figpack 0.2.27__py3-none-any.whl → 0.2.40__py3-none-any.whl

figpack/__init__.py

@@ -2,7 +2,7 @@
 figpack - A Python package for creating shareable, interactive visualizations in the browser
 """
 
-__version__ = "0.2.27"
+__version__ = "0.2.40"
 
 from .cli import view_figure
 from .core import FigpackView, FigpackExtension, ExtensionView

figpack/cli.py

@@ -18,6 +18,7 @@ import requests
 from . import __version__
 from .core._server_manager import CORSRequestHandler
 from .core._view_figure import serve_files, view_figure
+from .core._upload_bundle import _upload_bundle, get_figure_by_source_url
 from .extensions import ExtensionManager
 
 MAX_WORKERS_FOR_DOWNLOAD = 16
@@ -249,6 +250,188 @@ def handle_extensions_command(args):
         print("Use 'figpack extensions <command> --help' for more information.")
 
 
+def handle_upload_from_source_url(args) -> None:
+    """
+    Handle the upload-from-source-url command
+
+    Downloads a tar.gz/tgz file from a URL, extracts it, and uploads it as a new figure
+    with the source URL set.
+    """
+    import os
+
+    source_url = args.source_url
+    title = args.title if hasattr(args, "title") else None
+
+    # Get API key from environment variable
+    api_key = os.environ.get("FIGPACK_API_KEY")
+    if not api_key:
+        print(
+            "Error: FIGPACK_API_KEY environment variable must be set to upload figures."
+        )
+        sys.exit(1)
+
+    # Validate URL format
+    if not (source_url.endswith(".tar.gz") or source_url.endswith(".tgz")):
+        print(f"Error: Source URL must point to a .tar.gz or .tgz file: {source_url}")
+        sys.exit(1)
+
+    print(f"Downloading archive from: {source_url}")
+
+    try:
+        # Download the archive
+        response = requests.get(source_url, timeout=120, stream=True)
+        response.raise_for_status()
+
+        # Create temporary file for the archive
+        with tempfile.NamedTemporaryFile(
+            suffix=".tar.gz", delete=False
+        ) as temp_archive:
+            archive_path = temp_archive.name
+
+            # Download with progress indication
+            total_size = int(response.headers.get("content-length", 0))
+            downloaded_size = 0
+
+            for chunk in response.iter_content(chunk_size=8192):
+                if chunk:
+                    temp_archive.write(chunk)
+                    downloaded_size += len(chunk)
+                    if total_size > 0:
+                        progress = (downloaded_size / total_size) * 100
+                        print(
+                            f"Downloaded: {downloaded_size / (1024*1024):.2f} MB ({progress:.1f}%)",
+                            end="\r",
+                        )
+
+            if total_size > 0:
+                print()  # New line after progress
+            print(f"Download complete: {downloaded_size / (1024*1024):.2f} MB")
+
+        # Extract archive to temporary directory
+        print("Extracting archive...")
+        with tempfile.TemporaryDirectory() as extract_dir:
+            extract_path = pathlib.Path(extract_dir)
+
+            try:
+                with tarfile.open(archive_path, "r:gz") as tar:
+                    tar.extractall(path=extract_path)
+                print(f"Extracted to temporary directory")
+            except Exception as e:
+                print(f"Error: Failed to extract archive: {e}")
+                sys.exit(1)
+            finally:
+                # Clean up downloaded archive
+                import os
+
+                try:
+                    os.unlink(archive_path)
+                except Exception:
+                    pass
+
+            # Upload the extracted files
+            print(f"Uploading figure with source URL: {source_url}")
+            try:
+                figure_url = _upload_bundle(
+                    str(extract_path),
+                    api_key=api_key,
+                    title=title,
+                    source_url=source_url,
+                )
+                print(f"\nFigure uploaded successfully!")
+                print(f"Figure URL: {figure_url}")
+            except Exception as e:
+                print(f"Error: Failed to upload figure: {e}")
+                sys.exit(1)
+
+    except requests.exceptions.RequestException as e:
+        print(f"Error: Failed to download archive from {source_url}: {e}")
+        sys.exit(1)
+    except Exception as e:
+        print(f"Error: {e}")
+        sys.exit(1)
+
+
+def handle_find_by_source_url(args) -> None:
+    """
+    Handle the find-by-source-url command
+
+    Queries the API for a figure URL by its source URL.
+    """
+    source_url = args.source_url
+
+    print(f"Querying for figure with source URL: {source_url}")
+
+    try:
+        figure_url = get_figure_by_source_url(source_url)
+
+        if figure_url:
+            print(f"Figure found: {figure_url}")
+        else:
+            print(f"No figure found with source URL: {source_url}")
+            sys.exit(1)
+    except Exception as e:
+        print(f"Error: {e}")
+        sys.exit(1)
+
+
+def download_and_view_archive(url: str, port: int = None) -> None:
+    """
+    Download a tar.gz/tgz archive from a URL and view it
+
+    Args:
+        url: URL to the tar.gz or tgz file
+        port: Optional port number to serve on
+    """
+    if not (url.endswith(".tar.gz") or url.endswith(".tgz")):
+        print(f"Error: URL must point to a .tar.gz or .tgz file: {url}")
+        sys.exit(1)
+
+    print(f"Downloading archive from: {url}")
+
+    try:
+        response = requests.get(url, timeout=60, stream=True)
+        response.raise_for_status()
+
+        # Create a temporary file to store the downloaded archive
+        with tempfile.NamedTemporaryFile(suffix=".tar.gz", delete=False) as temp_file:
+            temp_path = temp_file.name
+
+            # Download with progress indication
+            total_size = int(response.headers.get("content-length", 0))
+            downloaded_size = 0
+
+            for chunk in response.iter_content(chunk_size=8192):
+                if chunk:
+                    temp_file.write(chunk)
+                    downloaded_size += len(chunk)
+                    if total_size > 0:
+                        progress = (downloaded_size / total_size) * 100
+                        print(
+                            f"Downloaded: {downloaded_size / (1024*1024):.2f} MB ({progress:.1f}%)",
+                            end="\r",
+                        )
+
+            if total_size > 0:
+                print()  # New line after progress
+            print(f"Download complete: {downloaded_size / (1024*1024):.2f} MB")
+
+        # Now view the downloaded file
+        try:
+            view_figure(temp_path, port=port)
+        finally:
+            # Clean up the temporary file after viewing
+            import os
+
+            try:
+                os.unlink(temp_path)
+            except Exception:
+                pass
+
+    except requests.exceptions.RequestException as e:
+        print(f"Error: Failed to download archive from {url}: {e}")
+        sys.exit(1)
+
+
 def main():
     """Main CLI entry point"""
     parser = argparse.ArgumentParser(
@@ -270,7 +453,7 @@ def main():
     view_parser = subparsers.add_parser(
         "view", help="Extract and serve a figure archive locally"
     )
-    view_parser.add_argument("archive", help="Path to the tar.gz archive file")
+    view_parser.add_argument("archive", help="Path or URL to the tar.gz archive file")
     view_parser.add_argument(
         "--port", type=int, help="Port number to serve on (default: auto-select)"
     )
@@ -312,14 +495,43 @@ def main():
         "extensions", nargs="+", help="Extension package names to uninstall"
     )
 
+    # Upload from URL command
+    upload_from_source_url_parser = subparsers.add_parser(
+        "upload-from-source-url",
+        help="Download a tar.gz/tgz file and upload it as a new figure",
+    )
+    upload_from_source_url_parser.add_argument(
+        "source_url",
+        help="URL to the tar.gz or tgz file (will be set as the figure's source URL)",
+    )
+    upload_from_source_url_parser.add_argument(
+        "--title", help="Optional title for the figure"
+    )
+
+    # Find by source URL command
+    find_by_source_url_parser = subparsers.add_parser(
+        "find-by-source-url", help="Get the figure URL for a given source URL"
+    )
+    find_by_source_url_parser.add_argument(
+        "source_url", help="The source URL to search for"
+    )
+
     args = parser.parse_args()
 
