docs.php

#!/usr/bin/env php
<?php
/* SPDX-License-Identifier: GPL-3.0-or-later */
 
date_default_timezone_set('UTC');
 
const HELP = <<<EOF
USAGE: php ./scripts/docs.php VERB [ARGS]
VERBS:
 build <lang> Build HTML from TeX using Pandoc for the given language.
 rebase Extract strings from the TeX files and re-create the base
 translation file.
 update <lang> Rebuild HTML from strings.xml for the given language.
 deploy [force] Rebuild HTML and deploy it to the GitHub pages.
 pdf Build PDF from TeX using pdflatex (English-only).
 debug Do experiments.
 
DEPENDENCIES: Pandoc, pandoc-crossref, minify
 
EOF;
 
require_once __DIR__ . "/utils.php";
 
const MAIN_TEX = 'main.tex';
const CUSTOM_CSS = 'custom.css';
const OUTPUT_FILENAME = 'index.html';
const STRINGS_XML = 'strings.xml';
const RAW_DIR = './docs/raw';
const BASE_DIR = RAW_DIR . '/en';
 
// Deployment
const DIST_DIR = './docs/dist';
const DIST_REPO = 'git@github.com:MuntashirAkon/AppManager.git';
const DIST_BRANCH = 'pages';
 
/**
 * Build and minify the outputs from TeX.
 *
 * @param string $lang Target language e.g. en, ru, ja, etc.
 */
function build_html(string $lang): void {
 $pwd = RAW_DIR . '/' . $lang;
 $main_tex = $pwd . '/' . MAIN_TEX;
 $output_file = $pwd . '/' . OUTPUT_FILENAME;
 // Fixed directories
 $base_dir = getcwd() . '/' . BASE_DIR;
 $custom_css = $base_dir . '/' . CUSTOM_CSS;
 $lua_dir = $base_dir . '/lua';
 // Sequence must be preserved
 $lua_files = [
 'toc_generator.lua',
 'img_to_object.lua',
 'header_with_hyperlinks.lua',
 'alert_fix.lua'
 ];
 $cmd = 'cd "' .$pwd. '" && pandoc "' . MAIN_TEX . '" -c "' . CUSTOM_CSS .'" -o "' . OUTPUT_FILENAME . '" -t html5'
 . ' -f latex -s --toc -N --section-divs --default-image-extension=png -i -F pandoc-crossref --citeproc'
 . ' --highlight-style=monochrome -M lang=' . get_IETF_language_tag($lang);
 foreach ($lua_files as $lua_script) {
 $cmd .= ' --lua-filter="' . $lua_dir . '/' . $lua_script . '"';
 }
 // Create variables first
 create_transient_tex($pwd, get_IETF_language_tag($lang));
 // Run command
 passthru($cmd, $ret_val);
 if ($ret_val != 0) {
 fprintf(STDERR, "Pandoc could not generate an HTML file.\n");
 exit(1);
 }
 
 // Read colours
 $main_contents = file_get_contents($main_tex);
 preg_match_all('/\\\definecolor\{(?<name>[^\}]+)\}\{HTML\}\{(?<color>[0-9a-fA-F]+)\}/', $main_contents, $matches);
 
 // Replace colours
 $to_search = array();
 $to_replace = array();
 foreach ($matches['name'] as $color_name) {
 array_push($to_search, "/style=\"background-color: $color_name\"/", "/style=\"color: $color_name\"/");
 }
 foreach ($matches['color'] as $color_value) {
 array_push($to_replace, "class=\"colorbox\" style=\"background-color: #$color_value\"", "style=\"color: #$color_value\"");
 }
 
 $to_search[] = '/href=\"custom\.css\"/';
 $to_replace[] = 'href="../css/custom.css"';
 $output_contents = file_get_contents($output_file);
 $output_contents = preg_replace($to_search, $to_replace, $output_contents);
 
 file_put_contents($output_file, $output_contents);
 
 // Minify CSS
 $cmd = "minify \"$custom_css\" -o \"$base_dir/../css/custom.css\"";
 system($cmd, $ret_val);
 if ($ret_val != 0) {
 fprintf(STDERR, "Could not minify custom.css\n");
 exit(1);
 }
 // Minify HTML
 $cmd = "minify \"$output_file\" -o \"$pwd/index.min.html\" && mv \"$pwd/index.min.html\" \"$output_file\"";
 system($cmd, $ret_val);
 if ($ret_val != 0) {
 fprintf(STDERR, "Could not minify index.html\n");
 exit(1);
 }
 // Replace custom.css with ../css/custom.css
}
 
