def _do(UnicodeTrafoInfo):
"""
PURPOSE: Writes converters for conversion towards UTF8/UTF16/UCS2/UCS4.
UnicodeTrafoInfo:
Provides the information about the relation of character codes in a particular
coding to unicode character codes. It is provided in the following form:
# Codec Values Unicode Values
[ (Source0_Begin, Source0_End, TargetInterval0_Begin),
(Source1_Begin, Source1_End, TargetInterval1_Begin),
(Source2_Begin, Source2_End, TargetInterval2_Begin),
...
]
"""
codec_name = make_safe_identifier(UnicodeTrafoInfo.name).lower()
utf8_epilog, utf8_function_body = ConverterWriterUTF8().do(UnicodeTrafoInfo)
utf16_prolog, utf16_function_body = ConverterWriterUTF16().do(UnicodeTrafoInfo)
dummy, utf32_function_body = ConverterWriterUTF32().do(UnicodeTrafoInfo)
# Provide only the constant which are necessary
FileName = os.path.normpath( QUEX_PATH
+ Lng["$code_base"]
+ "/converter_helper/TXT-from-codec-buffer.i")
codec_header = Setup.get_file_reference(Setup.output_buffer_codec_header)
txt_i = blue_print(get_file_content_or_die(FileName),
[["$$CODEC$$", codec_name],
["$$EPILOG$$", utf8_epilog],
["$$CODEC_HEADER$$", codec_header],
["$$BODY_UTF8$$", utf8_function_body],
["$$BODY_UTF16$$", utf16_function_body],
["$$BODY_UTF32$$", utf32_function_body]])
# A separate declaration header is required
FileName = os.path.normpath( QUEX_PATH
+ Lng["$code_base"]
+ "/converter_helper/TXT-from-codec-buffer")
template_h_txt = get_file_content_or_die(FileName)
txt_h = template_h_txt.replace("$$CODEC$$", codec_name)
return txt_h, txt_i
def _do(UnicodeTrafoInfo):
"""
PURPOSE: Writes converters for conversion towards UTF8/UTF16/UCS2/UCS4.
UnicodeTrafoInfo:
Provides the information about the relation of character codes in a particular
coding to unicode character codes. It is provided in the following form:
# Codec Values Unicode Values
[ (Source0_Begin, Source0_End, TargetInterval0_Begin),
(Source1_Begin, Source1_End, TargetInterval1_Begin),
(Source2_Begin, Source2_End, TargetInterval2_Begin),
...
]
"""
codec_name = make_safe_identifier(UnicodeTrafoInfo.name).lower()
utf8_epilog, utf8_function_body = ConverterWriterUTF8().do(
UnicodeTrafoInfo)
utf16_prolog, utf16_function_body = ConverterWriterUTF16().do(
UnicodeTrafoInfo)
dummy, utf32_function_body = ConverterWriterUTF32().do(UnicodeTrafoInfo)
# Provide only the constant which are necessary
FileName = os.path.normpath(QUEX_PATH + Lng["$code_base"] +
"/converter_helper/TXT-from-codec-buffer.i")
codec_header = Setup.get_file_reference(Setup.output_buffer_codec_header)
txt_i = blue_print(get_file_content_or_die(FileName),
[["$$CODEC$$", codec_name], ["$$EPILOG$$", utf8_epilog],
["$$CODEC_HEADER$$", codec_header],
["$$BODY_UTF8$$", utf8_function_body],
