/* PSPP - a program for statistical analysis.
- Copyright (C) 2007, 2009, 2010, 2011, 2012, 2013 Free Software Foundation, Inc.
+ Copyright (C) 2007, 2009, 2010, 2011, 2012, 2013, 2016 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
#include "spreadsheet-reader.h"
-#include "c-xvasprintf.h"
-
-#if !GNM_SUPPORT
+#if !GNM_READ_SUPPORT
struct casereader *
-gnumeric_open_reader (struct spreadsheet_read_info *gri, struct spreadsheet_read_options *opts, struct dictionary **dict)
+gnumeric_open_reader (const struct spreadsheet_read_options *opts, struct dictionary **dict)
{
msg (ME, _("Support for %s files was not compiled into this installation of PSPP"), "Gnumeric");
#include <libxml/xmlreader.h>
#include <zlib.h>
+#include "data/format.h"
+#include "data/data-in.h"
#include "data/case.h"
#include "data/casereader-provider.h"
#include "data/dictionary.h"
#include "gl/xalloc.h"
+
+/* Shamelessly lifted from the Gnumeric sources:
+ https://git.gnome.org/browse/gnumeric/tree/src/value.h
+ */
+enum gnm_value_type
+{
+ VALUE_EMPTY = 10,
+ VALUE_BOOLEAN = 20,
+ VALUE_INTEGER = 30, /* Note, this was removed from gnumeric in 2006 - old versions may of
+ course still be around. New ones are supposed to use float.*/
+ VALUE_FLOAT = 40,
+ VALUE_ERROR = 50,
+ VALUE_STRING = 60,
+ VALUE_CELLRANGE = 70,
+ VALUE_ARRAY = 80
+};
+
+
+
static void gnm_file_casereader_destroy (struct casereader *, void *);
static struct ccase *gnm_file_casereader_read (struct casereader *, void *);
+
static const struct casereader_class gnm_file_casereader_class =
{
gnm_file_casereader_read,
struct sheet_detail
{
- xmlChar *name;
+ /* The name of the sheet (utf8 encoding) */
+ char *name;
int start_col;
int stop_col;
int maxrow;
};
-
-struct gnumeric_reader
+struct state_data
{
- struct spreadsheet spreadsheet;
-
/* The libxml reader for this instance */
xmlTextReaderPtr xtr;
/* An internal state variable */
enum reader_state state;
+ int node_type;
+ int current_sheet;
+
int row;
int col;
+
int min_col;
- int node_type;
- int current_sheet;
+};
+
+
+static void
+state_data_destroy (struct state_data *sd)
+{
+ xmlFreeTextReader (sd->xtr);
+}
+
+
+struct gnumeric_reader
+{
+ struct spreadsheet spreadsheet;
+
+ struct state_data rsd;
+ struct state_data msd;
int start_col;
int stop_col;
struct dictionary *dict;
struct ccase *first_case;
bool used_first_case;
+
+ enum gnm_value_type vtype;
};
+void
+gnumeric_unref (struct spreadsheet *s)
+{
+ struct gnumeric_reader *r = (struct gnumeric_reader *) s;
+
+ if (0 == --s->ref_cnt)
+ {
+ int i;
+
+ for (i = 0; i < s->n_sheets; ++i)
+ {
+ xmlFree (r->sheets[i].name);
+ }
+
+
+ free (r->sheets);
+ state_data_destroy (&r->msd);
+
+ dict_destroy (r->dict);
+
+ free (s->file_name);
+
+ free (r);
+ }
+}
+
+
const char *
gnumeric_get_sheet_name (struct spreadsheet *s, int n)
{
struct gnumeric_reader *gr = (struct gnumeric_reader *) s;
assert (n < s->n_sheets);
