Merge pull request #32 from PlaneQuery/develop

Develop to Main: Fix Community Submission export. Fix CSV concatenation logic to prevent duplicates when there is no new ADSB.lol data.

.github/workflows/adsb-to-aircraft-for-day.yaml

@@ -49,11 +49,38 @@ jobs:
           python -m src.adsb.download_and_list_icaos --date "$DATE"
           ls -lah data/output/adsb_archives/"$DATE" || true
 
-      - name: Upload archives
+      - name: Upload archive part 0
         uses: actions/upload-artifact@v4
         with:
-          name: adsb-archives-${{ inputs.date }}
-          path: data/output/adsb_archives/${{ inputs.date }}
+          name: adsb-archive-${{ inputs.date }}-part-0
+          path: data/output/adsb_archives/${{ inputs.date }}/${{ inputs.date }}_part_0.tar.gz
+          retention-days: 1
+          compression-level: 0
+          if-no-files-found: error
+
+      - name: Upload archive part 1
+        uses: actions/upload-artifact@v4
+        with:
+          name: adsb-archive-${{ inputs.date }}-part-1
+          path: data/output/adsb_archives/${{ inputs.date }}/${{ inputs.date }}_part_1.tar.gz
+          retention-days: 1
+          compression-level: 0
+          if-no-files-found: error
+
+      - name: Upload archive part 2
+        uses: actions/upload-artifact@v4
+        with:
+          name: adsb-archive-${{ inputs.date }}-part-2
+          path: data/output/adsb_archives/${{ inputs.date }}/${{ inputs.date }}_part_2.tar.gz
+          retention-days: 1
+          compression-level: 0
+          if-no-files-found: error
+
+      - name: Upload archive part 3
+        uses: actions/upload-artifact@v4
+        with:
+          name: adsb-archive-${{ inputs.date }}-part-3
+          path: data/output/adsb_archives/${{ inputs.date }}/${{ inputs.date }}_part_3.tar.gz
           retention-days: 1
           compression-level: 0
           if-no-files-found: error
@@ -79,12 +106,22 @@ jobs:
           python -m pip install --upgrade pip
           pip install -r requirements.txt
 
-      - name: Download archives
+      - name: Download archive part
         uses: actions/download-artifact@v4
         with:
-          name: adsb-archives-${{ inputs.date }}
+          name: adsb-archive-${{ inputs.date }}-part-${{ matrix.part_id }}
           path: data/output/adsb_archives/${{ inputs.date }}
 
+      - name: Verify archive
+        run: |
+          FILE="data/output/adsb_archives/${{ inputs.date }}/${{ inputs.date }}_part_${{ matrix.part_id }}.tar.gz"
+          ls -lah data/output/adsb_archives/${{ inputs.date }}/
+          if [ ! -f "$FILE" ]; then
+            echo "::error::Archive not found: $FILE"
+            exit 1
+          fi
+          echo "Verified: $(du -h "$FILE")"
+
       - name: Process part
         env:
           DATE: ${{ inputs.date }}
@@ -140,6 +177,6 @@ jobs:
         uses: actions/upload-artifact@v4
         with:
           name: openairframes_adsb-${{ inputs.date }}
-          path: data/output/openairframes_adsb_${{ inputs.date }}*
+          path: data/output/openairframes_adsb_*
           retention-days: 30
           if-no-files-found: error

.github/workflows/openairframes-daily-release.yaml

@@ -266,7 +266,11 @@ jobs:
           
           # Find files from artifacts using find (handles nested structures)
           CSV_FILE_FAA=$(find artifacts/faa -name "openairframes_faa_*.csv" -type f 2>/dev/null | head -1)
-          CSV_FILE_ADSB=$(find artifacts/adsb -name "openairframes_adsb_*.csv.gz" -type f 2>/dev/null | head -1)
+          # Prefer concatenated file (with date range) over single-day file
+          CSV_FILE_ADSB=$(find artifacts/adsb -name "openairframes_adsb_*_*.csv.gz" -type f 2>/dev/null | head -1)
+          if [ -z "$CSV_FILE_ADSB" ]; then
+            CSV_FILE_ADSB=$(find artifacts/adsb -name "openairframes_adsb_*.csv.gz" -type f 2>/dev/null | head -1)
+          fi
           CSV_FILE_COMMUNITY=$(find artifacts/community -name "openairframes_community_*.csv" -type f 2>/dev/null | head -1)
           ZIP_FILE=$(find artifacts/faa -name "ReleasableAircraft_*.zip" -type f 2>/dev/null | head -1)
           JSON_FILE_ADSBX=$(find artifacts/adsbexchange -name "basic-ac-db_*.json.gz" -type f 2>/dev/null | head -1)

