def convert(options, args):
if options.test_mode:
import pkg_resources
print 'Entering testmode'
for fn in ('test1.html', 'test2.html', 'test3.html'):
tmpf = newTempfile()
print fn
print '-' * len(fn)
file(tmpf + '.html', 'wb').write(
pkg_resources.resource_string('zopyx.convert2.tests.data', fn))
for name in registry.availableConverters():
cls = registry.converter_registry[name]
print '%s: %s.html -> %s.%s' % (name, tmpf, tmpf,
cls.output_format)
C = Converter(tmpf + '.html', verbose=True)
try:
result = C(name,
output_filename=tmpf + '.' + cls.output_format)
print result
except Exception, e:
print 'FAILED (%s)' % e
print
def html2pdf(html_filename, output_filename=None, **options):
""" Convert a HTML file to PDF using FOP"""
if not output_filename:
output_filename = newTempfile(suffix='.pdf')
if not prince_available:
raise RuntimeError("The external PrinceXML converter isn't available")
cmd_options = list()
for k,v in options.items():
if v is None:
cmd_options.append('--%s ' % k)
else:
cmd_options.append('--%s="%s" ' % (k, v))
if sys.platform == 'win32':
raise NotImplementedError('No support for PrinceXML on Windows available')
else:
cmd = '%s "prince" "%s" %s -o "%s"' % \
(execution_shell, html_filename, ' '.join(cmd_options), output_filename)
status, output = runcmd(cmd)
if status != 0:
raise ConversionError('Error executing: %s' % cmd, output)
return dict(output_filename=output_filename,
status=status,
output=output)
def html2calibre(html_filename, output_filename=None, cmdopts='', **calibre_options):
""" Convert a HTML file using calibre """
if not html_filename.endswith('.html'):
shutil.copy(html_filename, html_filename + '.html')
html_filename += '.html'
if not output_filename:
output_filename = newTempfile(suffix='.epub')
if not calibre_available:
raise RuntimeError("The external calibre converter isn't available")
options = list()
for k,v in calibre_options.items():
if v is None:
options.append('--%s ' % k)
else:
options.append('--%s="%s" ' % (k, v))
if sys.platform == 'win32':
raise NotImplementedError('No support for using Calibre on Windows available')
else:
options = ' '.join(options)
options = options + ' ' + cmdopts
cmd = '"ebook-convert" "%s" "%s" %s' % (html_filename, output_filename, options)
status, output = runcmd(cmd)
if status != 0:
raise ConversionError('Error executing: %s' % cmd, output)
return dict(output_filename=output_filename,
status=status,
output=output)
def html2pdf(html_filename, output_filename=None, **options):
""" Convert a HTML file to PDF using FOP"""
if not output_filename:
output_filename = newTempfile(suffix='.pdf')
if not prince_available:
raise RuntimeError("The external PrinceXML converter isn't available")
cmd_options = list()
for k, v in options.items():
if v is None:
cmd_options.append('--%s ' % k)
else:
cmd_options.append('--%s="%s" ' % (k, v))
if sys.platform == 'win32':
raise NotImplementedError(
'No support for PrinceXML on Windows available')
else:
cmd = '%s "prince" "%s" %s -o "%s"' % \
(execution_shell, html_filename, ' '.join(cmd_options), output_filename)
status, output = runcmd(cmd)
if status != 0:
raise ConversionError('Error executing: %s' % cmd, output)
return dict(output_filename=output_filename, status=status, output=output)
def html2pdf(html_filename, output_filename=None, **options):
""" Convert a HTML file to PDF using FOP"""
if not output_filename:
output_filename = newTempfile(suffix='.pdf')
cmd = 'pisa --encoding utf-8 -d "%s" "%s"' % (html_filename,
output_filename)
status, output = runcmd(cmd)
return dict(output_filename=output_filename, status=0, output=output)
def html2pdf(html_filename, output_filename=None, **options):
""" Convert a HTML file to PDF using FOP"""
if not output_filename:
output_filename = newTempfile(suffix='.pdf')
cmd = 'pisa --encoding utf-8 -d "%s" "%s"' % (html_filename, output_filename)
status, output = runcmd(cmd)
return dict(output_filename=output_filename,
status=0,
output=output)
def html2pdf(html_filename, output_filename=None, **options):
""" Convert a HTML file to PDF using FOP"""
if not output_filename:
output_filename = newTempfile(suffix='.pdf')
fin = file(html_filename)
fout = file(output_filename, 'wb')
pisaDocument(fin, fout, encoding='utf-8', debug=True)
fin.close()
fout.close()
return dict(output_filename=output_filename, status=0, output='')
def convert(self, fo_filename, output_filename=None):
