def __iter__(self):
start, fields = self.__schema
proto = dict((k, dv) for k, (_, dv) in fields.iteritems())
default_field = (lambda x: x, None)
def record(rec):
rec_ = proto.copy()
rec_.update(rec)
for fn, err in ifilter(lambda i: isinstance(i[1], ErrorLevel),
rec_.iteritems()):
if err:
self._error(err, err.msg, rec, rec.name, fn)
rec_[fn] = None
return rec_
def accum(db, m):
valuator = fields.get(m.name, default_field)
try:
field = (m.name, valuator[0](m[0].rstrip() if m else ''))
except Exception, err:
self._error(level.Content,
err.msg if hasattr(err, 'msg') else unicode(err),
m)
field = (m.name, valuator[1])
if m.name == start:
if isinstance(field[1], ErrorLevel):
self._error(err, err.msg, m, m.name)
field = (m.name, '')
db.append(sfm.element(field[1], m.pos, content=[field]))
else:
db[-1].append(field)
return db
def __iter__(self):
start,fields = self.__schema
proto = dict((k,dv) for k,(_,dv) in fields.iteritems())
default_field = (lambda x:x, None)
def record(rec):
rec_ = proto.copy()
rec_.update(rec)
fn,err = next(ifilter(lambda i: isinstance(i[1], ErrorLevel),rec_.iteritems()),('',None))
if err:
self._error(err, err.msg, rec, rec.name, fn)
rec_[fn] = None
return rec_
def accum(db, m):
valuator = fields.get(m.name, default_field)
try:
field = (m.name, valuator[0](m[0].rstrip() if m else ''))
except Exception, err:
self._error(level.Content, err.msg if hasattr(err,'msg') else unicode(err), m)
field = (m.name, valuator[1])
if m.name == start:
if isinstance(field[1], ErrorLevel):
self._error(err, err.msg, m, m.name)
field = (m.name, '')
db.append(sfm.element(field[1], m.pos, content=[field]))
else:
db[-1].append(field)
return db
def _ChapterNumber_(self, chapter_marker):
tok = next(self._tokens)
chapter = self.numeric_re.match(tok)
if not chapter:
self._error(level.Content, 'missing chapter number after \\c',
chapter_marker)
chapter_marker.args = [u'\uFFFD']
else:
chapter_marker.args = [
unicode(tok[chapter.start(1):chapter.end(1)])
]
tok = tok[chapter.end():]
if tok and not self.sep_re.match(tok):
self._error(level.Content,
'missing space after chapter number \'{chapter}\'',
tok,
chapter=chapter_marker.args[0])
tok = tok.lstrip()
if tok:
if tok[0] == '\\':
self._tokens.put_back(tok)
else:
self._error(
level.Structure,
'text cannot follow chapter marker \'{0}\'',
tok,
chapter_marker,
)
chapter_marker.append(
sfm.element(None, meta=self.default_meta, content=[tok]))
tok = None
return self._default_(chapter_marker)
def elem(name, *content):
name, args = (name[0], list(name[1:])) if type(name) == tuple else (name,
[])
e = sfm.element(name,
args=args,
meta=usfm.default_stylesheet.get(name, {}))
e.extend(content)
return e
def accum(db, m):
valuator = fields.get(m.name, default_field)
try:
field = (m.name, valuator[0](m[0].rstrip() if m else ''))
except Exception as err:
self._error(level.Content,
err.msg if hasattr(err, 'msg') else unicode(err),
m)
field = (m.name, valuator[1])
if m.name == start:
if isinstance(field[1], ErrorLevel):
self._error(err.level, err.msg, m, m.name)
field = (m.name, '')
db.append(sfm.element(field[1], m.pos, content=[field]))
else:
db[-1].append(field)
return db
def _ChapterNumber_(self, chapter_marker):
tok = next(self._tokens)
chapter = self.numeric_re.match(tok)
if not chapter:
self._error(level.Content, 'missing chapter number after \\c',
chapter_marker)
chapter_marker.args = [u'\uFFFD']
else:
chapter_marker.args = [unicode(tok[chapter.start(1):chapter.end(1)])]
tok = tok[chapter.end():]
if tok and not self.sep_re.match(tok):
self._error(level.Content, 'missing space after chapter number \'{chapter}\'',
tok, chapter=chapter_marker.args[0])
tok = tok.lstrip()
if tok:
if tok[0] == '\\':
self._tokens.put_back(tok)
else:
self._error(level.Structure, 'text cannot follow chapter marker \'{0}\'', tok, chapter_marker, )