scripts/run_historical_adsb_action.py

@@ -194,7 +194,7 @@ def main():
     if triggered_runs and not args.dry_run:
         import json
         timestamp = datetime.now().strftime('%Y%m%d_%H%M%S')
-        runs_file = f"./triggered_runs_{timestamp}.json"
+        runs_file = f"./output/triggered_runs_{timestamp}.json"
         with open(runs_file, 'w') as f:
             json.dump({
                 'start_date': args.start_date,

src/adsb/concat_parquet_to_final.py

@@ -37,14 +37,31 @@ def main():
     if args.concat_with_latest_csv:
         print("Loading latest CSV from GitHub releases to concatenate with...")
         from src.get_latest_release import get_latest_aircraft_adsb_csv_df
-        df_latest_csv, csv_date = get_latest_aircraft_adsb_csv_df()
-        # Ensure column order matches before concatenating
-        df_latest_csv = df_latest_csv.select(CORRECT_ORDER_OF_COLUMNS)
-        from src.adsb.compress_adsb_to_aircraft_data import concat_compressed_dfs
-        df_final = concat_compressed_dfs(df_latest_csv, df)
-        df_final = df_final.select(CORRECT_ORDER_OF_COLUMNS)
-        final_csv_output_path = OUTPUT_DIR / f"openairframes_adsb_{csv_date}_{args.date}.csv.gz"
-        df_final.write_csv(final_csv_output_path, compression="gzip")
+        from datetime import datetime
+        
+        df_latest_csv, csv_start_date, csv_end_date = get_latest_aircraft_adsb_csv_df()
+        
+        # Compare dates: end_date is exclusive, so if csv_end_date > args.date, 
+        # the latest CSV already includes this day's data
+        csv_end_dt = datetime.strptime(csv_end_date, "%Y-%m-%d")
+        args_dt = datetime.strptime(args.date, "%Y-%m-%d")
+        
+        if csv_end_dt >= args_dt:
+            print(f"Latest CSV already includes data through {args.date} (end_date={csv_end_date} is exclusive)")
+            print("Writing latest CSV directly without concatenation to avoid duplicates")
+            final_csv_output_path = OUTPUT_DIR / f"openairframes_adsb_{csv_start_date}_{csv_end_date}.csv.gz"
+            df_latest_csv = df_latest_csv.select(CORRECT_ORDER_OF_COLUMNS)
+            df_latest_csv.write_csv(final_csv_output_path, compression="gzip")
+        else:
+            print(f"Concatenating latest CSV (through {csv_end_date}) with new data ({args.date})")
+            # Ensure column order matches before concatenating
+            df_latest_csv = df_latest_csv.select(CORRECT_ORDER_OF_COLUMNS)
+            from src.adsb.compress_adsb_to_aircraft_data import concat_compressed_dfs
+            df_final = concat_compressed_dfs(df_latest_csv, df)
+            df_final = df_final.select(CORRECT_ORDER_OF_COLUMNS)
+            final_csv_output_path = OUTPUT_DIR / f"openairframes_adsb_{csv_start_date}_{args.date}.csv.gz"
+            df_final.write_csv(final_csv_output_path, compression="gzip")
+        print(f"Final CSV written to {final_csv_output_path}")
 
 if __name__ == "__main__":
     main()

src/adsb/process_icao_chunk.py

@@ -123,7 +123,16 @@ def main():
     print(f"Processing part {args.part_id} for {args.date}")
     
     # Get specific archive file for this part
-    archive_path = os.path.join(OUTPUT_DIR, "adsb_archives", args.date, f"{args.date}_part_{args.part_id}.tar.gz")
+    archive_dir = os.path.join(OUTPUT_DIR, "adsb_archives", args.date)
+    archive_path = os.path.join(archive_dir, f"{args.date}_part_{args.part_id}.tar.gz")
+    
+    if not os.path.isfile(archive_path):
+        print(f"ERROR: Archive not found: {archive_path}")
+        if os.path.isdir(archive_dir):
+            print(f"Files in {archive_dir}: {os.listdir(archive_dir)}")
+        else:
+            print(f"Directory does not exist: {archive_dir}")
+        sys.exit(1)
     
     # Extract and collect trace files
     trace_map = build_trace_file_map(archive_path)

src/contributions/create_daily_community_release.py

@@ -24,7 +24,7 @@ def read_all_submissions(community_dir: Path) -> list[dict]:
     """Read all JSON submissions from the community directory."""
     all_submissions = []
     
-    for json_file in sorted(community_dir.glob("*.json")):
+    for json_file in sorted(community_dir.glob("**/*.json")):
         try:
             with open(json_file) as f:
                 data = json.load(f)

src/get_latest_release.py

@@ -27,6 +27,33 @@ def _http_get_json(url: str, headers: dict[str, str]) -> dict:
     return json.loads(data.decode("utf-8"))
 
