gitputra 0.1.0__tar.gz

gitputra-0.1.0/PKG-INFO

@@ -0,0 +1,18 @@
+Metadata-Version: 2.4
+Name: gitputra
+Version: 0.1.0
+Summary: AI-powered GitHub repo analyzer CLI — analyze, chat, and visualize any codebase.
+Author-email: Adityava Gangopadhyay <adityava49cse@gmail.com>
+License: MIT
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+Requires-Dist: click>=8.1
+Requires-Dist: python-dotenv>=1.0
+Requires-Dist: gitpython>=3.1
+Requires-Dist: chromadb>=0.5
+Requires-Dist: matplotlib>=3.8
+Requires-Dist: networkx>=3.2
+Requires-Dist: reportlab>=4.0
+Requires-Dist: google-generativeai>=0.5
+Requires-Dist: openai>=1.0
+Requires-Dist: anthropic>=0.25

gitputra-0.1.0/gitputra/cli.py

@@ -0,0 +1,132 @@
+"""
+cli.py — Gitputra CLI entry point.
+Usage:
+    gitputra analyze <url> --ai gemini --key sk-xxx
+    gitputra chat          --ai gemini --key sk-xxx
+"""
+
+import click
+from dotenv import load_dotenv
+from . import corelogic as core
+
+load_dotenv()
+
+
+@click.group()
+@click.version_option("0.1.0", prog_name="gitputra")
+def main():
+    """🔍 Gitputra — AI-powered GitHub repo analyzer."""
+    pass
+
+
+# ─────────────────────────────────────────────
+# ANALYZE
+# ─────────────────────────────────────────────
+@main.command()
+@click.argument("url")
+@click.option(
+    "--ai", "ai_choice",
+    type=click.Choice(["gemini", "openai", "claude"], case_sensitive=False),
+    default="gemini", show_default=True,
+    help="AI provider to use."
+)
+@click.option(
+    "--key",
+    envvar="API_KEY", required=True,
+    help="API key (or set API_KEY in .env)."
+)
+@click.option(
+    "--lang", default="English", show_default=True,
+    help="Output language for the report."
+)
+@click.option(
+    "--no-pdf", is_flag=True, default=False,
+    help="Skip PDF generation."
+)
+@click.option(
+    "--no-diagram", is_flag=True, default=False,
+    help="Skip diagram generation."
+)
+def analyze(url, ai_choice, key, lang, no_pdf, no_diagram):
+    """Clone a GitHub repo and generate an AI analysis report.
+
+    \b
+    Example:
+        gitputra analyze https://github.com/user/repo --ai gemini --key AIza...
+    """
+    core.init(ai_choice.lower(), key, lang)
+
+    repo_path = core.clone_repo(url)
+    files     = core.load_files(repo_path)
+
+    if not files:
+        click.secho("❌ No supported source files found.", fg="red")
+        raise SystemExit(1)
+
+    core.store_chunks(files)
+    analysis = core.analyze_repo(files)
+
+    click.echo("\n" + analysis + "\n")
+
+    if not no_diagram:
+        core.generate_diagram(files)
+        core.generate_mermaid(files)
+
+    if not no_pdf:
+        core.generate_pdf(analysis)
+
+    click.secho("✅ Done! Outputs saved in ./output/", fg="green")
+
+
+# ─────────────────────────────────────────────
+# CHAT
+# ─────────────────────────────────────────────
+@main.command()
+@click.option(
+    "--ai", "ai_choice",
+    type=click.Choice(["gemini", "openai", "claude"], case_sensitive=False),
+    default="gemini", show_default=True,
+)
+@click.option(
+    "--key",
+    envvar="API_KEY", required=True,
+    help="API key (or set API_KEY in .env)."
+)
+@click.option("--lang", default="English", show_default=True)
+def chat(ai_choice, key, lang):
+    """Start an interactive RAG chat about the last analyzed repo.
+
+    \b
+    Example:
+        gitputra chat --ai gemini --key AIza...
+        Type 'exit' or Ctrl+C to quit.
+    """
+    core.init(ai_choice.lower(), key, lang)
+
+    click.secho("\n💬 Chat mode — ask anything about the repo. Type 'exit' to quit.\n", fg="cyan")
+
+    while True:
+        try:
+            question = click.prompt(">>", prompt_suffix=" ")
+        except (KeyboardInterrupt, EOFError):
+            click.echo("\nBye!")
+            break
+
+        if question.strip().lower() in ("exit", "quit", "q"):
+            click.echo("Bye!")
+            break
+
+        answer = core.query_rag(question)
+        click.echo(f"\n{answer}\n")
+
+
+# ─────────────────────────────────────────────
+# CLEAR DB
+# ─────────────────────────────────────────────
+@main.command("clear-db")
+def clear_db():
+    """Wipe the local ChromaDB collection."""
+    import chromadb
+    client = chromadb.PersistentClient(path="./chroma_db")
+    client.delete_collection("repo")
+    click.secho("🗑️  ChromaDB collection cleared.", fg="yellow")