["$$BODY_UTF16$$", utf16_function_body],
["$$BODY_UTF32$$", utf32_function_body]])
# A separate declaration header is required
FileName = os.path.normpath(QUEX_PATH + Lng["$code_base"] +
"/converter_helper/TXT-from-codec-buffer")
template_h_txt = get_file_content_or_die(FileName)
txt_h = template_h_txt.replace("$$CODEC$$", codec_name)
return txt_h, txt_i
def do(ModeDescriptionDB):
IndentationSupportF = blackboard.required_support_indentation_count()
BeginOfLineSupportF = blackboard.required_support_begin_of_line()
LexerClassName = Setup.analyzer_class_name
ConfigurationTemplateFile =( QUEX_PATH \
+ Lng["$code_base"] \
+ "/analyzer/configuration/TXT").replace("//","/")
txt = get_file_content_or_die(ConfigurationTemplateFile)
# -- check if exit/entry handlers have to be active
entry_handler_active_f = False
exit_handler_active_f = False
for mode in ModeDescriptionDB.values():
entry_handler_active_f |= mode.incidence_db.has_key(E_IncidenceIDs.MODE_ENTRY)
exit_handler_active_f |= mode.incidence_db.has_key(E_IncidenceIDs.MODE_EXIT)
# Buffer filler converter (0x0 means: no buffer filler converter)
converter_new_str = "# define QUEX_SETTING_BUFFER_FILLERS_CONVERTER_NEW "
if Setup.converter_user_new_func != "":
converter_new_str += Setup.converter_user_new_func + "()"
else:
converter_new_str = "/* " + converter_new_str + " */"
# Token repetition support
token_repeat_test_txt = ""
for token_id_str in blackboard.token_repetition_token_id_list:
token_repeat_test_txt += "TokenID == %s || " % token_id_str
if token_repeat_test_txt != "":
token_repeat_test_txt = token_repeat_test_txt[:-3]
else:
token_repeat_test_txt = "false"
if Setup.analyzer_derived_class_name != "":
analyzer_derived_class_name = Setup.analyzer_derived_class_name
else:
analyzer_derived_class_name = Setup.analyzer_class_name
txt = __switch(txt, "QUEX_OPTION_COLUMN_NUMBER_COUNTING", Setup.count_column_number_f)
txt = __switch(txt, "QUEX_OPTION_COMPUTED_GOTOS", False)
txt = __switch(txt, "QUEX_OPTION_CONVERTER_ICONV", Setup.converter_iconv_f)
txt = __switch(txt, "QUEX_OPTION_CONVERTER_ICU", Setup.converter_icu_f)
txt = __switch(txt, "QUEX_OPTION_INCLUDE_STACK", Setup.include_stack_support_f)
txt = __switch(txt, "QUEX_OPTION_LINE_NUMBER_COUNTING", Setup.count_line_number_f)
txt = __switch(txt, "QUEX_OPTION_POST_CATEGORIZER", Setup.post_categorizer_f)
txt = __switch(txt, "QUEX_OPTION_RUNTIME_MODE_TRANSITION_CHECK", Setup.mode_transition_check_f)
txt = __switch(txt, "QUEX_OPTION_STRING_ACCUMULATOR", Setup.string_accumulator_f)
txt = __switch(txt, "QUEX_OPTION_TOKEN_POLICY_QUEUE", Setup.token_policy == "queue")
txt = __switch(txt, "QUEX_OPTION_TOKEN_POLICY_SINGLE", Setup.token_policy == "single")
