[Python-checkins] CVS: python/dist/src/Doc/tools/sgmlconv docfixer.py,1.25,1.26
Fred L. Drake
fdrake@users.sourceforge.net
Fri, 23 Mar 2001 09:01:50 -0800
Update of /cvsroot/python/python/dist/src/Doc/tools/sgmlconv
In directory usw-pr-cvs1:/tmp/cvs-serv30077
Modified Files:
docfixer.py
Log Message:
Lots of small changes to make this work with the Python DOM bindings
(minidom in particular); it was using PyDOM which is now obsolete.
Only write the output file on success -- this avoids updating the timestamp
on the file on failure, which confuses "make".
Index: docfixer.py
===================================================================
RCS file: /cvsroot/python/python/dist/src/Doc/tools/sgmlconv/docfixer.py,v
retrieving revision 1.25
retrieving revision 1.26
diff -C2 -r1.25 -r1.26
*** docfixer.py 2000/05/24 14:33:26 1.25
--- docfixer.py 2001/03/23 17:01:47 1.26
***************
*** 12,21 ****
import string
import sys
! import xml.dom.core
! from xml.dom.core import \
! ELEMENT, \
! ENTITY_REFERENCE, \
! TEXT
--- 12,21 ----
import string
import sys
! import xml.dom
! import xml.dom.minidom
! ELEMENT = xml.dom.Node.ELEMENT_NODE
! ENTITY_REFERENCE = xml.dom.Node.ENTITY_REFERENCE_NODE
! TEXT = xml.dom.Node.TEXT_NODE
***************
*** 50,79 ****
- # Workaround to deal with invalid documents (multiple root elements). This
- # does not indicate a bug in the DOM implementation.
- #
- def get_documentElement(doc):
- docelem = None
- for n in doc.childNodes:
- if n.nodeType == ELEMENT:
- docelem = n
- return docelem
-
- xml.dom.core.Document.get_documentElement = get_documentElement
-
-
- # Replace get_childNodes for the Document class; without this, children
- # accessed from the Document object via .childNodes (no matter how many
- # levels of access are used) will be given an ownerDocument of None.
- #
- def get_childNodes(doc):
- return xml.dom.core.NodeList(doc._node.children, doc._node)
-
- xml.dom.core.Document.get_childNodes = get_childNodes
-
-
def get_first_element(doc, gi):
for n in doc.childNodes:
! if n.get_nodeName() == gi:
return n
--- 50,56 ----
def get_first_element(doc, gi):
for n in doc.childNodes:
! if n.nodeName == gi:
return n
***************
*** 85,95 ****
def find_all_elements(doc, gi):
nodes = []
! if doc.get_nodeName() == gi:
nodes.append(doc)
for child in doc.childNodes:
if child.nodeType == ELEMENT:
! if child.get_tagName() == gi:
nodes.append(child)
for node in child.getElementsByTagName(gi):
--- 62,84 ----
+ def get_documentElement(node):
+ result = None
+ for child in node.childNodes:
+ if child.nodeType == ELEMENT:
+ result = child
+ return result
+
+
+ def set_tagName(elem, gi):
+ elem.nodeName = elem.tagName = gi
+
+
def find_all_elements(doc, gi):
nodes = []
! if doc.nodeName == gi:
nodes.append(doc)
for child in doc.childNodes:
if child.nodeType == ELEMENT:
! if child.tagName == gi:
nodes.append(child)
for node in child.getElementsByTagName(gi):
***************
*** 100,115 ****
nodes = []
for child in doc.childNodes:
! if child.get_nodeName() == gi:
nodes.append(child)
return nodes
def find_all_elements_from_set(doc, gi_set):
