[
https://issues.apache.org/jira/browse/SOLR-133?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12469603
]
J.J. Larrea commented on SOLR-133:
----------------------------------
It would be useful if there first were some consensus as to what the goals are
for making a change to the XML Update Handler; some possibilities I can think
of include:
1) To use standards-based rather than non-standards-based technologies as much
as possible
2) To use as few different XML technologies (and coding styles related to the
technology) as possible
3) To reduce as much as possible the complexity of code needed for interpreting
XML command and/or configuration streams
4) To lower resource consumption and limitations for XML handling, e.g.
stream-based rather than random-access
By all means add to that list, prioritize, and remove goals which are not seen
as important.
Then it seems to me the question would be how many of those goals are addressed
by changing XML Update Handler to stAX, vs. other technologies. One might at
the same time also want to look at other places where SOLR decodes XML such as
config files, to see if there can be more commonality rather than continued
isolation.
> change XmlUpdateRequestHandler to use StAX instead of XPP
> ---------------------------------------------------------
>
> Key: SOLR-133
> URL: https://issues.apache.org/jira/browse/SOLR-133
> Project: Solr
> Issue Type: Improvement
> Reporter: Hoss Man
>
> there has been discussion of using StAX for XML parsing of updates instead of
> XPP ... opening an issue to track it as a possible improvement (orriginally
> mentioned in SOLR-61, but that task was more specificly about refactoring the
> existing code)
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
