academia-mcp 1.2.2__py3-none-any.whl → 1.4.0__py3-none-any.whl

academia_mcp/files.py

@@ -5,6 +5,7 @@ from pathlib import Path
 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:

academia_mcp/server.py

@@ -12,7 +12,12 @@ from .tools.s2_citations import s2_get_citations, s2_get_references
 from .tools.hf_datasets_search import hf_datasets_search
 from .tools.anthology_search import anthology_search
 from .tools.document_qa import document_qa
-from .tools.md_to_pdf import md_to_pdf
+from .tools.latex import (
+    compile_latex_from_file,
+    compile_latex_from_str,
+    get_latex_template,
+    get_latex_templates_list,
+)
 from .tools.web_search import web_search, tavily_web_search, exa_web_search, brave_web_search
 from .tools.visit_webpage import visit_webpage
 from .tools.bitflip import (
@@ -33,7 +38,7 @@ def find_free_port() -> int:
                 return port
         except Exception:
             continue
-    return 5000
+    raise RuntimeError("No free port in range 5000-6000 found")
 
 
 def run(
@@ -42,6 +47,8 @@ def run(
     mount_path: str = "/",
     streamable_http_path: str = "/mcp",
     transport: Literal["stdio", "sse", "streamable-http"] = "streamable-http",
+    disable_web_search_tools: bool = False,
+    disable_llm_tools: bool = False,
 ) -> None:
     server = FastMCP(
         "Academia MCP",
@@ -56,21 +63,26 @@ def run(
     server.add_tool(s2_get_references)
     server.add_tool(hf_datasets_search)
     server.add_tool(anthology_search)
-    server.add_tool(md_to_pdf)
+    server.add_tool(compile_latex_from_file)
+    server.add_tool(compile_latex_from_str)
+    server.add_tool(get_latex_template)
+    server.add_tool(get_latex_templates_list)
     server.add_tool(visit_webpage)
-    server.add_tool(extract_bitflip_info)
-    server.add_tool(generate_research_proposal)
-    server.add_tool(score_research_proposals)
 
-    if os.getenv("TAVILY_API_KEY"):
-        server.add_tool(tavily_web_search)
-    if os.getenv("EXA_API_KEY"):
-        server.add_tool(exa_web_search)
-    if os.getenv("BRAVE_API_KEY"):
-        server.add_tool(brave_web_search)
-    if os.getenv("EXA_API_KEY") or os.getenv("BRAVE_API_KEY") or os.getenv("TAVILY_API_KEY"):
-        server.add_tool(web_search)
-    if os.getenv("OPENROUTER_API_KEY"):
+    if not disable_web_search_tools:
+        if os.getenv("TAVILY_API_KEY"):
+            server.add_tool(tavily_web_search)
+        if os.getenv("EXA_API_KEY"):
+            server.add_tool(exa_web_search)
+        if os.getenv("BRAVE_API_KEY"):
+            server.add_tool(brave_web_search)
+        if os.getenv("EXA_API_KEY") or os.getenv("BRAVE_API_KEY") or os.getenv("TAVILY_API_KEY"):
+            server.add_tool(web_search)
+
+    if not disable_llm_tools and os.getenv("OPENROUTER_API_KEY"):
+        server.add_tool(extract_bitflip_info)
+        server.add_tool(generate_research_proposal)
+        server.add_tool(score_research_proposals)
         server.add_tool(document_qa)
 
     if port is None:

academia_mcp/tools/__init__.py

@@ -4,7 +4,12 @@ from .arxiv_download import arxiv_download
 from .hf_datasets_search import hf_datasets_search
 from .s2_citations import s2_get_references, s2_get_citations
 from .document_qa import document_qa
-from .md_to_pdf import md_to_pdf
+from .latex import (
+    compile_latex_from_file,
+    compile_latex_from_str,
+    get_latex_template,
+    get_latex_templates_list,
+)
 from .web_search import web_search, tavily_web_search, exa_web_search, brave_web_search
 from .visit_webpage import visit_webpage
 from .bitflip import extract_bitflip_info, generate_research_proposal, score_research_proposals
@@ -18,7 +23,10 @@ __all__ = [
     "s2_get_citations",
     "hf_datasets_search",
     "document_qa",
-    "md_to_pdf",
+    "compile_latex_from_file",
+    "compile_latex_from_str",
+    "get_latex_template",
+    "get_latex_templates_list",
     "web_search",
     "tavily_web_search",
     "exa_web_search",

academia_mcp/tools/anthology_search.py

@@ -34,20 +34,13 @@ def _format_authors(authors: List[Any]) -> str:
     return result
 
 
-def _format_date(date_str: str) -> str:
-    try:
-        return datetime.strptime(date_str, "%Y").strftime("%B %d, %Y")
-    except ValueError:
-        return date_str
-
-
 def _clean_entry(entry: Any) -> Dict[str, Any]:
     return {
         "id": entry.full_id,
         "title": _format_text_field(entry.title.as_text()),
         "authors": _format_authors(entry.authors),
         "abstract": (_format_text_field(entry.abstract.as_text()) if entry.abstract else ""),
-        "published": _format_date(entry.year),
+        "published_year": entry.year,
         "categories": ", ".join(entry.venue_ids),
         "comment": entry.note if entry.note else "",
         "url": entry.pdf.url if entry.pdf else "",

academia_mcp/tools/arxiv_download.py

@@ -35,6 +35,8 @@ def parse_pdf_file(pdf_path: Path) -> List[str]:
     for page_number, page in enumerate(reader.pages, start=1):
         try:
             text = page.extract_text()
+            if not text:
+                continue
             prefix = f"## Page {page_number}\n\n"
             pages.append(prefix + text)
         except Exception:

academia_mcp/tools/document_qa.py

@@ -6,6 +6,7 @@ from dotenv import load_dotenv
 from pydantic import BaseModel
 
 from academia_mcp.llm import llm_acall
+from academia_mcp.utils import truncate_content
 
 load_dotenv()
 