return __find_all_elements_from_set(doc, gi_set, [])
def __find_all_elements_from_set(doc, gi_set, nodes):
! if doc.get_nodeName() in gi_set:
nodes.append(doc)
for child in doc.childNodes:
! if child.get_nodeType() == ELEMENT:
__find_all_elements_from_set(child, gi_set, nodes)
return nodes
--- 89,105 ----
nodes = []
for child in doc.childNodes:
! if child.nodeName == gi:
nodes.append(child)
return nodes
+
def find_all_elements_from_set(doc, gi_set):
return __find_all_elements_from_set(doc, gi_set, [])
def __find_all_elements_from_set(doc, gi_set, nodes):
! if doc.nodeName in gi_set:
nodes.append(doc)
for child in doc.childNodes:
! if child.nodeType == ELEMENT:
__find_all_elements_from_set(child, gi_set, nodes)
return nodes
***************
*** 130,134 ****
node = get_first_element(fragment, "document")
if node is not None:
! node._node.name = documentclass
while 1:
node = extract_first_element(fragment, "input")
--- 120,124 ----
node = get_first_element(fragment, "document")
if node is not None:
! set_tagName(node, documentclass)
while 1:
node = extract_first_element(fragment, "input")
***************
*** 144,148 ****
docelem.insertBefore(node, text)
docelem.insertBefore(doc.createTextNode("\n"), docelem.firstChild)
! while fragment.firstChild and fragment.firstChild.get_nodeType() == TEXT:
fragment.removeChild(fragment.firstChild)
--- 134,138 ----
docelem.insertBefore(node, text)
docelem.insertBefore(doc.createTextNode("\n"), docelem.firstChild)
! while fragment.firstChild and fragment.firstChild.nodeType == TEXT:
fragment.removeChild(fragment.firstChild)
***************
*** 154,160 ****
prevskip = skip
skip = 0
! if n.get_nodeType() == TEXT and not prevskip:
discards.append(n)
! elif n.get_nodeName() == "COMMENT":
skip = 1
for node in discards:
--- 144,150 ----
prevskip = skip
skip = 0
! if n.nodeType == TEXT and not prevskip:
discards.append(n)
! elif n.nodeName == "COMMENT":
skip = 1
for node in discards:
***************
*** 178,183 ****
children = container.childNodes
for child in children:
! if child.get_nodeType() == ELEMENT:
! tagName = child.get_tagName()
if tagName in DESCRIPTOR_ELEMENTS:
rewrite_descriptor(doc, child)
--- 168,173 ----
children = container.childNodes
for child in children:
! if child.nodeType == ELEMENT:
! tagName = child.tagName
if tagName in DESCRIPTOR_ELEMENTS:
rewrite_descriptor(doc, child)
***************
*** 201,210 ****
#
# 1.
! descname = descriptor.get_tagName()
index = 1
if descname[-2:] == "ni":
descname = descname[:-2]
descriptor.setAttribute("index", "no")
! descriptor._node.name = descname
index = 0
desctype = descname[:-4] # remove 'desc'
--- 191,200 ----
#
# 1.
! descname = descriptor.tagName
index = 1
if descname[-2:] == "ni":
descname = descname[:-2]
descriptor.setAttribute("index", "no")
! set_tagName(descriptor, descname)
index = 0
desctype = descname[:-4] # remove 'desc'
***************
*** 220,224 ****
descriptor.removeAttribute("name")
# 2a.
! if descriptor.attributes.has_key("var"):
if descname != "opcodedesc":
raise RuntimeError, \
--- 210,214 ----
descriptor.removeAttribute("name")
