[
https://issues.apache.org/jira/browse/PHOENIX-7881?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Andrew Kyle Purtell updated PHOENIX-7881:
-----------------------------------------
Description:
Now that we have infrastructure for parsing and validating EXPLAIN plan output
from PHOENIX-7879, we should use this instead of hardcoding exact text matches
in ~80 various IT and UT tests.
There are no changes made yet to the actual format of EXPLAIN text output as it
gets presented to the user. We also are not expanding (or reducing) existing
test coverage. However now we are beginning to expand ExplainPlanAttributes so
tests can use a fluent API to assert over ExplainPlanAttributes rather than
exact matching of text strings.
Now that we have expanded ExplainPlanAttributes to what we need, we also
reorganize its attribute display order for better logical grouping. This only
impacts future "EXPLAIN JSON" functionality that doesn't even exist yet.
Some exact matching of text remains so we don't lose coverage that used to
exist in QueryPlanTest.
Further test refactoring is likely down the road to fully rely on fluent
assertions using ExplainPlanAttributes, but that won't happen until the new
EXPLAIN logic is implemented and output grammar validation is built.
was:
Now that we have infrastructure for parsing and validating EXPLAIN plan output
from PHOENIX-7879, we should use this instead of hardcoding exact text matches
in ~80 various IT and UT tests.
There are no changes made yet to the actual format of EXPLAIN text output as it
gets presented to the user. We also are not expanding (or reducing) existing
test coverage. However now we are beginning to expand ExplainPlanAttributes so
tests can use a fluent API to assert over ExplainPlanAttributes rather than
exact matching of text strings.
Now that we have expanded ExplainPlanAttributes to what we need, we also
reorganize it for better logical grouping. This only impacts future "EXPLAIN
JSON" functionality that doesn't even exist yet.
Some exact matching of text remains so we don't lose coverage that used to
exist in QueryPlanTest.
Further test refactoring is likely down the road to fully rely on fluent
assertions using ExplainPlanAttributes, but that won't happen until the new
EXPLAIN logic is implemented and output grammar validation is built.
> Refactor UTs and ITs to assert on ExplainPlanAttributes
> -------------------------------------------------------
>
> Key: PHOENIX-7881
> URL: https://issues.apache.org/jira/browse/PHOENIX-7881
> Project: Phoenix
> Issue Type: Sub-task
> Reporter: Andrew Kyle Purtell
> Assignee: Andrew Kyle Purtell
> Priority: Major
> Fix For: 5.4.0, 5.3.2
>
>
> Now that we have infrastructure for parsing and validating EXPLAIN plan
> output from PHOENIX-7879, we should use this instead of hardcoding exact text
> matches in ~80 various IT and UT tests.
> There are no changes made yet to the actual format of EXPLAIN text output as
> it gets presented to the user. We also are not expanding (or reducing)
> existing test coverage. However now we are beginning to expand
> ExplainPlanAttributes so tests can use a fluent API to assert over
> ExplainPlanAttributes rather than exact matching of text strings.
> Now that we have expanded ExplainPlanAttributes to what we need, we also
> reorganize its attribute display order for better logical grouping. This only
> impacts future "EXPLAIN JSON" functionality that doesn't even exist yet.
> Some exact matching of text remains so we don't lose coverage that used to
> exist in QueryPlanTest.
> Further test refactoring is likely down the road to fully rely on fluent
> assertions using ExplainPlanAttributes, but that won't happen until the new
> EXPLAIN logic is implemented and output grammar validation is built.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
