diff options
author | Fred Drake <fdrake@acm.org> | 1999-07-29 22:42:27 (GMT) |
---|---|---|
committer | Fred Drake <fdrake@acm.org> | 1999-07-29 22:42:27 (GMT) |
commit | eac8abee4ecec2e88e0e3ad18b32de250e48a44a (patch) | |
tree | f533dc138e9d6550a6dee02baf9883cb9f4ecf88 /Doc | |
parent | 865e9ff98e2a5f77bf01aad6d4d7e7ec5ca7c9bd (diff) | |
download | cpython-eac8abee4ecec2e88e0e3ad18b32de250e48a44a.zip cpython-eac8abee4ecec2e88e0e3ad18b32de250e48a44a.tar.gz cpython-eac8abee4ecec2e88e0e3ad18b32de250e48a44a.tar.bz2 |
Ripped out the old conversion.
Diffstat (limited to 'Doc')
-rwxr-xr-x | Doc/tools/sgmlconv/latex2esis.py | 374 |
1 files changed, 7 insertions, 367 deletions
diff --git a/Doc/tools/sgmlconv/latex2esis.py b/Doc/tools/sgmlconv/latex2esis.py index 051c374..bc28562 100755 --- a/Doc/tools/sgmlconv/latex2esis.py +++ b/Doc/tools/sgmlconv/latex2esis.py @@ -135,239 +135,6 @@ class BaseConversion: class Conversion(BaseConversion): - def subconvert(self, endchar=None, depth=0): - stack = self.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: - # re-write to use the macro handler - line = r"\%s %s" % (m.group(1), line[m.end():]) - continue - m = _end_env_rx.match(line) - if m: - # end of environment - envname = m.group(1) - if envname == "document": - # special magic - for n in stack[1:]: - if n not in self.autoclosing: - self.err_write(stack) - raise LaTeXFormatError( - "open element on stack: " + `n`) - self.write(")document\n") - elif stack and envname == stack[-1]: - self.write(")%s\n" % envname) - del stack[-1] - popping(envname, "a", len(stack) + depth) - 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 == "verbatim": - # really magic case! - pos = string.find(line, "\\end{verbatim}") - text = line[m.end(1):pos] - self.write("(verbatim\n") - self.write("-%s\n" % encode(text)) - self.write(")verbatim\n") - line = line[pos + len("\\end{verbatim}"):] - continue - numbered = 1 - opened = 0 - if macroname[-1] == "*": - macroname = macroname[:-1] - numbered = 0 - if macroname in self.autoclosing and macroname in stack: - while stack[-1] != macroname: - top = stack.pop() - if top and top not in self.discards: - self.write(")%s\n-\\n\n" % top) - popping(top, "b", len(stack) + depth) - if macroname not in self.discards: - self.write("-\\n\n)%s\n-\\n\n" % macroname) - popping(macroname, "c", len(stack) + depth - 1) - del stack[-1] - # - if macroname in self.discards: - self.push_output(StringIO.StringIO()) - else: - self.push_output(self.ofp) - # - params, optional, empty, environ = self.start_macro(macroname) - if not numbered: - self.write("Anumbered TOKEN no\n") - # rip off the macroname - if params: - line = line[m.end(1):] - elif empty: - line = line[m.end(1):] - else: - line = line[m.end():] - # - # Very ugly special case to deal with \item[]. The catch - # is that this needs to occur outside the for loop that - # handles attribute parsing so we can 'continue' the outer - # loop. - # - if optional and type(params[0]) is TupleType: - # the attribute name isn't used in this special case - stack.append(macroname) - self.write("(%s\n" % macroname) - m = _start_optional_rx.match(line) - if m: - self.line = line[m.end():] - line = self.subconvert("]", depth + len(stack)) - line = "}" + line - continue - # handle attribute mappings here: - for attrname in params: - if optional: - optional = 0 - if type(attrname) is StringType: - m = _optional_rx.match(line) - if m: - line = line[m.end():] - self.write("A%s TOKEN %s\n" - % (attrname, encode(m.group(1)))) - elif type(attrname) is TupleType: - # This is a sub-element; but place the and attribute - # we found on the stack (\section-like); the - # content of the macro will become the content - # of the attribute element, and the macro will - # have to be closed some other way (such as - # auto-closing). - stack.append(macroname) - self.write("(%s\n" % macroname) - macroname = attrname[0] - m = _start_group_rx.match(line) - if m: - line = line[m.end():] - elif type(attrname) is ListType: - # A normal subelement: <macroname><attrname>...