def combobox_o_changed(self, event):
#log.debug(self.combobox_o.get())
formatD = self.combobox_o.get()
if formatD == noneItem:
return
format = Glossary.descFormat[formatD]
"""
if format=='Omnidic':
self.xml.get_widget('label_omnidic_o').show()
self.xml.get_widget('spinbutton_omnidic_o').show()
else:
self.xml.get_widget('label_omnidic_o').hide()
self.xml.get_widget('spinbutton_omnidic_o').hide()
if format=='Babylon':
self.xml.get_widget('label_enc').show()
self.xml.get_widget('comboentry_enc').show()
else:
self.xml.get_widget('label_enc').hide()
self.xml.get_widget('comboentry_enc').hide()
"""
if self.pref['auto_set_out']: #format==None:
pathI = toStr(self.entry_i.get())
pathO = toStr(self.entry_o.get())
formatOD = self.combobox_o.get()
if formatOD != None and not pathO and '.' in pathI:
extO = Glossary.descExt[formatOD]
pathO = ''.join(os.path.splitext(pathI)[:-1]) + extO
#self.entry_o.delete(0, 'end')
self.entry_o.insert(0, pathO)
def combobox_o_changed(self, event):
#log.debug(self.combobox_o.get())
formatD = self.combobox_o.get()
if formatD==noneItem:
return
format = Glossary.descFormat[formatD]
'''if format=='Omnidic':
self.xml.get_widget('label_omnidic_o').show()
self.xml.get_widget('spinbutton_omnidic_o').show()
else:
self.xml.get_widget('label_omnidic_o').hide()
self.xml.get_widget('spinbutton_omnidic_o').hide()
if format=='Babylon':
self.xml.get_widget('label_enc').show()
self.xml.get_widget('comboentry_enc').show()
else:
self.xml.get_widget('label_enc').hide()
self.xml.get_widget('comboentry_enc').hide()
if format=='Stardict':
self.xml.get_widget('checkb_o_ext').show()
else:
self.xml.get_widget('checkb_o_ext').hide()'''
if self.pref['auto_set_out']:#format==None:
pathI = toStr(self.entry_i.get())
pathO = toStr(self.entry_o.get())
formatOD = self.combobox_o.get()
if formatOD != None and not pathO and '.' in pathI:
extO=Glossary.descExt[formatOD]
pathO=''.join(os.path.splitext(pathI)[:-1])+extO
#self.entry_o.delete(0, 'end')
self.entry_o.insert(0, pathO)
def convert(self):
if len(self.glos)==0:
log.error('Input glossary has no word! Be sure to click "Load" before "Convert", '+\
'or just click "Apply" instead.')
return False
oPath = toStr(self.entry_o.get())
if not oPath:
log.critical('Output file path is empty!');return
formatD = self.combobox_o.get()
if formatD in (noneItem, ''):
log.critical('Output format is empty!');return
log.info('Converting to %s, please wait...'%formatD)
#while gtk.events_pending():#??????????
# gtk.main_iteration_do(False)
self.running = True
format = Glossary.descFormat[formatD]
t0 = time.time()
"""
if format=='Omnidic':
dicIndex=self.xml.get_widget('spinbutton_omnidic_o').get_value_as_int()
self.glos.writeOmnidic(oPath, dicIndex=dicIndex)
elif format=='Babylon':
encoding = self.xml.get_widget('comboentry_enc').get_active_text()
self.glos.writeBabylon(oPath, encoding=encoding)
else:"""##???????????????????????
self.glos.write(oPath, format=format)
#self.oFormat = format
self.oPath = oPath
log.info('writing %s file: "%s" done.'%(format, oPath))
log.info('time left = %3f seconds'%(time.time()-t0))
self.running = False
return True
def convert(self):
oPath = toStr(self.entry_o.get())
if not oPath:
log.critical('Output file path is empty!')
return
formatD = self.combobox_o.get()
if formatD in (noneItem, ''):
log.critical('Output format is empty!')
return
log.info('Converting to %s, please wait...' % formatD)
#while gtk.events_pending():#??????????