# 2a.
! if descriptor.hasAttribute("var"):
if descname != "opcodedesc":
raise RuntimeError, \
***************
*** 246,253 ****
pos = skip_leading_nodes(children, pos)
while pos < len(children) \
! and children[pos].get_nodeName() in (linename, "versionadded"):
! if children[pos].get_tagName() == linename:
# this is really a supplemental signature, create <signature>
! sig = methodline_to_signature(doc, children[pos])
newchildren.append(sig)
else:
--- 236,248 ----
pos = skip_leading_nodes(children, pos)
while pos < len(children) \
! and children[pos].nodeName in (linename, "versionadded"):
! if children[pos].tagName == linename:
# this is really a supplemental signature, create <signature>
! oldchild = children[pos].cloneNode(1)
! try:
! sig = methodline_to_signature(doc, children[pos])
! except KeyError:
! print oldchild.toxml()
! raise
newchildren.append(sig)
else:
***************
*** 302,306 ****
# must be called after simplfy() if document is multi-rooted to begin with
docelem = get_documentElement(fragment)
! toplevel = docelem.get_tagName() == "manual" and "chapter" or "section"
appendices = 0
nodes = []
--- 297,301 ----
# must be called after simplfy() if document is multi-rooted to begin with
docelem = get_documentElement(fragment)
! toplevel = docelem.tagName == "manual" and "chapter" or "section"
appendices = 0
nodes = []
***************
*** 334,338 ****
continue
parent = label.parentNode
! parentTagName = parent.get_tagName()
if parentTagName == "title":
parent.parentNode.setAttribute("id", id)
--- 329,333 ----
continue
parent = label.parentNode
! parentTagName = parent.tagName
if parentTagName == "title":
parent.parentNode.setAttribute("id", id)
***************
*** 353,358 ****
node = queue[0]
del queue[0]
! if wsmap.has_key(node.get_nodeName()):
! ws = wsmap[node.get_tagName()]
children = node.childNodes
children.reverse()
--- 348,353 ----
node = queue[0]
del queue[0]
! if wsmap.has_key(node.nodeName):
! ws = wsmap[node.tagName]
children = node.childNodes
children.reverse()
***************
*** 362,367 ****
children.reverse()
# hack to get the title in place:
! if node.get_tagName() == "title" \
! and node.parentNode.firstChild.get_nodeType() == ELEMENT:
node.parentNode.insertBefore(doc.createText("\n "),
node.parentNode.firstChild)
--- 357,362 ----
children.reverse()
# hack to get the title in place:
! if node.tagName == "title" \
! and node.parentNode.firstChild.nodeType == ELEMENT:
node.parentNode.insertBefore(doc.createText("\n "),
node.parentNode.firstChild)
***************
*** 389,393 ****
node = queue[0]
del queue[0]
! if rewrite_element(node.get_tagName()):
children = node.childNodes
if len(children) == 1 \
--- 384,388 ----
node = queue[0]
del queue[0]
! if rewrite_element(node.tagName):
children = node.childNodes
if len(children) == 1 \
***************
*** 412,416 ****
return 0
if nodeType == ELEMENT:
! if l.get_tagName() != r.get_tagName():
return 0
# should check attributes, but that's not a problem here
--- 407,411 ----
return 0
if nodeType == ELEMENT:
! if l.tagName != r.tagName:
return 0
# should check attributes, but that's not a problem here
***************
*** 431,435 ****
if node is None:
return
! node._node.name = "synopsis"
lastchild = node.childNodes[-1]
if lastchild.nodeType == TEXT \
--- 426,430 ----
if node is None:
return
! set_tagName(node, "synopsis")
lastchild = node.childNodes[-1]
if lastchild.nodeType == TEXT \
***************
*** 438,447 ****
modauthor = extract_first_element(section, "moduleauthor")
if modauthor:
! modauthor._node.name = "author"
modauthor.appendChild(doc.createTextNode(
modauthor.getAttribute("name")))
