def munge(document, template, linkrel, d, fullpath, ext, url, config):
fixRelativeLinks(template, linkrel)
addMtime(template, fullpath)
removeH1(document)
fixAPI(document, url)
fontifyPython(document)
addPyListings(document, d)
addHTMLListings(document, d)
fixLinks(document, ext)
putInToC(template, generateToC(document))
footnotes(document)
notes(document)
# Insert the document into the template
title = domhelpers.findNodesNamed(document, 'title')[0].childNodes
setTitle(template, title)
authors = domhelpers.findNodesNamed(document, 'link')
authors = [(n.getAttribute('title',''), n.getAttribute('href', ''))
for n in authors if n.getAttribute('rel', '') == 'author']
setAuthors(template, authors)
body = domhelpers.findNodesNamed(document, "body")[0]
tmplbody = domhelpers.findElementsWithAttribute(template, "class",
"body")[0]
tmplbody.childNodes = body.childNodes
tmplbody.setAttribute("class", "content")
def check_80_columns(self, dom, filename):
for node in domhelpers.findNodesNamed(dom, 'pre'):
# the ps/pdf output is in a font that cuts off at 80 characters,
# so this is enforced to make sure the interesting parts (which
# are likely to be on the right-hand edge) stay on the printed
# page.
for line in domhelpers.gatherTextNodes(node, 1).split('\n'):
if len(line.rstrip()) > 80:
self._reportError(filename, node,
'text wider than 80 columns in pre')
for node in domhelpers.findNodesNamed(dom, 'a'):
if node.getAttribute('class', '').endswith('listing'):
try:
fn = os.path.dirname(filename)
fn = os.path.join(fn, node.getAttribute('href'))
lines = open(fn,'r').readlines()
except:
self._reportError(filename, node,
'bad listing href: %r' %
node.getAttribute('href'))
continue
for line in lines:
if len(line.rstrip()) > 80:
self._reportError(filename, node,
'listing wider than 80 columns')
def check_lists(self, dom, filename):
for node in domhelpers.findNodesNamed(dom, "ul") + domhelpers.findNodesNamed(dom, "ol"):
if not node.childNodes:
self._reportError(filename, node, "empty list")
for child in node.childNodes:
if child.nodeName != "li":
self._reportError(filename, node, "only list items allowed in lists")
def templateMutate(self, document, parentCount=0):
if self.metaTemplate:
newDoc = self.metaTemplate.cloneNode(1)
if parentCount:
dotdot = parentCount * '../'
for ddname in 'href', 'src', 'action':
for node in domhelpers.findElementsWithAttribute(
newDoc, ddname):
node.setAttribute(ddname,
dotdot + node.getAttribute(ddname))
ttl = domhelpers.findNodesNamed(newDoc, "title")[0]
ttl2 = domhelpers.findNodesNamed(document, "title")[0]
ttl.childNodes[:] = []
for n in ttl2.childNodes:
ttl.appendChild(n)
body = domhelpers.findElementsWithAttribute(
newDoc, "class", "__BODY__")[0]
body2 = domhelpers.findNodesNamed(document, "body")[0]
ndx = body.parentNode.childNodes.index(body)
body.parentNode.childNodes[ndx:ndx + 1] = body2.childNodes
for n in body2.childNodes:
n.parentNode = body.parentNode
f = open("garbage.html", "wb")
f.write(newDoc.toprettyxml())
return newDoc
return document
def check_80_columns(self, dom, filename):
for node in domhelpers.findNodesNamed(dom, 'pre'):
# the ps/pdf output is in a font that cuts off at 80 characters,
# so this is enforced to make sure the interesting parts (which
# are likely to be on the right-hand edge) stay on the printed
# page.
for line in domhelpers.gatherTextNodes(node, 1).split('\n'):
if len(line.rstrip()) > 80:
self._reportError(filename, node,
'text wider than 80 columns in pre')
for node in domhelpers.findNodesNamed(dom, 'a'):
if node.getAttribute('class').endswith('listing'):
try:
fn = os.path.dirname(filename)
fn = os.path.join(fn, node.getAttribute('href'))
lines = open(fn, 'r').readlines()
except:
self._reportError(
filename, node,
'bad listing href: %r' % node.getAttribute('href'))
continue
for line in lines:
if len(line.rstrip()) > 80:
self._reportError(filename, node,
'listing wider than 80 columns')
def makeBook(dom, d):
body = microdom.Element('body')
body.appendChild(domhelpers.findNodesNamed(dom, 'h1')[0])
toc = domhelpers.findElementsWithAttribute(dom, 'class', 'toc')[0]
toc = domhelpers.findNodesNamed(toc, 'li')
for node in toc:
if (node.hasAttribute('class') and
node.getAttribute('class')=='tocignore'):
continue
parents = domhelpers.getParents(node)
nodeLevel = len([1 for parent in parents if hasattr(parent, 'tagName')
and parent.tagName in ('ol', 'ul')])
data = node.childNodes[0].data != ''
if not data:
node = node.childNodes[1]
newNode = lowerDocument(node.getAttribute('href'), d, nodeLevel)
for child in newNode.childNodes:
body.appendChild(child)
else:
text = microdom.Text(node.childNodes[0].data)
newNode = microdom.Element('h'+str(nodeLevel))
newNode.appendChild(text)
body.appendChild(newNode)
origBody = domhelpers.findNodesNamed(dom, 'body')[0]
origBody.parentNode.replaceChild(body, origBody)
def check_lists(self, dom, filename):