txt = __switch(txt, "QUEX_OPTION_TOKEN_REPETITION_SUPPORT", token_repeat_test_txt != "false")
txt = __switch(txt, "QUEX_OPTION_USER_MANAGED_TOKEN_MEMORY", Setup.token_memory_management_by_user_f)
txt = __switch(txt, "__QUEX_OPTION_BIG_ENDIAN", Setup.buffer_byte_order == "big")
txt = __switch(txt, "__QUEX_OPTION_CONVERTER_HELPER", Setup.converter_helper_required_f)
txt = __switch(txt, "__QUEX_OPTION_CONVERTER", Setup.converter_f)
txt = __switch(txt, "QUEX_OPTION_INDENTATION_TRIGGER", IndentationSupportF)
txt = __switch(txt, "__QUEX_OPTION_LITTLE_ENDIAN", Setup.buffer_byte_order == "little")
txt = __switch(txt, "__QUEX_OPTION_ON_ENTRY_HANDLER_PRESENT", entry_handler_active_f)
txt = __switch(txt, "__QUEX_OPTION_ON_EXIT_HANDLER_PRESENT", exit_handler_active_f)
txt = __switch(txt, "__QUEX_OPTION_PLAIN_C", Setup.language.upper() == "C")
txt = __switch(txt, "__QUEX_OPTION_SUPPORT_BEGIN_OF_LINE_PRE_CONDITION", BeginOfLineSupportF)
txt = __switch(txt, "__QUEX_OPTION_SYSTEM_ENDIAN", Setup.byte_order_is_that_of_current_system_f)
txt = __switch(txt, "QUEX_OPTION_BUFFER_BASED_ANALYZIS", Setup.buffer_based_analyzis_f)
txt = __switch(txt, "__QUEX_OPTION_ENGINE_RUNNING_ON_CODEC", Setup.buffer_codec.name != "unicode")
# -- token class related definitions
token_descr = blackboard.token_type_definition
# -- name of the character codec
codec_name = make_safe_identifier(Setup.buffer_codec.name).lower()
# Setup.buffer_element_size can be '-1'. This signals then that
# sizeof(QUEX_TYPE_CHARACTER) needs to be used. A numeric value
# is required here.
character_size_str = "%i" % Setup.buffer_element_size
def namespace(NameSpaceList):
result = Lng.NAMESPACE_REFERENCE(NameSpaceList, TrailingDelimiterF=False)
if len(result) == 0: return ""
assert Setup.language.upper() != "C++" or len(result) > 2, \
"Error while generating namespace reference '%s'" % result
return result
txt = blue_print(txt,
[
["$$BUFFER_LIMIT_CODE$$", "0x%X" % Setup.buffer_limit_code],
["$$QUEX_SETTING_CHARACTER_CODEC$$", codec_name],
["$$INCLUDE_GUARD_EXTENSION$$", get_include_guard_extension(Lng.NAMESPACE_REFERENCE(Setup.analyzer_name_space) + "__" + Setup.analyzer_class_name)],
["$$INITIAL_LEXER_MODE_ID$$", "QUEX_NAME(ModeID_%s)" % blackboard.initial_mode.get_pure_text()],
["$$LEXER_BUILD_DATE$$", time.asctime()],
["$$LEXER_CLASS_NAME$$", LexerClassName],
["$$LEXER_CLASS_NAME_SAFE$$", Setup.analyzer_name_safe],
["$$LEXER_DERIVED_CLASS_NAME$$", analyzer_derived_class_name],
["$$MAX_MODE_CLASS_N$$", repr(len(ModeDescriptionDB))],
["$$NAMESPACE_MAIN$$", namespace(Setup.analyzer_name_space)],
["$$NAMESPACE_MAIN_CLOSE$$", Lng.NAMESPACE_CLOSE(Setup.analyzer_name_space).replace("\n", "\\\n")],
["$$NAMESPACE_MAIN_OPEN$$", Lng.NAMESPACE_OPEN(Setup.analyzer_name_space).replace("\n", "\\\n")],
