You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@cassandra.apache.org by "Jonathan Ellis (JIRA)" <ji...@apache.org> on 2009/05/18 21:17:45 UTC
[jira] Created: (CASSANDRA-185) user-defined column ordering
user-defined column ordering
----------------------------
Key: CASSANDRA-185
URL: https://issues.apache.org/jira/browse/CASSANDRA-185
Project: Cassandra
Issue Type: New Feature
Reporter: Jonathan Ellis
name ordering by ascii String order isn't general enough.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Resolved: (CASSANDRA-185) user-defined column ordering
Posted by "Jonathan Ellis (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-185?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Jonathan Ellis resolved CASSANDRA-185.
--------------------------------------
Resolution: Fixed
Assignee: Jonathan Ellis
> explicitly have a validate() method instead
good idea.
committed with this modification.
> user-defined column ordering
> ----------------------------
>
> Key: CASSANDRA-185
> URL: https://issues.apache.org/jira/browse/CASSANDRA-185
> Project: Cassandra
> Issue Type: New Feature
> Components: Core
> Reporter: Jonathan Ellis
> Assignee: Jonathan Ellis
> Attachments: 0001-CASSANDRA-185-introduce-column-compatibility-layer-t.txt, 0002-r-m-time-sorted-columns.txt, 0003-change-String-column-names-to-byte-to-allow-user-def.txt, 0004-r-m-tests-that-won-t-pass-until-we-do-303.txt
>
>
> name ordering by ascii String order isn't general enough.
> we need to allow specifying a comparator class like we do (indirectly) for key ordering via the partitioner interface.
> per-CF would be best.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (CASSANDRA-185) user-defined column ordering
Posted by "Jonathan Ellis (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-185?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12733827#action_12733827 ]
Jonathan Ellis commented on CASSANDRA-185:
------------------------------------------
nope. :)
thanks for the catch, will fix.
> user-defined column ordering
> ----------------------------
>
> Key: CASSANDRA-185
> URL: https://issues.apache.org/jira/browse/CASSANDRA-185
> Project: Cassandra
> Issue Type: New Feature
> Components: Core
> Reporter: Jonathan Ellis
> Assignee: Jonathan Ellis
> Attachments: 0001-CASSANDRA-185-introduce-column-compatibility-layer-t.txt, 0002-r-m-time-sorted-columns.txt, 0003-change-String-column-names-to-byte-to-allow-user-def.txt, 0004-r-m-tests-that-won-t-pass-until-we-do-303.txt
>
>
> name ordering by ascii String order isn't general enough.
> we need to allow specifying a comparator class like we do (indirectly) for key ordering via the partitioner interface.
> per-CF would be best.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (CASSANDRA-185) user-defined column ordering
Posted by "Michael Greene (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-185?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12733718#action_12733718 ]
Michael Greene commented on CASSANDRA-185:
------------------------------------------
Did you mean to add that `logger_.debug(length)` in r796132 in FBUtilities.java?
It doesn't really make sense when reading the logs without context.
> user-defined column ordering
> ----------------------------
>
> Key: CASSANDRA-185
> URL: https://issues.apache.org/jira/browse/CASSANDRA-185
> Project: Cassandra
> Issue Type: New Feature
> Components: Core
> Reporter: Jonathan Ellis
> Assignee: Jonathan Ellis
> Attachments: 0001-CASSANDRA-185-introduce-column-compatibility-layer-t.txt, 0002-r-m-time-sorted-columns.txt, 0003-change-String-column-names-to-byte-to-allow-user-def.txt, 0004-r-m-tests-that-won-t-pass-until-we-do-303.txt
>
>
> name ordering by ascii String order isn't general enough.
> we need to allow specifying a comparator class like we do (indirectly) for key ordering via the partitioner interface.
> per-CF would be best.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (CASSANDRA-185) user-defined column ordering
Posted by "Sandeep Tata (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-185?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12733423#action_12733423 ]
Sandeep Tata commented on CASSANDRA-185:
----------------------------------------
+1
Minor comment: I saw that you're using getComparator.getString(path.columnName) and catching exceptions as a proxy for a validate() method for column names (eg in ColumnFamily.addColumn)... does it make sense to explicitly have a validate() method instead?
> user-defined column ordering
> ----------------------------
>
> Key: CASSANDRA-185
> URL: https://issues.apache.org/jira/browse/CASSANDRA-185
> Project: Cassandra
> Issue Type: New Feature
> Components: Core
> Reporter: Jonathan Ellis
> Attachments: 0001-CASSANDRA-185-introduce-column-compatibility-layer-t.txt, 0002-r-m-time-sorted-columns.txt, 0003-change-String-column-names-to-byte-to-allow-user-def.txt, 0004-r-m-tests-that-won-t-pass-until-we-do-303.txt
>
>
> name ordering by ascii String order isn't general enough.
> we need to allow specifying a comparator class like we do (indirectly) for key ordering via the partitioner interface.
