def testFunc(self, file_name=file_name):
assert file_name.endswith(".src.html")
base_path = file_name[:-len(".src.html")]
kwargs = {}
try:
options_file_name = base_path + ".options"
with open(options_file_name, "r") as options_file:
kwargs = json.load(options_file)
except IOError:
pass
default_processes = ["filter", "sub", "toc", "xref", "annotate"]
new_processes = kwargs.get("processes", [])
assert not set(default_processes) & set(new_processes)
kwargs["processes"] = default_processes + new_processes
# Sort attributes alphabetically by default.
kwargs["alphabetical_attributes"] = True
try:
output = StringIO.StringIO()
# Get the input
input = open(file_name, "rb")
tree = generator.fromFile(input, **kwargs)
input.close()
# Get the output
generator.toFile(tree, output, **kwargs)
# Get the expected result
expectedfp = open(base_path + ".html", "rb")
expected = expectedfp.read()
expectedfp.close()
# Run the test
self.assertEquals(output.getvalue(), expected)
except IOError as err:
self.fail(err)
def testFunc(self, file_name=file_name):
try:
# Get the input
input = open(file_name, "rb")
tree = generator.fromFile(input)
input.close()
# Get the output
output = StringIO.StringIO()
generator.toFile(tree, output)
# Get the expected result
expected = open(file_name[:-9] + ".html", "rb")
# Run the test
self.assertEquals(output.getvalue(), expected.read())
# Close the files
output.close()
expected.close()
except IOError, err:
self.fail(err)
'w3c_compat_xref_normalization': False,
}
print 'indexing'
filtered.seek(0)
tree = generator.fromFile(filtered, **opts)
filtered.close()
try:
os.makedirs('output/%s' % spec)
except:
pass
if spec == 'html':
from glob import glob
for name in glob('output/html/*.html'):
os.remove(name)
output = open('output/html/single-page.html', 'wb')
else:
output = open('output/%s/Overview.html' % spec, 'wb')
generator.toFile(tree, output, **opts)
output.close()
if spec == 'html':
print 'splitting'
import spec_splitter
spec_splitter.w3c = True
spec_splitter.main('output/html/single-page.html', 'output/html')
def main(spec, spec_dir, branch="master"):
conf = None
try:
conf = config.load_config()[spec]
except KeyError:
invoked_incorrectly()
if 'select' in conf:
select = conf['select']
else:
select = spec
try:
if not spec_dir:
spec_dir = os.path.join(conf["output"], spec)
except KeyError:
sys.stderr.write("error: Must specify output directory for %s! \
Check default-config.json.\n" % spec)
exit()
cur_dir = os.path.abspath(os.path.dirname(__file__))
os.chdir(conf["path"])
print "parsing"
source = open('source')
after_microsyntax = StringIO()
parser_microsyntax.main(source, after_microsyntax)
after_microsyntax.seek(0)
succint = StringIO()
bs.main(after_microsyntax, succint)
succint.seek(0)
filtered = StringIO()
try:
boilerplate.main(succint, filtered, select, branch)
except IOError:
sys.stderr.write("error: Problem loading boilerplate for %s. \
Are you on the correct branch?\n" % spec)
exit()
succint.close()
# See http://hg.gsnedders.com/anolis/file/tip/anolis
opts = {
'allow_duplicate_dfns': True,
'disable': None,
'escape_lt_in_attrs': False,
'escape_rcdata': False,
'force_html4_id': False,
'indent_char': u' ',
'inject_meta_charset': False,
'max_depth': 6,
'min_depth': 2,
'minimize_boolean_attributes': False,
'newline_char': u'\n',
