X-Git-Url: https://pintos-os.org/cgi-bin/gitweb.cgi?a=blobdiff_plain;f=src%2Flanguage%2Fdata-io%2Fcombine-files.c;h=aa6d3b98ec319b60f1efd31b3fcf299b7ed87654;hb=55e6e7ba37a30570f5a31e2d78c22dfa7b61a36f;hp=1a82ef3f1bf61bfcc6ce1e9a2559237f37fcc37d;hpb=99e37c4d062ac23f89070b578f28eb6d49eec632;p=pspp-builds.git diff --git a/src/language/data-io/combine-files.c b/src/language/data-io/combine-files.c index 1a82ef3f..aa6d3b98 100644 --- a/src/language/data-io/combine-files.c +++ b/src/language/data-io/combine-files.c @@ -1,5 +1,5 @@ /* PSPP - a program for statistical analysis. - Copyright (C) 1997-9, 2000, 2006, 2007, 2008, 2009 Free Software Foundation, Inc. + Copyright (C) 1997-9, 2000, 2006, 2007, 2008, 2009, 2010 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 @@ -170,7 +170,7 @@ combine_files (enum comb_command_type command, dict_set_case_limit (proc.dict, dict_get_case_limit (dataset_dict (ds))); - lex_match (lexer, '/'); + lex_match (lexer, T_SLASH); for (;;) { struct comb_file *file; @@ -185,7 +185,7 @@ combine_files (enum comb_command_type command, } else break; - lex_match (lexer, '='); + lex_match (lexer, T_EQUALS); if (proc.n_files >= allocated_files) proc.files = x2nrealloc (proc.files, &allocated_files, @@ -203,7 +203,7 @@ combine_files (enum comb_command_type command, file->in_name[0] = '\0'; file->in_var = NULL; - if (lex_match (lexer, '*')) + if (lex_match (lexer, T_ASTERISK)) { if (!proc_has_active_file (ds)) { @@ -230,7 +230,7 @@ combine_files (enum comb_command_type command, goto error; } - while (lex_match (lexer, '/')) + while (lex_match (lexer, T_SLASH)) if (lex_match_id (lexer, "RENAME")) { if (!parse_dict_rename (lexer, file->dict)) @@ -238,7 +238,7 @@ combine_files (enum comb_command_type command, } else if (lex_match_id (lexer, "IN")) { - lex_match (lexer, '='); + lex_match (lexer, T_EQUALS); if (lex_token (lexer) != T_ID) { lex_error (lexer, NULL); @@ -251,7 +251,7 @@ combine_files (enum comb_command_type command, "TABLE.")); goto error; } - strcpy (file->in_name, lex_tokid (lexer)); + strcpy (file->in_name, lex_tokcstr (lexer)); lex_get (lexer); } else if (lex_match_id (lexer, "SORT")) @@ -263,7 +263,7 @@ combine_files (enum comb_command_type command, merge_dictionary (proc.dict, file); } - while (lex_token (lexer) != '.') + while (lex_token (lexer) != T_ENDCMD) { if (lex_match (lexer, T_BY)) { @@ -278,7 +278,7 @@ combine_files (enum comb_command_type command, } saw_by = true; - lex_match (lexer, '='); + lex_match (lexer, T_EQUALS); if (!parse_sort_criteria (lexer, proc.dict, &proc.by_vars, &by_vars, NULL)) goto error; @@ -289,7 +289,7 @@ combine_files (enum comb_command_type command, struct comb_file *file = &proc.files[i]; size_t j; - for (j = 0; j < subcase_get_n_values (&proc.by_vars); j++) + for (j = 0; j < subcase_get_n_fields (&proc.by_vars); j++) { const char *name = var_get_name (by_vars[j]); struct variable *var = dict_lookup_var (file->dict, name); @@ -322,10 +322,10 @@ combine_files (enum comb_command_type command, goto error; } - lex_match (lexer, '='); + lex_match (lexer, T_EQUALS); if (!lex_force_id (lexer)) goto error; - strcpy (first_name, lex_tokid (lexer)); + strcpy (first_name, lex_tokcstr (lexer)); lex_get (lexer); } else if (command != COMB_UPDATE && lex_match_id (lexer, "LAST")) @@ -336,10 +336,10 @@ combine_files (enum comb_command_type command, goto error; } - lex_match (lexer, '='); + lex_match (lexer, T_EQUALS); if (!lex_force_id (lexer)) goto