Skip to content
GitLab
Commit fde7e064 authored by eckhart's avatar eckhart
Browse files

- some refinements

parent f6c14f6c
Hide whitespace changes
Inline Side-by-side
DHParser/ebnf.py
View file @ fde7e064
......@@ -581,7 +581,7 @@ class EBNFCompiler(Compiler):
defined_symbols = set(self.rules.keys()) | self.RESERVED_SYMBOLS
for symbol in self.symbols:
if symbol not in defined_symbols:
self.tree.add_error(self.symbols[symbol],
self.tree.new_error(self.symbols[symbol],
"Missing definition for symbol '%s'" % symbol)
# root_node.error_flag = True
......@@ -599,9 +599,9 @@ class EBNFCompiler(Compiler):
remove_connections(self.root_symbol)
for leftover in defined_symbols:
self.tree.add_error(self.rules[leftover][0],
('Rule "%s" is not connected to parser root "%s" !') %
(leftover, self.root_symbol), Error.WARNING)
self.tree.new_error(self.rules[leftover][0],
('Rule "%s" is not connected to parser root "%s" !') %
(leftover, self.root_symbol), Error.WARNING)
# set root_symbol parser and assemble python grammar definition
......@@ -640,19 +640,19 @@ class EBNFCompiler(Compiler):
if rule in self.rules:
first = self.rules[rule][0]
if not first.errors:
self.tree.add_error(first, 'First definition of rule "%s" '
self.tree.new_error(first, 'First definition of rule "%s" '
'followed by illegal redefinitions.' % rule)
self.tree.add_error(node, 'A rule "%s" has already been defined earlier.' % rule)
self.tree.new_error(node, 'A rule "%s" has already been defined earlier.' % rule)
elif rule in EBNFCompiler.RESERVED_SYMBOLS:
self.tree.add_error(node, 'Symbol "%s" is a reserved symbol.' % rule)
self.tree.new_error(node, 'Symbol "%s" is a reserved symbol.' % rule)
elif not sane_parser_name(rule):
self.tree.add_error(node, 'Illegal symbol "%s". Symbols must not start or '
self.tree.new_error(node, 'Illegal symbol "%s". Symbols must not start or '
' end with a doube underscore "__".' % rule)
elif rule in self.directives['tokens']:
self.tree.add_error(node, 'Symbol "%s" has already been defined as '
self.tree.new_error(node, 'Symbol "%s" has already been defined as '
'a preprocessor token.' % rule)
elif keyword.iskeyword(rule):
self.tree.add_error(node, 'Python keyword "%s" may not be used as a symbol. '
self.tree.new_error(node, 'Python keyword "%s" may not be used as a symbol. '
% rule + '(This may change in the future.)')
try:
self.current_symbols = [node]
......@@ -669,7 +669,7 @@ class EBNFCompiler(Compiler):
trace = str(extract_tb(error.__traceback__)[-1])
errmsg = "%s (TypeError: %s; %s)\n%s" \
% (EBNFCompiler.AST_ERROR, str(error), trace, node.as_sxpr())
self.tree.add_error(node, errmsg)
self.tree.new_error(node, errmsg)
rule, defn = rule + ':error', '"' + errmsg + '"'
return rule, defn
......@@ -685,7 +685,7 @@ class EBNFCompiler(Compiler):
try:
re.compile(rx)
except Exception as re_error:
self.tree.add_error(node, "malformed regular expression %s: %s" %
self.tree.new_error(node, "malformed regular expression %s: %s" %
(repr(rx), str(re_error)))
return rx
......@@ -696,7 +696,7 @@ class EBNFCompiler(Compiler):
if key not in self.REPEATABLE_DIRECTIVES:
if key in self.defined_directives:
self.tree.add_error(node, 'Directive "%s" has already been defined earlier. '
self.tree.new_error(node, 'Directive "%s" has already been defined earlier. '
% key + 'Later definition will be ignored!',
code=Error.REDEFINED_DIRECTIVE_WARNING)
return ""
......@@ -705,26 +705,26 @@ class EBNFCompiler(Compiler):
if key in {'comment', 'whitespace'}:
if node.children[1].parser.name == "list_":
if len(node.children[1].result) != 1:
self.tree.add_error(node, 'Directive "%s" must have one, but not %i values.'
self.tree.new_error(node, 'Directive "%s" must have one, but not %i values.'
