Update schema with new tags: airshows, dog_friendly, cat_friendly, notes

.github/ISSUE_TEMPLATE/community_submission.yaml

@@ -13,29 +13,42 @@ body:
         **Rules (enforced on review/automation):**
         - Each object must include **at least one** of:
           - `registration_number`
-          - `transponder_code_hex` (6 hex chars)
+          - `transponder_code_hex` (6 uppercase hex chars, e.g., `ABC123`)
           - `planequery_airframe_id`
         - Your contributor name (entered below) will be applied to all objects.
         - `contributor_uuid` is derived from your GitHub account automatically.
         - `creation_timestamp` is created by the system (you may omit it).
 
+        **Optional date scoping:**
+        - `start_date` - When the tags become valid (ISO 8601: `YYYY-MM-DD`)
+        - `end_date` - When the tags stop being valid (ISO 8601: `YYYY-MM-DD`)
+
         **Example: single object**
         ```json
         {
-          "transponder_code_hex": "a1b2c3"
+            "registration_number": "N12345",
+            "tags": {"owner": "John Doe"},
+            "start_date": "2025-01-01"
         }
         ```
-
         **Example: multiple objects (array)**
         ```json
         [
-          {
-            "registration_number": "N123AB"
-          },
-          {
-            "planequery_airframe_id": "cessna|172s|12345",
-            "transponder_code_hex": "0f1234"
-          }
+        {
+            "registration_number": "N12345",
+            "tags": {"internet": "starlink"},
+            "start_date": "2025-05-01"
+        },
+        {
+            "registration_number": "N12345",
+            "tags": {"owner": "John Doe"},
+            "start_date": "2025-01-01",
+            "end_date": "2025-07-20"
+        },
+        {
+            "transponder_code_hex": "ABC123",
+            "tags": {"internet": "viasat", "owner": "John Doe"}
+        }
         ]
         ```
 
@@ -52,9 +65,11 @@ body:
     id: submission_json
     attributes:
       label: Submission JSON
-      description: Paste either one JSON object or an array of JSON objects. Must be valid JSON. Do not include contributor_name or contributor_uuid in your JSON.
+      description: |
+        Paste JSON directly, OR drag-and-drop a .json file here.
+        Must be valid JSON. Do not include contributor_name or contributor_uuid.
       placeholder: |
-        Paste JSON here...
+        Paste JSON here, or drag-and-drop a .json file...
     validations:
       required: true
 

.github/workflows/approve-community-submission.yaml

@@ -38,9 +38,10 @@ jobs:
         env:
           GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
           GITHUB_REPOSITORY: ${{ github.repository }}
+          ISSUE_BODY: ${{ github.event.issue.body }}
         run: |
           python -m src.contributions.approve_submission \
             --issue-number ${{ github.event.issue.number }} \
-            --issue-body "${{ github.event.issue.body }}" \
+            --issue-body "$ISSUE_BODY" \
             --author "${{ steps.author.outputs.username }}" \
             --author-id ${{ steps.author.outputs.user_id }}

.github/workflows/historical-adsb.yaml

@@ -81,8 +81,22 @@ jobs:
       - name: Create tar of extracted data
         run: |
           cd data/output
-          tar -cf extracted_data.tar *-planes-readsb-prod-0.tar_0 icao_manifest_*.txt 2>/dev/null || echo "Some files may not exist"
-          ls -lah extracted_data.tar || echo "No tar created"
+          echo "=== Disk space before tar ==="
+          df -h .
+          echo "=== Files to tar ==="
+          ls -lah *-planes-readsb-prod-0.tar_0 icao_manifest_*.txt 2>/dev/null || echo "No files found"
+          
+          # Create tar with explicit error checking
+          if ls *-planes-readsb-prod-0.tar_0 1>/dev/null 2>&1; then
+            tar -cvf extracted_data.tar *-planes-readsb-prod-0.tar_0 icao_manifest_*.txt
+            echo "=== Tar file created ==="
+            ls -lah extracted_data.tar
+            # Verify tar integrity
+            tar -tf extracted_data.tar > /dev/null && echo "Tar integrity check passed" || { echo "Tar integrity check FAILED"; exit 1; }
+          else
+            echo "ERROR: No extracted directories found, cannot create tar"
+            exit 1
+          fi
 
       - name: Upload extracted data
         uses: actions/upload-artifact@v4
@@ -97,7 +111,7 @@ jobs:
     needs: [generate-matrix, adsb-extract]
     runs-on: ubuntu-24.04-arm
     strategy:
-      fail-fast: false
+      fail-fast: true
       matrix:
         chunk: ${{ fromJson(needs.generate-matrix.outputs.chunks) }}
         icao_chunk: [0, 1, 2, 3]
@@ -134,7 +148,12 @@ jobs:
         run: |
           cd data/output
           if [ -f extracted_data.tar ]; then
-            tar -xf extracted_data.tar
+            echo "=== Tar file info ==="
+            ls -lah extracted_data.tar
+            echo "=== Verifying tar integrity ==="
+            tar -tf extracted_data.tar > /dev/null || { echo "ERROR: Tar file is corrupted"; exit 1; }
+            echo "=== Extracting ==="
+            tar -xvf extracted_data.tar
             rm extracted_data.tar
             echo "has_data=true" >> "$GITHUB_OUTPUT"
             echo "=== Contents of data/output ==="
@@ -188,17 +207,19 @@ jobs:
 
