def write_esis(doc, ofp, knownempty):
for node in doc.childNodes:
nodeType = node.nodeType
if nodeType == ELEMENT:
gi = node.tagName
if knownempty(gi):
if node.hasChildNodes():
raise ValueError, \
"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"
else:
dtype = "CDATA"
ofp.write("A%s %s %s\n" % (k, dtype, esistools.encode(value)))
ofp.write("(%s\n" % gi)
write_esis(node, ofp, knownempty)
ofp.write(")%s\n" % gi)
elif nodeType == TEXT:
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
def dump_attr(self, pentry, value):
if not (pentry.name and value):
return
if _token_rx.match(value):
dtype = "TOKEN"
else:
dtype = "CDATA"
self.write("A%s %s %s\n" % (pentry.name, dtype, encode(value)))
def dump_attr(self, pentry, value):
if not (pentry.name and value):
return
if _token_rx.match(value):
dtype = "TOKEN"
else:
dtype = "CDATA"
self.write("A%s %s %s\n" % (pentry.name, dtype, encode(value)))
def subconvert(self, endchar=None, depth=0):
#
# Parses content, including sub-structures, until the character
# 'endchar' is found (with no open structures), or until the end
# of the input data is endchar is None.
#
stack = new_stack()
line = self.line
while line:
if line[0] == endchar and not stack:
self.line = line
return line
m = _comment_rx.match(line)
if m:
text = m.group(1)
if text:
self.write("(COMMENT\n- %s \n)COMMENT\n-\\n\n"
% encode(text))
line = line[m.end():]
continue
m = _begin_env_rx.match(line)
if m:
name = m.group(1)
entry = self.get_env_entry(name)
# re-write to use the macro handler
line = r"\%s %s" % (name, line[m.end():])
continue
m = _end_env_rx.match(line)
if m:
# end of environment
envname = m.group(1)
entry = self.get_entry(envname)
while stack and envname != stack[-1] \
and stack[-1] in entry.endcloses:
self.write(")%s\n" % stack.pop())
if stack and envname == stack[-1]:
self.write(")%s\n" % entry.outputname)
del stack[-1]
else:
raise LaTeXStackError(envname, stack)
line = line[m.end():]
continue
m = _begin_macro_rx.match(line)
if m:
# start of macro
macroname = m.group(1)
if macroname == "c":
# Ugh! This is a combining character...
endpos = m.end()
self.combining_char("c", line[endpos])
line = line[endpos + 1:]
continue
entry = self.get_entry(macroname)
if entry.verbatim:
# magic case!
pos = line.find("\\end{%s}" % macroname)
text = line[m.end(1):pos]
stack.append(entry.name)
self.write("(%s\n" % entry.outputname)
self.write("-%s\n" % encode(text))
self.write(")%s\n" % entry.outputname)
stack.pop()
line = line[pos + len("\\end{%s}" % macroname):]
continue
while stack and stack[-1] in entry.closes:
top = stack.pop()
topentry = self.get_entry(top)
if topentry.outputname:
self.write(")%s\n-\\n\n" % topentry.outputname)
#
if entry.outputname and entry.empty:
self.write("e\n")
#
params, optional, empty = self.start_macro(macroname)
# rip off the macroname
if params:
line = line[m.end(1):]
elif empty:
line = line[m.end(1):]
else:
line = line[m.end():]
opened = 0
implied_content = 0
# handle attribute mappings here:
for pentry in params:
if pentry.type == "attribute":
if pentry.optional:
m = _optional_rx.match(line)
if m and entry.outputname:
line = line[m.end():]
self.dump_attr(pentry, m.group(1))
elif pentry.text and entry.outputname:
# value supplied by conversion spec:
self.dump_attr(pentry, pentry.text)
else:
m = _parameter_rx.match(line)
if not m:
raise LaTeXFormatError(
"could not extract parameter %s for %s: %r"
% (pentry.name, macroname, line[:100]))
if entry.outputname:
self.dump_attr(pentry, m.group(1))
line = line[m.end():]
elif pentry.type == "child":
if pentry.optional:
m = _optional_rx.match(line)
if m:
line = line[m.end():]
if entry.outputname and not opened:
opened = 1
self.write("(%s\n" % entry.outputname)
stack.append(macroname)
stack.append(pentry.name)
self.write("(%s\n" % pentry.name)
self.write("-%s\n" % encode(m.group(1)))
self.write(")%s\n" % pentry.name)
stack.pop()
else:
if entry.outputname and not opened:
opened = 1