gitputra-0.1.0/gitputra/corelogic.py

@@ -0,0 +1,409 @@
+"""
+corelogic.py — All core logic for RepoLyzer.
+AI setup, repo cloning, file loading, chunking,
+embedding, ChromaDB, analysis, diagrams, PDF.
+"""
+
+import os
+import uuid
+import time
+import importlib
+
+from git import Repo
+import chromadb
+import matplotlib
+matplotlib.use("Agg")
+import matplotlib.pyplot as plt
+import networkx as nx
+from reportlab.platypus import SimpleDocTemplate, Paragraph, Spacer, Image
+from reportlab.lib.styles import ParagraphStyle
+from reportlab.lib import colors
+from reportlab.pdfbase import pdfmetrics
+from reportlab.pdfbase.ttfonts import TTFont
+
+# ─────────────────────────────────────────────
+# GLOBALS
+# ─────────────────────────────────────────────
+_ai_config  = {}
+_ai_lib     = None
+_claude_client = None
+_language   = "English"
+_ai_disabled = False
+
+CHROMA_CLIENT = chromadb.PersistentClient(path="./chroma_db")
+COLLECTION    = CHROMA_CLIENT.get_or_create_collection("repo")
+
+SUPPORTED_EXT = (".py", ".js", ".ts", ".java", ".cpp", ".c", ".go", ".rs")
+
+AI_OPTIONS = {
+    "gemini": {
+        "name": "Gemini",
+        "lib": "google.generativeai",
+        "text_model": "models/gemini-2.5-flash",
+        "embed_model": "models/text-embedding-004",
+    },
+    "openai": {
+        "name": "OpenAI",
+        "lib": "openai",
+        "text_model": "gpt-4",
+        "embed_model": "text-embedding-3-large",
+    },
+    "claude": {
+        "name": "Claude",
+        "lib": "anthropic",
+        "text_model": "claude-opus-4-5",
+        "embed_model": None,
+    },
+}
+
+
+# ─────────────────────────────────────────────
+# INIT
+# ─────────────────────────────────────────────
+def init(ai_choice: str, api_key: str, language: str = "English"):
+    """Bootstrap AI client and global settings. Call before anything else."""
+    global _ai_config, _ai_lib, _claude_client, _language
+
+    _language  = language
+    _ai_config = AI_OPTIONS[ai_choice]
+    _ai_lib    = importlib.import_module(_ai_config["lib"])
+
+    if ai_choice == "gemini":
+        _ai_lib.configure(api_key=api_key)
+
+    elif ai_choice == "openai":
+        _ai_lib.api_key = api_key
+
+    elif ai_choice == "claude":
+        _claude_client = _ai_lib.Anthropic(api_key=api_key)
+
+
+# ─────────────────────────────────────────────
+# AI — GENERATE
+# ─────────────────────────────────────────────
+def safe_generate(prompt: str, retries: int = 2) -> str | None:
+    global _ai_disabled
+
+    if _ai_disabled:
+        return None
+
+    for _ in range(retries):
+        try:
+            name = _ai_config["name"]
+
+            if name == "Gemini":
+                model = _ai_lib.GenerativeModel(_ai_config["text_model"])
+                res   = model.generate_content(prompt)
+                return res.text if res and hasattr(res, "text") else None
+
+            elif name == "OpenAI":
+                res = _ai_lib.ChatCompletion.create(
+                    model=_ai_config["text_model"],
+                    messages=[{"role": "user", "content": prompt}],
+                )
+                return res.choices[0].message.content
+
+            elif name == "Claude":
+                res = _claude_client.messages.create(
+                    model=_ai_config["text_model"],
+                    max_tokens=2048,
+                    messages=[{"role": "user", "content": prompt}],
+                )
+                return res.content[0].text
+
+        except Exception as e:
+            if "429" in str(e) or "quota" in str(e).lower():
+                print("⚠️  Quota exceeded → AI disabled for this session")
+                _ai_disabled = True
+                return None
+            time.sleep(2)
+
+    return None
+
+
+# ─────────────────────────────────────────────
+# AI — EMBED
+# ─────────────────────────────────────────────
+def embed_text(text: str) -> list[float]:
+    try:
+        name = _ai_config["name"]
+
+        if name == "Gemini":
+            return _ai_lib.embed_content(
+                model=_ai_config["embed_model"],