for node in (domhelpers.findNodesNamed(dom, 'ul') +
domhelpers.findNodesNamed(dom, 'ol')):
if not node.childNodes:
self._reportError(filename, node, 'empty list')
for child in node.childNodes:
if child.nodeName != 'li':
self._reportError(filename, node,
'only list items allowed in lists')
def check_lists(self, dom, filename):
for node in (domhelpers.findNodesNamed(dom, 'ul')+
domhelpers.findNodesNamed(dom, 'ol')):
if not node.childNodes:
self._reportError(filename, node, 'empty list')
for child in node.childNodes:
if child.nodeName != 'li':
self._reportError(filename, node,
'only list items allowed in lists')
def getTitleLink(url):
d = client.getPage("http://moshez.org/discuss/rss")
d.addCallback(microdom.parseString)
d.addCallback(lambda d: domhelpers.findNodesNamed(d, 'item')[0])
d.addCallback(lambda d: (
domhelpers.getNodeText(domhelpers.findNodesNamed(d, 'link')[0]),
domhelpers.getNodeText(domhelpers.findNodesNamed(d, 'title')[0]),
))
return d
def check_title(self, dom, filename):
doc = dom.documentElement
title = domhelpers.findNodesNamed(dom, 'title')
if len(title)!=1:
return self._reportError(filename, doc, 'not exactly one title')
h1 = domhelpers.findNodesNamed(dom, 'h1')
if len(h1)!=1:
return self._reportError(filename, doc, 'not exactly one h1')
if domhelpers.getNodeText(h1[0]) != domhelpers.getNodeText(title[0]):
self._reportError(filename, h1[0], 'title and h1 text differ')
def check_title(self, dom, filename):
doc = dom.documentElement
title = domhelpers.findNodesNamed(dom, 'title')
if len(title) != 1:
return self._reportError(filename, doc, 'not exactly one title')
h1 = domhelpers.findNodesNamed(dom, 'h1')
if len(h1) != 1:
return self._reportError(filename, doc, 'not exactly one h1')
if domhelpers.getNodeText(h1[0]) != domhelpers.getNodeText(title[0]):
self._reportError(filename, h1[0], 'title and h1 text differ')
def extractRSSFeeds(dom):
for bookmark in domhelpers.findNodesNamed(dom, "bookmark"):
titleNodes = domhelpers.findNodesNamed(bookmark, "title")
if titleNodes:
title = domhelpers.getNodeText(titleNodes[0])
else:
title = None
url = bookmark.getAttribute("href")
url = urllib.unquote(url)
yield title, url
def findRSSFolder(dom, folderName=None):
if folderName is None:
folderName = "RSS"
if folderName == "":
return dom
for folder in domhelpers.findNodesNamed(dom, "folder"):
for title in domhelpers.findNodesNamed(folder, "title"):
text = domhelpers.getNodeText(title)
if text == folderName:
return folder
return None
def numberDocument(document, chapterNumber):
"""
Number the sections of the given document.
A dot-separated chapter, section number is added to the beginning of each
section, as defined by C{h2} nodes.
This is probably intended to interact in a rather specific way with
L{getSectionNumber}.
@type document: A DOM Node or Document
@param document: The input document which contains all of the content to be
presented.
@type chapterNumber: C{int}
@param chapterNumber: The chapter number of this content in an overall
document.
@return: C{None}
"""
i = 1
for node in domhelpers.findNodesNamed(document, "h2"):
label = dom.Text()
label.data = "%s.%d " % (chapterNumber, i)
node.insertBefore(label, node.firstChild)
i += 1
def getDescription(self):
# http://purl.org/dc/elements/1.1/ description
l = domhelpers.findNodesNamed(self.dom, 'description')
if l:
return domhelpers.getNodeText(l[0])
else:
return None
def setTitle(template, title, chapterNumber):
"""
Add title and chapter number information to the template document.
The title is added to the end of the first C{title} tag and the end of the
first tag with a C{class} attribute set to C{title}. If specified, the
chapter is inserted before the title.
@type template: A DOM Node or Document
@param template: The output template which defines the presentation of the
version information.
@type title: C{list} of DOM Nodes
@param title: Nodes from the input document defining its title.
@type chapterNumber: C{int}
@param chapterNumber: The chapter number of this content in an overall
document. If not applicable, any C{False} value will result in this
information being omitted.
@return: C{None}
"""
for nodeList in (domhelpers.findNodesNamed(template, "title"),
domhelpers.findElementsWithAttribute(
template, "class", 'title')):
if nodeList:
if numberer.getNumberSections() and chapterNumber:
nodeList[0].childNodes.append(
microdom.Text('%s. ' % chapterNumber))
nodeList[0].childNodes.extend(title)
def footnotes(document):
"""
Find footnotes in the given document, move them to the end of the body, and
generate links to them.
A footnote is any node with a C{class} attribute set to C{footnote}.
Footnote links are generated as superscript. Footnotes are collected in a
C{ol} node at the end of the document.
@type document: A DOM Node or Document
@param document: The input document which contains all of the content to be
presented.
