view MoinMoin/parser/ @ 3911:348c9b6e0033

ParserBase (and other parsers deriving from it): cleanup / refactorings: source code reformattings / rearrangements, PEP8 fixes using booleans instead of 0/1, use raw strings for regexes reduce code duplication better attr names introduce tabwidth constant as a class member
author Thomas Waldmann <tw AT waldmann-edv DOT de>
date Sat, 02 Aug 2008 13:48:59 +0200
parents 01f05e74aa9c
children 01ba9b2e96f6
line wrap: on
line source
# -*- coding: iso-8859-1 -*-
    MoinMoin - Base Source Parser

    @copyright: 2002 by Taesu Pyo <>,
                2005 by Oliver Graf <>,
                2005-2008 MoinMoin:ThomasWaldmann

    @license: GNU GPL, see COPYING for details.

basic css:

pre.codearea     { font-style: sans-serif; color: #000000; }

pre.codearea span.ID       { color: #000000; }
pre.codearea span.Char     { color: #004080; }
pre.codearea span.Comment  { color: #808080; }
pre.codearea span.Number   { color: #008080; font-weight: bold; }
pre.codearea span.String   { color: #004080; }
pre.codearea span.SPChar   { color: #0000C0; }
pre.codearea span.ResWord  { color: #4040ff; font-weight: bold; }
pre.codearea span.ConsWord { color: #008080; font-weight: bold; }


import re, sha

from MoinMoin import config, wikiutil

class FormatTextBase:

class FormatText(FormatTextBase):

    def __init__(self, fmt):
        self.fmt = fmt

    def formatString(self, formatter, word):
        return (formatter.code_token(1, self.fmt) +
                formatter.text(word) +
                formatter.code_token(0, self.fmt))

class FormatTextID(FormatTextBase):

    def __init__(self, fmt, icase=False):
        if not isinstance(fmt, FormatText):
            fmt = FormatText(fmt)
        self._ignore_case = icase
        self.fmt = {}

    def setDefaultFormat(self, fmt):
        self._def_fmt = fmt

    def addFormat(self, word, fmt):
        if self._ignore_case:
            word = word.lower()
        self.fmt[word] = fmt

    def formatString(self, formatter, word):
        if self._ignore_case:
            sword = word.lower()
            sword = word
        return self.fmt.get(sword, self._def_fmt).formatString(formatter, word)

class FormattingRuleSingle:

    def __init__(self, name, str_re, icase=False): = name
        self.str_re = str_re

    def getStartRe(self):
        return self.str_re

    def getText(self, parser, hit):
        return hit

class FormattingRulePair:

    def __init__(self, name, str_begin, str_end, icase=False): = name
        self.str_begin = str_begin
        self.str_end = str_end
        re_flags = re.M
        if icase:
            re_flags |= re.I
        self.end_re = re.compile(str_end, re_flags)

    def getStartRe(self):
        return self.str_begin

    def getText(self, parser, hit):
        match =, parser.lastpos)
        if not match:
            next_lastpos = len(parser.line)
            next_lastpos = match.end() + (match.end() == parser.lastpos)
        r = parser.line[parser.lastpos:next_lastpos]
        parser.lastpos = next_lastpos
        return hit + r

# ------------------------------------------------------------------------

def parse_start_step(request, args):
    Parses common Colorizer parameters start, step, numbers.
    Uses L{wikiutil.parseAttributes} and sanitizes the results.

    Start and step must be a non negative number and default to 1,
    numbers might be on, off, or none and defaults to on. On or off
    means that numbers are switchable via JavaScript (html formatter),
    disabled means that numbers are disabled completely.

    attrdict is returned as last element in the tuple, to enable the
    calling parser to extract further arguments.