self.write("(%s\n" % entry.outputname)
stack.append(entry.name)
self.write("(%s\n" % pentry.name)
stack.append(pentry.name)
self.line = skip_white(line)[1:]
line = self.subconvert(
"}", len(stack) + depth + 1)[1:]
self.write(")%s\n" % stack.pop())
elif pentry.type == "content":
if pentry.implied:
implied_content = 1
else:
if entry.outputname and not opened:
opened = 1
self.write("(%s\n" % entry.outputname)
stack.append(entry.name)
line = skip_white(line)
if line[0] != "{":
raise LaTeXFormatError(
"missing content for " + macroname)
self.line = line[1:]
line = self.subconvert("}", len(stack) + depth + 1)
if line and line[0] == "}":
line = line[1:]
elif pentry.type == "text" and pentry.text:
if entry.outputname and not opened:
opened = 1
stack.append(entry.name)
self.write("(%s\n" % entry.outputname)
#dbgmsg("--- text: %r" % pentry.text)
self.write("-%s\n" % encode(pentry.text))
elif pentry.type == "entityref":
self.write("&%s\n" % pentry.name)
if entry.outputname:
if not opened:
self.write("(%s\n" % entry.outputname)
stack.append(entry.name)
if not implied_content:
self.write(")%s\n" % entry.outputname)
stack.pop()
continue
if line[0] == endchar and not stack:
self.line = line[1:]
return self.line
if line[0] == "}":
# end of macro or group
macroname = stack[-1]
if macroname:
conversion = self.table[macroname]
if conversion.outputname:
# otherwise, it was just a bare group
self.write(")%s\n" % conversion.outputname)
del stack[-1]
line = line[1:]
continue
if line[0] == "~":
# don't worry about the "tie" aspect of this command
line = line[1:]
self.write("- \n")
continue
if line[0] == "{":
stack.append("")
line = line[1:]
continue
if line[0] == "\\" and line[1] in ESCAPED_CHARS:
self.write("-%s\n" % encode(line[1]))
line = line[2:]
continue
if line[:2] == r"\\":
self.write("(BREAK\n)BREAK\n")
line = line[2:]
continue
if line[:2] == r"\_":
line = "_" + line[2:]
continue
if line[:2] in (r"\'", r'\"'):
# combining characters...
self.combining_char(line[1], line[2])
line = line[3:]
continue
m = _text_rx.match(line)
if m:
text = encode(m.group())
self.write("-%s\n" % text)
line = line[m.end():]
continue
# special case because of \item[]
# XXX can we axe this???
if line[0] == "]":
self.write("-]\n")
line = line[1:]
continue
# avoid infinite loops
extra = ""
if len(line) > 100:
extra = "..."
raise LaTeXFormatError("could not identify markup: %r%s"
% (line[:100], extra))
while stack:
entry = self.get_entry(stack[-1])
if entry.closes:
self.write(")%s\n-%s\n" % (entry.outputname, encode("\n")))
del stack[-1]
else:
break
if stack:
raise LaTeXFormatError("elements remain on stack: "
+ ", ".join(stack))
def subconvert(self, endchar=None, depth=0):
#
# Parses content, including sub-structures, until the character
# 'endchar' is found (with no open structures), or until the end
# of the input data is endchar is None.
#
stack = new_stack()
line = self.line
while line:
if line[0] == endchar and not stack:
self.line = line
return line
m = _comment_rx.match(line)
if m:
text = m.group(1)
if text:
self.write("(COMMENT\n- %s \n)COMMENT\n-\\n\n" %
encode(text))
line = line[m.end():]
continue
m = _begin_env_rx.match(line)
if m:
name = m.group(1)
entry = self.get_env_entry(name)
# re-write to use the macro handler
line = r"\%s %s" % (name, line[m.end():])
continue
m = _end_env_rx.match(line)
if m:
# end of environment
envname = m.group(1)
entry = self.get_entry(envname)
while stack and envname != stack[-1] \
and stack[-1] in entry.endcloses:
self.write(")%s\n" % stack.pop())
if stack and envname == stack[-1]:
self.write(")%s\n" % entry.outputname)
del stack[-1]
else:
raise LaTeXStackError(envname, stack)
line = line[m.end():]
continue
m = _begin_macro_rx.match(line)
if m:
# start of macro
macroname = m.group(1)
if macroname == "c":
# Ugh! This is a combining character...
endpos = m.end()
self.combining_char("c", line[endpos])
line = line[endpos + 1:]
continue
entry = self.get_entry(macroname)
if entry.verbatim:
# magic case!
pos = line.find("\\end{%s}" % macroname)
text = line[m.end(1):pos]
stack.append(entry.name)
self.write("(%s\n" % entry.outputname)
self.write("-%s\n" % encode(text))
self.write(")%s\n" % entry.outputname)
stack.pop()
line = line[pos + len("\\end{%s}" % macroname):]
continue
while stack and stack[-1] in entry.closes:
top = stack.pop()
topentry = self.get_entry(top)
if topentry.outputname:
self.write(")%s\n-\\n\n" % topentry.outputname)
#
if entry.outputname and entry.empty:
self.write("e\n")
#
params, optional, empty = self.start_macro(macroname)
# rip off the macroname
if params:
line = line[m.end(1):]
elif empty:
line = line[m.end(1):]
else:
line = line[m.end():]
opened = 0
implied_content = 0
# handle attribute mappings here:
for pentry in params:
if pentry.type == "attribute":
if pentry.optional:
m = _optional_rx.match(line)
if m and entry.outputname:
line = line[m.end():]
self.dump_attr(pentry, m.group(1))
elif pentry.text and entry.outputname:
# value supplied by conversion spec:
self.dump_attr(pentry, pentry.text)
else:
m = _parameter_rx.match(line)
if not m:
raise LaTeXFormatError(
"could not extract parameter %s for %s: %r"
% (pentry.name, macroname, line[:100]))
if entry.outputname:
self.dump_attr(pentry, m.group(1))
line = line[m.end():]
elif pentry.type == "child":
if pentry.optional:
m = _optional_rx.match(line)
if m:
line = line[m.end():]
if entry.outputname and not opened:
opened = 1
self.write("(%s\n" % entry.outputname)
stack.append(macroname)
stack.append(pentry.name)
self.write("(%s\n" % pentry.name)
self.write("-%s\n" % encode(m.group(1)))
self.write(")%s\n" % pentry.name)
stack.pop()
else:
if entry.outputname and not opened:
opened = 1
self.write("(%s\n" % entry.outputname)
stack.append(entry.name)
self.write("(%s\n" % pentry.name)
stack.append(pentry.name)
self.line = skip_white(line)[1:]
line = self.subconvert("}",
len(stack) + depth + 1)[1:]
self.write(")%s\n" % stack.pop())
elif pentry.type == "content":
if pentry.implied:
implied_content = 1
else:
if entry.outputname and not opened:
opened = 1
self.write("(%s\n" % entry.outputname)
stack.append(entry.name)
line = skip_white(line)
if line[0] != "{":
raise LaTeXFormatError("missing content for " +
macroname)
self.line = line[1:]
line = self.subconvert("}", len(stack) + depth + 1)
if line and line[0] == "}":
line = line[1:]
elif pentry.type == "text" and pentry.text:
if entry.outputname and not opened:
opened = 1
stack.append(entry.name)
self.write("(%s\n" % entry.outputname)
#dbgmsg("--- text: %r" % pentry.text)
self.write("-%s\n" % encode(pentry.text))
elif pentry.type == "entityref":
self.write("&%s\n" % pentry.name)
if entry.outputname:
if not opened:
self.write("(%s\n" % entry.outputname)
stack.append(entry.name)
if not implied_content:
self.write(")%s\n" % entry.outputname)
stack.pop()
continue
if line[0] == endchar and not stack:
self.line = line[1:]
return self.line
if line[0] == "}":
# end of macro or group
macroname = stack[-1]
if macroname:
conversion = self.table[macroname]
if conversion.outputname:
# otherwise, it was just a bare group
self.write(")%s\n" % conversion.outputname)
del stack[-1]
line = line[1:]
continue
if line[0] == "~":
# don't worry about the "tie" aspect of this command
line = line[1:]
self.write("- \n")
continue
if line[0] == "{":
stack.append("")
line = line[1:]
continue
if line[0] == "\\" and line[1] in ESCAPED_CHARS:
self.write("-%s\n" % encode(line[1]))
line = line[2:]
continue
if line[:2] == r"\\":
self.write("(BREAK\n)BREAK\n")
line = line[2:]
continue
if line[:2] == r"\_":
line = "_" + line[2:]
continue
if line[:2] in (r"\'", r'\"'):
# combining characters...
self.combining_char(line[1], line[2])
line = line[3:]
continue
m = _text_rx.match(line)
if m:
text = encode(m.group())
self.write("-%s\n" % text)
line = line[m.end():]
continue
# special case because of \item[]
# XXX can we axe this???
if line[0] == "]":
self.write("-]\n")
line = line[1:]
continue
# avoid infinite loops
extra = ""
if len(line) > 100:
extra = "..."
raise LaTeXFormatError("could not identify markup: %r%s" %
(line[:100], extra))
while stack:
entry = self.get_entry(stack[-1])
if entry.closes:
self.write(")%s\n-%s\n" % (entry.outputname, encode("\n")))
del stack[-1]
else:
break
if stack:
raise LaTeXFormatError("elements remain on stack: " +
", ".join(stack))