@return: C{None}
"""
footnotes = domhelpers.findElementsWithAttribute(document, "class", "footnote")
if not footnotes:
return
footnoteElement = microdom.Element("ol")
id = 1
for footnote in footnotes:
href = microdom.parseString('<a href="#footnote-%(id)d">' "<super>%(id)d</super></a>" % vars()).documentElement
text = " ".join(domhelpers.getNodeText(footnote).split())
href.setAttribute("title", text)
target = microdom.Element("a", attributes={"name": "footnote-%d" % id})
target.childNodes = [footnote]
footnoteContent = microdom.Element("li")
footnoteContent.childNodes = [target]
footnoteElement.childNodes.append(footnoteContent)
footnote.parentNode.replaceChild(href, footnote)
id += 1
body = domhelpers.findNodesNamed(document, "body")[0]
header = microdom.parseString("<h2>Footnotes</h2>").documentElement
body.childNodes.append(header)
body.childNodes.append(footnoteElement)
def testAwfulTagSoup(self):
s = """
<html>
<head><title> I send you this message to have your advice!!!!</titl e
</headd>
<body bgcolor alink hlink vlink>
<h1><BLINK>SALE</blINK> TWENTY MILLION EMAILS & FUR COAT NOW
FREE WITH `ENLARGER'</h1>
YES THIS WONDERFUL AWFER IS NOW HERER!!!
<script LANGUAGE="javascript">
function give_answers() {
if (score < 70) {
alert("I hate you");
}}
</script><a href=/foo.com/lalal name=foo>lalal</a>
</body>
</HTML>
"""
d = microdom.parseString(s, beExtremelyLenient=1)
l = domhelpers.findNodesNamed(d.documentElement, 'blink')
self.assertEquals(len(l), 1)
def test_awfulTagSoup(self):
s = """
<html>
<head><title> I send you this message to have your advice!!!!</titl e
</headd>
<body bgcolor alink hlink vlink>
<h1><BLINK>SALE</blINK> TWENTY MILLION EMAILS & FUR COAT NOW
FREE WITH `ENLARGER'</h1>
YES THIS WONDERFUL AWFER IS NOW HERER!!!
<script LANGUAGE="javascript">
function give_answers() {
if (score < 70) {
alert("I hate you");
}}
</script><a href=/foo.com/lalal name=foo>lalal</a>
</body>
</HTML>
"""
d = microdom.parseString(s, beExtremelyLenient=1)
l = domhelpers.findNodesNamed(d.documentElement, "blink")
self.assertEqual(len(l), 1)
def numberDocument(document, chapterNumber):
"""
Number the sections of the given document.
A dot-separated chapter, section number is added to the beginning of each
section, as defined by C{h2} nodes.
This is probably intended to interact in a rather specific way with
L{getSectionNumber}.
@type document: A DOM Node or Document
@param document: The input document which contains all of the content to be
presented.
@type chapterNumber: C{int}
@param chapterNumber: The chapter number of this content in an overall
document.
@return: C{None}
"""
i = 1
for node in domhelpers.findNodesNamed(document, "h2"):
label = dom.Text()
label.data = "%s.%d " % (chapterNumber, i)
node.insertBefore(label, node.firstChild)
i += 1
def check_pre_py_listing(self, dom, filename):
for node in domhelpers.findNodesNamed(dom, 'pre'):
if node.getAttribute('class') == 'python':
try:
text = domhelpers.getNodeText(node)
# Fix < and >
text = text.replace('>', '>').replace('<', '<')
# Strip blank lines
lines = filter(None,
[l.rstrip() for l in text.split('\n')])
# Strip leading space
while not [
1 for line in lines if line[:1] not in ('', ' ')
]:
lines = [line[1:] for line in lines]
text = '\n'.join(lines) + '\n'
try:
parser.suite(text)
except SyntaxError:
# Pretend the "..." idiom is syntactically valid
text = text.replace("...", "'...'")
parser.suite(text)
except SyntaxError as e:
self._reportError(filename, node,
'invalid python code:' + str(e))
def setAuthors(template, authors):
# First, similarly to setTitle, insert text into an <div class="authors">
text = ''
for name, href in authors:
# FIXME: Do proper quoting/escaping (is it ok to use
# xml.sax.saxutils.{escape,quoteattr}?)
anchor = '<a href="%s">%s</a>' % (href, name)
if (name, href) == authors[-1]:
if len(authors) == 1:
text = anchor
else:
text += 'and ' + anchor
else:
text += anchor + ','
childNodes = microdom.parseString('<span>' + text +'</span>').childNodes
for node in domhelpers.findElementsWithAttribute(template,
"class", 'authors'):
node.childNodes.extend(childNodes)
# Second, add appropriate <link rel="author" ...> tags to the <head>.
head = domhelpers.findNodesNamed(template, 'head')[0]
authors = [microdom.parseString('<link rel="author" href="%s" title="%s"/>'
% (href, name)).childNodes[0]
for name, href in authors]
head.childNodes.extend(authors)
def setTitle(template, title, chapterNumber):
"""
Add title and chapter number information to the template document.
The title is added to the end of the first C{title} tag and the end of the
first tag with a C{class} attribute set to C{title}. If specified, the
chapter is inserted before the title.
@type template: A DOM Node or Document
@param template: The output template which defines the presentation of the
version information.
@type title: C{list} of DOM Nodes
@param title: Nodes from the input document defining its title.
@type chapterNumber: C{int}
@param chapterNumber: The chapter number of this content in an overall
document. If not applicable, any C{False} value will result in this
information being omitted.
