X-Git-Url: http://git.sourceforge.jp/view?p=pf3gnuchains%2Fgcc-fork.git;a=blobdiff_plain;f=gcc%2Fcp%2FChangeLog;h=9ea3fe3e7bd731445a2a573747c4e5480b3e3543;hp=7124c5a1cc53853efa9342e0f1da1824bdac298c;hb=b9dd3954e0bd53cf675974ad9e18c6a43f102730;hpb=2e533eb1a8d2a7803c82f0446b1f146bd55ef111 diff --git a/gcc/cp/ChangeLog b/gcc/cp/ChangeLog index 7124c5a1cc5..9ea3fe3e7bd 100644 --- a/gcc/cp/ChangeLog +++ b/gcc/cp/ChangeLog @@ -1,3 +1,95 @@ +2004-09-23 Zack Weinberg + + * decl.c (grokfndecl): If ::main is found not to return int, + correct it after issuing a diagnostic. + (grokdeclarator): If the incoming type was error_mark_node, do + not complain about declaring something with no type. + (start_function): Change check for ::main not returning int to + an assertion, as grokfndecl now catches this when the user did it. + * init.c (perform_member_init, sort_mem_initializers) + (emit_mem_initializers): Make most diagnostics be issued on + the line of current_function_decl, not whatever the current + input line is. + * parser.c (cp_lexer_peek_token_emit_debug_info): Surround + definition and declaration with #ifdef ENABLE_CHECKING. + Avoid unnecessary use of fprintf. + (cp_lexer_print_token, cp_lexer_debug_stream): Adjust stub + definitions to avoid warnings. + (cp_lexer_new_main): Add assertion that first token is not a + padding token. + (cp_lexer_new_from_token_array): Fold into ... + (cp_lexer_new_from_tokens): ... here. Add assertion that + first token is not a padding token. + (cp_lexer_set_source_position_from_token): Move nearer to callers. + Remove unused lexer argument. + (cp_lexer_peek_token): Just print debugging report (if enabled) + and return lexer->next_token. + (cp_lexer_skip_purged_tokens): Delete. + (cp_lexer_next_token_is, cp_lexer_next_token_is_not): Make + inline, simplify bodies. + (cp_lexer_peek_nth_token): Add debugging report a la + cp_lexer_peek_token. + (cp_lexer_consume_token): Correct commentary. Advance over + purged tokens here. Set current source position here, from + token to be returned. Avoid unnecessary use of fprintf. + (cp_lexer_purge_token): Advance next_token pointer over this and + subsequent purged tokens. + (cp_parser_error): Adjust source position to that of the + peeked token. + (cp_parser_push_lexer_for_tokens, cp_parser_pop_lexer): New functions. + (cp_parser_string_literal): Remove some excessive cleverness. + (cp_parser_enum_specifier): Call start_enum before consuming + the opening brace. + (cp_parser_member_declaration): Make the "extra semicolon" + diagnostic consistently-worded with the other place this is + diagnosed. Explicitly set the diagnostic location to the + location of the offending semicolon. + (cp_parser_enclosed_template_argument_list): Use % quoting + in diagnostics. Do not use cp_parser_require. Set location + of diagnostics about improper use of '>>' to location of + offending token. + (cp_parser_late_parsing_for_member): + Use cp_parser_push_lexer_for_tokens and cp_parser_pop_lexer. + (cp_parser_late_parsing_default_args): Likewise. Manually + move some logic outside the loop. + +2004-09-23 Andrew Pinski + + PR c++/17618 + * cvt.c (cp_convert_to_pointer): Return early when the type is + an error_mark_node. + +2004-09-21 Fariborz Jahanian + + PR c++/13989 + PR c++/9844 + * decl.c (grokfndecl): Add new argument "attrlist", use it + to call cplus_decl_attributes. + (start_function): Remove call to cplus_decl_attributes. + * cvt.c (ocp_convert): Add support to use type conversion + function to vector type. + * parser.c (cp_parser_conversion_type_id): Add attributes, if any, + to the parsed type. + +2004-09-23 Paolo Bonzini + + PR c++/17596 + + * parser.c (cp_parser_token_tree_map_node, + cp_parser_pm_expression, cp_parser_additive_expression, + cp_parser_multiplicative_expression, cp_parser_shift_expression, + cp_parser_relational_expression, cp_parser_equality_expression, + cp_parser_and_expression, cp_parser_exclusive_or_expression, + cp_parser_inclusive_or_expression, + cp_parser_logical_and_expression, + cp_parser_logical_or_expression): Removed. + (enum cp_parser_prec, struct cp_parser_token_tree_map_node, + binops, binops_by_token): New. + (cp_parser_assignment_expression): Use cp_parser_binary_expression. + (cp_parser_new): Initialize binops_by_token. + (cp_parser_binary_expression): Rewritten. + (N_CP_TTYPES): New. + 2004-09-23 Kazu Hirata * parser.c: Fix a comment typo. @@ -31,7 +123,7 @@ class_hint_flags): Remove. (get_pseudo_ti_init): Use CLASSTYPE_REPEATED_BASE_P and CLASSTYPE_DIAMOND_SHAPED_P. - + 2004-09-21 Ziemowit Laski * cp-lang.c (LANG_HOOKS_FOLD_OBJ_TYPE_REF): Moved here from