/* PSPP - a program for statistical analysis.
- Copyright (C) 2011, 2012, 2013 Free Software Foundation, Inc.
+ Copyright (C) 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 <config.h>
-#include "libpspp/message.h"
-#include "libpspp/misc.h"
-#include "libpspp/assertion.h"
-
-#include "data/data-in.h"
-
-#include "gl/c-strtod.h"
-#include "gl/minmax.h"
-
-#include "gettext.h"
-#define _(msgid) gettext (msgid)
-#define N_(msgid) (msgid)
-
#include "ods-reader.h"
#include "spreadsheet-reader.h"
-#if !ODF_READ_SUPPORT
-
-struct casereader *
-ods_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"), "OpenDocument");
-
- return NULL;
-}
-
-#else
-
-#include "libpspp/zip-reader.h"
-
-
#include <assert.h>
#include <stdbool.h>
#include <errno.h>
#include <libxml/xmlreader.h>
#include <zlib.h>
-#include "data/format.h"
#include "data/case.h"
#include "data/casereader-provider.h"
+#include "data/data-in.h"
#include "data/dictionary.h"
+#include "data/format.h"
#include "data/identifier.h"
#include "data/value.h"
#include "data/variable.h"
+#include "libpspp/assertion.h"
#include "libpspp/i18n.h"
+#include "libpspp/message.h"
+#include "libpspp/misc.h"
#include "libpspp/str.h"
+#include "libpspp/zip-reader.h"
+#include "gl/c-strtod.h"
+#include "gl/minmax.h"
#include "gl/xalloc.h"
+#include "gettext.h"
+#define _(msgid) gettext (msgid)
+#define N_(msgid) (msgid)
+
static void ods_file_casereader_destroy (struct casereader *, void *);
static struct ccase *ods_file_casereader_read (struct casereader *, void *);
int col_span;
};
-struct ods_reader /* */
+static void
+state_data_destroy (struct state_data *sd)
+{
+ xmlFree (sd->current_sheet_name);
+ sd->current_sheet_name = NULL;
+
+ xmlFreeTextReader (sd->xtr);
+ sd->xtr = NULL;
+}
+
+struct ods_reader
{
struct spreadsheet spreadsheet;
struct zip_reader *zreader;
- int ref_cnt;
+
int target_sheet_index;
xmlChar *target_sheet_name;
-
- struct state_data foo;
+
+ /* State data for the meta data */
+ struct state_data msd;
+
+ /* State data for the reader */
struct state_data rsd;
int start_row;
bool read_names;
struct string ods_errs;
+
+ struct string zip_errs;
};
void
-ods_destroy (struct spreadsheet *s)
+ods_unref (struct spreadsheet *s)
{
-#if 0
struct ods_reader *r = (struct ods_reader *) s;
- if (--r->ref_cnt == 0)
+ if (--s->ref_cnt == 0)
{
int i;
+ state_data_destroy (&r->msd);
for (i = 0; i < r->n_allocated_sheets; ++i)
{
xmlFree (r->sheets[i].name);
}
+ dict_unref (r->dict);
+
zip_reader_destroy (r->zreader);
free (r->sheets);
+ free (s->file_name);
free (r);
}
-#endif
}
static bool
-reading_target_sheet (const struct ods_reader *r, const struct state_data *foo)
+reading_target_sheet (const struct ods_reader *r, const struct state_data *msd)
{
if (r->target_sheet_name != NULL)
{
- if ( 0 == xmlStrcmp (r->target_sheet_name, foo->current_sheet_name))
+ if ( 0 == xmlStrcmp (r->target_sheet_name, msd->current_sheet_name))
return true;
}
-
- if (r->target_sheet_index == foo->current_sheet + 1)
+
+ if (r->target_sheet_index == msd->current_sheet + 1)
return true;
return false;
ods_get_sheet_name (struct spreadsheet *s, int n)
{
struct ods_reader *r = (struct ods_reader *) s;
- struct state_data *or = &r->foo;
+ struct state_data *or = &r->msd;
assert (n < s->n_sheets);
- while (
