def convert(self, output_filename=None, **calibre_options):
# check for commandlineoptions.txt file
cmdopts = ''
cmdopts_filename = os.path.join(os.path.dirname(self.filename), 'commandlineoptions.txt')
if os.path.exists(cmdopts_filename):
cmdopts = file(cmdopts_filename).read()
cmdopts = cmdopts % dict(WORKDIR=os.path.dirname(self.filename))
tidy_filename = tidyhtml(self.filename, self.encoding)
result = html2calibre(tidy_filename, output_filename, cmdopts=cmdopts, **calibre_options)
os.unlink(tidy_filename)
return result
def convert(self, output_filename=None, **calibre_options):
# check for commandlineoptions.txt file
cmdopts = ''
cmdopts_filename = os.path.join(os.path.dirname(self.filename),
'commandlineoptions.txt')
if os.path.exists(cmdopts_filename):
cmdopts = file(cmdopts_filename).read()
cmdopts = cmdopts % dict(WORKDIR=os.path.dirname(self.filename))
tidy_filename = tidyhtml(self.filename, self.encoding)
result = html2calibre(tidy_filename,
output_filename,
cmdopts=cmdopts,
**calibre_options)
os.unlink(tidy_filename)
return result
def convert(self, output_filename=None, **options):
tidy_filename = tidyhtml(self.filename, self.encoding)
result = html2pdf(tidy_filename, output_filename, **options)
os.unlink(tidy_filename)
return result
def convert(self, output_filename=None, **options):
tidy_filename = tidyhtml(self.filename, self.encoding)
result = html2pdf(tidy_filename, output_filename, **options)
os.unlink(tidy_filename)
return result
def convert(self, filename, encoding='utf-8', tidy=True, output_filename=None, **kw):
""" Convert a HTML file stored as 'filename' to
FO using CSS2XSLFO.
"""
if tidy:
filename = tidyhtml(filename, encoding, strip_base=kw.get('strip_base', False))
if output_filename:
fo_filename = output_filename
else:
fo_filename = newTempfile(suffix='.fo')
csstoxslfo = os.path.abspath(os.path.join(dirname, 'lib', 'csstoxslfo', 'css2xslfo.jar'))
if not os.path.exists(csstoxslfo):
raise IOError('%s does not exist' % csstoxslfo)
cmd = '"%s"' % java + \
' -Duser.language=en -Xms256m -Xmx256m -jar "%(csstoxslfo)s" "%(filename)s" -fo "%(fo_filename)s"' % vars()
for k in kw:
cmd += ' %s="%s"' % (k, kw[k])
status, output = runcmd(cmd)
if status != 0:
raise ConversionError('Error executing: %s' % cmd, output)
# remove tidy-ed file
if tidy:
os.unlink(filename)
# remove some stuff from the generated FO file causing
# some conversion trouble either with XINC or XFC
E = parse(fo_filename)
ids_seen = list()
for node in E.getiterator():
get = node.attrib.get
# ensure that ID attributes are unique
node_id = get('id')
if node_id is not None:
if node_id in ids_seen:
del node.attrib['id']
ids_seen.append(node_id)
for k, v in (('footnote', 'reset'),
('unicode-bidi', 'embed'),
('writing-mode', 'lr-tb'),
('font-selection-strategy', 'character-by-character'),
('line-height-shift-adjustment', 'disregard-shifts'),
('page-break-after', 'avoid'),
('page-break-before', 'avoid'),
('page-break-inside', 'avoid')):
value = get(k)
if value == v:
del node.attrib[k]
for attr in ('margin-left', 'margin-right', 'margin-top', 'margin-bottom',
'padding-left', 'padding-right', 'padding-top', 'padding-bottom'):
value = get(attr)
if value == '0':
node.attrib[attr] = '0em'
if get('page-break-after') == 'always':
del node.attrib['page-break-after']
node.attrib['break-after'] = 'page'
if get('text-transform'):
del node.attrib['text-transform']
value = get('white-space')
if value == 'pre':
del node.attrib['white-space']
node.text = '\n' + node.text.lstrip()
for k,v in {'white-space-treatment' : 'preserve',
'white-space-collapse' : 'false',
'wrap-option' : 'no-wrap',
'linefeed-treatment' : 'preserve' }.items():
node.attrib[k] = v
fo_text = tostring(E.getroot())
fo_text = fo_text.replace('<ns0:block ' , '<ns0:block margin-top="0" margin-bottom="0" ') # avoid a linebreak through <li><p> (XFC)
# fo_text = fo_text.replace('<ns0:block/>', '') # causes a crash with XINC
fo_text = fo_text.replace('<ns0:block margin-top="0" margin-bottom="0" />', '')
file(fo_filename, 'wb').write(fo_text)
return fo_filename
def convert(self, filename, encoding="utf-8", tidy=True, output_filename=None, **kw):
""" Convert a HTML file stored as 'filename' to
FO using CSS2XSLFO.
