Chromium Code Reviews
chromiumcodereview-hr@appspot.gserviceaccount.com (chromiumcodereview-hr) | Please choose your nickname with Settings | Help | Chromium Project | Gerrit Changes | Sign out
(28)

Unified Diff: tools/lexer_generator/nfa.py

Issue 61893023: Experimental parser: split out NfaBuilder (Closed) Base URL: https://v8.googlecode.com/svn/branches/experimental/parser
Patch Set: Created 7 years, 1 month ago
Use n/p to move between diff chunks; N/P to move between comments. Draft comments are only viewable by you.
Jump to:
View side-by-side diff with in-line comments
Download patch
« no previous file with comments | « tools/lexer_generator/generator.py ('k') | tools/lexer_generator/nfa_builder.py » ('j') | no next file with comments »
Expand Comments ('e') | Collapse Comments ('c') | Show Comments Hide Comments ('s')
Index: tools/lexer_generator/nfa.py
diff --git a/tools/lexer_generator/nfa.py b/tools/lexer_generator/nfa.py
index cbc9a9f301334fa0803708be70a5acd1fb98bed9..bb60ef174876582811a43cb4d72f2f7d5da3385b 100644
--- a/tools/lexer_generator/nfa.py
+++ b/tools/lexer_generator/nfa.py
@@ -25,10 +25,8 @@
# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
-from types import TupleType
from transition_keys import TransitionKey
from automaton import *
-from inspect import getmembers
class NfaState(AutomatonState):
@@ -128,227 +126,6 @@ class NfaState(AutomatonState):
keys.discard(TransitionKey.epsilon())
return TransitionKey.disjoint_keys(keys)
-class NfaBuilder:
-
- def __init__(self):
- self.__node_number = 0
- self.__operation_map = {}
- self.__members = getmembers(self)
- self.__character_classes = {}
- self.__states = []
-
- def set_character_classes(self, classes):
- self.__character_classes = classes
-
- def __new_state(self):
- self.__node_number += 1
- return NfaState(self.__node_number - 1)
-
- def __or(self, graph):
- start = self.__new_state()
- ends = []
- for x in [self.__process(graph[1]), self.__process(graph[2])]:
- start.add_epsilon_transition(x[0])
- ends += x[1]
- start.close(None)
- return (start, ends)
-
- def __one_or_more(self, graph):
- (start, ends) = self.__process(graph[1])
- end = self.__new_state()
- end.add_epsilon_transition(start)
- self.__patch_ends(ends, end)
- return (start, [end])
-
- def __zero_or_more(self, graph):
- (node, ends) = self.__process(graph[1])
- start = self.__new_state()
- start.add_epsilon_transition(node)
- self.__patch_ends(ends, start)
- return (start, [start])
-
- def __zero_or_one(self, graph):
- (node, ends) = self.__process(graph[1])
- start = self.__new_state()
- start.add_epsilon_transition(node)
- return (start, ends + [start])
-
- def __repeat(self, graph):
- param_min = int(graph[1])
- param_max = int(graph[2])
- subgraph = graph[3]
- (start, ends) = self.__process(subgraph)
- for i in xrange(1, param_min):
- (start2, ends2) = self.__process(subgraph)
- self.__patch_ends(ends, start2)
- ends = ends2
- if param_min == param_max:
- return (start, ends)
-
- midpoints = []
- for i in xrange(param_min, param_max):
- midpoint = self.__new_state()
- self.__patch_ends(ends, midpoint)
- (start2, ends) = self.__process(subgraph)
- midpoint.add_epsilon_transition(start2)
- midpoints.append(midpoint)
-
- return (start, ends + midpoints)
-
- def __cat(self, graph):
- (left, right) = (self.__process(graph[1]), self.__process(graph[2]))
- self.__patch_ends(left[1], right[0])
- return (left[0], right[1])
-
- def __key_state(self, key):
- state = self.__new_state()
- state.add_unclosed_transition(key)
- return (state, [state])
-
- def __literal(self, graph):
- return self.__key_state(TransitionKey.single_char(graph[1]))
-
- def __class(self, graph):
- return self.__key_state(
- TransitionKey.character_class(graph, self.__character_classes))
-
- def __not_class(self, graph):
- return self.__key_state(
- TransitionKey.character_class(graph, self.__character_classes))
