From: Antonio Ospite Date: Sun, 11 Dec 2016 09:24:00 +0000 (+0100) Subject: Merge tag 'v1.0.0' into debian X-Git-Tag: debian/1.0.0-1~4 X-Git-Url: https://git.ao2.it/tweeper.git/commitdiff_plain/d8c771981d7c93b4b2b6fe6ac8df0d2b45a15479?hp=bb1004581214e6a562e3eb902751fae3e491d2ea Merge tag 'v1.0.0' into debian Release v1.0.0 --- diff --git a/Makefile b/Makefile index 4625aa8..eff450a 100644 --- a/Makefile +++ b/Makefile @@ -21,10 +21,11 @@ installdocs: docs install: installdocs install -d $(DESTDIR)$(TWEEPER_DIR) - install -m644 *.xsl $(DESTDIR)$(TWEEPER_DIR) install -m644 *.php $(DESTDIR)$(TWEEPER_DIR) install -m755 tweeper $(DESTDIR)$(TWEEPER_DIR) + install -d $(DESTDIR)$(TWEEPER_DIR)/src + install -m644 src/* $(DESTDIR)$(TWEEPER_DIR)/src install -d $(DESTDIR)$(BIN_DIR) - ln -sf $(TWEEPER_DIR)/tweeper $(DESTDIR)$(BIN_DIR)/tweeper + ln -rsf $(DESTDIR)$(TWEEPER_DIR)/tweeper $(DESTDIR)$(BIN_DIR)/tweeper @echo -e "\n\nINSTALLATION COMPLETE" - @echo -e "Make sure '$(PHP_SCRIPT_DIR)' is in PHP include_path!\n" + @echo -e "Make sure '$(DESTDIR)$(PHP_SCRIPT_DIR)' is in PHP include_path!\n" diff --git a/NEWS b/NEWS index d125dd5..6fccebb 100644 --- a/NEWS +++ b/NEWS @@ -1,3 +1,15 @@ +News for v1.0.0: +================ + + * Support "application/octet-stream" as an enclosure content type + * Support "application/pdf" as an enclosure content type + * Fix information leakage by validating the URL scheme + * Code restructuring to make it easier to use tweeper as a library in other + projects + * Allow installing tweeper via composer, the packagist page is at: + https://packagist.org/packages/ao2/tweeper + * Misc robustness fixes + News for v0.6: ============== diff --git a/TODO b/TODO index b305783..3c71811 100644 --- a/TODO +++ b/TODO @@ -1,7 +1,10 @@ -- write a better XSL stylesheet? I am not an XSL expert. -- evaluate the use of the RSS element. +- write better XSL stylesheets? I am not an XSL expert +- evaluate the use of the RSS element - show cards directly in RSS items for twitter.com - show direct links for videos in the Instagram feed - check the encoding of the tweets when UTF is used, maybe solvable with mb_convert_encoding()? See http://php.net/manual/en/domdocument.loadhtml.php + +- The dependencies on the symphony components in composer.json could be more + relaxed like ">=2.7.0", but for now sticking to "2.7.*" is good enough. diff --git a/autoload.php b/autoload.php new file mode 100644 index 0000000..4ba7832 --- /dev/null +++ b/autoload.php @@ -0,0 +1,82 @@ + + * + * This program is free software: you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation, either version 3 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +$package_name = 'ao2/tweeper'; + +if (file_exists(__DIR__ . '/vendor/autoload.php')) { + /* + * If "composer install" has been executed, use the composer autoloader. + * + * Using __DIR__ is OK as long as this file is on the same level of the + * project "vendor/" directory (usually the project root directory). + */ + require __DIR__ . '/vendor/autoload.php'; +} +elseif (preg_match('/' . preg_quote('/vendor/' . $package_name, '/') . '$/', __DIR__)) { + /* + * If running from a "vendor/" directory of another project use the + * autoloader of the parent project. + * + * This covers the case of running from a symlink in ./vendor/bin/ because + * __DIR__ contains the *real path* of this file. + * + * Note that using __DIR__ here and going back two levels is OK under the + * assumptions that this file is in the project root directory, and that the + * package name has the structure VENDOR/PROJECT_NAME. + */ + require __DIR__ . '/../../autoload.php'; +} +else { + /* + * Otherwise, run without composer: + * + * 1. register our own autoloader function for the Tweeper class + * + * The implementation follows the one suggested in: + * http://www.php-fig.org/psr/psr-4/ + */ + spl_autoload_register(function ($fully_qualified_class_name) { + /* This matches the data defined for the PSR-4 autoloader in composer.json */ + $namespace_prefix = 'Tweeper\\'; + $base_directory = 'src/'; + + $len = strlen($namespace_prefix); + if (strncmp($namespace_prefix, $fully_qualified_class_name, $len) !