🎉 New Destination Vector Database (powered by LangChain) #26184

flash1293 · 2023-05-17T10:28:09Z

How to test

Have local Airbyte running
Check out this PR
Go to airbyte-integrations/connectors/destination-langchain
Run docker build . -t airbyte/destination-langchain:dev
Add custom connector (no need for a custom registry here as evertyhing is happening on your machine) - repository name is airbyte/destination-langchain, image tag is dev
Configure a source, a langchain destination and start syncing data!

What

This PR adds the "Langchain" destination which combines the langchain helpers for processing, embedding and indexing to build a unified destination for various vector stores that takes care of the embedding on the fly and makes it super easy to get started with building integrations on top of langchain.

https://www.loom.com/share/db63e997d01d49118bf382dff876f7c1

How

General class setup implementing the main functionality:

Processor - turns records into langchain documents that can be indexed into vector stores
- Split into chunks of defined length
- Prepare metadata
Embedder (abstract super class) - provides means to embed documents, turning them into a vector
Indexer (abstract super class) - receives a batch of chunks and indexes them into the vector database. Uses the embedder instance
Batcher - collects records up to batch size, then processes them
Destination - initializes other components, passes records to the batcher and implements the main processing loop (call processor to get chunks, pass to indexer to push into vector store)

Unclear parts / TODOs

It's annoying text fields are configured on the global level - they are probably not known at this point of the setup and will apply to all streams at the same time. However, it's important to configure the right fields for the embedding
Basic multi-threading using queues between processing, embedding and indexing could speed up the throughput quite a bit, but would also complicate the setup - however it's probably worth it implementing (doesn't have to be part of the initial release)
Adding an option to do local embedding would be nice so it's possible to run the destination without singing up for any service (e.g. llama-cpp or fake embeddings)
Add more unit tests
Add basic integration tests

New Connector

Todos

github-actions · 2023-05-17T10:28:30Z

Before Merging a Connector Pull Request

Wow! What a great pull request you have here! 🎉

To merge this PR, ensure the following has been done/considered for each connector added or updated:

PR name follows PR naming conventions
Breaking changes are considered. If a Breaking Change is being introduced, ensure an Airbyte engineer has created a Breaking Change Plan and you've followed all steps in the Breaking Changes Checklist
Connector version has been incremented in the Dockerfile and metadata.yaml according to our Semantic Versioning for Connectors guidelines
Secrets in the connector's spec are annotated with airbyte_secret
All documentation files are up to date. (README.md, bootstrap.md, docs.md, etc...)
Changelog updated in docs/integrations/<source or destination>/<name>.md with an entry for the new version. See changelog example
You, or an Airbyter, have run /test successfully on this PR - or on a non-forked branch
You, or an Airbyter, have run /publish successfully on this PR - or on a non-forked branch
You've updated the connector's metadata.yaml file new!

If the checklist is complete, but the CI check is failing,

Check for hidden checklists in your PR description
Toggle the github label checklist-action-run on/off to re-run the checklist CI.

…-destination

sherifnada

I think this is a great start

My main advice would be to keep in mind how this should be structured to achieve maximum throughput since it seems the current bandwidths we're hitting are actually bad. Not something for v1 but probably an important thing to keep in mind.

I noticed a couple of things seem hard-bound to specific providers e.g: OpenAI embeddings, probably make our life easier if we distinguish between generic stuff and concrete implementations

Also we should have way more unit tests ;)

airbyte-integrations/connectors/destination-langchain/metadata.yaml

sherifnada · 2023-07-04T00:16:22Z

airbyte-integrations/connectors/destination-langchain/destination_langchain/config.py

+        schema_extra = {"group":"processing"}
+
+
+class EmbeddingConfigModel(BaseModel):


should probably be called OpenAI embedding Config or something

I added a second option for embeddings (fake embeddings that can be used to test the pipeline) and generalized these parts

sherifnada · 2023-07-04T00:16:51Z

airbyte-integrations/connectors/destination-langchain/destination_langchain/config.py

+import re
+
+
+class ProcessingConfigModel(BaseModel):


is spec.json generated from this file?