% (key, len(node.children[1].result)))
value = self.compile(node.children[1]).pop()
if key == 'whitespace' and value in EBNFCompiler.WHITESPACE:
value = EBNFCompiler.WHITESPACE[value] # replace whitespace-name by regex
else:
self.tree.add_error(node, 'Value "%s" not allowed for directive "%s".'
self.tree.new_error(node, 'Value "%s" not allowed for directive "%s".'
% (value, key))
else:
value = node.children[1].content.strip("~") # cast(str, node.children[
# 1].result).strip("~")
if value != node.children[1].content: # cast(str, node.children[1].result):
self.tree.add_error(node, "Whitespace marker '~' not allowed in definition "
self.tree.new_error(node, "Whitespace marker '~' not allowed in definition "
"of %s regular expression." % key)
if value[0] + value[-1] in {'""', "''"}:
value = escape_re(value[1:-1])
elif value[0] + value[-1] == '//':
value = self._check_rx(node, value[1:-1])
if key == 'whitespace' and not re.match(value, ''):
self.tree.add_error(node, "Implicit whitespace should always "
self.tree.new_error(node, "Implicit whitespace should always "
"match the empty string, /%s/ does not." % value)
self.directives[key] = value
......@@ -740,7 +740,7 @@ class EBNFCompiler(Compiler):
value = {item.lower() for item in self.compile(node.children[1])}
if ((value - {'left', 'right', 'both', 'none'})
or ('none' in value and len(value) > 1)):
self.tree.add_error(node, 'Directive "literalws" allows only `left`, `right`, '
self.tree.new_error(node, 'Directive "literalws" allows only `left`, `right`, '
'`both` or `none`, not `%s`' % ", ".join(value))
wsp = {'left', 'right'} if 'both' in value \
else {} if 'none' in value else value
......@@ -750,7 +750,7 @@ class EBNFCompiler(Compiler):
tokens = self.compile(node.children[1])
redeclared = self.directives['tokens'] & tokens
if redeclared:
self.tree.add_error(node, 'Tokens %s have already been declared earlier. '
self.tree.new_error(node, 'Tokens %s have already been declared earlier. '
% str(redeclared) + 'Later declaration will be ignored',
code=Error.REDECLARED_TOKEN_WARNING)
self.directives['tokens'] |= tokens - redeclared
......@@ -758,12 +758,12 @@ class EBNFCompiler(Compiler):
elif key.endswith('_filter'):
filter_set = self.compile(node.children[1])
if not isinstance(filter_set, set) or len(filter_set) != 1:
self.tree.add_error(node, 'Directive "%s" accepts exactly on symbol, not %s'
self.tree.new_error(node, 'Directive "%s" accepts exactly on symbol, not %s'
% (key, str(filter_set)))
self.directives['filter'][key[:-7]] = filter_set.pop()
else:
self.tree.add_error(node, 'Unknown directive %s ! (Known ones are %s .)' %
self.tree.new_error(node, 'Unknown directive %s ! (Known ones are %s .)' %
(key, ', '.join(list(self.directives.keys()))))
return ""
......@@ -794,10 +794,10 @@ class EBNFCompiler(Compiler):
if nd.parser.ptype == TOKEN_PTYPE and nd.content == "§":
mandatory_marker.append(len(filtered_children))
# if len(filtered_children) == 0:
# self.tree.add_error(nd.pos, 'First item of a series should not be mandatory.',
# self.tree.new_error(nd.pos, 'First item of a series should not be mandatory.',
# Error.WARNING)
if len(mandatory_marker) > 1:
self.tree.add_error(nd, 'One mandatory marker (§) sufficient to declare '
