Commit e9b10f1f authored by Eckhart Arnold's avatar Eckhart Arnold
Browse files

Refactorings: base.py removed in favor of erroy.py

parent 8229e7f3
......@@ -18,7 +18,7 @@ implied. See the License for the specific language governing
permissions and limitations under the License.
"""
from .base import *
from .error import *
from .dsl import *
from .ebnf import *
from .parser import *
......@@ -26,9 +26,10 @@ from .syntaxtree import *
# Flat namespace for the DHParser Package. Is this a good idea...?
from .toolkit import *
from .transform import *
# from .testing import *
from .testing import *
from .versionnumber import __version__
__author__ = "Eckhart Arnold <arnold@badw.de>"
__copyright__ = "http://www.apache.org/licenses/LICENSE-2.0"
# __all__ = ['toolkit', 'base', 'syntaxtree', 'parser', 'transform', 'ebnf', 'dsl', 'testing', 'versionnumber'] # flat namespace
# __all__ = ['toolkit', 'error', 'syntaxtree', 'parser', 'transform', 'ebnf', 'dsl', 'testing',
# 'versionnumber'] # flat namespace
"""base.py - various base classes that are used across several other
the DHParser-modules.
Copyright 2016 by Eckhart Arnold (arnold@badw.de)
Bavarian Academy of Sciences an Humanities (badw.de)
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
implied. See the License for the specific language governing
permissions and limitations under the License.
"""
import collections
from typing import Hashable, Iterable, Iterator, Optional, Tuple
__all__ = ('ParserBase',
'WHITESPACE_PTYPE',
'TOKEN_PTYPE',
'MockParser',
'ZombieParser',
'ZOMBIE_PARSER',
'Error',
'is_error',
'is_warning',
'has_errors',
'only_errors',
'StringView',
'EMPTY_STRING_VIEW')
#######################################################################
#
# parser base and mock parsers
#
#######################################################################
class ParserBase:
"""
ParserBase is the base class for all real and mock parser classes.
It is defined here, because Node objects require a parser object
for instantiation.
"""
def __init__(self, name=''): # , pbases=frozenset()):
self.name = name # type: str
self._ptype = ':' + self.__class__.__name__ # type: str
def __repr__(self):
return self.name + self.ptype
def __str__(self):
return self.name + (' = ' if self.name else '') + repr(self)
@property
def ptype(self) -> str:
return self._ptype
@property
def repr(self) -> str:
return self.name if self.name else repr(self)
WHITESPACE_PTYPE = ':Whitespace'
TOKEN_PTYPE = ':Token'
class MockParser(ParserBase):
"""
MockParser objects can be used to reconstruct syntax trees from a
serialized form like S-expressions or XML. Mock objects can mimic
different parser types by assigning them a ptype on initialization.
Mock objects should not be used for anything other than
syntax tree (re-)construction. In all other cases where a parser
object substitute is needed, chose the singleton ZOMBIE_PARSER.
"""
def __init__(self, name='', ptype=''): # , pbases=frozenset()):
assert not ptype or ptype[0] == ':'
super(MockParser, self).__init__(name)
self.name = name
self._ptype = ptype or ':' + self.__class__.__name__
class ZombieParser(MockParser):
"""
Serves as a substitute for a Parser instance.
``ZombieParser`` is the class of the singelton object
``ZOMBIE_PARSER``. The ``ZOMBIE_PARSER`` has a name and can be
called, but it never matches. It serves as a substitute where only
these (or one of these properties) is needed, but no real Parser-
object is instantiated.
"""
alive = False
def __init__(self):
super(ZombieParser, self).__init__("__ZOMBIE__")
assert not self.__class__.alive, "There can be only one!"
assert self.__class__ == ZombieParser, "No derivatives, please!"
self.__class__.alive = True
def __copy__(self):
return self
def __deepcopy__(self, memo):
return self
def __call__(self, text):
"""Better call Saul ;-)"""
return None, text
ZOMBIE_PARSER = ZombieParser()
#######################################################################
#
# error reporting
#
#######################################################################
class Error:
__slots__ = ['message', 'level', 'code', 'pos', 'line', 'column']
WARNING = 1
ERROR = 1000
HIGHEST = ERROR
def __init__(self, message: str, level: int=ERROR, code: Hashable=0):
self.message = message
assert level >= 0
self.level = level or Error.ERROR
self.code = code
self.pos = -1
self.line = -1
self.column = -1
def __str__(self):
prefix = ''
if self.line > 0:
prefix = "line: %3i, column: %2i, " % (self.line, self.column)
return prefix + "%s: %s" % (self.level_str, self.message)
@property
def level_str(self):
return "Warning" if is_warning(self.level) else "Error"
def is_warning(level: int) -> bool:
return level < Error.ERROR
def is_error(level: int) -> bool:
return level >= Error.ERROR
def has_errors(messages: Iterable[Error], level: int=Error.ERROR) -> bool:
"""
Returns True, if at least one entry in `messages` has at
least the given error `level`.