    @param request: a request instance
    @param args: the argument string

    @returns: numbers, start, step, attrdict
    nums, start, step = 1, 1, 1
    attrs, msg = wikiutil.parseAttributes(request, args)
    if not msg:
            start = int(attrs.get('start', '"1"')[1:-1])
        except ValueError:
            step = int(attrs.get('step', '"1"')[1:-1])
        except ValueError:
        if attrs.get('numbers', '"on"')[1:-1].lower() in ('off', 'false', 'no'):
            nums = 0
        elif attrs.get('numbers', '"on"')[1:-1].lower() in ('none', 'disable'):
            nums = -1
    return nums, start, step, attrs

class ParserBase:

    parsername = 'ParserBase'
    tabwidth = 4

    def __init__(self, raw, request, **kw):
        self.raw = raw
        self.request = request
        self.show_nums, self.num_start, self.num_step, attrs = parse_start_step(request, kw.get('format_args', ''))

        self._ignore_case = False
        self._formatting_rules = []
        self._formatting_rules_n2r = {}
        self._formatting_rule_index = 0
        self.rule_fmt = {}
        self.line_count = len(raw.split('\n')) + 1

    def setupRules(self):
        self.def_format = FormatText('Default')
        self.ID_format = FormatTextID('ID', self._ignore_case)
        self.addRuleFormat("ID", self.ID_format)
        self.reserved_word_format = FormatText('ResWord')
        self.constant_word_format = FormatText('ConsWord')

    def addRule(self, name, str_re):
        self._formatting_rule_index += 1
        n = "%s_%s" % (name, self._formatting_rule_index)
        f = FormattingRuleSingle(name, str_re, self._ignore_case)
        self._formatting_rules.append((n, f))
        self._formatting_rules_n2r[n] = f

    def addRulePair(self, name, start_re, end_re):
        self._formatting_rule_index += 1
        n = "%s_%s" % (name, self._formatting_rule_index)
        f = FormattingRulePair(name, start_re, end_re, self._ignore_case)
        self._formatting_rules.append((n, f))
        self._formatting_rules_n2r[n] = f

    def addWords(self, words, fmt):
        if not isinstance(fmt, FormatTextBase):
            fmt = FormatText(fmt)
        for w in words:
            self.ID_format.addFormat(w, fmt)

    def addReserved(self, words):
        self.addWords(words, self.reserved_word_format)

    def addConstant(self, words):
        self.addWords(words, self.constant_word_format)

    def addRuleFormat(self, name, fmt=None):
        if fmt is None:
            fmt = FormatText(name)
        self.rule_fmt[name] = fmt

    def format(self, formatter, form=None):
        """ Send the text.


        formatting_regexes = ["(?P<%s>%s)" % (n, f.getStartRe())
                              for n, f in self._formatting_rules]
        re_flags = re.M
        if self._ignore_case:
            re_flags |= re.I
        scan_re = re.compile("|".join(formatting_regexes), re_flags)

        self.lastpos = 0
        self.line = self.raw

        self._code_id =
        self.request.write(formatter.code_area(1, self._code_id, self.parsername, self.show_nums, self.num_start, self.num_step))

            #formatter, len('%d' % (self.line_count,)))

        match =

        while match and self.lastpos < len(self.line):
            # add the match we found
            self.lastpos = match.end() + (match.end() == self.lastpos)

            self.write_match(formatter, match)

            # search for the next one
            match =, self.lastpos)

        self.write_normal_text(formatter, self.line[self.lastpos:])

        self.request.write(formatter.code_area(0, self._code_id))

    def write_normal_text(self, formatter, text):
        first = True
        for line in text.expandtabs(self.tabwidth).split('\n'):
            if not first:
                first = False

    def write_match(self, formatter, match):
        for n, hit in match.groupdict().items():
            if not hit:
            r = self._formatting_rules_n2r[n]
            s = r.getText(self, hit)
            c = self.rule_fmt.get(, None)
            if not c:
                c = self.def_format
            first = True
            for line in s.expandtabs(self.tabwidth).split('\n'):
                if not first:
                    first = False
                self.request.write(c.formatString(formatter, line))