@return: C{None}
"""
for nodeList in (
domhelpers.findNodesNamed(template, "title"),
domhelpers.findElementsWithAttribute(template, "class", "title"),
):
if nodeList:
if numberer.getNumberSections() and chapterNumber:
nodeList[0].childNodes.append(microdom.Text("%s. " % chapterNumber))
nodeList[0].childNodes.extend(title)
def getFirstAncestorWithSectionHeader(entry):
"""Go up ancestors until one with at least one <h2> is found, then return the <h2> nodes"""
for a in domhelpers.getParents(entry)[1:]:
headers = domhelpers.findNodesNamed(a, "h2")
if len(headers) > 0:
return headers
return []
def setTitle(template, title, chapterNumber):
for nodeList in (domhelpers.findNodesNamed(template, "title"),
domhelpers.findElementsWithAttribute(template, "class",
'title')):
if nodeList:
if numberer.getNumberSections() and chapterNumber:
nodeList[0].childNodes.append(microdom.Text('%s. ' % chapterNumber))
nodeList[0].childNodes.extend(title)
def render_GET(self, request):
url = urlTemplate % request.args['user'][0]
client.getPage(url).addCallback(microdom.parseString).addCallback(
lambda t: domhelpers.findNodesNamed(t, 'item')
).addCallback(lambda itms: zip(
[domhelpers.findNodesNamed(x, 'title')[0] for x in itms
], [domhelpers.findNodesNamed(x, 'link')[0] for x in itms]
)).addCallback(
lambda itms: '<html><head></head><body><ul>%s</ul></body></html>' %
'\n'.join([
'<li><a href="%s">%s</a></li>' %
(domhelpers.getNodeText(link), domhelpers.getNodeText(title))
for (title, link) in itms
])).addCallback(lambda s: (request.write(s), request.finish(
))).addErrback(lambda e:
(request.write('Error: %s' % e), request.finish()))
return server.NOT_DONE_YET
def lowerDocument(href, d, nodeLevel):
newNode = microdom.parse(open(os.path.join(d, href)))
newNode = domhelpers.findNodesNamed(newNode, 'body')[0]
headers = domhelpers.findElements(newNode,
lambda x: len(x.tagName)==2 and x.tagName[0]=='h' and
x.tagName[1] in '123456')
for header in headers:
header.tagName = 'h'+str(int(header.tagName[1])+nodeLevel)
return newNode
def render_GET(self, request):
url = urlTemplate % request.args['user'][0]
client.getPage(url).addCallback(
microdom.parseString).addCallback(
lambda t: domhelpers.findNodesNamed(t, 'item')).addCallback(
lambda itms: zip([domhelpers.findNodesNamed(x, 'title')[0]
for x in itms],
[domhelpers.findNodesNamed(x, 'link')[0]
for x in itms]
)).addCallback(
lambda itms: '<html><head></head><body><ul>%s</ul></body></html>' %
'\n'.join(
['<li><a href="%s">%s</a></li>' % (
domhelpers.getNodeText(link), domhelpers.getNodeText(title))
for (title, link) in itms])
).addCallback(lambda s: (request.write(s),request.finish())).addErrback(
lambda e: (request.write('Error: %s' % e),request.finish()))
return server.NOT_DONE_YET
def check_texturl_matches_href(self, dom, filename):
for node in domhelpers.findNodesNamed(dom, "a"):
if not node.hasAttribute("href"):
continue
text = domhelpers.getNodeText(node)
proto = urlparse.urlparse(text)[0]
if proto and " " not in text:
if text != node.getAttribute("href"):
self._reportError(filename, node, "link text does not match href")
def testDeferredModel(self):
channel = FakeHTTPChannel()
channel.site = FakeSite(_TestPage())
request = channel.makeFakeRequest('/')
while not request.finished:
reactor.iterate()
dom = microdom.parseXMLString(request.written.getvalue())
spanElems = domhelpers.findNodesNamed(dom, 'span')
for spanElem in spanElems:
self.failUnlessEqual('The Result', spanElem.childNodes[0].data)
def check_texturl_matches_href(self, dom, filename):
for node in domhelpers.findNodesNamed(dom, 'a'):
if not node.hasAttribute('href'):
continue
text = domhelpers.getNodeText(node)
proto = urlparse.urlparse(text)[0]
if proto and ' ' not in text:
if text != node.getAttribute('href'):
self._reportError(filename, node,
'link text does not match href')
def setAuthors(template, authors):
"""
Add author information to the template document.
Names and contact information for authors are added to each node with a
C{class} attribute set to C{authors} and to the template head as C{link}
nodes.
@type template: A DOM Node or Document
@param template: The output template which defines the presentation of the
version information.
@type authors: C{list} of two-tuples of C{str}
@param authors: List of names and contact information for the authors of
the input document.
