X-Git-Url: https://pintos-os.org/cgi-bin/gitweb.cgi?a=blobdiff_plain;f=src%2Flanguage%2Fdata-io%2Finpt-pgm.c;h=462289987b4044c7568272abdab35a0a4b5cd687;hb=5c3291dc396b795696e94f47780308fd7ace6fc4;hp=94fa4044b291cb1c86997319c413665222d2bf7d;hpb=5729150f2c4a00caf688e7e3b514b10b1987e211;p=pspp-builds.git diff --git a/src/language/data-io/inpt-pgm.c b/src/language/data-io/inpt-pgm.c index 94fa4044..46228998 100644 --- a/src/language/data-io/inpt-pgm.c +++ b/src/language/data-io/inpt-pgm.c @@ -1,21 +1,18 @@ -/* PSPP - computes sample statistics. - Copyright (C) 1997-9, 2000 Free Software Foundation, Inc. - Written by Ben Pfaff . +/* PSPP - a program for statistical analysis. + Copyright (C) 1997-9, 2000, 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 @@ -24,9 +21,9 @@ #include #include -#include #include -#include +#include +#include #include #include #include @@ -36,18 +33,20 @@ #include #include #include -#include +#include #include #include #include #include #include +#include "xalloc.h" + #include "gettext.h" #define _(msgid) gettext (msgid) /* Private result codes for use within INPUT PROGRAM. */ -enum cmd_result_extensions +enum cmd_result_extensions { CMD_END_INPUT_PROGRAM = CMD_PRIVATE_FIRST, CMD_END_CASE @@ -58,22 +57,20 @@ enum value_init_type { INP_NUMERIC = 01, /* Numeric. */ INP_STRING = 0, /* String. */ - + INP_INIT_ONCE = 02, /* Initialize only once. */ INP_REINIT = 0, /* Reinitialize for each iteration. */ }; -struct input_program_pgm +struct input_program_pgm { struct trns_chain *trns_chain; + enum trns_result restart; - size_t case_nr; /* Incremented by END CASE transformation. */ - write_case_func *write_case;/* Called by END CASE. */ - write_case_data wc_data; /* Aux data used by END CASE. */ + casenumber case_nr; /* Incremented by END CASE transformation. */ - enum value_init_type *init; /* How to initialize each `union value'. */ - size_t init_cnt; /* Number of elements in inp_init. */ - size_t case_size; /* Size of case in bytes. */ + struct caseinit *init; + struct caseproto *proto; }; static void destroy_input_program (struct input_program_pgm *); @@ -82,230 +79,185 @@ static trns_proc_func reread_trns_proc; static trns_proc_func end_file_trns_proc; static trns_free_func reread_trns_free; -static const struct case_source_class input_program_source_class; +static const struct casereader_class input_program_casereader_class; static bool inside_input_program; /* Returns true if we're parsing the inside of a INPUT PROGRAM...END INPUT PROGRAM construct, false otherwise. */ bool -in_input_program (void) +in_input_program (void) { return inside_input_program; } /* Emits an END CASE transformation for INP. */ static void -emit_END_CASE (struct input_program_pgm *inp) +emit_END_CASE (struct dataset *ds, struct input_program_pgm *inp) { - add_transformation (end_case_trns_proc, NULL, inp); + add_transformation (ds, end_case_trns_proc, NULL, inp); } int -cmd_input_program (void) +cmd_input_program (struct lexer *lexer, struct dataset *ds) { struct input_program_pgm *inp; - size_t i; bool saw_END_CASE = false; - discard_variables (); - if (token != '.') - return lex_end_of_command (); + proc_discard_active_file (ds); + if (lex_token (lexer) != '.') + return lex_end_of_command (lexer); inp = xmalloc (sizeof *inp); inp->trns_chain = NULL; inp->init = NULL; - + inp->proto = NULL; + inside_input_program = true; - for (;;) + for (;;) { - enum cmd_result result = cmd_parse (CMD_STATE_INPUT_PROGRAM); + enum cmd_result result = cmd_parse_in_state (lexer, ds, CMD_STATE_INPUT_PROGRAM); if (result == CMD_END_INPUT_PROGRAM) break; - else if (result == CMD_END_CASE) + else if (result == CMD_END_CASE) { - emit_END_CASE (inp); - saw_END_CASE = true; + emit_END_CASE (ds, inp); + saw_END_CASE = true; } else if (cmd_result_is_failure (result) && result != CMD_FAILURE) { if (result == CMD_EOF) msg (SE, _("Unexpected end-of-file within INPUT PROGRAM.")); inside_input_program = false; - discard_variables (); + proc_discard_active_file (ds); destroy_input_program (inp); return result; } } if (!saw_END_CASE) - emit_END_CASE (inp); + emit_END_CASE (ds, inp); inside_input_program = false; - if (dict_get_next_value_idx (default_dict) == 0) + if (dict_get_next_value_idx (dataset_dict (ds)) == 0) { msg (SE, _("Input program did not create any variables.")); - discard_variables (); + proc_discard_active_file (ds); destroy_input_program (inp); return CMD_FAILURE; } - - inp->trns_chain = proc_capture_transformations (); + + inp->trns_chain = proc_capture_transformations (ds); trns_chain_finalize (inp->trns_chain); + inp->restart = TRNS_CONTINUE; + /* Figure out how to initialize each input case. */ - inp->init_cnt = dict_get_next_value_idx (default_dict); - inp->init = xnmalloc (inp->init_cnt, sizeof *inp->init); - for (i = 0; i < inp->init_cnt; i++) - inp->init[i] = -1; - for (i = 0; i < dict_get_var_cnt (default_dict); i++) - { - struct variable *var = dict_get_var (default_dict, i); - enum value_init_type value_init; - size_t j; - - value_init = var->type == NUMERIC ? INP_NUMERIC : INP_STRING; - value_init |= var->leave ? INP_INIT_ONCE : INP_REINIT; - - for (j = 0; j < var->nv; j++) - inp->init[j + var->fv] = value_init; - } - for (i = 0; i < inp->init_cnt; i++) - assert (inp->init[i] != -1); - inp->case_size = dict_get_case_size (default_dict); + inp->init = caseinit_create (); + caseinit_mark_for_init (inp->init, dataset_dict (ds)); + inp->proto = caseproto_ref (dict_get_proto (dataset_dict (ds))); - proc_set_source (create_case_source (&input_program_source_class, inp)); + proc_set_active_file_data ( + ds, casereader_create_sequential (NULL, inp->proto, CASENUMBER_MAX, + &input_program_casereader_class, inp)); return CMD_SUCCESS; } int -cmd_end_input_program (void) +cmd_end_input_program (struct lexer *lexer UNUSED, struct dataset *ds UNUSED) { assert (in_input_program ()); - return CMD_END_INPUT_PROGRAM; + return CMD_END_INPUT_PROGRAM; } -/* Initializes case C. Called before the first case is read. */ -static void -init_case (const struct input_program_pgm *inp, struct ccase *c) +/* Returns true if STATE is valid given the transformations that + are allowed within INPUT PROGRAM. */ +static bool +is_valid_state (enum trns_result state) { - size_t i; - - for (i = 0; i < inp->init_cnt; i++) - switch (inp->init[i]) - { - case INP_NUMERIC | INP_INIT_ONCE: - case_data_rw (c, i)->f = 0.0; - break; - case INP_NUMERIC | INP_REINIT: - case_data_rw (c, i)->f = SYSMIS; - break; - case INP_STRING | INP_INIT_ONCE: - case INP_STRING | INP_REINIT: - memset (case_data_rw (c, i)->s, ' ', sizeof case_data_rw (c, i)->s); - break; - default: - assert (0); - } + return (state == TRNS_CONTINUE + || state == TRNS_ERROR + || state == TRNS_END_FILE + || state >= 0); } -/* Clears case C. Called between reading successive records. */ -static void -clear_case (const struct input_program_pgm *inp, struct ccase *c) +/* Reads and returns one case. + Returns the case if successful, null at end of file or if an + I/O error occurred. */ +static struct ccase * +input_program_casereader_read (struct casereader *reader UNUSED, void *inp_) { - size_t i; - - for (i = 0; i < inp->init_cnt; i++) - switch (inp->init[i]) - { - case INP_NUMERIC | INP_INIT_ONCE: - break; - case INP_NUMERIC | INP_REINIT: - case_data_rw (c, i)->f = SYSMIS; - break; - case INP_STRING | INP_INIT_ONCE: - break; - case INP_STRING | INP_REINIT: - memset (case_data_rw (c, i)->s, ' ', sizeof case_data_rw (c, i)->s); - break; - default: - assert (0); - } -} - -/* Executes each transformation in turn on a `blank' case. - Returns true if successful, false if an I/O error occurred. */ -static bool -input_program_source_read (struct case_source *source, - struct ccase *c, - write_case_func *write_case, - write_case_data wc_data) -{ - struct input_program_pgm *inp = source->aux; + struct input_program_pgm *inp = inp_; + struct ccase *c = case_create (inp->proto); - inp->case_nr = 1; - inp->write_case = write_case; - inp->wc_data = wc_data; - for (init_case (inp, c); ; clear_case (inp, c)) + do { - enum trns_result result = trns_chain_execute (inp->trns_chain, c, - &inp->case_nr); - if (result == TRNS_ERROR) - return false; - else if (result == TRNS_END_FILE) - return true; + assert (is_valid_state (inp->restart)); + if (inp->restart == TRNS_ERROR || inp->restart == TRNS_END_FILE) + { + case_unref (c); + return NULL; + } + + c = case_unshare (c); + caseinit_init_vars (inp->init, c); + inp->restart = trns_chain_execute (inp->trns_chain, inp->restart, + &c, inp->case_nr); + assert (is_valid_state (inp->restart)); + caseinit_update_left_vars (inp->init, c); } + while (inp->restart < 0); + + return c; } static void -destroy_input_program (struct input_program_pgm *pgm) +destroy_input_program (struct