     if args.command == "download":
         download_figure(args.figure_url, args.dest)
     elif args.command == "view":
-        view_figure(args.archive, port=args.port)
+        # Check if archive argument is a URL
+        if args.archive.startswith("http://") or args.archive.startswith("https://"):
+            download_and_view_archive(args.archive, port=args.port)
+        else:
+            view_figure(args.archive, port=args.port)
     elif args.command == "extensions":
         handle_extensions_command(args)
+    elif args.command == "upload-from-source-url":
+        handle_upload_from_source_url(args)
+    elif args.command == "find-by-source-url":
+        handle_find_by_source_url(args)
     else:
         parser.print_help()
 

figpack/core/_bundle_utils.py

@@ -1,7 +1,7 @@
 import os
 import pathlib
 import json
-from typing import Set
+from typing import Optional, List
 
 import zarr
 
@@ -9,12 +9,13 @@ from .figpack_view import FigpackView
 from .figpack_extension import FigpackExtension
 from .extension_view import ExtensionView
 from .zarr import Group, _check_zarr_version
+from ._zarr_consolidate import consolidate_zarr_chunks
 
 thisdir = pathlib.Path(__file__).parent.resolve()
 
 
 def prepare_figure_bundle(
-    view: FigpackView, tmpdir: str, *, title: str, description: str = None
+    view: FigpackView, tmpdir: str, *, title: str, description: Optional[str] = None
 ) -> None:
     """
     Prepare a figure bundle in the specified temporary directory.
@@ -51,8 +52,8 @@ def prepare_figure_bundle(
     # because we only support version 2 on the frontend right now.
 
     if _check_zarr_version() == 3:
-        old_default_zarr_format = zarr.config.get("default_zarr_format")
-        zarr.config.set({"default_zarr_format": 2})
+        old_default_zarr_format = zarr.config.get("default_zarr_format")  # type: ignore
+        zarr.config.set({"default_zarr_format": 2})  # type: ignore
 
     try:
         # Write the view data to the Zarr group
@@ -72,15 +73,18 @@ def prepare_figure_bundle(
         # Generate extension manifest
         _write_extension_manifest(required_extensions, tmpdir)
 
+        # Create the .zmetadata file
         zarr.consolidate_metadata(zarr_group._zarr_group.store)
 
         # It's important that we remove all the metadata files except for the
-        # consolidated one, because otherwise we may get inconstencies
-        # once we start editing the zarr data from the browser.
+        # consolidated one so there is a single source of truth.
         _remove_metadata_files_except_consolidated(pathlib.Path(tmpdir) / "data.zarr")
+
+        # Consolidate zarr chunks into larger files to reduce upload count
+        consolidate_zarr_chunks(pathlib.Path(tmpdir) / "data.zarr")
     finally:
         if _check_zarr_version() == 3:
-            zarr.config.set({"default_zarr_format": old_default_zarr_format})
+            zarr.config.set({"default_zarr_format": old_default_zarr_format})  # type: ignore
 
 
 def _remove_metadata_files_except_consolidated(zarr_dir: pathlib.Path) -> None:
@@ -107,7 +111,7 @@ def _remove_metadata_files_except_consolidated(zarr_dir: pathlib.Path) -> None:
                     print(f"Warning: could not remove file {file_path}: {e}")
 
 
-def _discover_required_extensions(view: FigpackView) -> Set[str]:
+def _discover_required_extensions(view: FigpackView) -> List[str]:
     """
     Recursively discover all extensions required by a view and its children
 

figpack/core/_file_handler.py

@@ -42,6 +42,9 @@ class FileUploadCORSRequestHandler(CORSRequestHandler):
                 "Accept-Ranges, Content-Encoding, Content-Length, Content-Range",
             )
 