"""
for err_obj in messages:
if err_obj.level >= level:
return True
return False
def only_errors(messages: Iterable[Error], level: int=Error.ERROR) -> Iterator[Error]:
"""
Returns an Iterator that yields only those messages that have
at least the given error level.
"""
return (err for err in messages if err.level >= level)
#######################################################################
#
# string view
#
#######################################################################
class StringView(collections.abc.Sized):
""""A rudimentary StringView class, just enough for the use cases
in parser.py.
Slicing Python-strings always yields copies of a segment of the original
string. See: https://mail.python.org/pipermail/python-dev/2008-May/079699.html
However, this becomes costly (in terms of space and as a consequence also
time) when parsing longer documents. Unfortunately, Python's `memoryview`
does not work for unicode strings. Hence, the StringView class.
"""
__slots__ = ['text', 'begin', 'end', 'len', 'fullstring_flag']
def __init__(self, text: str, begin: Optional[int] = 0, end: Optional[int] = None) -> None:
self.text = text # type: str
self.begin = 0 # type: int
self.end = 0 # type: int
self.begin, self.end = StringView.real_indices(begin, end, len(text))
self.len = max(self.end - self.begin, 0)
self.fullstring_flag = (self.begin == 0 and self.len == len(self.text))
@staticmethod
def real_indices(begin, end, len):
def pack(index, len):
index = index if index >= 0 else index + len
return 0 if index < 0 else len if index > len else index
if begin is None: begin = 0
if end is None: end = len
return pack(begin, len), pack(end, len)
def __bool__(self):
return bool(self.text) and self.end > self.begin
def __len__(self):
return self.len
def __str__(self):
if self.fullstring_flag: # optimization: avoid slicing/copying
return self.text
return self.text[self.begin:self.end]
def __getitem__(self, index):
# assert isinstance(index, slice), "As of now, StringView only allows slicing."
# assert index.step is None or index.step == 1, \
# "Step sizes other than 1 are not yet supported by StringView"
start, stop = StringView.real_indices(index.start, index.stop, self.len)
return StringView(self.text, self.begin + start, self.begin + stop)
def __eq__(self, other):
return str(self) == str(other) # PERFORMANCE WARNING: This creates copies of the strings
def count(self, sub, start=None, end=None) -> int:
if self.fullstring_flag:
return self.text.count(sub, start, end)
elif start is None and end is None:
return self.text.count(sub, self.begin, self.end)
else:
start, end = StringView.real_indices(start, end, self.len)
return self.text.count(sub, self.begin + start, self.begin + end)
def find(self, sub, start=None, end=None) -> int:
if self.fullstring_flag:
return self.text.find(sub, start, end)
elif start is None and end is None:
return self.text.find(sub, self.begin, self.end) - self.begin
else:
start, end = StringView.real_indices(start, end, self.len)
return self.text.find(sub, self.begin + start, self.begin + end) - self.begin
def rfind(self, sub, start=None, end=None) -> int:
if self.fullstring_flag:
return self.text.rfind(sub, start, end)
if start is None and end is None:
return self.text.rfind(sub, self.begin, self.end) - self.begin
else:
start, end = StringView.real_indices(start, end, self.len)
return self.text.rfind(sub, self.begin + start, self.begin + end) - self.begin
def startswith(self, prefix: str, start:int = 0, end:Optional[int] = None) -> bool:
start += self.begin
end = self.end if end is None else self.begin + end
return self.text.startswith(prefix, start, end)
def match(self, regex):
return regex.match(self.text, pos=self.begin, endpos=self.end)
def index(self, absolute_index: int) -> int:
"""
Converts an index for a string watched by a StringView object
to an index relative to the string view object, e.g.:
>>> sv = StringView('xxIxx')[2:3]
>>> match = sv.match(re.compile('I'))
>>> match.end()
3
>>> sv.index(match.end())
1
"""
return absolute_index - self.begin
def indices(self, absolute_indices: Iterable[int]) -> Tuple[int, ...]:
"""Converts indices for a string watched by a StringView object
to indices relative to the string view object. See also: `sv_index()`
"""
return tuple(index - self.begin for index in absolute_indices)
def search(self, regex):
return regex.search(self.text, pos=self.begin, endpos=self.end)
EMPTY_STRING_VIEW = StringView('')
\ No newline at end of file
......@@ -36,7 +36,7 @@ from DHParser.ebnf import EBNFCompiler, grammar_changed, \
from DHParser.toolkit import logging, load_if_file, is_python_code, compile_python_object
from DHParser.parser import Grammar, Compiler, compile_source, nil_preprocessor, PreprocessorFunc
from DHParser.syntaxtree import Node, TransformationFunc
from DHParser.base import Error, is_error, has_errors, only_errors
from DHParser.error import Error, is_error, has_errors, only_errors
__all__ = ('GrammarError',
'CompilationError',
......
