IMPALA-4168: [DOCS] Adds Oracle-style hint placement for INSERT/UPSERT Change-Id: I43e0a782087c2e67f2e012424fb9261be445efc9 Reviewed-on: http://gerrit.cloudera.org:8080/9030 Reviewed-by: Alex Rodoni <arod...@cloudera.com> Reviewed-by: Kim Jin Chul <jinc...@gmail.com> Reviewed-by: John Russell <jruss...@cloudera.com> Tested-by: Impala Public Jenkins
Project: http://git-wip-us.apache.org/repos/asf/impala/repo Commit: http://git-wip-us.apache.org/repos/asf/impala/commit/e1123eb1 Tree: http://git-wip-us.apache.org/repos/asf/impala/tree/e1123eb1 Diff: http://git-wip-us.apache.org/repos/asf/impala/diff/e1123eb1 Branch: refs/heads/2.x Commit: e1123eb186ddc556bfb8db1d3165b1f287057ffe Parents: 80e385a Author: Jinchul <jinc...@gmail.com> Authored: Tue Jan 16 21:13:43 2018 +0900 Committer: Impala Public Jenkins <impala-public-jenk...@gerrit.cloudera.org> Committed: Wed Apr 11 22:56:00 2018 +0000 ---------------------------------------------------------------------- docs/topics/impala_hints.xml | 42 ++++++++++++++++++++++++++++++++++++++ docs/topics/impala_insert.xml | 9 ++++---- docs/topics/impala_upsert.xml | 2 +- 3 files changed, 48 insertions(+), 5 deletions(-) ---------------------------------------------------------------------- http://git-wip-us.apache.org/repos/asf/impala/blob/e1123eb1/docs/topics/impala_hints.xml ---------------------------------------------------------------------- diff --git a/docs/topics/impala_hints.xml b/docs/topics/impala_hints.xml index d489048..6cafcfb 100644 --- a/docs/topics/impala_hints.xml +++ b/docs/topics/impala_hints.xml @@ -77,6 +77,12 @@ INSERT <varname>insert_clauses</varname> [{ /* +SHUFFLE */ | /* +NOSHUFFLE */ }] [<ph rev="IMPALA-2522 2.8.0">/* +CLUSTERED */</ph>] SELECT <varname>remainder_of_query</varname>; + +<ph rev="2.12.0 IMPALA-4168"> +UPSERT [{ /* +SHUFFLE */ | /* +NOSHUFFLE */ }] + [<ph rev="IMPALA-2522 2.8.0">/* +CLUSTERED */</ph>] + <varname>upsert_clauses</varname> + SELECT <varname>remainder_of_query</varname>;</ph> </codeblock> <p rev="2.0.0"> @@ -112,6 +118,26 @@ INSERT <varname>insert_clauses</varname> -- +SHUFFLE|NOSHUFFLE SELECT <varname>remainder_of_query</varname>; +<ph rev="2.12.0 IMPALA-4168"> +INSERT /* +SHUFFLE|NOSHUFFLE */ + <varname>insert_clauses</varname> + SELECT <varname>remainder_of_query</varname>;</ph> + +<ph rev="2.12.0 IMPALA-4168"> +INSERT -- +SHUFFLE|NOSHUFFLE + <varname>insert_clauses</varname> + SELECT <varname>remainder_of_query</varname>;</ph> + +<ph rev="2.12.0 IMPALA-4168"> +UPSERT /* +SHUFFLE|NOSHUFFLE */ + <varname>upsert_clauses</varname> + SELECT <varname>remainder_of_query</varname>;</ph> + +<ph rev="2.12.0 IMPALA-4168"> +UPSERT -- +SHUFFLE|NOSHUFFLE + <varname>upsert_clauses</varname> + SELECT <varname>remainder_of_query</varname>;</ph> + <ph rev="IMPALA-2924">SELECT <varname>select_list</varname> FROM <varname>table_ref</varname> /* +{SCHEDULE_CACHE_LOCAL | SCHEDULE_DISK_LOCAL | SCHEDULE_REMOTE} @@ -125,6 +151,22 @@ INSERT <varname>insert_clauses</varname> INSERT <varname>insert_clauses</varname> /* +CLUSTERED */ SELECT <varname>remainder_of_query</varname>;</ph> + +<ph rev="2.12.0 IMPALA-4168">INSERT -- +CLUSTERED + <varname>insert_clauses</varname> + SELECT <varname>remainder_of_query</varname>; + +INSERT /* +CLUSTERED */ + <varname>insert_clauses</varname> + SELECT <varname>remainder_of_query</varname>; + +UPSERT -- +CLUSTERED + <varname>upsert_clauses</varname> + SELECT <varname>remainder_of_query</varname>; + +UPSERT /* +CLUSTERED */ + <varname>upsert_clauses</varname> + SELECT <varname>remainder_of_query</varname>;</ph> </codeblock> <p conref="../shared/impala_common.xml#common/usage_notes_blurb"/> http://git-wip-us.apache.org/repos/asf/impala/blob/e1123eb1/docs/topics/impala_insert.xml ---------------------------------------------------------------------- diff --git a/docs/topics/impala_insert.xml b/docs/topics/impala_insert.xml index a83692d..3880a70 100644 --- a/docs/topics/impala_insert.xml +++ b/docs/topics/impala_insert.xml @@ -49,7 +49,7 @@ under the License. <p conref="../shared/impala_common.xml#common/syntax_blurb"/> <codeblock>[<varname>with_clause</varname>] -INSERT { INTO | OVERWRITE } [TABLE] <varname>table_name</varname> + INSERT <ph rev="2.12.0 IMPALA-4168">[<varname>hint_clause</varname>]</ph> { INTO | OVERWRITE } [TABLE] <varname>table_name</varname> [(<varname>column_list</varname>)] [ PARTITION (<varname>partition_clause</varname>)] { @@ -204,9 +204,10 @@ hint_with_brackets ::= [SHUFFLE] | [NOSHUFFLE] </li> <li rev="1.2.2"> - An optional hint clause immediately before the <codeph>SELECT</codeph> keyword, to fine-tune the behavior - when doing an <codeph>INSERT ... SELECT</codeph> operation into partitioned Parquet tables. The hint - keywords are <codeph>[SHUFFLE]</codeph> and <codeph>[NOSHUFFLE]</codeph>, including the square brackets. + An optional hint clause immediately either before the <codeph>SELECT</codeph> keyword or after the + <codeph>INSERT</codeph> keyword, to fine-tune the behavior when doing an <codeph>INSERT ... SELECT</codeph> + operation into partitioned Parquet tables. The hint clause cannot be specified in multiple places. + The hint keywords are <codeph>[SHUFFLE]</codeph> and <codeph>[NOSHUFFLE]</codeph>, including the square brackets. Inserting into partitioned Parquet tables can be a resource-intensive operation because it potentially involves many files being written to HDFS simultaneously, and separate <ph rev="parquet_block_size">large</ph> memory buffers being allocated to buffer the data for each http://git-wip-us.apache.org/repos/asf/impala/blob/e1123eb1/docs/topics/impala_upsert.xml ---------------------------------------------------------------------- diff --git a/docs/topics/impala_upsert.xml b/docs/topics/impala_upsert.xml index 5830675..a95b0f8 100644 --- a/docs/topics/impala_upsert.xml +++ b/docs/topics/impala_upsert.xml @@ -68,7 +68,7 @@ under the License. <p conref="../shared/impala_common.xml#common/syntax_blurb"/> <codeblock> -UPSERT INTO [TABLE] [<varname>db_name</varname>.]<varname>table_name</varname> +UPSERT <ph rev="2.12.0 IMPALA-4168">[<varname>hint_clause</varname>]</ph> INTO [TABLE] [<varname>db_name</varname>.]<varname>table_name</varname> [(<varname>column_list</varname>)] { [<varname>hint_clause</varname>] <varname>select_statement</varname>
