Zotero

os.environ["LANGSMITH_API_KEY"] = getpass.getpass("Enter your LangSmith API key: ")
os.environ["LANGSMITH_TRACING"] = "true"

pip install -qU langchain-zotero-retriever pyzotero

from langchain_zotero_retriever.retrievers import ZoteroRetriever

retriever = ZoteroRetriever(
    k=10,
    library_id="2319375",  # a public group library that does not require an API key for access
    library_type="group",  # set this to "user" if you are using a personal library. Personal libraries require an API key
)

query = "Zuboff"

retriever.invoke(query)

tags = [
    "Surveillance",
    "Digital Capitalism",
]  # note that providing tags as a list will result in a logical AND operation

retriever.invoke("", tag=tags)

from typing import List, Optional, Union

from langchain_core.output_parsers import PydanticToolsParser
from langchain.tools import StructuredTool, tool
from langchain_openai import ChatOpenAI


def retrieve(
    query: str,
    itemType: Optional[str],
    tag: Optional[Union[str, List[str]]],
    qmode: str = "everything",
    since: Optional[int] = None,
):
    retrieved_docs = retriever.invoke(
        query, itemType=itemType, tag=tag, qmode=qmode, since=since
    )
    serialized_docs = "\n\n".join(
        (
            f"Metadata: { {key: doc.metadata[key] for key in doc.metadata if key != 'abstractNote'} }\n"
            f"Abstract: {doc.metadata['abstractNote']}\n"
        )
        for doc in retrieved_docs
    )

    return serialized_docs, retrieved_docs


description = """Search and return relevant documents from a Zotero library. The following search parameters can be used:

    Args:
        query: str: The search query to be used. Try to keep this specific and short, e.g. a specific topic or author name
        itemType: Optional. Type of item to search for (e.g. "book" or "journalArticle"). Multiple types can be passed as a string separated by "||", e.g. "book || journalArticle". Defaults to all types.
        tag: Optional. For searching over tags attached to library items. If documents tagged with multiple tags are to be retrieved, pass them as a list. If documents with any of the tags are to be retrieved, pass them as a string separated by "||", e.g. "tag1 || tag2"
        qmode: Search mode to use. Changes what the query searches over. "everything" includes full-text content. "titleCreatorYear" to search over title, authors and year. Defaults to "everything".
        since: Return only objects modified after the specified library version. Defaults to return everything.
    """

retriever_tool = StructuredTool.from_function(
    func=retrieve,
    name="retrieve",
    description=description,
    return_direct=True,
)


llm = ChatOpenAI(model="gpt-4o-mini-2024-07-18")

llm_with_tools = llm.bind_tools([retrieve])

q = "What journal articles do I have on Surveillance in the zotero library?"

chain = llm_with_tools | PydanticToolsParser(tools=[retrieve])

chain.invoke(q)