You are viewing a plain text version of this content. The canonical link for it is here.
Posted to solr-dev@lucene.apache.org by "Noble Paul (JIRA)" <ji...@apache.org> on 2009/12/12 05:58:18 UTC
[jira] Created: (SOLR-1647) Remove the option of setting solrconfig
from web.xml
Remove the option of setting solrconfig from web.xml
----------------------------------------------------
Key: SOLR-1647
URL: https://issues.apache.org/jira/browse/SOLR-1647
Project: Solr
Issue Type: Sub-task
Reporter: Noble Paul
Assignee: Noble Paul
Fix For: 1.5
with SOLR-1621 , it is not required to have an option to set solrconfig from web.xml. Moreover editing web.xml means hacking solr itself.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (SOLR-1647) Remove the option of setting
solrconfig from web.xml
Posted by "Noble Paul (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/SOLR-1647?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12794770#action_12794770 ]
Noble Paul commented on SOLR-1647:
----------------------------------
I wish to remove this feature because it is a bad one not because it is hard to keep it. I am sure there are very few users using it anyway.
We can deprecate it and remove it later. But the question is do we have to be that strict about everything?
> Remove the option of setting solrconfig from web.xml
> ----------------------------------------------------
>
> Key: SOLR-1647
> URL: https://issues.apache.org/jira/browse/SOLR-1647
> Project: Solr
> Issue Type: Improvement
> Reporter: Noble Paul
> Assignee: Noble Paul
> Fix For: 1.5
>
> Attachments: SOLR-1647.patch
>
>
> with SOLR-1621 , it is not required to have an option to set solrconfig from web.xml. Moreover editing web.xml means hacking solr itself.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (SOLR-1647) Remove the option of setting
solrconfig from web.xml
Posted by "Noble Paul (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/SOLR-1647?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12790594#action_12790594 ]
Noble Paul commented on SOLR-1647:
----------------------------------
this can be taken up as a separate issue . SOLR-1621 can be committed w/o this
> Remove the option of setting solrconfig from web.xml
> ----------------------------------------------------
>
> Key: SOLR-1647
> URL: https://issues.apache.org/jira/browse/SOLR-1647
> Project: Solr
> Issue Type: Improvement
> Reporter: Noble Paul
> Assignee: Noble Paul
> Fix For: 1.5
>
>
> with SOLR-1621 , it is not required to have an option to set solrconfig from web.xml. Moreover editing web.xml means hacking solr itself.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (SOLR-1647) Remove the option of setting solrconfig
from web.xml
Posted by "Noble Paul (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/SOLR-1647?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Noble Paul updated SOLR-1647:
-----------------------------
Issue Type: Improvement (was: Sub-task)
Parent: (was: SOLR-1621)
> Remove the option of setting solrconfig from web.xml
> ----------------------------------------------------
>
> Key: SOLR-1647
> URL: https://issues.apache.org/jira/browse/SOLR-1647
> Project: Solr
> Issue Type: Improvement
> Reporter: Noble Paul
> Assignee: Noble Paul
> Fix For: 1.5
>
>
> with SOLR-1621 , it is not required to have an option to set solrconfig from web.xml. Moreover editing web.xml means hacking solr itself.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (SOLR-1647) Remove the option of setting solrconfig
from web.xml
Posted by "Noble Paul (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/SOLR-1647?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Noble Paul updated SOLR-1647:
-----------------------------
Attachment: SOLR-1647.patch
> Remove the option of setting solrconfig from web.xml
> ----------------------------------------------------
>
> Key: SOLR-1647
> URL: https://issues.apache.org/jira/browse/SOLR-1647
> Project: Solr
> Issue Type: Improvement
> Reporter: Noble Paul
> Assignee: Noble Paul
> Fix For: 1.5
>
> Attachments: SOLR-1647.patch
>
>
> with SOLR-1621 , it is not required to have an option to set solrconfig from web.xml. Moreover editing web.xml means hacking solr itself.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (SOLR-1647) Remove the option of setting
solrconfig from web.xml
Posted by "Mark Miller (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/SOLR-1647?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12794401#action_12794401 ]
Mark Miller commented on SOLR-1647:
-----------------------------------
-1 on this patch.
1. There is no reason not to keep this for a release with a warning to users that its going to be removed. You put in a deprecation tag in an earlier patch, and now your just removing it before that was ever released - there is no downside to doing this more like a *real* deprecation and there is upside.
2. You have put no warnings or information to users about this change - you could be breaking things for some people and your doing it out of the blue - we shouldn't pull the rug out from under users like that.
3. Ive I have said more than once, the current patch *breaks tests* that rely on this feature.
I see no reason to remove it now. It should be deprecated with a warning in changes about its impending removal. And you need to provide a workaround for tests. I don't see why we don't just keep it for now. Its not hurting anything.