== 0) { + return; + } + + $class_relative = substr($fully_qualified_class_name, $len); + + $file_path = $base_directory . str_replace('\\', '/', $class_relative) . '.php'; + + require_once $file_path; + }); + + /* + * 2. load the system-wide autoloader from php-symphony-serializer + * + * This allows to run tweeper without composer, as long as the Symphony + * dependencies are available system-wide. + * + * For example, the Debian package takes care of that. + */ + require_once 'Symfony/Component/Serializer/autoload.php'; +} diff --git a/composer.json b/composer.json new file mode 100644 index 0000000..d490494 --- /dev/null +++ b/composer.json @@ -0,0 +1,29 @@ +{ + "name": "ao2/tweeper", + "type": "library", + "description": "Tweeper is a web scraper to convert popular social media sites to RSS (e.g. Twitter.com, Instagram.com).", + "keywords": ["Twitter", "Instagram", "Facebook", "RSS", "scraper"], + "homepage": "https://git.ao2.it/tweeper.git", + "license": "GPL-3.0+", + "authors": [ + { + "name": "Antonio Ospite", + "email": "ao2@ao2.it", + "homepage": "https://ao2.it", + "role": "Developer" + } + ], + "require": { + "php": ">=5.3.0", + "ext-curl": "*", + "ext-dom": "*", + "ext-json": "*", + "ext-xsl": "*", + "symfony/serializer": ">=2.7.0", + "symfony/property-access": ">=2.7.0" + }, + "autoload": { + "psr-4": { "Tweeper\\": "src/" } + }, + "bin": ["tweeper"] +} diff --git a/rss_converter_dilbert.com.xsl b/rss_converter_dilbert.com.xsl deleted file mode 100644 index b6d1975..0000000 --- a/rss_converter_dilbert.com.xsl +++ /dev/null @@ -1,115 +0,0 @@ - - - - - - - - - - - - - - - - - <xsl:variable name="title-length" select="140"/> - <!-- ellipsize, inspired from http://stackoverflow.com/questions/13622338 --> - <xsl:choose> - <xsl:when test="string-length($picture-title) > $title-length"> - <xsl:variable name="truncated-length" select="$title-length - 3"/> - <xsl:value-of select="substring($picture-title, 1, $truncated-length)"/> - <xsl:text>...</xsl:text> - </xsl:when> - <xsl:otherwise> - <xsl:value-of select="$picture-title"/> - </xsl:otherwise> - </xsl:choose> - - - - - - - - - - - - <![CDATA[ - {$picture-title} - ]]> - - - - - - - - - - - - - - - Tweeper - - <xsl:value-of select="$channel-title"/> - - - - - - - - - - <xsl:value-of select="$channel-title"/> - - - - - - - - - - - - - diff --git a/rss_converter_facebook.com.xsl b/rss_converter_facebook.com.xsl deleted file mode 100644 index 418b3d2..0000000 --- a/rss_converter_facebook.com.xsl +++ /dev/null @@ -1,141 +0,0 @@ - - - - - - - - - - https://facebook.com - - - - - - - - - - - - - - - <xsl:variable name="item-title" select="$item-content//p"/> - <xsl:variable name="title-length" select="140"/> - <!-- ellipsize, inspired from http://stackoverflow.com/questions/13622338 --> - <xsl:choose> - <xsl:when test="string-length($item-title) > $title-length"> - <xsl:variable name="truncated-length" select="$title-length - 3"/> - <xsl:value-of select="substring($item-title, 1, $truncated-length)"/> - <xsl:text>...</xsl:text> - </xsl:when> - <xsl:otherwise> - <xsl:value-of select="$item-title"/> - </xsl:otherwise> - </xsl:choose> - - - - - - - - - - - - - - - - - <![CDATA[ - - ]]> - - - - - - - - - - - - Tweeper - - <xsl:value-of select="$channel-title"/> - - - - - - <![CDATA[ - - ]]> - - - - <xsl:value-of select="$channel-title"/> - - - - - - - - - - - - - diff --git a/rss_converter_howtoons.com.xsl b/rss_converter_howtoons.com.xsl deleted file mode 100644 index 403b9ac..0000000 --- a/rss_converter_howtoons.com.xsl +++ /dev/null @@ -1,102 +0,0 @@ - - - - - - - - - - http://howtoons.com - - - - - - - <xsl:value-of select="normalize-space(.//div[@class='post-headline']//a)"/> - - - - - - - - - - - - - - - - - - <![CDATA[ - - ]]> - - - - - - - - - - - - Tweeper - - <xsl:value-of select="$channel-title"/> - - - - - - The world's greatest D.I.Y. comic website! Tools of mass construction! - - - - <xsl:value-of select="$channel-title"/> - - - - - - http://www.howtoons.com/wp-content/themes/atahualpa/images/header/tuck1000.png - - - - - - - diff --git a/rss_converter_identi.ca.xsl b/rss_converter_identi.ca.xsl deleted file mode 120000 index d8042a1..0000000 --- a/rss_converter_identi.ca.xsl +++ /dev/null @@ -1 +0,0 @@ -rss_converter_pump.io.xsl \ No newline at end of file diff --git a/rss_converter_instagram.com.xsl b/rss_converter_instagram.com.xsl deleted file mode 100644 index e869d7d..0000000 --- a/rss_converter_instagram.com.xsl +++ /dev/null @@ -1,135 +0,0 @@ - - - - - - - - - https://instagram.com - - - - - - - - - - - - - - - - - - - - - - - - <xsl:variable name="title-length" select="140"/> - <xsl:variable name="item-content-title" select="normalize-space(concat($user-name, ': ', $item-content-caption))"/> - <!-- ellipsize, inspired from http://stackoverflow.com/questions/13622338 --> - <xsl:choose> - <xsl:when test="string-length($item-content-title) > $title-length"> - <xsl:variable name="truncated-length" select="$title-length - 3"/> - <xsl:value-of select="substring($item-content-title, 1, $truncated-length)"/> - <xsl:text>...</xsl:text> - </xsl:when> - <xsl:otherwise> - <xsl:value-of select="$item-content-title"/> - </xsl:otherwise> - </xsl:choose> - - - - - - - - - - - - - <![CDATA[ -

