Add build script

author: David T. Sadler <davidtsadler@googlemail.com> 2021-05-07 17:11:05 +0100
committer: David T. Sadler <davidtsadler@googlemail.com> 2021-05-07 17:11:05 +0100
commit: 6b0d2b51f546a9964f18b58e794ade7bd7096475 (patch)
tree: f27bdb3fd5a04199a9ebc58e5a29588547a0c7c7
parent: c709d0c51f03b3010d8f2f19ee73f96c6a06ffe0 (diff)
4 files changed, 436 insertions, 0 deletions
diff --git a/html_templates/default.html b/html_templates/default.html
new file mode 100644
index 0000000..21abb9a
--- /dev/null
+++ b/html_templates/default.html
@@ -0,0 +1,12 @@
+<!doctype html>
+<html lang="en">
+    <head>
+        <meta charset="utf-8">
+        <meta name="viewport" content="width=device-width, initial-scale=1">
+        <title>{{ $title }}</title>
+        <link rel="shortcut icon" href="/images/favicon.png">
+    </head>
+  <body>
+  	<section>{{ $contents }}</section>
+  </body>
+</html>
diff --git a/scripts/build.php b/scripts/build.php
new file mode 100644
index 0000000..ecb0d9d
--- /dev/null
+++ b/scripts/build.php
@@ -0,0 +1,112 @@
+<?php declare(strict_types=1);
+
+require_once __DIR__.DIRECTORY_SEPARATOR.'functions.php';
+
+$hostname = null;
+$output = null;
+
+$opts = getopt(
+    'h:o:',
+    [
+        'hostname:',
+        'help::',
+        'output:',
+    ]
+);
+
+if ($opts === false) {
+    usage();
+}
+
+foreach($opts as $opt => $value) {
+    if ($value === false) {
+        continue;
+    }
+
+    switch ($opt)
+    {
+        case 'h':
+        case 'hostname':
+            $hostname = $value;
+            break;
+        case 'help':
+            usage();
+            break;
+        case 'o':
+        case 'output':
+            $output = $value;
+            break;
+    }
+}
+
+
+if (!$hostname || !$output) {
+    usage();
+}
+
+if (!is_dir($output)) {
+    echo "Unable to locate specified output directory $output.\n\n";
+    exit(1);
+}
+
+if (!is_writable($output)) {
+    echo "Output directory $output is not writable.\n\n";
+    exit(1);
+}
+
+$geminiOutput = "$output/gemini";
+
+if (file_exists($geminiOutput)) {
+    echo "Gemini build directory $geminiOutput already exists.\n\n";
+    exit(1);
+}
+
+$wwwOutput = "$output/www";
+
+if (file_exists($wwwOutput)) {
+    echo "WWW build directory $wwwOutput already exists.\n\n";
+    exit(1);
+}
+
+$src = __DIR__.DIRECTORY_SEPARATOR.'..'.DIRECTORY_SEPARATOR.'src';
+
+if (!is_dir($src)) {
+    echo "Unable to locate src directory $src.\n\n";
+    exit(1);
+}
+
+if (!is_readable($src)) {
+    echo "Src directory $src is not readable.\n\n";
+    exit(1);
+}
+
+$htmlTemplateDiretory = __DIR__.DIRECTORY_SEPARATOR.'../html_templates';
+
+if (!is_dir($htmlTemplateDiretory)) {
+    echo "Unable to locate html template directory $htmlTemplateDiretory.\n\n";
+    exit(1);
+}
+
+if (!is_readable($htmlTemplateDiretory)) {
+    echo "Src directory $htmlTemplateDiretory is not readable.\n\n";
+    exit(1);
+}
+
+$assetsDiretory = __DIR__.DIRECTORY_SEPARATOR.'../www_assets';
+
+if (!is_dir($assetsDiretory)) {
+    echo "Unable to locate assets directory $assetsDiretory.\n\n";
+    exit(1);
+}
+
+if (!is_readable($assetsDiretory)) {
+    echo "Src directory $assetsDiretorY is not readable.\n\n";
+    exit(1);
+}
+
+$siteMetaData = getSiteMetaData($src);
+
+buildGeminiSite($siteMetaData, $geminiOutput);
+
+buildWWWSite($siteMetaData, $wwwOutput, $htmlTemplateDiretory, $assetsDiretory);
+
diff --git a/scripts/functions.php b/scripts/functions.php
new file mode 100644
index 0000000..de563fc
--- /dev/null
+++ b/scripts/functions.php
@@ -0,0 +1,312 @@
+<?php declare(strict_types=1);
+
+const HEADER = '/^(#+)\s+(.*)/';
+const LINK = '/^=&gt;\s+(\S+)\s*(.*)/';
+const LIST_ITEM = '/^\*\s+(.*)/';
+const PRE = '/^```(.*)?/';
+const QUOTE = '/^&gt;\s+(.*)/';
+
+function usage(): void
+{
+    echo "Usage: php build.php [options]
+
+  -h --hostname <hostname> Hostname of site. Used when generating HTML.
