You are viewing a plain text version of this content. The canonical link for it is here.
Posted to derby-dev@db.apache.org by "Kathey Marsden (JIRA)" <ji...@apache.org> on 2008/05/22 18:05:57 UTC
[jira] Created: (DERBY-3686) convert predicatePushdown.sql to
JUnit
convert predicatePushdown.sql to JUnit
---------------------------------------
Key: DERBY-3686
URL: https://issues.apache.org/jira/browse/DERBY-3686
Project: Derby
Issue Type: Task
Components: Test
Affects Versions: 10.5.0.0
Reporter: Kathey Marsden
Assignee: Kathey Marsden
Priority: Minor
predicatePushdown.sql fails intermittently in the nightlies ( DERBY-1902) with a plan variance. It would be good to convert it to JUnit and extract only the relevant information from RuntimeStatistics output. This might make the test more stable.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Closed: (DERBY-3686) convert predicatePushdown.sql to JUnit
Posted by "Kathey Marsden (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/DERBY-3686?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Kathey Marsden closed DERBY-3686.
---------------------------------
Resolution: Fixed
Fix Version/s: 10.5.0.0
> convert predicatePushdown.sql to JUnit
> ---------------------------------------
>
> Key: DERBY-3686
> URL: https://issues.apache.org/jira/browse/DERBY-3686
> Project: Derby
> Issue Type: Task
> Components: Test
> Affects Versions: 10.5.0.0
> Reporter: Kathey Marsden
> Assignee: Kathey Marsden
> Priority: Minor
> Fix For: 10.5.0.0
>
> Attachments: derby-3686_complete_diff.txt, derby-3686_initial_conv_diff.txt, derby-3686_rts_diff.txt
>
>
> predicatePushdown.sql fails intermittently in the nightlies ( DERBY-1902) with a plan variance. It would be good to convert it to JUnit and extract only the relevant information from RuntimeStatistics output. This might make the test more stable.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (DERBY-3686) convert predicatePushdown.sql to
JUnit
Posted by "Dag H. Wanvik (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/DERBY-3686?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12601001#action_12601001 ]
Dag H. Wanvik commented on DERBY-3686:
--------------------------------------
My 0.02 cents:
Might be good to add a comment on the use of findString that the magic number 4
refers back to the fact that there should be scalar qualifiers for all 4 scans.
The SQL is reformatted to be less readable in the JUnit version making it a lot harder to read than the original
which seems a bit of a loss.
It would be nice if the statistics output was more well defined and parsable. The current
format is inherently brittle it seems... There may be a JIRA for this already; I didn't find one..
> convert predicatePushdown.sql to JUnit
> ---------------------------------------
>
> Key: DERBY-3686
> URL: https://issues.apache.org/jira/browse/DERBY-3686
> Project: Derby
> Issue Type: Task
> Components: Test
> Affects Versions: 10.5.0.0
> Reporter: Kathey Marsden
> Assignee: Kathey Marsden
> Priority: Minor
> Attachments: derby-3686_complete_diff.txt, derby-3686_initial_conv_diff.txt, derby-3686_rts_diff.txt
>
>
> predicatePushdown.sql fails intermittently in the nightlies ( DERBY-1902) with a plan variance. It would be good to convert it to JUnit and extract only the relevant information from RuntimeStatistics output. This might make the test more stable.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (DERBY-3686) convert predicatePushdown.sql to
JUnit
Posted by "Kathey Marsden (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/DERBY-3686?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12601157#action_12601157 ]
Kathey Marsden commented on DERBY-3686:
---------------------------------------
Thanks Dag for looking at the test. I'll add the comment. As for the reformatting of the sql, it seems to be a feature of the SQLToJUnit utility. perhaps that utility could be changed to keep existing line breaks if they are short enough to fit in the line. I don't think I'll go back and redo the whole test now, but it sounds like a good improvement for the tool. There is also a problem with the tool that it will not put in line breaks after single line comments. This might fix that problem too. I'll take a look and see if it is easy to do.
I don't think there is an issue for making the statistics easier to parse. If you have some ideas in this regard, please file an issue.