- return gr->sheets[n].name; // Kludge: Assumes the encoding is utf8
+ return gr->sheets[n].name;
}
-static void process_node (struct gnumeric_reader *r);
+static void process_node (struct gnumeric_reader *r, struct state_data *sd);
+
char *
while (
(gr->sheets[n].stop_col == -1)
&&
- (1 == (ret = xmlTextReaderRead (gr->xtr)))
+ (1 == (ret = xmlTextReaderRead (gr->msd.xtr)))
)
{
- process_node (gr);
+ process_node (gr, &gr->msd);
}
- return create_cell_ref (
+ return create_cell_range (
gr->sheets[n].start_col,
gr->sheets[n].start_row,
gr->sheets[n].stop_col,
gnm_file_casereader_destroy (struct casereader *reader UNUSED, void *r_)
{
struct gnumeric_reader *r = r_;
+
if ( r == NULL)
return ;
- if ( r->xtr)
- xmlFreeTextReader (r->xtr);
- r->xtr = NULL;
+ state_data_destroy (&r->rsd);
- if ( ! r->used_first_case )
+ if (r->first_case && ! r->used_first_case )
case_unref (r->first_case);
- caseproto_unref (r->proto);
+ if (r->proto)
+ caseproto_unref (r->proto);
-#if 0
- for (i = 0; i < r->spreadsheet.n_sheets; ++i)
- {
- xmlFree (r->sheets[i].name);
- }
-
- free (r->sheets);
-
-
- free (r);
-#endif
+ gnumeric_unref (&r->spreadsheet);
}
+
static void
-process_node (struct gnumeric_reader *r)
+process_node (struct gnumeric_reader *r, struct state_data *sd)
{
- xmlChar *name = xmlTextReaderName (r->xtr);
+ xmlChar *name = xmlTextReaderName (sd->xtr);
if (name == NULL)
name = xmlStrdup (_xml ("--"));
+ sd->node_type = xmlTextReaderNodeType (sd->xtr);
- r->node_type = xmlTextReaderNodeType (r->xtr);
-
- switch ( r->state)
+ switch (sd->state)
{
case STATE_PRE_INIT:
- r->current_sheet = -1;
+ sd->current_sheet = -1;
if (0 == xmlStrcasecmp (name, _xml("gnm:SheetNameIndex")) &&
- XML_READER_TYPE_ELEMENT == r->node_type)
+ XML_READER_TYPE_ELEMENT == sd->node_type)
{
- r->state = STATE_SHEET_COUNT;
+ sd->state = STATE_SHEET_COUNT;
}
break;
case STATE_SHEET_COUNT:
if (0 == xmlStrcasecmp (name, _xml("gnm:SheetName")) &&
- XML_READER_TYPE_ELEMENT == r->node_type)
+ XML_READER_TYPE_ELEMENT == sd->node_type)
{
- ++r->current_sheet;
- if (r->current_sheet + 1 > r->spreadsheet.n_sheets)
+ ++sd->current_sheet;
+ if (sd->current_sheet + 1 > r->spreadsheet.n_sheets)
{
- struct sheet_detail *sd ;
- r->sheets = xrealloc (r->sheets, (r->current_sheet + 1) * sizeof *r->sheets);
- sd = &r->sheets[r->current_sheet];
- sd->start_col = sd->stop_col = sd->start_row = sd->stop_row = -1;
- r->spreadsheet.n_sheets = r->current_sheet + 1;
+ struct sheet_detail *detail ;
+ r->sheets = xrealloc (r->sheets, (sd->current_sheet + 1) * sizeof *r->sheets);
+ detail = &r->sheets[sd->current_sheet];
+ detail->start_col = detail->stop_col = detail->start_row = detail->stop_row = -1;
+ detail->name = NULL;
+ r->spreadsheet.n_sheets = sd->current_sheet + 1;
}
}
else if (0 == xmlStrcasecmp (name, _xml("gnm:SheetNameIndex")) &&
- XML_READER_TYPE_END_ELEMENT == r->node_type)
+ XML_READER_TYPE_END_ELEMENT == sd->node_type)
{
- r->state = STATE_INIT;