......@@ -33,8 +33,8 @@ from DHParser.toolkit import load_if_file, escape_re, md5, sane_parser_name
from DHParser.parser import Grammar, mixin_comment, nil_preprocessor, Forward, RE, NegativeLookahead, \
Alternative, Series, Option, Required, OneOrMore, ZeroOrMore, Token, Compiler, \
PreprocessorFunc
from DHParser.syntaxtree import Node, TransformationFunc
from DHParser.base import WHITESPACE_PTYPE, TOKEN_PTYPE, Error
from DHParser.syntaxtree import Node, TransformationFunc, WHITESPACE_PTYPE, TOKEN_PTYPE
from DHParser.error import Error
from DHParser.transform import TransformationDict, traverse, remove_brackets, \
reduce_single_child, replace_by_single_child, remove_expendables, \
remove_tokens, flatten, forbid, assert_content, remove_infix_operator
......
"""error.py - error handling for DHParser
Copyright 2016 by Eckhart Arnold (arnold@badw.de)
Bavarian Academy of Sciences an Humanities (badw.de)
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
implied. See the License for the specific language governing
permissions and limitations under the License.
"""
import bisect
import functools
from typing import Hashable, Iterable, Iterator, Union, Tuple, List
from DHParser.toolkit import StringView
__all__ = ('Error',
'is_error',
'is_warning',
'has_errors',
'only_errors',
'linebreaks',
'line_col',
'error_messages')
class Error:
__slots__ = ['message', 'level', 'code', 'pos', 'line', 'column']
WARNING = 1
ERROR = 1000
HIGHEST = ERROR
def __init__(self, message: str, level: int=ERROR, code: Hashable=0):
self.message = message
assert level >= 0
self.level = level or Error.ERROR
self.code = code
self.pos = -1
self.line = -1
self.column = -1
def __str__(self):
prefix = ''
if self.line > 0:
prefix = "line: %3i, column: %2i, " % (self.line, self.column)
return prefix + "%s: %s" % (self.level_str, self.message)
@property
def level_str(self):
return "Warning" if is_warning(self.level) else "Error"
def is_warning(level: int) -> bool:
return level < Error.ERROR
def is_error(level: int) -> bool:
return level >= Error.ERROR
def has_errors(messages: Iterable[Error], level: int=Error.ERROR) -> bool:
"""
Returns True, if at least one entry in `messages` has at
least the given error `level`.
"""
for err_obj in messages:
if err_obj.level >= level:
return True
return False
def only_errors(messages: Iterable[Error], level: int=Error.ERROR) -> Iterator[Error]:
"""
Returns an Iterator that yields only those messages that have
at least the given error level.
"""
return (err for err in messages if err.level >= level)
def linebreaks(text: Union[StringView, str]):
lb = [-1]
i = text.find('\n', 0)
while i >= 0:
lb.append(i)
i = text.find('\n', i + 1)
lb.append(len(text))
return lb
@functools.singledispatch
def line_col(text: Union[StringView, str], pos: int) -> Tuple[int, int]:
"""Returns the position within a text as (line, column)-tuple.
"""
if pos < 0 or pos > len(text): # one character behind EOF is still an allowed position!
raise ValueError('Position %i outside text of length %s !' % (pos, len(text)))
# assert pos <= len(text), str(pos) + " > " + str(len(text))
line = text.count("\n", 0, pos) + 1
column = pos - text.rfind("\n", 0, pos)
return line, column
@line_col.register(list)
def _line_col(lbreaks: List[int], pos: int) -> Tuple[int, int]:
"""Returns the position within a text as (line, column)-tuple based
on a list of all line breaks, including -1 and EOF.
"""
if pos < 0 or pos > lbreaks[-1]: # one character behind EOF is still an allowed position!
raise ValueError('Position %i outside text of length %s !' % (pos, lbreaks[-1]))
line = bisect.bisect_left(lbreaks, pos)
column = pos - lbreaks[line - 1]
return line, column
def error_messages(source_text, errors) -> List[str]:
"""Returns the sequence or iterator of error objects as an intertor
of error messages with line and column numbers at the beginning.