["$$NAMESPACE_TOKEN$$", namespace(token_descr.name_space)],
["$$NAMESPACE_TOKEN_CLOSE$$", Lng.NAMESPACE_CLOSE(token_descr.name_space).replace("\n", "\\\n")],
["$$NAMESPACE_TOKEN_OPEN$$", Lng.NAMESPACE_OPEN(token_descr.name_space).replace("\n", "\\\n")],
["$$PATH_TERMINATION_CODE$$", "0x%X" % Setup.path_limit_code],
["$$QUEX_SETTING_BUFFER_FILLERS_CONVERTER_NEW$$", converter_new_str],
["$$QUEX_TYPE_CHARACTER$$", Setup.buffer_element_type],
["$$QUEX_SETTING_CHARACTER_SIZE$$", character_size_str],
["$$QUEX_NAMESPACE_LEXEME_NULL_OPEN$$", Lng.NAMESPACE_OPEN(Setup.lexeme_null_namespace).replace("\n", "\\\n")],
["$$QUEX_NAMESPACE_LEXEME_NULL_CLOSE$$", Lng.NAMESPACE_CLOSE(Setup.lexeme_null_namespace).replace("\n", "\\\n")],
["$$QUEX_LEXEME_NULL$$", Setup.lexeme_null_full_name_cpp],
["$$QUEX_LEXEME_NULL_SAFE$$", Setup.lexeme_null_name_safe],
["$$QUEX_LEXEME_NULL_IN_ITS_NAMESPACE$$", Setup.lexeme_null_name],
["$$QUEX_VERSION$$", QUEX_VERSION],
["$$TOKEN_CLASS$$", token_descr.class_name],
["$$TOKEN_CLASS_NAME_SAFE$$", token_descr.class_name_safe],
["$$TOKEN_COLUMN_N_TYPE$$", token_descr.column_number_type.get_pure_text()],
["$$TOKEN_ID_TYPE$$", token_descr.token_id_type.get_pure_text()],
["$$TOKEN_LINE_N_TYPE$$", token_descr.line_number_type.get_pure_text()],
["$$TOKEN_PREFIX$$", Setup.token_id_prefix],
["$$TOKEN_QUEUE_SAFETY_BORDER$$", repr(Setup.token_queue_safety_border)],
["$$TOKEN_QUEUE_SIZE$$", repr(Setup.token_queue_size)],
["$$TOKEN_REPEAT_TEST$$", token_repeat_test_txt],
["$$USER_LEXER_VERSION$$", Setup.user_application_version_id],
])
return txt
def do(ModeDB):
IndentationSupportF = blackboard.requires_indentation_count(ModeDB)
BeginOfLineSupportF = blackboard.requires_begin_of_line_condition_support(
ModeDB)
LanguageDB = Setup.language_db
LexerClassName = Setup.analyzer_class_name
ConfigurationTemplateFile =( QUEX_PATH \
+ Setup.language_db["$code_base"] \
+ "/analyzer/configuration/TXT").replace("//","/")
txt = get_file_content_or_die(ConfigurationTemplateFile)
# -- check if exit/entry handlers have to be active
entry_handler_active_f = False
exit_handler_active_f = False
for mode in ModeDB.values():
if len(mode.get_code_fragment_list("on_entry")) != 0:
entry_handler_active_f = True
if len(mode.get_code_fragment_list("on_exit")) != 0:
exit_handler_active_f = True
# Buffer filler converter (0x0 means: no buffer filler converter)
converter_new_str = "# define QUEX_SETTING_BUFFER_FILLERS_CONVERTER_NEW "
if Setup.converter_user_new_func != "":
converter_new_str += Setup.converter_user_new_func + "()"
else:
converter_new_str = "/* " + converter_new_str + " */"
# Token repetition support
token_repeat_test_txt = ""
