Removed all deprecated methods and attributes

Author: jamesbraza

src/paperqa/__init__.py

@@ -1,5 +1,3 @@
-import warnings
-
 from lmi import (
     EmbeddingModel,
     HybridEmbeddingModel,
@@ -21,17 +19,10 @@
     VectorStore,
 )
 from paperqa.settings import Settings, get_settings
-from paperqa.types import Answer, Context, Doc, DocDetails, Text
+from paperqa.types import Context, Doc, DocDetails, Text
 from paperqa.version import __version__
 
-# TODO: remove after refactoring all models to avoid using _* private vars
-warnings.filterwarnings(
-    "ignore", message="Valid config keys have changed in V2:", module="pydantic"
-)
-
-
 __all__ = [
-    "Answer",
     "Context",
     "Doc",
     "DocDetails",

src/paperqa/agents/search.py

@@ -9,7 +9,6 @@
 import pickle
 import re
 import sys
-import warnings
 import zlib
 from collections import Counter
 from collections.abc import AsyncIterator, Callable, Sequence
@@ -613,38 +612,21 @@ def progress_bar_update() -> None:
     return contextlib.nullcontext(), None
 
 
-async def get_directory_index(  # noqa: PLR0912
-    index_name: str | None = None,
-    sync_index_w_directory: bool = True,
-    settings: MaybeSettings = None,
-    build: bool = True,
+async def get_directory_index(
+    settings: MaybeSettings = None, build: bool = True
 ) -> SearchIndex:
     """
     Create a Tantivy index by reading from a directory of text files.
 
     This function only reads from the source directory, not edits or writes to it.
 
     Args:
-        index_name: Deprecated override on the name of the index. If unspecified,
-            the default behavior is to generate the name from the input settings.
-        sync_index_w_directory: Opt-out flag to sync the index (add or delete index
-            files) with the source paper directory.
         settings: Application settings.
         build: Opt-out flag (default is True) to read the contents of the source paper
             directory and if sync_index_w_directory is enabled also update the index.
     """
     _settings = get_settings(settings)
     index_settings = _settings.agent.index
-    if index_name:
-        warnings.warn(
-            "The index_name argument has been moved to"
-            f" {type(_settings.agent.index).__name__},"
-            " this deprecation will conclude in version 6.",
-            category=DeprecationWarning,
-            stacklevel=2,
-        )
-        index_settings.name = index_name
-    del index_name
 
     search_index = SearchIndex(
         fields=[*SearchIndex.REQUIRED_FIELDS, "title", "year"],
@@ -660,17 +642,6 @@ async def get_directory_index(  # noqa: PLR0912
             )
         return search_index
 
-    if not sync_index_w_directory:
-        warnings.warn(
-            "The sync_index_w_directory argument has been moved to"
-            f" {type(_settings.agent.index).__name__},"
-            " this deprecation will conclude in version 6.",
-            category=DeprecationWarning,
-            stacklevel=2,
-        )
-        index_settings.sync_with_paper_directory = sync_index_w_directory
-    del sync_index_w_directory
-
     paper_directory = anyio.Path(index_settings.paper_directory)
     manifest = await maybe_get_manifest(
         filename=await index_settings.finalize_manifest_file()

src/paperqa/agents/tools.py

@@ -47,10 +47,10 @@ def default_status(state: "EnvironmentState") -> str:
 class EnvironmentState(BaseModel):
     """State here contains documents and answer being populated."""
 
-    model_config = ConfigDict(extra="forbid", populate_by_name=True)
+    model_config = ConfigDict(extra="forbid")
 
     docs: Docs
-    session: PQASession = Field(..., alias="answer")
+    session: PQASession
     status_fn: Callable[[Self], str] | None = Field(
         default=None,
         description=(

src/paperqa/contrib/openreview_paper_helper.py

@@ -46,7 +46,7 @@ def __init__(
         password: str | None = None,
     ) -> None:
         self.settings = settings
-        Path(settings.paper_directory).mkdir(parents=True, exist_ok=True)
+        Path(settings.agent.index.paper_directory).mkdir(parents=True, exist_ok=True)
         if openreview is None:
             raise ImportError(
                 "openreview requires the 'openreview-py' extra. Please run: `pip"
@@ -122,7 +122,9 @@ async def _get_relevant_papers_chunk(self, question: str, chunk: str) -> list[An
 
     async def download_papers(self, submissions: list[Any]) -> None:
         """Download PDFs for given submissions."""
-        downloaded_papers = Path(self.settings.paper_directory).rglob("*.pdf")
+        downloaded_papers = Path(self.settings.agent.index.paper_directory).rglob(
+            "*.pdf"
+        )
         downloaded_ids = [p.stem for p in downloaded_papers]
         logger.info("Downloading PDFs for relevant papers.")
         for submission in submissions:
@@ -136,7 +138,7 @@ async def _download_pdf(self, submission: Any) -> bool:
             response = await client.get(pdf_link)
         if response.status_code == httpx.codes.OK.value:
             async with await anyio.open_file(
-                f"{self.settings.paper_directory}/{submission.id}.pdf", "wb"
+                f"{self.settings.agent.index.paper_directory}/{submission.id}.pdf", "wb"
             ) as f:
                 await f.write(response.content)
             return True
@@ -151,7 +153,7 @@ async def aadd_docs(
     ) -> Docs:
         if docs is None:
             docs = Docs()
-        for doc_path in Path(self.settings.paper_directory).rglob("*.pdf"):
+        for doc_path in Path(self.settings.agent.index.paper_directory).rglob("*.pdf"):
             sub = subs.get(doc_path.stem) if subs is not None else None
             if sub:
                 await docs.aadd(

src/paperqa/docs.py

@@ -6,7 +6,6 @@
 import re
 import tempfile
 import urllib.request
-import warnings
 from collections.abc import Callable, Sequence
 from datetime import datetime
 from io import BytesIO
@@ -32,7 +31,6 @@
 from paperqa.types import Doc, DocDetails, DocKey, PQASession, Text
 from paperqa.utils import (
     citation_to_docname,
-    get_loop,
     maybe_is_html,
     maybe_is_pdf,
     maybe_is_text,
@@ -90,35 +88,6 @@ def _get_unique_name(self, docname: str) -> str:
         docname += suffix
         return docname
 
-    def add_file(
-        self,
-        file: BinaryIO,
-        citation: str | None = None,
-        docname: str | None = None,
-        dockey: DocKey | None = None,
-        settings: MaybeSettings = None,
-        llm_model: LLMModel | None = None,
-        embedding_model: EmbeddingModel | None = None,
-    ) -> str | None:
-        warnings.warn(
-            "The synchronous `add_file` method is being deprecated in favor of the"
-            " asynchronous `aadd_file` method, this deprecation will conclude in"
-            " version 6.",
-            category=DeprecationWarning,
-            stacklevel=2,
-        )
-        return get_loop().run_until_complete(
-            self.aadd_file(
-                file,
-                citation=citation,
-                docname=docname,
-                dockey=dockey,
-                settings=settings,
-                llm_model=llm_model,
-                embedding_model=embedding_model,
-            )
-        )
-
     async def aadd_file(
         self,
         file: BinaryIO,
@@ -158,35 +127,6 @@ async def aadd_file(
                 **kwargs,
             )
 
-    def add_url(
-        self,
-        url: str,
-        citation: str | None = None,
-        docname: str | None = None,
-        dockey: DocKey | None = None,
-        settings: MaybeSettings = None,
-        llm_model: LLMModel | None = None,
-        embedding_model: EmbeddingModel | None = None,
-    ) -> str | None:
-        warnings.warn(
-            "The synchronous `add_url` method is being deprecated in favor of the"
-            " asynchronous `aadd_url` method, this deprecation will conclude in"
-            " version 6.",
-            category=DeprecationWarning,
-            stacklevel=2,
-        )
-        return get_loop().run_until_complete(
-            self.aadd_url(
-                url,
-                citation=citation,
-                docname=docname,
-                dockey=dockey,
-                settings=settings,
-                llm_model=llm_model,
-                embedding_model=embedding_model,
-            )
-        )
-
     async def aadd_url(
         self,
         url: str,
@@ -211,43 +151,6 @@ async def aadd_url(
                 embedding_model=embedding_model,
             )
 
-    def add(
-        self,
-        path: str | os.PathLike,
-        citation: str | None = None,
-        docname: str | None = None,
-        dockey: DocKey | None = None,
-        title: str | None = None,
-        doi: str | None = None,
-        authors: list[str] | None = None,
-        settings: MaybeSettings = None,
-        llm_model: LLMModel | None = None,
-        embedding_model: EmbeddingModel | None = None,
-        **kwargs,
-    ) -> str | None:
-        warnings.warn(
-            "The synchronous `add` method is being deprecated in favor of the"
-            " asynchronous `aadd` method, this deprecation will conclude in"
-            " version 6.",
-            category=DeprecationWarning,
-            stacklevel=2,
-        )
-        return get_loop().run_until_complete(
-            self.aadd(
-                path,
-                citation=citation,
-                docname=docname,
-                dockey=dockey,
-                title=title,
-                doi=doi,
-                authors=authors,
-                settings=settings,
-                llm_model=llm_model,
-                embedding_model=embedding_model,
-                **kwargs,
-            )
-        )
-
     async def aadd(  # noqa: PLR0912
         self,
         path: str | os.PathLike,
@@ -411,26 +314,6 @@ async def aadd(  # noqa: PLR0912
             return doc.docname
         return None
 
-    def add_texts(
-        self,
-        texts: list[Text],
-        doc: Doc,
-        settings: MaybeSettings = None,
-        embedding_model: EmbeddingModel | None = None,
-    ) -> bool:
-        warnings.warn(
-            "The synchronous `add_texts` method is being deprecated in favor of the"
-            " asynchronous `aadd_texts` method, this deprecation will conclude in"
-            " version 6.",
-            category=DeprecationWarning,
-            stacklevel=2,
-        )
-        return get_loop().run_until_complete(
-            self.aadd_texts(
-                texts, doc, settings=settings, embedding_model=embedding_model
-            )
-        )
-
     async def aadd_texts(
         self,
         texts: list[Text],
@@ -552,39 +435,9 @@ async def retrieve_texts(
         matches = [m for m in matches if m.doc.dockey not in self.deleted_dockeys]
         return matches[:k]
 
-    def get_evidence(
-        self,
-        query: PQASession | str,
-        exclude_text_filter: set[str] | None = None,
-        settings: MaybeSettings = None,
-        callbacks: Sequence[Callable] | None = None,
-        embedding_model: EmbeddingModel | None = None,
-        summary_llm_model: LLMModel | None = None,
-        partitioning_fn: Callable[[Embeddable], int] | None = None,
-    ) -> PQASession:
-        warnings.warn(
-            "The synchronous `get_evidence` method is being deprecated in favor of the"
-            " asynchronous `aget_evidence` method, this deprecation will conclude in"
-            " version 6.",
-            category=DeprecationWarning,
-            stacklevel=2,
-        )
-        return get_loop().run_until_complete(
-            self.aget_evidence(
-                query=query,
-                exclude_text_filter=exclude_text_filter,
-                settings=settings,
-                callbacks=callbacks,
-                embedding_model=embedding_model,
-                summary_llm_model=summary_llm_model,
-                partitioning_fn=partitioning_fn,
-            )
-        )
-
     async def aget_evidence(
         self,
         query: PQASession | str,
-        exclude_text_filter: set[str] | None = None,
         settings: MaybeSettings = None,
         callbacks: Sequence[Callable] | None = None,
         embedding_model: EmbeddingModel | None = None,
@@ -611,21 +464,6 @@ async def aget_evidence(
         if summary_llm_model is None:
             summary_llm_model = evidence_settings.get_summary_llm()
 
-        if exclude_text_filter is not None:
-            text_name = Text.__name__
-            warnings.warn(
-                (
-                    "The 'exclude_text_filter' argument did not work as intended"
-                    f" due to a mix-up in excluding {text_name}.name vs {text_name}."
-                    f" This bug enabled us to have 2+ contexts per {text_name}, so to"
-                    " first-class that capability and simplify our implementation,"
-                    " we're removing the 'exclude_text_filter' argument."
-                    " This deprecation will conclude in version 6"
-                ),
-                category=DeprecationWarning,
-                stacklevel=2,
-            )
-
         if answer_config.evidence_retrieval:
             matches = await self.retrieve_texts(
                 session.question,
@@ -684,35 +522,6 @@ async def aget_evidence(
         session.contexts += [r for r, _ in results]
         return session
 
-    def query(
-        self,
-        query: PQASession | str,
-        settings: MaybeSettings = None,
-        callbacks: Sequence[Callable] | None = None,
-        llm_model: LLMModel | None = None,
-        summary_llm_model: LLMModel | None = None,
-        embedding_model: EmbeddingModel | None = None,
-        partitioning_fn: Callable[[Embeddable], int] | None = None,
-    ) -> PQASession:
-        warnings.warn(
-            "The synchronous `query` method is being deprecated in favor of the"
-            " asynchronous `aquery` method, this deprecation will conclude in"
-            " version 6.",
-            category=DeprecationWarning,
-            stacklevel=2,
-        )
-        return get_loop().run_until_complete(
-            self.aquery(
-                query,
-                settings=settings,
-                callbacks=callbacks,
-                llm_model=llm_model,
-                summary_llm_model=summary_llm_model,
-                embedding_model=embedding_model,
-                partitioning_fn=partitioning_fn,
-            )
-        )
-
     async def aquery(
         self,
         query: PQASession | str,