From 95951ea7bb8728cf54ae4136cb59d0af9e8a06bd Mon Sep 17 00:00:00 2001 From: Carl Worth Date: Wed, 26 May 2010 15:57:10 -0700 Subject: Treat newlines as space when invoking a function-like macro invocation. This adds three new pieces of state to the parser, (is_control_line, newline_as_space, and paren_count), and a large amount of messy code. I'd definitely like to see a cleaner solution for this. With this fix, the "define-func-extra-newlines" now passes so we put it back to test #26 where it was originally (lately it has been known as test #55). Also, we tweak test 25 slightly. Previously this test was ending a file function-like macro name that was not actually a macro (not followed by a left parenthesis). As is, this fix was making that test fail because the text_line production expects to see a terminating NEWLINE, but that NEWLINE is now getting turned into a SPACE here. This seems unlikely to be a problem in the wild, (function macros being used in a non-macro sense seems rare enough---but more than likely they won't happen at the end of a file). Still, we document this shortcoming in the README. --- README | 4 +++ glcpp-parse.y | 61 ++++++++++++++++++++++++++++++++-- glcpp.h | 3 ++ tests/025-func-macro-as-non-macro.c | 2 +- tests/026-define-func-extra-newlines.c | 6 ++++ tests/055-define-func-extra-newlines.c | 6 ---- 6 files changed, 73 insertions(+), 9 deletions(-) create mode 100644 tests/026-define-func-extra-newlines.c delete mode 100644 tests/055-define-func-extra-newlines.c diff --git a/README b/README index f0f64c2..ab42a3f 100644 --- a/README +++ b/README @@ -24,3 +24,7 @@ parentheses. The #error, #pragma, #extension, #version, and #line macros are not yet supported. + +A file that ends with a function-like macro name as the last +non-whitespace token will result in a parse error, (where it should be +passed through as is). \ No newline at end of file diff --git a/glcpp-parse.y b/glcpp-parse.y index a809ebf..1346b65 100644 --- a/glcpp-parse.y +++ b/glcpp-parse.y @@ -856,6 +856,9 @@ glcpp_parser_create (void) hash_table_string_compare); parser->active = _string_list_create (parser); parser->space_tokens = 1; + parser->newline_as_space = 0; + parser->in_control_line = 0; + parser->paren_count = 0; parser->skip_stack = NULL; @@ -1274,8 +1277,62 @@ glcpp_parser_lex (glcpp_parser_t *parser) token_node_t *node; int ret; - if (parser->lex_from_list == NULL) - return glcpp_lex (parser->scanner); + if (parser->lex_from_list == NULL) { + ret = glcpp_lex (parser->scanner); + + /* XXX: This ugly block of code exists for the sole + * purpose of converting a NEWLINE token into a SPACE + * token, but only in the case where we have seen a + * function-like macro name, but have not yet seen its + * closing parenthesis. + * + * There's perhaps a more compact way to do this with + * mid-rule actions in the grammar. + * + * I'm definitely not pleased with the complexity of + * this code here. + */ + if (parser->newline_as_space) + { + if (ret == '(') { + parser->paren_count++; + } else if (ret == ')') { + parser->paren_count--; + if (parser->paren_count == 0) + parser->newline_as_space = 0; + } else if (ret == NEWLINE) { + ret = SPACE; + } else if (ret != SPACE) { + if (parser->paren_count == 0) + parser->newline_as_space = 0; + } + } + else if (parser->in_control_line) + { + if (ret == NEWLINE) + parser->in_control_line = 0; + } + else if (ret == HASH_DEFINE_OBJ || ret == HASH_DEFINE_FUNC || + ret == HASH_UNDEF || ret == HASH_IF || + ret == HASH_IFDEF || ret == HASH_IFNDEF || + ret == HASH_ELIF || ret == HASH_ELSE || + ret == HASH_ENDIF || ret == HASH) + { + parser->in_control_line = 1; + } + else if (ret == IDENTIFIER) + { + macro_t *macro; + macro = hash_table_find (parser->defines, + yylval.str); + if (macro && macro->is_function) { + parser->newline_as_space = 1; + parser->paren_count = 0; + } + } + + return ret; + } node = parser->lex_from_node; diff --git a/glcpp.h b/glcpp.h index e5be1a6..5c8c304 100644 --- a/glcpp.h +++ b/glcpp.h @@ -128,6 +128,9 @@ struct glcpp_parser { struct hash_table *defines; string_list_t *active; int space_tokens; + int newline_as_space; + int in_control_line; + int paren_count; skip_node_t *skip_stack; token_list_t *lex_from_list; token_node_t *lex_from_node; diff --git a/tests/025-func-macro-as-non-macro.c b/tests/025-func-macro-as-non-macro.c index 3dbe026..b433671 100644 --- a/tests/025-func-macro-as-non-macro.c +++ b/tests/025-func-macro-as-non-macro.c @@ -1,2 +1,2 @@ #define foo(bar) bar -foo +foo bar diff --git a/tests/026-define-func-extra-newlines.c b/tests/026-define-func-extra-newlines.c new file mode 100644 index 0000000..0d83740 --- /dev/null +++ b/tests/026-define-func-extra-newlines.c @@ -0,0 +1,6 @@ +#define foo(a) bar + +foo +( +1 +) diff --git a/tests/055-define-func-extra-newlines.c b/tests/055-define-func-extra-newlines.c deleted file mode 100644 index 0d83740..0000000 --- a/tests/055-define-func-extra-newlines.c +++ /dev/null @@ -1,6 +0,0 @@ -#define foo(a) bar - -foo -( -1 -) -- cgit v1.2.3