ETPlot/generate_gallery.py

"""
Scientific Gallery Generator

This module generates static HTML galleries from scientific plot collections.
It converts PDF plots to PNG thumbnails, creates responsive web interfaces,
and organizes plots into hierarchical directory structures.

Features:
- PDF to PNG conversion with configurable DPI
- Incremental updates (only converts when source is newer)
- Jinja2 templating for consistent HTML generation
- Support for nested folder structures
- Responsive grid layout with search and navigation
"""

import shutil
from pathlib import Path
from typing import Dict, Any, Optional
from datetime import datetime
from jinja2 import Environment, FileSystemLoader

from utils.config import Config
from utils.metadata import (
    load_folder_metadata,
    merge_metadata,
    save_metadata_cache,
)
from utils.processing import (
    process_plot_files,
    needs_update,
    render_gallery_page,
)


CONFIG = Config.from_yaml("config.yaml")


def datetime_from_timestamp(timestamp: float) -> datetime:
    """Convert a Unix timestamp to a datetime object."""
    return datetime.fromtimestamp(timestamp)


def strftime_filter(dt: datetime, fmt: str) -> str:
    """Format a datetime object using strftime."""
    return dt.strftime(fmt)


env = Environment(loader=FileSystemLoader("."))
env.filters['datetime_from_timestamp'] = datetime_from_timestamp
env.filters['strftime'] = strftime_filter
template = env.get_template("templates/gallery.html")


def build_gallery(
    source_dir: Path,
    web_dir: Path,
    relative_path: Path = None,
    inherited_metadata: Optional[Dict[str, Any]] = None,
) -> None:
    """
    Recursively build gallery structure from source directory.

    Processes all PDF files in the source directory, converts them to PNG,
    copies both to the web directory, and generates index.html files with
    navigation and thumbnails. Now includes metadata support.

    Args:
        source_dir: Source directory containing PDF files
        web_dir: Target web directory for gallery output
        relative_path: Relative path from gallery root (for navigation)
        inherited_metadata: Metadata inherited from parent directories
    """
    if relative_path is None:
        relative_path = Path(".")

    if inherited_metadata is None:
        inherited_metadata = {}

    folder_metadata = load_folder_metadata(source_dir)
    current_metadata = merge_metadata(inherited_metadata, folder_metadata)

    # Find all plot files (both PDF and HTML)
    pdf_files = list(source_dir.glob("*.pdf"))
    html_files = list(source_dir.glob("*.html"))
    plot_files = pdf_files + html_files

    items = []
    plot_metadata_cache = {}

    # Process all plot files (PDFs and HTMLs)
    for plot_file in plot_files:
        item = process_plot_files(
            CONFIG=CONFIG,
            plot_file=plot_file,
            web_dir=web_dir,
            current_metadata=current_metadata,
        )
        items.append(item)
        plot_metadata_cache[plot_file.stem] = item["metadata"]

    save_metadata_cache(web_dir, plot_metadata_cache)

    subdirs = [d for d in source_dir.iterdir() if d.is_dir()]
    subdir_names = []
    for subdir in subdirs:
        subdir_web = web_dir / subdir.name
        subdir_web.mkdir(exist_ok=True)
        subdir_relative = relative_path / subdir.name
        build_gallery(subdir, subdir_web, subdir_relative, current_metadata)
        subdir_names.append(subdir.name)

    render_gallery_page(
        CONFIG=CONFIG,
        template=template,
        web_dir=web_dir,
        items=items,
        subdirs=subdir_names,
        relative_path=relative_path,
        metadata=current_metadata
    )


def main(clean_first: bool = False) -> None:
    """
    Main entry point for gallery generation.

    Args:
        clean_first: If True, removes and recreates the gallery directory

    Processes all configured sources and generates the complete gallery
    structure in the web directory. Ensures assets are available.
    """
    gallery_root = Path(CONFIG.web_folder) / CONFIG.plot_root

    if clean_first and gallery_root.exists():
        print(f"Cleaning gallery directory {gallery_root}...")
        shutil.rmtree(gallery_root)

    gallery_root.mkdir(parents=True, exist_ok=True)
    assets_src = Path("assets")
    assets_dst = gallery_root.parent / "assets"

    if assets_src.exists():
        main_css_src = assets_src / "css" / "main.css"
        main_css_dst = assets_dst / "css" / "main.css"
        if not assets_dst.exists() or needs_update(main_css_src, main_css_dst):
            if assets_dst.exists():
                shutil.rmtree(assets_dst)
            shutil.copytree(assets_src, assets_dst)
            print(f"Updated assets from {assets_src} to {assets_dst}")
    else:
        print(f"Warning: Assets directory {assets_src} not found")

    source_subdirs = []

    for source in CONFIG.sources:
        source_path = Path(source.path)
        source_web_dir = gallery_root / source.name
        source_web_dir.mkdir(parents=True, exist_ok=True)
        source_subdirs.append(source.name)

        if source_path.is_file() and source_path.suffix == '.pdf':
            item = process_plot_files(
                CONFIG=CONFIG,
                plot_file=source_path,
                web_dir=source_web_dir,
            )
            render_gallery_page(
                CONFIG=CONFIG,
                template=template,
                web_dir=source_web_dir,
                items=[item],
                subdirs=[],
                relative_path=Path(source.name)
            )
        elif source_path.is_dir():
            build_gallery(source_path, source_web_dir, Path(source.name))
        else:
            print(
                f"Warning: Source {source.path} is neither a "
                f"directory nor a PDF file. Skipping."
            )

        print(f"Processed {source.name}: {source.path}")

    render_gallery_page(
        CONFIG=CONFIG,
        template=template,
        web_dir=gallery_root,
        items=[],
        subdirs=source_subdirs,
        relative_path=Path("."),
        title="Gallery Root"
    )


if __name__ == "__main__":
    import argparse

    parser = argparse.ArgumentParser(description='Generate gallery')
    parser.add_argument(
        '--clean',
        action='store_true',
        help='Clean gallery directory before generation'
    )
    args = parser.parse_args()
    main(clean_first=args.clean)