for token_id_str in blackboard.token_repetition_token_id_list:
token_repeat_test_txt += "TokenID == %s || " % token_id_str
if token_repeat_test_txt != "":
token_repeat_test_txt = token_repeat_test_txt[:-3]
else:
token_repeat_test_txt = "false"
if Setup.analyzer_derived_class_name != "":
analyzer_derived_class_name = Setup.analyzer_derived_class_name
else:
analyzer_derived_class_name = Setup.analyzer_class_name
txt = __switch(txt, "QUEX_OPTION_COLUMN_NUMBER_COUNTING",
Setup.count_column_number_f)
txt = __switch(txt, "QUEX_OPTION_COMPUTED_GOTOS", False)
txt = __switch(txt, "QUEX_OPTION_CONVERTER_ICONV", Setup.converter_iconv_f)
txt = __switch(txt, "QUEX_OPTION_CONVERTER_ICU", Setup.converter_icu_f)
txt = __switch(txt, "QUEX_OPTION_INCLUDE_STACK",
Setup.include_stack_support_f)
txt = __switch(txt, "QUEX_OPTION_LINE_NUMBER_COUNTING",
Setup.count_line_number_f)
txt = __switch(txt, "QUEX_OPTION_POST_CATEGORIZER",
Setup.post_categorizer_f)
txt = __switch(txt, "QUEX_OPTION_RUNTIME_MODE_TRANSITION_CHECK",
Setup.mode_transition_check_f)
txt = __switch(txt, "QUEX_OPTION_STRING_ACCUMULATOR",
Setup.string_accumulator_f)
txt = __switch(txt, "QUEX_OPTION_TOKEN_POLICY_QUEUE",
Setup.token_policy == "queue")
txt = __switch(txt, "QUEX_OPTION_TOKEN_POLICY_SINGLE",
Setup.token_policy == "single")
txt = __switch(txt, "QUEX_OPTION_TOKEN_REPETITION_SUPPORT",
token_repeat_test_txt != "false")
txt = __switch(txt, "QUEX_OPTION_USER_MANAGED_TOKEN_MEMORY",
Setup.token_memory_management_by_user_f)
txt = __switch(txt, "__QUEX_OPTION_BIG_ENDIAN",
Setup.buffer_byte_order == "big")
txt = __switch(txt, "__QUEX_OPTION_CONVERTER_HELPER",
Setup.converter_helper_required_f)
txt = __switch(txt, "__QUEX_OPTION_CONVERTER", Setup.converter_f)
txt = __switch(txt, "QUEX_OPTION_INDENTATION_TRIGGER", IndentationSupportF)
txt = __switch(txt, "__QUEX_OPTION_LITTLE_ENDIAN",
Setup.buffer_byte_order == "little")
txt = __switch(txt, "__QUEX_OPTION_ON_ENTRY_HANDLER_PRESENT",
entry_handler_active_f)
txt = __switch(txt, "__QUEX_OPTION_ON_EXIT_HANDLER_PRESENT",
exit_handler_active_f)
txt = __switch(txt, "__QUEX_OPTION_PLAIN_C", Setup.language.upper() == "C")
txt = __switch(txt, "__QUEX_OPTION_SUPPORT_BEGIN_OF_LINE_PRE_CONDITION",
BeginOfLineSupportF)
txt = __switch(txt, "__QUEX_OPTION_SYSTEM_ENDIAN",
Setup.byte_order_is_that_of_current_system_f)
txt = __switch(txt, "QUEX_OPTION_BUFFER_BASED_ANALYZIS",
Setup.buffer_based_analyzis_f)
txt = __switch(txt, "__QUEX_OPTION_ENGINE_RUNNING_ON_CODEC",
Setup.buffer_codec != "unicode")
# -- token class related definitions
token_descr = blackboard.token_type_definition
# -- name of the character codec
codec_name = "unicode"
if Setup.buffer_codec != "unicode":
codec_name = make_safe_identifier(Setup.buffer_codec).lower()
