readme_generator.py

from beet import Context, TextFile, configurable
import re
from pathlib import Path
from typing import Any
from gm4.plugins.manifest import write_credits, ManifestConfig

@configurable("gm4", validator=ManifestConfig)
def beet_default(ctx: Context, opts: ManifestConfig):
    """Loads the README.md and modifies:
        - converts local images to URLs pointed at the repo
        - download links for respective download sites
        - pulls credits from beet.yaml and contributors.json
        - pulls YT link from beet.yaml
        - performs some site-specific replacements
        - generates a BBCode translation of the markdown file"""
    readme_path = Path(ctx.project_id) / "README.md"
    if not readme_path.exists():
        return
    global_readme = TextFile(source_path=readme_path)
    global_contents = global_readme.text
    running_readme_gen = ctx.meta.get("readme-gen", False) # used to disable pmc replacements on normal release
    global_replacements: dict[str, str] = {}

    # Append standard footer content
    footer = "\n### More Info\n"
        # Youtube Info
    if opts.video is not None:
        footer += (
            f"[<img src=\"https://raw.githubusercontent.com/Gamemode4Dev/GM4_Datapacks/master/base/images/youtube_logo.png\" "
            f"alt=\"Youtube Logo\" width=\"40\" align=\"center\"/> "
            f"**Watch on Youtube**]({opts.video})"
            "\n\n"
        )

        # Wiki Info
    footer += (
        f"[<img src=\"https://raw.githubusercontent.com/Gamemode4Dev/GM4_Datapacks/master/base/images/gm4_wiki_logo.png\" "
        f"alt=\"Gamemode 4 Wiki Logo\" width=\"40\" align=\"center\"/> "
        f"**Read the Wiki**]({opts.wiki})"
        "\n\n"
    )

    footer += "### Credits\n"

        # Credits
    ctx.require(write_credits) # requires data from traversing credits files
    linked_credits = ctx.meta['linked_credits']
    credits_text = ""
    for title in linked_credits:
        credits_text += f"- {title}: {', '.join(linked_credits[title])}\n"
    footer += credits_text

        # Info Blurb
    footer += (
        "\n---\n"
        "## About Gamemode 4 <img src=\"https://raw.githubusercontent.com/Gamemode4Dev/GM4_Datapacks/master/base/images/gm4_logo.png\""
            " alt=\"Gamemode 4 Logo\" width=\"20\"/>\n"
        "Gamemode 4 is a series of command-powered creations that are designed to change and enhance the survival experience. "
        "All of our modules are designed to work together flawlessly, and are balanced for usage in a survival setting. "
        "Pick and choose your favorites from our [website](https://gm4.co), or wherever you get datapacks."
    )

    global_contents += footer
    
    # Local Images to raw.githubusercontent URLs
    global_replacements.update({
        r"([!<].*?[\"(])(.*?)([\")].*) *<!-- *\$localAssetToURL[ -]+?>":
            f"\\1https://raw.githubusercontent.com/Gamemode4Dev/GM4_Datapacks/master/{ctx.project_id}/\\2\\3"
    })

    # Header Title
    global_replacements.update({
        r'#? *(.+?) *<!-- *\$headerTitle.+?>' : (
            "# <img src=\"https://raw.githubusercontent.com/Gamemode4Dev/GM4_Datapacks/master/base/images/gm4_logo.png\""
                " alt=\"GM4 Logo\" width=\"32\" />"
            r" \1 by Gamemode 4"
        )
    })

    # Recommended modules dynamic linking; to gm4.co, serves as fallback link if modrinth post does not exist
    rec_modules = re.findall(r"\[(.+)\]\(\$dynamicLink:(\w+)\)", global_contents)
    for m in [m for _, m in rec_modules]:
        global_replacements.update({
            f"\\[(.+)\\]\\((\\$dynamicLink:{m})\\)": f"[\\1](https://gm4.co/modules/{m[4:].replace('_','-')})<!--$dynamicLink:{m}-->"
        })

    # Perform global replacements
    for pattern, repl in global_replacements.items():
        global_contents = re.sub(pattern, repl, global_contents)

    # download-site specific edits
    site_replacements: dict[str, dict[str, str]] = {
        "gm4": {},
        "modrinth": {},
        "smithed": {},
        "pmc": {}
    }

