suchi 0.1.1__tar.gz

suchi-0.1.1/.gitignore

@@ -0,0 +1,48 @@
+# Python
+__pycache__/
+*.py[cod]
+*.egg-info/
+dist/
+build/
+*.egg
+.venv/
+venv/
+
+# Node
+node_modules/
+frontend/dist/
+
+# IDE
+.vscode/
+.idea/
+*.swp
+*.swo
+*~
+
+# OS
+.DS_Store
+Thumbs.db
+
+# Suchi runtime
+.tantivy-index/
+.references-cache.json
+.pageindex.json
+.collection-index.json
+
+# Config (contains API keys)
+config.yaml
+gdrive-token.json
+.env
+
+# Test
+.pytest_cache/
+.coverage
+htmlcov/
+
+# Rust / Tauri
+src-tauri/target/
+src-tauri/gen/
+
+# Build
+*.whl
+*.tar.gz

suchi-0.1.1/PKG-INFO

@@ -0,0 +1,23 @@
+Metadata-Version: 2.4
+Name: suchi
+Version: 0.1.1
+Summary: सूची — CLI-first reference manager with AI-powered research tools
+Requires-Python: >=3.11
+Requires-Dist: citeproc-py>=0.6
+Requires-Dist: fastapi>=0.110
+Requires-Dist: httpx>=0.27
+Requires-Dist: pymupdf>=1.24
+Requires-Dist: python-multipart>=0.0.9
+Requires-Dist: pyyaml>=6.0
+Requires-Dist: rapidfuzz>=3.0
+Requires-Dist: rich>=13.0
+Requires-Dist: tantivy>=0.22
+Requires-Dist: typer>=0.9
+Requires-Dist: uvicorn>=0.27
+Provides-Extra: dev
+Requires-Dist: pytest-asyncio>=0.23; extra == 'dev'
+Requires-Dist: pytest>=8.0; extra == 'dev'
+Requires-Dist: ruff>=0.3; extra == 'dev'
+Provides-Extra: gdrive
+Requires-Dist: google-api-python-client>=2.0; extra == 'gdrive'
+Requires-Dist: google-auth-oauthlib>=1.0; extra == 'gdrive'

suchi-0.1.1/pyproject.toml

@@ -0,0 +1,47 @@
+[build-system]
+requires = ["hatchling"]
+build-backend = "hatchling.build"
+
+[project]
+name = "suchi"
+version = "0.1.1"
+description = "सूची — CLI-first reference manager with AI-powered research tools"
+requires-python = ">=3.11"
+dependencies = [
+    "typer>=0.9",
+    "fastapi>=0.110",
+    "uvicorn>=0.27",
+    "pyyaml>=6.0",
+    "rich>=13.0",
+    "httpx>=0.27",
+    "pymupdf>=1.24",
+    "python-multipart>=0.0.9",
+    "tantivy>=0.22",
+    "rapidfuzz>=3.0",
+    "citeproc-py>=0.6",
+]
+
+[project.optional-dependencies]
+gdrive = [
+    "google-api-python-client>=2.0",
+    "google-auth-oauthlib>=1.0",
+]
+dev = [
+    "pytest>=8.0",
+    "pytest-asyncio>=0.23",
+    "ruff>=0.3",
+]
+
+[project.scripts]
+suchi = "suchi.cli:app"
+
+[tool.hatch.build.targets.wheel]
+packages = ["src/suchi"]
+
+[tool.ruff]
+target-version = "py311"
+line-length = 100
+
+[tool.ruff.lint]
+# Allow compact one-liners (if x: y) and semicolons — they're intentional for conciseness
+ignore = ["E701", "E702", "E402"]

suchi-0.1.1/src/suchi/__init__.py

@@ -0,0 +1,3 @@
+"""Suchi (सूची): CLI-first reference manager with AI-powered research tools."""
+
+__version__ = "0.1.1"

suchi-0.1.1/src/suchi/api.py

@@ -0,0 +1,56 @@
+"""FastAPI application."""
+
+from contextlib import asynccontextmanager
+
+from fastapi import FastAPI
+from fastapi.middleware.cors import CORSMiddleware
+
+from .routes import entries, search, export, collections, settings, chat, citations, references, pdf_finder, annotations, discovery, sync
+from . import library as lib_module
+from .search import index_entry, remove_from_index, rebuild_index
+
+
+@asynccontextmanager
+async def lifespan(app: FastAPI):
+    # Register search index hooks so entries are indexed on create/update/delete
+    lib_module.set_index_hooks(
+        on_added=index_entry,
+        on_removed=remove_from_index,
+    )
+    # Build index on startup if needed
+    rebuild_index()
+    yield
+
+
+app = FastAPI(
+    title="Suchi",
+    description="सूची — CLI-first reference manager with AI-powered research tools",
+    version="0.1.1",
+    lifespan=lifespan,
+)
+
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=["*"],  # Tauri webview + browser extension
+    allow_credentials=True,
+    allow_methods=["*"],
+    allow_headers=["*"],
+)
+
+app.include_router(entries.router)
+app.include_router(search.router)
+app.include_router(export.router)
+app.include_router(collections.router)
+app.include_router(settings.router)
+app.include_router(chat.router)
+app.include_router(citations.router)
+app.include_router(references.router)
+app.include_router(pdf_finder.router)
+app.include_router(annotations.router)
+app.include_router(discovery.router)
+app.include_router(sync.router)
+
+
+@app.get("/api/health")
+def health():
+    return {"status": "ok", "version": "0.1.1"}