+        # Always send Accept-Ranges header to indicate byte-range support
+        self.send_header("Accept-Ranges", "bytes")
+
         # Prevent browser caching - important for when we are editing figures in place
         # This ensures the browser always fetches the latest version of files
         self.send_header("Cache-Control", "no-cache, no-store, must-revalidate")
@@ -187,6 +190,6 @@ class FileUploadCORSRequestHandler(CORSRequestHandler):
             self.send_error(500, f"Internal Server Error: {str(e)}")
             return False
 
-    def log_message(self, fmt, *args):
+    def log_message(self, format, *args):
         """Override to suppress default logging (same as parent class)."""
         pass

figpack/core/_save_figure.py

@@ -5,7 +5,9 @@ from ._bundle_utils import prepare_figure_bundle
 from .figpack_view import FigpackView
 
 
-def _save_figure(view: FigpackView, output_path: str, *, title: str):
+def _save_figure(
+    view: FigpackView, output_path: str, *, title: str, description: str = ""
+) -> None:
     """
     Save the figure to a folder or a .tar.gz file
 
@@ -13,19 +15,21 @@ def _save_figure(view: FigpackView, output_path: str, *, title: str):
         view: FigpackView instance to save
         output_path: Output path (destination folder or .tar.gz file path)
     """
-    output_path = pathlib.Path(output_path)
-    if (output_path.suffix == ".gz" and output_path.suffixes[-2] == ".tar") or (
-        output_path.suffix == ".tgz"
+    output_path_2 = pathlib.Path(output_path)
+    if (output_path_2.suffix == ".gz" and output_path_2.suffixes[-2] == ".tar") or (
+        output_path_2.suffix == ".tgz"
     ):
         # It's a .tar.gz file
         with tempfile.TemporaryDirectory(prefix="figpack_save_") as tmpdir:
-            prepare_figure_bundle(view, tmpdir, title=title)
+            prepare_figure_bundle(view, tmpdir, title=title, description=description)
             # Create tar.gz file
             import tarfile
 
-            with tarfile.open(output_path, "w:gz") as tar:
+            with tarfile.open(output_path_2, "w:gz") as tar:
                 tar.add(tmpdir, arcname=".")
     else:
         # It's a folder
-        output_path.mkdir(parents=True, exist_ok=True)
-        prepare_figure_bundle(view, str(output_path), title=title)
+        output_path_2.mkdir(parents=True, exist_ok=True)
+        prepare_figure_bundle(
+            view, str(output_path_2), title=title, description=description
+        )

figpack/core/_server_manager.py

@@ -29,6 +29,9 @@ class CORSRequestHandler(SimpleHTTPRequestHandler):
                 "Accept-Ranges, Content-Encoding, Content-Length, Content-Range",
             )
 
+        # Always send Accept-Ranges header to indicate byte-range support
+        self.send_header("Accept-Ranges", "bytes")
+
         # Prevent browser caching - important for when we are editing figures in place
         # This ensures the browser always fetches the latest version of files
         self.send_header("Cache-Control", "no-cache, no-store, must-revalidate")
@@ -45,7 +48,100 @@ class CORSRequestHandler(SimpleHTTPRequestHandler):
         """Reject PUT requests when file upload is not enabled."""
         self.send_error(405, "Method Not Allowed")
 
