gitlytics 0.1.6__tar.gz → 0.2.0__tar.gz

{gitlytics-0.1.6 → gitlytics-0.2.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: gitlytics
-Version: 0.1.6
+Version: 0.2.0
 Summary: Monitor and automate your GitHub repository traffic analytics.
 Author-email: Ameya Chopade <ameyaccod171@gmail.com>
 License: Apache-2.0
@@ -96,7 +96,7 @@ Dynamic: license-file
 
 The full Gitlytics ecosystem spans across a few repositories. If you are looking for the live web dashboard or the automation cron job, check out the links below:
 
-- **[Gitlytics Web Ecosystem](https://github.com/ameyac11/gitlytics-deployement)**: The production landing page, React Dashboard, and React Documentation site.
+- 📊 **[Gitlytics Live Dashboard](https://dashboard.gitlytics.dev)**: The production web interface to visualize repository traffic analytics, trends, and historical charts.
 - ⚙️ **[Gitlytics Automation](https://github.com/ameyac11/gitlytics-github-traffic-automation)**: The GitHub Action companion tool that automates fetching and saving to defeat GitHub's 14-day traffic limit.
 
 ---

{gitlytics-0.1.6 → gitlytics-0.2.0}/README.md

@@ -64,7 +64,7 @@
 
 The full Gitlytics ecosystem spans across a few repositories. If you are looking for the live web dashboard or the automation cron job, check out the links below:
 
-- **[Gitlytics Web Ecosystem](https://github.com/ameyac11/gitlytics-deployement)**: The production landing page, React Dashboard, and React Documentation site.
+- 📊 **[Gitlytics Live Dashboard](https://dashboard.gitlytics.dev)**: The production web interface to visualize repository traffic analytics, trends, and historical charts.
 - ⚙️ **[Gitlytics Automation](https://github.com/ameyac11/gitlytics-github-traffic-automation)**: The GitHub Action companion tool that automates fetching and saving to defeat GitHub's 14-day traffic limit.
 
 ---

{gitlytics-0.1.6 → gitlytics-0.2.0}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "gitlytics"
-version = "0.1.6"
+version = "0.2.0"
 description = "Monitor and automate your GitHub repository traffic analytics."
 readme = "README.md"
 requires-python = ">=3.9"

{gitlytics-0.1.6 → gitlytics-0.2.0}/src/gitlytics/__init__.py

@@ -8,7 +8,7 @@ import json
 
 # Single source of truth for the package version.
 # Mirrors the version in pyproject.toml — keep them in sync.
-__version__ = "0.1.6"
+__version__ = "0.2.0"
 
 __all__ = ["fetch_traffic", "sync", "serve_dashboard", "__version__"]
 
@@ -98,6 +98,11 @@ def sync(token: str, repo_name=None, data_dir: str = "./data", output_mode: str
             exported JSON — acts as a security firewall.
         metrics: Optional list of metrics to fetch (e.g., ``["views", "clones"]``).
     """
+    if data_dir and not os.path.isabs(data_dir) and not os.path.exists(data_dir):
+        parent_dir = os.path.join("..", data_dir)
+        if os.path.exists(parent_dir):
+            data_dir = parent_dir
+
     # Hand off to the automation engine — it handles deduplication and schema migration
     run_sync(
         token=token,
@@ -142,7 +147,17 @@ def serve_dashboard(host: str = "127.0.0.1", port: int = 8000, token: str = None
         if token:
             os.environ["GITLYTICS_TOKEN"] = token
         if data_dir:
-            os.environ["GITLYTICS_DATA_DIR"] = os.path.abspath(data_dir)
+            from pathlib import Path
+            abs_data_dir = os.path.abspath(data_dir)
+            if not os.path.exists(abs_data_dir) and not os.path.isabs(data_dir):
+                parent_dir = os.path.abspath(os.path.join("..", data_dir))
+                if os.path.exists(parent_dir):
+                    abs_data_dir = parent_dir
+            if not os.path.exists(abs_data_dir):
+                print(f"⚠️ Warning: The specified data directory '{data_dir}' (resolved to '{abs_data_dir}') does not exist.")
+            elif not any(Path(abs_data_dir).glob("traffic_*.csv")):
+                print(f"⚠️ Warning: No traffic_*.csv database files found in '{data_dir}' (resolved to '{abs_data_dir}').")
+            os.environ["GITLYTICS_DATA_DIR"] = abs_data_dir
         uvicorn.run("gitlytics.api:app", host=host, port=port, reload=False)
     finally:
         if _orig_token is None:

{gitlytics-0.1.6 → gitlytics-0.2.0}/src/gitlytics/api.py

@@ -135,22 +135,38 @@ def get_traffic(token: str = Body("", embed=True)):
     if not ok:
         raise HTTPException(status_code=401, detail="Invalid token")
     data_dir = os.environ.get("GITLYTICS_DATA_DIR")
+    dfs = []
     if data_dir:
         data_dir_path = Path(data_dir)
-        csv_files = list(data_dir_path.glob("traffic_*.csv")) if data_dir_path.exists() else []
-        dfs = []
-        for f in csv_files:
-            try:
-                dfs.append(pd.read_csv(f))
-            except Exception as exc:
-                logger.warning(f"Skipping unreadable CSV '{f}': {exc}")
-        if dfs:
-            df = pd.concat(dfs, ignore_index=True)
-            df = df.drop_duplicates(subset=["date", "repository"], keep="last")
+        if not data_dir_path.exists():
+            logger.warning(f"Data directory '{data_dir}' does not exist.")
         else:
-            df = fetch_traffic_data(active_token)
+            csv_files = list(data_dir_path.glob("traffic_*.csv"))
+            if not csv_files:
+                logger.warning(f"No traffic_*.csv files found in '{data_dir}'.")
+            for f in csv_files:
+                try:
+                    dfs.append(pd.read_csv(f))
+                except Exception as exc:
+                    logger.warning(f"Skipping unreadable CSV '{f}': {exc}")
+
+    try:
+        live_df = fetch_traffic_data(active_token)
+    except Exception as exc:
+        logger.warning(f"Failed to fetch live traffic: {exc}")
+        live_df = pd.DataFrame()
+
+    if dfs:
+        csv_df = pd.concat(dfs, ignore_index=True)
+        if not live_df.empty:
+            df = pd.concat([csv_df, live_df], ignore_index=True)
+        else:
+            df = csv_df
     else:
-        df = fetch_traffic_data(active_token)
+        df = live_df
+
+    if not df.empty:
+        df = df.drop_duplicates(subset=["date", "repository"], keep="last")
 
     df = df.replace([float('inf'), float('-inf')], None).where(pd.notnull(df), None)
 
@@ -171,8 +187,17 @@ def get_traffic(token: str = Body("", embed=True)):
 
 @app.post("/api/upload-csv")
 def upload_csv(file: UploadFile = File(...)):
-    # Accept a user-uploaded CSV — deep stats not available in CSV mode
     try:
+        data_dir = os.environ.get("GITLYTICS_DATA_DIR")
+        if data_dir:
+            data_dir_path = Path(data_dir)
+            data_dir_path.mkdir(parents=True, exist_ok=True)
+            file.file.seek(0)
+            content = file.file.read()
+            file.file.seek(0)
+            dest = data_dir_path / f"traffic_uploaded_{int(_time.time())}.csv"
+            with open(dest, "wb") as f:
+                f.write(content)
         df = process_uploaded_csv(file.file)
         df = df.replace([float('inf'), float('-inf')], None).where(pd.notnull(df), None)
         payload = build_react_payload(df, deep_stats=None)
@@ -192,7 +217,7 @@ def serve_index():
         return FileResponse(index_file)
     return JSONResponse(
         status_code=503,
-        content={"error": "Dashboard not found. Run 'npm run build' in the dashboard directory."}
+        content={"error": "Dashboard assets not found in the package installation."}
     )
 
 
@@ -209,7 +234,7 @@ def serve_spa_fallback(full_path: str):
 
     return JSONResponse(
         status_code=503,
-        content={"error": "Dashboard not found. Run 'npm run build' in the dashboard directory."}
+        content={"error": "Dashboard assets not found in the package installation."}
     )
 
 

{gitlytics-0.1.6 → gitlytics-0.2.0}/src/gitlytics/automation.py

@@ -130,13 +130,15 @@ def run_sync_cycle(token: str, repo_names=None, data_dir="./data", output_mode="
     else:
         existing_fields = new_fields
 