> convert predicatePushdown.sql to JUnit
> ---------------------------------------
>
> Key: DERBY-3686
> URL: https://issues.apache.org/jira/browse/DERBY-3686
> Project: Derby
> Issue Type: Task
> Components: Test
> Affects Versions: 10.5.0.0
> Reporter: Kathey Marsden
> Assignee: Kathey Marsden
> Priority: Minor
> Attachments: derby-3686_complete_diff.txt, derby-3686_initial_conv_diff.txt, derby-3686_rts_diff.txt
>
>
> predicatePushdown.sql fails intermittently in the nightlies ( DERBY-1902) with a plan variance. It would be good to convert it to JUnit and extract only the relevant information from RuntimeStatistics output. This might make the test more stable.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (DERBY-3686) convert predicatePushdown.sql to
JUnit
Posted by "Kathey Marsden (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/DERBY-3686?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Kathey Marsden updated DERBY-3686:
----------------------------------
Derby Info: [Patch Available]
> convert predicatePushdown.sql to JUnit
> ---------------------------------------
>
> Key: DERBY-3686
> URL: https://issues.apache.org/jira/browse/DERBY-3686
> Project: Derby
> Issue Type: Task
> Components: Test
> Affects Versions: 10.5.0.0
> Reporter: Kathey Marsden
> Assignee: Kathey Marsden
> Priority: Minor
> Attachments: derby-3686_initial_conv_diff.txt, derby-3686_rts_diff.txt
>
>
> predicatePushdown.sql fails intermittently in the nightlies ( DERBY-1902) with a plan variance. It would be good to convert it to JUnit and extract only the relevant information from RuntimeStatistics output. This might make the test more stable.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (DERBY-3686) convert predicatePushdown.sql to
JUnit
Posted by "Kathey Marsden (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/DERBY-3686?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Kathey Marsden updated DERBY-3686:
----------------------------------
Attachment: derby-3686_rts_diff.txt
Attached is a patch that adds the Runtime Statistics parsing for PredicatePushdownTest. There is still one case that I have not figured out how to parse yet.
// Nested unions with non-join predicates should work as
// usual (no change with DERBY-805). So should see scalar
// qualifiers on scans for all instances of T1 and T2.
I marked this with a //RESOLVE comment in the test.
Any tips on how to parse that would be appreciated. Also I would appreciate if someone could review the test to make sure I am checking the right things for the other queries. I based what I checked for on the comments and the predicatePushdown.out master file. Hopefully the checks are sufficient.
Kathey
> convert predicatePushdown.sql to JUnit
> ---------------------------------------
>
> Key: DERBY-3686
> URL: https://issues.apache.org/jira/browse/DERBY-3686
> Project: Derby
> Issue Type: Task
> Components: Test
> Affects Versions: 10.5.0.0
> Reporter: Kathey Marsden
> Assignee: Kathey Marsden
> Priority: Minor
> Attachments: derby-3686_initial_conv_diff.txt, derby-3686_rts_diff.txt
>
>
> predicatePushdown.sql fails intermittently in the nightlies ( DERBY-1902) with a plan variance. It would be good to convert it to JUnit and extract only the relevant information from RuntimeStatistics output. This might make the test more stable.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (DERBY-3686) convert predicatePushdown.sql to
JUnit
Posted by "Kathey Marsden (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/DERBY-3686?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Kathey Marsden updated DERBY-3686:
----------------------------------
Attachment: derby-3686_complete_diff.txt
Here is a new patch derby-3686_complete_diff.txt that replaces the previous one. For check for the four qualifiers I added findString method to RuntimeStatisticsParser that takes a number of instances to search for the string and called.
p.findString("Operator: <=", 4);
Hopefully that is a sufficient check.
This patch also removes the old test and adds the new one to the lang._Suite.
Please review.
> convert predicatePushdown.sql to JUnit
> ---------------------------------------
>
> Key: DERBY-3686
> URL: https://issues.apache.org/jira/browse/DERBY-3686
> Project: Derby
> Issue Type: Task
> Components: Test
> Affects Versions: 10.5.0.0
> Reporter: Kathey Marsden
> Assignee: Kathey Marsden
> Priority: Minor
> Attachments: derby-3686_complete_diff.txt, derby-3686_initial_conv_diff.txt, derby-3686_rts_diff.txt
>
>
> predicatePushdown.sql fails intermittently in the nightlies ( DERBY-1902) with a plan variance. It would be good to convert it to JUnit and extract only the relevant information from RuntimeStatistics output. This might make the test more stable.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (DERBY-3686) convert predicatePushdown.sql to
JUnit
Posted by "Kathey Marsden (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/DERBY-3686?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Kathey Marsden updated DERBY-3686:
----------------------------------
Attachment: derby-3686_initial_conv_diff.txt
Here is a patch for the initial test conversion with the SQLToJUnit utility. I ran into a few glitches with the tool which I patched up manually, but I will try to get a smaller repro case for those and maybe see if I can fix the tool. The test does not yet have the RuntimeStatistics analysis yet. I plan to check in the conversion so far and then add in the RuntimeStatistics parsing. The test currently passes but is incomplete.
> convert predicatePushdown.sql to JUnit
> ---------------------------------------
>
> Key: DERBY-3686
> URL: https://issues.apache.org/jira/browse/DERBY-3686
> Project: Derby
> Issue Type: Task
> Components: Test
> Affects Versions: 10.5.0.0
> Reporter: Kathey Marsden
> Assignee: Kathey Marsden
> Priority: Minor
> Attachments: derby-3686_initial_conv_diff.txt
>
>
> predicatePushdown.sql fails intermittently in the nightlies ( DERBY-1902) with a plan variance. It would be good to convert it to JUnit and extract only the relevant information from RuntimeStatistics output. This might make the test more stable.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.