def update_from_unit(self, translation, unit, pos, template=None):
"""
Process translation toolkit unit and stores/updates database entry.
"""
if template is None:
src = get_source(unit)
ctx = get_context(unit)
else:
src = get_target(template)
ctx = get_context(template)
checksum = msg_checksum(src, ctx)
# Try getting existing unit
from weblate.trans.models import Unit
dbunit = None
try:
dbunit = self.get(translation=translation, checksum=checksum)
force = False
except Unit.MultipleObjectsReturned:
# Some inconsistency (possibly race condition), try to recover
self.filter(translation=translation, checksum=checksum).delete()
except Unit.DoesNotExist:
pass
# Create unit if it does not exist
if dbunit is None:
dbunit = Unit(translation=translation, checksum=checksum, source=src, context=ctx)
force = True
# Update all details
dbunit.update_from_unit(unit, pos, force, template)
# Return result
return dbunit, force
def get_oracle(url):
source = util.get_source(url)
html = lxml.html.document_fromstring(source)
html.make_links_absolute(url, resolve_base_href=True)
util.save_file(lxml.html.tostring(html), 'oracle.html')
util.screenshot('oracle.html', 'oracle.png')
return html
def get_oracle(url): source = util.get_source(url) parser = lxml.etree.HTMLParser() etree = lxml.etree.parse(StringIO(source), parser) html = lxml.html.document_fromstring(source) html.make_links_absolute(url, resolve_base_href=True) html.doctype = etree.docinfo.doctype return html
def get_oracle(url):
source = util.get_source(url)
parser = lxml.etree.HTMLParser()
etree = lxml.etree.parse(StringIO(source), parser)
html = lxml.html.document_fromstring(source)
html.make_links_absolute(url, resolve_base_href=True)
html.doctype = etree.docinfo.doctype
return html
def main():
website = "https://nobaproject.com/textbooks/new-textbook-4782ff3c-3de1-4700-a262-9cc372550395/modules/sensation-and-perception"
# find out the source of the website and if it has a specific scraper for it
source = util.get_source(website)
scraper = util.get_scraper(source)
driver = util.init_driver(chrome_path)
# create article attributes based on the content scraped
article = fill_article_class(scraper, website, driver)
print(article.headline)
print(generate_content(article))
# parse to get the audio version
audio = get_audio(article, 'en')
audio.save('test.mp3')
winsound.PlaySound('test.mp3', winsound.SND_FILENAME | winsound.SND_ASYNC)
def update_from_unit(self, translation, unit, pos, template=None):
'''
Process translation toolkit unit and stores/updates database entry.
'''
if template is None:
src = get_source(unit)
ctx = get_context(unit)
else:
src = get_target(template)
ctx = get_context(template)
checksum = msg_checksum(src, ctx)
# Try getting existing unit
from weblate.trans.models import Unit
dbunit = None
try:
dbunit = self.get(translation=translation, checksum=checksum)
force = False
except Unit.MultipleObjectsReturned:
# Some inconsistency (possibly race condition), try to recover
self.filter(translation=translation, checksum=checksum).delete()
except Unit.DoesNotExist:
pass
# Create unit if it does not exist
if dbunit is None:
dbunit = Unit(translation=translation,
checksum=checksum,
source=src,
context=ctx)
force = True
# Update all details
dbunit.update_from_unit(unit, pos, force, template)
# Return result
return dbunit, force