@@ -62,9 +63,11 @@ async def document_qa(
     assert question and question.strip(), "Please provide non-empty 'question'"
     if isinstance(document, dict):
         document = json.dumps(document)
-
     assert document and document.strip(), "Please provide non-empty 'document'"
 
+    question = truncate_content(question, 10000)
+    document = truncate_content(document, 200000)
+
     model_name = os.getenv("DOCUMENT_QA_MODEL_NAME", "deepseek/deepseek-chat-v3-0324")
     prompt = PROMPT.format(question=question, document=document)
     content = await llm_acall(model_name=model_name, prompt=prompt)

academia_mcp/tools/latex.py

@@ -0,0 +1,151 @@
+import re
+import subprocess
+import shutil
+import tempfile
+import json
+from pathlib import Path
+
+
+from academia_mcp.files import get_workspace_dir, DEFAULT_LATEX_TEMPLATES_DIR_PATH
+
+
+def get_latex_templates_list() -> str:
+    """
+    Get the list of available latex templates.
+    Always use one of the templates from the list.
+
+    Returns a JSON list serialized to a string.
+    Use `json.loads` to deserialize the result.
+    """
+    return json.dumps([str(path.name) for path in DEFAULT_LATEX_TEMPLATES_DIR_PATH.glob("*")])
+
+
+def get_latex_template(template_name: str) -> str:
+    """
+    Get the latex template by name.
+
+    Returns a JSON object serialized to a string.
+    Use `json.loads` to deserialize the result.
+    The structure is: {"template": "...", "style": "..."}
+
+    Args:
+        template_name: The name of the latex template.
+    """
+    template_dir_path = DEFAULT_LATEX_TEMPLATES_DIR_PATH / template_name
+    if not template_dir_path.exists():
+        raise FileNotFoundError(
+            f"Template {template_name} not found in {DEFAULT_LATEX_TEMPLATES_DIR_PATH}"
+        )
+    template_path = template_dir_path / f"{template_name}.tex"
+    style_path = template_dir_path / f"{template_name}.sty"
+    if not template_path.exists():
+        raise FileNotFoundError(f"Template file {template_path} not found in {template_dir_path}")
+    if not style_path.exists():
+        raise FileNotFoundError(f"Style file {style_path} not found in {template_dir_path}")
+    return json.dumps({"template": template_path.read_text(), "style": style_path.read_text()})
+
+
+def compile_latex_from_file(
+    input_filename: str, output_filename: str = "output.pdf", timeout: int = 60
+) -> str:
+    """
+    Compile a latex file.
+
+    Returns a string with the result of the compilation.
+
+    Args:
+        input_filename: The path to the latex file.
+        output_filename: The path to the output pdf file.
+        timeout: The timeout for the compilation. 60 seconds by default.
+    """
+    with open(input_filename, "r", encoding="utf-8") as file:
+        latex_code = file.read()
+    return compile_latex_from_str(latex_code, output_filename, timeout)
+
+
+def compile_latex_from_str(
+    latex_code: str, output_filename: str = "output.pdf", timeout: int = 60
+) -> str:
+    """
+    Compile a latex code.
+
+    Returns a string with the result of the compilation.
+
+    Args:
+        latex_code: The latex code to compile.
+        output_filename: The path to the output pdf file.
+        timeout: The timeout for the compilation. 60 seconds by default.
+    """
+    if shutil.which("pdflatex") is None:
+        return "pdflatex is not installed or not found in PATH."
+
+    destination_name = (
+        output_filename if output_filename.lower().endswith(".pdf") else f"{output_filename}.pdf"
+    )
+
+    try:
+        with tempfile.TemporaryDirectory(
+            dir=str(get_workspace_dir()), prefix="temp_latex_"
+        ) as temp_dir:
+            temp_dir_path = Path(temp_dir)
+            tex_filename = "temp.tex"
+            pdf_filename = "temp.pdf"
+            tex_file_path = temp_dir_path / tex_filename
+            tex_file_path.write_text(latex_code, encoding="utf-8")
+
+            # Detect and copy local .sty packages referenced by \usepackage{...}
+            # Supports optional arguments: \usepackage[opts]{pkgA,pkgB}
+            try:
+                package_names: set[str] = set()
+                for match in re.finditer(r"\\usepackage(?:\[[^\]]*\])?\{([^}]+)\}", latex_code):
+                    for name in match.group(1).split(","):
+                        pkg = name.strip()
+                        if pkg:
+                            package_names.add(pkg)
+
+                for pkg in package_names:
+                    sty_name = f"{pkg}.sty"
+                    for candidate in DEFAULT_LATEX_TEMPLATES_DIR_PATH.rglob(sty_name):
+                        shutil.copyfile(candidate, temp_dir_path / sty_name)
+                        break
+            except Exception:
+                pass
+
+            try:
+                subprocess.run(
+                    [
+                        "pdflatex",
+                        "-interaction=nonstopmode",
+                        tex_filename,
+                    ],
+                    cwd=str(temp_dir_path),
+                    check=True,
+                    capture_output=True,
+                    text=True,
+                    timeout=timeout,
+                )
+            except subprocess.TimeoutExpired:
+                return f"Compilation timed out after {timeout} seconds"
+            except subprocess.CalledProcessError as e:
+                combined_output = (e.stdout or "") + "\n" + (e.stderr or "")
+                error_lines = [
+                    line
+                    for line in combined_output.split("\n")
+                    if ("error" in line.lower() or "!" in line)
+                ]
+                if error_lines:
+                    return "Compilation failed. LaTeX errors:\n" + "\n".join(error_lines)
+                return f"Compilation failed. Full LaTeX output:\n{combined_output}"
+
+            pdf_path = temp_dir_path / pdf_filename
+            output_pdf_path = Path(get_workspace_dir()) / destination_name
+
+            if pdf_path.exists():
+                shutil.move(str(pdf_path), str(output_pdf_path))
+                return f"Compilation successful! PDF file saved as {destination_name}"
+
+            return (
+                "Compilation completed, but PDF file was not created. Check LaTeX code for errors."
+            )
+    except Exception as e:
+        return f"Compilation failed due to an unexpected error: {e}"