self.tree.new_error(nd, 'One mandatory marker (§) sufficient to declare '
'the rest of the series as mandatory.', Error.WARNING)
else:
filtered_children.append(nd)
......@@ -822,7 +822,7 @@ class EBNFCompiler(Compiler):
assert len(node.children) == 2
arg = node.children[-1]
if arg.parser.name != 'symbol':
self.tree.add_error(node, ('Retrieve Operator "%s" requires a symbol, '
self.tree.new_error(node, ('Retrieve Operator "%s" requires a symbol, '
'and not a %s.') % (prefix, str(arg.parser)))
return str(arg.result)
if str(arg) in self.directives['filter']:
......@@ -856,14 +856,14 @@ class EBNFCompiler(Compiler):
break
if (nd.parser.name != "regexp" or nd.content[:1] != '/'
or nd.content[-1:] != '/'):
self.tree.add_error(node, "Lookbehind-parser can only be used with RegExp"
self.tree.new_error(node, "Lookbehind-parser can only be used with RegExp"
"-parsers, not: " + nd.parser.name + nd.parser.ptype)
if not result.startswith('RegExp('):
self.deferred_tasks.append(lambda: check(node))
return result
except KeyError:
self.tree.add_error(node, 'Unknown prefix "%s".' % prefix)
self.tree.new_error(node, 'Unknown prefix "%s".' % prefix)
return ""
......@@ -889,14 +889,14 @@ class EBNFCompiler(Compiler):
nd = node.children[0]
for child in nd.children:
if child.parser.ptype == TOKEN_PTYPE and nd.content == "§":
self.tree.add_error(node, "No mandatory items § allowed in Unordered sequences.")
self.tree.new_error(node, "No mandatory items § allowed in Unordered sequences.")
args = ', '.join(self.compile(child) for child in nd.children)
if nd.parser.name == "term":
return "AllOf(" + args + ")"
elif nd.parser.name == "expression":
return "SomeOf(" + args + ")"
else:
self.tree.add_error(node, "Unordered sequence or alternative "
self.tree.new_error(node, "Unordered sequence or alternative "
"requires at least two elements.")
return ""
......@@ -951,7 +951,7 @@ class EBNFCompiler(Compiler):
trace = str(extract_tb(error.__traceback__)[-1])
errmsg = "%s (AttributeError: %s; %s)\n%s" \
% (EBNFCompiler.AST_ERROR, str(error), trace, node.as_sxpr())
self.tree.add_error(node, errmsg)
self.tree.new_error(node, errmsg)
return '"' + errmsg + '"'
return parser + ', '.join([arg] + name) + ')'
......
DHParser/parse.py
View file @ fde7e064
......@@ -166,7 +166,7 @@ def add_parser_guard(parser_func):
except RecursionError:
node = Node(None, str(text[:min(10, max(1, text.find("\n")))]) + " ...")
node._pos = location
grammar.tree__.add_error(node, "maximum recursion depth of parser reached; "
grammar.tree__.new_error(node, "maximum recursion depth of parser reached; "
"potentially due to too many errors!")
rest = EMPTY_STRING_VIEW
......@@ -728,7 +728,7 @@ class Grammar:
result, _ = parser(rest)
if result is None:
result = Node(None, '').init_pos(0)
self.tree__.add_error(result,
self.tree__.new_error(result,
'Parser "%s" did not match empty document.' % str(parser))
while rest and len(stitches) < MAX_DROPOUTS:
result, rest = parser(rest)
......@@ -749,7 +749,7 @@ class Grammar:
if len(stitches) < MAX_DROPOUTS
else " too often! Terminating parser.")
stitches.append(Node(None, skip).init_pos(tail_pos(stitches)))
self.tree__.add_error(stitches[-1], error_msg)
self.tree__.new_error(stitches[-1], error_msg)
if self.history_tracking__:
# # some parsers may have matched and left history records with nodes != None.
# # Because these are not connected to the stitched root node, their pos-
......@@ -775,10 +775,10 @@ class Grammar:
# of the error will be the end of the text. Otherwise, the error
# message above ("...after end of parsing") would appear illogical.
error_node = Node(ZOMBIE_PARSER, '').init_pos(tail_pos(result.children))
self.tree__.add_error(error_node, error_str)
self.tree__.new_error(error_node, error_str)
result.result = result.children + (error_node,)
else:
self.tree__.add_error(result, error_str)
self.tree__.new_error(result, error_str)
# result.pos = 0 # calculate all positions
# result.collect_errors(self.document__)
self.tree__.swallow(result)
......@@ -862,19 +862,19 @@ class PreprocessorToken(Parser):
end = text.find(END_TOKEN, 1)
if end < 0:
node = Node(self, '')
self.grammar.tree__.add_error(node,
self.grammar.tree__.new_error(node,
'END_TOKEN delimiter missing from preprocessor token. '
'(Most likely due to a preprocessor bug!)') # type: Node
return node, text[1:]
elif end == 0:
node = Node(self, '')
self.grammar.tree__.add_error(node,
self.grammar.tree__.new_error(node,
'Preprocessor-token cannot have zero length. '
'(Most likely due to a preprocessor bug!)')
return node, text[2:]
elif text.find(BEGIN_TOKEN, 1, end) >= 0:
node = Node(self, text[len(self.name) + 1:end])
self.grammar.tree__.add_error(node,
self.grammar.tree__.new_error(node,
'Preprocessor-tokens must not be nested or contain '
'BEGIN_TOKEN delimiter as part of their argument. '
'(Most likely due to a preprocessor bug!)')
......@@ -1257,13 +1257,13 @@ class ZeroOrMore(Option):
if not node:
break
if len(text) == n:
err_msg = dsl_error_msg(self, 'Infinite Loop encountered.')
node.errors.append(Error(err_msg, Error.MESSAGE, node=node))
infinite_loop_error = Error(err_msg, node=node)
node.errors.append(Error("Infinite loop!", Error.MESSAGE, node=node))
infinite_loop_error = Error(dsl_error_msg(self, 'Infinite Loop encountered.'),
node=node)
results += (node,)
node = Node(self, results)
if infinite_loop_error:
self.grammar.tree__.add_error_obj(node, infinite_loop_error)
self.grammar.tree__.add_error(node, infinite_loop_error)
return node, text
def __repr__(self):
......@@ -1307,15 +1307,15 @@ class OneOrMore(UnaryOperator):
if not node:
break
if len(text_) == n:
err_msg = dsl_error_msg(self, 'Infinite Loop encountered.')
node.errors.append(Error(err_msg, Error.MESSAGE, node=node))
infinite_loop_error = Error(err_msg, node=node)
node.errors.append(Error("Infinite loop!", Error.MESSAGE, node=node))
infinite_loop_error = Error(dsl_error_msg(self, 'Infinite Loop encountered.'),
node=node)
results += (node,)
if results == ():
return None, text
node = Node(self, results)
if infinite_loop_error:
self.grammar.tree__.add_error_obj(node, infinite_loop_error)
self.grammar.tree__.add_error(node, infinite_loop_error)
return node, text_
def __repr__(self):
......@@ -1387,7 +1387,7 @@ class Series(NaryOperator):
assert len(results) <= len(self.parsers)
node = Node(self, results)
if mandatory_violation:
self.grammar.tree__.add_error_obj(node, mandatory_violation)
self.grammar.tree__.add_error(node, mandatory_violation)
return node, text_
def __repr__(self):
......@@ -1648,7 +1648,7 @@ def Required(parser: Parser) -> Parser:
# i = max(1, text.index(m.regs[1][0])) if m else 1
# node = Node(self, text[:i])
# text_ = text[i:]
# self.grammar.tree__.add_error(node,
# self.grammar.tree__.new_error(node,
# '%s expected; "%s" found!' % (str(self.parser),
# text[:10]), code=Error.MANDATORY_CONTINUATION)
# return node, text_
......@@ -1824,7 +1824,7 @@ class Retrieve(Parser):
value = self.filter(stack)
except (KeyError, IndexError):
node = Node(self, '')
self.grammar.tree__.add_error(
self.grammar.tree__.new_error(
node, dsl_error_msg(self, "'%s' undefined or exhausted." % self.symbol.name))
return node, text
if text.startswith(value):
......