'omit_optional_tags': False,
'output_encoding': 'utf-8',
'parser': 'html5lib',
'processes': set(['toc', 'xref', 'sub']),
'profile': False,
'quote_attr_values': True,
'serializer': 'html5lib',
'space_before_trailing_solidus': False,
'strip_whitespace': None,
'use_best_quote_char': False,
'use_trailing_solidus': False,
'w3c_compat_class_toc': False,
'w3c_compat_crazy_substitutions': False,
'w3c_compat_substitutions': False,
'w3c_compat': True,
'w3c_compat_xref_a_placement': False,
'w3c_compat_xref_elements': False,
'w3c_compat_xref_normalization': False,
}
if "anolis" in conf:
opts.update(conf["anolis"])
if spec == "srcset":
import html5lib
print 'munging (before anolis)'
filtered.seek(0)
pre_anolis_buffer = StringIO()
# Parse
parser = html5lib.html5parser.HTMLParser(tree = html5lib.treebuilders.getTreeBuilder('lxml'))
tree = parser.parse(filtered, encoding='utf-8')
# Move introduction above conformance requirements
introduction = tree.findall("//*[@id='introduction']")[0]
intro_ps = introduction.xpath("following-sibling::*")
target = tree.findall("//*[@id='conformance-requirements']")[0]
target.addprevious(introduction)
target = introduction
target.addnext(intro_ps[2])
target.addnext(intro_ps[1])
target.addnext(intro_ps[0])
# Serialize
tokens = html5lib.treewalkers.getTreeWalker('lxml')(tree)
serializer = html5lib.serializer.HTMLSerializer(quote_attr_values=True, inject_meta_charset=False)
for text in serializer.serialize(tokens, encoding='utf-8'):
pre_anolis_buffer.write(text)
filtered = pre_anolis_buffer
print 'indexing'
filtered.seek(0)
tree = generator.fromFile(filtered, **opts)
filtered.close()
# fixup nested dd's and dt's produced by lxml
for dd in tree.findall('//dd/dd'):
if list(dd) or dd.text.strip():
dd.getparent().addnext(dd)
else:
dd.getparent().remove(dd)
for dt in tree.findall('//dt/dt'):
if list(dt) or dt.text.strip():
dt.getparent().addnext(dt)
else:
dt.getparent().remove(dt)
if spec == "microdata":
print 'munging'
import lxml
# get the h3 for the misplaced section (it has no container)
section = tree.xpath("//h3[@id = 'htmlpropertiescollection']")[0]
# then get all of its following siblings that have the h2 for the next section as
# a following sibling themselves. Yeah, XPath doesn't suck.
section_content = section.xpath("following-sibling::*[following-sibling::h2[@id='introduction']]")
target = tree.xpath("//h2[@id = 'converting-html-to-other-formats']")[0].getparent()
target.addprevious(section)
for el in section_content: target.addprevious(el)
section.xpath("span")[0].text = "6.1 "
# move the toc as well
link = tree.xpath("//ol[@class='toc']//a[@href='#htmlpropertiescollection']")[0]
link.xpath("span")[0].text = "6.1 "
tree.xpath("//ol[@class='toc']/li[a[@href='#microdata-dom-api']]")[0].append(link.getparent().getparent())
if spec == "srcset":
print 'munging (after anolis)'
# In the WHATWG spec, srcset="" is simply an aspect of
# HTMLImageElement and not a separate feature. In order to keep
# the HTML WG's srcset="" spec organized, we have to move some
# things around in the final document.
# Move "The srcset IDL attribute must reflect..."
reflect_the_content_attribute = tree.findall("//div[@class='impl']")[0]
target = tree.find("//div[@class='note']")
target.addprevious(reflect_the_content_attribute)