-    # Overwrite existing rows with fresh data, and add brand-new day rows
+    # Merge fresh data into existing rows — preserves columns not present in this sync run
     new_records_added = 0
     for _, row in df.iterrows():
         key = (str(row["repository"]), str(row["date"]))
         if key not in existing_data:
             new_records_added += 1
-        existing_data[key] = row.to_dict()
+            existing_data[key] = row.to_dict()
+        else:
+            existing_data[key].update(row.to_dict())
 
     # Sort all rows by date and repo name before writing back to disk
     final_rows = []

{gitlytics-0.1.6 → gitlytics-0.2.0}/src/gitlytics/core.py

@@ -221,8 +221,8 @@ def get_deep_repo_stats(token: str, full_name: str) -> dict:
         "has_code_of_conduct": None,
     }
 
-    # Commit activity — GitHub computes this async and returns 202 when not ready
-    # We do not block the thread pool worker with sleep; accept None on 202 (C-1)
+    # Commit activity — GitHub computes this async and returns 202 when not ready.
+    # total_commits here is the trailing 52-week (12-month) sum, NOT lifetime.
     ca_url = f"{BASE}/repos/{full_name}/stats/commit_activity"
     ca_data, status = _safe_get(ca_url, h)
     if status == 202:
@@ -254,12 +254,35 @@ def get_deep_repo_stats(token: str, full_name: str) -> dict:
         stats["has_contributing"] = bool(files.get("contributing"))
         stats["has_code_of_conduct"] = bool(files.get("code_of_conduct"))
 
-    # Releases — count and most recent publish date
-    rel_data, _ = _safe_get(f"{BASE}/repos/{full_name}/releases", h, params={"per_page": 100})
-    if isinstance(rel_data, list):
-        stats["total_releases"] = len(rel_data)
-        if rel_data:
-            stats["last_release_at"] = rel_data[0].get("published_at") or rel_data[0].get("created_at")
+    # Releases — count real total using Link header pagination, then get latest date
+    try:
+        rel_resp = requests.get(
+            f"{BASE}/repos/{full_name}/releases",
+            headers=h,
+            params={"per_page": 1},
+            timeout=10,
+        )
+        if rel_resp.status_code == 200:
+            link = rel_resp.headers.get("Link", "")
+            if 'rel="last"' in link:
+                import re as _re
+                m = _re.search(r'page=(\d+)>; rel="last"', link)
+                stats["total_releases"] = int(m.group(1)) if m else 1
+            else:
+                # Only one page — count items in this page
+                items = rel_resp.json()
+                stats["total_releases"] = len(items) if isinstance(items, list) else 0
+            # Fetch the latest release separately for its date
+            latest_resp = requests.get(
+                f"{BASE}/repos/{full_name}/releases/latest",
+                headers=h,
+                timeout=10,
+            )
+            if latest_resp.status_code == 200:
+                latest = latest_resp.json()
+                stats["last_release_at"] = latest.get("published_at") or latest.get("created_at")
+    except Exception as exc:
+        logger.warning(f"Could not fetch releases for {full_name}: {exc}")
 
     return stats
 

{gitlytics-0.1.6 → gitlytics-0.2.0}/src/gitlytics/process.py

@@ -35,12 +35,12 @@ def build_json_payload(df: pd.DataFrame, return_format: str = "timeseries", expo
     for repo, group in df.groupby("repository"):
         group = group.sort_values("date")
 
-        r_views = int(group["views"].sum()) if "views" in group.columns else 0
-        r_clones = int(group["clones"].sum()) if "clones" in group.columns else 0
-        r_unique_v = int(group["unique_visitors"].sum()) if "unique_visitors" in group.columns else 0
-        r_unique_c = int(group["unique_cloners"].sum()) if "unique_cloners" in group.columns else 0
-        r_stars = int(group["stars"].dropna().iloc[-1]) if "stars" in group.columns and not group["stars"].dropna().empty else 0
-        r_forks = int(group["forks"].dropna().iloc[-1]) if "forks" in group.columns and not group["forks"].dropna().empty else 0
+        r_views = _safe_int(group["views"].sum()) if "views" in group.columns else 0
+        r_clones = _safe_int(group["clones"].sum()) if "clones" in group.columns else 0
+        r_unique_v = _safe_int(group["unique_visitors"].sum()) if "unique_visitors" in group.columns else 0
+        r_unique_c = _safe_int(group["unique_cloners"].sum()) if "unique_cloners" in group.columns else 0
+        r_stars = _safe_int(group["stars"].dropna().iloc[-1]) if "stars" in group.columns and not group["stars"].dropna().empty else 0
+        r_forks = _safe_int(group["forks"].dropna().iloc[-1]) if "forks" in group.columns and not group["forks"].dropna().empty else 0
         r_is_private = bool(group["is_private"].dropna().iloc[-1]) if "is_private" in group.columns and not group["is_private"].dropna().empty else False
 
