[
https://issues.apache.org/jira/browse/LUCENE-6188?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14286584#comment-14286584
]
Ramkumar Aiyengar commented on LUCENE-6188:
-------------------------------------------
Rob, the logic I have nuked is actually not as a duplicate of doclint (i just
didnt check that, and as you mention there might be differences) but the
checkJavadocLinks.py script which is run prior to this script in the
documentation-lint. That does the exact same check in Python, except it uses a
real parser rather than regex hacks..
> Remove HTML verification from checkJavaDocs.py
> ----------------------------------------------
>
> Key: LUCENE-6188
> URL: https://issues.apache.org/jira/browse/LUCENE-6188
> Project: Lucene - Core
> Issue Type: Improvement
> Components: general/javadocs
> Reporter: Ramkumar Aiyengar
> Assignee: Erick Erickson
> Priority: Minor
> Attachments: LUCENE-6188.patch, LUCENE-6188.patch
>
>
> Currently, the broken HTML verification in {{checkJavaDocs.py}} has issues in
> some cases (see SOLR-6902).
> On looking further to fix it with the {{html.parser}} package instead,
> noticed that there is broken HTML verification already present (using
> {{html.parser}}!)in {{checkJavadocLinks.py}} anyway which takes care of
> validation, and probably {{jTidy}} does it as well, going by the output
> (haven't verified it).
> Given this, the validation in {{checkJavaDocs.py}} doesn't seem to add any
> further value, so here's a patch to just nuke it instead.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
