gustavodemorais commented on code in PR #26687:
URL: https://github.com/apache/flink/pull/26687#discussion_r2153311633
##########
flink-table/flink-table-planner/src/test/resources/restore-tests/stream-exec-multi-join_1/three-way-inner-join-with-restore/plan/three-way-inner-join-with-restore.json:
##########
@@ -0,0 +1,273 @@
+{
+ "flinkVersion" : "2.1",
+ "nodes" : [ {
+ "id" : 1,
+ "type" : "stream-exec-table-source-scan_1",
+ "scanTableSource" : {
+ "table" : {
+ "identifier" : "`default_catalog`.`default_database`.`Users`",
+ "resolvedTable" : {
+ "schema" : {
+ "columns" : [ {
+ "name" : "user_id",
+ "dataType" : "VARCHAR(2147483647)"
+ }, {
+ "name" : "name",
+ "dataType" : "VARCHAR(2147483647)"
+ } ],
+ "watermarkSpecs" : [ ]
+ },
+ "partitionKeys" : [ ]
+ }
+ }
+ },
+ "outputType" : "ROW<`user_id` VARCHAR(2147483647), `name`
VARCHAR(2147483647)>",
+ "description" : "TableSourceScan(table=[[default_catalog,
default_database, Users]], fields=[user_id, name])",
+ "inputProperties" : [ ]
+ }, {
+ "id" : 2,
+ "type" : "stream-exec-table-source-scan_1",
+ "scanTableSource" : {
+ "table" : {
+ "identifier" : "`default_catalog`.`default_database`.`Orders`",
+ "resolvedTable" : {
+ "schema" : {
+ "columns" : [ {
+ "name" : "user_id",
+ "dataType" : "VARCHAR(2147483647)"
+ }, {
+ "name" : "order_id",
+ "dataType" : "VARCHAR(2147483647)"
+ } ],
+ "watermarkSpecs" : [ ]
+ },
+ "partitionKeys" : [ ]
+ }
+ }
+ },
+ "outputType" : "ROW<`user_id` VARCHAR(2147483647), `order_id`
VARCHAR(2147483647)>",
+ "description" : "TableSourceScan(table=[[default_catalog,
default_database, Orders]], fields=[user_id, order_id])",
+ "inputProperties" : [ ]
+ }, {
+ "id" : 3,
+ "type" : "stream-exec-table-source-scan_1",
+ "scanTableSource" : {
+ "table" : {
+ "identifier" : "`default_catalog`.`default_database`.`Payments`",
+ "resolvedTable" : {
+ "schema" : {
+ "columns" : [ {
+ "name" : "user_id",
+ "dataType" : "VARCHAR(2147483647)"
+ }, {
+ "name" : "payment_id",
+ "dataType" : "VARCHAR(2147483647)"
+ } ],
+ "watermarkSpecs" : [ ]
+ },
+ "partitionKeys" : [ ]
+ }
+ }
+ },
+ "outputType" : "ROW<`user_id` VARCHAR(2147483647), `payment_id`
VARCHAR(2147483647)>",
+ "description" : "TableSourceScan(table=[[default_catalog,
default_database, Payments]], fields=[user_id, payment_id])",
+ "inputProperties" : [ ]
+ }, {
+ "id" : 4,
+ "type" : "stream-exec-multi-join_1",
+ "joinTypes" : [ "INNER", "INNER", "INNER" ],
+ "joinConditions" : [ {
+ "kind" : "LITERAL",
+ "value" : true,
+ "type" : "BOOLEAN NOT NULL"
+ }, {
+ "kind" : "CALL",
+ "syntax" : "BINARY",
+ "internalName" : "$=$1",
+ "operands" : [ {
+ "kind" : "INPUT_REF",
+ "inputIndex" : 0,
+ "type" : "VARCHAR(2147483647)"
+ }, {
+ "kind" : "INPUT_REF",
+ "inputIndex" : 2,
+ "type" : "VARCHAR(2147483647)"
+ } ],
+ "type" : "BOOLEAN"
+ }, {
+ "kind" : "CALL",
+ "syntax" : "BINARY",
+ "internalName" : "$=$1",
+ "operands" : [ {
+ "kind" : "INPUT_REF",
+ "inputIndex" : 0,
+ "type" : "VARCHAR(2147483647)"
+ }, {
+ "kind" : "INPUT_REF",
+ "inputIndex" : 4,
+ "type" : "VARCHAR(2147483647)"
+ } ],
+ "type" : "BOOLEAN"
+ } ],
+ "multiJoinCondition" : null,
+ "joinAttributeMap" : {
+ "1" : [ {
+ "leftInputId" : 0,
+ "leftFieldIndex" : 0,
+ "rightInputId" : 1,
+ "rightFieldIndex" : 0
+ } ],
+ "2" : [ {
+ "leftInputId" : 0,
+ "leftFieldIndex" : 0,
+ "rightInputId" : 2,
+ "rightFieldIndex" : 0
+ } ]
+ },
+ "inputUpsertKeys" : [ [ ], [ ], [ ] ],
Review Comment:
In this case, I also added `@JsonInclude(JsonInclude.Include.NON_EMPTY)` to
inputUpsertkeys. However, we'll still have it in the compiled plan because it's
not an empty list and has the "relevant information" that we have three inputs.
We can add a workaround like using [value =
JsonInclude.Include.CUSTOM](https://github.com/apache/flink/blob/ed9aebe1680be720e23d6321623253f66b2032d4/flink-table/flink-table-planner/src/main/java/org/apache/flink/table/planner/plan/nodes/exec/ExecNodeBase.java#L104)
or some custom logic in the ExecNode but I wasn't sure if it's worth it for
the readability for this case. It's always going to be a relatively short list
of "empty lists".