@return: C{None}
"""
for node in domhelpers.findElementsWithAttribute(template, "class",
'authors'):
# First, similarly to setTitle, insert text into an <div
# class="authors">
container = dom.Element('span')
for name, href in authors:
anchor = dom.Element('a')
anchor.setAttribute('href', href)
anchorText = dom.Text()
anchorText.data = name
anchor.appendChild(anchorText)
if (name, href) == authors[-1]:
if len(authors) == 1:
container.appendChild(anchor)
else:
andText = dom.Text()
andText.data = 'and '
container.appendChild(andText)
container.appendChild(anchor)
else:
container.appendChild(anchor)
commaText = dom.Text()
commaText.data = ', '
container.appendChild(commaText)
node.appendChild(container)
# Second, add appropriate <link rel="author" ...> tags to the <head>.
head = domhelpers.findNodesNamed(template, 'head')[0]
authors = [
dom.parseString('<link rel="author" href="%s" title="%s"/>' %
(href, name)).childNodes[0] for name, href in authors
]
head.childNodes.extend(authors)
def setAuthors(template, authors):
"""
Add author information to the template document.
Names and contact information for authors are added to each node with a
C{class} attribute set to C{authors} and to the template head as C{link}
nodes.
@type template: A DOM Node or Document
@param template: The output template which defines the presentation of the
version information.
@type authors: C{list} of two-tuples of C{str}
@param authors: List of names and contact information for the authors of
the input document.
@return: C{None}
"""
for node in domhelpers.findElementsWithAttribute(template,
"class", 'authors'):
# First, similarly to setTitle, insert text into an <div
# class="authors">
container = dom.Element('span')
for name, href in authors:
anchor = dom.Element('a')
anchor.setAttribute('href', href)
anchorText = dom.Text()
anchorText.data = name
anchor.appendChild(anchorText)
if (name, href) == authors[-1]:
if len(authors) == 1:
container.appendChild(anchor)
else:
andText = dom.Text()
andText.data = 'and '
container.appendChild(andText)
container.appendChild(anchor)
else:
container.appendChild(anchor)
commaText = dom.Text()
commaText.data = ', '
container.appendChild(commaText)
node.appendChild(container)
# Second, add appropriate <link rel="author" ...> tags to the <head>.
head = domhelpers.findNodesNamed(template, 'head')[0]
authors = [dom.parseString('<link rel="author" href="%s" title="%s"/>'
% (href, name)).childNodes[0]
for name, href in authors]
head.childNodes.extend(authors)
def check_a_py_listing(self, dom, filename):
for node in domhelpers.findNodesNamed(dom, 'a'):
if node.getAttribute('class') == 'py-listing':
fn = os.path.join(os.path.dirname(filename),
node.getAttribute('href'))
lines = open(fn).readlines()
lines = lines[int(node.getAttribute('skipLines', 0)):]
for line, num in zip(lines, range(len(lines))):
if line.count('59 Temple Place, Suite 330, Boston'):
self._reportError(filename, node,
'included source file %s has licence boilerplate.'
' Use skipLines="%d".'
% (fn, int(node.getAttribute('skipLines',0))+num+1))
def testDeferredModel(self):
# Test that multiple uses of a deferred model work correctly.
channel = FakeHTTPChannel()
channel.site = FakeSite(_TestPage())
request = channel.makeFakeRequest('/')
while not request.finished:
reactor.iterate()
dom = microdom.parseXMLString(request.written.getvalue())
spanElems = domhelpers.findNodesNamed(dom, 'span')
for spanElem in spanElems:
self.failUnlessEqual('The Result', spanElem.childNodes[0].data)
def render_GET(self, request):
"""
Get an xml feed from LiveJournal and construct a new HTML page using the
'title' and 'link' parsed from the xml document.
"""
url = urlTemplate % request.args['user'][0]
client.getPage(url, timeout=30).addCallback(
microdom.parseString).addCallback(
lambda t: domhelpers.findNodesNamed(t, 'item')).addCallback(
lambda itms: zip([domhelpers.findNodesNamed(x, 'title')[0]
for x in itms],
[domhelpers.findNodesNamed(x, 'link')[0]
for x in itms]
)).addCallback(
lambda itms: '<html><head></head><body><ul>%s</ul></body></html>' %
'\n'.join(
['<li><a href="%s">%s</a></li>' % (
domhelpers.getNodeText(link), domhelpers.getNodeText(title))
for (title, link) in itms])
).addCallback(lambda s: (request.write(s),request.finish())).addErrback(
lambda e: (request.write('Error: %s' % e),request.finish()))
return server.NOT_DONE_YET
def check_80_columns(self, dom, filename):
for node in domhelpers.findNodesNamed(dom, 'pre'):
for line in domhelpers.getNodeText(node).split('\n'):
if len(line.rstrip()) > 80:
self._reportError(filename, node,
'text wider than 80 columns in pre')
for node in domhelpers.findNodesNamed(dom, 'a'):
if node.getAttribute('class', '').endswith('listing'):
try:
fn = os.path.dirname(filename)
fn = os.path.join(fn, node.getAttribute('href'))
lines = open(fn,'r').readlines()
except:
self._reportError(filename, node,
'bad listing href: %r' %
node.getAttribute('href'))
continue
for line in lines:
if len(line.rstrip()) > 80:
self._reportError(filename, node,
'listing wider than 80 columns')
def templateMutate(self, document, parentCount=0):
if self.metaTemplate:
newDoc = self.metaTemplate.cloneNode(1)