- - (Video) - - -


- - ]]> - - - - - - - - - - - - - - - Tweeper - - <xsl:value-of select="$channel-title"/> - - - - - - <![CDATA[ - - - - - - ]]> - - - - <xsl:value-of select="$channel-title"/> - - - - - - - - - - - - - diff --git a/rss_converter_pump.io.xsl b/rss_converter_pump.io.xsl deleted file mode 100644 index 1577dcf..0000000 --- a/rss_converter_pump.io.xsl +++ /dev/null @@ -1,99 +0,0 @@ - - - - - - - - - - - - - - - - - <xsl:value-of select="concat($user-name, ': ', normalize-space($item-content))"/> - - - - - - - - - - - - - <![CDATA[ - - ]]> - - - - - - - - - - - - - - - - - - - Tweeper - - <xsl:value-of select="$channel-title"/> - - - - - - - - - - <xsl:value-of select="$channel-title"/> - - - - - - - - - - - - - diff --git a/rss_converter_twitter.com.xsl b/rss_converter_twitter.com.xsl deleted file mode 100644 index c154141..0000000 --- a/rss_converter_twitter.com.xsl +++ /dev/null @@ -1,208 +0,0 @@ - - - - - - - - - https://twitter.com - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - <xsl:value-of select="concat($user-name, ': ')"/> - <xsl:if test="$item-has-video"> - <xsl:text>(Video) </xsl:text> - </xsl:if> - <!-- - Prepend a space in front of the URLs which are not - preceded by an open parenthesis, for aestethic reasons. - Also, regex, I know: http://xkcd.com/1171/ - --> - <xsl:variable - name="processed-title" - select="php:functionString('preg_replace', '@((?<!\()(?:http[s]?://|pic.twitter.com))@', ' \1', $item-content)"/> - <!-- Also strip   and … --> - <xsl:value-of select="normalize-space(translate($processed-title, ' …', ''))"/> - - - - - - - - - - - - - - <![CDATA[ - - (Video) - - - - ]]> - - - - - - - - - - - - - - - - - - - - - - - - - Tweeper - - <xsl:value-of select="$channel-title"/> - - - - - - - - - - <xsl:value-of select="$channel-title"/> - - - - - - - - - - - - - diff --git a/src/Tweeper.php b/src/Tweeper.php new file mode 100644 index 0000000..93ac9e0 --- /dev/null +++ b/src/Tweeper.php @@ -0,0 +1,363 @@ + + * + * This program is free software: you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation, either version 3 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +use DOMDocument; +use XSLTProcessor; + +use Symfony\Component\Serializer\Serializer; +use Symfony\Component\Serializer\Encoder\XmlEncoder; +use Symfony\Component\Serializer\Normalizer\ObjectNormalizer; + +date_default_timezone_set('UTC'); + +/** + * Scrape supported websites and perform conversion to RSS. + */ +class Tweeper { + + private static $userAgent = "Mozilla/5.0 (Windows NT 6.1; rv:22.0) Gecko/20130405 Firefox/22.0"; + + /** + * Constructor sets up {@link $generate_enclosure}. + */ + public function __construct($generate_enclosure = FALSE) { + $this->generate_enclosure = $generate_enclosure; + } + + /** + * Convert numeric Epoch to the date format expected in a RSS document. + */ + public static function epochToRssDate($timestamp) { + if (!is_numeric($timestamp) || is_nan($timestamp)) { + $timestamp = 0; + } + + return gmdate(DATE_RSS, $timestamp); + } + + /** + * Convert generic date string to the date format expected in a RSS document. + */ + public static function strToRssDate($date) { + $timestamp = strtotime($date); + if (FALSE === $timestamp) { + $timestamp = 0; + } + + return Tweeper::epochToRssDate($timestamp); + } + + /** + * Convert string to UpperCamelCase. + */ + public static function toUpperCamelCase($str, $delim = ' ') { + $str_upper = ucwords($str, $delim); + $str_camel_case = str_replace($delim, '', $str_upper); + return $str_camel_case; + } + + /** + * Get the contents from a URL. + */ + private static function getUrlContents($url) { + $ch = curl_init($url); + curl_setopt_array($ch, array( + CURLOPT_HEADER => FALSE, + // Follow http redirects to get the real URL. + CURLOPT_FOLLOWLOCATION => TRUE, + CURLOPT_RETURNTRANSFER => TRUE, + CURLOPT_SSL_VERIFYHOST => FALSE, + CURLOPT_SSL_VERIFYPEER => FALSE, + CURLOPT_HTTPHEADER => array('Accept-language: en'), + CURLOPT_USERAGENT => Tweeper::$userAgent, + )); + $contents = curl_exec($ch); + if (FALSE === $contents) { + trigger_error(curl_error($ch)); + } + curl_close($ch); + + return $contents; + } + + /** + * Get the headers from a URL. + */ + private static function getUrlInfo($url) { + $ch = curl_init($url); + curl_setopt_array($ch, array( + CURLOPT_HEADER => TRUE, + CURLOPT_NOBODY => TRUE, + // Follow http redirects to get the real URL. + CURLOPT_FOLLOWLOCATION => TRUE, + CURLOPT_RETURNTRANSFER => TRUE, + CURLOPT_SSL_VERIFYHOST => FALSE, + CURLOPT_SSL_VERIFYPEER => FALSE, + CURLOPT_USERAGENT => Tweeper::$userAgent, + )); + curl_exec($ch); + $url_info = curl_getinfo($ch); + if (FALSE === $url_info) { + trigger_error(curl_error($ch)); + } + curl_close($ch); + + return $url_info; + } + + /** + * Generate an RSS element. + */ + public static function generateEnclosure($url) { + $supported_content_types = array( + "application/octet-stream", + "application/ogg", + "application/pdf", + "audio/aac", + "audio/mp4", + "audio/mpeg", + "audio/ogg", + "audio/vorbis", + "audio/wav", + "audio/webm", + "audio/x-midi", + "image/gif", + "image/jpeg", + "image/png", + "video/avi", + "video/mp4", + "video/mpeg", + "video/ogg", + ); + + $url_info = Tweeper::getUrlInfo($url); + + $supported = in_array($url_info['content_type'], $supported_content_types); + if (!$supported) { + error_log("Unsupported enclosure content type \"" . $url_info['content_type'] . "\" for URL: " . $url_info['url']); + return ''; + } + + // The RSS specification says that the enclosure element URL must be http. + // See http://sourceforge.net/p/feedvalidator/bugs/72/ + $http_url = preg_replace("/^https/", "http", $url_info['url']); + + $dom = new DOMDocument(); + $enc = $dom->createElement('enclosure'); + $enc->setAttribute('url', $http_url); + $enc->setAttribute('length', $url_info['download_content_length']); + $enc->setAttribute('type', $url_info['content_type']); + + return $enc; + } + + /** + * Mimic the message from libxml.c::php_libxml_ctx_error_level() + */ + private static function logXmlError($error) { + $output = ""; + + switch ($error->level) { + case LIBXML_ERR_WARNING: + $output .= "Warning $error->code: "; + break; + + case LIBXML_ERR_ERROR: + $output .= "Error $error->code: "; + break; + + case LIBXML_ERR_FATAL: + $output .= "Fatal Error $error->code: "; + break; + } + + $output .= trim($error->message); + + if ($error->file) { + $output .= " in $error->file"; + } + else { + $output .= " in Entity,"; + } + + $output .= " line $error->line"; + + error_log($output); + } + + /** + * Convert json to XML. + */ + private static function jsonToXml($json, $root_node_name) { + // Apparently the ObjectNormalizer used afterwards is not able to handle + // the stdClass object created by json_decode() with the default setting + // $assoc = false; so use $assoc = true. + $data = json_decode($json, $assoc = TRUE); + if (!$data) { + return NULL; + } + + $encoder = new XmlEncoder(); + $normalizer = new ObjectNormalizer(); + $serializer = new Serializer(array($normalizer), array($encoder)); + + $serializer_options = array( + 'xml_encoding' => "UTF-8", + 'xml_format_output' => TRUE, + 'xml_root_node_name' => $root_node_name, + ); + + $xml_data = $serializer->serialize($data, 'xml', $serializer_options); + if (!$xml_data) { + trigger_error("Cannot serialize data", E_USER_ERROR); + return NULL; + } + + return $xml_data; + } + + /** + * Convert the Instagram content to XML. + */ + private function getXmlInstagramCom($html) { + // Extract the json data from the html code. + $json_match_expr = '/window._sharedData = (.*);/'; + $ret = preg_match($json_match_expr, $html, $matches); + if ($ret !== 1) { + trigger_error("Cannot match expression: $json_match_expr\n", E_USER_ERROR); + return NULL; + } + + return Tweeper::jsonToXml($matches[1], 'instagram'); + } + + /** + * Make the Facebook HTML processable. + */ + private function preprocessHtmlFacebookCom($html) { + $html = str_replace('', '', $html); + return $html; + } + + /** + * Convert the HTML retrieved from the site to XML. + */ + private function htmlToXml($html, $host) { + $xmlDoc = new DOMDocument(); + + // Handle warnings and errors when loading invalid HTML. + $xml_errors_value = libxml_use_internal_errors(TRUE); + + // If there is a host-specific method to get the XML data, use it! + $get_xml_host_method = 'getXml' . Tweeper::toUpperCamelCase($host, '.'); + if (method_exists($this, $get_xml_host_method)) { + $xml_data = call_user_func_array(array($this, $get_xml_host_method), array($html)); + $xmlDoc->loadXML($xml_data); + } + else { + $xmlDoc->loadHTML($html); + } + + foreach (libxml_get_errors() as $xml_error) { + Tweeper::logXmlError($xml_error); + } + libxml_clear_errors(); + libxml_use_internal_errors($xml_errors_value); + + return $xmlDoc; + } + + /** + * Load a stylesheet if the web site is supported. + */ + private function loadStylesheet($host) { + $stylesheet = "file://" . __DIR__ . "/rss_converter_" . $host . ".xsl"; + if (FALSE === file_exists($stylesheet)) { + trigger_error("Conversion to RSS not supported for $host ($stylesheet not found)", E_USER_ERROR); + return NULL; + } + + $stylesheet_contents = Tweeper::getUrlContents($stylesheet); + + $xslDoc = new