#!/usr/bin/python """ slurpIcalSpec.py -- extract formal view of iCalendar/vCard specs produces XHTML with various typed links @@TODO: make each marked up item visible via css (unless they're all linked). see also: rfc2html in dev.w3.org @@ """ __version__ = '$Id: slurpIcalSpec.py,v 1.27 2005/11/09 23:10:49 connolly Exp $' #see also: changelog at end import sys import re class Usage(Exception): """USAGE: python slurpIcalSpec.py NNNN rfcNNNN.html where NNNN is one of the supported RFCS (so far: 2425, 2426, and 2445) """ # rfcnum: (footerStarters, typos, example_tags) Specs = { 2445: (("RFC", "Dawson"), (('"EVENT"', '"VEVENT"'), # 4.8.7.3 Last Modified (' Purpose This value type', ' Purpose: This value type'), ('11. Full Copyright Statement', '11 Full Copyright Statement'), ('3.11 Contact for Further Information:', '3.11 Contact for Further Information'), ), ('Example',), ), 2426: (("RFC", "Dawson"), # a ref is split across lines; put it on one line (('Hence, this [MIME-', 'Hence, this [MIME-DIR]'), ('DIR] profile is', ' profile is'), ), ('Type example',), ), 2425: (("RFC", "Howes"), (), (), ) } def main(argv): if len(argv) == 2: try: rfcnum = int(argv[1]) pgbrks, typos, exampleTags = Specs[rfcnum] except (ValueError, KeyError): raise Usage() else: raise Usage() fp = sys.stdin lines = depaginate(fp, pgbrks, typos) sections = bySection(lines) head = sections.next() #print >>sys.stderr, "RFC header:", `head` title, rfcnum, category, date, authlines = titleEtc(head) print htmlTop(title, rfcnum, category, date, authlines, exampleTags) w = sys.stdout.write sections = list(sections) refs = findRefs(sections) for sec in sections: if sec[0].find("Table of Contents") >= 0: tocSect(w, sec) elif sec[0][0].isdigit(): numSect(w, sec, refs) else: flowSect(w, sec, refs) print htmlBot() def depaginate(fp, footerStarters, typos): """undo RFC pagination: generate sequence of lines skip lines at top of page 'Each page must be limited to 58 lines followed by a form feed on a line by itself.' --Instructions to RFC Authors Postel & Reynolds Oct 1997 http://www.ietf.org/rfc/rfc2223 """ sv = [] top = 1 while 1: line = fp.readline() if not line: break line = line.rstrip("\r\n") for err, fix in typos: if err in line: line = line.replace(err, fix) for s in footerStarters: if line.startswith(s): line = "\f" break if line == "\f": sv = [''] # leave a blank line where the pagebreak was top = 1 continue if len(line): if sv: for l in sv: yield l sv = [] yield line top = 0 else: if not top: sv.append(line) def bySection(lines): sec = [] for l in lines: if len(l): if l.startswith(" ") or \ l.startswith("Request for Comments") or \ l.startswith("Category:") or \ l.find(":") >= 0 or \ l.find("<") >= 0 or \ l.find("--") >= 0: sec.append(l) else: if len(sec): yield sec sec = [] sec.append(l) else: sec.append(l) if len(sec): yield sec def titleEtc(lines): sep = None for idx in range(0, len(lines)): if lines[idx].strip() == '': sep = idx break if sep is None: raise ValueError, \ "no blank line separating header from title: " + `lines` title = ' '.join(map(lambda l: l.strip(), lines[sep+1:])).strip() rfcnum = lines[1][:40].strip().split()[-1] category = lines[2][:40].strip().split(':')[1].strip() date = lines[sep-1].strip() authlines = [] for idx in range(0, sep-1): authlines.append(lines[idx][40:].strip()) return title, rfcnum, category, date, authlines def findRefs(sections): for sec in sections: if not sec[0].strip().endswith("References"): continue refs = [] state = '' for ln in sec[1:]: ln = ln.strip() if ln == '': state = '' continue if state == '': junk, ln = ln.split('[', 1) ref, ln = ln.split(']', 1) #print >>sys.stderr, "found ref:", ref refs.append(ref) state = 'inref' return refs raise ValueError, "no References section found" def htmlTop(title, rfcnum, category, date, authlines, exampleTags): css = "" for t in exampleTags: css = css + "dd.%s { border-style: solid; border-color: #d0dbe7; border-width: 0 0 0 .25em; padding-left: 0.5em;\n" % asClass(t) html = """

Network Working Group Request for Comments: %s Category: %s	%s
	%s

%s

\n") for ln in lines[1:]: ln = ln.strip() if ln == '': if dd: refEntry(w, ref, dd) state = '' continue if state == '': junk, ln = ln.split('[', 1) ref, dd = ln.split(']', 1) dd = dd + "\n" state = 'inref' elif state == 'inref': dd = dd + ln + "\n" if dd: refEntry(w, ref, dd) w("

")
        for l in lines[1:]:
            bodyText(w, l, refs)
            w("\n")
        w("

\n") secLabel = '%s Name' % secType dt = '' dd = [] idx = 2 while idx < len(lines): l = lines[idx] if ':' in l: hd, rest = l.lstrip().split(":", 1) if hd == secLabel or hd in Tags: if dd: subSect(w, secType, rdfClass, dt, dd, refs) dd = [] dt = hd dd.append(rest) else: dd.append(l) else: dd.append(l) idx += 1 if dd: subSect(w, secType, rdfClass, dt, dd, refs) w("

   %s\n"
          % (rel, name, name, rest))

        for l in dd[1:]:
            l = tokenRefs(w, l, 'allowed-type',
                          {'DATE': 'Value_DATE',
                           'DATE-TIME': 'Value_DATE-TIME',
                           'PERIOD': 'Value_PERIOD',
                           'BINARY': 'Value_BINARY'})
            bodyText(w, l, refs)
            w("\n")

    else:
        tokens = None
        rel = None
        
        if secType == "Property" and \
                        (dt == 'Conformance' or dt == 'Description'):
            rel = 'applies-to'
            tokens = {'VEVENT': 'Vevent',
                      'VTODO': 'Vtodo',
                      'VJOURNAL': 'Vjournal',
                      'VFREEBUSY': 'Vfreebusy',
                      'VTIMEZONE': 'Vtimezone',
                      'VALARM': 'Valarm'
                      }
        elif dt == 'Description' and secType == "Component":
            rel = 'def'
            tokens = {
                'STANDARD': 'standard',
                'DAYLIGHT': 'daylight',
                }
        elif dt == 'Description' and secType == "Value":
            rel = 'def'
            tokens = {
                'FREQ': 'freq',
                'UNTIL': 'until',
                'COUNT': 'count',
                'INTERVAL': 'interval',
                'BYSECOND': 'bysecond',
                'BYMINUTE': 'byminute',
                'BYHOUR': 'byhour',
                'BYDAY': 'byday',
                'BYMONTHDAY': 'bymonthday',
                'BYYEARDAY': 'byyearday',
                'BYWEEKNO': 'byweekno',
                'BYMONTH': 'bymonth',
                'BYSETPOS': 'bysetpos',
                'WKST': 'wkst',
                }

        w("
   ")

        for l in dd:
            if tokens: l = tokenRefs(w, l, rel, tokens)
            bodyText(w, l, refs)
            w("\n")

    w("
\n