if parentCount:
dotdot = parentCount * '../'
for ddname in 'href', 'src', 'action':
for node in domhelpers.findElementsWithAttribute(newDoc, ddname):
node.setAttribute(ddname, dotdot + node.getAttribute(ddname))
ttl = domhelpers.findNodesNamed(newDoc, "title")[0]
ttl2 = domhelpers.findNodesNamed(document, "title")[0]
ttl.childNodes[:] = []
for n in ttl2.childNodes:
ttl.appendChild(n)
body = domhelpers.findElementsWithAttribute(newDoc, "class", "__BODY__")[0]
body2 = domhelpers.findNodesNamed(document, "body")[0]
ndx = body.parentNode.childNodes.index(body)
body.parentNode.childNodes[ndx:ndx+1] = body2.childNodes
for n in body2.childNodes:
n.parentNode = body.parentNode
f = open("garbage.html", "wb")
f.write(newDoc.toprettyxml())
return newDoc
return document
def munge(document, template, linkrel, dir, fullpath, ext, url, config, outfileGenerator=getOutputFileName):
fixRelativeLinks(template, linkrel)
addMtime(template, fullpath)
removeH1(document)
if not config.get('noapi', False):
fixAPI(document, url)
fontifyPython(document)
fixLinks(document, ext)
addPyListings(document, dir)
addHTMLListings(document, dir)
addPlainListings(document, dir)
putInToC(template, generateToC(document))
footnotes(document)
notes(document)
setIndexLink(template, indexer.getIndexFilename())
setVersion(template, config.get('version', ''))
# Insert the document into the template
chapterNumber = htmlbook.getNumber(fullpath)
title = domhelpers.findNodesNamed(document, 'title')[0].childNodes
setTitle(template, title, chapterNumber)
if numberer.getNumberSections() and chapterNumber:
numberDocument(document, chapterNumber)
index(document, outfileGenerator(os.path.split(fullpath)[1], ext),
htmlbook.getReference(fullpath))
authors = domhelpers.findNodesNamed(document, 'link')
authors = [(node.getAttribute('title',''), node.getAttribute('href', ''))
for node in authors if node.getAttribute('rel', '') == 'author']
setAuthors(template, authors)
body = domhelpers.findNodesNamed(document, "body")[0]
tmplbody = domhelpers.findElementsWithAttribute(template, "class",
"body")[0]
tmplbody.childNodes = body.childNodes
tmplbody.setAttribute("class", "content")
def check_a_py_listing(self, dom, filename):
for node in domhelpers.findNodesNamed(dom, 'a'):
if node.getAttribute('class') == 'py-listing':
fn = os.path.join(os.path.dirname(filename),
node.getAttribute('href'))
lines = open(fn).readlines()
lines = lines[int(node.getAttribute('skipLines', 0)):]
for line, num in zip(lines, range(len(lines))):
if line.count('59 Temple Place, Suite 330, Boston'):
self._reportError(
filename, node,
'included source file %s has licence boilerplate.'
' Use skipLines="%d".' %
(fn,
int(node.getAttribute('skipLines', 0)) + num + 1))
def splitIntoSlides(document):
body = domhelpers.findNodesNamed(document, 'body')[0]
slides = []
slide = []
title = '(unset)'
for child in body.childNodes:
if isinstance(child, dom.Element) and child.tagName == 'h2':
if slide:
slides.append((title, slide))
slide = []
title = domhelpers.getNodeText(child)
else:
slide.append(child)
slides.append((title, slide))
return slides
def splitIntoSlides(document):
body = domhelpers.findNodesNamed(document, 'body')[0]
slides = []
slide = []
title = '(unset)'
for child in body.childNodes:
if isinstance(child, microdom.Element) and child.tagName == 'h2':
if slide:
slides.append((title, slide))
slide = []
title = domhelpers.getNodeText(child)
else:
slide.append(child)
slides.append((title, slide))
return slides
def testCaseSensitiveSoonCloser(self):
s = """
<HTML><BODY>
<P ALIGN="CENTER">
<A HREF="http://www.apache.org/"><IMG SRC="/icons/apache_pb.gif"></A>
</P>
<P>
This is an insane set of text nodes that should NOT be gathered under
the A tag above.
</P>
</BODY></HTML>
"""
d = microdom.parseString(s, beExtremelyLenient=1)
l = domhelpers.findNodesNamed(d.documentElement, 'a')
n = domhelpers.gatherTextNodes(l[0],1).replace(' ',' ')
self.assertEquals(n.find('insane'), -1)
def getFirstAncestorWithSectionHeader(entry):
"""
Visit the ancestors of C{entry} until one with at least one C{h2} child
node is found, then return all of that node's C{h2} child nodes.
@type entry: A DOM Node
@param entry: The node from which to begin traversal. This node itself is
excluded from consideration.
@rtype: C{list} of DOM Nodes
@return: All C{h2} nodes of the ultimately selected parent node.
"""
for a in domhelpers.getParents(entry)[1:]:
headers = domhelpers.findNodesNamed(a, "h2")
if len(headers) > 0:
return headers
return []
def getFirstAncestorWithSectionHeader(entry):
"""
Visit the ancestors of C{entry} until one with at least one C{h2} child
node is found, then return all of that node's C{h2} child nodes.
@type entry: A DOM Node
@param entry: The node from which to begin traversal. This node itself is
excluded from consideration.
@rtype: C{list} of DOM Nodes
@return: All C{h2} nodes of the ultimately selected parent node.
"""
for a in domhelpers.getParents(entry)[1:]:
headers = domhelpers.findNodesNamed(a, "h2")
if len(headers) > 0:
return headers
return []
def removeH1(document):
"""
Replace all C{h1} nodes in the given document with empty C{span} nodes.
