def save_cnxml(save_dir, cnxml, files):
# write CNXML output
save_and_backup_file(save_dir, 'index.cnxml', cnxml)
# write files
for filename, content in files:
filename = os.path.join(save_dir, filename)
f = open(filename, 'wb') # write binary, important!
f.write(content)
f.close()
# we generate the preview and save the error
conversionerror = None
try:
htmlpreview = cnxml_to_htmlpreview(cnxml)
except libxml2.parserError:
conversionerror = traceback.format_exc()
# Zip up all the files. This is done now, since we have all the files
# available, and it also allows us to post a simple download link.
# Note that we cannot use zipfile as context manager, as that is only
# available from python 2.7
# TODO: Do a filesize check xxxx
if conversionerror is None:
save_and_backup_file(save_dir, 'index.html', htmlpreview)
save_zip(save_dir, cnxml, htmlpreview, files)
else:
save_zip(save_dir, cnxml, None, files)
raise ConversionError(conversionerror)
def _download_module(self, module_url):
request = self.request
session = request.session
conn = sword2cnx.Connection(session['login'].service_document_url,
user_name=session['login'].username,
user_pass=session['login'].password,
always_authenticate=True,
download_service_document=False)
parts = urlparse.urlsplit(module_url)
path = parts.path.split('/')
path = path[:path.index('sword')]
module_url = '%s://%s%s' % (parts.scheme, parts.netloc, '/'.join(path))
# example: http://cnx.org/Members/user001/m17222/sword/editmedia
zip_file = conn.get_cnx_module(module_url = module_url,
packaging = 'zip')
save_dir = get_save_dir(request)
if save_dir is None:
request.session['upload_dir'], save_dir = create_save_dir(request)
extract_to_save_dir(zip_file, save_dir)
cnxml_file = open(os.path.join(save_dir, 'index.cnxml'), 'rb')
cnxml = cnxml_file.read()
cnxml_file.close()
conversionerror = None
try:
htmlpreview = cnxml_to_htmlpreview(cnxml)
save_and_backup_file(save_dir, 'index.html', htmlpreview)
files = get_files(save_dir)
save_zip(save_dir, cnxml, htmlpreview, files)
except libxml2.parserError:
conversionerror = traceback.format_exc()
raise ConversionError(conversionerror)
def process_import(save_dir_path, original_filepath, filename, download_url):
# convert from other office format to odt
odt_filename = '%s.odt' % filename
odt_filepath = str(os.path.join(save_dir_path, odt_filename))
# run jod service
converter = JOD.DocumentConverterClient()
try:
converter.convert(original_filepath, 'odt', odt_filepath)
except Exception as e:
raise e
# check file existed
try:
fp = open(odt_filepath, 'r')
fp.close()
except IOError as io:
# TODO: raise exception
raise io
# convert to cnxml
tree, files, errors = transform(odt_filepath)
cnxml = clean_cnxml(etree.tostring(tree))
# convert to html
html = cnxml_to_htmlpreview(cnxml)
# produce zipfile
ram = StringIO()
zip_archive = zipfile.ZipFile(ram, 'w')
zip_archive.writestr('index.html', html)
for fname, fdata in files.items():
zip_archive.writestr(fname, fdata)
zip_archive.close()
# save zipfile
zip_file_path = os.path.join(save_dir_path, '%s.zip' % filename)
if os.path.exists(zip_file_path):
os.rename(zip_file_path, zip_file_path + '~')
f = open(zip_file_path, 'wb')
f.write(ram.getvalue())
f.close()
return download_url
def process(self, zip_filename):
try:
self.zip_archive = zipfile.ZipFile(zip_filename, 'r')
# Unzip into transform directory
self.zip_archive.extractall(path=self.save_dir)
# Rename ZIP file so that the user can download it again
os.rename(zip_filename, os.path.join(self.save_dir, 'upload.zip'))
# Read CNXML
with open(os.path.join(self.save_dir, 'index.cnxml'), 'rt') as fp:
cnxml = fp.read()
# Convert the CNXML to XHTML for preview
html = cnxml_to_htmlpreview(cnxml)
with open(os.path.join(self.save_dir, 'index.xhtml'),
'w') as index:
index.write(html)
cnxml = clean_cnxml(cnxml)
validate_cnxml(cnxml)
except ConversionError as e:
return render_conversionerror(self.request, e.msg)
except Exception:
tb = traceback.format_exc()
self.write_traceback_to_zipfile(tb, form)
templatePath = 'templates/error.pt'
response = {'traceback': tb}
if ('title' in self.request.session):
del self.request.session['title']
return render_to_response(templatePath,
response,
request=self.request)
self.request.session.flash(self.message)
return HTTPFound(location=self.request.route_url(self.nextStep()))
def save_cnxml(save_dir, cnxml, files):
# write CNXML output
save_and_backup_file(save_dir, 'index.cnxml', cnxml)
# write files
for filename, content in files:
filename = os.path.join(save_dir, filename)
f = open(filename, 'wb') # write binary, important!