# gtk.main_iteration_do(False)
self.running = True
format = Glossary.descFormat[formatD]
t0 = time.time()
"""
if format=='Omnidic':
dicIndex=self.xml.get_widget('spinbutton_omnidic_o').get_value_as_int()
self.glos.writeOmnidic(oPath, dicIndex=dicIndex)
elif format=='Babylon':
encoding = self.xml.get_widget('comboentry_enc').get_active_text()
self.glos.writeBabylon(oPath, encoding=encoding)
else:"""##???????????????????????
self.glos.write(oPath, format=format)
#self.oFormat = format
self.oPath = oPath
log.info('writing %s file: "%s" done.' % (format, oPath))
log.info('time left = %3f seconds' % (time.time() - t0))
self.running = False
return True
def entry_changed(self, event=None):
#log.debug('entry_changed')
#char = event.keysym
pathI = toStr(self.entry_i.get())
if self.pathI != pathI:
formatD = self.combobox_i.get()
if len(pathI)>7:
if pathI[:7]=='file://':
pathI=urlToPath(pathI)
self.entry_i.delete(0, 'end')
self.entry_i.insert(0, pathI)
if self.pref['auto_set_for']:#format==noneItem:
ext = os.path.splitext(pathI)[-1].lower()
if ext in ('.gz', '.bz2', '.zip'):
ext = os.path.splitext(pathI[:-len(ext)])[-1].lower()
for i in xrange(len(Glossary.readExt)):
if ext in Glossary.readExt[i]:
self.combobox_i.set(Glossary.readDesc[i])
break
if self.pref['auto_set_out']:#format==noneItem:
#pathI = self.entry_i.get()
formatOD = self.combobox_o.get()
pathO = toStr(self.entry_o.get())
if formatOD != noneItem and not pathO and '.' in pathI:
extO=Glossary.descExt[formatOD]
pathO=''.join(os.path.splitext(pathI)[:-1])+extO
self.entry_o.delete(0, 'end')
self.entry_o.insert(0, pathO)
self.pathI = pathI
##############################################
pathO = toStr(self.entry_o.get())
if self.pathO!=pathO:
formatD = self.combobox_o.get()
if len(pathO)>7:
if pathO[:7]=='file://':
pathO=urlToPath(pathO)
self.entry_o.delete(0, 'end')
self.entry_o.insert(0, pathO)
if self.pref['auto_set_for']:#format==noneItem:
ext = os.path.splitext(pathO)[-1].lower()
if ext in ('.gz', '.bz2', '.zip'):
ext = os.path.splitext(pathO[:-len(ext)])[-1].lower()
for i in xrange(len(Glossary.writeExt)):
if ext in Glossary.writeExt[i]:
self.combobox_o.set(Glossary.writeDesc[i])
break
self.pathO = pathO
def entry_changed(self, event=None):
#log.debug('entry_changed')
#char = event.keysym
pathI = toStr(self.entry_i.get())
if self.pathI != pathI:
formatD = self.combobox_i.get()
if len(pathI) > 7:
if pathI[:7] == 'file://':
pathI = urlToPath(pathI)
self.entry_i.delete(0, 'end')
self.entry_i.insert(0, pathI)
if self.pref['auto_set_for']: #format==noneItem:
ext = os.path.splitext(pathI)[-1].lower()
if ext in ('.gz', '.bz2', '.zip'):
ext = os.path.splitext(pathI[:-len(ext)])[-1].lower()
for i in range(len(Glossary.readExt)):
if ext in Glossary.readExt[i]:
self.combobox_i.set(Glossary.readDesc[i])
break
if self.pref['auto_set_out']: #format==noneItem:
#pathI = self.entry_i.get()
formatOD = self.combobox_o.get()
pathO = toStr(self.entry_o.get())
if formatOD != noneItem and not pathO and '.' in pathI:
extO = Glossary.descExt[formatOD]
pathO = ''.join(os.path.splitext(pathI)[:-1]) + extO
self.entry_o.delete(0, 'end')
self.entry_o.insert(0, pathO)
self.pathI = pathI
##############################################
pathO = toStr(self.entry_o.get())
if self.pathO != pathO:
formatD = self.combobox_o.get()
if len(pathO) > 7:
if pathO[:7] == 'file://':
pathO = urlToPath(pathO)
self.entry_o.delete(0, 'end')
self.entry_o.insert(0, pathO)