DHParser/syntaxtree.py
View file @ fde7e064
......@@ -717,26 +717,28 @@ class RootNode(Node):
self._content = node._content
return self
def add_error_obj(self, node: Node, error: Error) -> 'RootNode':
def add_error(self, node: Node, error: Error) -> 'RootNode':
"""Adds an Error object to the tree, locating it at a specific node."""
self.all_errors.append(error)
self.error_flag = max(self.error_flag, error.code)
node.errors.append(error)
self.err_nodes_keep.append(node)
return self
def add_error(self,
def new_error(self,
node: Node,
message: str,
code: int = Error.ERROR) -> 'RootNode':
"""
Adds an error to this tree.
Adds an error to this tree, locating it at a specific node.
Parameters:
pos(int): The position of the error in the source text
message(str): A string with the error message.abs
code(int): An error code to identify the kind of error
"""
error = Error(message, code, node=node)
self.add_error_obj(node, error)
self.add_error(node, error)
return self
def collect_errors(self) -> List[Error]:
......@@ -748,7 +750,7 @@ class RootNode(Node):
self.all_errors.sort(key=lambda e: e.pos)
for node in self.err_nodes_keep: # redundant: consider removing Error.Error._node_keep
for error in node.errors:
# assert error._pos < 0 or node.pos <= error._pos <= node.len
assert error._pos < 0 or node.pos <= error._pos <= node.pos + len(node)
error._pos = node.pos
self.err_nodes_keep = []
errors = self.all_errors
......
DHParser/testing.py
View file @ fde7e064
......@@ -355,7 +355,7 @@ def grammar_unit(test_unit, parser_factory, transformer_factory, report=True, ve
cst = parser(test_code, parser_name)
except UnknownParserError as upe:
node = Node(ZOMBIE_PARSER, "").init_pos(0)
cst = RootNode().swallow(node).add_error(node, str(upe))
cst = RootNode(node).new_error(node, str(upe))
if not is_error(cst.error_flag):
errata.append('Fail test "%s" for parser "%s" yields match instead of '
'expected failure!' % (test_name, parser_name))
......
DHParser/transform.py
View file @ fde7e064
......@@ -888,7 +888,7 @@ assert_has_children = error_on(lambda nd: nd.children, 'Element "%s" has no chil
def assert_content(context: List[Node], regexp: str):
node = context[-1]
if not has_content(context, regexp):
context[0].add_error(node, 'Element "%s" violates %s on %s' %
context[0].new_error(node, 'Element "%s" violates %s on %s' %
(node.parser.name, str(regexp), node.content))
......@@ -897,7 +897,7 @@ def require(context: List[Node], child_tags: AbstractSet[str]):
node = context[-1]
for child in node.children:
if child.tag_name not in child_tags:
context[0].add_error(node, 'Element "%s" is not allowed inside "%s".' %
context[0].new_error(node, 'Element "%s" is not allowed inside "%s".' %
(child.parser.name, node.parser.name))
......@@ -906,6 +906,6 @@ def forbid(context: List[Node], child_tags: AbstractSet[str]):
node = context[-1]
for child in node.children:
if child.tag_name in child_tags:
context[0].add_error(node, 'Element "%s" cannot be nested inside "%s".' %
context[0].new_error(node, 'Element "%s" cannot be nested inside "%s".' %
(child.parser.name, node.parser.name))
documentation/ModuleReference.html
View file @ fde7e064
......@@ -526,7 +526,7 @@ S-Expression-output.</p>
<dl class="method">
<dt id="syntaxtree.Node.add_error">
<code class="descname">add_error</code><span class="sig-paren">(</span><em>message: str</em>, <em>code: int = 1000</em><span class="sig-paren">)</span> &#x2192; syntaxtree.Node<a class="reference internal" href="_modules/syntaxtree.html#Node.add_error"><span class="viewcode-link">[source]</span></a><a class="headerlink" href="#syntaxtree.Node.add_error" title="Permalink to this definition"></a></dt>
<code class="descname">add_error</code><span class="sig-paren">(</span><em>message: str</em>, <em>code: int = 1000</em><span class="sig-paren">)</span> &#x2192; syntaxtree.Node<a class="reference internal" href="_modules/syntaxtree.html#Node.new_error"><span class="viewcode-link">[source]</span></a><a class="headerlink" href="#syntaxtree.Node.new_error" title="Permalink to this definition"></a></dt>
<dd><p>Adds an error to this Node.
