Skip to content

G
Gpdb

Greenplum / Gpdb

通知 7
Star 1
Fork 0
G
Gpdb

体验新版 GitCode，发现更多精彩内容 >>

切换分支/标签
  1. 02 8月, 2019 4 次提交
    • B
      Support GIN Indexes with ORCA. · eae823f4
      由 Bhuvnesh Chaudhary 提交于 
      This commit adds the GPDB side changes required to support GIN Indexes with
ORCA.  It also adds a new test file gp_gin_indexes to test plans produced for
ORCA/planner.

GIN indexes are not supported with index expression or predicate constraints.
ORCA does not support it currently for other types of indexes too.
      eae823f4
    • I
      Unify backend/access/gin unittest infrastructure with other unit tests (#8275) · fc88b4ee
      由 Ivan Leskin 提交于 
      There is a pipeline for unit tests in GPDB used in most cases.

However, unit tests of src/backend/access/gin introduced by 99360f54
used a custom implementation of unit test build script. This led to
errors e.g. when a compiler different than GCC was used to build GPDB.

Rewrite Makefile in order to unify test infrastructure with common
pattern used in the backend while retaining test isolation from the
backend objects.

See also similar Makefile: src/backend/catalog/test/Makefile
at 122c79f2

Note the Makefile in src/backend/access/gin/test is different from
currently most used version of a backend unit test Makefile. These
differences and motivation for them is described in the README.

Run pgindent on ginpostlist_fakes.c
Reviewed-by: NAdam Berlin <aberlin@pivotal.io>
Reviewed-by: NDaniel Gustafsson <dgustafsson@pivotal.io>
Reviewed-by: NHeikki Linnakangas <hlinnakangas@pivotal.io>
Co-authored-by: NIvan Leskin <leskin.in@arenadata.io>
      fc88b4ee
    • I
      pg_upgrade: check no dependencies on partition array types exist (#8274) · 219f3a50
      由 Ivan Leskin 提交于 
      The following SQL snippet demonstrates dependencies on array types derived from partitions of partitioned tables:

```
CREATE TABLE table_part(a INT, b INT) PARTITION BY RANGE(b) (PARTITION part1 START(0) END(42));
CREATE TABLE replacement(LIKE table_part);
ALTER TABLE table_part EXCHANGE PARTITION part1 WITH TABLE replacement;
CREATE TABLE dependant(d table_part_1_prt_part1[]);
```

Such dependencies cannot be carried over to the upgraded cluster, and pg_upgrade should fail when they are present. The user should take extra steps to resolve the situation.

When check fails, the output is similar to the following:

```
Checking array types derived from partitions                fatal

Array types derived from partitions of a partitioned table must not have dependants.
OIDs of such types found and their original partitions:
16393 public.table_part_1_prt_part1
16471 public.table_part2_1_prt_part1
Failure, exiting
```
Reviewed-by: NSoumyadeep Chakraborty <sochakraborty@pivotal.io>
Co-authored-by: NIvan Leskin <leskin.in@arenadata.io>
      219f3a50
    • D
      Fix aocs table block version mismatch (#8202) · 41fd823a
      由 David Kimura 提交于 
      ALTER TABLE DROP COLUMN followed by reorganize leads to loss of column
encoding settings of the dropped column. When the column's compresstype
encoding is incorrect, we can encounter block version mismatch error
later during block info validation check of the dropped column.

One idea was to skip dropped columns when constructing AOCSScanDesc.
However, dropping all columns is a special case that is not easily handled
because it is not equivalent to deleted rows. Instead, the fix is to preserve
column encoding settings even for dropped columns.
Co-authored-by: NSoumyadeep Chakraborty <sochakraborty@pivotal.io>
Co-authored-by: NIvan Leskin <leskin.in@arenadata.io>
      41fd823a
  3. 01 8月, 2019 8 次提交
  5. 31 7月, 2019 18 次提交
  7. 30 7月, 2019 5 次提交
  9. 29 7月, 2019 5 次提交
    • D
      Add two more duplicates to mailmap · 188b2801
      由 Daniel Gustafsson 提交于 
      Fixing a small bug in my regexp searching for duplicates yielded two
additional hits. Fix by adding to the mailmap file.
      188b2801
    • D
      Fix duplicate function prototype · f722631e
      由 Daniel Gustafsson 提交于 
      It seems that the duplicate prototype snuck in during the 8.4 merge,
which should've been mostly a noop as I had synced pgcrypto with 8.4
in 16d26ee9. I had however placed the prototype on the wrong line
and so git merge was able to cleanly place the protype a few lines
down, creating a duplicate.
      f722631e
    • D
      Remove unused string substitutions · 57dfe4d2
      由 Daniel Gustafsson 提交于 
      The gpstringsubs program replaces tokens in input .source test files
when creating output .sql/.out files for pg_regress. There were no
more callsites for the gp_glob_connect token however so remove that,
and also remove tests for gpwhich_<binary> tokens which aren't in
use in tests and thus might start failing randomly, which is exactly
what happened in another patch where Perl programs were no longer
incorrectly installed in PATH. Without this, removing the Perl test
apps from PATH results in errors like this:

	which: no get_ereport.pl in ( .. )
Reviewed-by: NHeikki Linnakangas <hlinnakangas@pivotal.io>
      57dfe4d2
    • X
      Disbale test cases of gpfdist_ssl · a42f430b
      由 Xiaoran Wang 提交于 
      The test cases of gpfdist_ssl sometimes
fail on pipeline. We don't know the root
cause yet. So just disable them. And we will
enable them after fixing.
      a42f430b
    • C
      Fix quirk of isolation2 test (#8245) · 61325627
      由 Chen Mulong 提交于 
      * Fix quirk of isolation2 test

- Fix the line continuation behaviour and document
Instead of matching "; *$" to detect the end of a sql end, the original
code was using ";$". This creates a totally different output for tests
like:
a)
SELECT * FROM t1;
SELECT * FROM t2;
b)
SELECT * FROM t1;<space>
SELECT * FROM t2;
For a), two SELECT statements will be sent separately.
For b), they will be sent together as
        "SELECT * FROM t1; SELECT * FROM t2;"

Spaces are playing too much magic here, some of them could be a bug in
the test itself. For example, the oom_query test should NOT pass since
it misses error output in the expected.

This change unifies the behaviours by matching with "; *$". So if a line
is end with ';', no matter spaces follows it or not, the statement will
be sent at once.

- Document the known issue of multi SQL in one line.

This issue was found during debugging pipeline failure of #8211
Authored-by: NChen Mulong <muchen@pivotal.io>
      61325627