DOMDocument(); + $xslDoc->loadXML($stylesheet_contents); + + $xsltProcessor = new XSLTProcessor(); + $xsltProcessor->registerPHPFunctions(); + $xsltProcessor->setParameter('', 'generate-enclosure', $this->generate_enclosure); + $xsltProcessor->importStylesheet($xslDoc); + + return $xsltProcessor; + } + + /** + * Convert the site content to RSS. + */ + public function tweep($src_url) { + $url = parse_url($src_url); + if (FALSE === $url || empty($url["host"])) { + trigger_error("Invalid URL: $src_url", E_USER_ERROR); + return NULL; + } + + $scheme = $url["scheme"]; + if (!in_array($scheme, array("http", "https"))) { + trigger_error("unsupported scheme: $scheme", E_USER_ERROR); + return NULL; + } + + // Strip the leading www. to be more forgiving on input URLs. + $host = preg_replace('/^www\./', '', $url["host"]); + + $xsltProcessor = $this->loadStylesheet($host); + if (NULL === $xsltProcessor) { + return NULL; + } + + $html = Tweeper::getUrlContents($src_url); + if (FALSE === $html) { + return NULL; + } + + $preprocess_html_host_method = 'preprocessHtml' . Tweeper::toUpperCamelCase($host, '.'); + if (method_exists($this, $preprocess_html_host_method)) { + $html = call_user_func_array(array($this, $preprocess_html_host_method), array($html)); + } + + $xmlDoc = $this->htmlToXml($html, $host); + if (NULL === $xmlDoc) { + return NULL; + } + + $output = $xsltProcessor->transformToXML($xmlDoc); + + if (FALSE === $output) { + trigger_error('XSL transformation failed.', E_USER_ERROR); + return NULL; + } + return $output; + } + +} diff --git a/src/rss_converter_dilbert.com.xsl b/src/rss_converter_dilbert.com.xsl new file mode 100644 index 0000000..d340183 --- /dev/null +++ b/src/rss_converter_dilbert.com.xsl @@ -0,0 +1,115 @@ + + + + + + + + + + + + + + + + + <xsl:variable name="title-length" select="140"/> + <!-- ellipsize, inspired from http://stackoverflow.com/questions/13622338 --> + <xsl:choose> + <xsl:when test="string-length($picture-title) > $title-length"> + <xsl:variable name="truncated-length" select="$title-length - 3"/> + <xsl:value-of select="substring($picture-title, 1, $truncated-length)"/> + <xsl:text>...</xsl:text> + </xsl:when> + <xsl:otherwise> + <xsl:value-of select="$picture-title"/> + </xsl:otherwise> + </xsl:choose> + + + + + + + + + + + + <![CDATA[ + {$picture-title} + ]]> + + + + + + + + + + + + + + + Tweeper + + <xsl:value-of select="$channel-title"/> + + + + + + + + + + <xsl:value-of select="$channel-title"/> + + + + + + + + + + + + + diff --git a/src/rss_converter_facebook.com.xsl b/src/rss_converter_facebook.com.xsl new file mode 100644 index 0000000..933d3d2 --- /dev/null +++ b/src/rss_converter_facebook.com.xsl @@ -0,0 +1,141 @@ + + + + + + + + + + https://facebook.com + + + + + + + + + + + + + + + <xsl:variable name="item-title" select="$item-content//p"/> + <xsl:variable name="title-length" select="140"/> + <!-- ellipsize, inspired from http://stackoverflow.com/questions/13622338 --> + <xsl:choose> + <xsl:when test="string-length($item-title) > $title-length"> + <xsl:variable name="truncated-length" select="$title-length - 3"/> + <xsl:value-of select="substring($item-title, 1, $truncated-length)"/> + <xsl:text>...</xsl:text> + </xsl:when> + <xsl:otherwise> + <xsl:value-of select="$item-title"/> + </xsl:otherwise> + </xsl:choose> + + + + + + + + + + + + + + + + + <![CDATA[ + + ]]> + + + + + + + + + + + + Tweeper + + <xsl:value-of select="$channel-title"/> + + + + + + <![CDATA[ + + ]]> + + + + <xsl:value-of select="$channel-title"/> + + + + + + + + + + + + + diff --git a/src/rss_converter_howtoons.com.xsl b/src/rss_converter_howtoons.com.xsl new file mode 100644 index 0000000..35a6739 --- /dev/null +++ b/src/rss_converter_howtoons.com.xsl @@ -0,0 +1,102 @@ + + + + + + + + + + http://howtoons.com + + + + + + + <xsl:value-of select="normalize-space(.//div[@class='post-headline']//a)"/> + + + + + + + + + + + + + + + + + + <![CDATA[ + + ]]> + + + + + + + + + + + + Tweeper + + <xsl:value-of select="$channel-title"/> + + + + + + The world's greatest D.I.Y. comic website! Tools of mass construction! + + + + <xsl:value-of select="$channel-title"/> + + + + + + http://www.howtoons.com/wp-content/themes/atahualpa/images/header/tuck1000.png + + + + + + + diff --git a/src/rss_converter_identi.ca.xsl b/src/rss_converter_identi.ca.xsl new file mode 120000 index 0000000..d8042a1 --- /dev/null +++ b/src/rss_converter_identi.ca.xsl @@ -0,0 +1 @@ +rss_converter_pump.io.xsl \ No newline at end of file diff --git a/src/rss_converter_instagram.com.xsl b/src/rss_converter_instagram.com.xsl new file mode 100644 index 0000000..609be66 --- /dev/null +++ b/src/rss_converter_instagram.com.xsl @@ -0,0 +1,135 @@ + + + + + + + + + https://instagram.com + + + + + + + + + + + + + + + + + + + + + + + + <xsl:variable name="title-length" select="140"/> + <xsl:variable name="item-content-title" select="normalize-space(concat($user-name, ': ', $item-content-caption))"/> + <!