    # Remove lingering comments in main readme
    site_replacements["gm4"].update({r"\n?<!--.+?-->": ""})

    # Recommended modules dynamic linking; from gm4.co to modrinth/smithed/pmc
    manifest = ctx.cache["gm4_manifest"].json
    rec_modules = re.findall(r"\(.+\)<!--\$dynamicLink:(.+)-->", global_contents)
        # TODO relative links, if they are better
    for m in rec_modules:
        manifest_m_entry: dict[str, Any] = manifest["modules"].get(m, {})
        if (v:=manifest_m_entry.get('modrinth_id')):
            site_replacements["modrinth"].update({
                f"\\(.+\\)<!--\\$dynamicLink:{m}-->": f"(https://modrinth.com/datapack/{v})"
        })
        if (v:=manifest_m_entry.get('smithed_link')):
            site_replacements["smithed"].update({
                f"\\(.+\\)<!--\\$dynamicLink:{m}-->": f"(https://beta.smithed.dev/packs/{v})" # NOTE links to in-beta browser smithed access
        })
        if running_readme_gen and (v:=manifest_m_entry.get('pmc_link')):
            site_replacements["pmc"].update({
                f"\\(.+\\)<!--\\$dynamicLink:{m}-->": f"(https://planetminecraft.com/data-pack/{v})"
        })

    # Modrinth Youtube Video Embed
    if 'video' in ctx.meta['gm4'] and ctx.meta['gm4']['video'] is not None:
        embed_url = re.sub(r'https:\/\/www.youtube.com\/watch\?v=(\w+)', r'https://www.youtube.com/embed/\1', ctx.meta['gm4']['video'])

        site_replacements["modrinth"].update({
            r"(.+)<!-- *\$modrinth:replaceWithVideo[ -]+?>" : (
                "<iframe\n"
                "width=\"640\"\n"
                "height=\"480\"\n"
                f"src=\"{embed_url}\"\n"
                "frameborder=\"0\"\n"
                "allow=\"autoplay; encrypted-media\"\n"
                "allowfullscreen\n"
                "></iframe>"
            )
        })

    if running_readme_gen:
        # PMC Element Deletion
        site_replacements['pmc'].update({
            r".+<!--\$pmc:delete.+>\n\n": ""
        })

        # Table BBCode conversion
            # tables are gross to autodetect and capture, so we'll cheat with helper tags
        tables = re.findall(r'<!-- *\$pmc:startTable ?-->\n((?:.+\n)+)<!-- *\$pmc:endTable ?-->', global_contents)
        for table in tables:
            repl = "[table]\n"
            for line in table.strip('\n').split('\n'):
                if set(line).issubset(set("| -:")):
                    continue
                repl += "[tr]\n"
                for value in [v for v in line.split('|') if v != ""]:
                    repl += f"\t[td]{value.strip()}[/td]\n"
                repl += "[/tr]"
            repl += "[/table]"

            site_replacements['pmc'].update({
                f'<!-- *\\$pmc:startTable ?-->\n{re.escape(table)}<!-- *\\$pmc:endTable ?-->':
                    repl
            })

        # static BBCode translations
        site_replacements['pmc'].update({
            r"(.+)<!-- *\$pmc:headerSize.+>": r"[size=14px]\1[/size]",
            r"#{2,3} (.+)": r"[style b size=14px]\1[/style]",
            r"!\[(?!size|style|img|url)(.+)\]\((.+)\)": r"[img=\1]\2[/img]",
            r"\[(?!size|style|img|url)(.+?)\]\((.+?)\)": r"[url=\2]\1[/url]",
            r"<img src=\"(.+?)\" alt=(\".+?\") width=\"(.+?)\".+>": r"[img title=\2 width=\3]\1[/img]",
            r"\*\*(.+)\*\*": r"[b]\1[/b]",
            r"\*(.+)\*": r"[i]\1[/i]",
            r"_((?:https[^_\n])+?)_(?![\w])": r"[i]\1[/i]",
            r"__(.+)__": r"[u]\1[/u]",
            r"~~(.+)~~": r"[s]\1[/s]",
            r"`(.+)`": r"\1", # BBCode has no inline code blocks
            r"```(.+)```": r"[code]\1[/code]",
            r"---*\n": r"[hr]",
            r"\n?<!--.+?-->": ""
        })

    # Apply site-specific edits
    for site, replacements in site_replacements.items():
        site_contents = global_contents
        for pattern, repl in replacements.items():
            site_contents = re.sub(pattern, repl, site_contents)

        if site == "gm4":
            ctx.data.extra["README.md"] = TextFile(site_contents)
        else:
            ctx.meta[f'{site}_readme'] = TextFile(site_contents)


def libraries(ctx: Context):
    """Dumps the README (library docs) to the "smithed_readme" meta, so it ends up in the generated folder. No edits are made"""
    ctx.meta['smithed_readme'] = ctx.data.extra["README.md"] # relies on copy_files to load into datapack container.