[jira] [Comment Edited] (SOLR-12591) Ensure ParseDateFieldUpdateProcessorFactory can be used instead of ExtractionDateUtil

2018-08-11 Thread Bar Rotstein (JIRA)


[ 
https://issues.apache.org/jira/browse/SOLR-12591?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16577109#comment-16577109
 ] 

Bar Rotstein edited comment on SOLR-12591 at 8/11/18 9:59 AM:
--

There is another small detail that I am not sure about.
In ExtractionDateUtil the date is added to the document with type Instance, 
while the URP adds the date as a Date type.
The toString() method in each type slightly differ.
Is it OK to save the parsed fields as a Date, even though the XML returned by 
them is different? (since ExtractionDateUtil saves a String representation of 
Instant?)
Perhaps this could be addressed in 
[SOLR-12593|https://issues.apache.org/jira/browse/SOLR-12593], where we could 
use the .toInstant() method of the Date object in some way,
to mimic the behavior found in SolrContentLoader?
{code:java}
Instant date = ExtractionDateUtil.parseDate(val, dateFormats); // may throw
result = date.toString();//ISO format
{code}


was (Author: brot):
There is another small detail that I am not sure about.
In DateExtractionUtil the date is added to the document with type Instance, 
while the URP adds the date as a Date type.
The toString() method in each type slightly differ.
Is it OK to save the parsed fields as a Date, even though the XML returned by 
them is different? (since DateExtractionUtil saves a String representation of 
Instant?)
Perhaps this could be addressed in 
[SOLR-12593|https://issues.apache.org/jira/browse/SOLR-12593], where we could 
use the .toInstant() method of the Date object in some way,
to mimic the behavior found in SolrContentLoader?
{code:java}
Instant date = ExtractionDateUtil.parseDate(val, dateFormats); // may throw
result = date.toString();//ISO format
{code}

> Ensure ParseDateFieldUpdateProcessorFactory can be used instead of 
> ExtractionDateUtil
> -
>
> Key: SOLR-12591
> URL: https://issues.apache.org/jira/browse/SOLR-12591
> Project: Solr
>  Issue Type: Improvement
>  Security Level: Public(Default Security Level. Issues are Public) 
>Reporter: David Smiley
>Assignee: David Smiley
>Priority: Minor
> Fix For: master (8.0)
>
>
> ParseDateFieldUpdateProcessorFactory should ideally be able to handle the 
> cases that ExtractionDateUtil does in the "extraction" contrib module.  Tests 
> should be added, ported from patches in SOLR-12561 that enhance 
> TestExtractionDateUtil to similarly ensure the URP is tested.  I think in 
> this issue, I should switch out Joda time for java.time as well (though leave 
> the complete removal for SOLR-12586) if it any changes are actually necessary 
> – they probably will be.
> Once this issue is complete, it should be appropriate to gut date time 
> parsing out of the "extraction" contrib module – a separate issue. 



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

-
To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: dev-h...@lucene.apache.org



[jira] [Comment Edited] (SOLR-12591) Ensure ParseDateFieldUpdateProcessorFactory can be used instead of ExtractionDateUtil

2018-08-11 Thread Bar Rotstein (JIRA)


[ 
https://issues.apache.org/jira/browse/SOLR-12591?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16577109#comment-16577109
 ] 

Bar Rotstein edited comment on SOLR-12591 at 8/11/18 9:58 AM:
--

There is another small detail that I am not sure about.
In DateExtractionUtil the date is added to the document with type Instance, 
while the URP adds the date as a Date type.
The toString() method in each type slightly differ.
Is it OK to save the parsed fields as a Date, even though the XML returned by 
them is different? (since DateExtractionUtil saves a String representation of 
Instant?)
Perhaps this could be addressed in 
[SOLR-12593|https://issues.apache.org/jira/browse/SOLR-12593], where we could 
use the .toInstant() method of the Date object in some way,
to mimic the behavior found in SolrContentLoader?
{code:java}
Instant date = ExtractionDateUtil.parseDate(val, dateFormats); // may throw
result = date.toString();//ISO format
{code}


was (Author: brot):
There is another small detail that I am not sure about.
In DateExtractionUtil the date is added to the document with type Instance, 
while the URP adds the date as a Date type.
The toString() method in each type slightly differ.
Is it OK to save the parsed fields as a Date, even though the XML return by 
them is different?
Perhaps this could be addressed in 
[SOLR-12593|https://issues.apache.org/jira/browse/SOLR-12593], where we could 
use the .toInstant() method of the Date object in some way,
to mimic the behavior found in SolrContentLoader?
{code:java}
Instant date = ExtractionDateUtil.parseDate(val, dateFormats); // may throw
result = date.toString();//ISO format
{code}

> Ensure ParseDateFieldUpdateProcessorFactory can be used instead of 
> ExtractionDateUtil
> -
>
> Key: SOLR-12591
> URL: https://issues.apache.org/jira/browse/SOLR-12591
> Project: Solr
>  Issue Type: Improvement
>  Security Level: Public(Default Security Level. Issues are Public) 
>Reporter: David Smiley
>Assignee: David Smiley
>Priority: Minor
> Fix For: master (8.0)
>
>
> ParseDateFieldUpdateProcessorFactory should ideally be able to handle the 
> cases that ExtractionDateUtil does in the "extraction" contrib module.  Tests 
> should be added, ported from patches in SOLR-12561 that enhance 
> TestExtractionDateUtil to similarly ensure the URP is tested.  I think in 
> this issue, I should switch out Joda time for java.time as well (though leave 
> the complete removal for SOLR-12586) if it any changes are actually necessary 
> – they probably will be.
> Once this issue is complete, it should be appropriate to gut date time 
> parsing out of the "extraction" contrib module – a separate issue. 



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

-
To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: dev-h...@lucene.apache.org



[jira] [Comment Edited] (SOLR-12591) Ensure ParseDateFieldUpdateProcessorFactory can be used instead of ExtractionDateUtil

2018-08-07 Thread Bar Rotstein (JIRA)


[ 
https://issues.apache.org/jira/browse/SOLR-12591?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16572226#comment-16572226
 ] 

Bar Rotstein edited comment on SOLR-12591 at 8/7/18 7:54 PM:
-

Oh yes it will be a very interesting experiment.

Hopefully I'll have time to get to it this weekend.


was (Author: brot):
Oh yes it will be a very interesting experiment.

Hopefully I'll have time to get to it this wrekend.

> Ensure ParseDateFieldUpdateProcessorFactory can be used instead of 
> ExtractionDateUtil
> -
>
> Key: SOLR-12591
> URL: https://issues.apache.org/jira/browse/SOLR-12591
> Project: Solr
>  Issue Type: Improvement
>  Security Level: Public(Default Security Level. Issues are Public) 
>Reporter: David Smiley
>Assignee: David Smiley
>Priority: Minor
> Fix For: master (8.0)
>
>
> ParseDateFieldUpdateProcessorFactory should ideally be able to handle the 
> cases that ExtractionDateUtil does in the "extraction" contrib module.  Tests 
> should be added, ported from patches in SOLR-12561 that enhance 
> TestExtractionDateUtil to similarly ensure the URP is tested.  I think in 
> this issue, I should switch out Joda time for java.time as well (though leave 
> the complete removal for SOLR-12586) if it any changes are actually necessary 
> – they probably will be.
> Once this issue is complete, it should be appropriate to gut date time 
> parsing out of the "extraction" contrib module – a separate issue. 



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

-
To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: dev-h...@lucene.apache.org