f.write(content)
f.close()
# we generate the preview and save the error
conversionerror = None
try:
htmlpreview = cnxml_to_htmlpreview(cnxml)
except libxml2.parserError:
conversionerror = traceback.format_exc()
# Zip up all the files. This is done now, since we have all the files
# available, and it also allows us to post a simple download link.
# Note that we cannot use zipfile as context manager, as that is only
# available from python 2.7
# TODO: Do a filesize check xxxx
ram = StringIO()
zip_archive = zipfile.ZipFile(ram, 'w')
zip_archive.writestr('index.cnxml', cnxml)
if not conversionerror:
save_and_backup_file(save_dir, 'index.xhtml', htmlpreview)
zip_archive.writestr('index.xhtml', htmlpreview)
for filename, fileObj in files:
zip_archive.writestr(filename, fileObj)
zip_archive.close()
zip_filename = os.path.join(save_dir, 'upload.zip')
save_and_backup_file(save_dir, zip_filename, ram.getvalue(), mode='wb')
if conversionerror:
raise ConversionError(conversionerror)
def process(self, zip_filename):
try:
self.zip_archive = zipfile.ZipFile(zip_filename, 'r')
# Unzip into transform directory
self.zip_archive.extractall(path=self.save_dir)
# Rename ZIP file so that the user can download it again
os.rename(zip_filename,
os.path.join(self.save_dir, 'upload.zip'))
# Read CNXML
with open(os.path.join(self.save_dir, 'index.cnxml'), 'rt') as fp:
cnxml = fp.read()
# Convert the CNXML to XHTML for preview
html = cnxml_to_htmlpreview(cnxml)
with open(os.path.join(self.save_dir, 'index.xhtml'), 'w') as index:
index.write(html)
cnxml = clean_cnxml(cnxml)
validate_cnxml(cnxml)
except ConversionError as e:
return render_conversionerror(self.request, e.msg)
except Exception:
tb = traceback.format_exc()
self.write_traceback_to_zipfile(tb)
templatePath = 'templates/error.pt'
response = {'traceback': tb}
if('title' in self.request.session):
del self.request.session['title']
return render_to_response(templatePath, response, request=self.request)
self.request.session.flash(self.message)
return HTTPFound(location=self.request.route_url(self.nextStep()))
def choose_view(request):
check_login(request)
templatePath = 'templates/choose.pt'
form = Form(request, schema=UploadSchema)
field_list = [('upload', 'File')]
# clear the session
if 'transformerror' in request.session:
del request.session['transformerror']
if 'title' in request.session:
del request.session['title']
# Check for successful form completion
if form.validate():
try: # Catch-all exception block
# Create a directory to do the conversions
now_string = datetime.datetime.now().strftime('%Y%m%d-%H%M%S')
# TODO: This has a good chance of being unique, but even so...
temp_dir_name = '%s-%s' % (request.session['username'], now_string)
save_dir = os.path.join(
request.registry.settings['transform_dir'],
temp_dir_name
)
os.mkdir(save_dir)
# Keep the info we need for next uploads. Note that this
# might kill the ability to do multiple tabs in parallel,
# unless it gets offloaded onto the form again.
request.session['upload_dir'] = temp_dir_name
if form.data['upload'] is not None:
request.session['filename'] = form.data['upload'].filename
# Google Docs Conversion
# if we have a Google Docs ID and Access token.
if form.data['gdocs_resource_id']:
gdocs_resource_id = form.data['gdocs_resource_id']
gdocs_access_token = form.data['gdocs_access_token']
form.data['gdocs_resource_id'] = None
form.data['gdocs_access_token'] = None
(request.session['title'], request.session['filename']) = \
process_gdocs_resource(save_dir, \
gdocs_resource_id, \
gdocs_access_token)
# HTML URL Import:
elif form.data.get('url_text'):
url = form.data['url_text']
form.data['url_text'] = None
# Build a regex for Google Docs URLs
regex = re.compile("^https:\/\/docs\.google\.com\/.*document\/[^\/]\/([^\/]+)\/")
r = regex.search(url)
# Take special action for Google Docs URLs
if r:
gdocs_resource_id = r.groups()[0]
(request.session['title'], request.session['filename']) = \
process_gdocs_resource(save_dir, "document:" + gdocs_resource_id)
else:
# download html:
#html = urllib2.urlopen(url).read()
# Simple urlopen() will fail on mediawiki websites like e.g. Wikipedia!
import_opener = urllib2.build_opener()
import_opener.addheaders = [('User-agent', 'Mozilla/5.0')]
try:
import_request = import_opener.open(url)
html = import_request.read()
# transformation
cnxml, objects, html_title = htmlsoup_to_cnxml(
html, bDownloadImages=True, base_or_source_url=url)
request.session['title'] = html_title
cnxml = clean_cnxml(cnxml)
save_cnxml(save_dir, cnxml, objects.items())
# Keep the info we need for next uploads. Note that
# this might kill the ability to do multiple tabs in
# parallel, unless it gets offloaded onto the form
# again.
request.session['filename'] = "HTML Document"
validate_cnxml(cnxml)
except urllib2.URLError, e:
request['errors'] = ['The URL %s could not be opened' %url,]
response = {
'form': FormRenderer(form),
}
return render_to_response(templatePath, response, request=request)
# Office, CNXML-ZIP or LaTeX-ZIP file
else:
# Save the original file so that we can convert, plus keep it.
original_filename = os.path.join(
save_dir,
form.data['upload'].filename.replace(os.sep, '_'))
saved_file = open(original_filename, 'wb')
input_file = form.data['upload'].file
shutil.copyfileobj(input_file, saved_file)
saved_file.close()
input_file.close()
form.data['upload'] = None
# Check if it is a ZIP file with at least index.cnxml or a LaTeX file in it
try:
zip_archive = zipfile.ZipFile(original_filename, 'r')
is_zip_archive = ('index.cnxml' in zip_archive.namelist())
# Do we have a latex file?
if not is_zip_archive:
# incoming latex.zip must contain a latex.tex file, where "latex" is the base name.
(latex_head, latex_tail) = os.path.split(original_filename)
(latex_root, latex_ext) = os.path.splitext(latex_tail)
latex_basename = latex_root
latex_filename = latex_basename + '.tex'
is_latex_archive = (latex_filename in zip_archive.namelist())
except zipfile.BadZipfile:
is_zip_archive = False
is_latex_archive = False
# ZIP package from previous conversion
if is_zip_archive:
# Unzip into transform directory
zip_archive.extractall(path=save_dir)
# Rename ZIP file so that the user can download it again
os.rename(original_filename, os.path.join(save_dir, 'upload.zip'))
# Read CNXML
with open(os.path.join(save_dir, 'index.cnxml'), 'rt') as fp:
cnxml = fp.read()
# Convert the CNXML to XHTML for preview
html = cnxml_to_htmlpreview(cnxml)
with open(os.path.join(save_dir, 'index.xhtml'), 'w') as index:
index.write(html)
cnxml = clean_cnxml(cnxml)
validate_cnxml(cnxml)
# LaTeX
elif is_latex_archive:
f = open(original_filename)
latex_archive = f.read()
# LaTeX 2 CNXML transformation
cnxml, objects = latex_to_cnxml(latex_archive, original_filename)
cnxml = clean_cnxml(cnxml)
save_cnxml(save_dir, cnxml, objects.items())
validate_cnxml(cnxml)
# OOo / MS Word Conversion
else:
# Convert from other office format to odt if needed
odt_filename = original_filename
filename, extension = os.path.splitext(original_filename)
if(extension != '.odt'):
odt_filename= '%s.odt' % filename
command = '/usr/bin/soffice -headless -nologo -nofirststartwizard "macro:///Standard.Module1.SaveAsOOO(' + escape_system(original_filename)[1:-1] + ',' + odt_filename + ')"'
os.system(command)
try:
fp = open(odt_filename, 'r')
fp.close()
except IOError as io:
raise ConversionError("%s not found" %
original_filename)
# Convert and save all the resulting files.
tree, files, errors = transform(odt_filename)
cnxml = clean_cnxml(etree.tostring(tree))
save_cnxml(save_dir, cnxml, files.items())
# now validate with jing
validate_cnxml(cnxml)