+     --help                Displays this message.
+  -o --output <directory>  Directory where site will be built into.
+
+";
+  exit(1);
+}
+
+function getSiteMetaData(string $src): array
+{
+    $src = realpath($src);
+
+    $iter = new \RecursiveIteratorIterator(
+        new \RecursiveDirectoryIterator($src, \FilesystemIterator::SKIP_DOTS),
+        \RecursiveIteratorIterator::SELF_FIRST
+    );
+
+    $metaData = [];
+
+    foreach ($iter as $fileInfo) {
+        if (!is_file($fileInfo->getRealPath())) {
+            continue;
+        }
+        $metaData[] = getFileMetaData($src, $fileInfo);
+    }
+
+    return $metaData;
+}
+
+function getFileMetaData(string $src, \SplFileInfo $fileInfo): array
+{
+    $input = $fileInfo->getRealPath();
+
+    $url = str_replace($src, '', $input);
+
+    $urlData = parseUrl($url);
+    $contentData = parseContent(file_get_contents($input));
+
+    return [
+        'input'    => $input,
+        'url'      => $url,
+        'date'     => $urlData['date'],
+        'category' => $urlData['category'],
+        'isPost'   => $urlData['isPost'],
+        'title'    => $contentData['title'],
+        'author'   => $contentData['author'],
+    ];
+}
+
+function parseUrl(string $url): array
+{
+    $date = null;
+    $category = null;
+    $isPost = false;
+
+    /**
+     * Assume that only posts have both a date and a category.
+     */
+    if (preg_match('/\/posts\/(.+)\/(\d\d\d\d-\d\d-\d\d)\//', $url, $matches) === 1) {
+        $date = $matches[2];
+        $category = $matches[1];
+        $isPost = true;
+    }
+
+    return [
+        'date'     => $date,
+        'category' => $category,
+        'isPost'   => $isPost,
+    ];
+}
+
+function parseContent(string $content): array
+{
+    $title = null;
+    $author = null;
+
+    if (preg_match('/^# (.+)$/m', $content, $matches) === 1) {
+        $title = $matches[1];
+    }
+
+    if (preg_match('/^> .+ By (.+)$/m', $content, $matches) === 1) {
+        $author = $matches[1];
+    }
+
+    return [
+        'title'  => $title,
+        'author' => $author,
+    ];
+}
+
+function buildGeminiSite(array $siteMetaData, string $output): void
+{
+    foreach ($siteMetaData as $fileMetaData) {
+        $destFile = $output.DIRECTORY_SEPARATOR.$fileMetaData['url'];
+
+        $destDirectory = dirname($destFile);
+
+        if (!file_exists($destDirectory) && !mkdir($destDirectory, 0777, true)) {
+            echo "Unable to create Gemini site directory $destDirectory.";
+            exit(1);
+        }
+
+        copy($fileMetaData['input'], $destFile);
+    }
+}
+
+function buildWWWSite(array $siteMetaData, string $output, string $htmlTemplateDiretory, $assetsDiretory): void
+{
+    foreach ($siteMetaData as $fileMetaData) {
+        $destFile = $output.DIRECTORY_SEPARATOR.$fileMetaData['url'];
+
+        $destDirectory = dirname($destFile);
+
+        if (!file_exists($destDirectory) && !mkdir($destDirectory, 0777, true)) {
+            echo "Unable to create WWW site directory $destDirectory.";
+            exit(1);
+        }
+
+        file_put_contents(
+            str_replace('.gmi', '.html', $destFile),
+            buildHtmlFile(
+                $fileMetaData,
+                gemtext2hmtl(file_get_contents($fileMetaData['input'])),
+                $htmlTemplateDiretory
+            )
+        );
+    }
+
+    copyWWWAssets($assetsDiretory, $output);
+}
+
+function buildHtmlFile(array $fileMetaData, string $contents, string $htmlTemplateDiretory): string
+{
+    return str_replace(
+        [
+            '{{ $title }}',
+            '{{ $contents }}'
+        ],
+        [
+            $fileMetaData['title'],
+            $contents,
+        ],
+        file_get_contents($htmlTemplateDiretory.DIRECTORY_SEPARATOR.'default.html')
+    );
+}
+
+function gemtext2hmtl(string $gemtext): string
+{
+    return renderGemtextTokens(parseGemtext($gemtext));
+}
+
+function parseGemtext(string $gemtext): array
+{
+    $tokens = [];
+
+    $lines = preg_split('/\r?\n/', htmlspecialchars($gemtext));
+
+    $index = 0;
+    $numLines = count($lines);
+    $line = fn($index) => trim($lines[$index]);
+
+    while($index < $numLines) {
+        if (preg_match(HEADER, $line($index), $matches) === 1) {
+            [, $levels, $content] = $matches;
+
+            $tokens[] = [
+                'type'    => HEADER,
+                'level'   => strlen($levels),
+                'content' => $content,
