Hi everyone, I was wondering if we have idea when v0.30 will be released? I'm really looking for https://github.com/typesense/typesense/issues/1657 ;)

completely typesense unrelated; anyone here experienced with adding jsdoc typedefs to a third party library?

Hello team! I have 2 questions regarding the

typesense-java

library. Using

v1.3.0

1. Trying to use the JOIN feature, but found an obstacle in the way the referenced field is parsed: a. the naming of the referenced collection is namespaced, using this pattern -> namespace.x b. when referencing the field y, on the definition of the collection using

Field().reference("namespace.x.y")

, I get the error that collection “namespace” is not found c. I get that the string is split in the first dot, so can’t get the correct collection name, does anybody know any way to escape this? 2. Is there any way to enable async joins through the library api?

Dont know if its just anecdotal and temporary, but never experienced before. After I updated my local test environment from typesense/typesense:30.0.rca33 -> typesense/typesense:30.0.rca34, for some reason dropping collection has become incredible slow. Almost to the point where I am not sure if it will remove it at all, but then after 5 minutes it might be gone. So dont have any reproducable gist etc, and maybe its just a problem of size so need bigger collection to reproduce. but with exactly the same data it was never a problem, always took a few seconds before. Here I dropped a pretty small collection that normally takes 1 second.. dont know if logs makes any sense.. but I had to try delete several times, and it still was hanging around, and then all of sudden it was finally gone.

I20251212 09:23:07.337235  1379 raft_server.h:60] Peer refresh succeeded!
I20251212 09:23:17.337924  1177 raft_server.cpp:754] Term: 17, pending_queue: 0, last_index: 1653120, committed: 1653120, known_applied: 1653120, applying: 0, pending_writes: 0, queued_writes: 6, local_sequence: 4844127001
I20251212 09:23:17.338011  1408 raft_server.h:60] Peer refresh succeeded!
I20251212 09:23:27.338658  1177 raft_server.cpp:754] Term: 17, pending_queue: 0, last_index: 1653120, committed: 1653120, known_applied: 1653120, applying: 0, pending_writes: 0, queued_writes: 6, local_sequence: 4844750001
I20251212 09:23:27.338727  1235 raft_server.h:60] Peer refresh succeeded!
I20251212 09:23:33.298519  1178 batched_indexer.cpp:432] Running GC for aborted requests, req map size: 6, reference_q.size: 0
I20251212 09:23:37.339421  1177 raft_server.cpp:754] Term: 17, pending_queue: 0, last_index: 1653120, committed: 1653120, known_applied: 1653120, applying: 0, pending_writes: 0, queued_writes: 6, local_sequence: 4845430001
I20251212 09:23:37.339488  1410 raft_server.h:60] Peer refresh succeeded!
I20251212 09:23:38.383803  1321 http_data.h:387] event=slow_request, time=252687 ms, client_ip=172.19.0.1, endpoint=DELETE /collections/products_dk?collection=products_dk&x-typesense-user-id=172.19.0.1&, body=
I20251212 09:23:38.411680  1321 core_api.cpp:410] Invalidated schema prompt cache for collection: products_dk
I20251212 09:23:38.411854     1 http_data.h:387] event=slow_request, time=247613 ms, client_ip=172.19.0.1, endpoint=DELETE /collections/products_dk?collection=products_dk&x-typesense-user-id=172.19.0.1&, body=
I20251212 09:23:38.413399  1321 core_api.cpp:410] Invalidated schema prompt cache for collection: products_dk
I20251212 09:23:38.413451     1 http_data.h:387] event=slow_request, time=191666 ms, client_ip=172.19.0.1, endpoint=DELETE /collections/products_dk?collection=products_dk&x-typesense-user-id=172.19.0.1&, body=
I20251212 09:23:38.414960  1321 core_api.cpp:410] Invalidated schema prompt cache for collection: products_dk
I20251212 09:23:38.414989     1 http_data.h:387] event=slow_request, time=188705 ms, client_ip=172.19.0.1, endpoint=DELETE /collections/products_dk?collection=products_dk&x-typesense-user-id=172.19.0.1&, body=
I20251212 09:23:38.416502  1321 core_api.cpp:410] Invalidated schema prompt cache for collection: products_dk
I20251212 09:23:38.416528     1 http_data.h:387] event=slow_request, time=186568 ms, client_ip=172.19.0.1, endpoint=DELETE /collections/products_dk?collection=products_dk&x-typesense-user-id=172.19.0.1&, body=
I20251212 09:23:38.418022  1321 core_api.cpp:410] Invalidated schema prompt cache for collection: products_dk
I20251212 09:23:38.418047     1 http_data.h:387] event=slow_request, time=147989 ms, client_ip=172.19.0.1, endpoint=DELETE /collections/products_dk?collection=products_dk&x-typesense-user-id=172.19.0.1&, body=
I20251212 09:23:47.340145  1177 raft_server.cpp:754] Term: 17, pending_queue: 0, last_index: 1653120, committed: 1653120, known_applied: 1653120, applying: 0, pending_writes: 0, queued_writes: 0, local_sequence: 4845511705
I20251212 09:23:47.340224  1379 raft_server.h:60] Peer refresh succeeded!
I20251212 09:23:57.340984  1177 raft_server.cpp:754] Term: 17, pending_queue: 0, last_index: 1653120, committed: 1653120, known_applied: 1653120, applying: 0, pending_writes: 0, queued_writes: 0, local_sequence: 4845511705
I20251212 09:23:57.341075  1408 raft_server.h:60] Peer refresh succeeded!
I20251212 09:24:07.341699  1177 raft_server.cpp:754] Term: 17, pending_queue: 0, last_index: 1653120, committed: 1653120, known_applied: 1653120, applying: 0, pending_writes: 0, queued_writes: 0, local_sequence: 4845511705
I20251212 09:24:07.341790  1235 raft_server.h:60] Peer refresh succeeded!
I20251212 09:24:17.342546  1177 raft_server.cpp:754] Term: 17, pending_queue: 0, last_index: 1653120, committed: 1653120, known_applied: 1653120, applying: 0, pending_writes: 0, queued_writes: 0, local_sequence: 4845511705
I20251212 09:24:17.342624  1410 raft_server.h:60] Peer refresh succeeded!
I20251212 09:24:27.343371  1177 raft_server.cpp:754] Term: 17, pending_queue: 0, last_index: 1653120, committed: 1653120, known_applied: 1653120, applying: 0, pending_writes: 0, queued_writes: 0, local_sequence: 4845511705
I20251212 09:24:27.343441  1379 raft_server.h:60] Peer refresh succeeded!
I20251212 09:24:34.302376  1178 batched_indexer.cpp:432] Running GC for aborted requests, req map size: 0, reference_q.size: 0