       - name: Debug downloaded files
         run: |
+          echo "=== Disk space before processing ==="
+          df -h
           echo "=== Listing data/output/adsb_chunks/ ==="
-          find data/output/adsb_chunks/ -type f 2>/dev/null | head -50 || echo "No files found"
-          echo "=== Looking for parquet files ==="
-          find . -name "*.parquet" 2>/dev/null | head -20 || echo "No parquet files found"
+          find data/output/adsb_chunks/ -type f 2>/dev/null | wc -l
+          echo "=== Total parquet size ==="
+          du -sh data/output/adsb_chunks/ || echo "No chunks dir"
 
       - name: Combine chunks to CSV
         env:
           START_DATE: ${{ needs.generate-matrix.outputs.global_start }}
           END_DATE: ${{ needs.generate-matrix.outputs.global_end }}
         run: |
-          python -m src.adsb.combine_chunks_to_csv --chunks-dir data/output/adsb_chunks --start-date "$START_DATE" --end-date "$END_DATE" --skip-base
+          python -m src.adsb.combine_chunks_to_csv --chunks-dir data/output/adsb_chunks --start-date "$START_DATE" --end-date "$END_DATE" --skip-base --stream
           ls -lah data/planequery_aircraft/
 
       - name: Upload final artifact

.github/workflows/planequery-aircraft-daily-release.yaml

@@ -277,6 +277,15 @@ jobs:
           name: community-release
           path: artifacts/community
 
+      - name: Debug artifact structure
+        run: |
+          echo "=== FAA artifacts ==="
+          find artifacts/faa -type f 2>/dev/null || echo "No files found in artifacts/faa"
+          echo "=== ADS-B artifacts ==="
+          find artifacts/adsb -type f 2>/dev/null || echo "No files found in artifacts/adsb"
+          echo "=== Community artifacts ==="
+          find artifacts/community -type f 2>/dev/null || echo "No files found in artifacts/community"
+
       - name: Prepare release metadata
         id: meta
         run: |
@@ -312,6 +321,13 @@ jobs:
           echo "zip_basename=$ZIP_BASENAME" >> "$GITHUB_OUTPUT"
           echo "name=planequery-aircraft snapshot ($DATE)${BRANCH_SUFFIX}" >> "$GITHUB_OUTPUT"
 
+      - name: Delete existing release if exists
+        run: |
+          gh release delete "${{ steps.meta.outputs.tag }}" --yes 2>/dev/null || true
+          git push --delete origin "refs/tags/${{ steps.meta.outputs.tag }}" 2>/dev/null || true
+        env:
+          GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+
       - name: Create GitHub Release and upload assets
         uses: softprops/action-gh-release@v2
         with:

.github/workflows/update-community-prs.yaml

@@ -0,0 +1,77 @@
+name: Update Community PRs After Merge
+
+on:
+  push:
+    branches: [main]
+    paths:
+      - 'community/**'
+      - 'schemas/community_submission.v1.schema.json'
+
+permissions:
+  contents: write
+  pull-requests: write
+
+jobs:
+  update-open-prs:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+          token: ${{ secrets.GITHUB_TOKEN }}
+
+      - name: Setup Python
+        uses: actions/setup-python@v5
+        with:
+          python-version: "3.12"
+
+      - name: Install dependencies
+        run: pip install jsonschema
+
+      - name: Find and update open community PRs
+        env:
+          GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+        run: |
+          # Get list of open community PRs
+          prs=$(gh pr list --label community --state open --json number,headRefName --jq '.[] | "\(.number) \(.headRefName)"')
+          
+          if [ -z "$prs" ]; then
+            echo "No open community PRs found"
+            exit 0
+          fi
+          
+          echo "$prs" | while read pr_number branch_name; do
+            echo "Processing PR #$pr_number (branch: $branch_name)"
+            
+            # Checkout PR branch
+            git fetch origin "$branch_name"
+            git checkout "$branch_name"
+            
+            # Merge main into PR branch
+            git config user.name "github-actions[bot]"
+            git config user.email "github-actions[bot]@users.noreply.github.com"
+            
+            if git merge origin/main -m "Merge main to update schema"; then
+              # Regenerate schema for this PR's submission (adds any new tags)
+              python -m src.contributions.regenerate_pr_schema || true
+              
+              # If there are changes, commit and push
+              if [ -n "$(git status --porcelain schemas/)" ]; then
+                git add schemas/
+                git commit -m "Update schema with new tags"
+                git push origin "$branch_name"
+                echo "  Updated PR #$pr_number with schema changes"
+              else
+                git push origin "$branch_name"
+                echo "  Merged main into PR #$pr_number"
+              fi
+            else
+              echo "  Merge conflict in PR #$pr_number, adding comment"
+              gh pr comment "$pr_number" --body $'⚠️ **Merge Conflict**\n\nAnother community submission was merged and this PR has conflicts.\n\nA maintainer may need to:\n1. Close this PR\n2. Remove the `approved` label from the original issue\n3. Re-add the `approved` label to regenerate the PR'
+              git merge --abort
+              fi
+            fi
+            
+            git checkout main
+          done

.github/workflows/validate-community-submission.yaml

@@ -4,6 +4,9 @@ on:
   issues:
     types: [opened, edited]
 
+permissions:
+  issues: write
+
 jobs:
   validate:
     if: contains(github.event.issue.labels.*.name, 'submission')
@@ -20,11 +23,24 @@ jobs:
       - name: Install dependencies
         run: pip install jsonschema
 