It's generated on the flow, removed the old spec.json

airbyte-integrations/connectors/destination-langchain/destination_langchain/config.py

sherifnada · 2023-07-04T00:22:28Z

airbyte-integrations/connectors/destination-langchain/destination_langchain/destination.py

+    ) -> Iterable[AirbyteMessage]:
+        config_model = ConfigModel.parse_obj(config)
+        self._init_indexer(config_model)
+        self.processor = Processor(config_model.processing, configured_catalog, max_metadata_size=self.indexer.max_metadata_size)


nit: the name Processor is ambiguous

renamed to docment_processor

sherifnada · 2023-07-04T00:27:25Z

airbyte-integrations/connectors/destination-langchain/destination_langchain/processor.py

+        return relevant_fields
+
+    def _extract_metadata(self, record: AirbyteRecordMessage) -> dict:
+        metadata = record.data.copy()


this might be a heavy operation on performance (tomorrow problem though)

sherifnada · 2023-07-04T00:28:37Z

airbyte-integrations/connectors/destination-langchain/destination_langchain/processor.py

+            relevant_fields = record.data
+        return relevant_fields
+
+    def _extract_metadata(self, record: AirbyteRecordMessage) -> dict:


if i'm reading this correctly metadata is everything in the input record that isn't a text/data field?

right, we could also make it configurable. Technically you could use the column selection feature for this.

sherifnada · 2023-07-04T00:32:21Z

airbyte-integrations/connectors/destination-langchain/destination_langchain/measure_time.py

@@ -0,0 +1,26 @@
+import time
+
+def measure_time(func):


neat use of decorators

sherifnada · 2023-07-04T00:37:02Z

airbyte-integrations/connectors/destination-langchain/destination_langchain/processor.py

+METADATA_STREAM_FIELD = "_airbyte_stream"
+METADATA_NATURAL_ID_FIELD = "_natural_id"
+
+class Processor:


naive question. Why do we need this if embedding APIs can tokenize themselves?

Tokenization is just turning utf8 encoded text into a sequence of one-hot vector representations and as you are saying the embedding API takes care of that: https://help.openai.com/en/articles/4936856-what-are-tokens-and-how-to-count-them

What the processor is doing is turning the records into "documents":

Split them up appropriately

Separate text and metadata

Add bookkeeping metadata for incremental sync (id and stream name)

sherifnada · 2023-07-04T00:39:19Z

airbyte-integrations/connectors/destination-langchain/destination_langchain/embedder.py

+
+    @property
+    def embedding_dimensions(self) -> int:
+        return 1536


That's the size of the embedding vectors of the openai model, added a comment to explain

octavia-squidington-iii · 2023-07-05T12:29:22Z

destination-langchain test report (commit `bddd100c3d`) - ❌

⏲️ Total pipeline duration: 60 seconds

Step	Result
Validate airbyte-integrations/connectors/destination-langchain/metadata.yaml	❌
Connector version semver check	✅
Connector version increment check	✅
QA checks	❌
Code format checks	❌
Connector package install	❌

🔗 View the logs here

Please note that tests are only run on PR ready for review. Please set your PR to draft mode to not flood the CI engine and upstream service on following commits.
You can run the same pipeline locally on this branch with the airbyte-ci tool with the following command

airbyte-ci connectors --name=destination-langchain test

octavia-squidington-iii · 2023-07-05T12:33:36Z

destination-langchain test report (commit `6126f4f302`) - ❌

⏲️ Total pipeline duration: 30 seconds

Step	Result
Validate airbyte-integrations/connectors/destination-langchain/metadata.yaml	❌
Connector version semver check	✅
Connector version increment check	✅
QA checks	❌
Code format checks	❌
Connector package install	❌

🔗 View the logs here

Please note that tests are only run on PR ready for review. Please set your PR to draft mode to not flood the CI engine and upstream service on following commits.
You can run the same pipeline locally on this branch with the airbyte-ci tool with the following command

airbyte-ci connectors --name=destination-langchain test

octavia-squidington-iii · 2023-07-06T14:00:58Z

destination-langchain test report (commit `099ad208a9`) - ✅

⏲️ Total pipeline duration: 33.58s

Step	Result
Validate airbyte-integrations/connectors/destination-langchain/metadata.yaml	✅
Connector version semver check	✅
QA checks	✅
Code format checks	✅
Connector package install	✅
Build destination-langchain docker image for platform linux/x86_64	✅
Unit tests	✅
Integration tests	✅

🔗 View the logs here

Please note that tests are only run on PR ready for review. Please set your PR to draft mode to not flood the CI engine and upstream service on following commits.
You can run the same pipeline locally on this branch with the airbyte-ci tool with the following command

airbyte-ci connectors --name=destination-langchain test

octavia-squidington-iii · 2023-07-06T14:47:47Z

destination-langchain test report (commit `099ad208a9`) - ✅

⏲️ Total pipeline duration: 22.19s

Step	Result
Validate airbyte-integrations/connectors/destination-langchain/metadata.yaml	✅
Connector version semver check	✅
QA checks	✅
Code format checks	✅
Connector package install	✅
Build destination-langchain docker image for platform linux/x86_64	✅
Unit tests	✅
Integration tests	✅

