[
https://issues.apache.org/jira/browse/SOLR-534?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12796982#action_12796982
]
Lisa Carter commented on SOLR-534:
----------------------------------
I would argue that REALLY_BIG_NUMBER is actually significantly MORE dangerous
than a crash.
Here's why: A crash at least lets the programmer know something went wrong.
Missing data is a silent failure.
1) If the result set is too large for the client, it will run out of memory and
generate an exception. The programmer will immediately know they did something
wrong.
2) If the result set is too large for the network (unlikely) this will
disconnect and fail. The programmer will immediately know they did something
wrong.
3) If the result set is too large for solr, solr should not crash but rather
return a page with the standard error handler "result set too large"/"out of
memory". The programmer will immediately know they did something wrong. Solr
sure as heck better be checking this already--you never know when you'll run
into bizarre low memory conditions;allocations should ALWAYS be checked for.
But if you use the REALLY_BIG_NUMBER approach, the same bad programmer who
never thought he would get back more than a 1000 records will never check
whether the result set contains more than 1000 records either. If the
programmer was expecting the complete result set and the database now contains
1002 records instead of 999, they will not know there is a problem... the last
records in the set are simply truncated. The programmer who wrote the code may
not be the person maintaining the application, quite common in production
environments. The maintenance person may not know for weeks or months that a
problem even exists!
The -1 approach ensures immediate, loud failure.
The REALLY_BIG_NUMBER ensures only silent failure.
While it's impossible to idiot-proof everything, loud failure is always
preferable to silent failure. Barking loudly saves the poor soul who maintains
the idiot's code a lot of heartache.
> Return all query results with parameter rows=-1
> -----------------------------------------------
>
> Key: SOLR-534
> URL: https://issues.apache.org/jira/browse/SOLR-534
> Project: Solr
> Issue Type: New Feature
> Components: search
> Affects Versions: 1.3
> Environment: Tomcat 5.5
> Reporter: Lars Kotthoff
> Priority: Minor
> Attachments: solr-all-results.patch
>
>
> The searcher should return all results matching a query when the parameter
> rows=-1 is given.
> I know that it is a bad idea to do this in general, but as it explicitly
> requires a special parameter, people using this feature will be aware of what
> they are doing. The main use case for this feature is probably debugging, but
> in some cases one might actually need to retrieve all results because they
> e.g. are to be merged with results from different sources.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