academia_mcp/tools/visit_webpage.py

@@ -11,6 +11,17 @@ EXA_CONTENTS_URL = "https://api.exa.ai/contents"
 AVAILABLE_PROVIDERS = ("basic", "exa")
 
 
+def _exa_visit_webpage(url: str) -> str:
+    key = os.getenv("EXA_API_KEY", "")
+    assert key, "Error: EXA_API_KEY is not set and no api_key was provided"
+    payload = {
+        "urls": [url],
+        "text": True,
+    }
+    response = post_with_retries(EXA_CONTENTS_URL, payload=payload, api_key=key)
+    return json.dumps(response.json()["results"][0])
+
+
 def visit_webpage(url: str, provider: Optional[str] = "basic") -> str:
     """
     Visit a webpage and return the content.
@@ -28,17 +39,17 @@ def visit_webpage(url: str, provider: Optional[str] = "basic") -> str:
     ), f"Invalid provider: {provider}. Available providers: {AVAILABLE_PROVIDERS}"
 
     if provider == "exa":
-        key = os.getenv("EXA_API_KEY", "")
-        assert key, "Error: EXA_API_KEY is not set and no api_key was provided"
-        payload = {
-            "urls": [url],
-            "text": True,
-        }
-        response = post_with_retries(EXA_CONTENTS_URL, payload=payload, api_key=key)
-        return json.dumps(response.json()["results"][0])
+        return _exa_visit_webpage(url)
 
     assert provider == "basic"
     response = get_with_retries(url)
+    content_type = response.headers.get("content-type", "").lower()
+    if not content_type or (not content_type.startswith("text/") and "html" not in content_type):
+        if os.getenv("EXA_API_KEY"):
+            return _exa_visit_webpage(url)
+        return json.dumps(
+            {"id": url, "error": f"Unsupported content-type: {content_type or 'unknown'}"}
+        )
     markdown_content = markdownify(response.text).strip()
     markdown_content = re.sub(r"\n{3,}", "\n\n", markdown_content)
     return json.dumps({"id": url, "text": markdown_content})

academia_mcp/tools/web_search.py

@@ -17,7 +17,8 @@ def web_search(
     provider: Optional[str] = "tavily",
 ) -> str:
     """
-    Search the web using Exa Search or Tavily and return normalized results.
+    Search the web using Exa Search, Brave Search or Tavily and return normalized results.
+    If the specified provider is not available, the function will try to use the next available provider.
 
     Returns a JSON object serialized to a string. The structure is: {"results": [...]}
     Every item in the "results" has at least the following fields: ("title", "url")
@@ -47,13 +48,15 @@ def web_search(
                 provider = p
                 break
 
+    result = {}
     if provider == "exa":
-        return exa_web_search(query, limit)
+        result = json.loads(exa_web_search(query, limit))
     elif provider == "brave":
-        return brave_web_search(query, limit)
-
-    assert provider == "tavily"
-    return tavily_web_search(query, limit)
+        result = json.loads(brave_web_search(query, limit))
+    elif provider == "tavily":
+        result = json.loads(tavily_web_search(query, limit))
+    result["search_provider"] = provider
+    return json.dumps(result, ensure_ascii=False)
 
 
 def tavily_web_search(query: str, limit: Optional[int] = 20) -> str:

academia_mcp/utils.py

@@ -13,10 +13,11 @@ def post_with_retries(
     api_key: Optional[str] = None,
     timeout: int = 30,
     num_retries: int = 3,
+    backoff_factor: float = 3.0,
 ) -> requests.Response:
     retry_strategy = Retry(
         total=num_retries,
-        backoff_factor=3,
+        backoff_factor=backoff_factor,
         status_forcelist=[429, 500, 502, 503, 504],
         allowed_methods=["POST"],
     )
@@ -24,6 +25,7 @@ def post_with_retries(
     session = requests.Session()
     adapter = requests.adapters.HTTPAdapter(max_retries=retry_strategy)
     session.mount("https://", adapter)
+    session.mount("http://", adapter)
 