+      - name: Debug issue body
+        run: |
+          echo "=== Issue Body ==="
+          cat << 'ISSUE_BODY_EOF'
+          ${{ github.event.issue.body }}
+          ISSUE_BODY_EOF
+
+      - name: Save issue body to file
+        run: |
+          cat << 'ISSUE_BODY_EOF' > /tmp/issue_body.txt
+          ${{ github.event.issue.body }}
+          ISSUE_BODY_EOF
+
       - name: Validate submission
         env:
           GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
           GITHUB_REPOSITORY: ${{ github.repository }}
         run: |
           python -m src.contributions.validate_submission \
-            --issue-body "${{ github.event.issue.body }}" \
+            --issue-body-file /tmp/issue_body.txt \
             --issue-number ${{ github.event.issue.number }}

community/2026-02-12/ggman12_2026-02-12_58e320bb.json

@@ -0,0 +1,14 @@
+[
+  {
+    "contributor_name": "hellohello",
+    "contributor_uuid": "2981c3ee-8712-5f96-84bf-732eda515a3f",
+    "creation_timestamp": "2026-02-12T21:02:32.325360+00:00",
+    "registration_number": "N12345",
+    "tags": {
+      "airshows": true,
+      "cat_friendly": false,
+      "dog_friendly": true,
+      "notes": "is a pet carrier"
+    }
+  }
+]

requirements.txt

@@ -3,3 +3,4 @@ pandas==3.0.0
 pyarrow==23.0.0
 orjson==3.11.7
 polars==1.38.1
+jsonschema==4.26.0

schemas/community_submission.v1.schema.json

@@ -3,7 +3,6 @@
   "title": "PlaneQuery Aircraft Community Submission (v1)",
   "type": "object",
   "additionalProperties": false,
-
   "properties": {
     "registration_number": {
       "type": "string",
@@ -11,13 +10,12 @@
     },
     "transponder_code_hex": {
       "type": "string",
-      "pattern": "^[0-9A-Fa-f]{6}$"
+      "pattern": "^[0-9A-F]{6}$"
     },
     "planequery_airframe_id": {
       "type": "string",
       "minLength": 1
     },
-
     "contributor_uuid": {
       "type": "string",
       "format": "uuid"
@@ -28,14 +26,24 @@
       "maxLength": 150,
       "description": "Display name (may be blank)"
     },
-
     "creation_timestamp": {
       "type": "string",
       "format": "date-time",
       "description": "Set by the system when the submission is persisted/approved.",
       "readOnly": true
     },
-
+    "start_date": {
+      "type": "string",
+      "format": "date",
+      "pattern": "^\\d{4}-\\d{2}-\\d{2}$",
+      "description": "Optional start date for when this submission's tags are valid (ISO 8601, e.g., 2025-05-01)."
+    },
+    "end_date": {
+      "type": "string",
+      "format": "date",
+      "pattern": "^\\d{4}-\\d{2}-\\d{2}$",
+      "description": "Optional end date for when this submission's tags are valid (ISO 8601, e.g., 2025-07-03)."
+    },
     "tags": {
       "type": "object",
       "description": "Additional community-defined tags as key/value pairs (values may be scalar, array, or object).",
@@ -43,38 +51,76 @@
         "type": "string",
         "pattern": "^[a-z][a-z0-9_]{0,63}$"
       },
-      "additionalProperties": { "$ref": "#/$defs/tagValue" }
+      "additionalProperties": {
+        "$ref": "#/$defs/tagValue"
+      },
+      "properties": {
+        "airshows": {
+          "type": "boolean"
+        },
+        "cat_friendly": {
+          "type": "boolean"
+        },
+        "dog_friendly": {
+          "type": "boolean"
+        },
+        "notes": {
+          "type": "string"
+        }
+      }
     }
   },
-
   "allOf": [
     {
       "anyOf": [
-        { "required": ["registration_number"] },
-        { "required": ["transponder_code_hex"] },
-        { "required": ["planequery_airframe_id"] }
+        {
+          "required": [
+            "registration_number"
+          ]
+        },
+        {
+          "required": [
+            "transponder_code_hex"
+          ]
+        },
+        {
+          "required": [
+            "planequery_airframe_id"
+          ]
+        }
       ]
     }
   ],
-
   "$defs": {
     "tagScalar": {
-      "type": ["string", "number", "integer", "boolean", "null"]
+      "type": [
+        "string",
+        "number",
+        "integer",
+        "boolean",
+        "null"
+      ]
     },
     "tagValue": {
       "anyOf": [
-        { "$ref": "#/$defs/tagScalar" },
+        {
+          "$ref": "#/$defs/tagScalar"
+        },
         {
           "type": "array",
           "maxItems": 50,
-          "items": { "$ref": "#/$defs/tagScalar" }
+          "items": {
+            "$ref": "#/$defs/tagScalar"
+          }
         },
         {
           "type": "object",
           "maxProperties": 50,
-          "additionalProperties": { "$ref": "#/$defs/tagScalar" }
+          "additionalProperties": {
+            "$ref": "#/$defs/tagScalar"
+          }
         }
       ]
     }
   }
-}
+}

src/adsb/combine_chunks_to_csv.py

@@ -36,8 +36,13 @@ def get_target_day() -> datetime:
     return datetime.utcnow() - timedelta(days=1)
 
 
-def process_single_chunk(chunk_path: str) -> pl.DataFrame:
-    """Load and compress a single chunk parquet file."""
+def process_single_chunk(chunk_path: str, delete_after_load: bool = False) -> pl.DataFrame:
+    """Load and compress a single chunk parquet file.
+    
+    Args:
+        chunk_path: Path to parquet file
+        delete_after_load: If True, delete the parquet file after loading to free disk space
+    """
     print(f"Processing {os.path.basename(chunk_path)}... | {get_resource_usage()}")
     