# Move "The IDL attribute complete must return true..."
note_about_complete = tree.findall("//p[@class='note']")[5]
p_otherwise = note_about_complete.xpath("preceding-sibling::p[position()=1]")[0]
ul_conditions = p_otherwise.xpath("preceding-sibling::ul[position()=1]")[0]
p_start = ul_conditions.xpath("preceding-sibling::p[position()=1]")[0]
target.addnext(note_about_complete)
target.addnext(p_otherwise)
target.addnext(ul_conditions)
target.addnext(p_start)
try:
os.makedirs(spec_dir)
except:
pass
if spec == 'html':
print 'cleaning'
from glob import glob
for name in glob("%s/*.html" % spec_dir):
os.remove(name)
output = StringIO()
else:
output = open("%s/Overview.html" % spec_dir, 'wb')
generator.toFile(tree, output, **opts)
if spec != 'html':
output.close()
else:
value = output.getvalue()
if "<!--INTERFACES-->\n" in value:
print 'interfaces'
from interface_index import interface_index
output.seek(0)
index = StringIO()
interface_index(output, index)
value = value.replace("<!--INTERFACES-->\n", index.getvalue(), 1)
index.close()
output = open("%s/single-page.html" % spec_dir, 'wb')
output.write(value)
output.close()
value = ''
print 'splitting'
import spec_splitter
spec_splitter.w3c = True
spec_splitter.no_split_exceptions = conf.get("no_split_exceptions", False)
spec_splitter.minimal_split_exceptions = conf.get("minimal_split_exceptions", False)
spec_splitter.main("%s/single-page.html" % spec_dir, spec_dir)
print 'entities'
entities = open(os.path.join(cur_dir, "boilerplate/entities.inc"))
json = open("%s/entities.json" % spec_dir, 'w')
from entity_processor_json import entity_processor_json
entity_processor_json(entities, json)
entities.close()
json.close()
# copying dependencies
def copy_dependencies (targets):
import types
if not isinstance(targets, types.ListType): targets = [targets]
for target in targets:
os.system("/bin/csh -i -c '/bin/cp -R %s %s'" % (os.path.join(conf["path"], target), spec_dir))
print "copying"
if spec == "html":
copy_dependencies(["images", "fonts", "404/*", "switcher", "js"])
elif spec == "2dcontext":
copy_dependencies(["images", "fonts"])
else:
copy_dependencies("fonts")
# fix the styling of the 404
if spec == "html":
link = tree.xpath("//link[starts-with(@href, 'http://www.w3.org/StyleSheets/TR/')]")[0].get("href")
path = os.path.join(spec_dir, "404.html")
with open(path) as data: html404 = data.read()
html404 = html404.replace("http://www.w3.org/StyleSheets/TR/W3C-ED", link)
with open(path, "w") as data: data.write(html404)
def main(spec, spec_dir, branch="master"):
conf = None
try:
conf = config.load_config()[spec]
except KeyError:
invoked_incorrectly()
if "select" in conf:
select = conf["select"]
else:
select = spec
try:
if not spec_dir:
if conf.get("bareOutput", False):
spec_dir = conf["output"]
else:
spec_dir = os.path.join(conf["output"], spec)
except KeyError:
sys.stderr.write(
"error: Must specify output directory for %s! \
Check default-config.json.\n"
% spec
)
exit()
cur_dir = os.path.abspath(os.path.dirname(__file__))
os.chdir(conf["path"])
print "parsing"
source = open("source")
after_microsyntax = StringIO()
parser_microsyntax.main(source, after_microsyntax)
after_microsyntax.seek(0)
succint = StringIO()
bs.main(after_microsyntax, succint)
succint.seek(0)
filtered = StringIO()
if spec == "microdata":
md_content = succint.read()
md_content = re.sub(
'<h2 id="iana">IANA considerations</h2>',
'<!--BOILERPLATE microdata-extra-section--><h2 id="iana">IANA considerations</h2>',
md_content,
)
succint = StringIO()
succint.write(md_content)
succint.seek(0)
try:
boilerplate.main(succint, filtered, select, branch)
except IOError:
sys.stderr.write(
"error: Problem loading boilerplate for %s. \
Are you on the correct branch?\n"
% spec
)
exit()
succint.close()
# See http://hg.gsnedders.com/anolis/file/tip/anolis
opts = {
