mkdocs_llmstxt ¤

def on_config(self, config: MkDocsConfig) -> MkDocsConfig | None:
    """Save the global MkDocs configuration.

    Hook for the [`on_config` event](https://www.mkdocs.org/user-guide/plugins/#on_config).
    In this hook, we save the global MkDocs configuration into an instance variable,
    to re-use it later.

    Arguments:
        config: The MkDocs config object.

    Returns:
        The same, untouched config.
    """
    if config.site_url is None:
        raise ValueError("'site_url' must be set in the MkDocs configuration to be used with the 'llmstxt' plugin")
    self.mkdocs_config = config

    # A `defaultdict` could be used, but we need to retain the same order between `config.sections` and `md_pages`
    # (which wouldn't be guaranteed when filling `md_pages` in `on_page_content()`).
    self.md_pages = {section: [] for section in self.config.sections}
    return config

def on_files(self, files: Files, *, config: MkDocsConfig) -> Files | None:  # noqa: ARG002
    """Expand inputs for generated files.

    Hook for the [`on_files` event](https://www.mkdocs.org/user-guide/plugins/#on_files).
    In this hook we expand inputs for generated file (glob patterns using `*`).

    Parameters:
        files: The collection of MkDocs files.
        config: The MkDocs configuration.

    Returns:
        Modified collection or none.
    """
    page_uris = list(files.src_uris)
    self._sections = {
        section_name: self._expand_inputs(file_list, page_uris=page_uris)  # type: ignore[arg-type]
        for section_name, file_list in self.config.sections.items()
    }
    return files

def on_page_content(self, html: str, *, page: Page, **kwargs: Any) -> str | None:  # noqa: ARG002
    """Convert page content into a Markdown file and save the result to be processed in the `on_post_build` hook.

    Hook for the [`on_page_content` event](https://www.mkdocs.org/user-guide/plugins/#on_page_content).

    Parameters:
        html: The rendered HTML.
        page: The page object.
    """
    src_uri = page.file.src_uri
    for section_name, files in self._sections.items():
        if src_uri in files:
            path_md = Path(page.file.abs_dest_path).with_suffix(".md")
            page_md = _generate_page_markdown(
                html,
                should_autoclean=self.config.autoclean,
                preprocess=self.config.preprocess,
                path=str(path_md),
            )

            md_url = Path(page.file.dest_uri).with_suffix(".md").as_posix()
            # Apply the same logic as in the `Page.url` property.
            if md_url in (".", "./"):
                md_url = ""

            # Guaranteed to exist as we require `site_url` to be configured.
            base = cast("str", self.mkdocs_config.site_url)
            if not base.endswith("/"):
                base += "/"
            md_url = urljoin(base, md_url)

            self.md_pages[section_name].append(
                _MDPageInfo(
                    title=page.title if page.title is not None else src_uri,
                    path_md=path_md,
                    md_url=md_url,
                    content=page_md,
                    description=files[src_uri],
                ),
            )

    return html

def on_post_build(self, *, config: MkDocsConfig, **kwargs: Any) -> None:  # noqa: ARG002
    """Create the final `llms.txt` file and the MD files for all selected pages.

    Hook for the [`on_post_build` event](https://www.mkdocs.org/user-guide/plugins/#on_post_build).

    Parameters:
        config: MkDocs configuration.
    """
    output_file = Path(config.site_dir).joinpath("llms.txt")
    output_file.parent.mkdir(parents=True, exist_ok=True)
    markdown = f"# {config.site_name}\n\n"

    if config.site_description is not None:
        markdown += f"> {config.site_description}\n\n"

    if self.config.markdown_description is not None:
        markdown += f"{self.config.markdown_description}\n\n"

    full_markdown = markdown

    for section_name, file_list in self.md_pages.items():
        markdown += f"## {section_name}\n\n"
        for page_title, path_md, md_url, content, desc in file_list:
            path_md.write_text(content, encoding="utf8")
            _logger.debug(f"Generated MD file to {path_md}")
            markdown += f"- [{page_title}]({md_url}){(': ' + desc) if desc else ''}\n"
        markdown += "\n"

    output_file.write_text(markdown, encoding="utf8")
    _logger.debug("Generated file /llms.txt")

    if self.config.full_output is not None:
        full_output_file = Path(config.site_dir).joinpath(self.config.full_output)
        for section_name, file_list in self.md_pages.items():
            list_content = "\n".join(info.content for info in file_list)
            full_markdown += f"# {section_name}\n\n{list_content}"
        full_output_file.write_text(full_markdown, encoding="utf8")
        _logger.debug(f"Generated file /{self.config.full_output}.txt")