error; - strcpy (last_name, lex_tokid (lexer)); + strcpy (last_name, lex_tokcstr (lexer)); lex_get (lexer); } else if (lex_match_id (lexer, "MAP")) @@ -362,7 +362,7 @@ combine_files (enum comb_command_type command, goto error; } - if (!lex_match (lexer, '/') && lex_token (lexer) != '.') + if (!lex_match (lexer, T_SLASH) && lex_token (lexer) != T_ENDCMD) { lex_end_of_command (lexer); goto error; @@ -378,12 +378,12 @@ combine_files (enum comb_command_type command, } if (n_tables) { - msg (SE, _("BY is required when TABLE is specified.")); + msg (SE, _("BY is required when %s is specified."), "TABLE"); goto error; } if (saw_sort) { - msg (SE, _("BY is required when SORT is specified.")); + msg (SE, _("BY is required when %s is specified."), "SORT"); goto error; } } @@ -423,7 +423,7 @@ combine_files (enum comb_command_type command, } } - proc.output = autopaging_writer_create (dict_get_next_value_idx (proc.dict)); + proc.output = autopaging_writer_create (dict_get_proto (proc.dict)); taint = taint_clone (casewriter_get_taint (proc.output)); /* Set up case matcher. */ @@ -497,7 +497,10 @@ merge_dictionary (struct dictionary *const m, struct comb_file *f) m_docs = dict_get_documents (m); - /* If the input files have different encodings, then + /* FIXME: If the input files have different encodings, then + the result is undefined. + The correct thing to do would be to convert to an encoding + which can cope with all the input files (eg UTF-8). */ file_encoding = dict_get_encoding (f->dict); if ( file_encoding != NULL) @@ -559,7 +562,7 @@ merge_dictionary (struct dictionary *const m, struct comb_file *f) ds_put_format (&s, _("In an earlier file, %s was a string " "variable with width %d."), var_name, var_get_width (mv)); - msg (SE, ds_cstr (&s)); + msg (SE, "%s", ds_cstr (&s)); ds_destroy (&s); return false; } @@ -572,7 +575,7 @@ merge_dictionary (struct dictionary *const m, struct comb_file *f) var_set_label (mv, var_get_label (dv)); } else - mv = dict_clone_var_assert (m, dv, var_get_name (dv)); + mv = dict_clone_var_assert (m, dv); } return true; @@ -637,9 +640,14 @@ free_comb_proc (struct comb_proc *proc) dict_destroy (proc->dict); casewriter_destroy (proc->output); case_matcher_destroy (proc->matcher); + if (proc->prev_BY) + { + caseproto_destroy_values (subcase_get_proto (&proc->by_vars), + proc->prev_BY); + free (proc->prev_BY); + } subcase_destroy (&proc->by_vars); case_unref (proc->buffered_case); - free (proc->prev_BY); } static bool scan_table (struct comb_file *, union value by[]); @@ -786,7 +794,7 @@ create_output_case (const struct comb_proc *proc) struct ccase *output; size_t i; - output = case_create (dict_get_next_value_idx (proc->dict)); + output = case_create (dict_get_proto (proc->dict)); for (i = 0; i < n_vars; i++) { struct variable *v = dict_get_var (proc->dict, i); @@ -858,11 +866,15 @@ output_case (struct comb_proc *proc, struct ccase *output, union value by[]) if (new_BY) { - size_t n = (subcase_get_n_values (&proc->by_vars) - * sizeof (union value)); + size_t n_values = subcase_get_n_fields (&proc->by_vars); + const struct caseproto *proto = subcase_get_proto (&proc->by_vars); if (proc->prev_BY == NULL) - proc->prev_BY = xmalloc (n); - memcpy (proc->prev_BY, by, n); + { + proc->prev_BY = xmalloc (n_values * sizeof *proc->prev_BY); + caseproto_init_values (proto, proc->prev_BY); + } + caseproto_copy (subcase_get_proto (&proc->by_vars), 0, n_values, + proc->prev_BY, by); } } }