"allow_duplicate_dfns": True,
"disable": None,
"escape_lt_in_attrs": False,
"escape_rcdata": False,
"force_html4_id": False,
"indent_char": u" ",
"inject_meta_charset": False,
"max_depth": 6,
"min_depth": 2,
"minimize_boolean_attributes": False,
"newline_char": u"\n",
"omit_optional_tags": False,
"output_encoding": "utf-8",
"parser": "html5lib",
"processes": set(["toc", "xref", "sub"]),
"profile": False,
"quote_attr_values": True,
"serializer": "html5lib",
"space_before_trailing_solidus": False,
"strip_whitespace": None,
"use_best_quote_char": False,
"use_trailing_solidus": False,
"w3c_compat_class_toc": False,
"w3c_compat_crazy_substitutions": False,
"w3c_compat_substitutions": False,
"w3c_compat": True,
"w3c_compat_xref_a_placement": False,
"w3c_compat_xref_elements": False,
"w3c_compat_xref_normalization": False,
}
if "anolis" in conf:
opts.update(conf["anolis"])
if spec == "srcset":
print "munging (before anolis)"
filtered.seek(0)
pre_anolis_buffer = StringIO()
# Parse
parser = html5lib.html5parser.HTMLParser(tree=html5lib.treebuilders.getTreeBuilder("lxml"))
tree = parser.parse(filtered, encoding="utf-8")
# Move introduction above conformance requirements
introduction = tree.findall("//*[@id='introduction']")[0]
intro_ps = introduction.xpath("following-sibling::*")
target = tree.findall("//*[@id='conformance-requirements']")[0]
target.addprevious(introduction)
target = introduction
target.addnext(intro_ps[2])
target.addnext(intro_ps[1])
target.addnext(intro_ps[0])
# Serialize
tokens = html5lib.treewalkers.getTreeWalker("lxml")(tree)
serializer = html5lib.serializer.HTMLSerializer(quote_attr_values=True, inject_meta_charset=False)
for text in serializer.serialize(tokens, encoding="utf-8"):
pre_anolis_buffer.write(text)
filtered = pre_anolis_buffer
# replace data-x with data-anolis-xref
print "fixing xrefs"
filtered.seek(0)
# Parse
builder = treebuilders.getTreeBuilder("lxml", etree)
try:
parser = html5lib.HTMLParser(tree=builder, namespaceHTMLElements=False)
except TypeError:
parser = html5lib.HTMLParser(tree=builder)
tree = parser.parse(filtered, encoding="utf-8")
# Move introduction above conformance requirements
data_x = tree.findall("//*[@data-x]")
non_alphanumeric_spaces = re.compile(r"[^a-zA-Z0-9 \-\_\/\|]+")
for refel in data_x:
refel.attrib["data-anolis-xref"] = refel.get("data-x")
if refel.tag == "dfn" and not refel.get("id", False) and refel.attrib["data-anolis-xref"]:
refel.attrib["id"] = generateID(refel.attrib["data-anolis-xref"], refel)
del refel.attrib["data-x"]
# utils.ids = {}
print "indexing"
# filtered.seek(0)
# tree = generator.fromFile(filtered, **opts)
generator.process(tree, **opts)
filtered.close()
# fixup nested dd's and dt's produced by lxml
for dd in tree.findall("//dd/dd"):
if list(dd) or dd.text.strip():
dd.getparent().addnext(dd)
else:
dd.getparent().remove(dd)
for dt in tree.findall("//dt/dt"):
if list(dt) or dt.text.strip():
dt.getparent().addnext(dt)
else:
dt.getparent().remove(dt)
# remove unused references
print "processing references"
for dt in tree.findall("//dt[@id]"):
refID = dt.get("id")
if refID.startswith("refs") and len(tree.findall("//a[@href='#%s']" % refID)) == 0:
next = dt.getnext()
while next.tag != "dd":
next = next.getnext()
dt.getparent().remove(next)
dt.getparent().remove(dt)
elif refID.startswith("refs"):
dd = dt.getnext()
while dd.tag != "dd":
dd = dd.getnext()
links = dd.findall(".//a[@href]")
for link in links:
if link is not None:
wrap = link.getparent()
link.tail = " (URL: "
idx = wrap.index(link)
url = etree.Element("a", href=link.get("href"))
url.text = link.get("href")
wrap.insert(idx + 1, url)
url.tail = ")"
if spec == "microdata":
print "munging (after anolis)"
# get the h3 for the misplaced section (it has no container)
section = tree.xpath("//h3[@id = 'htmlpropertiescollection']")[0]