if not output_filename:
output_filename = newTempfile(suffix='.pdf')
cmd = 'fop -fo "%s" -pdf "%s"' % (fo_filename, output_filename)
status, output = runcmd(cmd)
if status != 0:
raise RuntimeError('Error executing: %s\n\n%s' % (cmd, output))
log.info("\n")
return output_filename
def tidyhtml(filename, encoding='utf-8', strip_base=False):
html = file(filename, 'rb').read()
# use BeautifulSoup for performing HTML checks
# and conversion to XHTML
soup = BeautifulSoup(html)
# check if all image files exist
for img in soup.findAll('img'):
src = img['src']
if not os.path.exists(src):
# try to find the image relative to the location of
# the HTML file
html_dirname = os.path.dirname(filename)
possible_img = os.path.join(html_dirname, src)
if os.path.exists(possible_img):
img['src'] = possible_img
else:
LOG.warn('No image file found: %s' % src)
html = soup.renderContents()
# replace the HTML tag
html = '<html xmlns="http://www.w3.org/1999/xhtml">' + \
html[html.find('<html') + 6:]
# add the XML preamble
html = '<?xml version="1.0" ?>\n' + html
# replace all HTML entities with numeric entities
def handler(mo):
""" Callback to convert entities """
e = mo.group(1)
v = e[1:-1]
if not v.startswith('#'):
codepoint = name2codepoint.get(v)
return codepoint and '&#%d;' % codepoint or ''
else:
return e
entity_reg = re.compile('(&.*?;)')
html = entity_reg.sub(handler, html)
# replace BASE tag
if strip_base:
base_reg = re.compile('(<base.*?>)', re.I)
html = base_reg.sub('', html)
filename = newTempfile()
file(filename, 'wb').write(str(html))
return filename
def tidyhtml(filename, encoding="utf-8", strip_base=False):
html = file(filename, "rb").read()
# use BeautifulSoup for performing HTML checks
# and conversion to XHTML
soup = BeautifulSoup(html)
# check if all image files exist
for img in soup.findAll("img"):
src = img["src"]
if not os.path.exists(src):
# try to find the image relative to the location of
# the HTML file
html_dirname = os.path.dirname(filename)
possible_img = os.path.join(html_dirname, src)
if os.path.exists(possible_img):
img["src"] = possible_img
else:
LOG.warn("No image file found: %s" % src)
html = soup.renderContents()
# replace the HTML tag
html = '<html xmlns="http://www.w3.org/1999/xhtml">' + html[html.find("<html") + 6 :]
# add the XML preamble
html = '<?xml version="1.0" ?>\n' + html
# replace all HTML entities with numeric entities
def handler(mo):
""" Callback to convert entities """
e = mo.group(1)
v = e[1:-1]
if not v.startswith("#"):
codepoint = name2codepoint.get(v)
return codepoint and "&#%d;" % codepoint or ""
else:
return e
entity_reg = re.compile("(&.*?;)")
html = entity_reg.sub(handler, html)
# replace BASE tag
if strip_base:
base_reg = re.compile("(<base.*?>)", re.I)
html = base_reg.sub("", html)
filename = newTempfile()
file(filename, "wb").write(str(html))
return filename
def html2pdf(html_filename, output_filename=None, **options):
""" Convert a HTML file to PDF using FOP"""
if not output_filename:
output_filename = newTempfile(suffix='.pdf')
if not pdfreactor_available:
raise RuntimeError("The external 'pdfreactor' converter isn't available")
cmd = '%s "pdfreactor" "%s" "%s"' % \
(execution_shell, html_filename, output_filename)
status, output = runcmd(cmd)
if status != 0:
raise ConversionError('Error executing: %s' % cmd, output)
return dict(output_filename=output_filename,
status=status,
output=output)
def fo2pdf(fo_filename, output_filename=None):
""" Convert a FO file to PDF using XINC """
if not output_filename:
output_filename = newTempfile(suffix=".pdf")
if not xinc_available:
raise RuntimeError("The external XINC converter isn't available")
if sys.platform == "win32":
cmd = '%s\\bin\\windows\\xinc.exe -fo "%s" -pdf "%s"' % (xinc_home, fo_filename, output_filename)
else:
cmd = '"%s/bin/unix/xinc" -fo "%s" -pdf "%s"' % (xinc_home, fo_filename, output_filename)
status, output = runcmd(cmd)
if status != 0:
raise ConversionError("Error executing: %s" % cmd, output)
return dict(output_filename=output_filename, status=status, output=output)
def html2pdf(html_filename, output_filename=None, **options):
""" Convert a HTML file to PDF using FOP"""
if not output_filename:
output_filename = newTempfile(suffix='.pdf')
fin = file(html_filename)
fout = file(output_filename, 'wb')
pisaDocument(fin,
fout,
encoding='utf-8',
debug=True)
fin.close()