/**
 * Recursively parse all the \input command and gather all the included tex files from main.tex.
 *
 * @param string[] $tex_files Relative links to the TeX files
 */
function collect_tex_files(array &$tex_files, string $base_dir = null, string $tex_file = null): void {
 if ($tex_file == null) {
 $base_dir = getcwd() . '/' . BASE_DIR;
 $tex_file = MAIN_TEX;
 }
 if (!file_exists($base_dir . '/' . $tex_file)) {
 echo "File $tex_file does not exist!";
 return;
 }
 $tex_files[] = $tex_file;
 $contents = file_get_contents($base_dir . '/' . $tex_file);
 preg_match_all('/\\\input\{(?<tex_file>[^\}]+)\}/', $contents, $matches);
 foreach ($matches['tex_file'] as $t) {
 collect_tex_files($tex_files, $base_dir, $t);
 }
}
 
/**
 * Parse the given TeX file and return the parsed contents as a key-value pair.
 */
function get_tex_contents_assoc(string $tex_file): array {
 $tex_file_contents = file_get_contents($tex_file);
 
 // Get all the titles
 preg_match_all('/((?<=section{)|(?<=subsection{)|(?<=subsubsection{)|(?<=chapter{)|(?<=caption{)|(?<=paragraph{))(?<raw_title>.*)(?<=%%##)(?<key>.*)(?=>>)/', $tex_file_contents, $matches);
 // Get titles from the raw titles
 $title_values = array();
 foreach ($matches['raw_title'] as $raw_title) {
 $c = 0; // number of extra {
 $len = strlen($raw_title);
 for ($i = 0; $i < $len; ++$i) {
 if ($raw_title[$i] == '{') {
 if ($i == 0 || $raw_title[$i - 1] != '\\') {
 // Unescaped {
 ++$c; // Increase { counter
 }
 } else if ($raw_title[$i] == '}') {
 if ($i == 0 || $raw_title[$i - 1] != '\\') {
 // Unescaped }
 --$c; // Decrease { counter since one match was found
 if ($c < 0) {
 // End of the title reached
 $title_values[] = substr($raw_title, 0, $i);
 break;
 }
 }
 }
 }
 }
 // Check keys for verification testing
 foreach ($matches['key'] as $key) {
 if (strlen($key) == 0) {
 echo "Warning: Empty raw title for key $key\n";
 continue;
 }
 if ($key[0] != '$') {
 echo "Warning: First letter of the title is not `$` (key: $key)\n";
 }
 if (!preg_match('/^\$[a-zA-Z0-9-_.]+$/', $key)) {
 echo "Warning: Key ($key) didn't match the required Regex\n";
 }
 }
 // Convert to key => value pair
 $titles = array_combine($matches['key'], $title_values);
 