modauthor.removeAttribute("name")
platform = extract_first_element(section, "platform")
! if section.get_tagName() == "section":
modinfo_pos = 2
modinfo = doc.createElement("moduleinfo")
--- 433,442 ----
modauthor = extract_first_element(section, "moduleauthor")
if modauthor:
! set_tagName(modauthor, "author")
modauthor.appendChild(doc.createTextNode(
modauthor.getAttribute("name")))
modauthor.removeAttribute("name")
platform = extract_first_element(section, "platform")
! if section.tagName == "section":
modinfo_pos = 2
modinfo = doc.createElement("moduleinfo")
***************
*** 468,472 ****
children = title.childNodes
if len(children) >= 2 \
! and children[0].get_nodeName() == "module" \
and children[0].childNodes[0].data == name:
# this is it; morph the <title> into <short-synopsis>
--- 463,467 ----
children = title.childNodes
if len(children) >= 2 \
! and children[0].nodeName == "module" \
and children[0].childNodes[0].data == name:
# this is it; morph the <title> into <short-synopsis>
***************
*** 474,478 ****
if first_data.data[:4] == " ---":
first_data.data = string.lstrip(first_data.data[4:])
! title._node.name = "short-synopsis"
if children[-1].nodeType == TEXT \
and children[-1].data[-1:] == ".":
--- 469,473 ----
if first_data.data[:4] == " ---":
first_data.data = string.lstrip(first_data.data[4:])
! set_tagName(title, "short-synopsis")
if children[-1].nodeType == TEXT \
and children[-1].data[-1:] == ".":
***************
*** 512,516 ****
for i in range(len(children)):
node = children[i]
! if node.get_nodeName() == "moduleinfo":
nextnode = children[i+1]
if nextnode.nodeType == TEXT:
--- 507,511 ----
for i in range(len(children)):
node = children[i]
! if node.nodeName == "moduleinfo":
nextnode = children[i+1]
if nextnode.nodeType == TEXT:
***************
*** 545,549 ****
for child in children:
if child.nodeType == ELEMENT:
! tagName = child.get_tagName()
if tagName == "hline" and prev_row is not None:
prev_row.setAttribute("rowsep", "1")
--- 540,544 ----
for child in children:
if child.nodeType == ELEMENT:
! tagName = child.tagName
if tagName == "hline" and prev_row is not None:
prev_row.setAttribute("rowsep", "1")
***************
*** 559,569 ****
if nodeType == TEXT:
if string.strip(child.data):
! raise ConversionError("unexpected free data in table")
table.removeChild(child)
continue
if nodeType == ELEMENT:
! if child.get_tagName() != "hline":
raise ConversionError(
! "unexpected <%s> in table" % child.get_tagName())
table.removeChild(child)
continue
--- 554,565 ----
if nodeType == TEXT:
if string.strip(child.data):
! raise ConversionError("unexpected free data in <%s>: %r"
! % (table.tagName, child.data))
table.removeChild(child)
continue
if nodeType == ELEMENT:
! if child.tagName != "hline":
raise ConversionError(
! "unexpected <%s> in table" % child.tagName)
table.removeChild(child)
continue
***************
*** 594,598 ****
nodes = []
for child in source.childNodes:
! if child.get_nodeName() == name:
nodes.append(child)
for node in nodes:
--- 590,594 ----
nodes = []
for child in source.childNodes:
! if child.nodeName == name:
nodes.append(child)
for node in nodes:
***************
*** 634,638 ****
def fixup_paras(doc, fragment):
for child in fragment.childNodes:
! if child.get_nodeName() in RECURSE_INTO_PARA_CONTAINERS:
fixup_paras_helper(doc, child)
descriptions = find_all_elements(fragment, "description")
--- 630,634 ----
def fixup_paras(doc, fragment):
for child in fragment.childNodes:
! if child.nodeName in RECURSE_INTO_PARA_CONTAINERS:
fixup_paras_helper(doc, child)
descriptions = find_all_elements(fragment, "description")
***************
*** 646,650 ****
start = skip_leading_nodes(children)