+ while (
(r->n_allocated_sheets <= n)
|| or->state != STATE_SPREADSHEET
)
ods_get_sheet_range (struct spreadsheet *s, int n)
{
struct ods_reader *r = (struct ods_reader *) s;
- struct state_data *or = &r->foo;
-
+ struct state_data *or = &r->msd;
+
assert (n < s->n_sheets);
- while (
+ while (
(r->n_allocated_sheets <= n)
- || (r->sheets[n].stop_row == -1)
+ || (r->sheets[n].stop_row == -1)
|| or->state != STATE_SPREADSHEET
)
{
process_node (r, or);
}
- return create_cell_ref (
+ return create_cell_range (
r->sheets[n].start_col,
r->sheets[n].start_row,
r->sheets[n].stop_col,
static void
ods_file_casereader_destroy (struct casereader *reader UNUSED, void *r_)
{
-#if 0
struct ods_reader *r = r_;
if ( r == NULL)
return ;
- if (r->xtr)
- xmlFreeTextReader (r->xtr);
- r->xtr = NULL;
+ state_data_destroy (&r->rsd);
if ( ! ds_is_empty (&r->ods_errs))
msg (ME, "%s", ds_cstr (&r->ods_errs));
ds_destroy (&r->ods_errs);
- if ( ! r->used_first_case )
+ if ( r->first_case && ! r->used_first_case )
case_unref (r->first_case);
+
caseproto_unref (r->proto);
+ r->proto = NULL;
- xmlFree (r->current_sheet_name);
xmlFree (r->target_sheet_name);
+ r->target_sheet_name = NULL;
+
- ods_destroy (&r->spreadsheet);
-#endif
+ ods_unref (&r->spreadsheet);
}
break;
case STATE_SPREADSHEET:
if (0 == xmlStrcasecmp (name, _xml("table:table"))
- &&
+ &&
(XML_READER_TYPE_ELEMENT == r->node_type))
{
xmlFree (r->current_sheet_name);
}
break;
case STATE_TABLE:
- if (0 == xmlStrcasecmp (name, _xml("table:table-row")) &&
+ if (0 == xmlStrcasecmp (name, _xml("table:table-row")) &&
(XML_READER_TYPE_ELEMENT == r->node_type))
{
xmlChar *value =
xmlTextReaderGetAttribute (r->xtr,
_xml ("table:number-rows-repeated"));
-
+
int row_span = value ? _xmlchar_to_int (value) : 1;
r->row += row_span;
r->col = 0;
-
+
if (! xmlTextReaderIsEmptyElement (r->xtr))
r->state = STATE_ROW;
xmlFree (value);
}
- else if (0 == xmlStrcasecmp (name, _xml("table:table")) &&
+ else if (0 == xmlStrcasecmp (name, _xml("table:table")) &&
(XML_READER_TYPE_END_ELEMENT == r->node_type))
{
r->state = STATE_SPREADSHEET;
break;
case STATE_ROW:
if ( (0 == xmlStrcasecmp (name, _xml ("table:table-cell")))
- &&
+ &&
(XML_READER_TYPE_ELEMENT == r->node_type))
{
xmlChar *value =
xmlTextReaderGetAttribute (r->xtr,
_xml ("table:number-columns-repeated"));
-
+
r->col_span = value ? _xmlchar_to_int (value) : 1;
r->col += r->col_span;
if (or->sheets[r->current_sheet].start_row == -1)
or->sheets[r->current_sheet].start_row = r->row - 1;
- if (
+ if (
(or->sheets[r->current_sheet].start_col == -1)
||
(or->sheets[r->current_sheet].start_col >= r->col - 1)
or->sheets[r->current_sheet].stop_row = r->row - 1;
- if ( or->sheets[r->current_sheet].stop_col < r->col - 1)
+ if ( or->sheets[r->current_sheet].stop_col < r->col - 1)
or->sheets[r->current_sheet].stop_col = r->col - 1;
if (XML_READER_TYPE_END_ELEMENT == r->node_type)
xmlFree (name);
}
-/*
- A struct containing the parameters of a cell's value
+/*
+ A struct containing the parameters of a cell's value
parsed from the xml
*/
struct xml_value
*/
static void
convert_xml_to_value (struct ccase *c, const struct variable *var,
- const struct xml_value *xmv)
+ const struct xml_value *xmv, int col, int row)
{
union value *v = case_data_rw (c, var);
const char *text = xmv->value ?
CHAR_CAST (const char *, xmv->value) : CHAR_CAST (const char *, xmv->text);
-
- data_in (ss_cstr (text), "UTF-8",
+ 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);
}
}
}
+static int
+xml_reader_for_zip_member (void *zm_, char *buffer, int len)
+{
+ struct zip_member *zm = zm_;
+ return zip_member_read (zm, buffer, len);
+}
/* Try to find out how many sheets there are in the "workbook" */
static int
if ( meta == NULL)
return -1;
- mxtr = xmlReaderForIO ((xmlInputReadCallback) zip_member_read,
- (xmlInputCloseCallback) NULL,
- meta, NULL, NULL, 0);
+ mxtr = xmlReaderForIO (xml_reader_for_zip_member, NULL, meta, NULL, NULL, 0);
while (1 == xmlTextReaderRead (mxtr))
{
int s = _xmlchar_to_int (attr);
xmlFreeTextReader (mxtr);
xmlFree (name);
- xmlFree (attr);
+ xmlFree (attr);
return s;
}
- xmlFree (attr);
+ xmlFree (attr);
}
- xmlFree (name);
+ xmlFree (name);
}
xmlFreeTextReader (mxtr);
UNUSED xmlParserSeverities sev, xmlTextReaderLocatorPtr loc)
{
struct ods_reader *r = ctx;
-
+
msg (MW, _("There was a problem whilst reading the %s file `%s' (near line %d): `%s'"),
"ODF",
r->spreadsheet.file_name,
if ( content == NULL)
return NULL;
- zip_member_ref (content);
- xtr = xmlReaderForIO ((xmlInputReadCallback) zip_member_read,
- (xmlInputCloseCallback) NULL,
- content, NULL, NULL,
+ xtr = xmlReaderForIO (xml_reader_for_zip_member, NULL, content, NULL, NULL,
report_errors ? 0 : (XML_PARSE_NOERROR | XML_PARSE_NOWARNING) );
if ( xtr == NULL)
r->spreadsheet.type = SPREADSHEET_ODS;
- if (report_errors)
+ if (report_errors)
xmlTextReaderSetErrorHandler (xtr, ods_error_handler, r);
return xtr;
}
+
struct spreadsheet *
ods_probe (const char *filename, bool report_errors)
{
- struct ods_reader *r;
- struct string errs = DS_EMPTY_INITIALIZER;
int sheet_count;
- struct zip_reader *zr = zip_reader_create (filename, &errs);
+ struct ods_reader *r = xzalloc (sizeof *r);
xmlTextReaderPtr xtr;
+ struct zip_reader *zr;
+
+ ds_init_empty (&r->zip_errs);
+
+ zr = zip_reader_create (filename, &r->zip_errs);
if (zr == NULL)
{
if (report_errors)
{
msg (ME, _("Cannot open %s as a OpenDocument file: %s"),
- filename, ds_cstr (&errs));
+ filename, ds_cstr (&r->zip_errs));
}
+ ds_destroy (&r->zip_errs);
+ free (r);
return NULL;
}
sheet_count = get_sheet_count (zr);
- r = xzalloc (sizeof *r);
r->zreader = zr;
- r->ref_cnt = 1;
+ r->spreadsheet.ref_cnt = 1;
xtr = init_reader (r, report_errors);
if (xtr == NULL)
{
goto error;
}
- r->foo.xtr = xtr;
- r->foo.row = 0;
- r->foo.col = 0;
- r->foo.current_sheet = 0;
- r->foo.state = STATE_INIT;
+ r->msd.xtr = xtr;
+ r->msd.row = 0;
+ r->msd.col = 0;
+ r->msd.current_sheet = 0;
+ r->msd.state = STATE_INIT;
r->spreadsheet.n_sheets = sheet_count;
r->n_allocated_sheets = 0;
r->sheets = NULL;
- ds_destroy (&errs);
-
- r->spreadsheet.file_name = filename;
+ r->spreadsheet.file_name = strdup (filename);
return &r->spreadsheet;
error:
+ ds_destroy (&r->zip_errs);
zip_reader_destroy (r->zreader);
- ds_destroy (&errs);
free (r);
return NULL;
}
struct casereader *
-ods_make_reader (struct spreadsheet *spreadsheet,
+ods_make_reader (struct spreadsheet *spreadsheet,
const struct spreadsheet_read_options *opts)
{
intf ret = 0;
assert (r);
r->read_names = opts->read_names;
ds_init_empty (&r->ods_errs);
- ++r->ref_cnt;
+ ++r->spreadsheet.ref_cnt;
xtr = init_reader (r, true);
if ( xtr == NULL)
r->rsd.current_sheet = 0;
r->rsd.state = STATE_INIT;
+ r->used_first_case = false;
+ r->first_case = NULL;
if (opts->cell_range)
{
r->target_sheet_index = opts->sheet_index;
/* Advance to the start of the cells for the target sheet */
- while ( ! reading_target_sheet (r, &r->rsd)
+ while ( ! reading_target_sheet (r, &r->rsd)
|| r->rsd.state != STATE_ROW || r->rsd.row <= r->start_row )
{
if (1 != (ret = xmlTextReaderRead (r->rsd.xtr)))
if ( opts->read_names)
{
- while (1 == (ret = xmlTextReaderRead (r->rsd.xtr)))
+ while (1 == xmlTextReaderRead (r->rsd.xtr))
{
- int idx;
-
process_node (r, &r->rsd);
/* If the row is finished then stop for now */
if (r->rsd.state == STATE_TABLE && r->rsd.row > r->start_row)
break;
- idx = r->rsd.col - r->start_col -1 ;
+ int idx = r->rsd.col - r->start_col - 1;
- if ( idx < 0)
+ if (idx < 0)
continue;
if (r->stop_col != -1 && idx > r->stop_col - r->start_col)
continue;
- if (r->rsd.state == STATE_CELL_CONTENT
+ if (r->rsd.state == STATE_CELL_CONTENT
&&
XML_READER_TYPE_TEXT == r->rsd.node_type)
{
xmlChar *value = xmlTextReaderValue (r->rsd.xtr);
-
- if ( idx >= n_var_specs)
+ if (idx >= n_var_specs)
{
var_spec = xrealloc (var_spec, sizeof (*var_spec) * (idx + 1));
/* xrealloc (unlike realloc) doesn't initialise its memory to 0 */
memset (var_spec + n_var_specs,
- 0,
+ 0,
(idx - n_var_specs + 1) * sizeof (*var_spec));
n_var_specs = idx + 1;
}
- var_spec[idx].firstval.text = 0;
- var_spec[idx].firstval.value = 0;
- var_spec[idx].firstval.type = 0;
-
- var_spec [idx].name = strdup (CHAR_CAST (const char *, value));
+ for (int i = 0; i < r->rsd.col_span; ++i)
+ {
+ var_spec[idx - i].firstval.text = 0;
+ var_spec[idx - i].firstval.value = 0;
+ var_spec[idx - i].firstval.type = 0;
+ var_spec[idx - i].name =
+ strdup (CHAR_CAST (const char *, value));
+ }
xmlFree (value);
}
{
var_spec = xrealloc (var_spec, sizeof (*var_spec) * (idx + 1));
memset (var_spec + n_var_specs,
- 0,
+ 0,
(idx - n_var_specs + 1) * sizeof (*var_spec));
var_spec [idx].name = NULL;
var_set_both_formats (var, &fmt);
}
- /* Create the first case, and cache it */
- r->used_first_case = false;
-
if ( n_var_specs == 0 )
{
msg (MW, _("Selected sheet or range of spreadsheet `%s' is empty."),
goto error;
}
+ /* Create the first case, and cache it */
r->proto = caseproto_ref (dict_get_proto (r->dict));
r->first_case = case_create (r->proto);
case_set_missing (r->first_case);
{
const struct variable *var = dict_get_var (r->dict, i);
- convert_xml_to_value (r->first_case, var, &var_spec[i].firstval);
+ convert_xml_to_value (r->first_case, var, &var_spec[i].firstval,
+ r->rsd.col - n_var_specs + i,
+ r->rsd.row - 1);
}
/* Read in the first row of data */
&ods_file_casereader_class, r);
error:
-
+
for ( i = 0 ; i < n_var_specs ; ++i )
{
free (var_spec[i].firstval.type);
free (var_spec);
- dict_destroy (r->spreadsheet.dict);
- r->spreadsheet.dict = NULL;
ods_file_casereader_destroy (NULL, r);
return NULL;
struct ccase *c = NULL;
struct ods_reader *r = r_;
+ xmlChar *val_string = NULL;
+ xmlChar *type = NULL;
+
if (!r->used_first_case)
{
r->used_first_case = true;
/* Advance to the start of a row. (If there is one) */
- while (r->rsd.state != STATE_ROW
+ while (r->rsd.state != STATE_ROW
&& 1 == xmlTextReaderRead (r->rsd.xtr)
)
{
}
- if ( ! reading_target_sheet (r, &r->rsd)
+ if ( ! reading_target_sheet (r, &r->rsd)
|| r->rsd.state < STATE_TABLE
|| (r->stop_row != -1 && r->rsd.row > r->stop_row + 1)
)
c = case_create (r->proto);
case_set_missing (c);
-
- xmlChar *val_string = NULL;
- xmlChar *type = NULL;
while (1 == xmlTextReaderRead (r->rsd.xtr))
{
val_string = xmlTextReaderGetAttribute (r->rsd.xtr, _xml ("office:value"));
}
- if (r->rsd.state == STATE_CELL_CONTENT &&
+ if (r->rsd.state == STATE_CELL_CONTENT &&
r->rsd.node_type == XML_READER_TYPE_TEXT)
{
int col;
struct xml_value *xmv = xzalloc (sizeof *xmv);
xmv->text = xmlTextReaderValue (r->rsd.xtr);
- xmv->value = val_string;
- xmv->type = type;
+ xmv->value = val_string;
val_string = NULL;
+ xmv->type = type;
+ type = NULL;
for (col = 0; col < r->rsd.col_span; ++col)
{
break;
var = dict_get_var (r->dict, idx);
- convert_xml_to_value (c, var, xmv);
+ convert_xml_to_value (c, var, xmv, idx + r->start_col, r->rsd.row - 1);
}
xmlFree (xmv->text);
break;
}
+ xmlFree (type);
+ xmlFree (val_string);
+
return c;
}
-#endif
projects / pspp / blobdiff