- r->current_sheet = -1;
+ sd->state = STATE_INIT;
+ sd->current_sheet = -1;
}
- else if (XML_READER_TYPE_TEXT == r->node_type)
+ else if (XML_READER_TYPE_TEXT == sd->node_type)
{
- r->sheets [r->spreadsheet.n_sheets - 1].name = xmlTextReaderValue (r->xtr);
+ if ( r->sheets [r->spreadsheet.n_sheets - 1].name == NULL)
+ r->sheets [r->spreadsheet.n_sheets - 1].name = CHAR_CAST (char *, xmlTextReaderValue (sd->xtr));
}
break;
case STATE_INIT:
if (0 == xmlStrcasecmp (name, _xml("gnm:Sheet")) &&
- XML_READER_TYPE_ELEMENT == r->node_type)
+ XML_READER_TYPE_ELEMENT == sd->node_type)
{
- ++r->current_sheet;
- r->state = STATE_SHEET_START;
+ ++sd->current_sheet;
+ sd->state = STATE_SHEET_START;
}
break;
case STATE_SHEET_START:
if (0 == xmlStrcasecmp (name, _xml("gnm:Name")) &&
- XML_READER_TYPE_ELEMENT == r->node_type)
+ XML_READER_TYPE_ELEMENT == sd->node_type)
{
- r->state = STATE_SHEET_NAME;
+ sd->state = STATE_SHEET_NAME;
}
break;
case STATE_SHEET_NAME:
if (0 == xmlStrcasecmp (name, _xml("gnm:Name")) &&
- XML_READER_TYPE_END_ELEMENT == r->node_type)
+ XML_READER_TYPE_END_ELEMENT == sd->node_type)
{
- r->state = STATE_INIT;
+ sd->state = STATE_INIT;
}
else if (0 == xmlStrcasecmp (name, _xml("gnm:Sheet")) &&
- XML_READER_TYPE_END_ELEMENT == r->node_type)
+ XML_READER_TYPE_END_ELEMENT == sd->node_type)
{
- r->state = STATE_INIT;
+ sd->state = STATE_INIT;
}
- else if (XML_READER_TYPE_TEXT == r->node_type)
+ else if (XML_READER_TYPE_TEXT == sd->node_type)
{
if ( r->target_sheet != NULL)
{
- xmlChar *value = xmlTextReaderValue (r->xtr);
+ xmlChar *value = xmlTextReaderValue (sd->xtr);
if ( 0 == xmlStrcmp (value, r->target_sheet))
- r->state = STATE_SHEET_FOUND;
+ sd->state = STATE_SHEET_FOUND;
free (value);
}
- else if (r->target_sheet_index == r->current_sheet + 1)
+ else if (r->target_sheet_index == sd->current_sheet + 1)
{
- r->state = STATE_SHEET_FOUND;
+ sd->state = STATE_SHEET_FOUND;
}
else if (r->target_sheet_index == -1)
{
- r->state = STATE_SHEET_FOUND;
+ sd->state = STATE_SHEET_FOUND;
}
}
break;
case STATE_SHEET_FOUND:
if (0 == xmlStrcasecmp (name, _xml("gnm:Cells")) &&
- XML_READER_TYPE_ELEMENT == r->node_type)
+ XML_READER_TYPE_ELEMENT == sd->node_type)
{
- r->min_col = INT_MAX;
- if (! xmlTextReaderIsEmptyElement (r->xtr))
- r->state = STATE_CELLS_START;
+ sd->min_col = INT_MAX;
+ if (! xmlTextReaderIsEmptyElement (sd->xtr))
+ sd->state = STATE_CELLS_START;
}
else if (0 == xmlStrcasecmp (name, _xml("gnm:MaxRow")) &&
- XML_READER_TYPE_ELEMENT == r->node_type)
+ XML_READER_TYPE_ELEMENT == sd->node_type)
{
- r->state = STATE_MAXROW;
+ sd->state = STATE_MAXROW;
}
else if (0 == xmlStrcasecmp (name, _xml("gnm:MaxCol")) &&
- XML_READER_TYPE_ELEMENT == r->node_type)
+ XML_READER_TYPE_ELEMENT == sd->node_type)
{
- r->state = STATE_MAXCOL;
+ sd->state = STATE_MAXCOL;
}
else if (0 == xmlStrcasecmp (name, _xml("gnm:Sheet")) &&
- XML_READER_TYPE_END_ELEMENT == r->node_type)
