X-Git-Url: https://pintos-os.org/cgi-bin/gitweb.cgi?a=blobdiff_plain;f=src%2Fdata%2Fdatasheet.c;h=da85d963c17dfb166ff93ab1e7882f1ee8105d42;hb=a94d5054363ca725a6810780b3621f064956929a;hp=63db4f36d38356293cd93d60b8e40db7408f3519;hpb=c532b2e8445401a2613391cda5ab485f9055b484;p=pspp-builds.git
diff --git a/src/data/datasheet.c b/src/data/datasheet.c
index 63db4f36..da85d963 100644
--- a/src/data/datasheet.c
+++ b/src/data/datasheet.c
@@ -1,20 +1,18 @@
-/* PSPP - computes sample statistics.
+/* PSPP - a program for statistical analysis.
Copyright (C) 2007 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
@@ -26,6 +24,7 @@
#include
#include
#include
+#include
#include
#include
#include
@@ -150,8 +149,7 @@ static bool rw_case (struct datasheet *ds, enum rw_op op,
/* Creates and returns a new datasheet.
If READER is nonnull, then the datasheet initially contains
- the contents of READER. READER become owned by the datasheet
- and the caller must not directly reference it again. */
+ the contents of READER. */
struct datasheet *
datasheet_create (struct casereader *reader)
{
@@ -357,6 +355,8 @@ datasheet_delete_columns (struct datasheet *ds, size_t start, size_t cnt)
{
size_t lcol;
+ assert ( start + cnt <= axis_get_size (ds->columns) );
+
/* Free up columns for reuse. */
for (lcol = start; lcol < start + cnt; lcol++)
{
@@ -429,6 +429,7 @@ bool
datasheet_get_value (const struct datasheet *ds, casenumber row, size_t column,
union value *value, int width)
{
+ assert ( row >= 0 );
return rw_case ((struct datasheet *) ds,
OP_READ, row, column, value_cnt_from_width (width), value);
}
@@ -1258,6 +1259,95 @@ hash_datasheet (const struct datasheet *ds)
return hash[0];
}
+/* Clones the structure and contents of ODS into a new datasheet,
+ and returns the new datasheet. */
+static struct datasheet *
+clone_datasheet (const struct datasheet *ods)
+{
+ struct datasheet *ds;
+ struct range_map_node *r;
+
+ ds = xmalloc (sizeof *ds);
+ ds->columns = axis_clone (ods->columns);
+ ds->rows = axis_clone (ods->rows);
+ range_map_init (&ds->sources);
+ for (r = range_map_first (&ods->sources); r != NULL;
+ r = range_map_next (&ods->sources, r))
+ {
+ const struct source_info *osi = source_info_from_range_map (r);
+ struct source_info *si = xmalloc (sizeof *si);
+ si->source = source_clone (osi->source);
+ range_map_insert (&ds->sources, range_map_node_get_start (r),
+ range_map_node_get_width (r), &si->column_range);
+ }
+ ds->column_min_alloc = ods->column_min_alloc;
+ ds->taint = taint_create ();
+
+ return ds;
+}
+
+/* lazy_casereader callback function to instantiate a casereader
+ from the datasheet. */
+static struct casereader *
+lazy_callback (void *ds_)
+{
+ struct datasheet *ds = ds_;
+ return datasheet_make_reader (ds);
+}
+
+/* Checks that READER contains the ROW_CNT rows and COLUMN_CNT
+ columns of data in ARRAY, reporting any errors via MC. */
+static void
+check_datasheet_casereader (struct mc *mc, struct casereader *reader,
+ double array[MAX_ROWS][MAX_COLS],
+ size_t row_cnt, size_t column_cnt)
+{
+ if (casereader_get_case_cnt (reader) != row_cnt)
+ {
+ if (casereader_get_case_cnt (reader) == CASENUMBER_MAX
+ && casereader_count_cases (reader) == row_cnt)
+ mc_error (mc, "datasheet casereader has unknown case count");
+ else
+ mc_error (mc, "casereader row count (%lu) does not match "
+ "expected (%zu)",
+ (unsigned long int) casereader_get_case_cnt (reader),
+ row_cnt);
+ }
+ else if (casereader_get_value_cnt (reader) != column_cnt)
+ mc_error (mc, "casereader column count (%zu) does not match "
+ "expected (%zu)",
+ casereader_get_value_cnt (reader), column_cnt);
+ else
+ {
+ struct ccase c;
+ size_t row;
+
+ for (row = 0; row < row_cnt; row++)
+ {
+ size_t col;
+
+ if (!casereader_read (reader, &c))
+ {
+ mc_error (mc, "casereader_read failed reading row %zu of %zu "
+ "(%zu columns)", row, row_cnt, column_cnt);
+ return;
+ }
+
+ for (col = 0; col < column_cnt; col++)
+ if (case_num_idx (&c, col) != array[row][col])
+ mc_error (mc, "element %zu,%zu (of %zu,%zu) differs: "
+ "%g != %g",
+ row, col, row_cnt, column_cnt,
+ case_num_idx (&c, col), array[row][col]);
+
+ case_destroy (&c);
+ }
+
+ if (casereader_read (reader, &c))
+ mc_error (mc, "casereader has extra cases (expected %zu)", row_cnt);
+ }
+}
+
/* Checks that datasheet DS contains has ROW_CNT rows, COLUMN_CNT
columns, and the same contents as ARRAY, reporting any
mismatches via mc_error. Then, adds DS to MC as a new state. */
@@ -1266,6 +1356,10 @@ check_datasheet (struct mc *mc, struct datasheet *ds,
double array[MAX_ROWS][MAX_COLS],
size_t row_cnt, size_t column_cnt)
{
+ struct datasheet *ds2;
+ struct casereader *reader;
+ unsigned long int serial = 0;
+
assert (row_cnt < MAX_ROWS);
assert (column_cnt < MAX_COLS);
@@ -1277,12 +1371,13 @@ check_datasheet (struct mc *mc, struct datasheet *ds,
return;
}
+ /* Check contents of datasheet via datasheet functions. */
if (row_cnt != datasheet_get_row_cnt (ds))
mc_error (mc, "row count (%lu) does not match expected (%zu)",
(unsigned long int) datasheet_get_row_cnt (ds), row_cnt);
else if (column_cnt != datasheet_get_column_cnt (ds))
- mc_error (mc, "column count (%lu) does not match expected (%zu)",
- (unsigned long int) datasheet_get_column_cnt (ds), column_cnt);
+ mc_error (mc, "column count (%zu) does not match expected (%zu)",
+ datasheet_get_column_cnt (ds), column_cnt);
else
{
size_t row, col;
@@ -1299,6 +1394,43 @@ check_datasheet (struct mc *mc, struct datasheet *ds,
}
}
+ /* Check that datasheet contents are correct when read through
+ casereader. */
+ ds2 = clone_datasheet (ds);
+ reader = datasheet_make_reader (ds2);
+ check_datasheet_casereader (mc, reader, array, row_cnt, column_cnt);
+ casereader_destroy (reader);
+
+ /* Check that datasheet contents are correct when read through
+ casereader with lazy_casereader wrapped around it. This is
+ valuable because otherwise there is no non-GUI code that
+ uses the lazy_casereader. */
+ ds2 = clone_datasheet (ds);
+ reader = lazy_casereader_create (column_cnt, row_cnt,
+ lazy_callback, ds2, &serial);
+ check_datasheet_casereader (mc, reader, array, row_cnt, column_cnt);
+ if (lazy_casereader_destroy (reader, serial))
+ {
+ /* Lazy casereader was never instantiated. This will
+ only happen if there are no rows (because in that case
+ casereader_read never gets called). */
+ datasheet_destroy (ds2);
+ if (row_cnt != 0)
+ mc_error (mc, "lazy casereader not instantiated, but should "
+ "have been (size %zu,%zu)", row_cnt, column_cnt);
+ }
+ else
+ {
+ /* Lazy casereader was instantiated. This is the common
+ case, in which some casereader operation
+ (casereader_read in this case) was performed on the
+ lazy casereader. */
+ casereader_destroy (reader);
+ if (row_cnt == 0)
+ mc_error (mc, "lazy casereader instantiated, but should not "
+ "have been (size %zu,%zu)", row_cnt, column_cnt);
+ }
+
mc_add_state (mc, ds);
}
@@ -1326,29 +1458,9 @@ extract_data (const struct datasheet *ds, double data[MAX_ROWS][MAX_COLS])
and the contents of ODATA into DATA. */
static void
clone_model (const struct datasheet *ods, double odata[MAX_ROWS][MAX_COLS],
- struct datasheet **ds_, double data[MAX_ROWS][MAX_COLS])
+ struct datasheet **ds, double data[MAX_ROWS][MAX_COLS])
{
- struct datasheet *ds;
- struct range_map_node *r;
-
- /* Clone ODS into DS. */
- ds = *ds_ = xmalloc (sizeof *ds);
- ds->columns = axis_clone (ods->columns);
- ds->rows = axis_clone (ods->rows);
- range_map_init (&ds->sources);
- for (r = range_map_first (&ods->sources); r != NULL;
- r = range_map_next (&ods->sources, r))
- {
- const struct source_info *osi = source_info_from_range_map (r);
- struct source_info *si = xmalloc (sizeof *si);
- si->source = source_clone (osi->source);
- range_map_insert (&ds->sources, range_map_node_get_start (r),
- range_map_node_get_width (r), &si->column_range);
- }
- ds->column_min_alloc = ods->column_min_alloc;
- ds->taint = taint_create ();
-
- /* Clone ODATA into DATA. */
+ *ds = clone_datasheet (ods);
memcpy (data, odata, MAX_ROWS * MAX_COLS * sizeof **data);
}