-    def log_message(self, fmt, *args):
+    def do_GET(self):
+        """Handle GET requests with support for Range requests."""
+        # Translate path and check if file exists
+        path = self.translate_path(self.path)
+
+        # Check if path is a file
+        if not os.path.isfile(path):
+            # Let parent class handle directories and 404s
+            return super().do_GET()
+
+        # Check for Range header
+        range_header = self.headers.get("Range")
+
+        if range_header is None:
+            # No range request, use parent's implementation
+            return super().do_GET()
+
+        # Parse range header
+        try:
+            # Range header format: "bytes=start-end"
+            if not range_header.startswith("bytes="):
+                # Invalid range format, ignore and serve full file
+                return super().do_GET()
+
+            range_spec = range_header[6:]  # Remove "bytes=" prefix
+
+            # Get file size
+            file_size = os.path.getsize(path)
+
+            # Parse range specification
+            if "-" not in range_spec:
+                # Invalid format
+                self.send_error(400, "Invalid Range header")
+                return
+
+            range_parts = range_spec.split("-", 1)
+
+            # Determine start and end positions
+            if range_parts[0]:  # Start position specified
+                start = int(range_parts[0])
+                if range_parts[1]:  # End position also specified
+                    end = int(range_parts[1])
+                else:  # Open-ended range (e.g., "1024-")
+                    end = file_size - 1
+            else:  # Suffix range (e.g., "-500" means last 500 bytes)
+                if not range_parts[1]:
+                    self.send_error(400, "Invalid Range header")
+                    return
+                suffix_length = int(range_parts[1])
+                start = max(0, file_size - suffix_length)
+                end = file_size - 1
+
+            # Validate range
+            if start < 0 or end >= file_size or start > end:
+                self.send_response(416, "Range Not Satisfiable")
+                self.send_header("Content-Range", f"bytes */{file_size}")
+                self.end_headers()
+                return
+
+            # Calculate content length
+            content_length = end - start + 1
+
+            # Guess content type
+            import mimetypes
+
+            content_type = mimetypes.guess_type(path)[0] or "application/octet-stream"
+
+            # Send 206 Partial Content response
+            self.send_response(206, "Partial Content")
+            self.send_header("Content-Type", content_type)
+            self.send_header("Content-Length", str(content_length))
+            self.send_header("Content-Range", f"bytes {start}-{end}/{file_size}")
+            self.end_headers()
+
+            # Send the requested byte range
+            with open(path, "rb") as f:
+                f.seek(start)
+                remaining = content_length
+                while remaining > 0:
+                    chunk_size = min(8192, remaining)
+                    chunk = f.read(chunk_size)
+                    if not chunk:
+                        break
+                    self.wfile.write(chunk)
+                    remaining -= len(chunk)
+
+        except ValueError:
+            # Invalid range values
+            self.send_error(400, "Invalid Range header")
+        except Exception as e:
+            # Log error and return 500
+            self.send_error(500, f"Internal Server Error: {str(e)}")
+
+    def log_message(self, format, *args):
         pass
 
 
@@ -189,6 +285,7 @@ class ProcessServerManager:
             and self._server_thread.is_alive()
             and (allow_origin is None or self._allow_origin == allow_origin)
         ):
+            assert self._port is not None
             return f"http://localhost:{self._port}", self._port
 
         # Stop existing server if settings are incompatible
@@ -209,7 +306,7 @@ class ProcessServerManager:
         if enable_file_upload:
             from ._file_handler import FileUploadCORSRequestHandler
 
-            def handler_factory(*args, **kwargs):
+            def handler_factory_enable_upload(*args, **kwargs):
                 return FileUploadCORSRequestHandler(
                     *args,
                     directory=str(temp_dir),
@@ -219,6 +316,11 @@ class ProcessServerManager:
                     **kwargs,
                 )
 
+            assert port is not None
+            self._server = ThreadingHTTPServer(
+                ("0.0.0.0", port), handler_factory_enable_upload
+            )
+
         else:
 
             def handler_factory(*args, **kwargs):
@@ -226,7 +328,8 @@ class ProcessServerManager:
                     *args, directory=str(temp_dir), allow_origin=allow_origin, **kwargs
                 )
 
-        self._server = ThreadingHTTPServer(("0.0.0.0", port), handler_factory)
+            assert port is not None
+            self._server = ThreadingHTTPServer(("0.0.0.0", port), handler_factory)
         self._port = port
         self._allow_origin = allow_origin
 

figpack/core/_show_view.py

@@ -19,7 +19,7 @@ def _is_in_notebook() -> bool:
     """
     try:
         # Check if IPython is available and we're in a notebook
-        from IPython import get_ipython
+        from IPython import get_ipython  # type: ignore
 
         ipython = get_ipython()
         if ipython is None: