def snap_term(stream, PatternDict): """term: primary primary term """ __debug_entry("term", stream) # -- primary result = snap_primary(stream, PatternDict) __debug_print("##primary(in term):", result) if result == None: return __debug_exit(None, stream) position_1 = stream.tell() # -- optional 'term' result_2 = snap_term(stream, PatternDict) __debug_print("##term(in term):", result_2) if result_2 == None: stream.seek(position_1) return __debug_exit(result, stream) ## print "##1:", result.get_string(NormalizeF=False) ## print "##2:", result_2.get_string(NormalizeF=False) result = sequentialize.do([result, result_2], MountToFirstStateMachineF=True, CloneRemainingStateMachinesF=False) return __debug_exit(construct.beautify(result), stream)
def snap_expression(stream, PatternDict): """expression: term term | expression """ __debug_entry("expression", stream) # -- term result = snap_term(stream, PatternDict) if result == None: return __debug_exit(None, stream) # -- optional '|' if not check(stream, '|'): return __debug_exit(result, stream) position_1 = stream.tell() __debug_print("'|' (in expression)") # -- expression result_2 = snap_expression(stream, PatternDict) __debug_print("expression(in expression):", result_2) if result_2 == None: stream.seek(position_1) return __debug_exit(result, stream) result = parallelize.do([result, result_2]) return __debug_exit(construct.beautify(result), stream)
def snap_primary(stream, PatternDict): """primary: " non_double_quote * " = character string [ non_rect_bracket_close ] = set of characters { identifier } = pattern replacement ( expression ) non_control_character+ = lonely characters primary repetition_cmd """ __debug_entry("primary", stream) x = stream.read(1) lookahead = stream.read(1) if x != "" and lookahead != "": stream.seek(-1, 1) if x == "": return __debug_exit(None, stream) # -- 'primary' primary if x == "\"": result = snap_character_string.do(stream) elif x == "[": stream.seek(-1, 1) result = character_set_expression.do(stream, PatternDict) elif x == "{": result = snap_replacement(stream, PatternDict) elif x == ".": result = create_ALL_BUT_NEWLINE_state_machine() elif x == "(": result = snap_bracketed_expression(stream, PatternDict) elif x.isspace(): # a lonestanding space ends the regular expression stream.seek(-1, 1) return __debug_exit(None, stream) elif x in ["*", "+", "?"]: raise RegularExpressionException( "lonely operator '%s' without expression proceeding." % x) elif x == "\\": if lookahead == "C": stream.read(1) result = snap_case_folded_pattern(stream, PatternDict) else: stream.seek(-1, 1) trigger_set = character_set_expression.snap_property_set(stream) if trigger_set == None: stream.seek( 1, 1) # snap_property_set() leaves tream right before '\\' char_code = snap_backslashed_character.do(stream) if char_code == None: raise RegularExpressionException( "Backslash followed by unrecognized character code.") trigger_set = char_code result = StateMachine() result.add_transition(result.init_state_index, trigger_set, AcceptanceF=True) elif x not in CONTROL_CHARACTERS: # NOTE: The '\' is not inside the control characters---for a reason. # It is used to define for example character codes using '\x' etc. stream.seek(-1, 1) result = snap_non_control_character(stream, PatternDict) else: # NOTE: This includes the '$' sign which means 'end of line' # because the '$' sign is in CONTROL_CHARACTERS, but is not checked # against. Thus, it it good to leave here on '$' because the # '$' sign is handled on the very top level. # this is not a valid primary stream.seek(-1, 1) return __debug_exit(None, stream) # -- optional repetition command? result_repeated = __snap_repetition_range(result, stream) ## print "##imr:", result.get_string(NormalizeF=False) if result_repeated != None: result = result_repeated return __debug_exit(construct.beautify(result), stream)