-- ellipsize, inspired from http://stackoverflow.com/questions/13622338 --> + <xsl:choose> + <xsl:when test="string-length($item-content-title) > $title-length"> + <xsl:variable name="truncated-length" select="$title-length - 3"/> + <xsl:value-of select="substring($item-content-title, 1, $truncated-length)"/> + <xsl:text>...</xsl:text> + </xsl:when> + <xsl:otherwise> + <xsl:value-of select="$item-content-title"/> + </xsl:otherwise> + </xsl:choose> + + + + + + + + + + + + + <![CDATA[ +

+ + (Video) + + +


+ + ]]> + + + + + + + + + + + + + + + Tweeper + + <xsl:value-of select="$channel-title"/> + + + + + + <![CDATA[ + + + + + + ]]> + + + + <xsl:value-of select="$channel-title"/> + + + + + + + + + + + + + diff --git a/src/rss_converter_pump.io.xsl b/src/rss_converter_pump.io.xsl new file mode 100644 index 0000000..bf9f674 --- /dev/null +++ b/src/rss_converter_pump.io.xsl @@ -0,0 +1,99 @@ + + + + + + + + + + + + + + + + + <xsl:value-of select="concat($user-name, ': ', normalize-space($item-content))"/> + + + + + + + + + + + + + <![CDATA[ + + ]]> + + + + + + + + + + + + + + + + + + + Tweeper + + <xsl:value-of select="$channel-title"/> + + + + + + + + + + <xsl:value-of select="$channel-title"/> + + + + + + + + + + + + + diff --git a/src/rss_converter_twitter.com.xsl b/src/rss_converter_twitter.com.xsl new file mode 100644 index 0000000..58539ae --- /dev/null +++ b/src/rss_converter_twitter.com.xsl @@ -0,0 +1,208 @@ + + + + + + + + + https://twitter.com + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + <xsl:value-of select="concat($user-name, ': ')"/> + <xsl:if test="$item-has-video"> + <xsl:text>(Video) </xsl:text> + </xsl:if> + <!-- + Prepend a space in front of the URLs which are not + preceded by an open parenthesis, for aestethic reasons. + Also, regex, I know: http://xkcd.com/1171/ + --> + <xsl:variable + name="processed-title" + select="php:functionString('preg_replace', '@((?<!\()(?:http[s]?://|pic.twitter.com))@', ' \1', $item-content)"/> + <!-- Also strip   and … --> + <xsl:value-of select="normalize-space(translate($processed-title, ' …', ''))"/> + + + + + + + + + + + + + + <![CDATA[ + + (Video) + + + + ]]> + + + + + + + + + + + + + + + + + + + + + + + + + Tweeper + + <xsl:value-of select="$channel-title"/> + + + + + + + + + + <xsl:value-of select="$channel-title"/> + + + + + + + + + + + + + diff --git a/tests/test_information_leakage.sh b/tests/test_information_leakage.sh new file mode 100755 index 0000000..061d917 --- /dev/null +++ b/tests/test_information_leakage.sh @@ -0,0 +1,54 @@ +#!/bin/sh + +set -e + +TWEEPER="/usr/share/php/tweeper/tweeper" +#TWEEPER="./tweeper" + +check_result() { + URL="$1" + FILE="$2" + RESULT="$3" + + echo "URL $URL" + if [ "$RESULT" ]; + then + echo "--> $FILE" + echo " exists" + else + echo "... $FILE" + echo " does not exist" + fi + echo +} + +file_exists() { + FILE="$1" + URL="file://twitter.com/$FILE" + OUTPUT=$($TWEEPER $URL) + check_result "$URL" "$FILE" "$OUTPUT" +} + +file_exists_on_server() { + SERVER="$1" + FILE="$2" + URL="file://twitter.com/$FILE" + OUTPUT=$(curl $SERVER/tweeper.php?src_url=$URL 2> /dev/null) + check_result "$URL" "$FILE on $SERVER" "$OUTPUT" +} + +file_exists /etc/passwd || true +file_exists /etc/file_with_an_unlikely_name || true + +echo "Staring a test server" +echo + +php -S localhost:8000 -t $(dirname $TWEEPER) > /dev/null 2>&1 & +SERVER_PID=$! +sleep 1 + +file_exists_on_server http://localhost:8000 /etc/passwd || true +file_exists_on_server http://localhost:8000 /etc/file_with_an_unlikely_name || true + +echo "Shutting down the test server" +kill $SERVER_PID diff --git a/tweeper b/tweeper index 6256e20..d4b04e3 100755 --- a/tweeper +++ b/tweeper @@ -6,4 +6,17 @@ * CLI file to run tweeper. */ -require dirname(__FILE__) . '/tweeper.php'; +if (preg_match('/' . preg_quote('/vendor/bin', '/') . '$/', __DIR__)) { + /* + * This covers the case of tweeper running from a "vendor/bin" directory in + * a composer setup, but with the tweeper executable _not_ being a symlink. + * + * This can happen when the filesystem does not support symlinks. + */ + $package_name = 'ao2/tweeper'; + require __DIR__ . '/../' . $package_name . '/tweeper.php'; +} +else { + /* For the other cases look at the autoload.php required by tweeper.php */ + require __DIR__ . '/tweeper.php'; +} diff --git a/tweeper.1.asciidoc b/tweeper.1.asciidoc index d2f1f50..ac1fdd1 100644 --- a/tweeper.1.asciidoc +++ b/tweeper.1.asciidoc @@ -108,7 +108,7 @@ Main web site: COPYING ------- -Copyright \(C) 2013-2015 Antonio Ospite +Copyright \(C) 2013-2016 Antonio Ospite This program is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by diff --git a/tweeper.php b/tweeper.php index 94ea05f..ff98ab7 100644 --- a/tweeper.php +++ b/tweeper.php @@ -3,7 +3,7 @@ * @file * Tweeper - a Twitter to RSS web scraper. * - * Copyright (C) 2013-2015 Antonio Ospite + * Copyright (C) 2013-2016 Antonio Ospite * * This program is free software: you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by @@ -19,332 +19,13 @@ * along with this program. If not, see . */ -require_once 'Symfony/Component/Serializer/autoload.php'; +require_once 'autoload.php'; -use Symfony\Component\Serializer\Serializer; -use Symfony\Component\Serializer\Encoder\XmlEncoder; -use Symfony\Component\Serializer\Normalizer\ObjectNormalizer; +use Tweeper\Tweeper; date_default_timezone_set('UTC'); /** - * Scrape supported websites and perform conversion to RSS. - */ -class Tweeper { - - private static $userAgent = "Mozilla/5.0 (Windows NT 6.1; rv:22.0) Gecko/20130405 Firefox/22.0"; - - /** - * Constructor sets up {@link $generate_enclosure}. - */ - public function __construct($generate_enclosure = FALSE) { - $this->generate_enclosure = $generate_enclosure; - } - - /** - * Convert numeric Epoch to the date format expected in a RSS document. - */ - public static function epochToRssDate($timestamp) { - if (!is_numeric($timestamp) || is_nan($timestamp)) { - $timestamp = 0; - } - - return gmdate(DATE_RSS, $timestamp); - } - - /** - * Convert generic date string to the date format expected in a RSS document. - */ - public static function strToRssDate($date) { - $timestamp = strtotime($date); - if (FALSE === $timestamp) { - $timestamp = 0; - } - - return Tweeper::epochToRssDate($timestamp); - } - - /** - * Convert string to UpperCamelCase. - */ - public static function toUpperCamelCase($str, $delim = ' ') { - $str_upper = ucwords($str, $delim); - $str_camel_case = str_replace($delim, '', $str_upper); - return $str_camel_case; - } - - /** - * Get the contents from a URL. - */ - private static function getUrlContents($url) { - $ch = curl_init($url); - curl_setopt_array($ch, array( - CURLOPT_HEADER => FALSE, - // Follow http redirects to get the real URL. - CURLOPT_FOLLOWLOCATION => TRUE, - CURLOPT_RETURNTRANSFER => TRUE, - CURLOPT_SSL_VERIFYHOST => FALSE, - CURLOPT_SSL_VERIFYPEER => FALSE, - CURLOPT_HTTPHEADER => array('Accept-language: en'), - CURLOPT_USERAGENT => Tweeper::$userAgent, - )); - $contents = curl_exec($ch); - curl_close($ch); - - return $contents; - } - - /** - * Get the headers from a URL. - */ - private static function getUrlInfo($url) { - $ch = curl_init($url); - curl_setopt_array($ch, array( - CURLOPT_HEADER => TRUE, - CURLOPT_NOBODY => TRUE, - // Follow http redirects to get the real URL. - CURLOPT_FOLLOWLOCATION => TRUE, - CURLOPT_RETURNTRANSFER => TRUE, - CURLOPT_SSL_VERIFYHOST => FALSE, - CURLOPT_SSL_VERIFYPEER => FALSE, - CURLOPT_USERAGENT => Tweeper::$userAgent, - )); - curl_exec($ch); - $url_info = curl_getinfo($ch); - curl_close($ch); - - return $url_info; - } - - /** - * Generate an RSS element. - */ - public static function generateEnclosure($url) { - $supported_content_types = array( - "application/ogg", - "audio/aac", - "audio/mp4", - "audio/mpeg", - "audio/ogg", - "audio/vorbis", - "audio/wav", - "audio/webm", - "audio/x-midi", - "image/gif", - "image/jpeg", - "image/png", - "video/avi", - "video/mp4", - "video/mpeg", - "video/ogg", - ); - - $url_info = Tweeper::getUrlInfo($url); - - $supported = in_array($url_info['content_type'], $supported_content_types); - if (!$supported) { - error_log("Unsupported enclosure content type \"" . $url_info['content_type'] . "\" for URL: " . $url_info['url']); - return ''; - } - - // The RSS specification says that the enclosure element URL must be http. - // See http://sourceforge.net/p/feedvalidator/bugs/72/ - $http_url = preg_replace("/^https/", "http", $url_info['url']); - - $dom = new DOMDocument(); - $enc = $dom->createElement('enclosure'); - $enc->setAttribute('url', $http_url); - $enc->setAttribute('length', $url_info['download_content_length']); - $enc->setAttribute('type', $url_info['content_type']); - - return $enc; - } - - /** - * Mimic the message from libxml.c::php_libxml_ctx_error_level() - */ - private static function logXmlError($error) { - $output = ""; - - switch ($error->level) { - case LIBXML_ERR_WARNING: - $output .= "Warning $error->code: "; - break; - - case LIBXML_ERR_ERROR: - $output .= "Error $error->code: "; - break; - - case LIBXML_ERR_FATAL: - $output .