Search Google Drive documents and retrieve contents #265
Conversation
…part of the document content, not its name)
… 'and' operators together
…rsions of arcade-ai
Codecov ReportAll modified and coverable lines are covered by tests ✅ 📢 Thoughts on this report? Let us know! |
|
@byrro I'm unable to run this locally. Looks like an issue with all of the double and single quotes in the
|
…kdown, HTML, and JSON format options
…terface abstracting away args that are specific to the Google API
|
@EricGustin pushed a new implementation of the tool, refactored |
byrro
added
the
toolkit: major
| document_contains: Optional[list[str]] = None, | ||
| document_not_contains: Optional[list[str]] = None, | ||
| ) -> str: | ||
| query = ["mimeType = 'application/vnd.google-apps.document' and trashed = false"] |
There was a problem hiding this comment.
Looking ahead, we will need to search for more file types beyond document. For example, searching for a spreadsheet by name. Perhaps mime type can be a parameter so that we don't have to worry about that debt in the future
There was a problem hiding this comment.
makes sense, just pushed a new version with mime_type as an argument to build files_list query / params.
byrro
changed the title
byrro
changed the title
| name_contains = keyword.replace("'", "\\'") | ||
| full_text_contains = keyword.replace("'", "\\'") | ||
| keyword_query = ( | ||
| f"name contains '{name_contains}' or fullText contains '{full_text_contains}'" |
There was a problem hiding this comment.
we need to group the left and right side of the or inside parentheses, otherwise Google interprets the query in a way that we don't intend. See slack dm for more details
There was a problem hiding this comment.
good catch! just pushed a fix
This tool will be useful in scenarios akin to RAG, where someone wants to ask questions or request the production of a summary, for instance, about a bunch of documents related to a particular topic. Currently, to fulfill such requests, the LLM needs to first
list_documents, thenget_document_by_idfor each document.We also implement a utility functions to return documents in Markdown and HTML, since the Drive API JSON is verbose and would waste too many tokens unnecessarily.
Limitations: the Markdown/HTML utilities do not handle table of contents (which I think aren't really useful here), headers, footers, or footnotes.
This PR deprecates
list_documentsand implementssearch_documents, apart fromsearch_and_retrieve_documents). This configuration makes it easier for LLMs to understand when to call each tool.Both tools had their interfaces refactored to remove Google API-specific arguments that were confusing LLMs sometimes, such as "corpora" and "support_all_drives". It now accepts arguments that better relate to expected user requests.