while len(children) > start:
! if children[start].get_nodeName() in RECURSE_INTO_PARA_CONTAINERS:
# Something to recurse into:
fixup_paras_helper(doc, children[start])
--- 642,646 ----
start = skip_leading_nodes(children)
while len(children) > start:
! if children[start].nodeName in RECURSE_INTO_PARA_CONTAINERS:
# Something to recurse into:
fixup_paras_helper(doc, children[start])
***************
*** 669,673 ****
nodeType = child.nodeType
if nodeType == ELEMENT:
! if child.get_tagName() in BREAK_ELEMENTS:
after = j
break
--- 665,669 ----
nodeType = child.nodeType
if nodeType == ELEMENT:
! if child.tagName in BREAK_ELEMENTS:
after = j
break
***************
*** 743,747 ****
# all whitespace, just skip
elif nodeType == ELEMENT:
! tagName = child.get_tagName()
if tagName in RECURSE_INTO_PARA_CONTAINERS:
return start
--- 739,743 ----
# all whitespace, just skip
elif nodeType == ELEMENT:
! tagName = child.tagName
if tagName in RECURSE_INTO_PARA_CONTAINERS:
return start
***************
*** 773,777 ****
def fixup_args(doc, arglist):
for child in arglist.childNodes:
! if child.get_nodeName() == "optional":
# found it; fix and return
arglist.insertBefore(doc.createTextNode("["), child)
--- 769,773 ----
def fixup_args(doc, arglist):
for child in arglist.childNodes:
! if child.nodeName == "optional":
# found it; fix and return
arglist.insertBefore(doc.createTextNode("["), child)
***************
*** 790,794 ****
section = sectauth.parentNode
section.removeChild(sectauth)
! sectauth._node.name = "author"
sectauth.appendChild(doc.createTextNode(
sectauth.getAttribute("name")))
--- 786,790 ----
section = sectauth.parentNode
section.removeChild(sectauth)
! set_tagName(sectauth, "author")
sectauth.appendChild(doc.createTextNode(
sectauth.getAttribute("name")))
***************
*** 796,800 ****
after = section.childNodes[2]
title = section.childNodes[1]
! if title.get_nodeName() != "title":
after = section.childNodes[0]
section.insertBefore(doc.createTextNode("\n "), after)
--- 792,796 ----
after = section.childNodes[2]
title = section.childNodes[1]
! if title.nodeName != "title":
after = section.childNodes[0]
section.insertBefore(doc.createTextNode("\n "), after)
***************
*** 807,815 ****
if child.nodeType == TEXT \
and string.lstrip(child.data)[:3] == ">>>":
! verbatim._node.name = "interactive-session"
def add_node_ids(fragment, counter=0):
! fragment._node.node_id = counter
for node in fragment.childNodes:
counter = counter + 1
--- 803,811 ----
if child.nodeType == TEXT \
and string.lstrip(child.data)[:3] == ">>>":
! set_tagName(verbatim, "interactive-session")
def add_node_ids(fragment, counter=0):
! fragment.node_id = counter
for node in fragment.childNodes:
counter = counter + 1
***************
*** 817,821 ****
counter = add_node_ids(node, counter)
else:
! node._node.node_id = counter
return counter + 1
--- 813,817 ----
counter = add_node_ids(node, counter)
else:
! node.node_id = counter
return counter + 1
***************
*** 832,836 ****
for node in nodes:
parent = node.parentNode
! d[parent._node.node_id] = parent
del nodes
map(fixup_refmodindexes_chunk, d.values())
--- 828,832 ----
for node in nodes:
parent = node.parentNode
! d[parent.node_id] = parent
del nodes
map(fixup_refmodindexes_chunk, d.values())
***************
*** 839,843 ****
def fixup_refmodindexes_chunk(container):
# node is probably a <para>; let's see how often it isn't:
! if container.get_tagName() != PARA_ELEMENT:
bwrite("--- fixup_refmodindexes_chunk(%s)\n" % container)
module_entries = find_all_elements(container, "module")
--- 835,839 ----
def fixup_refmodindexes_chunk(container):
