You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@lucene.apache.org by "Hoss Man (JIRA)" <ji...@apache.org> on 2017/05/11 00:07:04 UTC

[jira] [Created] (LUCENE-7821) classic parser range query parsing is sloppy about "TO"

Hoss Man created LUCENE-7821:
--------------------------------

             Summary: classic parser range query parsing is sloppy about "TO"
                 Key: LUCENE-7821
                 URL: https://issues.apache.org/jira/browse/LUCENE-7821
             Project: Lucene - Core
          Issue Type: Bug
            Reporter: Hoss Man


A post on the solr-user mailing list drew my attention to the fact that this is apparently a valid range query under the QueryParser.jj grammer (both for the classic parser and the solr variant -- i didn't check flexible)...

{noformat}
[x z]       // parsed the same as [x TO z]
{noformat}

it's parsed as a valid range query even though there is no {{ TO }} -- even though there is nothing in the docs to suggest that the {{ TO }} is intended to be optional.

Furthermore, in my experimenting i realized that how the grammer looks for the {{ TO }} keyword seems to be a bit sloppy, making some range queries that should be easy to validate (because they are unambiguous) fail to parse...

{noformat}
[TO TO z]     // fails
[a TO TO]     // fails
[a TO "TO"]   // works, but why should quoting be neccessary here?
{noformat}

----

If the "sloppy" parsing behavior is intentional, then the docs should reflect that the {{ TO }} is optional -- but it seems like in general we should make these other unambiguous cases parse cleanly.



--
This message was sent by Atlassian JIRA
(v6.3.15#6346)

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