X-Git-Url: https://pintos-os.org/cgi-bin/gitweb.cgi?a=blobdiff_plain;f=src%2Flanguage%2Flexer%2Flexer.c;h=8b3f2a48d00474811a09c090f8a82d1b75ed7546;hb=6b003fcaaf12c21fda3d44580154b54c54c6cd54;hp=ebb8a9ed1ec1bbcad5f22141466718c2cd708113;hpb=480a0746507ce73d26f528b56dc3ed80195096e0;p=pspp-builds.git diff --git a/src/language/lexer/lexer.c b/src/language/lexer/lexer.c index ebb8a9ed..8b3f2a48 100644 --- a/src/language/lexer/lexer.c +++ b/src/language/lexer/lexer.c @@ -1,40 +1,39 @@ -/* PSPP - computes sample statistics. - Copyright (C) 1997-9, 2000, 2006 Free Software Foundation, Inc. +/* PSPP - a program for statistical analysis. + Copyright (C) 1997-9, 2000, 2006, 2009 Free Software Foundation, Inc. - This program is free software; you can redistribute it and/or - modify it under the terms of the GNU General Public License as - published by the Free Software Foundation; either version 2 of the - License, or (at your option) any later version. + This program is free software: you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation, either version 3 of the License, or + (at your option) any later version. - This program is distributed in the hope that it will be useful, but - WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - General Public License for more details. + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. You should have received a copy of the GNU General Public License - along with this program; if not, write to the Free Software - Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA - 02110-1301, USA. */ + along with this program. If not, see . */ #include #include "lexer.h" #include -#include +#include +#include #include #include #include #include +#include #include -#include #include #include #include -#include #include #include #include +#include -#include "size_max.h" +#include "xalloc.h" #include "gettext.h" #define _(msgid) gettext (msgid) @@ -44,7 +43,8 @@ #define DUMP_TOKENS 0 -struct lexer + +struct lexer { struct string line_buffer; @@ -53,10 +53,10 @@ struct lexer int token; /* Current token. */ double tokval; /* T_POS_NUM, T_NEG_NUM: the token's value. */ - char tokid [LONG_NAME_LEN + 1]; /* T_ID: the identifier. */ + char tokid [VAR_NAME_LEN + 1]; /* T_ID: the identifier. */ struct string tokstr; /* T_ID, T_STRING: token string value. - For T_ID, this is not truncated as is + For T_ID, this is not truncated as is tokid. */ char *prog; /* Pointer to next token in line_buffer. */ @@ -64,7 +64,7 @@ struct lexer int put_token ; /* If nonzero, next token returned by lex_get(). Used only in exceptional circumstances. */ - + struct string put_tokstr; double put_tokval; }; @@ -73,7 +73,7 @@ struct lexer static int parse_id (struct lexer *); /* How a string represents its contents. */ -enum string_type +enum string_type { CHARACTER_STRING, /* Characters. */ BINARY_STRING, /* Binary digits. */ @@ -109,11 +109,23 @@ lex_get_source_stream (const struct lexer *lex) return lex->ss; } +enum syntax_mode +lex_current_syntax_mode (const struct lexer *lex) +{ + return source_stream_current_syntax_mode (lex->ss); +} + +enum error_mode +lex_current_error_mode (const struct lexer *lex) +{ + return source_stream_current_error_mode (lex->ss); +} + void lex_destroy (struct lexer *lexer) { - if ( NULL != lexer ) + if ( NULL != lexer ) { ds_destroy (&lexer->put_tokstr); ds_destroy (&lexer->tokstr); @@ -129,7 +141,7 @@ lex_destroy (struct lexer *lexer) /* Copies put_token, lexer->put_tokstr, put_tokval into token, tokstr, tokval, respectively, and sets tokid appropriately. */ static void -restore_token (struct lexer *lexer) +restore_token (struct lexer *lexer) { assert (lexer->put_token != 0); lexer->token = lexer->put_token; @@ -142,7 +154,7 @@ restore_token (struct lexer *lexer) /* Copies token, tokstr, lexer->tokval into lexer->put_token, put_tokstr, put_lexer->tokval respectively. */ static void -save_token (struct lexer *lexer) +save_token (struct lexer *lexer) { lexer->put_token = lexer->token; ds_assign_string (&lexer->put_tokstr, &lexer->tokstr); @@ -176,7 +188,7 @@ lex_get (struct lexer *lexer) for (;;) { /* Skip whitespace. */ - while (isspace ((unsigned char) *lexer->prog)) + while (c_isspace ((unsigned char) *lexer->prog)) lexer->prog++; if (*lexer->prog) @@ -214,7 +226,7 @@ lex_get (struct lexer *lexer) /* Actually parse the token. */ ds_clear (&lexer->tokstr); - + switch (*lexer->prog) { case '-': case '.': @@ -233,26 +245,26 @@ lex_get (struct lexer *lexer) if (*lexer->prog == '-') { ds_put_char (&lexer->tokstr, *lexer->prog++); - while (isspace ((unsigned char) *lexer->prog)) + while (c_isspace ((unsigned char) *lexer->prog)) lexer->prog++; - if (!isdigit ((unsigned char) *lexer->prog) && *lexer->prog != '.') + if (!c_isdigit ((unsigned char) *lexer->prog) && *lexer->prog != '.') { lexer->token = '-'; break; } lexer->token = T_NEG_NUM; } - else + else lexer->token = T_POS_NUM; - + /* Parse the number, copying it into tokstr. */ - while (isdigit ((unsigned char) *lexer->prog)) + while (c_isdigit ((unsigned char) *lexer->prog)) ds_put_char (&lexer->tokstr, *lexer->prog++); if (*lexer->prog == '.') { ds_put_char (&lexer->tokstr, *lexer->prog++); - while (isdigit ((unsigned char) *lexer->prog)) + while (c_isdigit ((unsigned char) *lexer->prog)) ds_put_char (&lexer->tokstr, *lexer->prog++); } if (*lexer->prog == 'e' || *lexer->prog == 'E') @@ -260,12 +272,12 @@ lex_get (struct lexer *lexer) ds_put_char (&lexer->tokstr, *lexer->prog++); if (*lexer->prog == '+' || *lexer->prog == '-') ds_put_char (&lexer->tokstr, *lexer->prog++); - while (isdigit ((unsigned char) *lexer->prog)) + while (c_isdigit ((unsigned char) *lexer->prog)) ds_put_char (&lexer->tokstr, *lexer->prog++); } /* Parse as floating point. */ - lexer->tokval = strtod (ds_cstr (&lexer->tokstr), &tail); + lexer->tokval = c_strtod (ds_cstr (&lexer->tokstr), &tail); if (*tail) { msg (SE, _("%s does not form a valid number."), @@ -284,6 +296,7 @@ lex_get (struct lexer *lexer) break; case '(': case ')': case ',': case '=': case '+': case '/': + case '[': case ']': lexer->token = *lexer->prog++; break; @@ -348,34 +361,34 @@ lex_get (struct lexer *lexer) else lexer->token = parse_id (lexer); break; - + case 'o': case 'O': if (lexer->prog[1] == '\'' || lexer->prog[1] == '"') lexer->token = parse_string (lexer, OCTAL_STRING); else lexer->token = parse_id (lexer); break; - + case 'x': case 'X': if (lexer->prog[1] == '\'' || lexer->prog[1] == '"') lexer->token = parse_string (lexer, HEX_STRING); else lexer->token = parse_id (lexer); break; - + default: - if (lex_is_id1 (*lexer->prog)) + if (lex_is_id1 (*lexer->prog)) { lexer->token = parse_id (lexer); - break; + break; } else { - if (isgraph ((unsigned char) *lexer->prog)) - msg (SE, _("Bad character in input: `%c'."), *lexer->prog++); - else - msg (SE, _("Bad character in input: `\\%o'."), *lexer->prog++); - continue; + unsigned char c = *lexer->prog++; + char *c_name = xasprintf (c_isgraph (c) ? "%c" : "\\%o", c); + msg (SE, _("Bad character in input: `%s'."), c_name); + free (c_name); + continue; } } break; @@ -390,7 +403,7 @@ lex_get (struct lexer *lexer) tokstr. Returns the correct token type. */ static int -parse_id (struct lexer *lexer) +parse_id (struct lexer *lexer) { struct substring rest_of_line = ss_substr (ds_ss (&lexer->line_buffer), @@ -408,7 +421,7 @@ parse_id (struct lexer *lexer) /* Reports an error to the effect that subcommand SBC may only be specified once. */ void -lex_sbc_only_once (const char *sbc) +lex_sbc_only_once (const char *sbc) { msg (SE, _("Subcommand %s may only be specified once."), sbc); } @@ -416,7 +429,7 @@ lex_sbc_only_once (const char *sbc) /* Reports an error to the effect that subcommand SBC is missing. */ void -lex_sbc_missing (struct lexer *lexer, const char *sbc) +lex_sbc_missing (struct lexer *lexer, const char *sbc) { lex_error (lexer, _("missing required subcommand %s"), sbc); } @@ -442,7 +455,7 @@ lex_error (struct lexer *lexer, const char *message, ...) { char buf[1024]; va_list args; - + va_start (args, message); vsnprintf (buf, 1024, message, args); va_end (args); @@ -473,11 +486,20 @@ lex_end_of_command (struct lexer *lexer) /* Returns true if the current token is a number. */ bool -lex_is_number (struct lexer *lexer) +lex_is_number (struct lexer *lexer) { return lexer->token == T_POS_NUM || lexer->token == T_NEG_NUM; } + +/* Returns true if the current token is a string. */ +bool +lex_is_string (struct lexer *lexer) +{ + return lexer->token == T_STRING; +} + + /* Returns the value of the current token, which must be a floating point number. */ double @@ -492,8 +514,7 @@ bool lex_is_integer (struct lexer *lexer) { return (lex_is_number (lexer) - && lexer->tokval != NOT_LONG - && lexer->tokval >= LONG_MIN + && lexer->tokval > LONG_MIN && lexer->tokval <= LONG_MAX && floor (lexer->tokval) == lexer->tokval); } @@ -506,7 +527,7 @@ lex_integer (struct lexer *lexer) assert (lex_is_integer (lexer)); return lexer->tokval; } - + /* Token matching functions. */ /* If TOK is the current token, skips it and returns true @@ -529,9 +550,19 @@ lex_match (struct lexer *lexer, int t) Otherwise, returns false. */ bool lex_match_id (struct lexer *lexer, const char *s) +{ + return lex_match_id_n (lexer, s, 3); +} + +/* If the current token is the identifier S, skips it and returns + true. The identifier may be abbreviated to its first N + letters. + Otherwise, returns false. */ +bool +lex_match_id_n (struct lexer *lexer, const char *s, size_t n) { if (lexer->token == T_ID - && lex_id_match (ss_cstr (s), ss_cstr (lexer->tokid))) + && lex_id_match_n (ss_cstr (s), ss_cstr (lexer->tokid), n)) { lex_get (lexer); return true; @@ -615,7 +646,7 @@ lex_force_int (struct lexer *lexer) return false; } } - + /* If this token is a number, does nothing and returns true. Otherwise, reports an error and returns false. */ bool @@ -627,7 +658,7 @@ lex_force_num (struct lexer *lexer) lex_error (lexer, _("expecting number")); return false; } - + /* If this token is an identifier, does nothing and returns true. Otherwise, reports an error and returns false. */ bool @@ -661,7 +692,7 @@ lex_look_ahead (struct lexer *lexer) for (;;) { - while (isspace ((unsigned char) *lexer->prog)) + while (c_isspace ((unsigned char) *lexer->prog)) lexer->prog++; if (*lexer->prog) break; @@ -671,7 +702,7 @@ lex_look_ahead (struct lexer *lexer) else if (!lex_get_line (lexer)) return 0; - if (lexer->put_token) + if (lexer->put_token) return lexer->put_token; } @@ -710,29 +741,33 @@ lex_put_back_id (struct lexer *lexer, const char *id) /* Returns the entire contents of the current line. */ const char * -lex_entire_line (struct lexer *lexer) +lex_entire_line (const struct lexer *lexer) { return ds_cstr (&lexer->line_buffer); } const struct string * -lex_entire_line_ds (struct lexer *lexer) +lex_entire_line_ds (const struct lexer *lexer) { return &lexer->line_buffer; } /* As lex_entire_line(), but only returns the part of the current line - that hasn't already been tokenized. - If END_DOT is non-null, stores nonzero into *END_DOT if the line - ends with a terminal dot, or zero if it doesn't. */ + that hasn't already been tokenized. */ const char * -lex_rest_of_line (struct lexer *lexer, int *end_dot) +lex_rest_of_line (const struct lexer *lexer) { - if (end_dot) - *end_dot = lexer->dot; return lexer->prog; } +/* Returns true if the current line ends in a terminal dot, + false otherwise. */ +bool +lex_end_dot (const struct lexer *lexer) +{ + return lexer->dot; +} + /* Causes the rest of the current input line to be ignored for tokenization purposes. */ void @@ -753,15 +788,15 @@ lex_discard_line (struct lexer *lexer) the user doesn't want to finish typing a command that will be ignored anyway. */ void -lex_discard_rest_of_command (struct lexer *lexer) +lex_discard_rest_of_command (struct lexer *lexer) { if (!getl_is_interactive (lexer->ss)) { while (lexer->token != T_STOP && lexer->token != '.') lex_get (lexer); } - else - lex_discard_line (lexer); + else + lex_discard_line (lexer); } /* Weird line reading functions. */ @@ -787,7 +822,7 @@ strip_comments (struct string *string) else if (*cp == '\'' || *cp == '"') quote = *cp; } - + /* If we're not inside a quotation, check for comment. */ if (quote == EOF) { @@ -806,7 +841,7 @@ strip_comments (struct string *string) continue; } } - + /* Check commenting. */ if (in_comment) *cp = ' '; @@ -820,20 +855,20 @@ strip_comments (struct string *string) *LINE_STARTS_COMMAND and *LINE_ENDS_COMMAND appropriately. */ void lex_preprocess_line (struct string *line, - enum getl_syntax syntax, + enum syntax_mode syntax, bool *line_starts_command, bool *line_ends_command) { strip_comments (line); ds_rtrim (line, ss_cstr (CC_SPACES)); - *line_ends_command = (ds_chomp (line, get_endcmd ()) - || (ds_is_empty (line) && get_nulline ())); + *line_ends_command = (ds_chomp (line, settings_get_endcmd ()) + || (ds_is_empty (line) && settings_get_nulline ())); *line_starts_command = false; if (syntax == GETL_BATCH) { int first = ds_first (line); - *line_starts_command = !isspace (first); - if (first == '+' || first == '-') + *line_starts_command = !c_isspace (first); + if (first == '+' || first == '-') *ds_data (line) = ' '; } } @@ -841,12 +876,13 @@ lex_preprocess_line (struct string *line, /* Reads a line, without performing any preprocessing. Sets *SYNTAX, if SYNTAX is non-null, to the line's syntax mode. */ -bool -lex_get_line_raw (struct lexer *lexer, enum getl_syntax *syntax) +bool +lex_get_line_raw (struct lexer *lexer) { - enum getl_syntax dummy; - bool ok = getl_read_line (lexer->ss, &lexer->line_buffer, - syntax != NULL ? syntax : &dummy); + bool ok = getl_read_line (lexer->ss, &lexer->line_buffer); + enum syntax_mode mode = lex_current_syntax_mode (lexer); + journal_write (mode == GETL_BATCH, ds_cstr (&lexer->line_buffer)); + return ok; } @@ -857,15 +893,15 @@ bool lex_get_line (struct lexer *lexer) { bool line_starts_command; - enum getl_syntax syntax; - if (!lex_get_line_raw (lexer, &syntax)) + if (!lex_get_line_raw (lexer)) { lexer->prog = NULL; - return false; + return false; } - lex_preprocess_line (&lexer->line_buffer, syntax, + lex_preprocess_line (&lexer->line_buffer, + lex_current_syntax_mode (lexer), &line_starts_command, &lexer->dot); if (line_starts_command) @@ -901,7 +937,7 @@ char * lex_token_representation (struct lexer *lexer) { char *token_rep; - + switch (lexer->token) { case T_ID: @@ -916,12 +952,12 @@ lex_token_representation (struct lexer *lexer) char *sp, *dp; for (sp = ds_cstr (&lexer->tokstr); sp < ds_end (&lexer->tokstr); sp++) - if (!isprint ((unsigned char) *sp)) + if (!c_isprint ((unsigned char) *sp)) { hexstring = 1; break; } - + token_rep = xmalloc (2 + ds_length (&lexer->tokstr) * 2 + 1 + 1); dp = token_rep; @@ -944,7 +980,7 @@ lex_token_representation (struct lexer *lexer) } *dp++ = '\''; *dp = '\0'; - + return token_rep; } break; @@ -960,7 +996,7 @@ lex_token_representation (struct lexer *lexer) default: return xstrdup (lex_token_name (lexer->token)); } - + NOT_REACHED (); } @@ -982,20 +1018,20 @@ lex_negative_to_dash (struct lexer *lexer) lexer->token = '-'; } } - + /* Skip a COMMENT command. */ void lex_skip_comment (struct lexer *lexer) { for (;;) { - if (!lex_get_line (lexer)) + if (!lex_get_line (lexer)) { lexer->put_token = T_STOP; lexer->prog = NULL; return; } - + if (lexer->put_token == '.') break; @@ -1012,7 +1048,7 @@ lex_skip_comment (struct lexer *lexer) hex digits, according