Commit f7fbe190 authored by Eckhart Arnold's avatar Eckhart Arnold
Browse files

small corrections

parent 5b1b56c4
......@@ -17,10 +17,30 @@
"""
Module ``ebnf`` provides a self-hosting parser for EBNF-Grammars as
well as an EBNF-compiler that compiles an EBNF-Grammar into a
DHParser based Grammar class that can be executed to parse source text
Module ``ebnf`` provides an EBNF-parser-generator that compiles an
EBNF-Grammar into avPython-code that can be executed to parse source text
conforming to this grammar into concrete syntax trees.
With DHParser, Grammars can be specified either directly in Python-code
(see :py:mod:`DHParser.parse`) or in one of several EBNF-dialects. (Yes,
DHParser supports several different variants of EBNF! This makes it easy
to crate a parser directly from Grammars found in external sources.)
"EBNF" stands for the "Extended-Backus-Naur-Form" which is a common
formalism for specifying Grammars for context-free-languages.
(see https://en.wikipedia.org/wiki/Extended_Backus%E2%80%93Naur_form)
The recommended way of compiling grammars with DHParser is to either
write the EBNF-specification for that Grammar into a text-file and then
compile EBNF-source to an executable as well as importable Python-module
with the help of the "dhparser"-skript. Or, for bigger projects, to
create a new domain-specific-language-project with the DHParser-skript
as described in the step-by-step-guide.
However, here we will show how to compile an EBNF-specified grammar
from within Python-code and how to execute the parser that was
generated by compiling the grammar.
"""
......@@ -42,7 +62,7 @@ from DHParser.parse import Parser, Grammar, mixin_comment, mixin_nonempty, Forwa
Text, Capture, Retrieve, Pop, optional_last_value, GrammarError, Whitespace, Always, Never, \
INFINITE, matching_bracket, ParseFunc, update_scanner
from DHParser.preprocess import nil_preprocessor, PreprocessorFunc
from DHParser.syntaxtree import Node, RootNode, WHITESPACE_PTYPE, TOKEN_PTYPE, EMPTY_NODE
from DHParser.syntaxtree import Node, RootNode, WHITESPACE_PTYPE, TOKEN_PTYPE
from DHParser.toolkit import load_if_file, escape_re, escape_control_characters, md5, \
sane_parser_name, re, expand_table, unrepr, compile_python_object, DHPARSER_PARENTDIR, \
RX_NEVER_MATCH, cython
......@@ -492,7 +512,7 @@ class EBNFGrammar(Grammar):
set_parsefunc(self.free_char, self.free_char_parsefunc__)
set_parsefunc(self.regex_heuristics, never)
set_parsefunc(self.char_range_heuristics, always)
elif mode == 'regex-like':
elif mode == 'regex-like':
set_parsefunc(self.free_char, self.free_char_parsefunc__)
set_parsefunc(self.regex_heuristics, always)
set_parsefunc(self.char_range_heuristics, always)
......@@ -978,10 +998,10 @@ WHITESPACE_TYPES = {'horizontal': r'[\t ]*', # default: horizontal
'linefeed': r'[ \t]*\n?(?!\s*\n)[ \t]*',
'vertical': r'\s*'}
DROP_STRINGS = 'strings'
DROP_WSPC = 'whitespace'
DROP_REGEXP = 'regexps'
DROP_VALUES = {DROP_STRINGS, DROP_WSPC, DROP_REGEXP}
DROP_STRINGS = 'strings'
DROP_WSPC = 'whitespace'
DROP_REGEXP = 'regexps'
DROP_VALUES = {DROP_STRINGS, DROP_WSPC, DROP_REGEXP}
# Representation of Python code or, rather, something that will be output as Python code
ReprType = Union[str, unrepr]
......@@ -1495,7 +1515,7 @@ class EBNFCompiler(Compiler):
"""Returns the recursive paths from symbol to itself. If
sym is not recursive, the returned tuple (of paths) will be empty.
This method exists only for debugging (so far...)."""
path = []
path = [] # type: List[str]
recursive_paths = set() # type: Set[Tuple[str]]
def gather(sym: str):
......@@ -2342,7 +2362,7 @@ class EBNFCompiler(Compiler):
"-parsers, not: " + nd.tag_name)
if not result[:7] == 'RegExp(':
self.deferred_tasks.append(lambda :verify(node))
self.deferred_tasks.append(lambda: verify(node))
return result
......@@ -2365,8 +2385,8 @@ class EBNFCompiler(Compiler):
if self.anonymous_regexp.match(arg):
self.tree.new_error(
node, ('Retrieve operator "%s" does not work with anonymous parsers like %s')
% (prefix, arg))
node, 'Retrieve operator "%s" does not work with anonymous parsers like %s'
% (prefix, arg))
return arg
custom_args = [] # type: List[str]
......@@ -2424,9 +2444,9 @@ class EBNFCompiler(Compiler):
"""
assert node.tag_name == 'counted'
assert len(node.children) == 2
range = node.get('range', None)
if range:
r = self.extract_range(range)
rng = node.get('range', None)
if rng:
r = self.extract_range(rng)
what = node.children[0]
assert what.tag_name != 'range'
else: # multiplier specified instead of range
......
......@@ -36,7 +36,7 @@ from typing import Callable, cast, List, Tuple, Set, AbstractSet, Dict, \
DefaultDict, Sequence, Union, Optional, Iterator
from DHParser.configuration import get_config_value
from DHParser.error import Error, ErrorCode, is_error, MANDATORY_CONTINUATION, \
from DHParser.error import Error, ErrorCode, MANDATORY_CONTINUATION, \
UNDEFINED_RETRIEVE, PARSER_LOOKAHEAD_FAILURE_ONLY, \
PARSER_LOOKAHEAD_MATCH_ONLY, PARSER_STOPPED_BEFORE_END, PARSER_NEVER_TOUCHES_DOCUMENT, \
MALFORMED_ERROR_STRING, MANDATORY_CONTINUATION_AT_EOF, DUPLICATE_PARSERS_IN_ALTERNATIVE, \
......
Supports Markdown
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment