On 3/4/22 1:33 PM, 6harat wrote:
Hi,
I am not able to hit any limit in terms of uploading a 100MB file and was able to search
the relevant fields inside the doc too.
I hope:
1. The json file that you are trying to upload has a root level key named "docs"
2. You are not trying to fetch the entire document when using solr admin UI.
Thanks for this - #1 - I had a copy/paste error in my load script, and didn't intend to
have that /docs in the wildcard string.
That said, there may be an opportunity to improve this documentation a bit:
https://solr.apache.org/guide/6_6/transforming-and-indexing-custom-json.html
Until now, I didn't realize that that is what f=*/docs*/* was doing in this context -
relating to the structure of the json itself. In retrospect, its pretty confusing to use
/docs in the example there, when the rest endpoint also uses /docs and solr is capturing
documents itself. Plus, the curl example given has json content that doesn't contain a
root level key of docs.
How do you typically handle documentation updates? I think this section would be much
more clear if f=*/docs*/* was changed to something else, and if the json example given
actually had something that demonstrated this.
I can deal with the scaling issues on the hit in the monster document, this is just proof
of concept / demo ware.
Thanks,
Dan