         top_ref = str(group["top_referrer"].dropna().iloc[-1]) if "top_referrer" in group.columns and not group["top_referrer"].dropna().empty else ""
@@ -103,6 +103,11 @@ def build_json_payload(df: pd.DataFrame, return_format: str = "timeseries", expo
 def process_uploaded_csv(uploaded_file) -> pd.DataFrame:
     """Reads a user-uploaded CSV and normalises column names to match our tidy schema."""
     raw_df = pd.read_csv(uploaded_file)
+
+    # Normalise capitalized "Date" column before any other checks
+    if "Date" in raw_df.columns and "date" not in raw_df.columns:
+        raw_df = raw_df.rename(columns={"Date": "date"})
+
     if "repository" not in raw_df.columns:
         if "repo_name" in raw_df.columns:
             raw_df = raw_df.rename(columns={"repo_name": "repository"})
@@ -113,7 +118,7 @@ def process_uploaded_csv(uploaded_file) -> pd.DataFrame:
             })
         else:
             raise ValueError("Invalid CSV format: missing 'repository' column")
-    # M-5: validate date column exists after renaming so callers get 400 not a 500 KeyError later
+
     if "date" not in raw_df.columns:
         raise ValueError("Invalid CSV format: missing required 'date' column")
     return raw_df
@@ -161,10 +166,10 @@ def build_react_payload(df: pd.DataFrame, deep_stats: dict = None) -> list:
     for repo, group in df.groupby("repository"):
         group = group.sort_values("date")
 
-        r_views = int(group["views"].sum()) if "views" in group.columns else 0
-        r_clones = int(group["clones"].sum()) if "clones" in group.columns else 0
-        r_unique_v = int(group["unique_visitors"].sum()) if "unique_visitors" in group.columns else 0
-        r_unique_c = int(group["unique_cloners"].sum()) if "unique_cloners" in group.columns else 0
+        r_views = _safe_int(group["views"].sum()) if "views" in group.columns else 0
+        r_clones = _safe_int(group["clones"].sum()) if "clones" in group.columns else 0
+        r_unique_v = _safe_int(group["unique_visitors"].sum()) if "unique_visitors" in group.columns else 0
+        r_unique_c = _safe_int(group["unique_cloners"].sum()) if "unique_cloners" in group.columns else 0
         r_stars = _safe_int(group["stars"].dropna().iloc[-1]) if "stars" in group.columns and not group["stars"].dropna().empty else 0
         r_forks = _safe_int(group["forks"].dropna().iloc[-1]) if "forks" in group.columns and not group["forks"].dropna().empty else 0
         r_is_private = bool(group["is_private"].dropna().iloc[-1]) if "is_private" in group.columns and not group["is_private"].dropna().empty else False
@@ -194,13 +199,13 @@ def build_react_payload(df: pd.DataFrame, deep_stats: dict = None) -> list:
             date_str = str(row["date"])
             daily_views.append({
                 "timestamp": date_str,
-                "count": int(row.get("views", 0)),
-                "uniques": int(row.get("unique_visitors", 0))
+                "count": _safe_int(row.get("views", 0)),
+                "uniques": _safe_int(row.get("unique_visitors", 0))
             })
             daily_clones.append({
                 "timestamp": date_str,
-                "count": int(row.get("clones", 0)),
-                "uniques": int(row.get("unique_cloners", 0))  # fixed typo
+                "count": _safe_int(row.get("clones", 0)),
+                "uniques": _safe_int(row.get("unique_cloners", 0))  # fixed typo
             })
 
         raw_refs_val = group["_raw_referrers"].iloc[-1] if "_raw_referrers" in group.columns else None