I'd also like to be more clear on what deprecation means in Solr - I know it doesn't mean deprecating and removing before a release even goes out (that makes no sense) - and I'm not sure if it means removing in the next dot release (as Ive see that occurring recently). Can someone pipe in on whether Solr has an (in)formal policy on when deprecations should/can be removed?
> Remove the option of setting solrconfig from web.xml
> ----------------------------------------------------
>
> Key: SOLR-1647
> URL: https://issues.apache.org/jira/browse/SOLR-1647
> Project: Solr
> Issue Type: Improvement
> Reporter: Noble Paul
> Assignee: Noble Paul
> Fix For: 1.5
>
> Attachments: SOLR-1647.patch
>
>
> with SOLR-1621 , it is not required to have an option to set solrconfig from web.xml. Moreover editing web.xml means hacking solr itself.
--
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: (SOLR-1647) Remove the option of
setting solrconfig from web.xml
Posted by "Mark Miller (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/SOLR-1647?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12795009#action_12795009 ]
Mark Miller edited comment on SOLR-1647 at 12/29/09 3:14 AM:
-------------------------------------------------------------
bq. I'm definitely not going to commit the same patch which is attached. I will ensure that all tests pass before this goes in.
I guess this is a communication error. I took {quote}I plan to commit this shortly.{quote} as "you are going to commit the current patch".
I guessed that you might do a few things before committing, but I have no way of knowing. When someone says that they plan to commit something shortly,
I take it to mean something along the lines of the patch posted. As you are missing two things that are pretty major pieces to this patch (the deprecation/non deprecation approach and a good workaround
for the tests), I voiced my opposition to the current approach shown. Its hard for me to guess what changes you will make to this patch before you commit soon - I have to assume when you say that you are committing shortly that perhaps you will address both things correctly and perhaps you won't - you could just commit the current patch, who knows - I can't really rely on you doing anything unless you post the patch first, but you mention nothing of another patch, just of committing. Thats why I brought up the issues that I did. I can see making some last minutes changes to a patch, but these two things are fairly important to this issue I feel, and not really last minute tweaks before a commit.
I'd like the opportunity to take a look at how you are going to address these two issues and (fwiw) possibly provide feedback.
was (Author: markrmiller@gmail.com):
bq. I'm definitely not going to commit the same patch which is attached. I will ensure that all tests pass before this goes in.
I guess this is a communication error. I took {quote}I plan to commit this shortly.{quote} as "you are going to commit the current patch".
I guessed that you might do a few things before committing, but I have no way of knowing. When someone says that they plan to commit something shortly,
I take it to mean something along the lines of the patch posted. As you are missing two things that are pretty major pieces to this patch (the deprecation/non deprecation approach and a good workaround
for the tests), I voiced my opposition to the current approach shown. Its hard for me to guess what changes you will make to this patch before you commit soon - I have to assume when you say that you are committing shortly that perhaps you will address both things correctly and perhaps you won't - you could just commit the current patch, who knows - I can't really rely on you doing anything unless you post the patch first, but you mention nothing of another patch, just of committing. Thats why I brought up the issues that I did. I can say making some last minutes changes to a patch, but these two things are fairly important to this issue I feel, and not really last minute tweaks before a commit.
I'd like the opportunity to take a look at the how you are going to address these two issues and (fwiw) possibly provide feedback.
> Remove the option of setting solrconfig from web.xml
> ----------------------------------------------------
>
> Key: SOLR-1647
> URL: https://issues.apache.org/jira/browse/SOLR-1647
> Project: Solr
> Issue Type: Improvement
> Reporter: Noble Paul
> Assignee: Noble Paul
> Fix For: 1.5
>
> Attachments: SOLR-1647.patch
>
>
> with SOLR-1621 , it is not required to have an option to set solrconfig from web.xml. Moreover editing web.xml means hacking solr itself.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (SOLR-1647) Remove the option of setting
solrconfig from web.xml
Posted by "Mark Miller (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/SOLR-1647?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12794806#action_12794806 ]
Mark Miller commented on SOLR-1647:
-----------------------------------
bq. But the question is do we have to be that strict about everything?
Personally, I would say no, we don't (and many times we have not been). I'm not -1'ing the removal of this. I think we should keep it and say its going away first, but if we didn't, I wouldn't be too upset about it.
The weight of my opposition comes from: you said your about to commit it but there is no evidence of what your going to do about warning users - you are removing a feature some may currently be counting on - and there is no work around for the tests you are breaking that rely on this feature. One fails, and others that count on it that don't fail now, will confuse people when they try and change the config for the tests, but the specified config is not actually being used.