fout.close()
return dict(output_filename=output_filename,
status=0,
output='')
def fo2pdf(fo_filename, output_filename=None):
""" Convert a FO file to PDF using XINC """
if not output_filename:
output_filename = newTempfile(suffix='.pdf')
if not xinc_available:
raise RuntimeError("The external XINC converter isn't available")
if sys.platform == 'win32':
cmd = '%s\\bin\\windows\\xinc.exe -fo "%s" -pdf "%s"' % (
xinc_home, fo_filename, output_filename)
else:
cmd = '"%s/bin/unix/xinc" -fo "%s" -pdf "%s"' % (
xinc_home, fo_filename, output_filename)
status, output = runcmd(cmd)
if status != 0:
raise ConversionError('Error executing: %s' % cmd, output)
return dict(output_filename=output_filename, status=status, output=output)
def fo2pdf(fo_filename, output_filename=None):
""" Convert a FO file to PDF using FOP"""
if not output_filename:
output_filename = newTempfile(suffix='.pdf')
if not fop_available:
raise RuntimeError("The external FOP converter isn't available")
if sys.platform == 'win32':
cmd = '%s\\fop.bat -fo "%s" -pdf "%s"' % (fop_home, fo_filename,
output_filename)
else:
cmd = '%s "%s/fop" -fo "%s" -pdf "%s"' % \
(execution_shell, fop_home, fo_filename, output_filename)
status, output = runcmd(cmd)
if status != 0:
raise ConversionError('Error executing: %s' % cmd, output)
return dict(output_filename=output_filename, status=status, output=output)
def fo2xfc(fo_filename, format="rtf", output_filename=None):
""" Convert a FO file to some format support
through XFC-4.0.
"""
if not format in ("rtf", "docx", "wml", "odt"):
raise ValueError("Unsupported format: %s" % format)
if not output_filename:
output_filename = newTempfile(suffix=".%s" % format)
if sys.platform == "win32":
cmd = '"%s\\fo2%s.bat" "%s" "%s"' % (xfc_dir, format, fo_filename, output_filename)
else:
cmd = '"%s/fo2%s" "%s" "%s"' % (xfc_dir, format, fo_filename, output_filename)
status, output = runcmd(cmd)
if status != 0:
raise ConversionError("Error executing: %s" % cmd, output)
return dict(output_filename=output_filename, status=status, output=output)
def fo2pdf(fo_filename, output_filename=None):
""" Convert a FO file to PDF using FOP"""
if not output_filename:
output_filename = newTempfile(suffix='.pdf')
if not fop_available:
raise RuntimeError("The external FOP converter isn't available")
if sys.platform == 'win32':
cmd = '%s\\fop.bat -fo "%s" -pdf "%s"' % (fop_home, fo_filename, output_filename)
else:
cmd = '%s "%s/fop" -fo "%s" -pdf "%s"' % \
(execution_shell, fop_home, fo_filename, output_filename)
status, output = runcmd(cmd)
if status != 0:
raise ConversionError('Error executing: %s' % cmd, output)
return dict(output_filename=output_filename,
status=status,
output=output)
def html2calibre(html_filename,
output_filename=None,
cmdopts='',
**calibre_options):
""" Convert a HTML file using calibre """
if not html_filename.endswith('.html'):
shutil.copy(html_filename, html_filename + '.html')
html_filename += '.html'
if not output_filename:
output_filename = newTempfile(suffix='.epub')
if not calibre_available:
raise RuntimeError("The external calibre converter isn't available")
options = list()
for k, v in calibre_options.items():
if v is None:
options.append('--%s ' % k)
else:
options.append('--%s="%s" ' % (k, v))
if sys.platform == 'win32':
raise NotImplementedError(
'No support for using Calibre on Windows available')
else:
options = ' '.join(options)
options = options + ' ' + cmdopts
cmd = '"ebook-convert" "%s" "%s" %s' % (html_filename, output_filename,
options)
status, output = runcmd(cmd)
if status != 0:
raise ConversionError('Error executing: %s' % cmd, output)
return dict(output_filename=output_filename, status=status, output=output)
def convert(options, args):
if options.test_mode:
import pkg_resources
print 'Entering testmode'
for fn in ('test1.html', 'test2.html', 'test3.html'):
tmpf = newTempfile()
print fn
print '-'*len(fn)
file(tmpf + '.html', 'wb').write(pkg_resources.resource_string('zopyx.convert2.tests.data', fn))
for name in registry.availableConverters():
cls = registry.converter_registry[name]
print '%s: %s.html -> %s.%s' % (name, tmpf, tmpf, cls.output_format)
C = Converter(tmpf + '.html', verbose=True)
try:
result = C(name, output_filename=tmpf + '.' + cls.output_format)
print result
except Exception, e:
print 'FAILED (%s)' % e
print
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