= "Fatal Error $error->code: "; - break; - } - - $output .= trim($error->message); - - if ($error->file) { - $output .= " in $error->file"; - } - else { - $output .= " in Entity,"; - } - - $output .= " line $error->line"; - - error_log($output); - } - - /** - * Convert json to XML. - */ - private static function jsonToXml($json, $root_node_name) { - // Apparently the ObjectNormalizer used afterwards is not able to handle - // the stdClass object created by json_decode() with the default setting - // $assoc = false; so use $assoc = true. - $data = json_decode($json, $assoc = TRUE); - if (!$data) { - return NULL; - } - - $encoder = new XmlEncoder(); - $normalizer = new ObjectNormalizer(); - $serializer = new Serializer(array($normalizer), array($encoder)); - - $serializer_options = array( - 'xml_encoding' => "UTF-8", - 'xml_format_output' => TRUE, - 'xml_root_node_name' => $root_node_name, - ); - - $xml_data = $serializer->serialize($data, 'xml', $serializer_options); - if (!$xml_data) { - trigger_error("Cannot serialize data", E_USER_ERROR); - return NULL; - } - - return $xml_data; - } - - /** - * Convert the Instagram content to XML. - */ - private function getXmlInstagramCom($html) { - // Extract the json data from the html code. - $json_match_expr = '/window._sharedData = (.*);/'; - $ret = preg_match($json_match_expr, $html, $matches); - if ($ret !== 1) { - trigger_error("Cannot match expression: $json_match_expr\n", E_USER_ERROR); - return NULL; - } - - return Tweeper::jsonToXml($matches[1], 'instagram'); - } - - /** - * Make the Facebook HTML processable. - */ - private function preprocessHtmlFacebookCom($html) { - $html = str_replace('', '', $html); - return $html; - } - - /** - * Convert the HTML retrieved from the site to XML. - */ - private function htmlToXml($html, $host) { - $xmlDoc = new DOMDocument(); - - // Handle warnings and errors when loading invalid HTML. - $xml_errors_value = libxml_use_internal_errors(TRUE); - - // If there is a host-specific method to get the XML data, use it! - $get_xml_host_method = 'getXml' . Tweeper::toUpperCamelCase($host, '.'); - if (method_exists($this, $get_xml_host_method)) { - $xml_data = call_user_func_array(array($this, $get_xml_host_method), array($html)); - $xmlDoc->loadXML($xml_data); - } - else { - $xmlDoc->loadHTML($html); - } - - foreach (libxml_get_errors() as $xml_error) { - Tweeper::logXmlError($xml_error); - } - libxml_clear_errors(); - libxml_use_internal_errors($xml_errors_value); - - return $xmlDoc; - } - - /** - * Load a stylesheet if the web site is supported. - */ - private function loadStylesheet($host) { - $stylesheet = "file://" . __DIR__ . "/rss_converter_" . $host . ".xsl"; - if (FALSE === file_exists($stylesheet)) { - trigger_error("Conversion to RSS not supported for $host ($stylesheet not found)", E_USER_ERROR); - return NULL; - } - - $stylesheet_contents = Tweeper::getUrlContents($stylesheet); - - $xslDoc = new DOMDocument(); - $xslDoc->loadXML($stylesheet_contents); - - $xsltProcessor = new XSLTProcessor(); - $xsltProcessor->registerPHPFunctions(); - $xsltProcessor->setParameter('', 'generate-enclosure', $this->generate_enclosure); - $xsltProcessor->importStylesheet($xslDoc); - - return $xsltProcessor; - } - - /** - * Convert the site content to RSS. - */ - public function tweep($src_url) { - $url = parse_url($src_url); - if (FALSE === $url || empty($url["host"])) { - trigger_error("Invalid URL: $src_url", E_USER_ERROR); - return NULL; - } - - // Strip the leading www. to be more forgiving on input URLs. - $host = preg_replace('/^www\./', '', $url["host"]); - - $xsltProcessor = $this->loadStylesheet($host); - if (NULL === $xsltProcessor) { - return NULL; - } - - $html = Tweeper::getUrlContents($src_url); - if (FALSE === $html) { - return NULL; - } - - $preprocess_html_host_method = 'preprocessHtml' . Tweeper::toUpperCamelCase($host, '.'); - if (method_exists($this, $preprocess_html_host_method)) { - $html = call_user_func_array(array($this, $preprocess_html_host_method), array($html)); - } - - $xmlDoc = $this->htmlToXml($html, $host); - if (NULL === $xmlDoc) { - return NULL; - } - - $output = $xsltProcessor->transformToXML($xmlDoc); - - if (FALSE === $output) { - trigger_error('XSL transformation failed.', E_USER_ERROR); - return NULL; - } - return $output; - } - -} - -/** * Check if the script is being run from the command line. */ function is_cli() { @@ -434,4 +115,8 @@ if (!isset($options['src_url'])) { } $tweeper = new Tweeper($options['generate_enclosure']); -echo $tweeper->tweep($options['src_url']); +$output = $tweeper->tweep($options['src_url']); +if (is_null($output)) { + exit(1); +} +echo $output;