 // Extract all the contents
 preg_match_all('/(?<=%%!!)(?<key>.*)(?=<<)/', $tex_file_contents, $matches);
 $content_values = array();
 $offset = 0;
 foreach ($matches['key'] as $key) {
 $start_magic = '%%!!' . $key . "<<\n";
 $start_pos = strpos($tex_file_contents, $start_magic, $offset);
 if ($start_pos === false) {
 fprintf(STDERR, "Error: Could not find key $key in $tex_file\n");
 exit(1);
 }
 $start_pos += strlen($start_magic);
 $end_pos = strpos($tex_file_contents, "\n%%!!>>", $offset);
 if ($end_pos === false) {
 $supposed_block_start_pos = strpos($tex_file_contents, "\n%%!!", $start_pos);
 if ($supposed_block_start_pos !== false) {
 syntax_error_with_position($tex_file, $tex_file_contents, $supposed_block_start_pos, "Could not locate %%!!>> for key: $key");
 } else {
 syntax_error_with_position($tex_file, $tex_file_contents, strlen($tex_file_contents), "EOF reached before matching %%!!>> for key: $key");
 }
 }
 // Ensure there is no %%!!<string><< or %%##$<string>>> between start and end since nesting isn't allowed
 $mal_pos1 = strpos($tex_file_contents, "\n%%!!", $start_pos);
 $mal_pos2 = strpos($tex_file_contents, "\n%%##", $start_pos);
 if ($mal_pos1 !== $end_pos) {
 syntax_error_with_position($tex_file, $tex_file_contents, $mal_pos1);
 }
 if ($mal_pos2 !== false && $mal_pos2 < $end_pos) {
 syntax_error_with_position($tex_file, $tex_file_contents, $mal_pos2);
 }
 $offset = $end_pos + 7;
 $content_values[] = substr($tex_file_contents, $start_pos, $end_pos - $start_pos);
 // Ensure that the next key is available before running into %%!!>> again
 $mal_pos1 = strpos($tex_file_contents, "\n%%!!", $offset);
 $mal_pos2 = strpos($tex_file_contents, "\n%%!!>>", $offset);
 if ($mal_pos1 !== false && $mal_pos1 === $mal_pos2) {
 // Invalid %%!!>> i.e. the end position considered earlier is wrong
 syntax_error_with_position($tex_file, $tex_file_contents, $end_pos);
 }
 }
 foreach ($matches['key'] as $key) {
 if (strlen($key) == 0) {
 echo "Warning: Empty raw title for key $key\n";
 continue;
 }
 if (!preg_match('/^[a-zA-Z0-9-_.]+$/', $key)) {
 echo "Warning: Key ($key) didn't match the required Regex\n";
 }
 }
 $contents = array_combine($matches['key'], $content_values);
 
 return array_merge($titles, $contents);
}
 
/**
 * Update base strings.xml
 */
function rebase_strings(): void {
 $base_dir = getcwd() . '/' . BASE_DIR;
 $strings_file = $base_dir . '/' . STRINGS_XML;
 $tex_files = array();
 // Gather all the tex files
 collect_tex_files($tex_files);
 $xml = new XMLWriter();
 $xml->openUri($strings_file);
 $xml->setIndent(true);
 $xml->setIndentString(' ');
 $xml->startDocument('1.0', 'utf-8');
 $xml->writeComment('This file is auto-generated by ./scripts/docs.php. DO NOT EDIT THIS FILE.');
 $xml->startElement('resources');
 $xml->writeAttribute('xmlns:xliff', 'urn:oasis:names:tc:xliff:document:1.2');
 foreach ($tex_files as $tex_file) {
 $contents = get_tex_contents_assoc($base_dir . '/' . $tex_file);
 // Replace `/` and `.tex` with `$`
 $key_prefix = preg_replace(['/\//', '/\.tex$/'], '$', $tex_file);
 foreach ($contents as $key => $val) {
 $xml->startElement('string');
 $xml->writeAttribute('name', $key_prefix . $key);
 $xml->writeRaw(android_escape_slash_newline(ltrim($val)));
 $xml->endElement(); // string
 }
 }
 $xml->endElement(); // resources
}
 