C{h1} nodes mark up document sections and the output template is given an
opportunity to present this information in a different way.
@type document: A DOM Node or Document
@param document: The input document which contains all of the content to be
presented.
@return: C{None}
"""
h1 = domhelpers.findNodesNamed(document, 'h1')
empty = dom.Element('span')
for node in h1:
node.parentNode.replaceChild(empty, node)
def setAuthors(template, authors):
"""
Add author information to the template document.
Names and contact information for authors are added to each node with a
C{class} attribute set to C{authors} and to the template head as C{link}
nodes.
@type template: A DOM Node or Document
@param template: The output template which defines the presentation of the
version information.
@type authors: C{list} of two-tuples of C{str}
@param authors: List of names and contact information for the authors of
the input document.
@return: C{None}
"""
# First, similarly to setTitle, insert text into an <div class="authors">
text = ''
for name, href in authors:
# FIXME: Do proper quoting/escaping (is it ok to use
# xml.sax.saxutils.{escape,quoteattr}?)
anchor = '<a href="%s">%s</a>' % (href, name)
if (name, href) == authors[-1]:
if len(authors) == 1:
text = anchor
else:
text += 'and ' + anchor
else:
text += anchor + ','
childNodes = microdom.parseString('<span>' + text + '</span>').childNodes
for node in domhelpers.findElementsWithAttribute(template, "class",
'authors'):
node.childNodes.extend(childNodes)
# Second, add appropriate <link rel="author" ...> tags to the <head>.
head = domhelpers.findNodesNamed(template, 'head')[0]
authors = [
microdom.parseString('<link rel="author" href="%s" title="%s"/>' %
(href, name)).childNodes[0]
for name, href in authors
]
head.childNodes.extend(authors)
def test_caseSensitiveSoonCloser(self):
s = """
<HTML><BODY>
<P ALIGN="CENTER">
<A HREF="http://www.apache.org/"><IMG SRC="/icons/apache_pb.gif"></A>
</P>
<P>
This is an insane set of text nodes that should NOT be gathered under
the A tag above.
</P>
</BODY></HTML>
"""
d = microdom.parseString(s, beExtremelyLenient=1)
l = domhelpers.findNodesNamed(d.documentElement, "a")
n = domhelpers.gatherTextNodes(l[0], 1).replace(" ", " ")
self.assertEqual(n.find("insane"), -1)
def numberDocument(document, chapterNumber):
"""
Number the sections of the given document.
A dot-separated chapter, section number is added to the beginning of each
section, as defined by C{h2} nodes.
@type document: A DOM Node or Document
@param document: The input document which contains all of the content to be
presented.
@type chapterNumber: C{int}
@param chapterNumber: The chapter number of this content in an overall
document.
@return: C{None}
"""
i = 1
for node in domhelpers.findNodesNamed(document, "h2"):
node.childNodes = [microdom.Text("%s.%d " % (chapterNumber, i))] + node.childNodes
i += 1
def footnotes(document):
"""
Find footnotes in the given document, move them to the end of the body, and
generate links to them.
A footnote is any node with a C{class} attribute set to C{footnote}.
Footnote links are generated as superscript. Footnotes are collected in a
C{ol} node at the end of the document.
@type document: A DOM Node or Document
@param document: The input document which contains all of the content to be
presented.
@return: C{None}
"""
footnotes = domhelpers.findElementsWithAttribute(document, "class",
"footnote")
if not footnotes:
return
footnoteElement = dom.Element('ol')
id = 1
for footnote in footnotes:
href = dom.parseString('<a href="#footnote-%(id)d">'
'<super>%(id)d</super></a>' %
vars()).documentElement
text = ' '.join(domhelpers.getNodeText(footnote).split())
href.setAttribute('title', text)
target = dom.Element('a')
target.setAttribute('name', 'footnote-%d' % (id, ))
target.childNodes = [footnote]
footnoteContent = dom.Element('li')
footnoteContent.childNodes = [target]
footnoteElement.childNodes.append(footnoteContent)
footnote.parentNode.replaceChild(href, footnote)
id += 1
body = domhelpers.findNodesNamed(document, "body")[0]
header = dom.parseString('<h2>Footnotes</h2>').documentElement
body.childNodes.append(header)
body.childNodes.append(footnoteElement)
def setTitle(template, title, chapterNumber):
"""
Add title and chapter number information to the template document.
The title is added to the end of the first C{title} tag and the end of the
first tag with a C{class} attribute set to C{title}. If specified, the
chapter is inserted before the title.
@type template: A DOM Node or Document
@param template: The output template which defines the presentation of the
version information.
@type title: C{list} of DOM Nodes
@param title: Nodes from the input document defining its title.
@type chapterNumber: C{int}
@param chapterNumber: The chapter number of this content in an overall
document. If not applicable, any C{False} value will result in this
information being omitted.