</>... - attrname = attrname[0] - if not opened: - opened = 1 - self.write("(%s\n" % macroname) - pushing(macroname, "c", len(stack) + depth) - self.write("(%s\n" % attrname) - pushing(attrname, "sub-elem", len(stack) + depth + 1) - self.line = skip_white(line)[1:] - line = self.subconvert("}", len(stack) + depth + 1)[1:] - popping(attrname, "sub-elem", len(stack) + depth + 1) - self.write(")%s\n" % attrname) - else: - m = _parameter_rx.match(line) - if not m: - raise LaTeXFormatError( - "could not extract parameter %s for %s: %s" - % (attrname, macroname, `line[:100]`)) - value = m.group(1) - if _token_rx.match(value): - dtype = "TOKEN" - else: - dtype = "CDATA" - self.write("A%s %s %s\n" - % (attrname, dtype, encode(value))) - line = line[m.end():] - if params and type(params[-1]) is StringType \ - and (not empty) and not environ: - # attempt to strip off next '{' - m = _start_group_rx.match(line) - if not m: - raise LaTeXFormatError( - "non-empty element '%s' has no content: %s" - % (macroname, line[:12])) - line = line[m.end():] - if not opened: - self.write("(%s\n" % macroname) - pushing(macroname, "d", len(stack) + depth) - if empty: - line = "}" + line - stack.append(macroname) - self.pop_output() - 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] - conversion = self.table.get(macroname) - if macroname \ - and macroname not in self.discards \ - and type(conversion) is not StringType: - # otherwise, it was just a bare group - self.write(")%s\n" % stack[-1]) - popping(macroname, "d", len(stack) + depth - 1) - del stack[-1] - line = line[1:] - continue - if line[0] == "{": - pushing("", "e", len(stack) + depth) - 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 - 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[] - 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: %s%s" - % (`line[:100]`, extra)) - while stack and stack[-1] in self.autoclosing: - self.write("-\\n\n") - self.write(")%s\n" % stack[-1]) - popping(stack.pop(), "e", len(stack) + depth - 1) - if stack: - raise LaTeXFormatError("elements remain on stack: " - + string.join(stack, ", ")) - # otherwise we just ran out of input here... - - def start_macro(self, name): - conversion = self.table.get(name, ([], 0, 0, 0, 0)) - params, optional, empty, environ, nocontent = conversion - if empty: - self.write("e\n") - elif nocontent: - empty = 1 - return params, optional, empty, environ - - -class NewConversion(BaseConversion): def __init__(self, ifp, ofp, table={}): BaseConversion.__init__(self, ifp, ofp, table) self.discards = [] @@ -635,17 +402,8 @@ class NewConversion(BaseConversion): self.write("A%s %s %s\n" % (pentry.name, dtype, encode(value))) -def old_convert(ifp, ofp, table={}, discards=(), autoclosing=()): - c = Conversion(ifp, ofp, table, discards, autoclosing) - try: - c.convert() - except IOError, (err, msg): - if err != errno.EPIPE: - raise - - -def new_convert(ifp, ofp, table={}, discards=(), autoclosing=()): - c = NewConversion(ifp, ofp, table) +def convert(ifp, ofp, table): + c = Conversion(ifp, ofp, table) try: c.convert() except IOError, (err, msg): @@ -768,17 +526,9 @@ def load_table(fp): def main(): global DEBUG # - convert = new_convert - newstyle = 1 - opts, args = getopt.getopt(sys.argv[1:], "Dn", ["debug", "new"]) + opts, args = getopt.getopt(sys.argv[1:], "D", ["debug"]) for opt, arg in opts: - if opt in ("-n", "--new"): - convert = new_convert - newstyle = 1 - elif opt in ("-o", "--old"): - convert = old_convert - newstyle = 0 - elif opt in ("-D", "--debug"): + if opt in ("-D", "--debug"): DEBUG = DEBUG + 1 if len(args) == 0: ifp = sys.stdin @@ -792,119 +542,9 @@ def main(): else: usage() sys.exit(2) - table = { - # entries have the form: - # name: ([attribute names], is1stOptional, isEmpty, isEnv, nocontent) - # attribute names can be: - # "string" -- normal attribute - # ("string",) -- sub-element with content of macro; like for \section - # ["string"] -- sub-element - "bifuncindex": (["name"], 0, 1, 0, 0), - "cfuncdesc": (["type", "name", ("args",)], 0, 0, 1, 0), - "chapter": ([("title",)], 0, 0, 0, 0), - "chapter*": ([("title",)], 0, 0, 0, 