##########
flink-table/flink-table-planner/src/test/resources/restore-tests/stream-exec-multi-join_1/three-way-inner-join-with-restore/plan/three-way-inner-join-with-restore.json:
##########
@@ -0,0 +1,273 @@
+{
+ "flinkVersion" : "2.1",
+ "nodes" : [ {
+ "id" : 1,
+ "type" : "stream-exec-table-source-scan_1",
+ "scanTableSource" : {
+ "table" : {
+ "identifier" : "`default_catalog`.`default_database`.`Users`",
+ "resolvedTable" : {
+ "schema" : {
+ "columns" : [ {
+ "name" : "user_id",
+ "dataType" : "VARCHAR(2147483647)"
+ }, {
+ "name" : "name",
+ "dataType" : "VARCHAR(2147483647)"
+ } ],
+ "watermarkSpecs" : [ ]
+ },
+ "partitionKeys" : [ ]
+ }
+ }
+ },
+ "outputType" : "ROW<`user_id` VARCHAR(2147483647), `name`
VARCHAR(2147483647)>",
+ "description" : "TableSourceScan(table=[[default_catalog,
default_database, Users]], fields=[user_id, name])",
+ "inputProperties" : [ ]
+ }, {
+ "id" : 2,
+ "type" : "stream-exec-table-source-scan_1",
+ "scanTableSource" : {
+ "table" : {
+ "identifier" : "`default_catalog`.`default_database`.`Orders`",
+ "resolvedTable" : {
+ "schema" : {
+ "columns" : [ {
+ "name" : "user_id",
+ "dataType" : "VARCHAR(2147483647)"
+ }, {
+ "name" : "order_id",
+ "dataType" : "VARCHAR(2147483647)"
+ } ],
+ "watermarkSpecs" : [ ]
+ },
+ "partitionKeys" : [ ]
+ }
+ }
+ },
+ "outputType" : "ROW<`user_id` VARCHAR(2147483647), `order_id`
VARCHAR(2147483647)>",
+ "description" : "TableSourceScan(table=[[default_catalog,
default_database, Orders]], fields=[user_id, order_id])",
+ "inputProperties" : [ ]
+ }, {
+ "id" : 3,
+ "type" : "stream-exec-table-source-scan_1",
+ "scanTableSource" : {
+ "table" : {
+ "identifier" : "`default_catalog`.`default_database`.`Payments`",
+ "resolvedTable" : {
+ "schema" : {
+ "columns" : [ {
+ "name" : "user_id",
+ "dataType" : "VARCHAR(2147483647)"
+ }, {
+ "name" : "payment_id",
+ "dataType" : "VARCHAR(2147483647)"
+ } ],
+ "watermarkSpecs" : [ ]
+ },
+ "partitionKeys" : [ ]
+ }
+ }
+ },
+ "outputType" : "ROW<`user_id` VARCHAR(2147483647), `payment_id`
VARCHAR(2147483647)>",
+ "description" : "TableSourceScan(table=[[default_catalog,
default_database, Payments]], fields=[user_id, payment_id])",
+ "inputProperties" : [ ]
+ }, {
+ "id" : 4,
+ "type" : "stream-exec-multi-join_1",
+ "joinTypes" : [ "INNER", "INNER", "INNER" ],
+ "joinConditions" : [ {
+ "kind" : "LITERAL",
+ "value" : true,
+ "type" : "BOOLEAN NOT NULL"
+ }, {
+ "kind" : "CALL",
+ "syntax" : "BINARY",
+ "internalName" : "$=$1",
+ "operands" : [ {
+ "kind" : "INPUT_REF",
+ "inputIndex" : 0,
+ "type" : "VARCHAR(2147483647)"
+ }, {
+ "kind" : "INPUT_REF",
+ "inputIndex" : 2,
+ "type" : "VARCHAR(2147483647)"
+ } ],
+ "type" : "BOOLEAN"
+ }, {
+ "kind" : "CALL",
+ "syntax" : "BINARY",
+ "internalName" : "$=$1",
+ "operands" : [ {
+ "kind" : "INPUT_REF",
+ "inputIndex" : 0,
+ "type" : "VARCHAR(2147483647)"
+ }, {
+ "kind" : "INPUT_REF",
+ "inputIndex" : 4,
+ "type" : "VARCHAR(2147483647)"
+ } ],
+ "type" : "BOOLEAN"
+ } ],
+ "multiJoinCondition" : null,
+ "joinAttributeMap" : {
+ "1" : [ {
+ "leftInputId" : 0,
+ "leftFieldIndex" : 0,
+ "rightInputId" : 1,
+ "rightFieldIndex" : 0
+ } ],
+ "2" : [ {
+ "leftInputId" : 0,
+ "leftFieldIndex" : 0,
+ "rightInputId" : 2,
+ "rightFieldIndex" : 0
+ } ]
+ },
+ "inputUpsertKeys" : [ [ ], [ ], [ ] ],
Review Comment:
In this case, I also added `@JsonInclude(JsonInclude.Include.NON_EMPTY)` to
inputUpsertkeys. However, we'll still have it in the compiled plan because it's
not an empty list and has the "relevant information" that we have three inputs.
We can add a workaround like using [value =
JsonInclude.Include.CUSTOM](https://github.com/apache/flink/blob/ed9aebe1680be720e23d6321623253f66b2032d4/flink-table/flink-table-planner/src/main/java/org/apache/flink/table/planner/plan/nodes/exec/ExecNodeBase.java#L104)
or some custom logic in the ExecNode but I wasn't sure if it's worth it for
the readability for this case. It's always going to be a relatively short list
of "empty lists". Wdyt?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
