WT-2605 Add C tests used in testing joins (#2787)

4 files changed, 416 insertions, 0 deletions

diff --git a/dist/s_string.ok b/dist/s_string.ok
index 48290ad1675..420006ccc6c 100644
--- a/dist/s_string.ok
+++ b/dist/s_string.ok
@@ -698,6 +698,7 @@ ibackup
 icount
 idx
 ifdef's
+iiu
 ikey
 im
 impl
diff --git a/test/csuite/Makefile.am b/test/csuite/Makefile.am
index 6058a05431b..f842bc1316f 100644
--- a/test/csuite/Makefile.am
+++ b/test/csuite/Makefile.am
@@ -13,6 +13,12 @@ noinst_PROGRAMS += test_wt2246_col_append
 test_wt2535_insert_race_SOURCES = wt2535_insert_race/main.c
 noinst_PROGRAMS += test_wt2535_insert_race
 
+test_wt2447_join_main_table_SOURCES = wt2447_join_main_table/main.c
+noinst_PROGRAMS += test_wt2447_join_main_table
+
+test_wt2592_join_schema_SOURCES = wt2592_join_schema/main.c
+noinst_PROGRAMS += test_wt2592_join_schema
+
 # Run this during a "make check" smoke test.
 TESTS = $(noinst_PROGRAMS)
 LOG_COMPILER = $(TEST_WRAPPER)
diff --git a/test/csuite/wt2447_join_main_table/main.c b/test/csuite/wt2447_join_main_table/main.c
new file mode 100644
index 00000000000..a797a4223a2
--- /dev/null
+++ b/test/csuite/wt2447_join_main_table/main.c
@@ -0,0 +1,188 @@
+/*-
+ * Public Domain 2014-2016 MongoDB, Inc.
+ * Public Domain 2008-2014 WiredTiger, Inc.
+ *
+ * This is free and unencumbered software released into the public domain.
+ *
+ * Anyone is free to copy, modify, publish, use, compile, sell, or
+ * distribute this software, either in source code form or as a compiled
+ * binary, for any purpose, commercial or non-commercial, and by any
+ * means.
+ *
+ * In jurisdictions that recognize copyright laws, the author or authors
+ * of this software dedicate any and all copyright interest in the
+ * software to the public domain. We make this dedication for the benefit
+ * of the public at large and to the detriment of our heirs and
+ * successors. We intend this dedication to be an overt act of
+ * relinquishment in perpetuity of all present and future rights to this
+ * software under copyright law.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
+ * IN NO EVENT SHALL THE AUTHORS BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+#include "test_util.h"
+
+/*
+ * JIRA ticket reference: WT-2447
+ *
+ * Test case description: This test case is adapted from the submitted test
+ * program in the JIRA ticket. We create a database of 10,000 entries, with
+ * every key i having pair of values (i, i).  Create indices on both values,
+ * and establish a join: table.v1 >= 5000 AND table.v2 < 5001.  There's a
+ * Bloom filter on v2.  We expect that although we iterate from 5000 to
+ * 10000, we'll only have accesses to the main table for key 5000, as
+ * 5001-10000 will generally not be in the Bloom filter.  For key 5000,
+ * we technically have two accesses to the main table - one occurs when we
+ * see key 5000 is in the Bloom filter, and we need to do a full test, we
+ * make an access to the projection table:tablename(v2), that's just to get
+ * the value of v2, which we'll check by comparison to the cursor at 5001.
+ * That counts as a main table access, and when we see it is satisfied and
+ * return the complete set of values, we'll access the main table with the
+ * full projection (that's the second main table access).
+ *
+ * Failure mode: Before fixes of WT-2447, we saw lots of accesses to the main
+ * table.
+ */
+
+void (*custom_die)(void) = NULL;
+
+static int N_RECORDS = 10000;
+
+static int
+get_stat_total(WT_SESSION *session, WT_CURSOR *jcursor, const char *descmatch,
+    uint64_t *pval)
+{
+	WT_CURSOR *statcursor;
+	char *desc, *valstr;
+	uint64_t val;
+	int ret;
+	bool match;
+
+	match = false;
+	*pval = 0;
+	testutil_check(session->open_cursor(session, "statistics:join", jcursor,
+	    NULL, &statcursor));
+
+	while ((ret = statcursor->next(statcursor)) == 0) {
+		statcursor->get_value(statcursor, &desc, &valstr, &val);
+
+		printf("statistics: %s: %s: %" PRIu64 "\n", desc, valstr, val);
+
+		if (strstr(desc, descmatch) != NULL) {
+			*pval += val;
+			match = true;
+		}
+	}
+	testutil_assert(ret == WT_NOTFOUND);
+	testutil_check(statcursor->close(statcursor));
+	testutil_assert(match);
+	return (0);
+}
+
+int
+main(int argc, char *argv[])
+{
+	TEST_OPTS *opts, _opts;
+	WT_CURSOR *cursor1, *cursor2, *jcursor;
+	WT_ITEM d;
+	WT_SESSION *session;
+	uint64_t maincount;
+	int half, i, j;
+	const char *tablename;
+	char bloom_cfg[128], index1uri[256], index2uri[256], joinuri[256];
+
+	opts = &_opts;
+	memset(opts, 0, sizeof(*opts));
+	testutil_check(testutil_parse_opts(argc, argv, opts));
+	testutil_make_work_dir(opts->home);
+
+	testutil_assert((tablename = strchr(opts->uri, ':')) != 0);
+	tablename++;
+	snprintf(index1uri, sizeof(index1uri), "index:%s:index1", tablename);
+	snprintf(index2uri, sizeof(index2uri), "index:%s:index2", tablename);
+	snprintf(joinuri, sizeof(joinuri), "join:%s", opts->uri);
+
+	testutil_check(wiredtiger_open(opts->home, NULL,
+	    "statistics=(all),create", &opts->conn));
+	testutil_check(
+	    opts->conn->open_session(opts->conn, NULL, NULL, &session));
+
+	testutil_check(session->create(session, opts->uri,
+	    "key_format=i,value_format=iiu,columns=(k,v1,v2,d)"));
+	testutil_check(session->create(session, index1uri, "columns=(v1)"));
+	testutil_check(session->create(session, index2uri, "columns=(v2)"));
+
+	testutil_check(session->open_cursor(session, opts->uri, NULL, NULL,
+	    &cursor1));
+
+	d.size = 4100;
+	d.data = malloc(d.size);
+	memset((char *)d.data, 7, d.size);
+
+	for (i = 0; i < N_RECORDS; ++i)
+	{
+		cursor1->set_key(cursor1, i);
+		cursor1->set_value(cursor1, i, i, &d);
+		testutil_check(cursor1->insert(cursor1));
+	}
+
+	free((void*)d.data);
+
+	testutil_check(opts->conn->close(opts->conn, NULL));
+	testutil_check(wiredtiger_open(opts->home, NULL,
+	    "statistics=(all),create,cache_size=1GB", &opts->conn));
+	testutil_check(opts->conn->open_session(opts->conn, NULL, NULL,
+	    &session));
+
+	testutil_check(session->open_cursor(session, index1uri, NULL, NULL,
+	    &cursor1));
+	testutil_check(session->open_cursor(session, index2uri, NULL, NULL,
+	    &cursor2));
+
+	half = N_RECORDS / 2;
+	cursor1->set_key(cursor1, half);
+	testutil_check(cursor1->search(cursor1));
+
+	cursor2->set_key(cursor2, half + 1);
+	testutil_check(cursor2->search(cursor2));
+
+	sprintf(bloom_cfg, "compare=lt,strategy=bloom,count=%u", half);
+
+	testutil_check(session->open_cursor(session, joinuri, NULL, NULL,
+	    &jcursor));
+	testutil_check(session->join(session, jcursor, cursor1, "compare=ge"));
+	testutil_check(session->join(session, jcursor, cursor2, bloom_cfg));
+
+	/* Expect one value returned */
+	testutil_assert(jcursor->next(jcursor) == 0);
+	i = 0;
+	jcursor->get_key(jcursor, &i);
+	testutil_assert(i == (int)half);
+	i = j = 0;
+	memset(&d, 0, sizeof(d));
+	jcursor->get_value(jcursor, &i, &j, &d);
+	testutil_assert(i == (int)half);
+	testutil_assert(j == (int)half);
+	testutil_assert(d.size == 4100);
+	for (i = 0; i < 4100; i++)
+		testutil_assert(((char *)d.data)[i] == 7);
+
+	testutil_assert(jcursor->next(jcursor) == WT_NOTFOUND);
+
+	/*
+	 * Make sure there have been 2 accesses to the main table,
+	 * explained in the discussion above.
+	 */
+	get_stat_total(session, jcursor, "accesses to the main table",
+	    &maincount);
+	testutil_assert(maincount == 2);
+
+	testutil_cleanup(opts);
+
+	return (0);
+}
diff --git a/test/csuite/wt2592_join_schema/main.c b/test/csuite/wt2592_join_schema/main.c
new file mode 100644
index 00000000000..7a735fd5d94
--- /dev/null
+++ b/test/csuite/wt2592_join_schema/main.c
@@ -0,0 +1,221 @@
+/*-
+ * Public Domain 2014-2016 MongoDB, Inc.
+ * Public Domain 2008-2014 WiredTiger, Inc.
+ *
+ * This is free and unencumbered software released into the public domain.
+ *
+ * Anyone is free to copy, modify, publish, use, compile, sell, or
+ * distribute this software, either in source code form or as a compiled
+ * binary, for any purpose, commercial or non-commercial, and by any
+ * means.
+ *
+ * In jurisdictions that recognize copyright laws, the author or authors
+ * of this software dedicate any and all copyright interest in the
+ * software to the public domain. We make this dedication for the benefit
+ * of the public at large and to the detriment of our heirs and
+ * successors. We intend this dedication to be an overt act of
+ * relinquishment in perpetuity of all present and future rights to this
+ * software under copyright law.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
+ * IN NO EVENT SHALL THE AUTHORS BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+#include "test_util.h"
+
+/*
+ * JIRA ticket reference: WT-2592
+ * Test case description: This is an adaptation of the join parts of
+ * ex_schema.c, but written as a test. Though we have join tests in the
+ * Python test suite, the Python API uses raw mode for cursors, so errors
+ * that are specific to non-raw mode are undetected in Python.
+ * Failure mode: The failure seen in WT-2592 was that no items were returned
+ * by a join.
+ */
+#include <inttypes.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+
+#include <wiredtiger.h>
+
+/* The C struct for the data we are storing in a WiredTiger table. */
+typedef struct {
+	char country[5];
+	uint16_t year;
+	uint64_t population;
+} POP_RECORD;
+
+static POP_RECORD pop_data[] = {
+	{ "AU",  1900,	  4000000 },
+	{ "AU",  1950,	  8267337 },
+	{ "AU",  2000,	 19053186 },
+	{ "CAN", 1900,	  5500000 },
+	{ "CAN", 1950,	 14011422 },
+	{ "CAN", 2000,	 31099561 },
+	{ "UK",  1900,	369000000 },
+	{ "UK",  1950,	 50127000 },
+	{ "UK",  2000,	 59522468 },
+	{ "USA", 1900,	 76212168 },
+	{ "USA", 1950,	150697361 },
+	{ "USA", 2000,	301279593 },
+	{ "", 0, 0 }
+};
+
+void (*custom_die)(void) = NULL;
+
+int
+main(int argc, char *argv[])
+{
+	POP_RECORD *p;
+	TEST_OPTS *opts, _opts;
+	WT_CURSOR *country_cursor, *country_cursor2, *cursor, *join_cursor,
+	    *subjoin_cursor, *year_cursor;
+	WT_SESSION *session;
+	const char *basename, *country;
+	char countryuri[256], joinuri[256], yearuri[256];
+	uint64_t recno, population;
+	uint16_t year;
+	int count, ret;
+
+	opts = &_opts;
+	memset(opts, 0, sizeof(*opts));
+	testutil_check(testutil_parse_opts(argc, argv, opts));
+	testutil_make_work_dir(opts->home);
+
+	testutil_assert((basename = strchr(opts->uri, ':')) != 0);
+	basename++;
+	snprintf(countryuri, sizeof(countryuri), "index:%s:country", basename);
+	snprintf(yearuri, sizeof(yearuri), "index:%s:year", basename);
+	snprintf(joinuri, sizeof(joinuri), "join:%s", opts->uri);
+
+	testutil_check(wiredtiger_open(opts->home, NULL,
+	    "create,cache_size=200M", &opts->conn));
+	testutil_check(
+	    opts->conn->open_session(opts->conn, NULL, NULL, &session));
+	testutil_check(session->create(session, opts->uri,
+	    "key_format=i,"
+	    "value_format=5sHQ,"
+	    "columns=(id,country,year,population)"));
+
+	/* Create an index with a simple key. */
+	testutil_check(session->create(session,
+	    countryuri, "columns=(country)"));
+
+	/* Create an immutable index. */
+	testutil_check(session->create(session,
+	    yearuri, "columns=(year),immutable"));
+
+	/* Insert the records into the table. */
+	testutil_check(session->open_cursor(
+	    session, opts->uri, NULL, "append", &cursor));
+	count = 1;
+	for (p = pop_data; p->year != 0; p++) {
+		cursor->set_key(cursor, count);
+		cursor->set_value(cursor, p->country, p->year, p->population);
+		testutil_check(cursor->insert(cursor));
+		count++;
+	}
+	testutil_check(cursor->close(cursor));
+
+	/* Open cursors needed by the join. */
+	testutil_check(session->open_cursor(session,
+	    joinuri, NULL, NULL, &join_cursor));
+	testutil_check(session->open_cursor(session,
+	    countryuri, NULL, NULL, &country_cursor));
+	testutil_check(session->open_cursor(session,
+	    yearuri, NULL, NULL, &year_cursor));
+
+	/* select values WHERE country == "AU" AND year > 1900 */
+	country_cursor->set_key(country_cursor, "AU\0\0\0");
+	testutil_check(country_cursor->search(country_cursor));
+	testutil_check(session->join(session, join_cursor, country_cursor,
+	    "compare=eq,count=10"));
+	year_cursor->set_key(year_cursor, (uint16_t)1900);
+	testutil_check(year_cursor->search(year_cursor));
+	testutil_check(session->join(session, join_cursor, year_cursor,
+	    "compare=gt,count=10,strategy=bloom"));
+
+	count = 0;
+	/* List the values that are joined */
+	while ((ret = join_cursor->next(join_cursor)) == 0) {
+		testutil_check(join_cursor->get_key(join_cursor, &recno));
+		testutil_check(join_cursor->get_value(join_cursor, &country,
+		    &year, &population));
+		printf("ID %" PRIu64, recno);
+		printf(
+		    ": country %s, year %" PRIu16 ", population %" PRIu64 "\n",
+		    country, year, population);
+		count++;
+	}
+	testutil_assert(ret == WT_NOTFOUND);
+	testutil_assert(count == 2);
+
+	testutil_check(join_cursor->close(join_cursor));
+	testutil_check(year_cursor->close(year_cursor));
+	testutil_check(country_cursor->close(country_cursor));
+
+	/* Open cursors needed by the join. */
+	testutil_check(session->open_cursor(session,
+	    joinuri, NULL, NULL, &join_cursor));
+	testutil_check(session->open_cursor(session,
+	    joinuri, NULL, NULL, &subjoin_cursor));
+	testutil_check(session->open_cursor(session,
+	    countryuri, NULL, NULL, &country_cursor));
+	testutil_check(session->open_cursor(session,
+	    countryuri, NULL, NULL, &country_cursor2));
+	testutil_check(session->open_cursor(session,
+	    yearuri, NULL, NULL, &year_cursor));
+
+	/*
+	 * select values WHERE (country == "AU" OR country == "UK")
+	 *                     AND year > 1900
+	 *
+	 * First, set up the join representing the country clause.
+	 */
+	country_cursor->set_key(country_cursor, "AU\0\0\0");
+	testutil_check(country_cursor->search(country_cursor));
+	testutil_check(session->join(session, subjoin_cursor, country_cursor,
+	    "operation=or,compare=eq,count=10"));
+	country_cursor2->set_key(country_cursor2, "UK\0\0\0");
+	testutil_check(country_cursor2->search(country_cursor2));
+	testutil_check(session->join(session, subjoin_cursor, country_cursor2,
+	    "operation=or,compare=eq,count=10"));
+
+	/* Join that to the top join, and add the year clause */
+	testutil_check(session->join(session, join_cursor, subjoin_cursor,
+	    NULL));
+	year_cursor->set_key(year_cursor, (uint16_t)1900);
+	testutil_check(year_cursor->search(year_cursor));
+	testutil_check(session->join(session, join_cursor, year_cursor,
+	    "compare=gt,count=10,strategy=bloom"));
+
+	count = 0;
+	/* List the values that are joined */
+	while ((ret = join_cursor->next(join_cursor)) == 0) {
+		testutil_check(join_cursor->get_key(join_cursor, &recno));
+		testutil_check(join_cursor->get_value(join_cursor, &country,
+		    &year, &population));
+		printf("ID %" PRIu64, recno);
+		printf(
+		    ": country %s, year %" PRIu16 ", population %" PRIu64 "\n",
+		    country, year, population);
+		count++;
+	}
+	testutil_assert(ret == WT_NOTFOUND);
+	testutil_assert(count == 4);
+
+	testutil_check(join_cursor->close(join_cursor));
+	testutil_check(subjoin_cursor->close(subjoin_cursor));
+	testutil_check(country_cursor->close(country_cursor));
+	testutil_check(country_cursor2->close(country_cursor2));
+	testutil_check(year_cursor->close(year_cursor));
+	testutil_check(session->close(session, NULL));
+
+	testutil_cleanup(opts);
+	return (EXIT_SUCCESS);
+}