chapter_marker.append(sfm.element(None, meta=self.default_meta, content=[tok]))
tok = None
return self._default_(chapter_marker)
class parser(sfm.parser):
'''
>>> from pprint import pprint
>>> import warnings
>>> doc = """\\Marker toc1
... \\Name toc1 - File - Long Table of Contents Text
... \\OccursUnder h h1 h2 h3
... \\FontSize 12
... \\Bold"""
>>> with warnings.catch_warnings():
... warnings.simplefilter("ignore")
... pprint(list(parser(doc.splitlines(True), schema('Marker',{}))))
[{},
{u'Bold': '',
u'FontSize': text(u'12'),
u'Marker': text(u'toc1'),
u'Name': text(u'toc1 - File - Long Table of Contents Text'),
u'OccursUnder': text(u'h h1 h2 h3')}]
>>> demo_schema = schema('Marker',
... {'Marker' : (str, UnrecoverableError('Start of record marker: {0} missing')),
... 'Name' : (str, StructureError('Marker {0} defintion missing: {1}')),
... 'Description' : (str, ''),
... 'OccursUnder' : (unique(sequence(str)), [None]),
... 'FontSize' : (int, None),
... 'Bold' : (flag, False)})
>>> with warnings.catch_warnings():
... warnings.simplefilter("error")
... pprint(list(parser(doc.splitlines(True), demo_schema)))
[{},
{'Bold': True,
'Description': '',
'FontSize': 12,
'Marker': 'toc1',
'Name': 'toc1 - File - Long Table of Contents Text',
'OccursUnder': set(['h', 'h1', 'h2', 'h3'])}]
>>> with warnings.catch_warnings():
... warnings.simplefilter("error")
... pprint(list(parser("""
... \\Description this goes in the header since it's before the
... key marker Marker as does the following marker.
... \\FontSize 15
... \\Marker toc1
... \\Name toc1 - File - Long Table of Contents Text
... \\FontSize 12""".splitlines(True), demo_schema)))
[{u'Description': "this goes in the header since it's before the \\nkey marker Marker as does the following marker.",
u'FontSize': 15},
{'Bold': False,
'Description': '',
'FontSize': 12,
'Marker': 'toc1',
'Name': 'toc1 - File - Long Table of Contents Text',
'OccursUnder': [None]}]
>>> with warnings.catch_warnings():
... warnings.simplefilter("error")
... pprint(list(parser("""\Marker toc1
... \FontSize 12""".splitlines(True), demo_schema)))
Traceback (most recent call last):
...
SyntaxError: <string>: line 1,1: Marker toc1 defintion missing: Name
'''
def __init__(self, source, schema, error_level=level.Content):
if not isinstance(schema, _schema):
raise TypeError('arg 2 must a \'schema\' not {0!r}'.format(schema))
metas = dict(
(k, super(parser, self).default_meta) for k in schema.fields)
super(parser, self).__init__(source, metas, error_level=error_level)
self.__schema = schema
def __iter__(self):
start, fields = self.__schema
proto = dict((k, dv) for k, (_, dv) in fields.iteritems())
default_field = (lambda x: x, None)
def record(rec):
rec_ = proto.copy()
rec_.update(rec)
for fn, err in ifilter(lambda i: isinstance(i[1], ErrorLevel),
rec_.iteritems()):
if err:
self._error(err, err.msg, rec, rec.name, fn)
rec_[fn] = None
return rec_
def accum(db, m):
valuator = fields.get(m.name, default_field)
try:
field = (m.name, valuator[0](m[0].rstrip() if m else ''))
except Exception, err:
self._error(level.Content,
err.msg if hasattr(err, 'msg') else unicode(err),
m)
field = (m.name, valuator[1])
if m.name == start:
if isinstance(field[1], ErrorLevel):
self._error(err, err.msg, m, m.name)
field = (m.name, '')
db.append(sfm.element(field[1], m.pos, content=[field]))
else:
db[-1].append(field)
return db
es = super(parser, self).__iter__()
fs = ifilter(lambda v: isinstance(v, sfm.element), es)
fgs = reduce(accum, fs, [sfm.element('header')])
return chain((dict(fgs[0]), ), imap(record, fgs[1:]))
def elem(name, *content):
name, args = (name[0],list(name[1:])) if type(name) == tuple else (name, [])
e = sfm.element(name, args=args, meta=usfm.default_stylesheet.get(name,{}))
e.extend(content)
return e