input_program_pgm *pgm) { - if (pgm != NULL) + if (pgm != NULL) { trns_chain_destroy (pgm->trns_chain); - free (pgm->init); + caseinit_destroy (pgm->init); + caseproto_unref (pgm->proto); free (pgm); } } -/* Destroys an INPUT PROGRAM source. */ +/* Destroys the casereader. */ static void -input_program_source_destroy (struct case_source *source) +input_program_casereader_destroy (struct casereader *reader UNUSED, void *inp_) { - struct input_program_pgm *inp = source->aux; - + struct input_program_pgm *inp = inp_; + if (inp->restart == TRNS_ERROR) + casereader_force_error (reader); destroy_input_program (inp); } -static const struct case_source_class input_program_source_class = +static const struct casereader_class input_program_casereader_class = { - "INPUT PROGRAM", + input_program_casereader_read, + input_program_casereader_destroy, + NULL, NULL, - input_program_source_read, - input_program_source_destroy, }; int -cmd_end_case (void) +cmd_end_case (struct lexer *lexer, struct dataset *ds UNUSED) { assert (in_input_program ()); - if (token == '.') + if (lex_token (lexer) == '.') return CMD_END_CASE; - return lex_end_of_command (); + return lex_end_of_command (lexer); } -/* Sends the current case as the source's output. */ +/* Outputs the current case */ int -end_case_trns_proc (void *inp_, struct ccase *c, int case_nr UNUSED) +end_case_trns_proc (void *inp_, struct ccase **c UNUSED, + casenumber case_nr UNUSED) { struct input_program_pgm *inp = inp_; - - if (!inp->write_case (inp->wc_data)) - return TRNS_ERROR; - inp->case_nr++; - clear_case (inp, c); - return TRNS_CONTINUE; + return TRNS_END_CASE; } /* REREAD transformation. */ @@ -317,7 +269,7 @@ struct reread_trns /* Parses REREAD command. */ int -cmd_reread (void) +cmd_reread (struct lexer *lexer, struct dataset *ds) { struct file_handle *fh; /* File to be re-read. */ struct expression *e; /* Expression for column to set. */ @@ -325,52 +277,54 @@ cmd_reread (void) fh = fh_get_default_handle (); e = NULL; - while (token != '.') + while (lex_token (lexer) != '.') { - if (lex_match_id ("COLUMN")) + if (lex_match_id (lexer, "COLUMN")) { - lex_match ('='); - + lex_match (lexer, '='); + if (e) { msg (SE, _("COLUMN subcommand multiply specified.")); expr_free (e); return CMD_CASCADING_FAILURE; } - - e = expr_parse (default_dict, EXPR_NUMBER); + + e = expr_parse (lexer, ds, EXPR_NUMBER); if (!e) return CMD_CASCADING_FAILURE; } - else if (lex_match_id ("FILE")) + else if (lex_match_id (lexer, "FILE")) { - lex_match ('='); - fh = fh_parse (FH_REF_FILE | FH_REF_INLINE); + lex_match (lexer, '='); + fh_unref (fh); + fh = fh_parse (lexer, FH_REF_FILE | FH_REF_INLINE); if (fh == NULL) { expr_free (e); return CMD_CASCADING_FAILURE; } - lex_get (); } else { - lex_error (NULL); + lex_error (lexer, NULL); expr_free (e); + return CMD_CASCADING_FAILURE; } } t = xmalloc (sizeof *t); - t->reader = dfm_open_reader (fh); + t->reader = dfm_open_reader (fh, lexer); t->column = e; - add_transformation (reread_trns_proc, reread_trns_free, t); + add_transformation (ds, reread_trns_proc, reread_trns_free, t); + fh_unref (fh); return CMD_SUCCESS; } /* Executes a REREAD transformation. */ static int -reread_trns_proc (void *t_, struct ccase *c, int case_num) +reread_trns_proc (void *t_, struct ccase **c, casenumber case_num) { struct reread_trns *t = t_; @@ -378,8 +332,8 @@ reread_trns_proc (void *t_, struct ccase *c, int case_num) dfm_reread_record (t->reader, 1); else { - double column = expr_evaluate_num (t->column, c, case_num); - if (!finite (column) || column < 1) + double column = expr_evaluate_num (t->column, *c, case_num); + if (!isfinite (column) || column < 1) { msg (SE, _("REREAD: Column numbers must be positive finite " "numbers. Column set to 1.")); @@ -404,19 +358,19 @@ reread_trns_free (void *t_) /* Parses END FILE command. */ int -cmd_end_file (void) +cmd_end_file (struct lexer *lexer, struct dataset *ds) { assert (in_input_program ()); - add_transformation (end_file_trns_proc, NULL, NULL); + add_transformation (ds, end_file_trns_proc, NULL, NULL); - return lex_end_of_command (); + return lex_end_of_command (lexer); } /* Executes an END FILE transformation. */ static int -end_file_trns_proc (void *trns_ UNUSED, struct ccase *c UNUSED, - int case_num UNUSED) +end_file_trns_proc (void *trns_ UNUSED, struct ccase **c UNUSED, + casenumber case_num UNUSED) { return TRNS_END_FILE; }