     # Load chunk - only columns we need
@@ -45,6 +50,14 @@ def process_single_chunk(chunk_path: str) -> pl.DataFrame:
     df = pl.read_parquet(chunk_path, columns=needed_columns)
     print(f"  Loaded {len(df)} rows")
     
+    # Delete file immediately after loading to free disk space
+    if delete_after_load:
+        try:
+            os.remove(chunk_path)
+            print(f"  Deleted {chunk_path} to free disk space")
+        except Exception as e:
+            print(f"  Warning: Failed to delete {chunk_path}: {e}")
+    
     # Compress to aircraft records (one per ICAO) using shared function
     compressed = compress_multi_icao_df(df, verbose=True)
     print(f"  Compressed to {len(compressed)} aircraft records")
@@ -156,6 +169,7 @@ def main():
     parser.add_argument("--chunks-dir", type=str, default=DEFAULT_CHUNK_DIR, help="Directory containing chunk parquet files")
     parser.add_argument("--skip-base", action="store_true", help="Skip downloading and merging base release")
     parser.add_argument("--keep-chunks", action="store_true", help="Keep chunk files after merging")
+    parser.add_argument("--stream", action="store_true", help="Delete parquet files immediately after loading to save disk space")
     args = parser.parse_args()
     
     # Determine output ID and filename based on mode
@@ -190,9 +204,10 @@ def main():
     print(f"Found {len(chunk_files)} chunk files")
     
     # Process each chunk separately to save memory
+    # With --stream, delete parquet files immediately after loading to save disk space
     compressed_chunks = []
     for chunk_path in chunk_files:
-        compressed = process_single_chunk(chunk_path)
+        compressed = process_single_chunk(chunk_path, delete_after_load=args.stream)
         compressed_chunks.append(compressed)
         gc.collect()
     

src/contributions/approve_submission.py

@@ -21,12 +21,14 @@ import urllib.request
 import urllib.error
 from datetime import datetime, timezone
 
-from .schema import extract_json_from_issue_body, extract_contributor_name_from_issue_body, parse_and_validate
+from .schema import extract_json_from_issue_body, extract_contributor_name_from_issue_body, parse_and_validate, load_schema, SCHEMAS_DIR
 from .contributor import (
     generate_contributor_uuid,
     generate_submission_filename,
     compute_content_hash,
 )
+from .update_schema import generate_updated_schema, check_for_new_tags, get_existing_tag_definitions
+from .read_community_data import build_tag_type_registry
 
 
 def github_api_request(
@@ -54,7 +56,11 @@ def github_api_request(
     
     try:
         with urllib.request.urlopen(req) as response:
-            return json.loads(response.read())
+            response_body = response.read()
+            # DELETE requests return empty body (204 No Content)
+            if not response_body:
+                return {}
+            return json.loads(response_body)
     except urllib.error.HTTPError as e:
         error_body = e.read().decode() if e.fp else ""
         print(f"GitHub API error: {e.code} {e.reason}: {error_body}", file=sys.stderr)
@@ -94,14 +100,30 @@ def create_branch(branch_name: str, sha: str) -> None:
             raise
 
 
+def get_file_sha(path: str, branch: str) -> str | None:
+    """Get the SHA of an existing file, or None if it doesn't exist."""
+    try:
+        response = github_api_request("GET", f"/contents/{path}?ref={branch}")
+        return response.get("sha")
+    except Exception:
+        return None
+
+
 def create_or_update_file(path: str, content: str, message: str, branch: str) -> None:
     """Create or update a file in the repository."""
     content_b64 = base64.b64encode(content.encode()).decode()
-    github_api_request("PUT", f"/contents/{path}", {
+    payload = {
         "message": message,
         "content": content_b64,
         "branch": branch,
-    })
+    }
+    
+    # If file exists, we need to include its SHA to update it
+    sha = get_file_sha(path, branch)
+    if sha:
+        payload["sha"] = sha
+    
+    github_api_request("PUT", f"/contents/{path}", payload)
 
 
 def create_pull_request(title: str, head: str, base: str, body: str) -> dict:
@@ -144,21 +166,19 @@ def process_submission(
         return False
     
     data, errors = parse_and_validate(json_str)
-    if errors:
-        error_list = "\n".join(f"- {e}" for e in errors)
+    if errors or data is None:
+        error_list = "\n".join(f"- {e}" for e in errors) if errors else "Unknown error"
         add_issue_comment(issue_number, f"❌ **Validation Failed**\n\n{error_list}")
         return False
     
     # Normalize to list
-    submissions = data if isinstance(data, list) else [data]
+    submissions: list[dict] = data if isinstance(data, list) else [data]
     
     # Generate contributor UUID from GitHub ID
     contributor_uuid = generate_contributor_uuid(author_id)
     
-    # Extract contributor name from issue form (or default to GitHub username)
+    # Extract contributor name from issue form (None means user opted out of attribution)
     contributor_name = extract_contributor_name_from_issue_body(issue_body)
-    if not contributor_name:
-        contributor_name = f"@{author_username}"
     
     # Add metadata to each submission
     now = datetime.now(timezone.utc)
@@ -167,14 +187,15 @@ def process_submission(
     
     for submission in submissions:
         submission["contributor_uuid"] = contributor_uuid
-        submission["contributor_name"] = contributor_name
+        if contributor_name:
+            submission["contributor_name"] = contributor_name
         submission["creation_timestamp"] = timestamp_str
     
     # Generate unique filename
     content_json = json.dumps(submissions, indent=2, sort_keys=True)
     content_hash = compute_content_hash(content_json)
     filename = generate_submission_filename(author_username, date_str, content_hash)
-    file_path = f"community/{filename}"
+    file_path = f"community/{date_str}/{filename}"
     
     # Create branch
     branch_name = f"community-submission-{issue_number}"
@@ -185,14 +206,53 @@ def process_submission(
     commit_message = f"Add community submission from @{author_username} (closes #{issue_number})"
     create_or_update_file(file_path, content_json, commit_message, branch_name)
     
+    # Update schema with any new tags (modifies v1 in place)
+    schema_updated = False
+    new_tags = []
+    try:
+        # Build tag registry from new submissions
+        tag_registry = build_tag_type_registry(submissions)
+        
+        # Get current schema and merge existing tags
+        current_schema = load_schema()
+        existing_tags = get_existing_tag_definitions(current_schema)
+        
+        # Merge existing tags into registry
+        for tag_name, tag_def in existing_tags.items():
+            if tag_name not in tag_registry:
+                tag_type = tag_def.get("type", "string")
+                tag_registry[tag_name] = tag_type
+        
+        # Check for new tags
+        new_tags = check_for_new_tags(tag_registry, current_schema)
+        
+        if new_tags:
+            # Generate updated schema
+            updated_schema = generate_updated_schema(current_schema, tag_registry)
+            schema_json = json.dumps(updated_schema, indent=2) + "\n"
+            
+            create_or_update_file(
+                "schemas/community_submission.v1.schema.json",
+                schema_json,
+                f"Update schema with new tags: {', '.join(new_tags)}",
+                branch_name
+            )
+            schema_updated = True
+    except Exception as e:
+        print(f"Warning: Could not update schema: {e}", file=sys.stderr)
+    
     # Create PR
+    schema_note = ""
+    if schema_updated:
+        schema_note = f"\n**Schema Updated:** Added new tags: `{', '.join(new_tags)}`\n"
+    
     pr_body = f"""## Community Submission
 
 Adds {len(submissions)} submission(s) from @{author_username}.
 
 **File:** `{file_path}`
 **Contributor UUID:** `{contributor_uuid}`
-
+{schema_note}
 Closes #{issue_number}
 
 ---

src/contributions/read_community_data.py

@@ -30,7 +30,8 @@ def read_all_submissions(community_dir: Path | None = None) -> list[dict]:
     
     all_submissions = []
     
-    for json_file in sorted(community_dir.glob("*.json")):
+    # Search both root directory and date subdirectories (e.g., 2026-02-12/)
+    for json_file in sorted(community_dir.glob("**/*.json")):
         try:
             with open(json_file) as f:
                 data = json.load(f)
@@ -50,6 +51,52 @@ def read_all_submissions(community_dir: Path | None = None) -> list[dict]:
     return all_submissions
 
 
+def get_python_type_name(value) -> str:
+    """Get a normalized type name for a value."""
+    if value is None:
+        return "null"
+    if isinstance(value, bool):
+        return "boolean"
+    if isinstance(value, int):
+        return "integer"
+    if isinstance(value, float):
+        return "number"
+    if isinstance(value, str):
+        return "string"
+    if isinstance(value, list):
+        return "array"
+    if isinstance(value, dict):
+        return "object"
+    return type(value).__name__
+
+
+def build_tag_type_registry(submissions: list[dict]) -> dict[str, str]:
+    """
+    Build a registry of tag names to their expected types from existing submissions.
+    
+    Args:
+        submissions: List of existing submission dictionaries
+        
+    Returns:
+        Dict mapping tag name to expected type (e.g., {"internet": "string", "year_built": "integer"})
+    """
+    tag_types = {}
+    
+    for submission in submissions:
+        tags = submission.get("tags", {})
+        if not isinstance(tags, dict):
+            continue
+        
+        for key, value in tags.items():
+            inferred_type = get_python_type_name(value)
+            
+            if key not in tag_types:
+                tag_types[key] = inferred_type
+            # If there's a conflict, keep the first type (it's already in use)
+    
+    return tag_types
+
+
 def group_by_identifier(submissions: list[dict]) -> dict[str, list[dict]]:
     """
     Group submissions by their identifier (registration, transponder, or airframe ID).

src/contributions/regenerate_pr_schema.py

@@ -0,0 +1,66 @@
+#!/usr/bin/env python3
+"""
+Regenerate schema for a PR branch after main has been merged in.
+This script looks at the submission files in this branch and updates
+the schema if new tags were introduced.
+
+Usage: python -m src.contributions.regenerate_pr_schema
+"""
+
+import json
+import sys
+from pathlib import Path
+
+# Add parent to path for imports when running as script
+sys.path.insert(0, str(Path(__file__).parent.parent.parent))
+
+from src.contributions.read_community_data import read_all_submissions, build_tag_type_registry
+from src.contributions.update_schema import (
+    get_existing_tag_definitions,
+    check_for_new_tags,
+    generate_updated_schema,
+)
+from src.contributions.schema import load_schema, SCHEMAS_DIR
+
+
+def main():
+    """Main entry point."""
+    # Load current schema
+    current_schema = load_schema()
+    
+    # Get existing tag definitions from schema
+    existing_tags = get_existing_tag_definitions(current_schema)
+    
+    # Read all submissions (including ones from this PR branch)
+    submissions = read_all_submissions()
+    
+    if not submissions:
+        print("No submissions found")
+        return
+    
+    # Build tag registry from all submissions
+    tag_registry = build_tag_type_registry(submissions)
+    
+    # Check for new tags not in the current schema
+    new_tags = check_for_new_tags(tag_registry, current_schema)
+    
+    if new_tags:
+        print(f"Found new tags: {new_tags}")
+        print("Updating schema...")
+        
+        # Generate updated schema
+        updated_schema = generate_updated_schema(current_schema, tag_registry)
+        
+        # Write updated schema (in place)
+        schema_path = SCHEMAS_DIR / "community_submission.v1.schema.json"
+        with open(schema_path, 'w') as f:
+            json.dump(updated_schema, f, indent=2)
+            f.write("\n")
+        
+        print(f"Updated {schema_path}")
+    else:
+        print("No new tags found, schema is up to date")
+
+
+if __name__ == "__main__":
+    main()

src/contributions/schema.py

@@ -10,12 +10,59 @@ except ImportError:
     Draft202012Validator = None
 
 
-SCHEMA_PATH = Path(__file__).parent.parent.parent / "schemas" / "community_submission.v1.schema.json"
+SCHEMAS_DIR = Path(__file__).parent.parent.parent / "schemas"
+
+# For backwards compatibility
+SCHEMA_PATH = SCHEMAS_DIR / "community_submission.v1.schema.json"
 
 
-def load_schema() -> dict:
-    """Load the community submission schema."""
-    with open(SCHEMA_PATH) as f:
+def get_latest_schema_version() -> int:
+    """
+    Find the latest schema version number.
+    
+    Returns:
+        Latest version number (e.g., 1, 2, 3)
+    """
+    import re
+    pattern = re.compile(r"community_submission\.v(\d+)\.schema\.json$")
+    max_version = 0
+    
+    for path in SCHEMAS_DIR.glob("community_submission.v*.schema.json"):
+        match = pattern.search(path.name)
+        if match:
+            version = int(match.group(1))
+            max_version = max(max_version, version)
+    
+    return max_version
+
+
+def get_schema_path(version: int | None = None) -> Path:
+    """
+    Get path to a specific schema version, or latest if version is None.
+    
+    Args:
+        version: Schema version number, or None for latest
+        
+    Returns:
+        Path to schema file
+    """
+    if version is None:
+        version = get_latest_schema_version()
+    return SCHEMAS_DIR / f"community_submission.v{version}.schema.json"
+
+
+def load_schema(version: int | None = None) -> dict:
+    """
+    Load the community submission schema.
+    
+    Args:
+        version: Schema version to load. If None, loads the latest version.
+        
+    Returns:
+        Schema dict
+    """
+    schema_path = get_schema_path(version)
+    with open(schema_path) as f:
         return json.load(f)
 
 
@@ -50,11 +97,36 @@ def validate_submission(data: dict | list, schema: dict | None = None) -> list[s
     return errors
 
 
+def download_github_attachment(url: str) -> str | None:
+    """
+    Download content from a GitHub attachment URL.
+    
+    Args:
+        url: GitHub attachment URL (e.g., https://github.com/user-attachments/files/...)
+        
+    Returns:
+        File content as string, or None if download failed
+    """
+    import urllib.request
+    import urllib.error
+    
+    try:
+        req = urllib.request.Request(url, headers={"User-Agent": "PlaneQuery-Bot"})
+        with urllib.request.urlopen(req, timeout=30) as response:
+            return response.read().decode("utf-8")
+    except (urllib.error.URLError, urllib.error.HTTPError, UnicodeDecodeError) as e:
+        print(f"Failed to download attachment from {url}: {e}")
+        return None
+
+
 def extract_json_from_issue_body(body: str) -> str | None:
     """
     Extract JSON from GitHub issue body.
     
-    Looks for JSON in the 'Submission JSON' section wrapped in code blocks.
+    Looks for JSON in the 'Submission JSON' section, either:
+    - A GitHub file attachment URL (drag-and-drop .json file)
+    - Wrapped in code blocks (```json ... ``` or ``` ... ```)
+    - Or raw JSON after the header
     
     Args:
         body: The issue body text
@@ -62,13 +134,49 @@ def extract_json_from_issue_body(body: str) -> str | None:
     Returns:
         Extracted JSON string or None if not found
     """
-    # Match JSON in "### Submission JSON" section
-    pattern = r"### Submission JSON\s*\n\s*```(?:json)?\s*\n([\s\S]*?)\n\s*```"
-    match = re.search(pattern, body)
+    # Try: GitHub attachment URL in the Submission JSON section
+    # Format: [filename.json](https://github.com/user-attachments/files/...)
+    # Or just the raw URL
+    pattern_attachment = r"### Submission JSON\s*\n[\s\S]*?(https://github\.com/(?:user-attachments/files|.*?/files)/[^\s\)\]]+\.json)"
+    match = re.search(pattern_attachment, body)
+    if match:
+        url = match.group(1)
+        content = download_github_attachment(url)
+        if content:
+            return content.strip()
     
+    # Also check for GitHub user-attachments URL anywhere in submission section
+    pattern_attachment_alt = r"\[.*?\.json\]\((https://github\.com/[^\)]+)\)"
+    match = re.search(pattern_attachment_alt, body)
+    if match:
+        url = match.group(1)
+        if ".json" in url or "user-attachments" in url:
+            content = download_github_attachment(url)
+            if content:
+                return content.strip()
+    
+    # Try: JSON in code blocks after "### Submission JSON"
+    pattern_codeblock = r"### Submission JSON\s*\n\s*```(?:json)?\s*\n([\s\S]*?)\n\s*```"
+    match = re.search(pattern_codeblock, body)
     if match:
         return match.group(1).strip()
     
+    # Try: Raw JSON after "### Submission JSON" until next section or end
+    pattern_raw = r"### Submission JSON\s*\n\s*([\[{][\s\S]*?[\]}])(?=\n###|\n\n###|$)"
+    match = re.search(pattern_raw, body)
+    if match:
+        return match.group(1).strip()
+    
+    # Try: Any JSON object/array in the body (fallback)
+    pattern_any = r"([\[{][\s\S]*?[\]}])"
+    for match in re.finditer(pattern_any, body):
+        candidate = match.group(1).strip()
+        # Validate it looks like JSON
+        if candidate.startswith('{') and candidate.endswith('}'):
+            return candidate
+        if candidate.startswith('[') and candidate.endswith(']'):
+            return candidate
+    
     return None
 
 

src/contributions/update_schema.py

@@ -0,0 +1,154 @@
+#!/usr/bin/env python3
+"""
+Update the schema with tag type definitions from existing submissions.
+
+This script reads all community submissions and generates a new schema version
+that includes explicit type definitions for all known tags.
+
+When new tags are introduced, a new schema version is created (e.g., v1 -> v2 -> v3).
+
+Usage:
+    python -m src.contributions.update_schema
+    python -m src.contributions.update_schema --check  # Check if update needed
+"""
+import argparse
+import json
+import sys
+from pathlib import Path
+
+from .read_community_data import read_all_submissions, build_tag_type_registry
+from .schema import SCHEMAS_DIR, get_latest_schema_version, get_schema_path, load_schema
+
+
+def get_existing_tag_definitions(schema: dict) -> dict[str, dict]:
+    """Extract existing tag property definitions from schema."""
+    tags_props = schema.get("properties", {}).get("tags", {}).get("properties", {})
+    return tags_props
+
+
+def type_name_to_json_schema(type_name: str) -> dict:
+    """Convert a type name to a JSON Schema type definition."""
+    type_map = {
+        "string": {"type": "string"},
+        "integer": {"type": "integer"},
+        "number": {"type": "number"},
+        "boolean": {"type": "boolean"},
+        "null": {"type": "null"},
+        "array": {"type": "array", "items": {"$ref": "#/$defs/tagScalar"}},
+        "object": {"type": "object", "additionalProperties": {"$ref": "#/$defs/tagScalar"}},
+    }
+    return type_map.get(type_name, {"$ref": "#/$defs/tagValue"})
+
+
+def generate_updated_schema(base_schema: dict, tag_registry: dict[str, str]) -> dict:
+    """
+    Generate an updated schema with explicit tag definitions.
+    
+    Args:
+        base_schema: The current schema to update
+        tag_registry: Dict mapping tag name to type name
+        
+    Returns:
+        Updated schema dict
+    """
+    schema = json.loads(json.dumps(base_schema))  # Deep copy
+    
+    # Build tag properties with explicit types
+    tag_properties = {}
+    for tag_name, type_name in sorted(tag_registry.items()):
+        tag_properties[tag_name] = type_name_to_json_schema(type_name)
+    
+    # Only add/update the properties key within tags, preserve everything else
+    if "properties" in schema and "tags" in schema["properties"]:
+        schema["properties"]["tags"]["properties"] = tag_properties
+    
+    return schema
+
+
+def check_for_new_tags(tag_registry: dict[str, str], current_schema: dict) -> list[str]:
+    """
+    Check which tags in the registry are not yet defined in the schema.
+    
+    Returns:
+        List of new tag names
+    """
+    existing_tags = get_existing_tag_definitions(current_schema)
+    return [tag for tag in tag_registry if tag not in existing_tags]
+
+
+def update_schema_file(
+    tag_registry: dict[str, str],
+    check_only: bool = False
+) -> tuple[bool, list[str]]:
+    """
+    Update the v1 schema file with new tag definitions.
+    
+    Args:
+        tag_registry: Dict mapping tag name to type name
+        check_only: If True, only check if update is needed without writing
+        
+    Returns:
+        Tuple of (was_updated, list_of_new_tags)
+    """
+    current_schema = load_schema()
+    
+    # Find new tags
+    new_tags = check_for_new_tags(tag_registry, current_schema)
+    
+    if not new_tags:
+        return False, []
+    
+    if check_only:
+        return True, new_tags
+    
+    # Generate and write updated schema (in place)
+    updated_schema = generate_updated_schema(current_schema, tag_registry)
+    schema_path = get_schema_path()
+    
+    with open(schema_path, "w") as f:
+        json.dump(updated_schema, f, indent=2)
+        f.write("\n")
+    
+    return True, new_tags
+
+
+def update_schema_from_submissions(check_only: bool = False) -> tuple[bool, list[str]]:
+    """
+    Read all submissions and update the schema if needed.
+    
+    Args:
+        check_only: If True, only check if update is needed without writing
+        
+    Returns:
+        Tuple of (was_updated, list_of_new_tags)
+    """
+    submissions = read_all_submissions()
+    tag_registry = build_tag_type_registry(submissions)
+    return update_schema_file(tag_registry, check_only)
+
+
+def main():
+    parser = argparse.ArgumentParser(description="Update schema with tag definitions")
+    parser.add_argument("--check", action="store_true", help="Check if update needed without writing")
+    
+    args = parser.parse_args()
+    
+    was_updated, new_tags = update_schema_from_submissions(check_only=args.check)
+    
+    if args.check:
+        if was_updated:
+            print(f"Schema update needed. New tags: {', '.join(new_tags)}")
+            sys.exit(1)
+        else:
+            print("Schema is up to date")
+            sys.exit(0)
+    else:
+        if was_updated:
+            print(f"Updated {get_schema_path()}")
+            print(f"Added tags: {', '.join(new_tags)}")
+        else:
+            print("No update needed")
+
+
+if __name__ == "__main__":
+    main()

src/contributions/validate_submission.py

@@ -7,6 +7,7 @@ submissions when issues are opened or edited.
 
 Usage:
     python -m src.contributions.validate_submission --issue-body "..."
+    python -m src.contributions.validate_submission --issue-body-file /path/to/body.txt
     python -m src.contributions.validate_submission --file submission.json
     echo '{"registration_number": "N12345"}' | python -m src.contributions.validate_submission --stdin
     
@@ -23,6 +24,7 @@ import urllib.request
 import urllib.error
 
 from .schema import extract_json_from_issue_body, parse_and_validate, load_schema
+from .read_community_data import read_all_submissions, build_tag_type_registry, get_python_type_name
 
 
 def github_api_request(method: str, endpoint: str, data: dict | None = None) -> dict:
@@ -65,6 +67,40 @@ def remove_issue_label(issue_number: int, label: str) -> None:
         pass  # Label might not exist
 
 
+def validate_tag_consistency(data: dict | list, tag_registry: dict[str, str]) -> list[str]:
+    """
+    Check that tag types in new submissions match existing tag types.
+    
+    Args:
+        data: Single submission dict or list of submissions
+        tag_registry: Dict mapping tag name to expected type
+        
+    Returns:
+        List of error messages. Empty list means validation passed.
+    """
+    errors = []
+    submissions = data if isinstance(data, list) else [data]
+    
+    for i, submission in enumerate(submissions):
+        prefix = f"[{i}] " if len(submissions) > 1 else ""
+        tags = submission.get("tags", {})
+        
+        if not isinstance(tags, dict):
+            continue
+        
+        for key, value in tags.items():
+            actual_type = get_python_type_name(value)
+            
+            if key in tag_registry:
+                expected_type = tag_registry[key]
+                if actual_type != expected_type:
+                    errors.append(
+                        f"{prefix}tags.{key}: expected type '{expected_type}', got '{actual_type}'"
+                    )
+    
+    return errors
+
+
 def validate_and_report(json_str: str, issue_number: int | None = None) -> bool:
     """
     Validate JSON and optionally report to GitHub issue.
@@ -90,6 +126,33 @@ def validate_and_report(json_str: str, issue_number: int | None = None) -> bool:
         
         return False
     
+    # Check tag type consistency against existing submissions
+    if data is not None:
+        try:
+            existing_submissions = read_all_submissions()
+            tag_registry = build_tag_type_registry(existing_submissions)
+            tag_errors = validate_tag_consistency(data, tag_registry)
+            
+            if tag_errors:
+                error_list = "\n".join(f"- {e}" for e in tag_errors)
+                message = (
+                    f"❌ **Tag Type Mismatch**\n\n"
+                    f"Your submission uses tags with types that don't match existing submissions:\n\n"
+                    f"{error_list}\n\n"
+                    f"Please use the same type as existing tags, or use a different tag name."
+                )
+                
+                print(message, file=sys.stderr)
+                
+                if issue_number:
+                    add_issue_comment(issue_number, message)
+                    remove_issue_label(issue_number, "validated")
+                
+                return False
+        except Exception as e:
+            # Don't fail validation if we can't read existing submissions
+            print(f"Warning: Could not check tag consistency: {e}", file=sys.stderr)
+    
     count = len(data) if isinstance(data, list) else 1
     message = f"✅ **Validation Passed**\n\n{count} submission(s) validated successfully against the schema.\n\nA maintainer can approve this submission by adding the `approved` label."
     
@@ -106,6 +169,7 @@ def main():
     parser = argparse.ArgumentParser(description="Validate community submission JSON")
     source_group = parser.add_mutually_exclusive_group(required=True)
     source_group.add_argument("--issue-body", help="Issue body text containing JSON")
+    source_group.add_argument("--issue-body-file", help="File containing issue body text")
     source_group.add_argument("--file", help="JSON file to validate")
     source_group.add_argument("--stdin", action="store_true", help="Read JSON from stdin")
     
@@ -125,6 +189,20 @@ def main():
                     "Please ensure your JSON is in the 'Submission JSON' field wrapped in code blocks."
                 )
             sys.exit(1)
+    elif args.issue_body_file:
+        with open(args.issue_body_file) as f:
+            issue_body = f.read()
+        json_str = extract_json_from_issue_body(issue_body)
+        if not json_str:
+            print("❌ Could not extract JSON from issue body", file=sys.stderr)
+            print(f"Issue body:\n{issue_body}", file=sys.stderr)
+            if args.issue_number:
+                add_issue_comment(
+                    args.issue_number,
+                    "❌ **Validation Failed**\n\nCould not extract JSON from submission. "
+                    "Please ensure your JSON is in the 'Submission JSON' field."
+                )
+            sys.exit(1)
     elif args.file:
         with open(args.file) as f:
             json_str = f.read()