Thank you Ishan, but I'm not interested in contributing to core. I'll try
to do stuff in plugins.

So, my remaining questions:

1. a recommended guide/tutorial/blog-post/book on plugin-dev or plugin that
I can see/duplicate (i'm only seeing very old blog stuff for maximum solr5)

2. Is it possible to create a new router in a plugin ?

3. Is it possible to change how limit() works and implement that in a
plugin ?

Thank You,
Dorian


On Tue, Apr 18, 2017 at 9:30 PM, Ishan Chattopadhyaya <
[email protected]> wrote:

> https://wiki.apache.org/solr/HowToContribute
>
> On Wed, Apr 19, 2017 at 12:41 AM, Dorian Hoxha <[email protected]>
> wrote:
>
>> Hey friends,
>>
>> I'll be having some free time in the next 2 weeks and would like to get
>> up to speed on hacking on solr. I've done very little java on university so
>> I'd like to get on speed on java too together with that.
>>
>> So if you can give me some high level pointers if possible on:::
>>
>> 1. getting up to speed on java
>> 2. environment (like i have idea ide, ubuntu 16.04, java 1.8)
>> 3. I'd like to embed solr but only to make distributed requests and not
>> store data. So, my java-webapp calls a solr-function in the same jvm, which
>> doesn't store data itself but has the state of all other nodes and it can
>> do the merging of results. This way I make 1 less http-request for each
>> distributed-search. Something like on elasticsearch with search-only-nodes
>> but also to include it in my java webapp.
>> 4. A way to create a module to subclass router.compositeid to create my
>> own logic on routing documents *(most important). *
>> 5. A way to create a module (or something else?) so I can change the
>> sorting of documents to not limit() docs if the last ones have the same
>> sort-value (so doing this on each shard, and also on the merging). Meaning
>> if we have limit(10), but docs 10-20 have the same sort-value, we return 20
>> (as long as 20 < max_limit(), which does the same cut as limit() currently
>> does). I think currently it uses a min-heap, so maybe a similar data
>> structure that also keeps duplicates at the end.
>> 6. How to test all this stuff ? Or just create test cases and make sure
>> they keep working ?
>>
>> Hope I'm not asking too much.
>>
>> Regards,
>> Dorian
>>
>
>

Reply via email to