Args:
source_text (str): The source text on which the errors occurred.
(Needed in order to determine the line and column numbers.)
errors (list): The list of errors as returned by the method
``collect_errors()`` of a Node object
Returns:
a list that contains all error messages in string form. Each
string starts with "line: [Line-No], column: [Column-No]
"""
for err in errors:
if err.pos >= 0 and err.line <= 0:
err.line, err.column = line_col(source_text, err.pos)
return [str(err) for err in sorted(errors, key=lambda err: err.pos)]
......@@ -74,11 +74,12 @@ try:
except ImportError:
from .typing34 import Any, Callable, cast, Dict, Iterator, List, Set, Tuple, Union, Optional
from DHParser.toolkit import is_logging, log_dir, logfile_basename, escape_re, sane_parser_name
from DHParser.syntaxtree import Node, TransformationFunc
from DHParser.base import ParserBase, WHITESPACE_PTYPE, TOKEN_PTYPE, ZOMBIE_PARSER, Error, is_error, has_errors, \
from DHParser.toolkit import is_logging, log_dir, logfile_basename, escape_re, sane_parser_name, \
StringView, EMPTY_STRING_VIEW
from DHParser.toolkit import load_if_file, error_messages, line_col
from DHParser.syntaxtree import Node, TransformationFunc, ParserBase, WHITESPACE_PTYPE, TOKEN_PTYPE, \
ZOMBIE_PARSER
from DHParser.error import Error, is_error, has_errors, line_col, error_messages
from DHParser.toolkit import load_if_file
__all__ = ('PreprocessorFunc',
'HistoryRecord',
......
......@@ -32,14 +32,112 @@ except ImportError:
from .typing34 import AbstractSet, Any, ByteString, Callable, cast, Container, Dict, \
Iterator, Iterable, List, NamedTuple, Sequence, Union, Text, Tuple, Hashable
from DHParser.toolkit import is_logging, log_dir, linebreaks, line_col, identity
from DHParser.base import MockParser, ZOMBIE_PARSER, Error, StringView
__all__ = ('Node',
from DHParser.toolkit import is_logging, log_dir, identity, StringView
from DHParser.error import Error, linebreaks, line_col
__all__ = ('ParserBase',
'WHITESPACE_PTYPE',
'TOKEN_PTYPE',
'MockParser',
'ZombieParser',
'ZOMBIE_PARSER',
'Node',
'mock_syntax_tree',
'TransformationFunc')
#######################################################################
#
# parser base and mock parsers
#
#######################################################################
class ParserBase:
"""
ParserBase is the base class for all real and mock parser classes.
It is defined here, because Node objects require a parser object
for instantiation.
"""
def __init__(self, name=''): # , pbases=frozenset()):
self.name = name # type: str
self._ptype = ':' + self.__class__.__name__ # type: str
def __repr__(self):
return self.name + self.ptype
def __str__(self):
return self.name + (' = ' if self.name else '') + repr(self)
@property
def ptype(self) -> str:
return self._ptype
@property
def repr(self) -> str:
return self.name if self.name else repr(self)
WHITESPACE_PTYPE = ':Whitespace'
TOKEN_PTYPE = ':Token'
class MockParser(ParserBase):
"""
MockParser objects can be used to reconstruct syntax trees from a
serialized form like S-expressions or XML. Mock objects can mimic
different parser types by assigning them a ptype on initialization.
Mock objects should not be used for anything other than
syntax tree (re-)construction. In all other cases where a parser
object substitute is needed, chose the singleton ZOMBIE_PARSER.
"""
def __init__(self, name='', ptype=''): # , pbases=frozenset()):
assert not ptype or ptype[0] == ':'
super(MockParser, self).__init__(name)
self.name = name
self._ptype = ptype or ':' + self.__class__.__name__
class ZombieParser(MockParser):
"""
Serves as a substitute for a Parser instance.
``ZombieParser`` is the class of the singelton object
``ZOMBIE_PARSER``. The ``ZOMBIE_PARSER`` has a name and can be
called, but it never matches. It serves as a substitute where only
these (or one of these properties) is needed, but no real Parser-
object is instantiated.
"""
alive = False
def __init__(self):
super(ZombieParser, self).__init__("__ZOMBIE__")
assert not self.__class__.alive, "There can be only one!"
assert self.__class__ == ZombieParser, "No derivatives, please!"
self.__class__.alive = True
def __copy__(self):
return self
def __deepcopy__(self, memo):
return self
def __call__(self, text):
"""Better call Saul ;-)"""
return None, text
ZOMBIE_PARSER = ZombieParser()
#######################################################################
#
# syntaxtree nodes
#