> Remove the option of setting solrconfig from web.xml
> ----------------------------------------------------
>
> Key: SOLR-1647
> URL: https://issues.apache.org/jira/browse/SOLR-1647
> Project: Solr
> Issue Type: Improvement
> Reporter: Noble Paul
> Assignee: Noble Paul
> Fix For: 1.5
>
> Attachments: SOLR-1647.patch
>
>
> with SOLR-1621 , it is not required to have an option to set solrconfig from web.xml. Moreover editing web.xml means hacking solr itself.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (SOLR-1647) Remove the option of setting
solrconfig from web.xml
Posted by "Yonik Seeley (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/SOLR-1647?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12795001#action_12795001 ]
Yonik Seeley commented on SOLR-1647:
------------------------------------
bq. I'd also like to be more clear on what deprecation means in Solr - I know it doesn't mean deprecating and removing before a release even goes out (that makes no sense) - and I'm not sure if it means removing in the next dot release (as Ive see that occurring recently). Can someone pipe in on whether Solr has an (in)formal policy on when deprecations should/can be removed?
We don't really have a policy, but it's been a low priority to remove deprecations if they don't get in the way. I think it would depend a lot on what was deprecated (how expert-level or not).
Specific to this feature, if it isn't causing problems with development of other features, there's no reason to not deprecate it first.
> Remove the option of setting solrconfig from web.xml
> ----------------------------------------------------
>
> Key: SOLR-1647
> URL: https://issues.apache.org/jira/browse/SOLR-1647
> Project: Solr
> Issue Type: Improvement
> Reporter: Noble Paul
> Assignee: Noble Paul
> Fix For: 1.5
>
> Attachments: SOLR-1647.patch
>
>
> with SOLR-1621 , it is not required to have an option to set solrconfig from web.xml. Moreover editing web.xml means hacking solr itself.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (SOLR-1647) Remove the option of setting
solrconfig from web.xml
Posted by "Noble Paul (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/SOLR-1647?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12794327#action_12794327 ]
Noble Paul commented on SOLR-1647:
----------------------------------
I plan to commit this shortly.
> Remove the option of setting solrconfig from web.xml
> ----------------------------------------------------
>
> Key: SOLR-1647
> URL: https://issues.apache.org/jira/browse/SOLR-1647
> Project: Solr
> Issue Type: Improvement
> Reporter: Noble Paul
> Assignee: Noble Paul
> Fix For: 1.5
>
> Attachments: SOLR-1647.patch
>
>
> with SOLR-1621 , it is not required to have an option to set solrconfig from web.xml. Moreover editing web.xml means hacking solr itself.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (SOLR-1647) Remove the option of setting
solrconfig from web.xml
Posted by "Noble Paul (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/SOLR-1647?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12794839#action_12794839 ]
Noble Paul commented on SOLR-1647:
----------------------------------
I'm definitely not going to commit the same patch which is attached. I will ensure that all tests pass before this goes in.
I was trying to find what do others have to say on this.
The warning is going to go in CHANGES.TXT. Do we have to put it anywhere else?
> Remove the option of setting solrconfig from web.xml
> ----------------------------------------------------
>
> Key: SOLR-1647
> URL: https://issues.apache.org/jira/browse/SOLR-1647
> Project: Solr
> Issue Type: Improvement
> Reporter: Noble Paul
> Assignee: Noble Paul
> Fix For: 1.5
>
> Attachments: SOLR-1647.patch
>
>
> with SOLR-1621 , it is not required to have an option to set solrconfig from web.xml. Moreover editing web.xml means hacking solr itself.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (SOLR-1647) Remove the option of setting
solrconfig from web.xml
Posted by "Mark Miller (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/SOLR-1647?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12795009#action_12795009 ]
Mark Miller commented on SOLR-1647:
-----------------------------------
bq. I'm definitely not going to commit the same patch which is attached. I will ensure that all tests pass before this goes in.
I guess this is a communication error. I took {quote}I plan to commit this shortly.{quote} as "you are going to commit the current patch".
I guessed that you might do a few things before committing, but I have no way of knowing. When someone says that they plan to commit something shortly,
I take it to mean something along the lines of the patch posted. As you are missing two things that are pretty major pieces to this patch (the deprecation/non deprecation approach and a good workaround
for the tests), I voiced my opposition to the current approach shown. Its hard for me to guess what changes you will make to this patch before you commit soon - I have to assume when you say that you are committing shortly that perhaps you will address both things correctly and perhaps you won't - you could just commit the current patch, who knows - I can't really rely on you doing anything unless you post the patch first, but you mention nothing of another patch, just of committing. Thats why I brought up the issues that I did. I can say making some last minutes changes to a patch, but these two things are fairly important to this issue I feel, and not really last minute tweaks before a commit.
I'd like the opportunity to take a look at the how you are going to address these two issues and (fwiw) possibly provide feedback.
> Remove the option of setting solrconfig from web.xml
> ----------------------------------------------------
>
> Key: SOLR-1647
> URL: https://issues.apache.org/jira/browse/SOLR-1647
> Project: Solr
> Issue Type: Improvement
> Reporter: Noble Paul
> Assignee: Noble Paul
> Fix For: 1.5
>
> Attachments: SOLR-1647.patch
>
>
> with SOLR-1621 , it is not required to have an option to set solrconfig from web.xml. Moreover editing web.xml means hacking solr itself.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.