# Setup.buffer_element_size can be '-1'. This signals then that
# sizeof(QUEX_TYPE_CHARACTER) needs to be used. A numeric value
# is required here.
character_size_str = "%i" % Setup.buffer_element_size
def namespace(NameSpaceList):
result = Setup.language_db.NAMESPACE_REFERENCE(NameSpaceList)
if result == "::": return ""
assert Setup.language.upper() != "C++" or len(result) > 2, \
"Error while generating namespace reference '%s'" % result
return result[:-2]
txt = blue_print(txt, [
["$$BUFFER_LIMIT_CODE$$",
"0x%X" % Setup.buffer_limit_code],
["$$QUEX_SETTING_CHARACTER_CODEC$$", codec_name],
[
"$$INCLUDE_GUARD_EXTENSION$$",
get_include_guard_extension(
LanguageDB.NAMESPACE_REFERENCE(Setup.analyzer_name_space) +
"__" + Setup.analyzer_class_name)
],
[
"$$INITIAL_LEXER_MODE_ID$$",
"QUEX_NAME(ModeID_%s)" % blackboard.initial_mode.get_pure_code()
],
["$$LEXER_BUILD_DATE$$", time.asctime()],
["$$LEXER_CLASS_NAME$$", LexerClassName],
["$$LEXER_CLASS_NAME_SAFE$$", Setup.analyzer_name_safe],
["$$LEXER_DERIVED_CLASS_NAME$$", analyzer_derived_class_name],
["$$MAX_MODE_CLASS_N$$", repr(len(ModeDB))],
["$$NAMESPACE_MAIN$$",
namespace(Setup.analyzer_name_space)],
[
"$$NAMESPACE_MAIN_CLOSE$$",
LanguageDB.NAMESPACE_CLOSE(Setup.analyzer_name_space).replace(
"\n", "\\\n")
],
[
"$$NAMESPACE_MAIN_OPEN$$",
LanguageDB.NAMESPACE_OPEN(Setup.analyzer_name_space).replace(
"\n", "\\\n")
],
["$$NAMESPACE_TOKEN$$",
namespace(token_descr.name_space)],
[
"$$NAMESPACE_TOKEN_CLOSE$$",
LanguageDB.NAMESPACE_CLOSE(token_descr.name_space).replace(
"\n", "\\\n")
],
[
"$$NAMESPACE_TOKEN_OPEN$$",
LanguageDB.NAMESPACE_OPEN(token_descr.name_space).replace(
"\n", "\\\n")
],
["$$PATH_TERMINATION_CODE$$",
"0x%X" % Setup.path_limit_code],
["$$QUEX_SETTING_BUFFER_FILLERS_CONVERTER_NEW$$", converter_new_str],
["$$QUEX_TYPE_CHARACTER$$", Setup.buffer_element_type],
["$$QUEX_SETTING_CHARACTER_SIZE$$", character_size_str],
[
"$$QUEX_NAMESPACE_LEXEME_NULL_OPEN$$",
LanguageDB.NAMESPACE_OPEN(Setup.lexeme_null_namespace).replace(
"\n", "\\\n")
],
[
"$$QUEX_NAMESPACE_LEXEME_NULL_CLOSE$$",
LanguageDB.NAMESPACE_CLOSE(Setup.lexeme_null_namespace).replace(
"\n", "\\\n")
],
["$$QUEX_LEXEME_NULL$$", Setup.lexeme_null_full_name_cpp],
["$$QUEX_LEXEME_NULL_SAFE$$", Setup.lexeme_null_name_safe],
["$$QUEX_LEXEME_NULL_IN_ITS_NAMESPACE$$", Setup.lexeme_null_name],
["$$QUEX_VERSION$$", QUEX_VERSION],
["$$TOKEN_CLASS$$", token_descr.class_name],
["$$TOKEN_CLASS_NAME_SAFE$$", token_descr.class_name_safe],
[
"$$TOKEN_COLUMN_N_TYPE$$",
token_descr.column_number_type.get_pure_code()
],
["$$TOKEN_ID_TYPE$$",
token_descr.token_id_type.get_pure_code()],
[
"$$TOKEN_LINE_N_TYPE$$",
token_descr.line_number_type.get_pure_code()
],
["$$TOKEN_PREFIX$$", Setup.token_id_prefix],
[
"$$TOKEN_QUEUE_SAFETY_BORDER$$",
repr(Setup.token_queue_safety_border)
],
["$$TOKEN_QUEUE_SIZE$$",
repr(Setup.token_queue_size)],
["$$TOKEN_REPEAT_TEST$$", token_repeat_test_txt],
["$$USER_LEXER_VERSION$$", Setup.user_application_version_id],
])
return txt