+ XML_READER_TYPE_END_ELEMENT == sd->node_type)
{
- r->state = STATE_INIT;
+ sd->state = STATE_INIT;
}
break;
case STATE_MAXROW:
if (0 == xmlStrcasecmp (name, _xml("gnm:MaxRow")) &&
- XML_READER_TYPE_END_ELEMENT == r->node_type)
+ XML_READER_TYPE_END_ELEMENT == sd->node_type)
{
- r->state = STATE_SHEET_FOUND;
+ sd->state = STATE_SHEET_FOUND;
}
- else if (r->node_type == XML_READER_TYPE_TEXT)
+ else if (sd->node_type == XML_READER_TYPE_TEXT)
{
- xmlChar *value = xmlTextReaderValue (r->xtr);
- r->sheets[r->current_sheet].maxrow = _xmlchar_to_int (value);
+ xmlChar *value = xmlTextReaderValue (sd->xtr);
+ r->sheets[sd->current_sheet].maxrow = _xmlchar_to_int (value);
xmlFree (value);
}
break;
case STATE_MAXCOL:
if (0 == xmlStrcasecmp (name, _xml("gnm:MaxCol")) &&
- XML_READER_TYPE_END_ELEMENT == r->node_type)
+ XML_READER_TYPE_END_ELEMENT == sd->node_type)
{
- r->state = STATE_SHEET_FOUND;
+ sd->state = STATE_SHEET_FOUND;
}
- else if (r->node_type == XML_READER_TYPE_TEXT)
+ else if (sd->node_type == XML_READER_TYPE_TEXT)
{
- xmlChar *value = xmlTextReaderValue (r->xtr);
- r->sheets[r->current_sheet].maxcol = _xmlchar_to_int (value);
+ xmlChar *value = xmlTextReaderValue (sd->xtr);
+ r->sheets[sd->current_sheet].maxcol = _xmlchar_to_int (value);
xmlFree (value);
}
break;
case STATE_CELLS_START:
if (0 == xmlStrcasecmp (name, _xml ("gnm:Cell")) &&
- XML_READER_TYPE_ELEMENT == r->node_type)
+ XML_READER_TYPE_ELEMENT == sd->node_type)
{
xmlChar *attr = NULL;
- attr = xmlTextReaderGetAttribute (r->xtr, _xml ("Col"));
- r->col = _xmlchar_to_int (attr);
+ attr = xmlTextReaderGetAttribute (sd->xtr, _xml ("Col"));
+ sd->col = _xmlchar_to_int (attr);
free (attr);
- if (r->col < r->min_col)
- r->min_col = r->col;
+ if (sd->col < sd->min_col)
+ sd->min_col = sd->col;
- attr = xmlTextReaderGetAttribute (r->xtr, _xml ("Row"));
- r->row = _xmlchar_to_int (attr);
+ attr = xmlTextReaderGetAttribute (sd->xtr, _xml ("Row"));
+ sd->row = _xmlchar_to_int (attr);
free (attr);
- if (r->sheets[r->current_sheet].start_row == -1)
+ if (r->sheets[sd->current_sheet].start_row == -1)
{
- r->sheets[r->current_sheet].start_row = r->row;
+ r->sheets[sd->current_sheet].start_row = sd->row;
}
- if (r->sheets[r->current_sheet].start_col == -1)
+ if (r->sheets[sd->current_sheet].start_col == -1)
{
- r->sheets[r->current_sheet].start_col = r->col;
+ r->sheets[sd->current_sheet].start_col = sd->col;
}
- if (! xmlTextReaderIsEmptyElement (r->xtr))
- r->state = STATE_CELL;
+ if (! xmlTextReaderIsEmptyElement (sd->xtr))
+ sd->state = STATE_CELL;
}
- else if ( (0 == xmlStrcasecmp (name, _xml("gnm:Cells"))) && (XML_READER_TYPE_END_ELEMENT == r->node_type) )
+ else if ( (0 == xmlStrcasecmp (name, _xml("gnm:Cells"))) && (XML_READER_TYPE_END_ELEMENT == sd->node_type) )
{
- r->sheets[r->current_sheet].stop_col = r->col;
- r->sheets[r->current_sheet].stop_row = r->row;
- r->state = STATE_SHEET_NAME;
+ r->sheets[sd->current_sheet].stop_col = sd->col;