+            ];
+        } elseif (preg_match(LINK, $line($index), $matches) === 1) {
+            [, $href, $content] = $matches;
+
+            $tokens[] = [
+                'type'    => LINK,
+                'href'    => $href,
+                'content' => $content != '' ? $content : $href,
+            ];
+        } elseif (preg_match(LIST_ITEM, $line($index), $matches) === 1) {
+            $items = [];
+
+            while ($index < $numLines) {
+                if (preg_match(LIST_ITEM, $line($index), $matches) === 0) {
+                    break;
+                }
+
+                [, $content] = $matches;
+
+                $items[] = $content;
+
+                $index++;
+
+            }
+
+            $index--;
+
+            $tokens[] = [
+                'type'  => LIST_ITEM,
+                'items' => $items,
+            ];
+        } elseif (preg_match(PRE, $line($index), $matches) === 1) {
+            [, $alt] = $matches;
+
+            $items = [];
+
+            $index++;
+
+            while ($index < $numLines) {
+                $item = $line($index);
+
+                if (preg_match(PRE, $item, $matches) === 1) {
+                    break;
+                }
+
+                $items[] = $item;
+
+                $index++;
+            }
+
+            $tokens[] = [
+                'type'  => PRE,
+                'alt'   => $alt,
+                'items' => $items,
+            ];
+        } elseif (preg_match(QUOTE, $line($index), $matches) === 1) {
+            [, $content] = $matches;
+
+            $tokens[] = [
+                'type'    => QUOTE,
+                'content' => $content,
+            ];
+        } else {
+            $tokens[] = [
+                'type'    => null,
+                'content' => $line($index),
+            ];
+        }
+
+        $index++;
+    }
+
+    return $tokens;
+}
+
+function renderGemtextTokens(array $tokens): string
+{
+    return implode("\n", array_filter(array_map(function ($token) {
+        switch ($token['type']) {
+            case HEADER:
+                return sprintf('<h%s>%s</h%s>', $token['level'], $token['content'], $token['level']);
+            case LINK:
+                return sprintf('<a href="%s">%s</a>', $token['href'], $token['content']);
+            case LIST_ITEM:
+                return sprintf(
+                    "<ul>\n%s\n</ul>",
+                    implode("\n", array_map(function ($item) { return sprintf("<li>%s</li>", $item);}, $token['items']))
+                );
+            case PRE:
+                return $token['alt'] ?
+                    sprintf("<pre><code class=\"%s\">\n%s\n</code></pre>", $token['alt'], implode("\n", $token['items']))
+                    :
+                    sprintf("<pre>\n%s\n</pre>", $token['alt'], implode('', $token['items']));
+            case QUOTE:
+                return sprintf('<blockquote>%s</blockquote>', $token['content']);
+            default:
+                return $token['content'] !== '' ? sprintf('<p>%s</p>', $token['content']) : null;
+        }
+    }, $tokens)));
+}
+
+function copyWWWAssets(string $assetsDiretory, string $output): void
+{
+    $assetsDiretory = realpath($assetsDiretory);
+
+    $iter = new \RecursiveIteratorIterator(
+        new \RecursiveDirectoryIterator($assetsDiretory, \FilesystemIterator::SKIP_DOTS),
+        \RecursiveIteratorIterator::SELF_FIRST
+    );
+
+    foreach ($iter as $fileInfo) {
+        if (!is_file($fileInfo->getRealPath())) {
+            continue;
+        }
+
+        $input = $fileInfo->getRealPath();
+        $url = str_replace($assetsDiretory, '', $input);
+
+        $destFile = $output.DIRECTORY_SEPARATOR.$url;
+
+        $destDirectory = dirname($destFile);
+
+        if (!file_exists($destDirectory) && !mkdir($destDirectory, 0777, true)) {
+            echo "Unable to create WWW asset directory $destDirectory.";
+            exit(1);
+        }
+
+        copy($input, $destFile);
+    }
+}
+
diff --git a/www_assets/images/favicon.png b/www_assets/images/favicon.png
new file mode 100644
index 0000000..4909792
--- /dev/null
+++ b/www_assets/images/favicon.png
author	David T. Sadler <davidtsadler@googlemail.com>	2021-05-07 17:11:05 +0100
committer	David T. Sadler <davidtsadler@googlemail.com>	2021-05-07 17:11:05 +0100
commit	6b0d2b51f546a9964f18b58e794ade7bd7096475 (patch)
tree	f27bdb3fd5a04199a9ebc58e5a29588547a0c7c7
parent	c709d0c51f03b3010d8f2f19ee73f96c6a06ffe0 (diff)