X-Git-Url: https://pintos-os.org/cgi-bin/gitweb.cgi?a=blobdiff_plain;f=src%2Fdata%2Fgnumeric-reader.c;h=6cc7b3e1c254c0d920a93f2a1e1275b95a3b20d4;hb=4785da39b9cb392c0a6a2c38a8abac199a72d615;hp=c379717505cb940de485dbee8c4d05059bb53e0c;hpb=5c3291dc396b795696e94f47780308fd7ace6fc4;p=pspp diff --git a/src/data/gnumeric-reader.c b/src/data/gnumeric-reader.c index c379717505..6cc7b3e1c2 100644 --- a/src/data/gnumeric-reader.c +++ b/src/data/gnumeric-reader.c @@ -1,5 +1,5 @@ /* PSPP - a program for statistical analysis. - Copyright (C) 2007, 2009 Free Software Foundation, Inc. + Copyright (C) 2007, 2009, 2010, 2011, 2012 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 @@ -14,52 +14,52 @@ You should have received a copy of the GNU General Public License along with this program. If not, see . */ - - #include -#include -#include +#include "libpspp/message.h" +#include "libpspp/misc.h" -#include "minmax.h" +#include "gl/minmax.h" +#include "gl/c-strtod.h" #include "gettext.h" #define _(msgid) gettext (msgid) #define N_(msgid) (msgid) +#include "spreadsheet-reader.h" + +#include "c-xvasprintf.h" #if !GNM_SUPPORT struct casereader * -gnumeric_open_reader (struct gnumeric_read_info *gri, struct dictionary **dict) +gnumeric_open_reader (struct spreadsheet_read_info *gri, struct spreadsheet_read_options *opts, struct dictionary **dict) { - msg (ME, _("Support for Gnumeric files was not compiled into this installation of PSPP")); + msg (ME, _("Support for %s files was not compiled into this installation of PSPP"), "Gnumeric"); return NULL; } #else -#include -#include -#include -#include -#include -#include -#include +#include "data/gnumeric-reader.h" +#include +#include #include #include #include -#include -#include -#include - -#include "gnumeric-reader.h" -#include -#include +#include "data/case.h" +#include "data/casereader-provider.h" +#include "data/dictionary.h" +#include "data/identifier.h" +#include "data/value.h" +#include "data/variable.h" +#include "libpspp/i18n.h" +#include "libpspp/str.h" +#include "gl/xalloc.h" static void gnm_file_casereader_destroy (struct casereader *, void *); @@ -73,73 +73,6 @@ static const struct casereader_class gnm_file_casereader_class = NULL, }; -/* Convert a string, which is an integer encoded in base26 - IE, A=0, B=1, ... Z=25 to the integer it represents. - ... except that in this scheme, digits with an exponent - greater than 1 are implicitly incremented by 1, so - AA = 0 + 1*26, AB = 1 + 1*26, - ABC = 2 + 2*26 + 1*26^2 .... -*/ -static int -pseudo_base26 (const char *str) -{ - int i; - int multiplier = 1; - int result = 0; - int len = strlen (str); - - for ( i = len - 1 ; i >= 0; --i) - { - int mantissa = (str[i] - 'A'); - - if ( mantissa < 0 || mantissa > 25 ) - return -1; - - if ( i != len - 1) - mantissa++; - - result += mantissa * multiplier; - - multiplier *= 26; - } - - return result; -} - - - -/* Convert a cell reference in the form "A1:B2", to - integers. A1 means column zero, row zero. - B1 means column 1 row 0. AA1 means column 26, row 0. -*/ -static bool -convert_cell_ref (const char *ref, - int *col0, int *row0, - int *coli, int *rowi) -{ - char startcol[5]; - char stopcol [5]; - - int startrow; - int stoprow; - - int n = sscanf (ref, "%4[a-zA-Z]%d:%4[a-zA-Z]%d", - startcol, &startrow, - stopcol, &stoprow); - if ( n != 4) - return false; - - str_uppercase (startcol); - *col0 = pseudo_base26 (startcol); - str_uppercase (stopcol); - *coli = pseudo_base26 (stopcol); - *row0 = startrow - 1; - *rowi = stoprow - 1 ; - - return true; -} - - enum reader_state { STATE_INIT = 0, /* Initial state */ @@ -159,6 +92,7 @@ struct gnumeric_reader enum reader_state state; int row; int col; + int min_col; int node_type; int sheet_index; @@ -179,9 +113,6 @@ struct gnumeric_reader static void process_node (struct gnumeric_reader *r); -#define _xml(X) (const xmlChar *)(X) - -#define _xmlchar_to_int(X) atoi((const char *)X) static void gnm_file_casereader_destroy (struct casereader *reader UNUSED, void *r_) @@ -217,6 +148,7 @@ process_node (struct gnumeric_reader *r) if (0 == xmlStrcasecmp (name, _xml("gnm:Sheet")) && XML_READER_TYPE_ELEMENT == r->node_type) { + ++r->sheet_index; r->state = STATE_SHEET_START; } break; @@ -226,21 +158,15 @@ process_node (struct gnumeric_reader *r) { r->state = STATE_SHEET_NAME; } - else if (0 == xmlStrcasecmp (name, _xml("gnm:Name")) && - XML_READER_TYPE_END_ELEMENT == r->node_type) - { - r->state = STATE_INIT; - } break; case STATE_SHEET_NAME: if (0 == xmlStrcasecmp (name, _xml("gnm:Name")) && XML_READER_TYPE_END_ELEMENT == r->node_type) { - r->state = STATE_SHEET_START; + r->state = STATE_INIT; } else if (XML_READER_TYPE_TEXT == r->node_type) { - ++r->sheet_index; if ( r->target_sheet != NULL) { xmlChar *value = xmlTextReaderValue (r->xtr); @@ -258,6 +184,7 @@ process_node (struct gnumeric_reader *r) if (0 == xmlStrcasecmp (name, _xml("gnm:Cells")) && XML_READER_TYPE_ELEMENT == r->node_type) { + r->min_col = INT_MAX; if (! xmlTextReaderIsEmptyElement (r->xtr)) r->state = STATE_CELLS_START; } @@ -289,6 +216,9 @@ process_node (struct gnumeric_reader *r) r->col = _xmlchar_to_int (attr); free (attr); + if (r->col < r->min_col) + r->min_col = r->col; + attr = xmlTextReaderGetAttribute (r->xtr, _xml ("Row")); r->row = _xmlchar_to_int (attr); free (attr); @@ -301,7 +231,9 @@ process_node (struct gnumeric_reader *r) case STATE_CELL: if (0 == xmlStrcasecmp (name, _xml("gnm:Cell")) && XML_READER_TYPE_END_ELEMENT == r->node_type) - r->state = STATE_CELLS_START; + { + r->state = STATE_CELLS_START; + } break; default: break; @@ -318,23 +250,19 @@ static void convert_xml_string_to_value (struct ccase *c, const struct variable *var, const xmlChar *xv) { - int n_bytes = 0; union value *v = case_data_rw (c, var); - const char *text = (const char *) xv; - - if ( text) - n_bytes = MIN (var_get_width (var), strlen (text)); - - if ( var_is_alpha (var)) - { - memcpy (value_str_rw (v, var_get_width (var)), text, n_bytes); - } + if (xv == NULL) + value_set_missing (v, var_get_width (var)); + else if ( var_is_alpha (var)) + value_copy_str_rpad (v, var_get_width (var), xv, ' '); else { + const char *text = CHAR_CAST (const char *, xv); char *endptr; + errno = 0; - v->f = strtod (text, &endptr); + v->f = c_strtod (text, &endptr); if ( errno != 0 || endptr == text) v->f = SYSMIS; } @@ -348,7 +276,9 @@ struct var_spec }; struct casereader * -gnumeric_open_reader (struct gnumeric_read_info *gri, struct dictionary **dict) +gnumeric_open_reader (const struct spreadsheet_read_info *gri, + struct spreadsheet_read_options *opts, + struct dictionary **dict) { unsigned long int vstart = 0; int ret; @@ -363,7 +293,7 @@ gnumeric_open_reader (struct gnumeric_read_info *gri, struct dictionary **dict) if ( NULL == gz) { - msg (ME, _("Error opening \"%s\" for reading as a gnumeric file: %s."), + msg (ME, _("Error opening `%s' for reading as a Gnumeric file: %s."), gri->file_name, strerror (errno)); goto error; @@ -371,34 +301,35 @@ gnumeric_open_reader (struct gnumeric_read_info *gri, struct dictionary **dict) r = xzalloc (sizeof *r); - r->xtr = xmlReaderForIO ((xmlInputReadCallback) gzread, gzclose, gz, + r->xtr = xmlReaderForIO ((xmlInputReadCallback) gzread, + (xmlInputCloseCallback) gzclose, gz, NULL, NULL, 0); - if ( r->xtr == NULL) + if ( r->xtr == NULL ) goto error; - if ( gri->cell_range ) + if ( opts->cell_range ) { - if ( ! convert_cell_ref (gri->cell_range, + if ( ! convert_cell_ref (opts->cell_range, &r->start_col, &r->start_row, &r->stop_col, &r->stop_row)) { - msg (SE, _("Invalid cell range \"%s\""), - gri->cell_range); + msg (SE, _("Invalid cell range `%s'"), + opts->cell_range); goto error; } } else { - r->start_col = 0; + r->start_col = -1; r->start_row = 0; r->stop_col = -1; r->stop_row = -1; } r->state = STATE_INIT; - r->target_sheet = BAD_CAST gri->sheet_name; - r->target_sheet_index = gri->sheet_index; + r->target_sheet = BAD_CAST opts->sheet_name; + r->target_sheet_index = opts->sheet_index; r->row = r->col = -1; r->sheet_index = 0; @@ -420,7 +351,7 @@ gnumeric_open_reader (struct gnumeric_read_info *gri, struct dictionary **dict) /* If a range has been given, then use that to calculate the number of cases */ - if ( gri->cell_range) + if ( opts->cell_range) { n_cases = MIN (n_cases, r->stop_row - r->start_row + 1); } @@ -451,17 +382,21 @@ gnumeric_open_reader (struct gnumeric_read_info *gri, struct dictionary **dict) if ( idx >= n_var_specs ) { + int i; + var_spec = xrealloc (var_spec, sizeof (*var_spec) * (idx + 1)); + for (i = n_var_specs; i <= idx; ++i) + { + var_spec [i].name = NULL; + var_spec [i].width = -1; + var_spec [i].first_value = NULL; + } n_var_specs = idx + 1 ; - var_spec = xrealloc (var_spec, sizeof (*var_spec) * n_var_specs); - var_spec [idx].name = NULL; - var_spec [idx].width = -1; - var_spec [idx].first_value = NULL; } if ( r->node_type == XML_READER_TYPE_TEXT ) { xmlChar *value = xmlTextReaderValue (r->xtr); - const char *text = (const char *) value; + const char *text = CHAR_CAST (const char *, value); if ( r->row < r->start_row) { @@ -476,7 +411,7 @@ gnumeric_open_reader (struct gnumeric_read_info *gri, struct dictionary **dict) if (-1 == var_spec [idx].width ) var_spec [idx].width = (gri->asw == -1) ? - ROUND_UP (strlen(text), MAX_SHORT_STRING) : gri->asw; + ROUND_UP (strlen(text), SPREADSHEET_DEFAULT_WIDTH) : gri->asw; } free (value); @@ -497,29 +432,29 @@ gnumeric_open_reader (struct gnumeric_read_info *gri, struct dictionary **dict) } } - - /* Create the dictionary and populate it */ - *dict = r->dict = dict_create (); - - dict_set_encoding (r->dict, (const char *) xmlTextReaderConstEncoding (r->xtr)); + { + const xmlChar *enc = xmlTextReaderConstEncoding (r->xtr); + if ( enc == NULL) + goto error; + /* Create the dictionary and populate it */ + *dict = r->dict = dict_create (CHAR_CAST (const char *, enc)); + } for (i = 0 ; i < n_var_specs ; ++i ) { - char name[VAR_NAME_LEN + 1]; + char *name; + + if ( (var_spec[i].name == NULL) && (var_spec[i].first_value == NULL)) + continue; /* Probably no data exists for this variable, so allocate a default width */ if ( var_spec[i].width == -1 ) - var_spec[i].width = MAX_SHORT_STRING; - - if ( ! dict_make_unique_var_name (r->dict, var_spec[i].name, - &vstart, name)) - { - msg (ME, _("Cannot create variable name from %s"), var_spec[i].name); - goto error; - } + var_spec[i].width = SPREADSHEET_DEFAULT_WIDTH; + name = dict_make_unique_var_name (r->dict, var_spec[i].name, &vstart); dict_create_var (r->dict, name, var_spec[i].width); + free (name); } /* Create the first case, and cache it */ @@ -527,7 +462,7 @@ gnumeric_open_reader (struct gnumeric_read_info *gri, struct dictionary **dict) if ( n_var_specs == 0 ) { - msg (MW, _("Selected sheet or range of spreadsheet \"%s\" is empty."), + msg (MW, _("Selected sheet or range of spreadsheet `%s' is empty."), gri->file_name); goto error; } @@ -536,9 +471,13 @@ gnumeric_open_reader (struct gnumeric_read_info *gri, struct dictionary **dict) r->first_case = case_create (r->proto); case_set_missing (r->first_case); + int x = 0; for ( i = 0 ; i < n_var_specs ; ++i ) { - const struct variable *var = dict_get_var (r->dict, i); + if ( (var_spec[i].name == NULL) && (var_spec[i].first_value == NULL)) + continue; + + const struct variable *var = dict_get_var (r->dict, x++); convert_xml_string_to_value (r->first_case, var, var_spec[i].first_value); @@ -551,7 +490,17 @@ gnumeric_open_reader (struct gnumeric_read_info *gri, struct dictionary **dict) } free (var_spec); - + + + if (opts->cell_range == NULL) + { + opts->cell_range = c_xasprintf ("%c%d:%c%d", + r->start_col + 'A', + r->start_row, + r->stop_col + 'A' + caseproto_get_n_widths (r->proto), + r->start_row + n_cases); + } + return casereader_create_sequential (NULL, r->proto, @@ -568,6 +517,7 @@ gnumeric_open_reader (struct gnumeric_read_info *gri, struct dictionary **dict) free (var_spec); dict_destroy (*dict); + *dict = NULL; gnm_file_casereader_destroy (NULL, r); @@ -595,6 +545,9 @@ gnm_file_casereader_read (struct casereader *reader UNUSED, void *r_) c = case_create (r->proto); case_set_missing (c); + if (r->start_col == -1) + r->start_col = r->min_col; + while ((r->state == STATE_CELL || r->state == STATE_CELLS_START ) && r->row == current_row && (ret = xmlTextReaderRead (r->xtr))) {