+ r->sheets[sd->current_sheet].stop_row = sd->row;
+ sd->state = STATE_SHEET_NAME;
}
break;
case STATE_CELL:
- if (0 == xmlStrcasecmp (name, _xml("gnm:Cell")) && XML_READER_TYPE_END_ELEMENT == r->node_type)
+ if (0 == xmlStrcasecmp (name, _xml("gnm:Cell")) && XML_READER_TYPE_END_ELEMENT == sd->node_type)
{
- r->state = STATE_CELLS_START;
+ sd->state = STATE_CELLS_START;
}
break;
default:
*/
static void
convert_xml_string_to_value (struct ccase *c, const struct variable *var,
- const xmlChar *xv)
+ const xmlChar *xv, enum gnm_value_type type, int col, int row)
{
union value *v = case_data_rw (c, var);
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
+ else if (type == VALUE_FLOAT || type == VALUE_INTEGER)
{
const char *text = CHAR_CAST (const char *, xv);
char *endptr;
if ( errno != 0 || endptr == text)
v->f = SYSMIS;
}
+ else
+ {
+ const char *text = CHAR_CAST (const char *, xv);
+
+ const struct fmt_spec *fmt = var_get_write_format (var);
+
+ char *m = data_in (ss_cstr (text), "UTF-8",
+ fmt->type,
+ v,
+ var_get_width (var),
+ "UTF-8");
+
+ if (m)
+ {
+ char buf [FMT_STRING_LEN_MAX + 1];
+ char *cell = create_cell_ref (col, row);
+
+ msg (MW, _("Cannot convert the value in the spreadsheet cell %s to format (%s): %s"),
+ cell, fmt_to_string (fmt, buf), m);
+ free (cell);
+ }
+ free (m);
+ }
}
struct var_spec
char *name;
int width;
xmlChar *first_value;
+ int first_type;
};
-void
-gnumeric_destroy (struct spreadsheet *s)
-{
- gnm_file_casereader_destroy (NULL, s);
-}
-
-
static void
gnumeric_error_handler (void *ctx, const char *mesg,
UNUSED xmlParserSeverities sev, xmlTextReaderLocatorPtr loc)
{
struct gnumeric_reader *r = ctx;
- msg (MW, _("There was a problem whilst reading the Gnumeric file `%s' (near line %d): `%s'"),
+ msg (MW, _("There was a problem whilst reading the %s file `%s' (near line %d): `%s'"),
+ "Gnumeric",
r->spreadsheet.file_name,
xmlTextReaderLocatorLineNumber (loc),
mesg);
static struct gnumeric_reader *
gnumeric_reopen (struct gnumeric_reader *r, const char *filename, bool show_errors)
{
- int ret;
+ int ret = -1;
+ struct state_data *sd;
xmlTextReaderPtr xtr;
gzFile gz;
assert (r == NULL || filename == NULL);
- if (r && r->xtr)
- xmlFreeTextReader (r->xtr);
-
if (filename)
- gz = gzopen (filename, "r");
+ {
+ gz = gzopen (filename, "r");
+ }
else
- gz = gzopen ( r->spreadsheet.file_name, "r");
+ {
+ gz = gzopen (r->spreadsheet.file_name, "r");
+ }
if (NULL == gz)
return NULL;
{
r = xzalloc (sizeof *r);
r->spreadsheet.n_sheets = -1;
- r->spreadsheet.file_name = filename;
+ r->spreadsheet.file_name = strdup (filename);
+ sd = &r->msd;
+ }
+ else
+ {
+ sd = &r->rsd;
}
if (show_errors)
r->target_sheet = NULL;
r->target_sheet_index = -1;
- r->row = r->col = -1;
- r->state = STATE_PRE_INIT;
- r->xtr = xtr;
+ sd->row = sd->col = -1;
+ sd->state = STATE_PRE_INIT;
+ sd->xtr = xtr;
+ r->spreadsheet.ref_cnt++;
+
/* Advance to the start of the workbook.