+                content=text,
+            )["embedding"]
+
+        elif name == "OpenAI":
+            return _ai_lib.Embedding.create(
+                model=_ai_config["embed_model"],
+                input=text,
+            )["data"][0]["embedding"]
+
+        # Claude has no embedding API — use zero vector as fallback
+        return [0.0] * 768
+
+    except Exception:
+        return [0.0] * 768
+
+
+# ─────────────────────────────────────────────
+# REPO — CLONE & LOAD
+# ─────────────────────────────────────────────
+def clone_repo(url: str) -> str:
+    os.makedirs("repos", exist_ok=True)
+    name = url.rstrip("/").split("/")[-1].replace(".git", "")
+    path = os.path.join("repos", name)
+
+    if not os.path.exists(path):
+        print(f"📥 Cloning {url} ...")
+        Repo.clone_from(url, path)
+    else:
+        print(f"✅ Repo already exists at {path}")
+
+    return path
+
+
+def load_files(repo_path: str) -> list[tuple[str, str]]:
+    files_data = []
+
+    for root, _, files in os.walk(repo_path):
+        # skip hidden dirs like .git
+        if any(part.startswith(".") for part in root.split(os.sep)):
+            continue
+        for file in files:
+            if file.endswith(SUPPORTED_EXT):
+                try:
+                    with open(os.path.join(root, file), "r", errors="ignore") as f:
+                        content = f.read().strip()
+                        if content:
+                            files_data.append((file, content))
+                except Exception:
+                    pass
+
+    print(f"📂 Loaded {len(files_data)} source files")
+    return files_data
+
+
+# ─────────────────────────────────────────────
+# CHUNKING
+# ─────────────────────────────────────────────
+def chunk_by_file(files: list[tuple[str, str]], max_chars: int = 3000) -> list[str]:
+    chunks, current = [], ""
+
+    for fname, content in files:
+        piece = f"\n# FILE: {fname}\n{content}\n"
+        if len(current) + len(piece) > max_chars:
+            if current:
+                chunks.append(current)
+            current = piece
+        else:
+            current += piece
+
+    if current:
+        chunks.append(current)
+
+    return chunks
+
+
+# ─────────────────────────────────────────────
+# CHROMADB — STORE & QUERY
+# ─────────────────────────────────────────────
+def store_chunks(files: list[tuple[str, str]]):
+    print("💾 Storing embeddings...")
+    for fname, content in files:
+        emb = embed_text(content)
+        COLLECTION.add(
+            documents=[content],
+            embeddings=[emb],
+            ids=[str(uuid.uuid4())],
+            metadatas=[{"file": fname}],
+        )
+
+
+def query_rag(question: str) -> str:
+    emb = embed_text(question)
+    res = COLLECTION.query(query_embeddings=[emb], n_results=5)
+    context = "\n".join(res["documents"][0])
+
+    prompt = f"""Answer in {_language}.
+
+Context from codebase:
+{context}
+
+Question:
+{question}
+"""
+    return safe_generate(prompt) or "⚠️ AI unavailable."
+
+
+# ─────────────────────────────────────────────
+# ANALYSIS
+# ─────────────────────────────────────────────
+def analyze_repo(files: list[tuple[str, str]]) -> str:
+    print("\n🔍 Analyzing repo...\n")
+    chunks    = chunk_by_file(files)
+    summaries = []
+
+    for i, chunk in enumerate(chunks, 1):
+        print(f"  Chunk {i}/{len(chunks)}")
+        res = safe_generate(f"""Summarize this code chunk in {_language}.
+Focus on logic, structure, and key parts.
+
+Code:
+{chunk}
+""")
+        if res:
+            summaries.append(res)
+
+    if not summaries:
+        return "⚠️ AI unavailable — no analysis generated."
+
+    return safe_generate(f"""Respond ONLY in {_language}.
+
+Using the summaries below, write ONE structured report with these sections:
+
+### Summary
+### Architecture
+### Tech Stack
+### Issues
+### Suggested Improvements
+
+No repetition. Be concise.
+
+Summaries:
+{chr(10).join(summaries)}
+""") or "⚠️ Final synthesis failed."
+
+
+# ─────────────────────────────────────────────
+# DIAGRAMS
+# ─────────────────────────────────────────────