# then get all of its following siblings that have the h2 for the next section as
# a following sibling themselves. Yeah, XPath doesn't suck.
section_content = section.xpath("following-sibling::*[following-sibling::h2[@id='introduction']]")
target = tree.xpath("//h2[@id = 'converting-html-to-other-formats']")[0].getparent()
target.addprevious(section)
for el in section_content:
target.addprevious(el)
section.xpath("span")[0].text = "6.1 "
# move the toc as well
link = tree.xpath("//ol[@class='toc']//a[@href='#htmlpropertiescollection']")[0]
link.xpath("span")[0].text = "6.1 "
tree.xpath("//ol[@class='toc']/li[a[@href='#microdata-dom-api']]")[0].append(link.getparent().getparent())
if spec == "srcset":
print "munging (after anolis)"
# In the WHATWG spec, srcset="" is simply an aspect of
# HTMLImageElement and not a separate feature. In order to keep
# the HTML WG's srcset="" spec organized, we have to move some
# things around in the final document.
# Move "The srcset IDL attribute must reflect..."
reflect_the_content_attribute = tree.findall("//div[@class='impl']")[0]
target = tree.find("//div[@class='note']")
target.addprevious(reflect_the_content_attribute)
# Move "The IDL attribute complete must return true..."
note_about_complete = tree.findall("//p[@class='note']")[4]
p_otherwise = note_about_complete.xpath("preceding-sibling::p[position()=1]")[0]
ul_conditions = p_otherwise.xpath("preceding-sibling::ul[position()=1]")[0]
p_start = ul_conditions.xpath("preceding-sibling::p[position()=1]")[0]
target.addnext(note_about_complete)
target.addnext(p_otherwise)
target.addnext(ul_conditions)
target.addnext(p_start)
try:
os.makedirs(spec_dir)
except:
pass
if spec == "html":
print "cleaning"
from glob import glob
for name in glob("%s/*.html" % spec_dir):
os.remove(name)
output = StringIO()
else:
output = open("%s/Overview.html" % spec_dir, "wb")
generator.toFile(tree, output, **opts)
if spec != "html":
output.close()
else:
value = output.getvalue()
if "<!--INTERFACES-->\n" in value:
print "interfaces"
from interface_index import interface_index
output.seek(0)
index = StringIO()
interface_index(output, index)
value = value.replace("<!--INTERFACES-->\n", index.getvalue(), 1)
index.close()
output = open("%s/single-page.html" % spec_dir, "wb")
output.write(value)
output.close()
value = ""
print "splitting"
import spec_splitter
spec_splitter.w3c = True
spec_splitter.no_split_exceptions = conf.get("no_split_exceptions", False)
spec_splitter.minimal_split_exceptions = conf.get("minimal_split_exceptions", False)
spec_splitter.main("%s/single-page.html" % spec_dir, spec_dir)
print "entities"
entities = open(os.path.join(cur_dir, "boilerplate/entities.inc"))
json = open("%s/entities.json" % spec_dir, "w")
from entity_processor_json import entity_processor_json
entity_processor_json(entities, json)
entities.close()
json.close()
# copying dependencies
def copy_dependencies(targets):
import types
if not isinstance(targets, types.ListType):
targets = [targets]
if os.name == "nt":
for target in targets:
os.system("xcopy /s %s %s" % (os.path.join(conf["path"], target), spec_dir))
else:
for target in targets:
os.system("/bin/csh -i -c '/bin/cp -R %s %s'" % (os.path.join(conf["path"], target), spec_dir))
print "copying"
if spec == "html":
if os.name == "nt":
dirs = ["images", "fonts", "404", "switcher", "js"]
else:
dirs = ["images", "fonts", "404/*", "switcher", "js"]
copy_dependencies(dirs)
elif spec == "2dcontext":
copy_dependencies(["images", "fonts"])
else:
copy_dependencies("fonts")
# fix the styling of the 404
if spec == "html":
link = tree.xpath("//link[starts-with(@href, 'http://www.w3.org/StyleSheets/TR/')]")[0].get("href")
path = os.path.join(spec_dir, "404.html")
with open(path) as data:
html404 = data.read()
html404 = html404.replace("http://www.w3.org/StyleSheets/TR/W3C-ED", link)
with open(path, "w") as data:
data.write(html404)