This gives us some confidence that we are actually dealing with a gnumeric
spreadsheet.
*/
- while ( (r->state != STATE_INIT )
- && 1 == (ret = xmlTextReaderRead (r->xtr)))
+ while ( (sd->state != STATE_INIT )
+ && 1 == (ret = xmlTextReaderRead (sd->xtr)))
{
- process_node (r);
+ process_node (r, sd);
}
+
if ( ret != 1)
{
/* Does not seem to be a gnumeric file */
- xmlFreeTextReader (r->xtr);
- free (r);
+ gnumeric_unref (&r->spreadsheet);
return NULL;
}
r->spreadsheet.type = SPREADSHEET_GNUMERIC;
+ if (show_errors)
+ {
+ const xmlChar *enc = xmlTextReaderConstEncoding (sd->xtr);
+ xmlCharEncoding xce = xmlParseCharEncoding (CHAR_CAST (const char *, enc));
+
+ if ( XML_CHAR_ENCODING_UTF8 != xce)
+ {
+ /* I have been told that ALL gnumeric files are UTF8 encoded. If that is correct, this
+ can never happen. */
+ msg (MW, _("The gnumeric file `%s' is encoded as %s instead of the usual UTF-8 encoding. "
+ "Any non-ascii characters will be incorrectly imported."),
+ r->spreadsheet.file_name,
+ enc);
+ }
+ }
+
return r;
}
struct casereader *
gnumeric_make_reader (struct spreadsheet *spreadsheet,
- const struct spreadsheet_read_info *gri,
- struct spreadsheet_read_options *opts)
+ const struct spreadsheet_read_options *opts)
{
+ int type = 0;
+ int x = 0;
struct gnumeric_reader *r = NULL;
unsigned long int vstart = 0;
int ret;
r = (struct gnumeric_reader *) (spreadsheet);
- if (r->row != -1)
- r = gnumeric_reopen (r, NULL, true);
+ r = gnumeric_reopen (r, NULL, true);
if ( opts->cell_range )
{
r->target_sheet = BAD_CAST opts->sheet_name;
r->target_sheet_index = opts->sheet_index;
- r->row = r->col = -1;
- r->current_sheet = -1;
+ r->rsd.row = r->rsd.col = -1;
+ r->rsd.current_sheet = -1;
+ r->first_case = NULL;
+ r->proto = NULL;
/* Advance to the start of the cells for the target sheet */
- while ( (r->state != STATE_CELL || r->row < r->start_row )
- && 1 == (ret = xmlTextReaderRead (r->xtr)))
+ while ( (r->rsd.state != STATE_CELL || r->rsd.row < r->start_row )
+ && 1 == (ret = xmlTextReaderRead (r->rsd.xtr)))
{
xmlChar *value ;
- process_node (r);
- value = xmlTextReaderValue (r->xtr);
+ process_node (r, &r->rsd);
+ value = xmlTextReaderValue (r->rsd.xtr);
- if ( r->state == STATE_MAXROW && r->node_type == XML_READER_TYPE_TEXT)