:param message: A string with the error message.abs
:type message: str
......
documentation/_modules/syntaxtree.html
View file @ fde7e064
......@@ -633,7 +633,7 @@
<span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_errors</span><span class="o">.</span><span class="n">copy</span><span class="p">()</span>
<div class="viewcode-block" id="Node.add_error"><a class="viewcode-back" href="../ModuleReference.html#syntaxtree.Node.add_error">[docs]</a> <span class="k">def</span> <span class="nf">add_error</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span>
<div class="viewcode-block" id="Node.add_error"><a class="viewcode-back" href="../ModuleReference.html#syntaxtree.Node.new_error">[docs]</a> <span class="k">def</span> <span class="nf">new_error</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span>
<span class="n">message</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
<span class="n">code</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="n">Error</span><span class="o">.</span><span class="n">ERROR</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="s1">&#39;Node&#39;</span><span class="p">:</span>
<span class="sd">&quot;&quot;&quot;</span>
......
documentation/genindex.html
View file @ fde7e064
......@@ -199,7 +199,7 @@
<h2 id="A">A</h2>
<table style="width: 100%" class="indextable genindextable"><tr>
<td style="width: 33%; vertical-align: top;"><ul>
<li><a href="ModuleReference.html#syntaxtree.Node.add_error">add_error() (Node method)</a>
<li><a href="ModuleReference.html#syntaxtree.Node.add_error">new_error() (Node method)</a>
</li>
<li><a href="ModuleReference.html#error.adjust_error_locations">adjust_error_locations() (in module error)</a>
</li>
......
examples/LaTeX/LaTeXCompiler.py
View file @ fde7e064
......@@ -690,7 +690,7 @@ class LaTeXCompiler(Compiler):
if 'language' in node.attributes:
self.metadata['language'] = it
else:
self.tree.add_error(node, 'Only one document language supported. '
self.tree.new_error(node, 'Only one document language supported. '
'Using %s, ignoring %s.'
% (self.metadata['language'], it), Error.WARNING)
if node['text'] in self.KNOWN_DOCUMENT_CLASSES:
......
test/test_compile.py
View file @ fde7e064
......@@ -30,8 +30,8 @@ from DHParser import parse_sxpr, Compiler
# B = next(tree.select(lambda node: str(node) == "1"))
# D = next(tree.select(lambda node: node.parser.name == "D"))
# F = next(tree.select(lambda node: str(node) == "3"))
# B.add_error("Error in child node")
# F.add_error("Error in child's child node")
# B.new_error("Error in child node")
# F.new_error("Error in child's child node")
# Compiler.propagate_error_flags(tree, lazy=True)
# assert A.error_flag
# assert not D.error_flag
......
test/test_syntaxtree.py
View file @ fde7e064
......@@ -133,8 +133,8 @@ class TestRootNode:
tree = parse_sxpr('(A (B D) (C E))')
tree.init_pos(0)
root = RootNode()
root.add_error(tree.children[1], "error C")
root.add_error(tree.children[0], "error B")
root.new_error(tree.children[1], "error C")
root.new_error(tree.children[0], "error B")
root.swallow(tree)
assert root.error_flag
errors = root.collect_errors()
......@@ -152,7 +152,7 @@ class TestRootNode:
# def find_h(context):
# node = context[-1]
# if node.result == "h":
# node.add_error("an error deep inside the syntax tree")
# node.new_error("an error deep inside the syntax tree")
#
# assert not tree.error_flag
# traverse(tree, {"*": find_h})
......@@ -164,8 +164,8 @@ class TestRootNode:
# B = next(tree.select(lambda node: str(node) == "1"))
# D = next(tree.select(lambda node: node.parser.name == "D"))
# F = next(tree.select(lambda node: str(node) == "3"))
# B.add_error("Error in child node")
# F.add_error("Error in child's child node")
# B.new_error("Error in child node")
# F.new_error("Error in child's child node")
# tree.error_flag = Error.ERROR
# errors = tree.collect_errors()
# assert len(errors) == 2, str(errors)
......
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

LRZ Homepage | Datenschutz | Dokumentation und Betriebsbedingungen | Impressum