suchi-0.1.1/src/suchi/citations/processor.py

@@ -0,0 +1,163 @@
+"""Citation formatter using citeproc-py with CSL styles.
+
+Supports 10,000+ CSL citation styles. Ships with the most common ones;
+users can add more .csl files to the styles directory.
+
+Usage:
+    from suchi.citations.processor import format_citation, format_bibliography, list_styles
+
+    # Format a single entry
+    citation = format_citation(entry, style="apa")
+
+    # Format a bibliography from multiple entries
+    bib = format_bibliography(entries, style="chicago-author-date")
+
+    # List available styles
+    styles = list_styles()
+"""
+
+from pathlib import Path
+
+from citeproc import CitationStylesStyle, CitationStylesBibliography
+from citeproc import Citation, CitationItem
+from citeproc.source.json import CiteProcJSON
+
+
+STYLES_DIR = Path(__file__).parent / "styles"
+
+
+def list_styles() -> list[dict]:
+    """List available citation styles."""
+    styles = []
+    for f in sorted(STYLES_DIR.glob("*.csl")):
+        styles.append({
+            "id": f.stem,
+            "name": f.stem.replace("-", " ").title(),
+            "file": f.name,
+        })
+    return styles
+
+
+def _entry_to_csl(entry: dict) -> dict:
+    """Convert a Suchi entry dict to CSL-JSON format for citeproc."""
+    csl = {
+        "id": entry.get("id", "unknown"),
+        "type": _map_to_csl_type(entry.get("type", "article")),
+        "title": entry.get("title", ""),
+    }
+
+    # Authors
+    authors = entry.get("author", [])
+    if authors:
+        csl["author"] = [
+            {"family": a.get("family", ""), "given": a.get("given", "")}
+            for a in authors
+        ]
+
+    # Date
+    date = entry.get("date", "")
+    if date:
+        parts = date.split("-")
+        date_parts = []
+        for p in parts:
+            try:
+                date_parts.append(int(p))
+            except ValueError:
+                break
+        if date_parts:
+            csl["issued"] = {"date-parts": [date_parts]}
+
+    # Other fields
+    field_map = {
+        "doi": "DOI",
+        "isbn": "ISBN",
+        "url": "URL",
+        "abstract": "abstract",
+        "volume": "volume",
+        "issue": "issue",
+        "pages": "page",
+        "publisher": "publisher",
+    }
+    for src, dst in field_map.items():
+        val = entry.get(src)
+        if val:
+            csl[dst] = str(val)
+
+    if entry.get("journal"):
+        csl["container-title"] = entry["journal"]
+
+    return csl
+
+
+def format_citation(entry: dict, style: str = "apa") -> str:
+    """Format a single entry as an inline citation (e.g., '(Smith, 2024)')."""
+    csl_data = [_entry_to_csl(entry)]
+    source = CiteProcJSON(csl_data)
+
+    style_path = STYLES_DIR / f"{style}.csl"
+    if not style_path.exists():
+        raise ValueError(f"Style not found: {style}. Available: {[s['id'] for s in list_styles()]}")
+
+    bib_style = CitationStylesStyle(str(style_path), validate=False)
+    bibliography = CitationStylesBibliography(bib_style, source)
+
+    citation = Citation([CitationItem(entry.get("id", "unknown"))])
+    bibliography.register(citation)
+
+    # Get inline citation
+    result = bibliography.cite(citation, lambda _: None)
+    return str(result)
+
+
+def format_bibliography(entries: list[dict], style: str = "apa") -> str:
+    """Format multiple entries as a formatted bibliography."""
+    if not entries:
+        return ""
+
+    csl_data = [_entry_to_csl(e) for e in entries]
+    source = CiteProcJSON(csl_data)
+
+    style_path = STYLES_DIR / f"{style}.csl"
+    if not style_path.exists():
+        raise ValueError(f"Style not found: {style}. Available: {[s['id'] for s in list_styles()]}")
+
+    bib_style = CitationStylesStyle(str(style_path), validate=False)
+    bibliography = CitationStylesBibliography(bib_style, source)
+
+    # Register all citations
+    for entry in entries:
+        citation = Citation([CitationItem(entry.get("id", "unknown"))])
+        bibliography.register(citation)
+
+    # Render bibliography
+    bib_items = bibliography.bibliography()
+    if not bib_items:
+        return ""
+
+    lines = []
+    for item in bib_items:
+        text = str(item).strip()
+        if text:
+            lines.append(text)
+
+    return "\n\n".join(lines)
+
+
+def format_entry_full(entry: dict, style: str = "apa") -> dict:
+    """Format both inline citation and bibliography entry for a single entry."""
+    return {
+        "citation": format_citation(entry, style),
+        "bibliography": format_bibliography([entry], style),
+    }
+
+
+def _map_to_csl_type(entry_type: str) -> str:
+    return {
+        "article": "article-journal",
+        "book": "book",
+        "inbook": "chapter",
+        "inproceedings": "paper-conference",
+        "thesis": "thesis",
+        "report": "report",
+        "dataset": "dataset",
+    }.get(entry_type, "article")