/**
 * Update translation from strings.xml for the given language. It replaces the strings available in the strings.xml and
 * then rebuilds the HTML file.
 *
 * @param string $lang Target language e.g. en, ru, ja, etc.
 */
function update_translations(string $lang): void {
 $base_dir = getcwd() . '/' . BASE_DIR;
 $pwd = $base_dir . '/../' . $lang;
 $strings_file = $pwd . '/' . STRINGS_XML;
 // Read strings.xml: Get tex file and key
 $dom = new DOMDocument();
 $dom->loadXML(file_get_contents($strings_file));
 $string_nodes = $dom->getElementsByTagName('string');
 $strings = array(); // tex_file => [ key => value ]
 foreach ($string_nodes as $node) {
 $raw_key = $node->getAttribute('name');
 $pos = strrpos($raw_key, '$');
 if ($raw_key[$pos - 1] == '$') --$pos; // This $ was part of the title
 $tex_file = str_replace('$', '/', substr($raw_key, 0, $pos) . '.tex');
 $key = substr($raw_key, $pos + 1);
 if (strlen($tex_file) == 0 || strlen($key) == 0) {
 fprintf(STDERR, "Invalid TeX filename or key (raw: $raw_key)\n");
 exit(1);
 }
 if (!isset($strings[$tex_file])) $strings[$tex_file] = array();
 $strings[$tex_file][$key] = get_trimmed_content($node->textContent);
 }
 // Gather all the tex files
 $tex_files = array();
 collect_tex_files($tex_files);
 foreach ($tex_files as $tex_file) {
 $target_path = $pwd . '/' . $tex_file;
 // Create directories if not exists
 $dir = substr($target_path, 0, strrpos($target_path, '/'));
 if (!is_dir($dir)) {
 if (file_exists($dir)) unlink($dir);
 if (!mkdir($dir, 0777, true)) {
 fprintf(STDERR, "Error: Could not create $dir\n");
 exit(1);
 }
 }
 if (isset($strings[$tex_file])) { // Matched a tex file
 $contents = $strings[$tex_file];
 $tex_file_contents = file_get_contents($base_dir . '/' . $tex_file);
 foreach ($contents as $key => $val) {
 if (strlen($key) == 0) {
 echo "Warning: Empty key (file: $tex_file)";
 continue;
 }
 if ($key[0] == '$') {
 // Fetch raw title and its offset
 $magic = preg_quote('%%##' . $key . '>>', '/');
 preg_match('/((?<=section{)|(?<=subsection{)|(?<=subsubsection{)|(?<=chapter{)|(?<=caption{)|(?<=paragraph{))(?<raw_title>.*)'. $magic .'/', $tex_file_contents, $matches, PREG_OFFSET_CAPTURE);
 if (!isset($matches['raw_title'])) {
 echo "Warning: Could not find magic $magic in $tex_file\n";
 continue;
 }
 // Sanitize raw title to real title
 $raw_title = $matches['raw_title'][0];
 $start_pos = $matches['raw_title'][1];
 $title = null;
 $c = 0; // number of extra {
 $len = strlen($raw_title);
 for ($i = 0; $i < $len; ++$i) {
 if ($raw_title[$i] == '{') {
 if ($i == 0 || $raw_title[$i - 1] != '\\') {
 // Unescaped {
 ++$c; // Increase { counter
 }
 } else if ($raw_title[$i] == '}') {
 if ($i == 0 || $raw_title[$i - 1] != '\\') {
 // Unescaped }
 --$c; // Decrease { counter since one match was found
 if ($c < 0) {
 // End of the title reached
 $title = substr($raw_title, 0, $i);
 break;
 }
 }
 }
 }
 // Replace title with translated title
 $tex_file_contents = substr_replace($tex_file_contents, $val, $start_pos, strlen($title));
 continue;
 }
 // Replace TeX contents
 $start_magic = '%%!!' . $key . "<<\n";
 $start_pos = strpos($tex_file_contents, $start_magic);
 if ($start_pos === false) {
 echo "Warning: Key not found (file: $tex_file, key: $key)\n";
 continue;
 }
 $start_pos += strlen($start_magic);
 $end_pos = strpos($tex_file_contents, "\n%%!!>>", $start_pos);
 $tex_file_contents = substr_replace($tex_file_contents, $val, $start_pos, $end_pos - $start_pos);
 }
 // Store the files in pwd
 file_put_contents($pwd . '/' . $tex_file, $tex_file_contents);
 // Check if curly braces are closed correctly
 $brace_count = 0;
 $len = strlen($tex_file_contents);
 for ($i = 0; $i < $len; ++$i) {
 if ($tex_file_contents[$i] == '{' && ($i == 0 || $tex_file_contents[$i - 1] != '\\')) {
 // Unescaped {
 ++$brace_count;
 } else if ($tex_file_contents[$i] == '}' && ($i == 0 || $tex_file_contents[$i - 1] != '\\')) {
 // Unescaped }
 --$brace_count;
 if ($brace_count < 0) {
 syntax_error_with_position($pwd . '/' . $tex_file, $tex_file_contents, $i, "Syntax error '$tex_file_contents[$i]'");
 }
 }
 }
 if ($brace_count > 0) {
 $brace_count = 0;
 for ($i = $len - 1; $i >= 0; --$i) {
 if ($tex_file_contents[$i] == '{' && ($i == 0 || $tex_file_contents[$i - 1] != '\\')) {
 // Unescaped {
 --$brace_count;
 if ($brace_count < 0) {
 syntax_error_with_position($pwd . '/' . $tex_file, $tex_file_contents, $i, "Syntax error '$tex_file_contents[$i]'");
 }
 } else if ($tex_file_contents[$i] == '}' && ($i == 0 || $tex_file_contents[$i - 1] != '\\')) {
 // Unescaped }
 ++$brace_count;
 }
 }
 }
 // Begin/end checks FIXME: This check is only covers a few things. It completely ignores checks like nesting
 preg_match_all('/(?<=\\begin\{)(?<env>.*)(?=})/', $tex_file_contents, $matches);
 $env_begin = $matches['env'];
 sort($env_begin);
 preg_match_all('/(?<=\\end\{)(?<env>.*)(?=})/', $tex_file_contents, $matches);
 $env_end = $matches['env'];
 sort($env_end);
 if (count($env_begin) != count($env_end)) {
 fprintf(STDERR, "Error: Invalid number of begin and ending of environments.\n");
 exit(1);
 }
 for ($i = 0; $i < count($env_begin); ++$i) {
 if ($env_begin[$i] != $env_end[$i]) {
 fprintf(STDERR, "Error: Could not find the end of the environment: $env_begin[$i].\n");
 exit(1);
 }
 }
 } else { // Didn't match any translation
 // Simply copy the file
 copy($base_dir . '/' . $tex_file, $pwd . '/' . $tex_file);
 }
 }
 // Build HTML
 build_html($lang);
 // Delete all except index.html and strings.xml
 $skip_delete = [OUTPUT_FILENAME, STRINGS_XML];
 $paths = list_files_recursive($pwd);
 rsort($paths);
 foreach ($paths as $path) {
 if (!in_array($path, $skip_delete)) {
 $full_path = $pwd . '/' . $path;
 if (is_dir($full_path)) rmdir($full_path);
 else unlink($full_path);
 }
 }
}
 