"""
if tidy:
filename = tidyhtml(filename, encoding, strip_base=kw.get("strip_base", False))
if output_filename:
fo_filename = output_filename
else:
fo_filename = newTempfile(suffix=".fo")
csstoxslfo = os.path.abspath(os.path.join(dirname, "lib", "csstoxslfo", "css2xslfo.jar"))
if not os.path.exists(csstoxslfo):
raise IOError("%s does not exist" % csstoxslfo)
cmd = (
'"%s"' % java
+ ' -Duser.language=en -Xms256m -Xmx256m -jar "%(csstoxslfo)s" "%(filename)s" -fo "%(fo_filename)s"'
% vars()
)
for k in kw:
cmd += ' %s="%s"' % (k, kw[k])
status, output = runcmd(cmd)
if status != 0:
raise ConversionError("Error executing: %s" % cmd, output)
# remove tidy-ed file
if tidy:
os.unlink(filename)
# remove some stuff from the generated FO file causing
# some conversion trouble either with XINC or XFC
E = parse(fo_filename)
ids_seen = list()
for node in E.getiterator():
get = node.attrib.get
# ensure that ID attributes are unique
node_id = get("id")
if node_id is not None:
if node_id in ids_seen:
del node.attrib["id"]
ids_seen.append(node_id)
for k, v in (
("footnote", "reset"),
("unicode-bidi", "embed"),
("writing-mode", "lr-tb"),
("font-selection-strategy", "character-by-character"),
("line-height-shift-adjustment", "disregard-shifts"),
("page-break-after", "avoid"),
("page-break-before", "avoid"),
("page-break-inside", "avoid"),
):
value = get(k)
if value == v:
del node.attrib[k]
for attr in (
"margin-left",
"margin-right",
"margin-top",
"margin-bottom",
"padding-left",
"padding-right",
"padding-top",
"padding-bottom",
):
value = get(attr)
if value == "0":
node.attrib[attr] = "0em"
if get("page-break-after") == "always":
del node.attrib["page-break-after"]
node.attrib["break-after"] = "page"
if get("text-transform"):
del node.attrib["text-transform"]
value = get("white-space")
if value == "pre":
del node.attrib["white-space"]
node.text = "\n" + node.text.lstrip()
for k, v in {
"white-space-treatment": "preserve",
"white-space-collapse": "false",
"wrap-option": "no-wrap",
"linefeed-treatment": "preserve",
}.items():
node.attrib[k] = v
fo_text = tostring(E.getroot())
fo_text = fo_text.replace(
"<ns0:block ", '<ns0:block margin-top="0" margin-bottom="0" '
) # avoid a linebreak through <li><p> (XFC)
# fo_text = fo_text.replace('<ns0:block/>', '') # causes a crash with XINC
fo_text = fo_text.replace('<ns0:block margin-top="0" margin-bottom="0" />', "")
file(fo_filename, "wb").write(fo_text)
return fo_filename
def convert(self,
filename,
encoding='utf-8',
tidy=True,
output_filename=None,
**kw):
""" Convert a HTML file stored as 'filename' to
FO using CSS2XSLFO.
"""
if tidy:
filename = tidyhtml(filename,
encoding,
strip_base=kw.get('strip_base', False))
if output_filename:
fo_filename = output_filename
else:
fo_filename = newTempfile(suffix='.fo')
csstoxslfo = os.path.abspath(
os.path.join(dirname, 'lib', 'csstoxslfo', 'css2xslfo.jar'))
if not os.path.exists(csstoxslfo):
raise IOError('%s does not exist' % csstoxslfo)
cmd = '"%s"' % java + \
' -Duser.language=en -Xms256m -Xmx256m -jar "%(csstoxslfo)s" "%(filename)s" -fo "%(fo_filename)s"' % vars()
for k in kw:
cmd += ' %s="%s"' % (k, kw[k])
status, output = runcmd(cmd)
if status != 0:
raise ConversionError('Error executing: %s' % cmd, output)
# remove tidy-ed file
if tidy:
os.unlink(filename)
# remove some stuff from the generated FO file causing
# some conversion trouble either with XINC or XFC
E = parse(fo_filename)
ids_seen = list()
for node in E.getiterator():
get = node.attrib.get
# ensure that ID attributes are unique
node_id = get('id')
if node_id is not None:
if node_id in ids_seen:
del node.attrib['id']
ids_seen.append(node_id)
for k, v in (('footnote', 'reset'), ('unicode-bidi', 'embed'),
('writing-mode', 'lr-tb'), ('font-selection-strategy',
'character-by-character'),
('line-height-shift-adjustment',
'disregard-shifts'), ('page-break-after', 'avoid'),
('page-break-before', 'avoid'), ('page-break-inside',
'avoid')):
value = get(k)
if value == v:
del node.attrib[k]
for attr in ('margin-left', 'margin-right', 'margin-top',
'margin-bottom', 'padding-left', 'padding-right',
'padding-top', 'padding-bottom'):
value = get(attr)
if value == '0':
node.attrib[attr] = '0em'
if get('page-break-after') == 'always':
del node.attrib['page-break-after']
node.attrib['break-after'] = 'page'
if get('text-transform'):
del node.attrib['text-transform']
value = get('white-space')
if value == 'pre':
del node.attrib['white-space']
node.text = '\n' + node.text.lstrip()
for k, v in {
'white-space-treatment': 'preserve',
'white-space-collapse': 'false',
'wrap-option': 'no-wrap',
'linefeed-treatment': 'preserve'
}.items():
node.attrib[k] = v
fo_text = tostring(E.getroot())
fo_text = fo_text.replace(
'<ns0:block ', '<ns0:block margin-top="0" margin-bottom="0" '
) # avoid a linebreak through <li><p> (XFC)
# fo_text = fo_text.replace('<ns0:block/>', '') # causes a crash with XINC
fo_text = fo_text.replace(
'<ns0:block margin-top="0" margin-bottom="0" />', '')
file(fo_filename, 'wb').write(fo_text)
return fo_filename