academia-mcp 1.11.2__py3-none-any.whl → 1.11.4__py3-none-any.whl

academia_mcp/files.py

@@ -1,30 +1,15 @@
-from typing import Optional
 from pathlib import Path
 
 from academia_mcp.settings import settings
 
 DIR_PATH = Path(__file__).parent
 ROOT_PATH = DIR_PATH.parent
-DEFAULT_WORKSPACE_DIR_PATH: Path = DIR_PATH / "workdir"
 DEFAULT_LATEX_TEMPLATES_DIR_PATH: Path = DIR_PATH / "latex_templates"
 
 
-class WorkspaceDirectory:
-    workspace_dir: Optional[Path] = None
-
-    @classmethod
-    def get_dir(cls) -> Path:
-        if cls.workspace_dir is None:
-            return Path(settings.WORKSPACE_DIR)
-        return cls.workspace_dir
-
-    @classmethod
-    def set_dir(cls, workspace_dir: Path) -> None:
-        cls.workspace_dir = workspace_dir
-
-
 def get_workspace_dir() -> Path:
-    directory = WorkspaceDirectory.get_dir()
+    assert settings.WORKSPACE_DIR is not None, "Please set the WORKSPACE_DIR environment variable"
+    directory = Path(settings.WORKSPACE_DIR)
     if not directory.exists():
         directory.mkdir(parents=True, exist_ok=True)
     return directory

academia_mcp/server.py

@@ -40,7 +40,7 @@ from academia_mcp.tools.bitflip import (
     score_research_proposals,
 )
 from academia_mcp.tools.review import review_pdf_paper, download_pdf_paper
-from academia_mcp.tools.show_image import show_image, describe_image
+from academia_mcp.tools.image_processing import show_image, describe_image
 from academia_mcp.tools.speech_to_text import speech_to_text
 from academia_mcp.tools.yt_transcript import yt_transcript
 

academia_mcp/tools/__init__.py

@@ -14,7 +14,7 @@ from .web_search import web_search, tavily_web_search, exa_web_search, brave_web
 from .visit_webpage import visit_webpage
 from .bitflip import extract_bitflip_info, generate_research_proposals, score_research_proposals
 from .review import review_pdf_paper, download_pdf_paper, review_pdf_paper_by_url
-from .show_image import show_image, describe_image
+from .image_processing import show_image, describe_image
 from .speech_to_text import speech_to_text
 from .yt_transcript import yt_transcript
 

academia_mcp/tools/{show_image.py → image_processing.py}

@@ -1,16 +1,28 @@
+import asyncio
 import base64
-from pathlib import Path
+import contextlib
+import json
+import logging
+import os
+import threading
 from io import BytesIO
-from typing import Dict, Optional
+from pathlib import Path
 from textwrap import dedent
+from typing import Dict, List, Optional, Any
 
 import httpx
+from paddleocr import PaddleOCR  # type: ignore
 from PIL import Image
+from pydantic import BaseModel
 
 from academia_mcp.files import get_workspace_dir
+from academia_mcp.llm import ChatMessage, llm_acall
 from academia_mcp.settings import settings
-from academia_mcp.llm import llm_acall, ChatMessage
 
+paddlex_logger = logging.getLogger("paddlex")
+paddleocr_logger = logging.getLogger("paddleocr")
+paddlex_logger.setLevel(logging.ERROR)
+paddleocr_logger.setLevel(logging.ERROR)
 
 DESCRIBE_PROMPTS = {
     "general": "Provide a general description of this image. Focus on the main subjects, colors, and overall scene.",
@@ -37,16 +49,64 @@ DESCRIBE_PROMPTS = {
         - If layout is multi-column or tabular, reconstruct lines top-to-bottom, left-to-right; use line breaks between blocks.
         - For any uncertain or low-confidence characters, mark with a '?' and include a note.
         - After the raw extraction, provide a clean, normalized version (fixing obvious OCR artifacts) as a separate section.
-        Return two sections:
+        Return three sections:
+        [GENERAL IMAGE DESCRIPTION]
+        ...
         [RAW TRANSCRIPTION]
         ...
-        [NORMALIZED]
+        [NORMALIZED TRANSCRIPTION]
         ...
         """
     ),
 }
 
 
+class OCRBox(BaseModel):  # type: ignore
+    poly: List[List[float]]
+    text: str
+    score: float
+
+
+class OCRSingleton:
+    instance: Optional[PaddleOCR] = None
+    lock: threading.Lock = threading.Lock()
+
+    @classmethod
+    def get(cls) -> PaddleOCR:
+        if cls.instance is not None:
+            return cls.instance
+        with cls.lock:
+            if cls.instance is None:
+                with open(os.devnull, "w") as devnull:
+                    with contextlib.redirect_stderr(devnull):
+                        cls.instance = PaddleOCR(
+                            use_doc_orientation_classify=False,
+                            use_doc_unwarping=False,
+                            use_textline_orientation=False,
+                        )
+        return cls.instance
+
+
+async def _run_ocr(path: str) -> Dict[str, Any]:
+    def _sync_ocr(path: str) -> Dict[str, Any]:
+        try:
+            ocr = OCRSingleton.get()
+            with open(os.devnull, "w") as devnull:
+                with contextlib.redirect_stderr(devnull):
+                    result = ocr.predict(input=path)[0]
+            rec_texts = result["rec_texts"]
+            rec_scores = result["rec_scores"]
+            rec_polys = result["rec_polys"]
+        except Exception as e:
+            return {"error": str(e)}
+        items = []
+        for poly, text, score in zip(rec_polys, rec_texts, rec_scores):
+            items.append(OCRBox(poly=poly, text=text, score=score).model_dump())
+        return {"boxes": items}
+
+    return await asyncio.to_thread(_sync_ocr, path)
+
+
 def show_image(path: str) -> Dict[str, str]:
     """
     Reads an image from the specified URL or from the current work directory.
@@ -91,7 +151,7 @@ async def describe_image(
             - "general": General description of the image
             - "detailed": Detailed analysis of the image
             - "chess": Analysis of a chess position
-            - "text": Extract and describe text or numbers from the image
+            - "text": Extract and describe text or numbers with an OCR pipeline.
             - "custom": Custom description based on user prompt
     """
     image_base64 = show_image(path)["image_base64"]
@@ -116,4 +176,12 @@ async def describe_image(
         messages=[ChatMessage(role="user", content=content)],
         **llm_kwargs,
     )
+    if description_type == "text":
+        ocr_response = await _run_ocr(path)
+        response = json.dumps(
+            {
+                "vlm_response": response,
+                "ocr_response": ocr_response if ocr_response else [],
+            }
+        )
     return response

{academia_mcp-1.11.2.dist-info → academia_mcp-1.11.4.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: academia-mcp
-Version: 1.11.2
+Version: 1.11.4
 Summary: MCP server that provides different tools to search for scientific publications
 Author-email: Ilya Gusev <phoenixilya@gmail.com>
 Project-URL: Homepage, https://github.com/IlyaGusev/academia_mcp
@@ -12,16 +12,14 @@ Description-Content-Type: text/markdown
 License-File: LICENSE
 Requires-Dist: mcp>=1.10.1
 Requires-Dist: xmltodict>=0.14.0
-Requires-Dist: types-xmltodict>=0.14.0
 Requires-Dist: requests>=2.32.0
-Requires-Dist: types-requests>=2.32.0
 Requires-Dist: pypdf>=5.1.0
 Requires-Dist: beautifulsoup4>=4.12.0
-Requires-Dist: types-beautifulsoup4>=4.12.0
 Requires-Dist: markdownify==0.14.1
+Requires-Dist: types-xmltodict>=0.14.0
+Requires-Dist: types-requests>=2.32.0
+Requires-Dist: types-beautifulsoup4>=4.12.0
 Requires-Dist: acl-anthology==0.5.2
-Requires-Dist: markdown==3.7.0
-Requires-Dist: types-markdown==3.7.0.20250322
 Requires-Dist: huggingface-hub>=0.32.4
 Requires-Dist: fire>=0.7.0
 Requires-Dist: openai>=1.97.1
@@ -31,6 +29,8 @@ Requires-Dist: pymupdf>=1.26.4
 Requires-Dist: pillow>=11.3.0
 Requires-Dist: pydantic-settings>=2.6.0
 Requires-Dist: youtube-transcript-api>=1.2.2
+Requires-Dist: paddlepaddle>=3.2.0
+Requires-Dist: paddleocr>=3.2.0
 Dynamic: license-file
 
 # Academia MCP
@@ -70,12 +70,16 @@ make install
 ### Quickstart
 - Run over HTTP (default transport):
 ```bash
+python -m academia_mcp --transport streamable-http
+# OR
 uv run -m academia_mcp --transport streamable-http
 ```
 
 - Run over stdio (for local MCP clients like Claude Desktop):
 ```bash
 python -m academia_mcp --transport stdio
+# OR
+uv run -m academia_mcp --transport stdio
 ```
 
 Notes:
@@ -122,7 +126,7 @@ Availability notes:
 - Set one or more of `EXA_API_KEY`, `BRAVE_API_KEY`, `TAVILY_API_KEY` to enable `web_search` and provider tools.
 
 ### Environment variables
-Set as needed depending on which tools you use:
+Set as needed, depending on which tools you use:
 
 - `OPENROUTER_API_KEY`: required for LLM-related tools.
 - `BASE_URL`: override OpenRouter base URL.
@@ -152,7 +156,7 @@ docker run --rm -p 5056:5056 \
   academia_mcp
 ```
 
-Or use existing image: `phoenix120/academia_mcp`
+Or use existing image: [`phoenix120/academia_mcp`](https://hub.docker.com/repository/docker/phoenix120/academia_mcp)
 
 ### Examples
 - [Comprehensive report screencast (YouTube)](https://www.youtube.com/watch?v=4bweqQcN6w8)

{academia_mcp-1.11.2.dist-info → academia_mcp-1.11.4.dist-info}/RECORD

@@ -1,33 +1,33 @@
 academia_mcp/__init__.py,sha256=2Ru2I5u4cE7DrkkAsibDUEF1K6sYtqppb9VyFrRoQKI,94
 academia_mcp/__main__.py,sha256=rcmsOtJd3SA82exjrcGBuxuptcoxF8AXI7jNjiVq2BY,59
-academia_mcp/files.py,sha256=ynIt0XbU1Z7EPWkv_hVX0pGKsLlmjYv-MVJLOfi6yzs,817
+academia_mcp/files.py,sha256=J_81COWWryBp2bHVe_9dg9c0rbVIzv7-zWkJ7CS9lr4,485
 academia_mcp/llm.py,sha256=zpGkuJFf58Ofgys_fi28-47_wJ1a7sIs_yZvI1Si6z0,993
 academia_mcp/pdf.py,sha256=9PlXzHGhb6ay3ldbTdxCcTWvH4TkET3bnb64mgoh9i0,1273
 academia_mcp/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-academia_mcp/server.py,sha256=uJNV5YStMJ3sEbx4waOclSw_hN0Gkz5PJ-q8VjX1kfA,6390
+academia_mcp/server.py,sha256=sxr_M1JQif29hSuduXvoknPgidwoau8WKszT_P9mEVI,6396
 academia_mcp/settings.py,sha256=c5s4dI8V_cWmMED-jKDmHjfdIaBcxwEK4HdHNQ3WUIg,1096
 academia_mcp/utils.py,sha256=ixtYI7qidFJpc8Tzc5sokseCLx4r0yFFgbktKaY-Ixo,4904
 academia_mcp/latex_templates/agents4science_2025/agents4science_2025.sty,sha256=hGcEPCYBJS4vdhWvN_yEaJC4GvT_yDroI94CfY2Oguk,12268
 academia_mcp/latex_templates/agents4science_2025/agents4science_2025.tex,sha256=Tl1QkHXHRopw9VEfWrD3Layr5JP_0gIzVQjL4KXIWqc,15814
-academia_mcp/tools/__init__.py,sha256=Z30vULZwUeUX5nDz5wcv0znhAeBtZRa0dvz7vD8SUYE,1555
+academia_mcp/tools/__init__.py,sha256=cCS073dW2S5T_YWrbwd-H9QRiervZ7bof_2sdvzR8JQ,1561
 academia_mcp/tools/anthology_search.py,sha256=rhFpJZqGLABgr0raDuH0CARBiAJNJtEI4dlMrKNHfDQ,7669
 academia_mcp/tools/arxiv_download.py,sha256=4zl9QkWF7uU2BYz4XH7Fu_51htolSYtO7a2v4_ikhxg,10633
 academia_mcp/tools/arxiv_search.py,sha256=p4DeCvV3TUpj58etx0QOxm-GYKAWkP9AGjLv4HUUqUc,8857
 academia_mcp/tools/bitflip.py,sha256=0OPxQRU5VLWrcIlj4Ubjeq4gIq0pR5FDhXDHuLR0ppw,12759
 academia_mcp/tools/document_qa.py,sha256=Wb2nEEVu9UyPp8ktHWeT9wS2JBle8fb9zRjTNVIDdBE,2463
 academia_mcp/tools/hf_datasets_search.py,sha256=KiBkqT4rXjEN4oc1AWZOPnqN_Go90TQogY5-DUm3LQo,2854
+academia_mcp/tools/image_processing.py,sha256=BFj5D0lYbe6OCAL9LSHjS1adr8C0BrtWBXD4xozkC7I,6776
 academia_mcp/tools/latex.py,sha256=B1Leqt1FHY6H3DlUgeYse4LMFpf4-K1FQViXl5MKk8A,6144
 academia_mcp/tools/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 academia_mcp/tools/review.py,sha256=Va0lFJJKuk-NvWhKS3UZ-Dnuk7CyuDQ4S1nd70D-ffE,11117
 academia_mcp/tools/s2.py,sha256=ykJOkpHnyZRlfyDnCIL9m-Rnu5dBecoKxg0SjlzdvLk,6457
-academia_mcp/tools/show_image.py,sha256=DWSnYMTn_dJpGTLL1r_sbX5XsB6p9z-vClApDANz84s,4534
 academia_mcp/tools/speech_to_text.py,sha256=YZzMqdvunzXkpcadP_mYhm6cs4qH1Y_42SfY-7eX4O4,1601
 academia_mcp/tools/visit_webpage.py,sha256=uJZx9vBGS8q-J-VH4Pr7T9lNtDsWU83gJhlotcd1ajg,3788
 academia_mcp/tools/web_search.py,sha256=CHgco8DufTFwtVecgDOOMylIY99iUmCdb0oZtpGntx0,8646
 academia_mcp/tools/yt_transcript.py,sha256=ilfOpX14moC1bKHbFmOVvZ8-_NxuQQUoQbV28e9FBaE,1217
-academia_mcp-1.11.2.dist-info/licenses/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
-academia_mcp-1.11.2.dist-info/METADATA,sha256=-Nbs00tnl2H1LhBJrVriuLsvbwANaqAHwQsbAn9u79k,6356
-academia_mcp-1.11.2.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-academia_mcp-1.11.2.dist-info/entry_points.txt,sha256=gxkiKJ74w2FwJpSECpjA3XtCfI5ZfrM6N8cqnwsq4yY,51
-academia_mcp-1.11.2.dist-info/top_level.txt,sha256=CzGpRFsRRJRqWEb1e3SUlcfGqRzOxevZGaJWrtGF8W0,13
-academia_mcp-1.11.2.dist-info/RECORD,,
+academia_mcp-1.11.4.dist-info/licenses/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
+academia_mcp-1.11.4.dist-info/METADATA,sha256=j4_CAs-0UIfhs0ai-XdrSFJcCvid4AkVbcqNnvhrQrg,6517
+academia_mcp-1.11.4.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+academia_mcp-1.11.4.dist-info/entry_points.txt,sha256=gxkiKJ74w2FwJpSECpjA3XtCfI5ZfrM6N8cqnwsq4yY,51
+academia_mcp-1.11.4.dist-info/top_level.txt,sha256=CzGpRFsRRJRqWEb1e3SUlcfGqRzOxevZGaJWrtGF8W0,13
+academia_mcp-1.11.4.dist-info/RECORD,,