0), - "classdesc": (["name", ("args",)], 0, 0, 1, 0), - "ctypedesc": (["name"], 0, 0, 1, 0), - "cvardesc": (["type", "name"], 0, 0, 1, 0), - "datadesc": (["name"], 0, 0, 1, 0), - "declaremodule": (["id", "type", "name"], 1, 1, 0, 0), - "deprecated": (["release"], 0, 0, 0, 0), - "documentclass": (["classname"], 0, 1, 0, 0), - "excdesc": (["name"], 0, 0, 1, 0), - "funcdesc": (["name", ("args",)], 0, 0, 1, 0), - "funcdescni": (["name", ("args",)], 0, 0, 1, 0), - "funcline": (["name"], 0, 0, 0, 0), - "funclineni": (["name"], 0, 0, 0, 0), - "geq": ([], 0, 1, 0, 0), - "hline": ([], 0, 1, 0, 0), - "include": (["source"], 0, 1, 0, 0), - "indexii": (["ie1", "ie2"], 0, 1, 0, 0), - "indexiii": (["ie1", "ie2", "ie3"], 0, 1, 0, 0), - "indexiv": (["ie1", "ie2", "ie3", "ie4"], 0, 1, 0, 0), - "indexname": ([], 0, 0, 0, 0), - "input": (["source"], 0, 1, 0, 0), - "item": ([("leader",)], 1, 0, 0, 0), - "label": (["id"], 0, 1, 0, 0), - "labelwidth": ([], 0, 1, 0, 0), - "large": ([], 0, 1, 0, 0), - "LaTeX": ([], 0, 1, 0, 0), - "leftmargin": ([], 0, 1, 0, 0), - "leq": ([], 0, 1, 0, 0), - "lineii": ([["entry"], ["entry"]], 0, 0, 0, 1), - "lineiii": ([["entry"], ["entry"], ["entry"]], 0, 0, 0, 1), - "lineiv": ([["entry"], ["entry"], ["entry"], ["entry"]], 0, 0, 0, 1), - "localmoduletable": ([], 0, 1, 0, 0), - "makeindex": ([], 0, 1, 0, 0), - "makemodindex": ([], 0, 1, 0, 0), - "maketitle": ([], 0, 1, 0, 0), - "manpage": (["name", "section"], 0, 1, 0, 0), - "memberdesc": (["class", "name"], 1, 0, 1, 0), - "memberdescni": (["class", "name"], 1, 0, 1, 0), - "methoddesc": (["class", "name", ("args",)], 1, 0, 1, 0), - "methoddescni": (["class", "name", ("args",)], 1, 0, 1, 0), - "methodline": (["class", "name"], 1, 0, 0, 0), - "methodlineni": (["class", "name"], 1, 0, 0, 0), - "moduleauthor": (["name", "email"], 0, 1, 0, 0), - "opcodedesc": (["name", "var"], 0, 0, 1, 0), - "par": ([], 0, 1, 0, 0), - "paragraph": ([("title",)], 0, 0, 0, 0), - "refbimodindex": (["name"], 0, 1, 0, 0), - "refexmodindex": (["name"], 0, 1, 0, 0), - "refmodindex": (["name"], 0, 1, 0, 0), - "refstmodindex": (["name"], 0, 1, 0, 0), - "refmodule": (["ref"], 1, 0, 0, 0), - "renewcommand": (["macro"], 0, 0, 0, 0), - "rfc": (["num"], 0, 1, 0, 0), - "section": ([("title",)], 0, 0, 0, 0), - "sectionauthor": (["name", "email"], 0, 1, 0, 0), - "seemodule": (["ref", "name"], 1, 0, 0, 0), - "stindex": (["type"], 0, 1, 0, 0), - "subparagraph": ([("title",)], 0, 0, 0, 0), - "subsection": ([("title",)], 0, 0, 0, 0), - "subsubsection": ([("title",)], 0, 0, 0, 0), - "list": (["bullet", "init"], 0, 0, 1, 0), - "tableii": (["colspec", "style", - ["entry"], ["entry"]], 0, 0, 1, 0), - "tableiii": (["colspec", "style", - ["entry"], ["entry"], ["entry"]], 0, 0, 1, 0), - "tableiv": (["colspec", "style", - ["entry"], ["entry"], ["entry"], ["entry"]], 0, 0, 1, 0), - "version": ([], 0, 1, 0, 0), - "versionadded": (["version"], 0, 1, 0, 0), - "versionchanged": (["version"], 0, 1, 0, 0), - "withsubitem": (["text"], 0, 0, 0, 0), - # - "ABC": ([], 0, 1, 0, 0), - "ASCII": ([], 0, 1, 0, 0), - "C": ([], 0, 1, 0, 0), - "Cpp": ([], 0, 1, 0, 0), - "EOF": ([], 0, 1, 0, 0), - "e": ([], 0, 1, 0, 0), - "ldots": ([], 0, 1, 0, 0), - "NULL": ([], 0, 1, 0, 0), - "POSIX": ([], 0, 1, 0, 0), - "UNIX": ([], 0, 1, 0, 0), - # - # Things that will actually be going away! - # - "appendix": ([], 0, 1, 0, 0), - "catcode": ([], 0, 1, 0, 0), - "fi": ([], 0, 1, 0, 0), - "ifhtml": ([], 0, 1, 0, 0), - "makeindex": ([], 0, 1, 0, 0), - "makemodindex": ([], 0, 1, 0, 0), - "maketitle": ([], 0, 1, 0, 0), - "noindent": ([], 0, 1, 0, 0), - "protect": ([], 0, 1, 0, 0), - "tableofcontents": ([], 0, 1, 0, 0), - } - if newstyle: - table = load_table(open(os.path.join(sys.path[0], 'conversion.xml'))) - convert(ifp, ofp, table, - discards=["fi", "ifhtml", "makeindex", "makemodindex", "maketitle", - "noindent", "tableofcontents"], - autoclosing=["chapter", "section", "subsection", "subsubsection", - "paragraph", "subparagraph", ]) + + table = load_table(open(os.path.join(sys.path[0], 'conversion.xml'))) + convert(ifp, ofp, table) if __name__ == "__main__": |