lucene-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Alexandre Rafalovitch (JIRA)" <>
Subject [jira] [Commented] (LUCENE-6247) artifacts are half a gigabyte
Date Mon, 16 Feb 2015 17:01:13 GMT


Alexandre Rafalovitch commented on LUCENE-6247:

Could we have a parent blocker and children for individual issues to discuss? My strong vote
(if I had any) would be to dump Javadoc from 5.0 (yes, with a re-spin). Explaining it going
away in 5.1 would be a lot more awkward. It's a clear issue with simple solution (just stop
building/bundling). Nothing will break, because nothing relies on those Javadocs. I am already
doing Javadocs online with search built-in (
), so we know that the shipped Javadocs are not actually all that good or useful.

Bundling as a zip/tgz-only is also easy to do (disable the task), but choosing which one to
keep is harder and there might be downstream tasks that rely on a particular format's presence.
But that should not hold up the Javadocs issue. So, maybe that should be for 5.1 with a warning/discussion
on the dev list outside of JIRA.

Other artifacts would probably take even more discussion. Target that for 5.1 or 5.2. I have
absolutely zero opinions on those as I only use Lucene downloads to build the Javadocs. But
I strongly agree on making things smaller. I suspect this discussion would also re-trigger
the discussion on modularity/plugins. Both in relations to ElasticSearch implementation of
plugins and with recent work on Solr towards plugins. Sparks could fly.

> artifacts are half a gigabyte
> -----------------------------
>                 Key: LUCENE-6247
>                 URL:
>             Project: Lucene - Core
>          Issue Type: Bug
>            Reporter: Robert Muir
>            Priority: Blocker
>             Fix For: 5.0
> This is a growing problem and now, its spun out of control.
> The latest release artifacts are half a gigabyte. sorry, I am against adding more retries
to the smoke tester and continuing down the same path (LUCENE-6231).
> Instead I open this blocker issue to discuss fixing this. Whenever i tried to fix it
before (e.g. removing zips), people complained at me "what are you trying to fix" and wouldn't
let me minimize it in the slightest.
> Now the problem is clear, we have a blocker issue to discuss it.

This message was sent by Atlassian JIRA

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message