Do with this info as you will.. maybe anyone from development team have some idea in case something changed between rca33 and rca34 that might be relevant. So just leavning it here if its useful.

The collection I dropped here (products_dk) was not big.. dont remember, maybe 10.000 records or so.. the other I dropped earlier that took very long time was maybe 500.000 records and has some embeddings etc..

But point is more that identical size as really fast before, that is the alarming part.. so maybe it just waits longer for something to complete now that happened async before, or its really some problem.

A bit strange.. now it feels like its a little bit more OK for some other collections.. so maybe its embedding collections specifically that have problem? Anyway.. I will re-create and fill with data and report back if it behaves the same or not again. Just wanted to share in case it ringed some bell for devs if any part of code is touched that can affect this somehow between rca releases.

Hey Folks - has anyone had any luck setting up an MCP integration with Typesense?

Hey peeps, wanted to ask for a Cloud cluster, what should I expect the RAM behaviour to be if I drop a collection? In order to facilitate some schema changes, fitting the modified schema required upgrading the cluster twice. I was hoping to downgrade after that but it seems the RAM usage hasn't gone down enough where I could reasonably downgrade. Before starting, we were on a 2GB cluster. Now on 8 because the data juuuuuust just went over 4GB. I've dropped the old collection and migrated usage to the new one, but haven't yet seen a significant RAM drop in ~30min. Just want to make sure I'm not being impatient with it. Collection is roughly 900k records

Hey all, I'm setting up TypeSense for ecommerce product search. I have it up on a dev copy of the live database. I am running some tests on a copy of our live data, which includes historic sales. Some of the results are great, and the speed is great. I have a few issues though on trying to work out the best way to slightly promote popular products though (without pinning). I have tried to read up on the official website some techniques but I am struggling to get the results I need. For some Very generic searches which bring back a lot of products (say about 5-10k), The search scores of the tops results are Very similar for a large range of products, but hundreds if not thousands of the 1st ones have no sales at all and so are likely not the best fit. What I would like is to be able to influence the results without doing anything drastic like forcing sold products to the top. Is there a way to make sales stats increase ranking without taking over it? My concerns are: • Result size varies a lot from thousands to 10s, so fixed rules on X size buckets seems unlikely to fit all • Just because a product has sold doesnt mean it should be at the top, so I dont want to blanket push all to the top. Ideally I would be able to say something like 10% of score is based on sold and 90% is based on relevance for example I have tried buckets but they dont seem to give desired results in my tests and doesnt scale well with variable and unknown result sizes. I tried looking here: https://typesense.org/docs/guide/ranking-and-relevance.html#ranking-based-on-relevance-and-popularity Thanks for any advice

Im trying to work out Dynamic filtering. If I add a filter with { "rule": { "query": "{attr_brand}", "match": "contains" }, "filter_by": "attr_brand:={attr_brand}", "remove_matched_tokens": true } It does nothing. In contrast without dynamic.. eg this "works" { "rule": { "query": "something", "match": "contains" }, "filter_by": "attr_brand:=something", "remove_matched_tokens": true } What could I be doing wrong? if it matters attr_brand is a dynamic property ie in the schema I have attr_*. This is the fields properties: {"facet":true,"index":true,"infix":false,"locale":"","name":"attr_brand","optional":true,"sort":false,"stem":false,"stem_dictionary":"","store":true,"type":"string"}

Any tips for improving the time it takes to reload/reindex when a node in a cluster restarts? We have about 75 million documents roughly 1k each and it takes over an hour for a node to be healthy after it restarts. We have a three node cluster with 40GB RAM and 8 CPU for each node. Looking at the CPU and memory usage when the node is coming back up we are no where CPU or memory limits although it does seem like a single could be close to maxed out while the others are mostly idle. Top shows the brpc_wkr process using 75%. We tried breaking up the collection into eight smaller collections and also tried increasing num-documents-parallel-load but neither made a difference. Anything else we should try?

Hello, i am currently testing typesense locally on my mac (i am still new to it). I installed it and migrated my data. In my schema i have this field:

{ name: "sku_raw", type: "string", infix: true },

i am testing a search query like this:

const res = await client.collections("products").documents().search({

"gdx",

query_by: "sku",

prefix: true,

per_page: 20,

page,

});

in my collection i have documents with sku like "gdx1234" and others like "ngdx1234". My expectation is to get all documents that have these patterns or at least get the ones that have sku field starting with "gdx". i am getting only 10 results even though i have 2557 documents with sku field starting with "gdx".

Hey guys, I am trying setup no hits query analytics, and it doesn’t work. Here is my rule:

{
  "rules": [
    {
      "name": "listings_no_hits_rule",
      "params": {
        "destination": {
          "collection": "listings_no_hits_queries"
        },
        "limit": 1000,
        "source": {
          "collections": [
            "listings"
          ]
        }
      },
      "type": "nohits_queries"
    }
  ]
}

I am not sure if it because we don’t have collections named listings(it’s an alias to another collection). Where is a way to debug it?

hi, is there a good way to deal with a specific term I want to deal with in a special way.. Say there is a brand name made up of 2 words "Banana Computers". Is there a way for me to not hit documents with the whole phrase "Banana Computers" when searching just "Banana" ? The only thing I can think of is replacing all occurrences of "Banana Computers" before indexing with say "BananaComputers" and then likewise on searching. I don't think I can use synonyms here right as its 2 words ? Thanks

Hi, I'm facing a bit of an issue with q and query_by parameters. I have a set of records in my collection with "Sample dispenser" in the description field and I've been experiencing varying match patterns, in relation to the content of query_by. • When I use query_by="description", everything works as expected. • When I use query_by="*" , the number of matches becomes quite random

Hi again, I have a question regarding Multisearch with Union and pagination. Today the documentation gives this example:

curl '<http://localhost:8108/multi_search?page=1&per_page=10>' -X POST \
     -H "X-TYPESENSE-API-KEY: ${TYPESENSE_API_KEY}" -d '
{
  "union": true,
  "searches": [
    {
      "collection": "products",
      "q": "wireless headphones",
      "query_by": "name,description"
    },
    {
      "collection": "articles",
      "q": "wireless headphones",
      "query_by": "title,content"
    }
  ]
}'

Passing page/per_page in the URL looks a bit strange in that payload IMHO. Is there a way to pass them in the JSON payload?

Hi, Got this error

Number of multi searches exceeds limit_multi_searches parameter.

, is it possible to increase the limit on our side?

Good morning, quick question about Union (federated) search + `group_by`: when I use

group_by

to aggregate results in a union search the

found

-integer returns an incorrect number of results. In my case it is

84

, whereas the actual (grouped) hits only has a size of

7

. I can't use pagination on our website without the correct number of results. Is there a way to obtain the correct number of actual results?

I was wondering if all queries are diacritic-insensitive with default locales of "en" both in the query and in the search index. I am noticing that I am not getting results that don't exactly match diacritics even though I never set the locale of my fields.

Hello @Kishore Nallan @Jason Bosco - Good Morning/Evening. Currently I'm facing an issue when trying to filter documents using Starts With (ID_KEY:=CS*) along with other filters (STATUS:=[

ACTIVE

] && ARCHIVED:=[

NO

]); I've tried with and without max_filter_by_candidates parameter; Still *no luck*; But, If I've filetered with Starts with (ID_KEY:=CS*) and max_filter_by_candidates then it returns records; If I've filtered with (STATUS:=[

ACTIVE

] && ARCHIVED:=[

NO

]) alone then also I'm getting the reccords When I've combined both, I'm not getting the expected results; But, I'm sure that, we have records for the given condition. Tested with Version 29.0 Please let me know if I've missing something here; Your advice on this would be really appreciated. Looking forward to hear from you. Here is the sample payload.

{
  "searches": [
    {
      "query_by": "*",
      "sort_by": "ID_KEY:asc",
      "highlight_full_fields": "*",
      "collection": "CASE_DOCS",
      "q": "*",
      "filter_by": "STATUS:=[`ACTIVE`] && ARCHIVED:=[`NO`] && ID_KEY:=CS*",
      "max_filter_by_candidates": 10000,
      "page": 1,
      "per_page": 10
    }
  ]
}

Hey guys! is it possible to use the _eval sorting with more than 1 group? Something like:

_eval([(category:shoes):1, (category:shirts):2] && [(brand:nike):1, (brand:adidas):2]):asc

I've only found examples with 1 group. When I tried with 2, I got the error: `Could not find a field named `` in the schema for sorting.`

Hi, I'm using the go library to fetch from Typesense, however I've hit a problem. When doing a standard multi search the results return and map fine, however, with union:true, I get no results. I think this is due to the different format of the response not being handled by the existing

api.MultiSearchResponse

. Is this a known issue or am I using it incorrectly?

In

multi_search

, when using

q="*"

, reference joins used in

filter_by

are not returned in the response even if explicitly requested via

include_fields

(e.g.

*, $Ref_collection(*)

). The same query works in single search, and also works in

multi_search

when a non-wildcard or search query keyword is used. Is join hydration intentionally skipped for wildcard queries in

multi_search

, or is there a way to force it / should this be documented? @Jason Bosco

Morning, I'm looking to self-host a typesense cluster in kubernetes. Does anyone have any experience? What are the pitfalls and things to look out for? I've been using the tyko operator and so far so good but I did run into an issue with resource limits and statefulsets.

urgent issue, scaled our typsense collection and now one of our collections is unreachable, with an error about a missing synonym set and curation - i think scaling the size up might have deleted the curation and synonym sets and maybe even the connection?

Hi, I was wondering if there are any plan to support more than 3 parameters in sorting?

Hi Team Greetings of the day Cluster Id hd1q4bkjatlm5pgzp We are getting an out-of-memory error in the development environment, even though the data is low. We are working on it. Please do the needful as soon as possible

Hi Team, We are trying to use faceting support on joined collection in

typesense/typesense:30.0.rca35

. But, facing an issue with returned facet counts. On running multisearch, faceting on joined collection fields

($collection(field))

returns incorrect facet counts. It appears to be counting unique facet values (2 distinct types) rather than the actual document counts per facet value. Faceting on native fields works correctly, it returns correct facet counts. Our Usecase: There are two collections, Collection A and Collection B, with a one-to-many relationship from A to B, where the foreign key/reference is stored in Collection B. Our queries are executed on Collection B, joined with Collection A, and specific fields from Collection A are exposed as facets.

Hello ! quick question, is there an official typesense dashboard ?