one_char],qc.category())
IMC.needSpellCheck = True # after a census this is true
IMC.staleCensus = 0 # but this is no longer true
IMC.needMetadataSave |= IMC.wordlistsChanged
def loadStarts(self):
pqMsgs.startBar(100,"Loading HTML")
def loadStarts(self):
pqMsgs.startBar(100, "Loading HTML")
def doCensus(self, page=False):
global reLineSep, reTokens, reLang, qcLess
# Clear the current census values
IMC.wordCensus.clear()
IMC.charCensus.clear()
# Count chars locally for speed
local_char_census = defaultdict(int)
# Name of current alternate dictionary
alt_dict = QString() # isEmpty when none
# Tag from which we set an alternate dict
alt_dict_tag = QString()
# Start the progress bar based on the number of lines in the document
pqMsgs.startBar(self.document().blockCount(),
"Counting words and chars...")
# Find the first text block of interest, skipping an HTML header file
qtb = self.document().begin() # first text block
if IMC.bookType.startsWith(QString(u"htm")) \
and qtb.text().startsWith(QString(u"<!DOCTYPE")) :
while (qtb != self.document().end()) \
and (not qtb.text().startsWith(QString(u"<body"))) :
qtb = qtb.next()
# Scan all lines of the document to the end.
while qtb != self.document().end():
qsLine = qtb.text() # text of line as qstring
dbg = qsLine.size()
dbg2 = qtb.length()
if reLineSep.exactMatch(qsLine): # this is a page separator line
if page:
# We are doing page seps, it's for Open with no .meta seen,
# the page table has been cleared. Store the page sep
# data in the page table, with a textCursor to its start.
qsfilenum = reLineSep.cap(1) # xxx from "File: xxx.png"
qsproofers = reLineSep.cap(2) # \who\x\blah\etc
# proofer names can contain spaces, replace with en-space char
qsproofers.replace(QChar(" "), QChar(0x2002))
# create a new TextCursor instance
tcursor = QTextCursor(self.document())
# point it to this text block
tcursor.setPosition(qtb.position())
# dump all that in the page table
IMC.pageTable.loadPsep(tcursor, qsfilenum, qsproofers)
# else not doing pages, just ignore this psep line
else: # not psep, ordinary text line, count chars and words
pyLine = unicode(qsLine) # move into Python space to count
for c in pyLine:
local_char_census[c] += 1
j = 0
while True:
j = reTokens.indexIn(qsLine, j)
if j < 0: # no more word-like units
break
qsWord = reTokens.cap(0)
j += qsWord.size()
if qsWord.startsWith(qcLess):
# Examine a captured HTML production.
if not reTokens.cap(2).isEmpty():
# HTML open tag, look for lang='dict'
if 0 <= reLang.indexIn(reTokens.cap(3)):
# found it: save tag and dict name
alt_dict_tag = QString(reTokens.cap(2))
alt_dict = QString(reLang.cap(1))
alt_dict.prepend(u'/') # make "/en_GB"
# else no lang= attribute
else:
# HTML close tag, see if it closes alt dict use
if reTokens.cap(5) == alt_dict_tag:
# yes, matches open-tag for dict, clear it
alt_dict_tag = QString()
alt_dict = QString()
# else no alt dict in use, or didn't match
else: # did not start with "<", process as a word
# Set the property flags, which is harder now we don't
# look at every character. Use the QString facilities
# rather than python because python .isalnum fails
# for a hyphenated number "1850-1910".
flag = 0
if 0 != qsWord.compare(qsWord.toLower()):
flag |= IMC.WordHasUpper
if 0 != qsWord.compare(qsWord.toUpper()):
flag |= IMC.WordHasLower
if qsWord.contains(qcHyphen):
flag |= IMC.WordHasHyphen
if qsWord.contains(qcApostrophe) or qsWord.contains(
qcCurlyApostrophe):
flag |= IMC.WordHasApostrophe
if qsWord.contains(reDigit):
flag |= IMC.WordHasDigit
IMC.wordCensus.count(qsWord.append(alt_dict), flag)
# end "while any more words in this line"
# end of not-a-psep-line processing
qtb = qtb.next() # move on to next block
if (0 == (qtb.blockNumber() & 255)): #every 256th block
pqMsgs.rollBar(qtb.blockNumber()) # roll the bar
QApplication.processEvents()
# end of scanning all text blocks in the doc
pqMsgs.endBar()
# we accumulated the char counts in localCharCensus. Now read it out
# in sorted order and stick it in the IMC.charCensus list.
for one_char in sorted(local_char_census.keys()):
qc = QChar(ord(one_char)) # get to QChar for category() method
IMC.charCensus.append(QString(qc), local_char_census[one_char],
qc.category())
IMC.needSpellCheck = True # after a census this is true
IMC.staleCensus = 0 # but this is no longer true
IMC.needMetadataSave |= IMC.wordlistsChanged