if self.pref['auto_set_for']: #format==noneItem:
ext = os.path.splitext(pathO)[-1].lower()
if ext in ('.gz', '.bz2', '.zip'):
ext = os.path.splitext(pathO[:-len(ext)])[-1].lower()
for i in range(len(Glossary.writeExt)):
if ext in Glossary.writeExt[i]:
self.combobox_o.set(Glossary.writeDesc[i])
break
self.pathO = pathO
def get_prefix(self, word: str) -> str:
length = self._group_by_prefix_length
if not word:
return None
word = toStr(word)
if "Z" < word[0] < "a":
return "SPECIAL"
return word[:length]
def samplesDumpFileWrite(self, text):
text = toStr(text)
if self.samplesDumpFile:
offset = self.samplesDumpFile.tell()
self.samplesDumpFile.write("\noffset = {0:#X}\n" % offset)
self.samplesDumpFile.write(text + "\n")
else:
log.debug(text)
def samplesDumpFileWrite(self, text):
text = toStr(text)
if self.samplesDumpFile:
offset = self.samplesDumpFile.tell()
self.samplesDumpFile.write(f"\noffset = {offset:#02x}\n")
self.samplesDumpFile.write(text + "\n")
else:
log.debug(text)
def msgLogFileWrite(self, text):
text = toStr(text)
if self.msgLogFile:
offset = self.msgLogFile.tell()
# print offset in the log file to facilitate navigating this
# log in hex editor
# intended usage:
# the log file is opened in a text editor and hex editor
# use text editor to read error messages, use hex editor to
# inspect char codes offsets allows to quickly jump to the right
# place of the file hex editor
self.msgLogFile.write(f"\noffset = {offset:#02x}\n")
self.msgLogFile.write(text + "\n")
else:
log.debug(text)
def get_prefix(word, length):
"""
Return the prefix for the given word,
of length length.
:param word: the word string
:type word: unicode
:param length: prefix length
:type length: int
:rtype: unicode
"""
if not word:
return None
word = toStr(word)
if "Z" < word[0] < "a":
return "SPECIAL"
return word[:length] ## return a unicode? FIXME
def load(self):
iPath = toStr(self.entry_i.get())
if not iPath:
printAsError('Input file path is empty!')
return
formatD = self.combobox_i.get()
if formatD == noneItem:
#printAsError('Input format is empty!');return
format = ''
print('Please wait...')
else:
format = Glossary.descFormat[formatD]
print('Reading from %s, please wait...' % formatD)
#while gtk.events_pending():#??????????????
# gtk.main_iteration_do(False)
t0 = time.time()
'''
if formatD[:7]=='Omnidic':
dicIndex=self.xml.get_widget('spinbutton_omnidic_i').get_value_as_int()
ex = self.glos.readOmnidic(iPath, dicIndex=dicIndex)
elif formatD[:8]=='StarDict' and self.checkb_i_ext.get_active():
ex = self.glos.readStardict_ext(iPath)
else:'''
ex = self.glos.read(iPath, format=format)
if ex:
print('reading %s file: "%s" done.\n%d words found.' % (
format,
iPath,
len(self.glos.data),
))
else:
print('reading %s file: "%s" failed.' % (format, iPath))
return False
#self.iFormat = format
self.iPath = iPath
#self.button_conv.set_sensitive(True)
self.glos.uiEdit()
self.progress(1.0, 'Loading Comleted')
if self.checkb_o_det.get(): #?????????
print('time left = %3f seconds' % (time.time() - t0))
for x in self.glos.info:
print('%s="%s"' % (x[0], x[1]))
return True
def load(self):
iPath = toStr(self.entry_i.get())
if not iPath:
log.error('Input file path is empty!');return
formatD = self.combobox_i.get()
if formatD==noneItem:
#log.error('Input format is empty!');return
format=''
log.info('Please wait...')
else:
format = Glossary.descFormat[formatD]
log.info('Reading from %s, please wait...'%formatD)
#while gtk.events_pending():#??????????????