-
- def __any(self, graph):
- return self.__key_state(TransitionKey.any())
-
- def __epsilon(self, graph):
- start = self.__new_state()
- end = self.__new_state()
- start.close(end)
- return (start, [end])
-
- def __action(self, graph):
- (start, ends) = self.__process(graph[1])
- action = graph[2]
- end = self.__new_state()
- self.__patch_ends(ends, end)
- end.set_action(action)
- return (start, [end])
-
- def __continue(self, graph):
- (start, ends) = self.__process(graph[1])
- state = self.__peek_state()
- if not state['start_node']:
- state['start_node'] = self.__new_state()
- self.__patch_ends(ends, state['start_node'])
- return (start, [])
-
- def __catch_all(self, graph):
- return self.__key_state(TransitionKey.unique('catch_all'))
-
- def __join(self, graph):
- (graph, name, subgraph, modifier) = graph[1:]
- subgraphs = self.__peek_state()['subgraphs']
- if not name in subgraphs:
- subgraphs[name] = self.__nfa(subgraph)
- (subgraph_start, subgraph_end, nodes_in_subgraph) = subgraphs[name]
- (start, ends) = self.__process(graph)
- if modifier:
- assert modifier == 'ZERO_OR_MORE'
- for end in ends:
- end.add_epsilon_transition(subgraph_end)
- self.__patch_ends(ends, subgraph_start)
- end = self.__new_state()
- subgraph_end.add_epsilon_transition(end)
- return (start, [end])
-
- def __process(self, graph):
- assert type(graph) == TupleType
- method = "_NfaBuilder__" + graph[0].lower()
- if not method in self.__operation_map:
- matches = filter(lambda (name, func): name == method, self.__members)
- assert len(matches) == 1
- self.__operation_map[method] = matches[0][1]
- return self.__operation_map[method](graph)
-
- def __patch_ends(self, ends, new_end):
- for end in ends:
- end.close(new_end)
-
- def __push_state(self):
- self.__states.append({
- 'start_node' : None,
- 'subgraphs' : {},
- 'unpatched_ends' : [],
- })
-
- def __pop_state(self):
- return self.__states.pop()
-
- def __peek_state(self):
- return self.__states[len(self.__states) - 1]
-
- def __nfa(self, graph):
- start_node_number = self.__node_number
- self.__push_state()
- (start, ends) = self.__process(graph)
- state = self.__pop_state()
- if state['start_node']:
- state['start_node'].close(start)
- start = state['start_node']
- for k, subgraph in state['subgraphs'].items():
- subgraph[1].close(None)
- end = self.__new_state()
- if self.__states:
- self.__peek_state()['unpatched_ends'] += state['unpatched_ends']
- else:
- self.__patch_ends(state['unpatched_ends'], end)
- self.__patch_ends(ends, end)
- return (start, end, self.__node_number - start_node_number)
-
- def nfa(self, graph):
- (start, end, nodes_created) = self.__nfa(graph)
- end.close(None)
- return Nfa(start, end, nodes_created)
-
- @staticmethod
- def add_action(graph, action):
- return ('ACTION', graph, action)
-
- @staticmethod
- def add_continue(graph):
- return ('CONTINUE', graph)
-
- @staticmethod
- def catch_all():
- return ('CATCH_ALL',)
-
- @staticmethod
- def epsilon():
- return ('EPSILON',)
-
- @staticmethod
- def join_subgraph(graph, name, subgraph, modifier):
- if modifier:
- modifier = NfaBuilder.__modifer_map[modifier]
- return ('JOIN', graph, name, subgraph, modifier)
-
- @staticmethod
- def or_graphs(graphs):
- return reduce(lambda acc, g: ('OR', acc, g), graphs)
-
- @staticmethod
- def cat_graphs(graphs):
- return reduce(lambda acc, g: ('CAT', acc, g), graphs)
-
- __modifer_map = {
- '+': 'ONE_OR_MORE',
- '?': 'ZERO_OR_ONE',
- '*': 'ZERO_OR_MORE',
- }
-
- @staticmethod
- def apply_modifier(modifier, graph):
- return (NfaBuilder.__modifer_map[modifier], graph)
-
class Nfa(Automaton):
def __init__(self, start, end, nodes_created):
« no previous file with comments | « tools/lexer_generator/generator.py ('k') | tools/lexer_generator/nfa_builder.py » ('j') | no next file with comments »

Powered by Google App Engine
This is Rietveld 408576698