diff options
| -rw-r--r-- | Makefile | 8 | ||||
| -rw-r--r-- | lib.c | 11 | ||||
| -rw-r--r-- | parse.c | 10 | ||||
| -rw-r--r-- | pre-process.c | 51 | ||||
| -rw-r--r-- | test-lexing.c | 1 | ||||
| -rw-r--r-- | test-parsing.c | 2 | ||||
| -rw-r--r-- | token.h | 14 | ||||
| -rw-r--r-- | tokenize.c | 18 |
8 files changed, 93 insertions, 22 deletions
@@ -4,11 +4,11 @@ PROGRAMS=test-lexing test-parsing all: $(PROGRAMS) -test-lexing: test-lexing.o tokenize.o lib.o - gcc -o $@ test-lexing.o tokenize.o lib.o +test-lexing: test-lexing.o tokenize.o pre-process.o lib.o + gcc -o $@ test-lexing.o tokenize.o pre-process.o lib.o -test-parsing: test-parsing.o parse.o tokenize.o symbol.o lib.o - gcc -o $@ test-parsing.o parse.o tokenize.o symbol.o lib.o +test-parsing: test-parsing.o parse.o tokenize.o symbol.o pre-process.o lib.o + gcc -o $@ test-parsing.o parse.o tokenize.o symbol.o pre-process.o lib.o test-parsing.o: token.h parse.h test-lexing.o: token.h @@ -36,6 +36,17 @@ struct allocator_struct { unsigned int allocations, total_bytes, useful_bytes; }; +void drop_all_allocations(struct allocator_struct *desc) +{ + struct allocation_blob *blob = desc->blobs; + + while (blob) { + struct allocation_blob *next = blob->next; + free(blob); + blob = next; + } +} + void *allocate(struct allocator_struct *desc, unsigned int size) { unsigned long alignment = desc->alignment; @@ -71,16 +71,6 @@ struct statement *alloc_statement(struct token * token, int type) return stmt; } -static int match_op(struct token *token, int op) -{ - return token->type == TOKEN_SPECIAL && token->special == op; -} - -static int match_ident(struct token *token, struct ident *id) -{ - return token->type == TOKEN_IDENT && token->ident == id; -} - static int match_oplist(int op, ...) { va_list args; diff --git a/pre-process.c b/pre-process.c new file mode 100644 index 00000000..7a7a7db0 --- /dev/null +++ b/pre-process.c @@ -0,0 +1,51 @@ +/* + * Do C preprocessing, based on a token list gathered by + * the tokenizer. + * + * This may not be the smartest preprocessor on the planet. + */ +#include <stdio.h> +#include <stdlib.h> +#include <stdarg.h> +#include <stddef.h> +#include <string.h> +#include <ctype.h> +#include <unistd.h> + +#include "lib.h" +#include "token.h" +#include "symbol.h" + +static struct token *preprocessor_line(struct token * head) +{ + struct token *token = head->next; // hash-mark + + do { + fprintf(stderr, "%s ", show_token(token)); + token = token->next; + } while (!token->newline && !eof_token(token)); + fprintf(stderr, "\n"); + head->next = token; + return head; +} + +static void do_preprocess(struct token *head) +{ + do { + struct token *next = head->next; + if (next->newline && match_op(next, '#')) { + head = preprocessor_line(head); + continue; + } + head = next; + } while (!eof_token(head)); +} + +struct token * preprocess(struct token *token) +{ + struct token header = { 0, }; + + header.next = token; + do_preprocess(&header); + return header.next; +} diff --git a/test-lexing.c b/test-lexing.c index 263f2e60..72206dbd 100644 --- a/test-lexing.c +++ b/test-lexing.c @@ -17,6 +17,7 @@ int main(int argc, char **argv) die("No such file: %s", argv[1]); token = tokenize(argv[1], fd); + token = preprocess(token); while (!eof_token(token)) { struct token *next = token->next; char separator = '\n'; diff --git a/test-parsing.c b/test-parsing.c index 94e5e703..7fa39bd4 100644 --- a/test-parsing.c +++ b/test-parsing.c @@ -21,7 +21,7 @@ int main(int argc, char **argv) die("No such file: %s", argv[1]); init_symbols(); token = tokenize(argv[1], fd); - + token = preprocess(token); translation_unit(token, &list); show_symbol_list(list); @@ -95,7 +95,8 @@ struct string { struct token { unsigned int type:8, stream:8, - pos:16; + pos:15, + newline:1; unsigned int line; struct token *next; union { @@ -127,5 +128,16 @@ extern struct token * tokenize(const char *, int); extern void die(const char *, ...); extern void warn(struct token *, const char *, ...); extern void show_identifier_stats(void); +extern struct token *preprocess(struct token *); + +static inline int match_op(struct token *token, int op) +{ + return token->type == TOKEN_SPECIAL && token->special == op; +} + +static inline int match_ident(struct token *token, struct ident *id) +{ + return token->type == TOKEN_IDENT && token->ident == id; +} #endif @@ -145,7 +145,7 @@ struct token * alloc_token(int stream, int line, int pos) #define BUFSIZE (4096) typedef struct { - int fd, line, pos, offset, size; + int fd, line, pos, offset, size, newline; struct token **tokenlist; struct token *token; unsigned char buffer[BUFSIZE]; @@ -170,6 +170,7 @@ static int nextchar(action_t *action) action->pos++; if (c == '\n') { action->line++; + action->newline = 1; action->pos = 0; } return c; @@ -196,10 +197,8 @@ static void add_token(action_t *action) static void drop_token(action_t *action) { - struct token *token = action->token; - + action->newline |= action->token->newline; action->token = NULL; - free(token); } static int get_base_number(unsigned int base, char **p, int next, action_t *action) @@ -663,6 +662,7 @@ struct token * tokenize(const char *name, int fd) action.tokenlist = &retval; action.token = NULL; action.line = 1; + action.newline = 1; action.pos = 0; action.fd = fd; action.offset = 0; @@ -670,9 +670,15 @@ struct token * tokenize(const char *name, int fd) c = nextchar(&action); while (c != EOF) { + if (c == '\\') { + c = nextchar(&action); + action.newline = 0; + } if (!isspace(c)) { - action.token = alloc_token(stream, action.line, action.pos); - + struct token *token = alloc_token(stream, action.line, action.pos); + token->newline = action.newline; + action.newline = 0; + action.token = token; c = get_one_token(c, &action); continue; } |