+ if ( r->rsd.state == STATE_MAXROW && r->rsd.node_type == XML_READER_TYPE_TEXT)
{
n_cases = 1 + _xmlchar_to_int (value) ;
}
n_cases = MIN (n_cases, r->stop_row - r->start_row + 1);
}
- if ( gri->read_names )
+ if ( opts->read_names )
{
r->start_row++;
n_cases --;
}
+
/* Read in the first row of cells,
including the headers if read_names was set */
while (
- (( r->state == STATE_CELLS_START && r->row <= r->start_row) || r->state == STATE_CELL )
- && (ret = xmlTextReaderRead (r->xtr))
+ (( r->rsd.state == STATE_CELLS_START && r->rsd.row <= r->start_row) || r->rsd.state == STATE_CELL )
+ && (ret = xmlTextReaderRead (r->rsd.xtr))
)
{
int idx;
- process_node (r);
- if ( r->row > r->start_row ) break;
+ if (r->rsd.state == STATE_CELL && r->rsd.node_type == XML_READER_TYPE_TEXT)
+ {
+ xmlChar *attr =
+ xmlTextReaderGetAttribute (r->rsd.xtr, _xml ("ValueType"));
+
+ type = _xmlchar_to_int (attr);
+
+ xmlFree (attr);
+ }
+
+ process_node (r, &r->rsd);
+
+ if ( r->rsd.row > r->start_row )
+ {
+ xmlChar *attr =
+ xmlTextReaderGetAttribute (r->rsd.xtr, _xml ("ValueType"));
+
+ r->vtype = _xmlchar_to_int (attr);
+
+ xmlFree (attr);
+ break;
+ }
- if ( r->col < r->start_col ||
- (r->stop_col != -1 && r->col > r->stop_col))
+ if ( r->rsd.col < r->start_col ||
+ (r->stop_col != -1 && r->rsd.col > r->stop_col))
continue;
- idx = r->col - r->start_col;
+ idx = r->rsd.col - r->start_col;
if ( idx >= n_var_specs )
{
var_spec [i].name = NULL;
var_spec [i].width = -1;
var_spec [i].first_value = NULL;
+ var_spec [i].first_type = -1;
}
n_var_specs = idx + 1 ;
}
- if ( r->node_type == XML_READER_TYPE_TEXT )
+ var_spec [idx].first_type = type;
+
+ if ( r->rsd.node_type == XML_READER_TYPE_TEXT )
{
- xmlChar *value = xmlTextReaderValue (r->xtr);
+ xmlChar *value = xmlTextReaderValue (r->rsd.xtr);
const char *text = CHAR_CAST (const char *, value);
- if ( r->row < r->start_row)
+ if ( r->rsd.row < r->start_row)
{
- if ( gri->read_names )
+ if ( opts->read_names )
{
var_spec [idx].name = xstrdup (text);
}
var_spec [idx].first_value = xmlStrdup (value);
if (-1 == var_spec [idx].width )
- var_spec [idx].width = (gri->asw == -1) ?
- ROUND_UP (strlen(text), SPREADSHEET_DEFAULT_WIDTH) : gri->asw;
+ var_spec [idx].width = (opts->asw == -1) ?