# node is probably a <para>; let's see how often it isn't:
! if container.tagName != PARA_ELEMENT:
bwrite("--- fixup_refmodindexes_chunk(%s)\n" % container)
module_entries = find_all_elements(container, "module")
***************
*** 850,854 ****
if len(children) != 0:
bwrite("--- unexpected number of children for %s node:\n"
! % entry.get_tagName())
ewrite(entry.toxml() + "\n")
continue
--- 846,850 ----
if len(children) != 0:
bwrite("--- unexpected number of children for %s node:\n"
! % entry.tagName)
ewrite(entry.toxml() + "\n")
continue
***************
*** 874,878 ****
for node in nodes:
parent = node.parentNode
! d[parent._node.node_id] = parent
del nodes
map(fixup_bifuncindexes_chunk, d.values())
--- 870,874 ----
for node in nodes:
parent = node.parentNode
! d[parent.node_id] = parent
del nodes
map(fixup_bifuncindexes_chunk, d.values())
***************
*** 906,910 ****
parent = queue.pop()
i = 0
! children = parent.get_childNodes()
nchildren = len(children)
while i < (nchildren - 1):
--- 902,906 ----
parent = queue.pop()
i = 0
! children = parent.childNodes
nchildren = len(children)
while i < (nchildren - 1):
***************
*** 915,919 ****
child = children[i]
nextchild = children[i+1]
! nextchildren = nextchild.get_childNodes()
while len(nextchildren):
node = nextchildren[0]
--- 911,915 ----
child = children[i]
nextchild = children[i+1]
! nextchildren = nextchild.childNodes
while len(nextchildren):
node = nextchildren[0]
***************
*** 933,937 ****
nodeType = node.nodeType
if nodeType == ELEMENT:
! gi = node.get_tagName()
if knownempty(gi):
if node.hasChildNodes():
--- 929,933 ----
nodeType = node.nodeType
if nodeType == ELEMENT:
! gi = node.tagName
if knownempty(gi):
if node.hasChildNodes():
***************
*** 939,944 ****
"declared-empty node <%s> has children" % gi
ofp.write("e\n")
! for k, v in node.attributes.items():
! value = v.value
if _token_rx.match(value):
dtype = "TOKEN"
--- 935,939 ----
"declared-empty node <%s> has children" % gi
ofp.write("e\n")
! for k, value in node.attributes.items():
if _token_rx.match(value):
dtype = "TOKEN"
***************
*** 952,956 ****
ofp.write("-%s\n" % esistools.encode(node.data))
elif nodeType == ENTITY_REFERENCE:
! ofp.write("&%s\n" % node.get_nodeName())
else:
raise RuntimeError, "unsupported node type: %s" % nodeType
--- 947,951 ----
ofp.write("-%s\n" % esistools.encode(node.data))
elif nodeType == ENTITY_REFERENCE:
! ofp.write("&%s\n" % node.nodeName)
else:
raise RuntimeError, "unsupported node type: %s" % nodeType
***************
*** 958,965 ****
def convert(ifp, ofp):
! p = esistools.ExtendedEsisBuilder()
! p.feed(ifp.read())
! doc = p.document
! fragment = p.fragment
normalize(fragment)
simplify(doc, fragment)
--- 953,961 ----
def convert(ifp, ofp):
! events = esistools.parse(ifp)
! toktype, doc = events.getEvent()
! fragment = doc.createDocumentFragment()
! events.expandNode(fragment)
!
normalize(fragment)
simplify(doc, fragment)
***************
*** 995,1000 ****
#
d = {}
! for gi in p.get_empties():
d[gi] = gi
if d.has_key("rfc"):
del d["rfc"]
--- 991,998 ----
#
d = {}
! for gi in events.parser.get_empties():
d[gi] = gi
+ if d.has_key("author"):
+ del d["author"]
if d.has_key("rfc"):
del d["rfc"]
***************
*** 1020,1028 ****
elif len(sys.argv) == 3:
ifp = open(sys.argv[1])
! ofp = open(sys.argv[2], "w")
else:
usage()
sys.exit(2)
convert(ifp, ofp)
--- 1018,1032 ----
elif len(sys.argv) == 3:
ifp = open(sys.argv[1])
! import StringIO
! ofp = StringIO.StringIO()
else:
usage()
sys.exit(2)
convert(ifp, ofp)
+ if len(sys.argv) == 3:
+ fp = open(sys.argv[2], "w")
+ fp.write(ofp.getvalue())
+ fp.close()
+ ofp.close()