Обсуждение: [COMMITTERS] pgsql: Identity columns
Identity columns This is the SQL standard-conforming variant of PostgreSQL's serial columns. It fixes a few usability issues that serial columns have: - CREATE TABLE / LIKE copies default but refers to same sequence - cannot add/drop serialness with ALTER TABLE - dropping default does not drop sequence - need to grant separate privileges to sequence - other slight weirdnesses because serial is some kind of special macro Reviewed-by: Vitaly Burovoy <vitaly.burovoy@gmail.com> Branch ------ master Details ------- http://git.postgresql.org/pg/commitdiff/3217327053638085d24dd4d276e7c1f7ac2c4c6b Modified Files -------------- doc/src/sgml/catalogs.sgml | 11 + doc/src/sgml/information_schema.sgml | 11 +- doc/src/sgml/ref/alter_table.sgml | 47 +++- doc/src/sgml/ref/copy.sgml | 7 + doc/src/sgml/ref/create_table.sgml | 65 ++++- doc/src/sgml/ref/insert.sgml | 41 +++ src/backend/access/common/tupdesc.c | 6 + src/backend/catalog/dependency.c | 7 + src/backend/catalog/genbki.pl | 7 +- src/backend/catalog/heap.c | 15 +- src/backend/catalog/index.c | 1 + src/backend/catalog/information_schema.sql | 17 +- src/backend/catalog/pg_depend.c | 52 ++-- src/backend/catalog/sql_features.txt | 12 +- src/backend/commands/sequence.c | 101 +++++-- src/backend/commands/tablecmds.c | 295 ++++++++++++++++++- src/backend/executor/execExpr.c | 12 + src/backend/executor/execExprInterp.c | 23 ++ src/backend/nodes/copyfuncs.c | 23 ++ src/backend/nodes/equalfuncs.c | 18 ++ src/backend/nodes/nodeFuncs.c | 11 + src/backend/nodes/outfuncs.c | 9 + src/backend/nodes/readfuncs.c | 1 + src/backend/parser/analyze.c | 2 + src/backend/parser/gram.y | 134 ++++++++- src/backend/parser/parse_utilcmd.c | 360 +++++++++++++++++++----- src/backend/rewrite/rewriteHandler.c | 56 +++- src/backend/utils/adt/ruleutils.c | 8 + src/backend/utils/cache/lsyscache.c | 32 +++ src/backend/utils/cache/relcache.c | 1 + src/backend/utils/errcodes.txt | 1 + src/bin/pg_dump/pg_dump.c | 95 ++++++- src/bin/pg_dump/pg_dump.h | 3 + src/bin/pg_dump/t/002_pg_dump.pl | 87 ++++++ src/bin/psql/describe.c | 27 +- src/bin/psql/tab-complete.c | 18 +- src/include/catalog/catversion.h | 2 +- src/include/catalog/dependency.h | 8 +- src/include/catalog/pg_attribute.h | 24 +- src/include/catalog/pg_class.h | 2 +- src/include/commands/sequence.h | 2 + src/include/executor/execExpr.h | 8 + src/include/nodes/nodes.h | 1 + src/include/nodes/parsenodes.h | 27 +- src/include/nodes/primnodes.h | 14 + src/include/parser/kwlist.h | 2 + src/include/utils/lsyscache.h | 1 + src/test/regress/expected/create_table_like.out | 47 ++++ src/test/regress/expected/identity.out | 322 +++++++++++++++++++++ src/test/regress/expected/sequence.out | 4 +- src/test/regress/expected/truncate.out | 30 ++ src/test/regress/parallel_schedule | 5 + src/test/regress/serial_schedule | 1 + src/test/regress/sql/create_table_like.sql | 14 + src/test/regress/sql/identity.sql | 192 +++++++++++++ src/test/regress/sql/sequence.sql | 2 +- src/test/regress/sql/truncate.sql | 18 ++ 57 files changed, 2140 insertions(+), 202 deletions(-)
Peter Eisentraut <peter_e@gmx.net> writes: > Identity columns This commit is causing a compiler warning for me: tablecmds.c: In function 'ATExecSetIdentity': tablecmds.c:5936: warning: 'address.objectSubId' may be used uninitialized in this function tablecmds.c:5936: warning: 'address.objectId' may be used uninitialized in this function I'm not sure why it's not complaining about all three fields, because AFAICS, the function returns a totally undefined ObjectAddress when generatedEl is not set. What is the intention there? (If the function were adequately documented, maybe I could divine that for myself, but heaven help the reader who would like to know what this function is supposed to do.) regards, tom lane
On 4/6/17 10:24, Tom Lane wrote: > This commit is causing a compiler warning for me: > > tablecmds.c: In function 'ATExecSetIdentity': > tablecmds.c:5936: warning: 'address.objectSubId' may be used uninitialized in this function > tablecmds.c:5936: warning: 'address.objectId' may be used uninitialized in this function > > I'm not sure why it's not complaining about all three fields, because > AFAICS, the function returns a totally undefined ObjectAddress when > generatedEl is not set. Yeah, that was an oversight. It's depressing that not more compilers warn about an obvious case like this. I have pushed a fix. > What is the intention there? (If the function > were adequately documented, maybe I could divine that for myself, but > heaven help the reader who would like to know what this function is > supposed to do.) Added some comments, too. -- Peter Eisentraut http://www.2ndQuadrant.com/ PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services
On 7 April 2017 at 00:44, Peter Eisentraut <peter_e@gmx.net> wrote: > Identity columns > > This is the SQL standard-conforming variant of PostgreSQL's serial > columns. It fixes a few usability issues that serial columns have: > > - CREATE TABLE / LIKE copies default but refers to same sequence > - cannot add/drop serialness with ALTER TABLE > - dropping default does not drop sequence > - need to grant separate privileges to sequence > - other slight weirdnesses because serial is some kind of special macro Attached is a small patch which fixes up a warning for compilers not smart enough to know the elog(ERROR) does not return. -- David Rowley http://www.2ndQuadrant.com/ PostgreSQL Development, 24x7 Support, Training & Services
Вложения
On 4/6/17 23:13, David Rowley wrote: > On 7 April 2017 at 00:44, Peter Eisentraut <peter_e@gmx.net> wrote: >> Identity columns >> >> This is the SQL standard-conforming variant of PostgreSQL's serial >> columns. It fixes a few usability issues that serial columns have: >> >> - CREATE TABLE / LIKE copies default but refers to same sequence >> - cannot add/drop serialness with ALTER TABLE >> - dropping default does not drop sequence >> - need to grant separate privileges to sequence >> - other slight weirdnesses because serial is some kind of special macro > > Attached is a small patch which fixes up a warning for compilers not > smart enough to know the elog(ERROR) does not return. fixed -- Peter Eisentraut http://www.2ndQuadrant.com/ PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services