def findTags(text):
ret = []
PTTSTATESWITCH = {":": GOTCCOL, "'": SQUOT, '"': DQUOT, "`": BQUOT}
VALID_PTTEXT = (string.letters + string.digits + string.whitespace +
'[]_=/\\#.*')
QUOTE_CLOSE = {SQUOT: "'", DQUOT: '"', BQUOT: '`'}
offset = 0
lentext = len(text)
state = INIT
backslash_save_state = 0
textoffset = 0
tagoffset = 0
while offset < lentext:
c = text[offset]
#print 'c=%s state=%s' % (c, _statenumtostr[state])
if state == INIT:
if c == '<':
tagoffset = offset
state = GOTLT
elif state == GOTLT:
if c == ':':
state = BEGINCOMMENTCHECK
#put in text token
ret.append(
(textoffset, offset - 1, 0, text[textoffset:offset - 1]))
else:
state = INIT
elif state == BEGINCOMMENTCHECK:
if c != '*':
state = PLAINTAGTEXT
continue
state = INCOMMENT
elif state == INCOMMENT:
if c == '*':
state = INCOMMENTSTAR
elif state == INCOMMENTSTAR:
if c == ':':
state = INCOMMENTSTARCOL
else:
state = INCOMMENT
elif state == INCOMMENTSTARCOL:
if c == '>':
state = INIT
textoffset = offset + 1
else:
state = INCOMMENT
elif state == PLAINTAGTEXT:
nstate = PTTSTATESWITCH.get(c)
if nstate:
state = nstate
elif c == '\\':
backslash_save_state = state
state = BACKSLASH
elif c not in VALID_PTTEXT:
if c != '<': #not beginning of next tag?
raise DTLexicalError(lineno=getLineno(text, offset),
msg="invalid character |%s| in tag" %
c)
else:
raise DTLexicalError(lineno=getLineno(text, offset),
msg="found < in tag")
elif state == GOTCCOL:
if c != ">":
#raise DTLexicalError ( lineno = getLineno(text, offset),
# msg = "got : in tag but not followed by >" )
state = PLAINTAGTEXT
else:
textoffset = offset + 1
# put in tag token
ret.append(
[tagoffset, offset + 1, 1, text[tagoffset + 2:offset - 1]])
state = INIT
elif state in (SQUOT, DQUOT, BQUOT):
if c == '\\':
#print 'entering backslash save_state=', _statenumtostr[state]
backslash_save_state = state
state = BACKSLASH
elif c == QUOTE_CLOSE[state]:
state = PLAINTAGTEXT
elif state == BACKSLASH:
#print 'inbackslash'
#print
state = backslash_save_state
offset = offset + 1
if state != INIT:
md = {
PLAINTAGTEXT: "text",
SQUOT: "in single quoted area",
DQUOT: "in double quoted area",
BQUOT: "in expression area",
GOTCCOL: "after closing colon",
BACKSLASH: "after backslash",
INCOMMENT: "in comment"
}
raise DTLexicalError(
msg="hit end of file %s in tag, beginning of "
"tag %s at" % (md[state], text[tagoffset:tagoffset + 10] + "..."),
lineno=getLineno(text, tagoffset))
else:
ret.append([textoffset, offset, 0, text[textoffset:offset]])
return ret
def processTag(tagString, text, st, end):
#print 'tag is <%s>' % tagString
tupargs = []
dictargs = {}
tagname = ''
off = 0
tnmatch = None
while not tnmatch:
wsmatch = wsre.match(tagString, off, len(tagString), pcre.ANCHORED)
if not wsmatch:
tnmatch = plainre.match(tagString, off, len(tagString),
pcre.ANCHORED)
if not tnmatch:
raise DTLexicalError(msg='invalid tag name',
tagtext=tagString,
lineno=getLineno(text, st))
else:
break
off = wsmatch[0][1]
tagname = tagString[tnmatch[0][0]:tnmatch[0][1]]
#print 'tagname is ', tagname
off = tnmatch[0][1]
lts = len(tagString)
while off < lts:
match = tagContentsRe.match(tagString, off, lts, pcre.ANCHORED)
if match is None:
raise DTLexicalError(msg='invalid tag text',
tagtext=tagString,
lineno=getLineno(text, st))
if match[WHITESPACE] != (-1, -1):
mb, me = match[WHITESPACE]
#print 'Got whitespace <%s>' % tagString[mb:me]
off = me
elif match[PLAIN] != (-1, -1):
mb, me = match[PLAIN]
#print 'Got plain <%s>' % tagString[mb:me]
tupargs.append(tagString[mb:me])
off = me
elif match[PLAINEQ] != (-1, -1):
mb, me = match[PLAINEQ]
#print 'Got plaineq <%s>' % tagString[mb:me]
tmb, tme = match[PLAINEQVAL]
#print 'Got plaineqval <%s>' % tagString[tmb:tme]
dictargs[tagString[mb:me]] = tagString[tmb:tme]
off = tme
elif match[PLAINEQQ] != (-1, -1):
mb, me = match[PLAINEQQ]
#print 'Got plaineqq <%s>' % tagString[mb:me]
tmb, tme = match[PLAINEQQVAL]
#print 'Got plaineqqval <%s>' % tagString[tmb:tme]
dictargs[tagString[mb:me]] = dequote(tagString[tmb:tme])
off = tme
elif match[QUOTONLY] != (-1, -1):
mb, me = match[QUOTONLY]
#print 'Got quotonly <%s>' % tagString[mb:me]
tupargs.append(dequote(tagString[mb:me]))
off = me
else:
raise DTLexicalError(msg='unknown text in tag',
tagtext=tagString[off:off + 10],
lineno=getLineno(text, st))
return tagname, tupargs, dictargs
def processTag(tagString, text, st, end):
tupargs = []
dictargs = {}
tagname = ''
off = 0
tnmatch = None
while not tnmatch:
wsmatch = wsre.match(tagString, off, len(tagString))
if not wsmatch:
tnmatch = plainre.match(tagString, off, len(tagString))
if not tnmatch:
raise DTLexicalError(msg='invalid tag name',
tagtext=tagString,
lineno=getLineno(text, st))
else:
break
off = wsmatch.end()
tagname = tnmatch.group()
off = tnmatch.end()
lts = len(tagString)
while off < lts:
match = tagContentsRe.match(tagString, off, lts)
if match is None:
raise DTLexicalError(msg='invalid tag text',
tagtext=tagString,
lineno=getLineno(text, st))
g = match.groupdict()
tmp = g[WHITESPACE]
if tmp is not None:
off = match.end(WHITESPACE)
continue
tmp = g[PLAIN]
if tmp is not None:
tupargs.append(tmp)
off = match.end(PLAIN)
continue
tmp = g[PLAINEQ]
tmp2 = g[PLAINEQVAL]
if tmp is not None:
assert tmp2 is not None
dictargs[tmp] = tmp2
off = match.end(PLAINEQVAL)
continue
tmp = g[PLAINEQQ]
tmp2 = g[PLAINEQQVAL]
if tmp is not None:
assert tmp2 is not None
dictargs[tmp] = dequote(tmp2)
off = match.end(PLAINEQQVAL)
continue
tmp = g[QUOTONLY]
if tmp is not None:
tupargs.append(dequote(tmp))
off = match.end(QUOTONLY)
continue
raise DTLexicalError(msg='unknown text in tag',
tagtext=tagString[off:off + 10],
lineno=getLineno(text, st))
return tagname, tupargs, dictargs