melpomene/zooms_generator.py

# Melpomene webcomic reader JSON/JS/HTML generator
# Version 1.0.0_RC1
# CC-BY-NC-SA https://git.aribaud.net/caribaud/melpomene/

import sys
from argparse import ArgumentParser
from xml.etree import ElementTree
from xml.etree.ElementTree import Element
from typing import Any
from pathlib import Path


HTML_TEMPLATE = Path(__file__).parent / "melpomene.html"
HTML_TO_REPLACE = "<!-- your img tags here, see documentation -->"


def get_val_has_str(elem: Element, attrib: str, filepath: str | Path) -> str:
    value = elem.get(attrib)

    if value is None:
        sys.exit(f"Attribute '{attrib}' is not valid in file {filepath}")

    return str(value)


def extract_zooms(src_folder) -> dict[int, Any]:
    folder = Path(src_folder)

    pages_zooms: dict[int, Any] = {}

    idx = 0

    for svg_path in folder.glob("*.svg"):
        idx += 1

        print(f"page {idx} : {svg_path.name}")

        # Setting up default values
        pages_zooms[idx] = {
            "name": svg_path.stem,
            "width": 0,
            "height": 0,
            "zooms": [],
        }

        tree = ElementTree.parse(svg_path)
        root = tree.getroot()

        width = get_val_has_str(root, "width", svg_path)
        height = get_val_has_str(root, "height", svg_path)

        if "." in width:
            print(
                f"WARNING: file {svg_path} has a floating width, it will be rounded",
                file=sys.stderr,
            )
        pages_zooms[idx]["width"] = round(float(width))

        if "." in height:
            print(
                f"WARNING: file {svg_path} has a floating height, it will be rounded",
                file=sys.stderr,
            )
        pages_zooms[idx]["height"] = round(float(height))

        zooms = []
        for area in root.findall(".//{*}rect"):
            zooms.append(
                [
                    float(get_val_has_str(area, "width", svg_path)),
                    float(get_val_has_str(area, "height", svg_path)),
                    float(get_val_has_str(area, "x", svg_path)),
                    float(get_val_has_str(area, "y", svg_path)),
                ]
            )

        pages_zooms[idx]["zooms"] = zooms

    return pages_zooms


def write_json_or_js(zooms, dest_file, is_js) -> None:
    with open(dest_file, "w", encoding="UTF-8") as data_file:
        if is_js:
            data_file.write("PAGES_ZOOMS = ")
        data_file.write("[\n")
        first_coma_skiped = False
        for page_idx in sorted(zooms.keys()):
            for zoom in zooms[page_idx]["zooms"]:
                if zoom[2] < 0 or zoom[3] < 0:
                    print(
                        f"WARNING: negative pos x / pos y in page {page_idx} for "
                        f"zoom {zoom} (is the rectangle flipped?)"
                    )

                if first_coma_skiped:
                    data_file.write(",\n")
                else:
                    first_coma_skiped = True
                data_file.write(f"    {[page_idx] + zoom}")
        data_file.write("\n]\n")


def write_html(zooms, dest_file, prefix, extention) -> None:
    img_tags = ""
    for page_idx in sorted(zooms.keys()):
        img_url = f"{prefix}{zooms[page_idx]['name']}.{extention}"
        zoom_html_data = [
            ",".join([str(zoom) for zoom in page_zooms])
            for page_zooms in zooms[page_idx]["zooms"]
        ]
        zoom_html_str = ";".join(zoom_html_data)
        img_tags = (
            img_tags
            + "            "
            + f'<img loading="lazy" height="{zooms[page_idx]["height"]}" '
            + f'width="{zooms[page_idx]["width"]}" src="{img_url}" '
            + f'data-zooms="{zoom_html_str}"/>\n'
        )

    img_tags = img_tags.strip()

    with open(HTML_TEMPLATE, "r", encoding="UTF-8") as template_file, open(
        dest_file, "w", encoding="UTF-8"
    ) as data_file:
        data = template_file.read().replace(HTML_TO_REPLACE, img_tags)

        data_file.write(data)


def generate_argparse() -> ArgumentParser:
    """Generate Melpomene's generator input parser"""

    parser = ArgumentParser(
        description=(
            "Helper that can generate JSON / JS / "
            "HTML files for Melpomene webcomic reader"
        )
    )

    parser.add_argument(
        "output_format",
        choices=["html", "json", "js"],
        help="The type of output to generate",
    )
    parser.add_argument("svg_folders", help="Path of the folder containing the SVGs")
    parser.add_argument(
        "-o", metavar="dest_file", help="Where to write the generator output to"
    )
    parser.add_argument(
        "-p",
        default="",
        metavar="img_url_prefix",
        help="What to prefix the URL of the images when using HTML format.",
    )
    parser.add_argument(
        "-e",
        default="png",
        metavar="img_ext",
        help="What extention to use in the URL of the images when using HTML format.",
    )

    return parser


def run():
    args = generate_argparse().parse_args()

    # Get the final outout name
    output = None

    if not args.o:
        output = "melpomene_data"
    else:
        output = args.o

    if args.output_format == "html" and not output.endswith(".html"):
        output += ".html"

    elif args.output_format == "json" and not output.endswith(".json"):
        output += ".json"

    elif args.output_format == "js" and not output.endswith(".js"):
        output += ".js"

    zooms = extract_zooms(args.svg_folders)

    if args.output_format == "html":
        write_html(zooms, output, args.p, args.e)

    elif args.output_format == "json":
        write_json_or_js(zooms, output, False)

    elif args.output_format == "js":
        write_json_or_js(zooms, output, True)


if __name__ == "__main__":
    run()