# gtk.main_iteration_do(False)
t0=time.time()
'''
if formatD[:7]=='Omnidic':
dicIndex=self.xml.get_widget('spinbutton_omnidic_i').get_value_as_int()
ex = self.glos.readOmnidic(iPath, dicIndex=dicIndex)
elif formatD[:8]=='StarDict' and self.checkb_i_ext.get_active():
ex = self.glos.readStardict_ext(iPath)
else:'''
ex = self.glos.read(iPath, format=format)
if ex:
log.info('reading %s file: "%s" done.\n%d words found.'%(
format,
iPath,
len(self.glos.data),
))
else:
log.error('reading %s file: "%s" failed.'%(format, iPath))
return False
#self.iFormat = format
self.iPath = iPath
#self.button_conv.set_sensitive(True)
self.glos.uiEdit()
self.progress(1.0, 'Loading Comleted')
if self.checkb_o_det.get():#?????????
log.info('time left = %3f seconds'%(time.time()-t0))
for x in self.glos.info:
log.info('%s="%s"'%(x[0], x[1]))
return True
def processDefiStat(self, fields, b_defi, b_key):
if fields.singleEncoding:
self.findAndPrintCharSamples(
fields.b_defi,
f"defi, key = {b_key}",
fields.encoding,
)
if self.metadata2:
self.metadata2.defiProcessedCount += 1
if isASCII(toStr(fields.b_defi)):
self.metadata2.defiAsciiCount += 1
try:
fields.b_defi.decode("utf-8")
except UnicodeError:
pass
else:
self.metadata2.defiUtf8Count += 1
if self.metadata2 and self.metadata2.isDefiASCII:
if not isASCII(fields.u_defi):
self.metadata2.isDefiASCII = False
def convert(self):
if len(self.glos.data) == 0:
printAsError('Input glossary has no word! Be sure to click "Load" before "Convert", '+\
'or just click "Apply" instead.')
return False
oPath = toStr(self.entry_o.get())
if not oPath:
printAsError('Output file path is empty!')
return
formatD = self.combobox_o.get()
if formatD in (noneItem, ''):
printAsError('Output format is empty!')
return
print('Converting to %s, please wait...' % formatD)
#while gtk.events_pending():#??????????
# gtk.main_iteration_do(False)
self.running = True
format = Glossary.descFormat[formatD]
t0 = time.time()
'''
if format=='Stardict':
if self.xml.get_widget('checkb_o_ext').get_active():
self.glos.writeStardict(oPath)
else:
self.glos.writeStardict_int(oPath)
elif format=='Omnidic':
dicIndex=self.xml.get_widget('spinbutton_omnidic_o').get_value_as_int()
self.glos.writeOmnidic(oPath, dicIndex=dicIndex)
elif format=='Babylon':
encoding = self.xml.get_widget('comboentry_enc').get_active_text()
self.glos.writeBabylon(oPath, encoding=encoding)
else:'''##???????????????????????
self.glos.write(oPath, format=format)
#self.oFormat = format
self.oPath = oPath
print('writing %s file: "%s" done.' % (format, oPath))
if self.checkb_o_det.get(): #???????
print('time left = %3f seconds' % (time.time() - t0))
self.running = False
return True
def _mktitle(title_element, include_opts=()):
title = title_element.text
opt_i = -1
for c in title_element:
if c.tag == 'nu' and c.tail:
if title:
title += c.tail
else:
title = c.tail
if c.tag == 'opt':
opt_i += 1
if opt_i in include_opts:
if title:
title += c.text
else:
title = c.text
if c.tail:
if title:
title += c.tail
else:
title = c.tail
return toStr(title.strip())
def rawDumpFileWriteData(self, text):
text = toStr(text)
# the next function escapes too many chars, for example, it escapes äöü
# self.rawDumpFile.write(text.encode("unicode_escape"))
if self.rawDumpFile:
self.rawDumpFile.write(text)
def rawDumpFileWriteText(self, text): # FIXME
text = toStr(text)
if self.rawDumpFile:
self.rawDumpFile.write(text)
def format_clean_content(title, body, BeautifulSoup):
# heavily integrated with output of dsl reader plugin!
# and with xdxf also.
"""
:param title: str | None
"""
# class="sec" => d:priority="2"
# style="color:steelblue" => class="ex"
# class="p" style="color:green" => class="p"
# style="color:green" => class="c"
# style="margin-left:{}em" => class="m{}"
# <s> => <del>
# xhtml is strict
if BeautifulSoup:
soup = BeautifulSoup.BeautifulSoup(body, "lxml", from_encoding='utf-8')
# difference between 'lxml' and 'html.parser'
if soup.body:
soup = soup.body
for tag in soup(class_='sec'):
tag['class'].remove('sec')
if not tag['class']:
del tag['class']
tag['d:priority'] = "2"
for tag in soup(lambda x: 'color:steelblue' in x.get('style', '')):
remove_style(tag, 'color:steelblue')
if 'ex' not in tag.get('class', []):
tag['class'] = tag.get('class', []) + ['ex']
for tag in soup(is_green):
remove_style(tag, 'color:green')
if 'p' not in tag.get('class', ''):
tag['class'] = tag.get('class', []) + ['c']
for tag in soup(True):
if 'style' in tag.attrs:
m = margin_re.search(tag['style'])
if m:
remove_style(tag, m.group(0))
tag['class'] = tag.get('class', []) + ['m' + m.group(1)]
for tag in soup.select('[href]'):
href = tag['href']
if not (href.startswith('http:') or href.startswith('https:')):
tag['href'] = 'x-dictionary:d:%s' % href
for tag in soup('u'):
tag.name = 'span'
tag['class'] = tag.get('class', []) + ['u']
for tag in soup('s'):
tag.name = 'del'
if title:
h1 = BeautifulSoup.Tag(name='h1')
h1.string = title
soup.insert(0, h1)
# hence the name BeautifulSoup
content = toStr(soup.encode_contents())
else:
# somewhat analogue to what BeautifulSoup suppose to do
body = em0_9_re.sub(em0_9_sub, body)
body = em0_9_ex_re.sub(em0_9_ex_sub, body)
body = href_re.sub(href_sub, body)
body = body \
.replace('<i style="color:green">', '<i class="c">') \
.replace('<i class="p" style="color:green">', '<i class="p">') \
.replace('<span class="ex" style="color:steelblue">', '<span class="ex">') \
.replace('<span class="sec ex" style="color:steelblue">', '<span class="sec ex">') \
.replace('<u>', '<span class="u">').replace('</u>', '</span>') \
.replace('<s>', '<del>').replace('</s>', '</del>')
# nice header to display
content = '<h1>%s</h1>%s' % (title, body) if title else body
content = close_tag.sub('<\g<1> />', content)
content = img_tag.sub('<img \g<1>/>', content)
content = content.replace(' ', ' ')
content = nonprintable.sub('', content)
return content
def prepare_content_with_soup(
title: "Optional[str]",
body: str,
BeautifulSoup: "Any",
) -> str:
soup = BeautifulSoup.BeautifulSoup(body, features="lxml")
# difference between "lxml" and "html.parser"
if soup.body:
soup = soup.body
for tag in soup(class_="sec"):
tag["class"].remove("sec")
if not tag["class"]:
del tag["class"]
tag["d:priority"] = "2"
for tag in soup(lambda x: "color:steelblue" in x.get("style", "")):
remove_style(tag, "color:steelblue")
if "ex" not in tag.get("class", []):
tag["class"] = tag.get("class", []) + ["ex"]
for tag in soup(is_green):
remove_style(tag, "color:green")
if "p" not in tag.get("class", ""):
tag["class"] = tag.get("class", []) + ["c"]
for tag in soup(True):
if "style" in tag.attrs:
m = re_margin.search(tag["style"])
if m:
remove_style(tag, m.group(0))
tag["class"] = tag.get("class", []) + ["m" + m.group(1)]
for tag in soup(lambda x: "xhtml:" in x.name):
old_tag_name = tag.name
tag.name = old_tag_name[len("xhtml:"):]
if tag.string:
tag.string = f"{tag.string} "
for tag in soup.select("[href]"):
href = tag["href"]
href = cleanup_link_target(href)
if href.startswith("sound:"):
fix_sound_link(href, tag)
elif href.startswith("phonetics") or href.startswith("help:phonetics"):
# for oxford9
log.debug(f"phonetics: tag={tag}")
if tag.audio and "name" in tag.audio.attrs:
tag["onmousedown"] = f"this.lastChild.play(); return false;"
src_name = tag.audio["name"].replace("#", "_")
tag.audio["src"] = f"{src_name}.mp3"
elif not link_is_url(href):
tag["href"] = f"x-dictionary:d:{href}"
for thumb in soup.find_all("div", "pic_thumb"):
thumb["onclick"] = 'this.setAttribute("style", "display:none"); ' \
'this.nextElementSibling.setAttribute("style", "display:block")'
for pic in soup.find_all("div", "big_pic"):
pic["onclick"] = 'this.setAttribute("style", "display:none"), ' \
'this.previousElementSibling.setAttribute("style", "display:block")'
# to unfold(expand) and fold(collapse) blocks
for pos in soup.find_all("pos", onclick="toggle_infl(this)"):
# TODO: simplify this!
pos["onclick"] = (
r'var e = this.parentElement.parentElement.parentElement'
r'.querySelector("res-g vp-gs"); style = window.'
r'getComputedStyle(e), display = style.getPropertyValue'
r'("display"), "none" === e.style.display || "none" === display'
r' ? e.style.display = "block" : e.style.display = "none", '
r'this.className.match(/(?:^|\s)Clicked(?!\S)/) ? this.'
r'className = this.className.replace('
r'/(?:^|\s)Clicked(?!\S)/g, "") : this.setAttribute('
r'"class", "Clicked")'
)
for tag in soup.select("[src]"):
src = tag["src"]
if src.startswith("/"):
tag["src"] = src[1:]
for tag in soup("u"):
tag.name = "span"
tag["class"] = tag.get("class", []) + ["u"]
for tag in soup("s"):
tag.name = "del"
if title and "<h" not in body:
h1 = BeautifulSoup.Tag(name="h1")
h1.string = title
soup.insert(0, h1)
# hence the name BeautifulSoup
# soup.insert(0,head)
content = toStr(soup.encode_contents())
return content
def format_clean_content(title, body, BeautifulSoup):
# heavily integrated with output of dsl reader plugin!
# and with xdxf also.
"""
:param title: str | None
"""
# class="sec" => d:priority="2"
# style="color:steelblue" => class="ex"
# class="p" style="color:green" => class="p"
# style="color:green" => class="c"
# style="margin-left:{}em" => class="m{}"
# <s> => <del>
# xhtml is strict
if BeautifulSoup:
soup = BeautifulSoup.BeautifulSoup(body, "lxml", from_encoding='utf-8')
# difference between 'lxml' and 'html.parser'
if soup.body:
soup = soup.body
for tag in soup(class_='sec'):
tag['class'].remove('sec')
if not tag['class']:
del tag['class']
tag['d:priority'] = "2"
for tag in soup(lambda x: 'color:steelblue' in x.get('style', '')):
remove_style(tag, 'color:steelblue')
if 'ex' not in tag.get('class', []):
tag['class'] = tag.get('class', []) + ['ex']
for tag in soup(is_green):
remove_style(tag, 'color:green')
if 'p' not in tag.get('class', ''):
tag['class'] = tag.get('class', []) + ['c']
for tag in soup(True):
if 'style' in tag.attrs:
m = margin_re.search(tag['style'])
if m:
remove_style(tag, m.group(0))
tag['class'] = tag.get('class', []) + ['m' + m.group(1)]
for tag in soup.select('[href]'):
href = tag['href']
if not (href.startswith('http:') or href.startswith('https:')):
tag['href'] = 'x-dictionary:d:%s' % href
for tag in soup('u'):
tag.name = 'span'
tag['class'] = tag.get('class', []) + ['u']
for tag in soup('s'):
tag.name = 'del'
if title:
h1 = BeautifulSoup.Tag(name='h1')
h1.string = title
soup.insert(0, h1)
# hence the name BeautifulSoup
content = toStr(soup.encode_contents())
else:
# somewhat analogue to what BeautifulSoup suppose to do
body = em0_9_re.sub(em0_9_sub, body)
body = em0_9_ex_re.sub(em0_9_ex_sub, body)
body = href_re.sub(href_sub, body)
body = body \
.replace('<i style="color:green">', '<i class="c">') \
.replace('<i class="p" style="color:green">', '<i class="p">') \
.replace('<span class="ex" style="color:steelblue">', '<span class="ex">') \
.replace('<span class="sec ex" style="color:steelblue">', '<span class="sec ex">') \
.replace('<u>', '<span class="u">').replace('</u>', '</span>') \
.replace('<s>', '<del>').replace('</s>', '</del>')
# nice header to display
content = '<h1>%s</h1>%s' % (title, body) if title else body
content = close_tag.sub('<\g<1> />', content)
content = img_tag.sub('<img \g<1>/>', content)
content = content.replace(' ', ' ')
content = nonprintable.sub('', content)
return content
