feat: astra db chunks deletion based on metadata field #5537
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
… document management - Introduced a new 'deletion_field' input to specify a metadata field for deleting documents before loading new data. - Enhanced the _add_documents_to_vector_store method to handle document deletion based on the specified field, improving data management capabilities.
dosubot
bot
added
size:S
CodSpeed Performance ReportMerging #5537 will degrade performances by 25.38%Comparing Summary
Benchmarks breakdown
|
…ove readability. - Optimized the deletion logic by using a set comprehension to eliminate duplicates when gathering delete values from documents.
smatiolids
changed the title
smatiolids
changed the title
github-actions
bot
added
enhancement
smatiolids
changed the title
github-actions
bot
added
enhancement
msmygit
reviewed
Jan 6, 2025
| @@ -607,6 +616,18 @@ def _add_documents_to_vector_store(self, vector_store) -> None: | |||
| msg = "Vector Store Inputs must be Data objects." | |||
| raise TypeError(msg) | |||
|
|
|||
| if documents and self.deletion_field: | |||
| self.log(f"Deleting documents where {self.deletion_field}") | |||
There was a problem hiding this comment.
nit: should we remove this log line?
There was a problem hiding this comment.
@msmygit I can see the argument to delete it for sure - i'd say we can keep it for now... I don't think it exposes any unnecessary information, and might help debug some issues.
Co-authored-by: Madhavan <msmygit@users.noreply.github.com>
github-actions
bot
added
enhancement
github-actions
bot
added
enhancement
erichare
approved these changes
Jan 6, 2025
There was a problem hiding this comment.
Agreed with @msmygit 's suggestions, just left the log message in for now. I commited the change to the error log message suggested. Everything else looks great - i think this is a good addition, though it will have to eventually be nicely documented to be used to its full potential. But i'm going to approve! Thanks!
|
I should add - the team is actively investigating a better ingestion experience for situations like this - but i think this is a solid solution for the current time. |
ogabrielluiz
pushed a commit
that referenced
this pull request
Jan 8, 2025
* feat: Add deletion_field parameter to AstraDBVectorStoreComponent for document management - Introduced a new 'deletion_field' input to specify a metadata field for deleting documents before loading new data. - Enhanced the _add_documents_to_vector_store method to handle document deletion based on the specified field, improving data management capabilities. * Merging with main * [autofix.ci] apply automated fixes * [autofix.ci] apply automated fixes * - Enhanced the info string for the 'deletion_field' parameter to improve readability. - Optimized the deletion logic by using a set comprehension to eliminate duplicates when gathering delete values from documents. * [autofix.ci] apply automated fixes * Update src/backend/base/langflow/components/vectorstores/astradb.py Co-authored-by: Madhavan <msmygit@users.noreply.github.com> * [autofix.ci] apply automated fixes --------- Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com> Co-authored-by: Eric Hare <ericrhare@gmail.com> Co-authored-by: Madhavan <msmygit@users.noreply.github.com>
ogabrielluiz
pushed a commit
to raphaelchristi/langflow
that referenced
this pull request
Jan 10, 2025
) * feat: Add deletion_field parameter to AstraDBVectorStoreComponent for document management - Introduced a new 'deletion_field' input to specify a metadata field for deleting documents before loading new data. - Enhanced the _add_documents_to_vector_store method to handle document deletion based on the specified field, improving data management capabilities. * Merging with main * [autofix.ci] apply automated fixes * [autofix.ci] apply automated fixes * - Enhanced the info string for the 'deletion_field' parameter to improve readability. - Optimized the deletion logic by using a set comprehension to eliminate duplicates when gathering delete values from documents. * [autofix.ci] apply automated fixes * Update src/backend/base/langflow/components/vectorstores/astradb.py Co-authored-by: Madhavan <msmygit@users.noreply.github.com> * [autofix.ci] apply automated fixes --------- Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com> Co-authored-by: Eric Hare <ericrhare@gmail.com> Co-authored-by: Madhavan <msmygit@users.noreply.github.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Purpose
This PR addresses the need to reload specific documents without affecting others. To achieve this, a new option, "deletion_field", has been introduced.
Functionality
When "deletion_field" is set (e.g., "file_path"), the system will delete all documents in the target collection where metadata["file_path"] matches the corresponding value in the incoming documents.
This ensures that chunks from the specific file are removed before reloading it, preventing duplicates or conflicts.