+def _clean(name: str) -> str:
+    return (
+        name.replace(".py", "_py")
+            .replace(".", "_")
+            .replace("-", "_")
+            .replace("/", "_")
+    )
+
+
+def generate_diagram(files: list[tuple[str, str]]):
+    G = nx.DiGraph()
+
+    for fname, content in files:
+        for line in content.splitlines():
+            line = line.strip()
+            if line.startswith(("import ", "from ", "#include")):
+                G.add_edge(fname, line)
+
+    os.makedirs("output", exist_ok=True)
+    plt.figure(figsize=(12, 9))
+    nx.draw(G, with_labels=True, node_size=800, font_size=7, arrows=True)
+    plt.tight_layout()
+    plt.savefig("output/diagram.png", dpi=150)
+    plt.close()
+    print("📊 Saved output/diagram.png")
+
+
+def generate_mermaid(files: list[tuple[str, str]]):
+    lines, seen = ["graph TD"], set()
+
+    for fname, content in files:
+        file_node = _clean(fname)
+
+        for line in content.splitlines():
+            line = line.strip()
+            module = None
+
+            if line.startswith("import "):
+                parts = line.replace(",", " ").split()
+                if len(parts) >= 2:
+                    module = _clean(parts[1])
+
+            elif line.startswith("from "):
+                parts = line.split()
+                if len(parts) >= 2:
+                    module = _clean(parts[1])
+
+            elif line.startswith("#include"):
+                module = _clean(line.replace("#include", "").strip())
+
+            if module:
+                edge = f"{file_node} --> {module}"
+                if edge not in seen:
+                    lines.append(f"    {edge}")
+                    seen.add(edge)
+
+    os.makedirs("output", exist_ok=True)
+    with open("output/mermaid.txt", "w", encoding="utf-8") as f:
+        f.write("\n".join(lines))
+    print("🗺️  Saved output/mermaid.txt")
+
+
+# ─────────────────────────────────────────────
+# PDF
+# ─────────────────────────────────────────────
+_BASE_DIR = os.path.dirname(os.path.abspath(__file__))
+
+def _register_fonts():
+    font_dir = os.path.join(_BASE_DIR, "fonts")
+    mappings = {
+        "NotoLatin":      "NotoSans-Regular.ttf",
+        "NotoBengali":    "NotoSansBengali-Regular.ttf",
+        "NotoDevanagari": "NotoSansDevanagari-Regular.ttf",
+    }
+    for font_name, filename in mappings.items():
+        path = os.path.join(font_dir, filename)
+        if os.path.exists(path):
+            pdfmetrics.registerFont(TTFont(font_name, path))
+
+
+def _get_font(language: str) -> str:
+    lang = language.lower()
+    if lang in ("bengali", "bangla"):
+        font = "NotoBengali"
+    elif lang == "hindi":
+        font = "NotoDevanagari"
+    else:
+        font = "NotoLatin"
+
+    try:
+        pdfmetrics.getFont(font)
+        return font
+    except KeyError:
+        return "Helvetica"
+
+
+def generate_pdf(text: str):
+    os.makedirs("output", exist_ok=True)
+    _register_fonts()
+    font = _get_font(_language)
+
+    styles = {
+        "title": ParagraphStyle("title", fontName=font, fontSize=20, textColor=colors.darkblue),
+        "head":  ParagraphStyle("head",  fontName=font, fontSize=14, textColor=colors.blue),
+        "body":  ParagraphStyle("body",  fontName=font, fontSize=11),
+    }
+
+    story = [
+        Paragraph(f"Repo Analysis Report ({_language})", styles["title"]),
+        Spacer(1, 12),
+    ]
+
+    for section in text.split("###"):
+        lines = section.strip().splitlines()
+        if not lines or not lines[0].strip():
+            continue
+        story.append(Paragraph(lines[0].strip(), styles["head"]))
+        body_text = "<br/>".join(l for l in lines[1:] if l.strip())
+        if body_text:
+            story.append(Paragraph(body_text, styles["body"]))
+        story.append(Spacer(1, 10))
+
+    diagram_path = "output/diagram.png"
+    if os.path.exists(diagram_path):
+        story.append(Image(diagram_path, width=420, height=300))
+
+    SimpleDocTemplate("output/report.pdf").build(story)
+    print("📄 Saved output/report.pdf")