@return: C{None}
"""
if numberer.getNumberSections() and chapterNumber:
titleNode = dom.Text()
# This is necessary in order for cloning below to work. See Python
# isuse 4851.
titleNode.ownerDocument = template.ownerDocument
titleNode.data = '%s. ' % (chapterNumber, )
title.insert(0, titleNode)
for nodeList in (domhelpers.findNodesNamed(template, "title"),
domhelpers.findElementsWithAttribute(
template, "class", 'title')):
if nodeList:
for titleNode in title:
nodeList[0].appendChild(titleNode.cloneNode(True))
def test_findNodesNamed(self):
doc1 = microdom.parseString('<doc><foo/><bar/><foo>a</foo></doc>')
node_list = domhelpers.findNodesNamed(doc1, 'foo')
actual = len(node_list)
expected = 2
assert actual == expected, 'expected %d, got %d' % (expected, actual)
def munge(document,
template,
linkrel,
dir,
fullpath,
ext,
url,
config,
outfileGenerator=getOutputFileName):
"""
Mutate C{template} until it resembles C{document}.
@type document: A DOM Node or Document
@param document: The input document which contains all of the content to be
presented.
@type template: A DOM Node or Document
@param template: The template document which defines the desired
presentation format of the content.
@type linkrel: C{str}
@param linkrel: An prefix to apply to all relative links in C{src} or
C{href} attributes in the input document when generating the output
document.
@type dir: C{str}
@param dir: The directory in which to search for source listing files.
@type fullpath: C{str}
@param fullpath: The file name which contained the input document.
@type ext: C{str}
@param ext: The extension to use when selecting an output file name. This
replaces the extension of the input file name.
@type url: C{str}
@param url: A string which will be interpolated with the fully qualified
Python name of any API reference encountered in the input document, the
result of which will be used as a link to API documentation for that name
in the output document.
@type config: C{dict}
@param config: Further specification of the desired form of the output.
Valid keys in this dictionary::
noapi: If present and set to a True value, links to API documentation
will not be generated.
version: A string which will be included in the output to indicate the
version of this documentation.
@type outfileGenerator: Callable of C{str}, C{str} returning C{str}
@param outfileGenerator: Output filename factory. This is invoked with the
intput filename and C{ext} and the output document is serialized to the
file with the name returned.
@return: C{None}
"""
fixRelativeLinks(template, linkrel)
addMtime(template, fullpath)
removeH1(document)
if not config.get('noapi', False):
fixAPI(document, url)
fontifyPython(document)
fixLinks(document, ext)
addPyListings(document, dir)
addHTMLListings(document, dir)
addPlainListings(document, dir)
putInToC(template, generateToC(document))
footnotes(document)
notes(document)
setIndexLink(template, indexer.getIndexFilename())
setVersion(template, config.get('version', ''))
# Insert the document into the template
chapterNumber = htmlbook.getNumber(fullpath)
title = domhelpers.findNodesNamed(document, 'title')[0].childNodes
setTitle(template, title, chapterNumber)
if numberer.getNumberSections() and chapterNumber:
numberDocument(document, chapterNumber)
index(document, outfileGenerator(os.path.split(fullpath)[1], ext),
htmlbook.getReference(fullpath))
authors = domhelpers.findNodesNamed(document, 'link')
authors = [(node.getAttribute('title') or '', node.getAttribute('href')
or '') for node in authors
if node.getAttribute('rel') == 'author']
setAuthors(template, authors)
body = domhelpers.findNodesNamed(document, "body")[0]
tmplbody = domhelpers.findElementsWithAttribute(template, "class",
"body")[0]
tmplbody.childNodes = body.childNodes
tmplbody.setAttribute("class", "content")
import re
if __name__ == '__main__':
files = {
'lomVocab':
Path('exe') / 'webui' / 'schemas' / 'scorm2004' / 'common' /
'vocabValues.xsd',
'lomesVocab':
Path('exe') / 'webui' / 'schemas' / 'scorm2004' / 'vocab' /
'lomesvocab.xsd'
}
response = ''
vocab = {}
for varname, f in files.items():
document = parseString(f.bytes(), escapeAttributes=0)
nodes = findNodesNamed(document, 'xs:simpletype')
for node in nodes:
name = node.getAttribute('name', str())
enumerations = findNodesNamed(node, 'xs:enumeration')
vocab[name] = []
for enumeration in enumerations:
vocab[name].append([
enumeration.getAttribute('value'),
'_(%s)' % enumeration.getAttribute('value')
])
response += '%s = %s;\n\n' % (varname, json.dumps(
vocab, indent=4).encode('utf-8'))
outfile = Path('exe') / 'jsui' / 'scripts' / 'lomvocab.js'
response = re.sub('"_\(', '_("', response)
response = re.sub('\)"', '")', response)
outfile.write_bytes(response)
def test_findNodesNamed(self):
doc1 = self.dom.parseString('<doc><foo/><bar/><foo>a</foo></doc>')
node_list = domhelpers.findNodesNamed(doc1, 'foo')
actual = len(node_list)
self.assertEqual(actual, 2)
def check_style(self, dom, filename):
for node in domhelpers.findNodesNamed(dom, 'style'):
if domhelpers.getNodeText(node) != '':
self._reportError(filename, node, 'hand hacked style')
def check(_):
dom = microdom.parseXMLString(request.written.getvalue())
spanElems = domhelpers.findNodesNamed(dom, 'span')
for spanElem in spanElems:
self.failUnlessEqual('The Result', spanElem.childNodes[0].data)
def check_anchor_in_heading(self, dom, filename):
headingNames = ['h%d' % n for n in range(1, 7)]
for hname in headingNames:
for node in domhelpers.findNodesNamed(dom, hname):
if domhelpers.findNodesNamed(node, 'a'):
self._reportError(filename, node, 'anchor in heading')