to TYPE. The string is converted to characters having the specified values. */ static void -convert_numeric_string_to_char_string (struct lexer *lexer, +convert_numeric_string_to_char_string (struct lexer *lexer, enum string_type type) { const char *base_name; @@ -1022,7 +1058,7 @@ convert_numeric_string_to_char_string (struct lexer *lexer, size_t i; char *p; - switch (type) + switch (type) { case BINARY_STRING: base_name = _("binary"); @@ -1042,10 +1078,10 @@ convert_numeric_string_to_char_string (struct lexer *lexer, default: NOT_REACHED (); } - + byte_cnt = ds_length (&lexer->tokstr) / chars_per_byte; if (ds_length (&lexer->tokstr) % chars_per_byte) - msg (SE, _("String of %s digits has %d characters, which is not a " + msg (SE, _("String of %s digits has %zu characters, which is not a " "multiple of %d."), base_name, ds_length (&lexer->tokstr), chars_per_byte); @@ -1054,7 +1090,7 @@ convert_numeric_string_to_char_string (struct lexer *lexer, { int value; int j; - + value = 0; for (j = 0; j < chars_per_byte; j++, p++) { @@ -1089,7 +1125,7 @@ convert_numeric_string_to_char_string (struct lexer *lexer, buffer pointer lexer->prog must point to the initial single or double quote. TYPE indicates the type of string to be parsed. Returns token type. */ -static int +static int parse_string (struct lexer *lexer, enum string_type type) { if (type != CHARACTER_STRING) @@ -1101,7 +1137,7 @@ parse_string (struct lexer *lexer, enum string_type type) { /* Single or double quote. */ int c = *lexer->prog++; - + /* Accumulate section. */ for (;;) { @@ -1111,7 +1147,7 @@ parse_string (struct lexer *lexer, enum string_type type) msg (SE, _("Unterminated string constant.")); goto finish; } - + /* Double quote characters to embed them in strings. */ if (*lexer->prog == c) { @@ -1130,7 +1166,7 @@ parse_string (struct lexer *lexer, enum string_type type) break; for (;;) { - while (isspace ((unsigned char) *lexer->prog)) + while (c_isspace ((unsigned char) *lexer->prog)) lexer->prog++; if (*lexer->prog) break; @@ -1152,7 +1188,7 @@ parse_string (struct lexer *lexer, enum string_type type) break; for (;;) { - while (isspace ((unsigned char) *lexer->prog)) + while (c_isspace ((unsigned char) *lexer->prog)) lexer->prog++; if (*lexer->prog) break; @@ -1183,14 +1219,14 @@ finish: if (ds_length (&lexer->tokstr) > 255) { - msg (SE, _("String exceeds 255 characters in length (%d characters)."), + msg (SE, _("String exceeds 255 characters in length (%zu characters)."), ds_length (&lexer->tokstr)); ds_truncate (&lexer->tokstr, 255); } - + return T_STRING; } - + #if DUMP_TOKENS /* Reads one token from the lexer and writes a textual representation on stdout for debugging purposes. */ @@ -1206,7 +1242,7 @@ dump_token (struct lexer *lexer) if (curfn) fprintf (stderr, "%s:%d\t", curfn, curln); } - + switch (lexer->token) { case T_ID: @@ -1247,13 +1283,13 @@ dump_token (struct lexer *lexer) /* Token Accessor Functions */ -int +int lex_token (const struct lexer *lexer) { return lexer->token; } -double +double lex_tokval (const struct lexer *lexer) { return lexer->tokval; @@ -1270,3 +1306,28 @@ lex_tokstr (const struct lexer *lexer) { return &lexer->tokstr; } + +/* If the lexer is positioned at the (pseudo)identifier S, which + may contain a hyphen ('-'), skips it and returns true. Each + half of the identifier may be abbreviated to its first three + letters. + Otherwise, returns false. */ +bool +lex_match_hyphenated_word (struct lexer *lexer, const char *s) +{ + const char *hyphen = strchr (s, '-'); + if (hyphen == NULL) + return lex_match_id (lexer, s); + else if (lexer->token != T_ID + || !lex_id_match (ss_buffer (s, hyphen - s), ss_cstr (lexer->tokid)) + || lex_look_ahead (lexer) != '-') + return false; + else + { + lex_get (lexer); + lex_force_match (lexer, '-'); + lex_force_match_id (lexer, hyphen + 1); + return true; + } +} +