function deploy(bool $force = false): void {
 $languages = collect_languages();
 // Rebuild HTML
 foreach ($languages as $language) {
 $output_file = RAW_DIR . '/' . $language . '/' . OUTPUT_FILENAME;
 $strings_file = RAW_DIR . '/' . $language . '/' . STRINGS_XML;
 if (!$force && !need_update($output_file, $strings_file)) {
 fprintf(STDERR, "Skipped updating HTML for language $language\n");
 continue;
 }
 if ($language == 'en') {
 // For en, only build HTML
 build_html($language);
 } else {
 update_translations($language);
 }
 }
 if (!is_dir(DIST_DIR)) {
 mkdir(DIST_DIR);
 }
 // Copy HTML files to dist dir
 foreach ($languages as $language) {
 $src_html = RAW_DIR . '/' . $language . '/index.html';
 $dst_dir = DIST_DIR . '/' . $language;
 if (!is_dir($dst_dir)) {
 mkdir($dst_dir);
 }
 copy($src_html, $dst_dir . '/index.html');
 }
 // Copy other files
 // - images
 $dir = RAW_DIR . '/images';
 $files = list_files_recursive($dir);
 mkdir(DIST_DIR . '/images');
 foreach ($files as $file) {
 copy($dir . '/' . $file, DIST_DIR . '/images/' . $file);
 }
 // - css
 $dir = RAW_DIR . '/css';
 $files = list_files_recursive($dir);
 mkdir(DIST_DIR . '/css');
 foreach ($files as $file) {
 copy($dir . '/' . $file, DIST_DIR . '/css/' . $file);
 }
 // Generate index.html
 $js_lang_html = array();
 foreach ($languages as $language) {
 $lang_code = get_IETF_language_tag($language);
 $js_lang_html[] = " <a class=\"link\" href=\"$language/\" onclick=\"return setLanguage('$language')\">" . trim(Locale::getDisplayName($lang_code, $lang_code)) . "</a>";
 }
 $html_contents = file_get_contents(RAW_DIR . '/index.html');
 $html_contents = str_replace('PLACEHOLDER_LANGUAGES_AS_ARRAY', implode('\', \'', $languages), $html_contents);
 $html_contents = str_replace('<!-- PLACEHOLDER_LANGUAGES_AS_HTML -->', "\n" . implode(" &#x2022;\n", $js_lang_html) . "\n", $html_contents);
 file_put_contents(DIST_DIR . '/index.html', $html_contents);
 // Ignore .DS_Store files
 file_put_contents(DIST_DIR . '/' . '.gitignore', '*.DS_Store');
 // Commit changes
 passthru('cd ' . DIST_DIR . ' && git init && git add -A && git commit && git push -f ' . DIST_REPO . ' master:' . DIST_BRANCH);
 // Delete dist dir
 passthru('rm -rf ' . DIST_DIR);
}
 