> per-CF would be best.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (CASSANDRA-185) user-defined column ordering
Posted by "Jonathan Ellis (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-185?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Jonathan Ellis updated CASSANDRA-185:
-------------------------------------
Attachment: 0004-r-m-tests-that-won-t-pass-until-we-do-303.txt
0003-change-String-column-names-to-byte-to-allow-user-def.txt
0002-r-m-time-sorted-columns.txt
0001-CASSANDRA-185-introduce-column-compatibility-layer-t.txt
> user-defined column ordering
> ----------------------------
>
> Key: CASSANDRA-185
> URL: https://issues.apache.org/jira/browse/CASSANDRA-185
> Project: Cassandra
> Issue Type: New Feature
> Components: Core
> Reporter: Jonathan Ellis
> Attachments: 0001-CASSANDRA-185-introduce-column-compatibility-layer-t.txt, 0002-r-m-time-sorted-columns.txt, 0003-change-String-column-names-to-byte-to-allow-user-def.txt, 0004-r-m-tests-that-won-t-pass-until-we-do-303.txt
>
>
> name ordering by ascii String order isn't general enough.
> we need to allow specifying a comparator class like we do (indirectly) for key ordering via the partitioner interface.
> per-CF would be best.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (CASSANDRA-185) user-defined column ordering
Posted by "Jonathan Ellis (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-185?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12730497#action_12730497 ]
Jonathan Ellis commented on CASSANDRA-185:
------------------------------------------
for backwards compatibility on-disk, we'll need to limit the byte[] length to 64k
see e.g. http://java.sun.com/javase/6/docs/api/java/io/DataInput.html#readUTF()
> user-defined column ordering
> ----------------------------
>
> Key: CASSANDRA-185
> URL: https://issues.apache.org/jira/browse/CASSANDRA-185
> Project: Cassandra
> Issue Type: New Feature
> Reporter: Jonathan Ellis
>
> name ordering by ascii String order isn't general enough.
> we need to allow specifying a comparator class like we do (indirectly) for key ordering via the partitioner interface.
> per-CF would be best.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Issue Comment Edited: (CASSANDRA-185) user-defined column
ordering
Posted by "Sandeep Tata (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-185?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12733378#action_12733378 ]
Sandeep Tata edited comment on CASSANDRA-185 at 7/20/09 3:18 PM:
-----------------------------------------------------------------
Ah .. just noticed the happy side-effect that EfficientBidiMap is now replaced by a single ConcurrentSkipListMap (See CASSANDRA-51 for context)
was (Author: sandeep_tata):
Ah .. just noticed the happy side-effect that EfficientBidiMap is now replaced by a single ConcurrentSkipListMap ...
> user-defined column ordering
> ----------------------------
>
> Key: CASSANDRA-185
> URL: https://issues.apache.org/jira/browse/CASSANDRA-185
> Project: Cassandra
> Issue Type: New Feature
> Components: Core
> Reporter: Jonathan Ellis
> Attachments: 0001-CASSANDRA-185-introduce-column-compatibility-layer-t.txt, 0002-r-m-time-sorted-columns.txt, 0003-change-String-column-names-to-byte-to-allow-user-def.txt, 0004-r-m-tests-that-won-t-pass-until-we-do-303.txt
>
>
> name ordering by ascii String order isn't general enough.
> we need to allow specifying a comparator class like we do (indirectly) for key ordering via the partitioner interface.
> per-CF would be best.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (CASSANDRA-185) user-defined column ordering
Posted by "Sandeep Tata (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-185?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12733378#action_12733378 ]
Sandeep Tata commented on CASSANDRA-185:
----------------------------------------
Ah .. just noticed the happy side-effect that EfficientBidiMap is now replaced by a single ConcurrentSkipListMap ...
> user-defined column ordering
> ----------------------------
>
> Key: CASSANDRA-185
> URL: https://issues.apache.org/jira/browse/CASSANDRA-185
> Project: Cassandra
> Issue Type: New Feature
> Components: Core
> Reporter: Jonathan Ellis
> Attachments: 0001-CASSANDRA-185-introduce-column-compatibility-layer-t.txt, 0002-r-m-time-sorted-columns.txt, 0003-change-String-column-names-to-byte-to-allow-user-def.txt, 0004-r-m-tests-that-won-t-pass-until-we-do-303.txt
>
>
> name ordering by ascii String order isn't general enough.
> we need to allow specifying a comparator class like we do (indirectly) for key ordering via the partitioner interface.
> per-CF would be best.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (CASSANDRA-185) user-defined column ordering
Posted by "Jonathan Ellis (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-185?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Jonathan Ellis updated CASSANDRA-185:
-------------------------------------
Description:
name ordering by ascii String order isn't general enough.
we need to allow specifying a comparator class like we do (indirectly) for key ordering via the partitioner interface.
per-CF would be best.
was:name ordering by ascii String order isn't general enough.
> user-defined column ordering
> ----------------------------
>
> Key: CASSANDRA-185
> URL: https://issues.apache.org/jira/browse/CASSANDRA-185
> Project: Cassandra
> Issue Type: New Feature
> Reporter: Jonathan Ellis
>
> name ordering by ascii String order isn't general enough.
> we need to allow specifying a comparator class like we do (indirectly) for key ordering via the partitioner interface.
> per-CF would be best.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.