     headers = {
         "x-api-key": api_key,
@@ -42,11 +44,12 @@ def get_with_retries(
     api_key: Optional[str] = None,
     timeout: int = 30,
     num_retries: int = 3,
+    backoff_factor: float = 3.0,
     params: Optional[Dict[str, Any]] = None,
 ) -> requests.Response:
     retry_strategy = Retry(
         total=num_retries,
-        backoff_factor=30,
+        backoff_factor=backoff_factor,
         status_forcelist=[429, 500, 502, 503, 504],
         allowed_methods=["GET"],
     )
@@ -54,6 +57,7 @@ def get_with_retries(
     session = requests.Session()
     adapter = requests.adapters.HTTPAdapter(max_retries=retry_strategy)
     session.mount("https://", adapter)
+    session.mount("http://", adapter)
 
     headers = {}
     if api_key:
@@ -145,3 +149,19 @@ def extract_json(text: str) -> Any:
 def encode_prompt(template: str, **kwargs: Any) -> str:
     template_obj = Template(template)
     return template_obj.render(**kwargs).strip()
+
+
+def truncate_content(
+    content: str,
+    max_length: int,
+) -> str:
+    disclaimer = (
+        f"\n\n..._This content has been truncated to stay below {max_length} characters_...\n\n"
+    )
+    half_length = max_length // 2
+    if len(content) <= max_length:
+        return content
+
+    prefix = content[:half_length]
+    suffix = content[-half_length:]
+    return prefix + disclaimer + suffix

{academia_mcp-1.2.2.dist-info → academia_mcp-1.4.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: academia-mcp
-Version: 1.2.2
+Version: 1.4.0
 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
@@ -22,16 +22,11 @@ Requires-Dist: markdownify==0.14.1
 Requires-Dist: acl-anthology==0.5.2
 Requires-Dist: markdown==3.7.0
 Requires-Dist: types-markdown==3.7.0.20250322
-Requires-Dist: black==25.1.0
-Requires-Dist: mypy==1.16.0
-Requires-Dist: flake8==7.2.0
 Requires-Dist: huggingface-hub>=0.32.4
 Requires-Dist: fire>=0.7.0
-Requires-Dist: pytest>=8.4.1
 Requires-Dist: openai>=1.97.1
 Requires-Dist: jinja2>=3.1.6
 Requires-Dist: datasets>=4.0.0
-Requires-Dist: pytest-asyncio>=1.1.0
 Dynamic: license-file
 
 # Academia MCP
@@ -44,16 +39,26 @@ A collection of MCP tools related to the search of scientific papers:
 - Web search: Exa/Brave/Tavily
 - Page crawler
 
-Install:
+## Install
+
+- Using pip (end users):
 ```
 pip3 install academia-mcp
 ```
 
+- For development (uv + Makefile):
+```
+uv venv .venv
+make install
+```
+
+## Examples
 Comprehensive report screencast: https://www.youtube.com/watch?v=4bweqQcN6w8
 
 Single paper screencast: https://www.youtube.com/watch?v=IAAPMptJ5k8
 
-Claude Desktop config:
+
+## Claude Desktop config
 ```
 {
   "mcpServers": {
@@ -69,3 +74,41 @@ Claude Desktop config:
   }
 }
 ```
+
+## Running the server (CLI)
+
+```
+uv run -m academia_mcp --transport streamable-http
+```
+
+Notes:
+- Transports supported: `stdio`, `sse`, `streamable-http`.
+- Host/port are used for HTTP transports; for `stdio` they are ignored.
+
+## Makefile targets
+
+- `make install`: install the package in editable mode with uv.
+- `make validate`: run black, flake8, and mypy (strict).
+- `make test`: run the test suite with pytest.
+- `make publish`: build and publish using uv.
+
+## Environment variables
+
+Set as needed depending on which tools you use:
+
+- `TAVILY_API_KEY`: enables Tavily in `web_search`.
+- `EXA_API_KEY`: enables Exa in `web_search` and `visit_webpage`.
+- `BRAVE_API_KEY`: enables Brave in `web_search`.
+- `OPENROUTER_API_KEY`: required for `document_qa`.
+- `BASE_URL`: override OpenRouter base URL for `document_qa` and bitflip tools.
+- `DOCUMENT_QA_MODEL_NAME`: override default model for `document_qa`.
+- `BITFLIP_MODEL_NAME`: override default model for bitflip tools.
+- `WORKSPACE_DIR`: directory for generated files (PDFs, temp artifacts).
+
+## md_to_pdf requirements
+
+The `md_to_pdf` tool invokes `pdflatex`. Ensure a LaTeX distribution is installed and `pdflatex` is on PATH. On Debian/Ubuntu:
+
+```
+sudo apt install texlive-latex-base texlive-fonts-recommended texlive-latex-extra texlive-science
+```

academia_mcp-1.4.0.dist-info/RECORD

@@ -0,0 +1,25 @@
+academia_mcp/__init__.py,sha256=2Ru2I5u4cE7DrkkAsibDUEF1K6sYtqppb9VyFrRoQKI,94
+academia_mcp/__main__.py,sha256=rcmsOtJd3SA82exjrcGBuxuptcoxF8AXI7jNjiVq2BY,59
+academia_mcp/files.py,sha256=tvt3OPr5q6pAPCZ0XvRHHL9ZWuTXINRZvqjeRFmx5YE,815
+academia_mcp/llm.py,sha256=o84FQNSbjjVSk9DlvFXWsUDiz5IOaavYU6kOqnPEG7E,1071
+academia_mcp/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+academia_mcp/server.py,sha256=FRrPAacAPs1IZ3LRKDFZi7copAqzy_aPGYd2RLsA01U,2974
+academia_mcp/utils.py,sha256=P9U3RjYzcztE0KxXvJSy5wSBaUg2CM9tpByljYrsrl4,4607
+academia_mcp/tools/__init__.py,sha256=u_6HkChV2P46zXxGp92s1cTSjkkd42udhCM3BFDYQ_c,1137
+academia_mcp/tools/anthology_search.py,sha256=rhFpJZqGLABgr0raDuH0CARBiAJNJtEI4dlMrKNHfDQ,7669
+academia_mcp/tools/arxiv_download.py,sha256=soa9nPDHV1-ZgTLtqh-Fj69WkRsrb1hRKouWDa2ePng,11269
+academia_mcp/tools/arxiv_search.py,sha256=pzM18qrF3QL03A53w003kE7hQi3s3QKtjgw0m7K88UY,8355
+academia_mcp/tools/bitflip.py,sha256=u0hSOPWbnCDu2EbA_RkueX496SvTKz9QhZcXugshSfI,10949
+academia_mcp/tools/document_qa.py,sha256=04pJpiYCg27EFiZhfmTaMjeobu8SMT0Dls7OAFDoH00,2392
+academia_mcp/tools/hf_datasets_search.py,sha256=KiBkqT4rXjEN4oc1AWZOPnqN_Go90TQogY5-DUm3LQo,2854
+academia_mcp/tools/latex.py,sha256=rM6Xm4dXkpbhOMuxx-lK83dzCEAs_bWhRaK44hyGwuU,5639
+academia_mcp/tools/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+academia_mcp/tools/s2_citations.py,sha256=dqrBp76RrX1zH2XzcMAoWBbvbtyhxLeF-xnqOKD_JiM,4852
+academia_mcp/tools/visit_webpage.py,sha256=OZdqDkVPIbANyFw5o5jIjU5Rr_dolxrGDs63Ud-GmRM,1966
+academia_mcp/tools/web_search.py,sha256=mobKm4iqKppn8pduZYMzWRo1MQBjkAqmMtrFLI5XY2Y,6296
+academia_mcp-1.4.0.dist-info/licenses/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
+academia_mcp-1.4.0.dist-info/METADATA,sha256=6A2AjFm3IplORaP02xXTgBdTNY00Bz56nU5t1w5gNg8,3172
+academia_mcp-1.4.0.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+academia_mcp-1.4.0.dist-info/entry_points.txt,sha256=gxkiKJ74w2FwJpSECpjA3XtCfI5ZfrM6N8cqnwsq4yY,51
+academia_mcp-1.4.0.dist-info/top_level.txt,sha256=CzGpRFsRRJRqWEb1e3SUlcfGqRzOxevZGaJWrtGF8W0,13
+academia_mcp-1.4.0.dist-info/RECORD,,

academia_mcp/tools/md_to_pdf.py

@@ -1,404 +0,0 @@
-import markdown
-import re
-import os
-import subprocess
-import shutil
-import xml.dom.minidom
-import xml.etree.ElementTree as etree
-from typing import Optional, Any
-
-from markdown.core import Markdown
-
-from academia_mcp.files import get_workspace_dir
-
-
-START_SINGLE_QUOTE_RE = re.compile(r"(^|\s|\")'")
-START_DOUBLE_QUOTE_RE = re.compile(r"(^|\s|'|`)\"")
-END_DOUBLE_QUOTE_RE = re.compile(r'"(,|\.|\s|$)')
-
-
-MAIN_TEMPLATE = """\\documentclass{{article}}
-\\usepackage[utf8]{{inputenc}}
-\\usepackage[T1]{{fontenc}}
-\\usepackage{{textcomp}}
-\\usepackage{{amsmath}}
-\\usepackage{{float}}
-\\usepackage{{graphicx}}
-\\usepackage{{enumitem}}
-\\usepackage{{quoting}}
-\\usepackage{{booktabs}}
-\\usepackage{{caption}}
-\\usepackage{{siunitx}}
-\\sisetup{{
-  group-separator = {{,}},
-  output-decimal-marker = {{.}}
-}}
-\\usepackage{{hyperref}}
-
-\\author{{Holosophos}}
-
-\\begin{{document}}
-
-{latex_content}
-
-\\end{{document}}"""
-
-
-IMAGE_TEMPLATE = """\\begin{{figure}}[H]
-\\centering
-\\includegraphics[width=\\linewidth]{{{src}}}
-\\caption{{{alt}}}
-\\end{{figure}}"""
-
-
-TABLE_TEMPLATE = """
-\\begin{{table}}[h]
-\\begin{{tabular}}{{{descriptor}}}
-{core}
-\\hline
-\\end{{tabular}}
-\\\\[5pt]
-\\caption{{{caption}}}
-\\end{{table}}
-"""
-
-ITEMIZE_TEMPLATE = """
-\\begin{{itemize}}
-{content}
-\\end{{itemize}}
-"""
-
-QUOTE_TEMPLATE = """
-\\begin{{quotation}}
-{content}
-\\end{{quotation}}
-"""
-
-VERBATIM_TEMPLATE = """
-\\begin{{verbatim}}
-{content}
-\\end{{verbatim}}
-"""
-
-
-MAKETITLE = """
-% ----------------------------------------------------------------
-\\maketitle
-% ----------------------------------------------------------------
-"""
-
-
-def inline_html_latex(text: str) -> str:
-    out = text
-    if re.search(r"“.*?”", text, flags=re.DOTALL):
-        out = out.replace("“", "\\enquote{").replace("”", "}")
-    if re.search(r"‘.*?’", text, flags=re.DOTALL):
-        out = out.replace("‘", "\\enquote{").replace("’", "}")
-    if re.search(r"“.*?“", text, flags=re.DOTALL):
-        out = out.replace("“", "\\enquote{", 1).replace("“", "}", 1)
-    if re.search(r"«.*?»", text, flags=re.DOTALL):
-        out = out.replace("«", "\\enquote{").replace("»", "}")
-    out = out.replace("...", "\\dots")
-    out = out.replace("…", "\\dots")
-    out = out.replace("–", "--")
-    out = out.replace("—", "---")
-    out = out.replace("\\|", "|")
-    return out
-
-
-def unescape_html_entities(text: str) -> str:
-    mapping = {
-        "&": "&",
-        "&lt;": "<",
-        "&gt;": ">",
-        "&quot;": '"',
-    }
-    for k, v in mapping.items():
-        text = text.replace(k, v)
-    return text
-
-
-def escape_latex_entities(text: str) -> str:
-    out = unescape_html_entities(text)
-    out = out.replace("%", "\\%")
-    out = out.replace("&", "\\&")
-    out = out.replace("#", "\\#")
-    out = START_SINGLE_QUOTE_RE.sub(r"\g<1>`", out)
-    out = START_DOUBLE_QUOTE_RE.sub(r"\g<1>``", out)
-    out = END_DOUBLE_QUOTE_RE.sub(r"''\g<1>", out)
-    return out
-
-
-class LaTeXExtension(markdown.Extension):
-    def __init__(self, configs: Optional[Any] = None) -> None:
-        self.reset()
-
-    def extendMarkdown(self, md: Markdown) -> None:
-        self.md = md
-        latex_tp = LaTeXTreeProcessor()
-        math_pp = MathTextPostProcessor()
-        table_pp = TableTextPostProcessor()
-        image_pp = ImageTextPostProcessor()
-        link_pp = LinkTextPostProcessor()
-        unescape_html_pp = UnescapeHtmlTextPostProcessor()
-
-        md.treeprocessors.register(latex_tp, "latex", 20)
-        md.postprocessors.register(unescape_html_pp, "unescape_html", 20)
-        md.postprocessors.register(math_pp, "math", 20)
-        md.postprocessors.register(image_pp, "image", 20)
-        md.postprocessors.register(table_pp, "table", 20)
-        md.postprocessors.register(link_pp, "link", 20)
-
-    def reset(self) -> None:
-        pass
-
-
-class LaTeXTreeProcessor(markdown.treeprocessors.Treeprocessor):
-    def run(self, doc: etree.Element) -> None:
-        latex_text = self.tolatex(doc)
-        doc.clear()
-        latex_node = etree.Element("plaintext")
-        latex_node.text = latex_text
-        doc.append(latex_node)
-
-    def tolatex(self, ournode: etree.Element) -> str:
-        buf = ""
-        subcontent = ""
-
-        if ournode.text:
-            subcontent += escape_latex_entities(ournode.text)
-
-        for child in list(ournode):
-            subcontent += self.tolatex(child)
-
-        tag = ournode.tag
-        if tag == "h1":
-            buf += "\n\\title{%s}\n" % subcontent
-            buf += MAKETITLE
-        elif tag == "h2":
-            buf += "\n\n\\section{%s}\n" % subcontent
-        elif tag == "h3":
-            buf += "\n\n\\subsection{%s}\n" % subcontent
-        elif tag == "h4":
-            buf += "\n\\subsubsection{%s}\n" % subcontent
-        elif tag == "hr":
-            buf += "\\noindent\\makebox[\\linewidth]{\\rule{\\linewidth}{0.4pt}}"
-        elif tag == "ul":
-            buf += ITEMIZE_TEMPLATE.format(content=subcontent.strip())
-        elif tag == "ol":
-            buf += " \\begin{enumerate}"
-            if "start" in ournode.attrib:
-                start = int(ournode.attrib["start"]) - 1
-                buf += "\\setcounter{enumi}{" + str(start) + "}"
-            buf += f"\n{subcontent}\n\\end{{enumerate}}"
-        elif tag == "li":
-            buf += "\n  \\item %s" % subcontent.strip()
-        elif tag == "blockquote":
-            buf += QUOTE_TEMPLATE.format(content=subcontent.strip())
-        elif tag == "pre":
-            buf += VERBATIM_TEMPLATE.format(content=subcontent.strip())
-        elif tag == "q":
-            buf += "`%s'" % subcontent.strip()
-        elif tag == "p":
-            buf += "\n%s\n" % subcontent.strip()
-        elif tag == "sup":
-            buf += "\\footnote{%s}" % subcontent.strip()
-        elif tag == "strong":
-            buf += "\\textbf{%s}" % subcontent.strip()
-        elif tag == "em":
-            buf += "\\emph{%s}" % subcontent.strip()
-        elif tag == "table":
-            buf += "\n\n<table>%s</table>\n\n" % subcontent
-        elif tag == "thead":
-            buf += "<thead>%s</thead>" % subcontent
-        elif tag == "tbody":
-            buf += "<tbody>%s</tbody>" % subcontent
-        elif tag == "tr":
-            buf += "<tr>%s</tr>" % subcontent
-        elif tag == "th":
-            buf += "<th>%s</th>" % subcontent
-        elif tag == "td":
-            buf += "<td>%s</td>" % subcontent
-        elif tag == "img":
-            buf += '<img src="%s" alt="%s" />' % (
-                ournode.get("src"),
-                ournode.get("alt"),
-            )
-        elif tag == "a":
-            href = ournode.get("href")
-            assert href
-            buf += '<a href="%s">%s</a>' % (
-                escape_latex_entities(href),
-                subcontent,
-            )
-        else:
-            buf = subcontent
-
-        if ournode.tail:
-            buf += escape_latex_entities(ournode.tail)
-
-        return buf
-
-
-class Table2Latex:
-    def convert_markdown_table(self, instr: str) -> str:
-        lines = instr.split("\n")
-        headers = lines[0].strip("|").split("|")
-        cols = len(headers)
-        buf = (
-            "\\begin{table}[h]\n\\centering\n\\begin{tabular}{|"
-            + "|".join(["l"] * cols)
-            + "|}\n\\hline\n"
-        )
-        buf += (
-            " & ".join([f"\\textbf{{{header.strip()}}}" for header in headers]) + " \\\\\n\\hline\n"
-        )
-        for line in lines[2:]:
-            cells = line.strip("|").split("|")
-            buf += " & ".join([cell.strip() for cell in cells]) + " \\\\\n\\hline\n"
-        buf += "\\end{tabular}\n\\end{table}"
-        return buf
-
-
-class Img2Latex:
-    def convert(self, instr: str) -> str:
-        dom = xml.dom.minidom.parseString(instr)
-        img = dom.documentElement
-        assert img is not None
-        src = img.getAttribute("src")
-        alt = img.getAttribute("alt")
-        return IMAGE_TEMPLATE.format(src=src, alt=alt)
-
-
-class Link2Latex:
-    def convert(self, instr: str) -> str:
-        dom = xml.dom.minidom.parseString(instr)
-        link = dom.documentElement
-        assert link is not None
-        href = link.getAttribute("href")
-        matches = re.search(r">([^<]+)", instr)
-        desc = ""
-        if matches:
-            desc = matches.group(1)
-        return r"\href{%s}{%s}" % (href, desc) if href != desc else r"\url{%s}" % href
-
-
-class ImageTextPostProcessor(markdown.postprocessors.Postprocessor):
-    def run(self, instr: str) -> str:
-        converter = Img2Latex()
-        new_blocks = []
-        for block in instr.split("\n\n"):
-            stripped = block.strip()
-            if stripped.startswith("<img"):
-                stripped = re.sub(r"<\/?plaintext[^>]*>", "", stripped, flags=re.IGNORECASE)
-                new_blocks.append(converter.convert(stripped).strip())
-            else:
-                new_blocks.append(block)
-        return "\n\n".join(new_blocks)
-
-
-class LinkTextPostProcessor(markdown.postprocessors.Postprocessor):
-    def run(self, instr: str) -> str:
-        converter = Link2Latex()
-        new_blocks = []
-        for block in instr.split("\n\n"):
-            stripped = block.strip()
-            matches = re.findall(r"<a[^>]*>[^<]+</a>", stripped)
-            if matches:
-                for match in matches:
-                    stripped = stripped.replace(match, converter.convert(match).strip())
-                new_blocks.append(stripped)
-            else:
-                new_blocks.append(block)
-        return "\n\n".join(new_blocks)
-
-
-class UnescapeHtmlTextPostProcessor(markdown.postprocessors.Postprocessor):
-    def run(self, text: str) -> str:
-        return unescape_html_entities(inline_html_latex(text))
-
-
-class MathTextPostProcessor(markdown.postprocessors.Postprocessor):
-    def run(self, instr: str) -> str:
-        instr = re.sub(r"\$\$([^\$]*)\$\$", r"\\[\1\\]", instr)
-        instr = re.sub(r"\$([^\$]*)\$", r"\\(\1\\)", instr)
-        instr = instr.replace("\\lt", "<").replace(" * ", " \\cdot ").replace("\\del", "\\partial")
-        return instr
-
-
-class TableTextPostProcessor(markdown.postprocessors.Postprocessor):
-    def run(self, instr: str) -> str:
-        converter = Table2Latex()
-        new_blocks = []
-        for block in instr.split("\n\n"):
-            stripped = block.strip()
-            if re.match(r"\|.*\|", stripped):  # Check for Markdown table
-                new_blocks.append(converter.convert_markdown_table(stripped).strip())
-            else:
-                new_blocks.append(block)
-        return "\n\n".join(new_blocks)
-
-
-def convert_md_to_latex(md_content: str) -> str:
-    md = markdown.Markdown(extensions=[LaTeXExtension()])
-    latex_content = md.convert(md_content)
-    latex_content = re.sub(r"<\/?plaintext[^>]*>", "", latex_content, flags=re.IGNORECASE)
-    return MAIN_TEMPLATE.format(latex_content=latex_content)
-
-
-def md_to_pdf(markdown_text: str, output_filename: str = "output") -> str:
-    """
-    Convert Markdown to PDF via LaTeX.
-
-    Args:
-        markdown_text: Markdown text
-        output_filename: Output filename (without extension)
-
-    Returns:
-        Message about the compilation result
-    """
-
-    latex_code = convert_md_to_latex(markdown_text)
-
-    temp_dir = get_workspace_dir() / "temp_latex"
-    temp_dir.mkdir(parents=True, exist_ok=True)
-
-    tex_file_path = temp_dir / "temp.tex"
-    with open(tex_file_path, "w", encoding="utf-8") as f:
-        f.write(latex_code)
-
-    try:
-        subprocess.run(
-            [
-                "pdflatex",
-                "-interaction=nonstopmode",
-                "-output-directory",
-                temp_dir,
-                tex_file_path,
-            ],
-            check=True,
-            stdout=subprocess.PIPE,
-            stderr=subprocess.PIPE,
-            timeout=30,
-        )
-
-    except subprocess.TimeoutExpired:
-        return "Compilation timed out after 30 seconds"
-    except subprocess.CalledProcessError as e:
-        error_msg = e.stdout.decode("utf-8")
-        error_lines = [
-            line for line in error_msg.split("\n") if "error" in line.lower() or "!" in line
-        ]
-        if error_lines:
-            return "Compilation failed. LaTeX errors:\n" + "\n".join(error_lines)
-        return f"Compilation failed. Full LaTeX output:\n{error_msg}"
-
-    pdf_path = os.path.join(temp_dir, "temp.pdf")
-    output_pdf_path = os.path.join(get_workspace_dir(), f"{output_filename}.pdf")
-
-    if os.path.exists(pdf_path):
-        shutil.move(pdf_path, output_pdf_path)
-        shutil.rmtree(temp_dir, ignore_errors=True)
-        return f"Compilation successful! PDF file saved as {output_filename}.pdf"
-
-    return "Compilation completed, but PDF file was not created. Check LaTeX code for errors."