gitputra-0.1.0/gitputra.egg-info/PKG-INFO

@@ -0,0 +1,18 @@
+Metadata-Version: 2.4
+Name: gitputra
+Version: 0.1.0
+Summary: AI-powered GitHub repo analyzer CLI — analyze, chat, and visualize any codebase.
+Author-email: Adityava Gangopadhyay <adityava49cse@gmail.com>
+License: MIT
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+Requires-Dist: click>=8.1
+Requires-Dist: python-dotenv>=1.0
+Requires-Dist: gitpython>=3.1
+Requires-Dist: chromadb>=0.5
+Requires-Dist: matplotlib>=3.8
+Requires-Dist: networkx>=3.2
+Requires-Dist: reportlab>=4.0
+Requires-Dist: google-generativeai>=0.5
+Requires-Dist: openai>=1.0
+Requires-Dist: anthropic>=0.25

gitputra-0.1.0/gitputra.egg-info/SOURCES.txt

@@ -0,0 +1,12 @@
+pyproject.toml
+gitputra/cli.py
+gitputra/corelogic.py
+gitputra.egg-info/PKG-INFO
+gitputra.egg-info/SOURCES.txt
+gitputra.egg-info/dependency_links.txt
+gitputra.egg-info/entry_points.txt
+gitputra.egg-info/requires.txt
+gitputra.egg-info/top_level.txt
+gitputra/fonts/NotoSans-Regular.ttf
+gitputra/fonts/NotoSansBengali-Regular.ttf
+gitputra/fonts/NotoSansDevanagari-Regular.ttf

gitputra-0.1.0/gitputra.egg-info/dependency_links.txt

@@ -0,0 +1 @@
+

gitputra-0.1.0/gitputra.egg-info/entry_points.txt

@@ -0,0 +1,2 @@
+[console_scripts]
+gitputra = gitputra.cli:main

gitputra-0.1.0/gitputra.egg-info/requires.txt

@@ -0,0 +1,10 @@
+click>=8.1
+python-dotenv>=1.0
+gitpython>=3.1
+chromadb>=0.5
+matplotlib>=3.8
+networkx>=3.2
+reportlab>=4.0
+google-generativeai>=0.5
+openai>=1.0
+anthropic>=0.25

gitputra-0.1.0/gitputra.egg-info/top_level.txt

@@ -0,0 +1 @@
+gitputra

gitputra-0.1.0/pyproject.toml

@@ -0,0 +1,37 @@
+[build-system]
+requires = ["setuptools>=61"]
+build-backend = "setuptools.build_meta"
+
+[project]
+name = "gitputra"
+version = "0.1.0"
+description = "AI-powered GitHub repo analyzer CLI — analyze, chat, and visualize any codebase."
+readme = "README.md"
+license = { text = "MIT" }
+requires-python = ">=3.10"
+authors = [
+    { name = "Adityava Gangopadhyay", email = "adityava49cse@gmail.com" }
+]
+
+dependencies = [
+    "click>=8.1",
+    "python-dotenv>=1.0",
+    "gitpython>=3.1",
+    "chromadb>=0.5",
+    "matplotlib>=3.8",
+    "networkx>=3.2",
+    "reportlab>=4.0",
+    "google-generativeai>=0.5",
+    "openai>=1.0",
+    "anthropic>=0.25",
+]
+
+[project.scripts]
+gitputra = "gitputra.cli:main"
+
+[tool.setuptools.packages.find]
+where = ["."]
+include = ["gitputra*"]
+
+[tool.setuptools.package-data]
+gitputra = ["fonts/*.ttf"]

gitputra-0.1.0/setup.cfg

@@ -0,0 +1,4 @@
+[egg_info]
+tag_build = 
+tag_date = 0
+