🔗 View the logs here

Please note that tests are only run on PR ready for review. Please set your PR to draft mode to not flood the CI engine and upstream service on following commits.
You can run the same pipeline locally on this branch with the airbyte-ci tool with the following command

airbyte-ci connectors --name=destination-langchain test

…-destination

octavia-squidington-iii · 2023-07-07T08:06:05Z

destination-langchain test report (commit `0a571f10d8`) - ✅

⏲️ Total pipeline duration: 01mn42s

Step	Result
Validate airbyte-integrations/connectors/destination-langchain/metadata.yaml	✅
Connector version semver check	✅
QA checks	✅
Code format checks	✅
Connector package install	✅
Build destination-langchain docker image for platform linux/x86_64	✅
Unit tests	✅
Integration tests	✅

🔗 View the logs here

Please note that tests are only run on PR ready for review. Please set your PR to draft mode to not flood the CI engine and upstream service on following commits.
You can run the same pipeline locally on this branch with the airbyte-ci tool with the following command

airbyte-ci connectors --name=destination-langchain test

maxi297

Just a couple of comments here and there while taking a break. I can recheck later if you want

airbyte-integrations/connectors/destination-langchain/destination_langchain/batcher.py

airbyte-integrations/connectors/destination-langchain/destination_langchain/destination.py

airbyte-integrations/connectors/destination-langchain/unit_tests/destination_test.py

airbyte-integrations/connectors/destination-langchain/destination_langchain/batcher.py

...te-integrations/connectors/destination-langchain/destination_langchain/document_processor.py

airbyte-integrations/connectors/destination-langchain/unit_tests/docarray_indexer_test.py

flash1293 · 2023-07-10T06:30:21Z

@maxi297 Thanks for your review, addressed your points. Could you take another look?

octavia-squidington-iii · 2023-07-10T06:31:28Z

destination-langchain test report (commit `92693970e0`) - ❌

⏲️ Total pipeline duration: 01mn37s

Step	Result
Validate airbyte-integrations/connectors/destination-langchain/metadata.yaml	✅
Connector version semver check	✅
QA checks	✅
Code format checks	❌
Connector package install	✅
Build destination-langchain docker image for platform linux/x86_64	✅
Unit tests	✅
Integration tests	✅

🔗 View the logs here

Please note that tests are only run on PR ready for review. Please set your PR to draft mode to not flood the CI engine and upstream service on following commits.
You can run the same pipeline locally on this branch with the airbyte-ci tool with the following command

airbyte-ci connectors --name=destination-langchain test

octavia-squidington-iii · 2023-07-10T06:33:42Z

destination-langchain test report (commit `b52a36420c`) - ✅

⏲️ Total pipeline duration: 58.09s

Step	Result
Validate airbyte-integrations/connectors/destination-langchain/metadata.yaml	✅
Connector version semver check	✅
QA checks	✅
Code format checks	✅
Connector package install	✅
Build destination-langchain docker image for platform linux/x86_64	✅
Unit tests	✅
Integration tests	✅

🔗 View the logs here

Please note that tests are only run on PR ready for review. Please set your PR to draft mode to not flood the CI engine and upstream service on following commits.
You can run the same pipeline locally on this branch with the airbyte-ci tool with the following command

airbyte-ci connectors --name=destination-langchain test

airbyte-integrations/connectors/destination-langchain/README.md

girarda · 2023-07-10T23:28:54Z

...te-integrations/connectors/destination-langchain/destination_langchain/document_processor.py

+        """
+        Generate documents from records.
+        :param records: List of AirbyteRecordMessages
+        :return: Tuple of (List of document chunks, Natural id to delete if applicable)


can you comment, either here or in destination.py, what the IDs to delete are about? It's not clear what natural ids are

It's an identifier of a record so we can remove documents belonging to a record before inserting new documents when in append-dedup mode. As one record might be split across multiple documents, it has to be made part of the metadata. It's leveraged here:

airbyte/airbyte-integrations/connectors/destination-langchain/destination_langchain/indexer.py

Line 75 in b3b556f

self.pinecone_index.delete(filter={METADATA_NATURAL_ID_FIELD: {"$in": delete_ids}})

I renamed it to record_id instead of natural id to make it easier to understand and added a comment.

girarda · 2023-07-10T23:42:23Z

airbyte-integrations/connectors/destination-langchain/destination_langchain/embedder.py

+        try:
+            self.embeddings.embed_query("test")
+        except Exception as e:
+            return str(e)


it's generally better to use repr than str because it returns the "printable representation of the object".

Example with a ValueError:

>>> try: ... raise ValueError("hello") ... except Exception as e: ... print(str(e)) ... hello >>> try: ... raise ValueError("hello") ... except Exception as e: ... print(repr(e)) ... ValueError('hello')

That being said, we should be able to do even better and return the stacktrace. Here's how we do it in sources

honestly I feel like in this case, str(e) plus a stacktrace is more readable? the ValueError piece makes it harder to read for an end user

Good call, added this

girarda · 2023-07-10T23:53:53Z

docs/integrations/destinations/langchain.md

+
+:::caution
+
+DocArrayHnswSearch is meant to be used on a local workstation and won't work on Kubernetes


doesn't have to be done now, but we'll need a way to disable DocArrayHnswSearch before we can add this destination to Cloud

Good callout, how does that work? Similarly to the *-strict-encrypt variants on connectors? Or can we adjust the spec somehow?

airbyte-integrations/connectors/destination-langchain/destination_langchain/indexer.py

airbyte-integrations/connectors/destination-langchain/destination_langchain/config.py

...te-integrations/connectors/destination-langchain/destination_langchain/document_processor.py

airbyte-integrations/connectors/destination-langchain/destination_langchain/indexer.py

clnoll

Awesome @flash1293! Left a couple of comments.

I ran a langchain destination locally per the instructions in the PR description, using DocArrayHnswSearch + OpenAI and got a green sync, but fyi there were a couple of unexpected error messages in the logs.

errors: $.mode: must be a constant value pinecone, $.mode: does not have a value in the enumeration [pinecone], $.pinecone_key: is missing but it is required, $.pinecone_environment: is missing but it is required, $.index: is missing but it is required
errors: $.openai_key: object found, string expected
errors: $.mode: must be a constant value fake, $.mode: does not have a value in the enumeration [fake]

clnoll · 2023-07-10T23:42:49Z

airbyte-integrations/connectors/destination-langchain/destination_langchain/indexer.py

+                raise Exception(
+                    f"DocArrayHnswSearchIndexer only supports overwrite mode, got {stream.destination_sync_mode} for stream {stream.stream.name}"
+                )
+        for file in os.listdir(self.config.destination_path):


Looks like there's some asymmetry between this and the PineconeIndexer, which only deletes an index for the specific stream that's being re-synced. Is it worth putting a warning in the connector UI that the files in this directory will be deleted before every sync?

clnoll · 2023-07-11T00:07:08Z

airbyte-integrations/connectors/destination-langchain/destination_langchain/embedder.py

+
+    @property
+    def embedding_dimensions(self) -> int:
+        # vector size produced by text-embedding-ada-002 model


Should the vector size and/or model be configurable? Is text-embedding-ada-002 the default? I'm not seeing it mentioned anywhere else.

It is mentioned in the documentation and as a description as part of the spec:

airbyte/airbyte-integrations/connectors/destination-langchain/destination_langchain/config.py

Line 44 in 0a571f1

"description": "Use the OpenAI API to embed text. This option is using the text-embedding-ada-002 model with 1536 embedding dimensions."

airbyte/docs/integrations/destinations/langchain.md

Line 10 in 0a571f1

* Embedding - convert the text into a vector representation using a pre-trained model (currently only OpenAI `text-embedding-ada-002` is supported)

I think it makes sense to start like this (OpenAI recently deprecated all other embedding models as this one is preferable in all situations), we probably want to add other embeddings at a later stage though that might have different numbers of dimensions (the number of dimensions is an intrinsic part of the embedding model).

flash1293 · 2023-07-11T06:28:09Z

@clnoll

I ran a langchain destination locally per the instructions in the PR description, using DocArrayHnswSearch + OpenAI and got a green sync, but fyi there were a couple of unexpected error messages in the logs.