+ ROUND_UP (strlen(text), SPREADSHEET_DEFAULT_WIDTH) : opts->asw;
}
free (value);
}
- else if ( r->node_type == XML_READER_TYPE_ELEMENT
- && r->state == STATE_CELL)
+ else if ( r->rsd.node_type == XML_READER_TYPE_ELEMENT
+ && r->rsd.state == STATE_CELL)
{
- if ( r->row == r->start_row )
+ if ( r->rsd.row == r->start_row )
{
xmlChar *attr =
- xmlTextReaderGetAttribute (r->xtr, _xml ("ValueType"));
+ xmlTextReaderGetAttribute (r->rsd.xtr, _xml ("ValueType"));
- if ( NULL == attr || 60 != _xmlchar_to_int (attr))
+ if ( NULL == attr || VALUE_STRING != _xmlchar_to_int (attr))
var_spec [idx].width = 0;
free (attr);
}
{
- const xmlChar *enc = xmlTextReaderConstEncoding (r->xtr);
+ const xmlChar *enc = xmlTextReaderConstEncoding (r->rsd.xtr);
if ( enc == NULL)
goto error;
/* Create the dictionary and populate it */
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;
+
if ( (var_spec[i].name == NULL) && (var_spec[i].first_value == NULL))
continue;
- const struct variable *var = dict_get_var (r->dict, x++);
-
+ var = dict_get_var (r->dict, x++);
+
convert_xml_string_to_value (r->first_case, var,
- var_spec[i].first_value);
+ var_spec[i].first_value,
+ var_spec[i].first_type,
+ r->rsd.col + i - 1,
+ r->rsd.row - 1);
}
for ( i = 0 ; i < n_var_specs ; ++i )
free (var_spec);
-
- if (opts->cell_range == NULL)
- {
- opts->cell_range = c_xasprintf ("%c%d:%c%ld",
- 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,
}
free (var_spec);
- dict_destroy (spreadsheet->dict);
- spreadsheet->dict = NULL;
gnm_file_casereader_destroy (NULL, r);
int ret = 0;
struct gnumeric_reader *r = r_;
- int current_row = r->row;
+ int current_row = r->rsd.row;
if ( !r->used_first_case )
{
case_set_missing (c);
if (r->start_col == -1)
- r->start_col = r->min_col;
+ r->start_col = r->rsd.min_col;
+
- while ((r->state == STATE_CELL || r->state == STATE_CELLS_START )
- && r->row == current_row && (ret = xmlTextReaderRead (r->xtr)))
+ while ((r->rsd.state == STATE_CELL || r->rsd.state == STATE_CELLS_START )
+ && r->rsd.row == current_row && (ret = xmlTextReaderRead (r->rsd.xtr)))
{
- process_node (r);
+ process_node (r, &r->rsd);
- if ( r->col < r->start_col || (r->stop_col != -1 &&
- r->col > r->stop_col))
+ if (r->rsd.state == STATE_CELL && r->rsd.node_type == XML_READER_TYPE_ELEMENT)
+ {
+ xmlChar *attr =
+ xmlTextReaderGetAttribute (r->rsd.xtr, _xml ("ValueType"));
+
+ r->vtype = _xmlchar_to_int (attr);
+
+ xmlFree (attr);
+ }
+
+ if ( r->rsd.col < r->start_col || (r->stop_col != -1 &&
+ r->rsd.col > r->stop_col))
continue;
- if ( r->col - r->start_col >= caseproto_get_n_widths (r->proto))
+ if ( r->rsd.col - r->start_col >= caseproto_get_n_widths (r->proto))
continue;
- if ( r->stop_row != -1 && r->row > r->stop_row)
+ if ( r->stop_row != -1 && r->rsd.row > r->stop_row)
break;
- if ( r->node_type == XML_READER_TYPE_TEXT )
- {
- xmlChar *value = xmlTextReaderValue (r->xtr);
-
- const int idx = r->col - r->start_col;
+ if ( r->rsd.node_type == XML_READER_TYPE_TEXT )
+ {
+ xmlChar *value = xmlTextReaderValue (r->rsd.xtr);
+ const int idx = r->rsd.col - r->start_col;
const struct variable *var = dict_get_var (r->dict, idx);
- convert_xml_string_to_value (c, var, value);
+ convert_xml_string_to_value (c, var, value, r->vtype,
+ r->rsd.col, r->rsd.row);
- free (value);
+ xmlFree (value);
}
-
}
if (ret == 1)
}
-#endif /* GNM_SUPPORT */
+#endif /* GNM_READ_SUPPORT */
projects / pspp / blobdiff