 
+def get_releases(repo: str = REPO, github_token: Optional[str] = None, per_page: int = 30) -> list[dict]:
+    """Get a list of releases from the repository."""
+    url = f"https://api.github.com/repos/{repo}/releases?per_page={per_page}"
+    headers = {
+        "Accept": "application/vnd.github+json",
+        "User-Agent": "openairframes-downloader/1.0",
+    }
+    if github_token:
+        headers["Authorization"] = f"Bearer {github_token}"
+
+    return _http_get_json(url, headers=headers)
+
+
+def get_release_assets_from_release_data(release_data: dict) -> list[ReleaseAsset]:
+    """Extract assets from a release data dictionary."""
+    assets = []
+    for a in release_data.get("assets", []):
+        assets.append(
+            ReleaseAsset(
+                name=a["name"],
+                download_url=a["browser_download_url"],
+                size=int(a.get("size", 0)),
+            )
+        )
+    return assets
+
+
 def get_latest_release_assets(repo: str = REPO, github_token: Optional[str] = None) -> list[ReleaseAsset]:
     url = f"https://api.github.com/repos/{repo}/releases/latest"
     headers = {
@@ -37,16 +64,7 @@ def get_latest_release_assets(repo: str = REPO, github_token: Optional[str] = No
         headers["Authorization"] = f"Bearer {github_token}"
 
     payload = _http_get_json(url, headers=headers)
-    assets = []
-    for a in payload.get("assets", []):
-        assets.append(
-            ReleaseAsset(
-                name=a["name"],
-                download_url=a["browser_download_url"],
-                size=int(a.get("size", 0)),
-            )
-        )
-    return assets
+    return get_release_assets_from_release_data(payload)
 
 
 def pick_asset(
@@ -155,7 +173,8 @@ def download_latest_aircraft_adsb_csv(
     repo: str = REPO,
 ) -> Path:
     """
-    Download the latest openairframes_adsb_*.csv file from the latest GitHub release.
+    Download the latest openairframes_adsb_*.csv file from GitHub releases.
+    If the latest release doesn't have the file, searches previous releases.
 
     Args:
         output_dir: Directory to save the downloaded file (default: "downloads")
@@ -166,15 +185,33 @@ def download_latest_aircraft_adsb_csv(
         Path to the downloaded file
     """
     output_dir = Path(output_dir)
-    assets = get_latest_release_assets(repo, github_token=github_token)
-    asset = pick_asset(assets, name_regex=r"^openairframes_adsb_.*\.csv(\.gz)?$")
-    saved_to = download_asset(asset, output_dir / asset.name, github_token=github_token)
-    print(f"Downloaded: {asset.name} ({asset.size} bytes) -> {saved_to}")
-    return saved_to
+    
+    # Get multiple releases
+    releases = get_releases(repo, github_token=github_token, per_page=30)
+    
+    # Try each release until we find one with the matching asset
+    for release in releases:
+        assets = get_release_assets_from_release_data(release)
+        try:
+            asset = pick_asset(assets, name_regex=r"^openairframes_adsb_.*\.csv(\.gz)?$")
+            saved_to = download_asset(asset, output_dir / asset.name, github_token=github_token)
+            print(f"Downloaded: {asset.name} ({asset.size} bytes) -> {saved_to}")
+            return saved_to
+        except FileNotFoundError:
+            # This release doesn't have the matching asset, try the next one
+            continue
+    
+    raise FileNotFoundError(
+        f"No release in the last 30 releases has an asset matching 'openairframes_adsb_.*\\.csv(\\.gz)?$'"
+    )
 
 import polars as pl
 def get_latest_aircraft_adsb_csv_df():
-    """Download and load the latest ADS-B CSV from GitHub releases."""
+    """Download and load the latest ADS-B CSV from GitHub releases.
+    
+    Returns:
+        tuple: (df, start_date, end_date) where dates are in YYYY-MM-DD format
+    """
     import re
     
     csv_path = download_latest_aircraft_adsb_csv()
@@ -198,17 +235,19 @@ def get_latest_aircraft_adsb_csv_df():
         if df[col].dtype == pl.Utf8:
             df = df.with_columns(pl.col(col).fill_null(""))
     
-    # Extract start date from filename pattern: openairframes_adsb_{start_date}_{end_date}.csv[.gz]
-    match = re.search(r"openairframes_adsb_(\d{4}-\d{2}-\d{2})_", str(csv_path))
+    # Extract start and end dates from filename pattern: openairframes_adsb_{start_date}_{end_date}.csv[.gz]
+    match = re.search(r"openairframes_adsb_(\d{4}-\d{2}-\d{2})_(\d{4}-\d{2}-\d{2})\.csv", str(csv_path))
     if not match:
-        raise ValueError(f"Could not extract date from filename: {csv_path.name}")
+        raise ValueError(f"Could not extract dates from filename: {csv_path.name}")
     
-    date_str = match.group(1)
+    start_date = match.group(1)
+    end_date = match.group(2)
     print(df.columns)
     print(df.dtypes)
-    return df, date_str
+    return df, start_date, end_date
 
 
 
 if __name__ == "__main__":
     download_latest_aircraft_csv()
+    download_latest_aircraft_adsb_csv()