academia_mcp-1.2.2.dist-info/RECORD

@@ -1,25 +0,0 @@
-academia_mcp/__init__.py,sha256=2Ru2I5u4cE7DrkkAsibDUEF1K6sYtqppb9VyFrRoQKI,94
-academia_mcp/__main__.py,sha256=rcmsOtJd3SA82exjrcGBuxuptcoxF8AXI7jNjiVq2BY,59
-academia_mcp/files.py,sha256=hI5dj4h0fX8V3DXKI_C8vs1fte2uc9gsBXC6prLV4o4,745
-academia_mcp/llm.py,sha256=o84FQNSbjjVSk9DlvFXWsUDiz5IOaavYU6kOqnPEG7E,1071
-academia_mcp/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-academia_mcp/server.py,sha256=q9bhacVm-8uuIMc_xSeymmVaIB8pQCqfTQx5GU8hhpM,2492
-academia_mcp/utils.py,sha256=tkHBho-NfzAR8rplFaiRYq4sBmQ9V3JPjDJTDrz58Xs,4041
-academia_mcp/tools/__init__.py,sha256=8_8QWPRYmPiUjdiNrQilEEnCRR-UBU7g-56jT52V3VQ,934
-academia_mcp/tools/anthology_search.py,sha256=_5s8EzdV7NQD_F3bjVH4XlKKHOJlFtWlQVrPbODuc3I,7847
-academia_mcp/tools/arxiv_download.py,sha256=xanzt77TZBQRngzGbKCRz4Hp-Mwfe_q-46eRW23TpVs,11219
-academia_mcp/tools/arxiv_search.py,sha256=pzM18qrF3QL03A53w003kE7hQi3s3QKtjgw0m7K88UY,8355
-academia_mcp/tools/bitflip.py,sha256=u0hSOPWbnCDu2EbA_RkueX496SvTKz9QhZcXugshSfI,10949
-academia_mcp/tools/document_qa.py,sha256=lWtzRNFKwQpQO8vPNvYYbNnGh3LsBmGl6f9vSRIuqaw,2245
-academia_mcp/tools/hf_datasets_search.py,sha256=KiBkqT4rXjEN4oc1AWZOPnqN_Go90TQogY5-DUm3LQo,2854
-academia_mcp/tools/md_to_pdf.py,sha256=Ovc_-8j7gIZNEM1d0ZDH-8qbtfZLSaNmCm5DQjrtM0k,12810
-academia_mcp/tools/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-academia_mcp/tools/s2_citations.py,sha256=dqrBp76RrX1zH2XzcMAoWBbvbtyhxLeF-xnqOKD_JiM,4852
-academia_mcp/tools/visit_webpage.py,sha256=0zAZYeQxPDu0OjgAAvbMLZh0ttaS5q-_4WhgsEPrbsI,1542
-academia_mcp/tools/web_search.py,sha256=P74HF5agF-cWOGUYmrDcr_R-Ay3xz9fNdhFov8vtt9o,6025
-academia_mcp-1.2.2.dist-info/licenses/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
-academia_mcp-1.2.2.dist-info/METADATA,sha256=SC1JQUbH4sbFFZFcuTeC-iwEVIhMbQqXlSL2sB6riqM,1899
-academia_mcp-1.2.2.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-academia_mcp-1.2.2.dist-info/entry_points.txt,sha256=gxkiKJ74w2FwJpSECpjA3XtCfI5ZfrM6N8cqnwsq4yY,51
-academia_mcp-1.2.2.dist-info/top_level.txt,sha256=CzGpRFsRRJRqWEb1e3SUlcfGqRzOxevZGaJWrtGF8W0,13
-academia_mcp-1.2.2.dist-info/RECORD,,