[
https://issues.apache.org/jira/browse/SOLR-10329?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15948301#comment-15948301
]
Alexandre Rafalovitch commented on SOLR-10329:
----------------------------------------------
Decisions to make if/when making examples uniform:
* Should multivalued fields be plural? Currently we have *cat* and *features*,
both multivalued
* CamelCase or under_score? Currently we have *inStock* and *content_type*, as
well as *\_src_*
* Do we order definitions in the schema file as uniqueKey, fields, types? Or
types, fields, uniqueKey? Or as whatever rewriting schema does just with more
spaces and comments?
* What are the space rules for XML fie offsets, line continuations, comments,
multi-line comments?
* How much of current over-commented solrconfig.xml to actually keep, now much
to keep by reference to the Reference Guide and how much not bother mentioning
at all?
* Where should examples be stored? Should the *configset* directory be renamed
as we are not treating examples in them as configsets (not shared)
* Should we allow downloading example config directories from the web as part
of *bin/solr* command?
> Rebuild Solr examples
> ---------------------
>
> Key: SOLR-10329
> URL: https://issues.apache.org/jira/browse/SOLR-10329
> Project: Solr
> Issue Type: Wish
> Components: examples
> Reporter: Alexandre Rafalovitch
> Labels: gsoc2017
>
> Apache Solr ships with a number of examples. They evolved from a kitchen sync
> example and are rather large. When new Solr features are added, they are
> often shoehorned into the most appropriate example and sometimes are not
> represented at all.
> Often, for new users, it is hard to tell what part of example is relevant,
> what part is default and what part is demonstrating something completely
> different.
> It would take significant (and very appreciated) effort to review all the
> examples and rebuild them to provide clean way to showcase best practices
> around base and most recent features.
> Specific issues are around kitchen sync vs. minimal examples, better approach
> to "schemaless" mode and creating examples and datasets that allow to create
> both "hello world" and more-advanced tutorials.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