function collect_languages() : array {
 $files = array_diff(list_files(RAW_DIR), array('css', 'images'));
 $languages = array();
 foreach ($files as $file) {
 if (is_dir(RAW_DIR . '/' . $file)) {
 $languages[] = $file;
 }
 }
 return $languages;
}
 
function need_update(string $html_file, string $strings_file) : bool {
 if (!is_file($html_file) || !is_file($strings_file)) {
 return true;
 }
 $html_time = filemtime($html_file);
 $strings_time = filemtime($strings_file);
 if ($html_time === false || $strings_time === false) {
 return true;
 }
 return $strings_time > $html_time;
}
 
function get_trimmed_content(string $content) : string {
 $len = strlen($content);
 if ($len > 0 && $content[0] == '"' && $content[$len - 1] == '"') {
 // Remove starting and ending quotations
 $content = substr($content, 1, strlen($content) - 2);
 }
 // Remove all newline literals
 return android_escape_slash_newline_reverse(str_replace("\n", '', $content));
}
 
function get_IETF_language_tag(string $lang): string {
 if (!str_contains($lang, '-')) {
 return $lang;
 }
 $lang_parts = explode('-', $lang);
 if ($lang_parts[1][0] == 'r') {
 // Skip r
 $lang_parts[1] = substr($lang_parts[1], 1, strlen($lang_parts[1]) - 1);
 }
 if ($lang_parts[1] == 'CN') $lang_parts[1] = 'Hans';
 else if ($lang_parts[1] == 'TW') $lang_parts[1] = 'Hant';
 return implode('-', $lang_parts);
}
 
function create_transient_tex(string $target_dir, string $ietf_lang = 'en'): void {
 $am_version = system("grep -m1 versionName ./app/build.gradle | awk -F \\\" '{print $2}'", $ret_val);
 if ($ret_val != 0) {
 fprintf(STDERR, "Could not get the versionName from ./app/build.gradle\n");
 exit(1);
 }
 $fmt = new IntlDateFormatter(
 $ietf_lang,
 IntlDateFormatter::FULL,
 IntlDateFormatter::FULL,
 'UTC',
 IntlDateFormatter::GREGORIAN,
 'd MMMM yyyy'
 );
 $today = $fmt->format(new DateTime());
 $content = <<<EOF
\\newcommand{\\version}{v{$am_version}}
\\ifdefined\\Vanilla\\else
 \\renewcommand{\\today}{{$today}}
\\fi
EOF;
 
 file_put_contents($target_dir . '/transient.tex', $content);
}
 
// MAIN //
if ($argc < 2) {
 fprintf(STDERR, "Invalid number of arguments.\n\n");
 fprintf(STDERR, HELP);
 exit(1);
}
 
$verb = $argv[1];
 
switch($verb) {
 case 'build':
 if (!isset($argv[2])) {
 fprintf(STDERR, "build <lang>\n");
 exit(1);
 }
 build_html($argv[2]);
 break;
 case 'rebase':
 rebase_strings();
 break;
 case 'update':
 if (!isset($argv[2])) {
 fprintf(STDERR, 'update <lang>\n');
 exit(1);
 }
 if ($argv[2] == 'en') {
 // For en, we only rebase string and update HTML (for some reason)
 rebase_strings();
 build_html($argv[2]);
 } else {
 update_translations($argv[2]);
 }
 break;
 case 'deploy':
 $force = isset($argv[2]) && $argv[2] == 'force';
 deploy($force);
 break;
 case 'pdf':
 create_transient_tex(BASE_DIR);
 passthru('cd ' . BASE_DIR . ' && pdflatex -shell-escape main_vanilla.tex', $return_code);
 if ($return_code == 0) {
 echo 'Built pdf: ' . BASE_DIR . '/main_vanilla.pdf' . "\n";
 } else {
 // Error
 exit($return_code);
 }
 break;
 case 'debug':
 echo "Nothing to do.\n";
 break;
 case 'help':
 echo HELP;
 exit(0);
 default:
 fprintf(STDERR, "Invalid verb $verb\n\n");
 fprintf(STDERR, HELP);
 exit(1);
}