Berlioz Doc Parser

Berlioz Doc Parser is a PHP library to parse documentation and allow to integrate this on website for example.

Installation

Composer

You can install Berlioz Doc Parser with Composer, it's the recommended installation.

$ composer require berlioz/doc-parser

Dependencies

• PHP ^8.0
• PHP libraries:
  • mbstring
• Packages:
  • berlioz/html-selector
  • berlioz/http-message
  • league/flysystem

To parse files, you need additional package:

• For MarkDown files: league/commonmark
• For reStructuredText files: gregwar/rst

Usage

Library use league/flysystem library to manipulate files. This library permit to use some adapter like Local files, GitLab, Google... that's very useful for documentation source.

Documentation generation

You need to define the parser of you documentation, in the example: Berlioz\DocParser\Parser\Markdown.

use Berlioz\DocParser\DocGenerator;
use Berlioz\DocParser\Parser\Markdown;
use League\Flysystem\Filesystem;
use League\Flysystem\Local\LocalFilesystemAdapter;

$version = '1.0';

$docGenerator = new DocGenerator();
$docGenerator->addParser(new Markdown());
$documentation =
    $docGenerator->handle(
        $version,
        new Filesystem(new LocalFilesystemAdapter('/path-of-project/doc'))
    );

Documentation files

You can access to documentation files easily:

/** @var \Berlioz\DocParser\Doc\Documentation $documentation */
$files = $documentation->getFiles();

And filter the files by type:

use Berlioz\DocParser\Doc\File\FileInterface;
use Berlioz\DocParser\Doc\File\Page;

/** @var \Berlioz\DocParser\Doc\Documentation $documentation */
$pages = $documentation->getFiles(fn(FileInterface $file) => $file instanceof Page);

Two file type exists:

• Berlioz\DocParser\Doc\File\Page: page with parsed content
• Berlioz\DocParser\Doc\File\RawFile: raw file like images

Documentation handle

When your documentation is generate, you can use Documentation::handle() method to get pages and files.

use Berlioz\DocParser\Doc\Documentation;
use Berlioz\DocParser\Doc\File\Page;

/** @var Documentation $documentation */
$file = $documentation->handle('path/of/my/page');

if (null === $file) {
   // Show not found error
}

// Raw file?
if (!$file instanceof Page) {
    // Return \Psr\Http\Message\ResponseInterface object
    return $file->response();
}

// Show HTML content of my page
echo $file->getContents();

Treatments

A concept of treatment is implement to manipulate files after generation.

Some treatments are activate by default:

• Berlioz\DocParser\Treatment\DocSummaryTreatment: generate the summary of documentation
• Berlioz\DocParser\Treatment\PageSummaryTreatment: generate the pages summaries
• Berlioz\DocParser\Treatment\PathTreatment: resolve all links to transform in relative links
• Berlioz\DocParser\Treatment\TitleTreatment: extract title of page and set to Page object

Optionals treatments are available:

• Berlioz\DocParser\Treatment\BootstrapTreatment: add CSS classes to HTML elements to beautify them for Bootstrap framework

You can also create your treatment class and add this to generator:

use Berlioz\DocParser\DocGenerator;
use Berlioz\DocParser\Treatment\BootstrapTreatment;

/** @var DocGenerator $docGenerator */
$docGenerator->addTreatment(new BootstrapTreatment($docGenerator));

Formats

MarkDown

A specific tag is for indexation of documentation.

```index
title: Title of page
slug: Slug name (replace only the filename, not path)
breadcrumb: Category; Sub category; My page
summary-order: 1
summary-visible: true
```

reStructuredText

A specific directive is for indexation of documentation.

...index:
    :title: Title of page
    :slug: Slug name (replace only the filename, not path)
    :breadcrumb: Category; Sub category; My page
    :summary-order: 1
    :summary-visible: true

Options available for indexation

• title: replace the title of page, the title treatment does not replace this title
• slug: replace the filename of page, not the complete path, it's an url encoded representation of value
• breadcrumb: the breadcrumb of page into the documentation summary, if not present, the page will not in the summary
• summary-order: the order in summary section
• summary-visible: default to true, but you can define the page to not visible in summary

You can define your own options, for your treatments for example, options are available with Page::getMeta(...) method.

Summary

Doc summary

Doc summary references all pages with a breadcrumb. It's a hierarchy of Berlioz\DocParser\Doc\Summary\Entry objects.

You can get path of entry with method Entry::getPath(), it's the absolute path from root directory of documentation.

Page summary

Page summary references all headings in the page content. It's a hierarchy of Berlioz\DocParser\Doc\Summary\Entry objects.

No path given to the entry, but an id Entry::getId() linked to the corresponding heading.

Parser

Two parser are available by default:

• Berlioz\DocParser\Parser\Markdown: to parse MarkDown files, use league/commonmark package
• Berlioz\DocParser\Parser\reStructuredText: to parse reStructuredText files, use gregwar/rst package

You can also create your own parser, you need only to implement Berlioz\DocParser\Parser\ParserInterface interface.

If you need to add an extension to a specific parser, a getter method is available to access to the original parser. Example for MarkDown parser: \Berlioz\DocParser\Parser\Markdown::getCommonMarkConverter().

You can also pass to the Markdown and reStructuredText constructors the same parameters that corresponding libraries.

Cache

Generation of documentation take time, and it's not acceptable to generate documentation at every request.

Berlioz\DocParser\DocCacheGenerator class generate a cache of generated and treated documentation to reuse it quickly. The cache use league/flysystem package to store cache files.

Example of usage:

use Berlioz\DocParser\DocCacheGenerator;
use Berlioz\DocParser\DocGenerator;
use Berlioz\DocParser\Parser\Markdown;
use League\Flysystem\Filesystem;
use League\Flysystem\Local\LocalFilesystemAdapter;

$version = '1.0';
$cacheFilesystem = new Filesystem(new LocalFilesystemAdapter('/path-of-project/cache'));
$docCacheGenerator = new DocCacheGenerator($cacheFilesystem);

if (null === ($documentation = $docCacheGenerator->get($version))) {
    $docGenerator = new DocGenerator();
    $docGenerator->addParser(new Markdown());
    $documentation =
        $docGenerator->handle(
            $version,
            new Filesystem(new LocalFilesystemAdapter('/path-of-project/doc'))
        );

    $docCacheGenerator->save($documentation);
}

$documentation->handle('path/of/my/page');