I've noticed that too, but for me those show up on other syncs as well (e.g. this one is from slack to local json), so I didn't think it's a problem:

2023-07-06 13:40:21 INFO i.a.v.j.JsonSchemaValidator(test):121 - JSON schema validation failed. 
errors: $.client_id: is missing but it is required, $.client_secret: is missing but it is required, $.access_token: is missing but it is required, $.option_title: must be a constant value Default OAuth2.0 authorization
2023-07-06 13:40:21 INFO i.a.v.j.JsonSchemaValidator(test):121 - JSON schema validation failed.

octavia-squidington-iii · 2023-07-11T09:24:22Z

destination-langchain test report (commit `74dc691848`) - ✅

⏲️ Total pipeline duration: 03mn06s

Step	Result
Validate airbyte-integrations/connectors/destination-langchain/metadata.yaml	✅
Connector version semver check	✅
QA checks	✅
Code format checks	✅
Connector package install	✅
Build destination-langchain docker image for platform linux/x86_64	✅
Unit tests	✅
Integration tests	✅

🔗 View the logs here

Please note that tests are only run on PR ready for review. Please set your PR to draft mode to not flood the CI engine and upstream service on following commits.
You can run the same pipeline locally on this branch with the airbyte-ci tool with the following command

airbyte-ci connectors --name=destination-langchain test

girarda

code looks good to me 🎸

be sure to go through and update the PR checklist before merging https://github.com/airbytehq/airbyte/actions/runs/5518023163/jobs/10061428323?pr=26184

girarda · 2023-07-11T22:20:40Z

airbyte-integrations/connectors/destination-langchain/README.md

+* required for the testing need to go to `TEST_REQUIREMENTS` list
+
+### Publishing a new version of the connector
+You've checked out the repo, implemented a million dollar feature, and you're ready to share your changes with the world. Now what?


girarda · 2023-07-12T00:49:30Z

...yte-integrations/connectors/destination-langchain/integration_tests/base_integration_test.py

@@ -0,0 +1,42 @@
+#


do we want to add acceptance tests for this connector?

octavia-squidington-iii · 2023-07-12T15:46:37Z

destination-langchain test report (commit `7b9da895b2`) - ✅

⏲️ Total pipeline duration: 40.78s

Step	Result
Validate airbyte-integrations/connectors/destination-langchain/metadata.yaml	✅
Connector version semver check	✅
QA checks	✅
Code format checks	✅
Connector package install	✅
Build destination-langchain docker image for platform linux/x86_64	✅
Unit tests	✅
Integration tests	✅

🔗 View the logs here

Please note that tests are only run on PR ready for review. Please set your PR to draft mode to not flood the CI engine and upstream service on following commits.
You can run the same pipeline locally on this branch with the airbyte-ci tool with the following command

airbyte-ci connectors --name=destination-langchain test

slavakurilyak · 2023-07-15T19:09:41Z

Excellent work on this integration!

It would be great to see this destination appear in the Connectors list of Airbyte. Consider categorizing this destination under 'LangChain' and 'Pinecone' since Pinecone is unlocked thanks to LangChain.

flash1293 · 2023-07-17T08:35:11Z

Thanks for the feedback @slavakurilyak , we have a bunch of plans in this direction, stay tuned!

…6184) * basic version * polish * iterate * keep working * fix spec * wip * improve destination * basic unit tests * move embedding dimensionality into embedder * improve several things * adjust documentation * remove unnecessary call * add some debug information * fix local destination * various small fixes * bring tests into order * document and add batching to pinecone * checklist * improve performance a bit and add test * fix formatting * fix metadata * install C++ 11 on python base * no more alpine for ci-connector-ops * remove hard-to-run test * more documentation * better documentation * add icon * some small adjustments * review comments * format * review comments --------- Co-authored-by: alafanechere <augustin.lafanechere@gmail.com> Co-authored-by: Augustin <augustin@airbyte.io>

basic version

e42d138

octavia-squidington-iii added area/connectors Connector related issues connectors/destination/langchain labels May 17, 2023

Joe Reuter added 4 commits May 18, 2023 05:40

polish

1b015d1

iterate

bd84085

Merge remote-tracking branch 'origin/master' into flash1293/langchain…

d04f75e

…-destination

keep working

f7d39ca

octavia-squidington-iii added the area/documentation Improvements or additions to documentation label Jun 29, 2023

Joe Reuter added 8 commits June 29, 2023 13:27

fix spec

7fafaa8

wip

bea9cd0

improve destination

8684ee8

basic unit tests

3a5650d

move embedding dimensionality into embedder

a94532d

improve several things

1670eb4

adjust documentation

abc2bd7

remove unnecessary call

409d0f5

sherifnada reviewed Jul 4, 2023

View reviewed changes

Joe Reuter added 3 commits July 4, 2023 11:42

add some debug information

21ca42c

fix local destination

b8e2275

various small fixes

91c9df7

flash1293 mentioned this pull request Jul 4, 2023

Langchain destination multithreading #27945

Closed

Joe Reuter added 3 commits July 5, 2023 08:52

bring tests into order

3ecb04d

document and add batching to pinecone

29b3b0a

checklist

bddd100

flash1293 changed the title ~~Langchain destination~~ 🎉 New Destination Vector Database (powered by LangChain) Jul 5, 2023

flash1293 marked this pull request as ready for review July 5, 2023 12:23

improve performance a bit and add test

6126f4f

add icon

099ad20

Joe Reuter added 2 commits July 7, 2023 08:48

Merge remote-tracking branch 'origin/master' into flash1293/langchain…

b3b556f

…-destination

some small adjustments

0a571f1

maxi297 reviewed Jul 7, 2023

View reviewed changes

Joe Reuter added 2 commits July 10, 2023 08:28

review comments

9269397

format

b52a364

girarda reviewed Jul 11, 2023

View reviewed changes

clnoll reviewed Jul 11, 2023

View reviewed changes

review comments

74dc691

flash1293 requested review from clnoll and girarda July 11, 2023 09:09

girarda reviewed Jul 12, 2023

View reviewed changes

girarda approved these changes Jul 12, 2023

View reviewed changes

Merge branch 'master' into flash1293/langchain-destination

7b9da89

flash1293 merged commit b52e88a into master Jul 12, 2023
20 checks passed

flash1293 deleted the flash1293/langchain-destination branch July 12, 2023 18:07

alafanechere mentioned this pull request Jul 17, 2023

connectors-ci: revert system deps changes introduced for langchain #28375

Closed

		schema_extra = {"group":"processing"}


		class EmbeddingConfigModel(BaseModel):


		:::caution

		DocArrayHnswSearch is meant to be used on a local workstation and won't work on Kubernetes

🎉 New Destination Vector Database (powered by LangChain) #26184

🎉 New Destination Vector Database (powered by LangChain) #26184

Conversation

flash1293 commented May 17, 2023 • edited

How to test

What

How

Unclear parts / TODOs

New Connector

Todos

github-actions bot commented May 17, 2023 • edited by flash1293

Before Merging a Connector Pull Request

sherifnada left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

flash1293 Jul 4, 2023 • edited

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

octavia-squidington-iii commented Jul 5, 2023

destination-langchain test report (commit bddd100c3d) - ❌

octavia-squidington-iii commented Jul 5, 2023

destination-langchain test report (commit 6126f4f302) - ❌

octavia-squidington-iii commented Jul 6, 2023

destination-langchain test report (commit 099ad208a9) - ✅

octavia-squidington-iii commented Jul 6, 2023

destination-langchain test report (commit 099ad208a9) - ✅

octavia-squidington-iii commented Jul 7, 2023

destination-langchain test report (commit 0a571f10d8) - ✅

maxi297 left a comment

Choose a reason for hiding this comment

flash1293 commented Jul 10, 2023

octavia-squidington-iii commented Jul 10, 2023

destination-langchain test report (commit 92693970e0) - ❌

octavia-squidington-iii commented Jul 10, 2023

destination-langchain test report (commit b52a36420c) - ✅

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

clnoll left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

flash1293 commented Jul 11, 2023

octavia-squidington-iii commented Jul 11, 2023

destination-langchain test report (commit 74dc691848) - ✅

girarda left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

octavia-squidington-iii commented Jul 12, 2023

destination-langchain test report (commit 7b9da895b2) - ✅

slavakurilyak commented Jul 15, 2023

flash1293 commented Jul 17, 2023

flash1293 commented May 17, 2023 •

edited

github-actions bot commented May 17, 2023 •

edited by flash1293

flash1293 Jul 4, 2023 •

edited

destination-langchain test report (commit `bddd100c3d`) - ❌

destination-langchain test report (commit `6126f4f302`) - ❌

destination-langchain test report (commit `099ad208a9`) - ✅

destination-langchain test report (commit `099ad208a9`) - ✅

destination-langchain test report (commit `0a571f10d8`) - ✅

destination-langchain test report (commit `92693970e0`) - ❌

destination-langchain test report (commit `b52a36420c`) - ✅

destination-langchain test report (commit `74dc691848`) - ✅

destination-langchain test report (commit `7b9da895b2`) - ✅