Page MenuHomeDevCentral
Files F24895534

No OneTemporary
Actions

View Options

diff --git a/page.php b/page.php
index 4f6c72d..d8b1d9a 100644
--- a/page.php
+++ b/page.php
@@ -1,364 +1,364 @@
<?php
define('LONG_DATE_FORMAT', '%e %B %Y');
define('USER_AGENT', 'WikimediaTools/SourceTemplatesGenerator/0.1');
define('USER_AGENT_FALLBACK', 'Mozilla/5.0');
define('USER_AGENT_FALLBACK_FULL', 'Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8.1.13) Gecko/20080311 Firefox/2.0.0.13');
require_once('helpers/Encoding.php');
class Page {
/*
* @var string The page URL
*/
public $url;
/**
* @var array Meta tags
*/
public $meta_tags;
/**
* @var string The page content
*/
public $data;
/**
* @var string The page title
*/
public $title;
/**
* @var string The page author
*/
public $author;
/**
* @var Array The page coauthors
*/
- public $coauthors;
+ public $coauthors = [];
/**
* @var string The site ISSN
*/
public $issn;
//If we use the parameters yyyy mm dd, we describe CONTENT date:
/**
* @var int The page content's year
*/
public $yyyy;
/**
* @var int The page content's month
*/
public $mm;
/**
* @var int The page content's day
*/
public $dd;
//If not, we describe ONLINE RESOURCE PUBLISH date:
/**
* @var string The page publication date in relevant locale
*/
public $date;
/**
* @var int The page publication unixtime
*/
public $unixtime;
/**
* @var bool Indicates if we have to skip year/month/date template parameters
*/
public $skipYMD = false;
/**
* @var bool Indicates if we have to skip month/date (but maybe keep year) template parameters
*/
public $skipMD = false;
/**
* @var bool Indicates if we have to skip author template parameter
*/
public $skipAuthor;
/**
* @var mixed If not null, contains an array for anotheser service to use
*/
public $switchTo = null;
/**
* @var string The last error occured while opening and parsing the page
*/
public $error;
/**
* Initializes a new Page instance. If an error occured, you can read it in $this->error.
*
* @param string $url the page URL
*/
function __construct ($url) {
$this->url = $url;
$this->get_data();
if ($this->data) {
$this->analyse();
}
}
function get_data () {
ini_set('user_agent', USER_AGENT);
$data = file_get_contents($this->url);
if (!$data) {
ini_set('user_agent', USER_AGENT_FALLBACK);
if (!$data = @file_get_contents($this->url)) {
$this->error = "Can't read URL";
return;
}
}
$encoding = mb_detect_encoding($data, "ISO-8859-15, ISO-8859-1, UTF-8, ASCII, auto");
if ($encoding && $encoding != 'UTF-8') {
$this->data = Encoding::toUTF8($data);
} else {
$this->data = $data;
}
}
/**
* Return a new Page instance, or if such class exists, an instance class specialized for your site.
*
* @param $url the page URL
*/
static function load ($url) {
//Classes list are stored in pages/index.dat file
//Each line contains the URL beginning, a tabulation, and the page analyser name
// * class is this name, appended by 'Page'
// * source file is the lowercase version of this name, appended by '.php'
$pages = file('pages/index.dat', true);
foreach ($pages as $line) {
$page = explode("\t", $line);
if (substr($url, 0, strlen($page[0])) == $page[0]) {
$file = strtolower(trim($page[1])) . '.php';
$class = trim($page[1]) . 'Page';
require("pages/$file");
return new $class($url);
}
}
return new Page($url);
}
/**
* Analyses metatags to process content
*/
function analyse () {
//Meta tags (including <meta property="" value=""> and <meta itemprop="" value="" syntax)
$this->meta_tags = $this->get_meta_tags();
$t = $this->meta_tags;
//Title
$this->title = $this->get_title();
//Date
if ($date = $this->getMetaTag($t, 'date', 'pubdate', 'content_create_date')) {
$date = date_parse($date);
$this->yyyy = $date['year'];
$this->mm = $date['month'];
$this->dd = $date['day'];
}
//Site name
$this->site = $this->getMetaTag($t, 'og:site_name');
//Author
$this->author = $this->getMetaTag($t, 'author');
}
/**
* Gets page metatags
*
* @return array an array where the keys are the metatags' names and the values the metatags' values
*/
function get_meta_tags () {
return $this::get_all_meta_tags($this->url);
}
/**
* Gets all metatags, including those using meta property= and meta itemprop= syntax
*
* @return array an array where the keys are the metatags' names and the values the metatags' values
*/
function get_all_meta_tags () {
//Thank you to Michael Knapp and Mariano
//See http://php.net/manual/en/function.get-meta-tags.php comments
preg_match_all('/<[\s]*meta[\s]+.*?\b(name|property|itemprop)\b="?' . '([^>"]*)"?[\s]*' . 'content="?([^>"]*)"?[\s]*[\/]?[\s]*>/si', $this->data, $match);
if (isset($match) && is_array($match) && count($match) == 4) {
$originals = $match[0];
$names = $match[2];
$values = $match[3];
if (count($originals) == count($names) && count($names) == count($values)) {
$metaTags = array();
for ($i = 0, $limiti = count($names) ; $i < $limiti ; $i++) {
$key = $names[$i];
$value = $values[$i];
//Sets an unique scalar value, or if several identical tag names are offered, an array of values.
//Some publishers offer several times the same tag to list several values (see T241).
if (array_key_exists($key, $metaTags)) {
$currentValue = $metaTags[$key];
if ($currentValue == $value) {
continue;
}
if (is_array($currentValue)) {
$metaTags[$key][] = $value;
} else {
//Scalar -> array
$metaTags[$key] = [ $currentValue, $value ];
}
} else {
$metaTags[$key] = $value;
}
}
}
}
array_walk($metaTags, [ self, clean_tag ]);
return $metaTags;
}
/**
* Cleans a tag value (callback for array_walk)
*
* @param mixed &$value array item's value
* @param string $key array item's key
*/
static function clean_tag (&$item, $key) {
if (is_array($item)) {
$item = join("; ", $item);
}
return trim($item);
}
/**
* Gets title
*
* @return string The page title
*/
function get_title () {
$title = $this->getMetaTag($this->meta_tags, 'title', 'og:title', 'DC.title', 'Title');
return $title ?: ((preg_match("#<title>(.+)<\/title>#iU", $this->data, $title)) ? trim($title[1]) : '');
}
/**
* Determines if the current page is an article published in a journal.
*
* @return bool true if the current page is an article ; otherwise, false
*/
function is_article () {
return
(array_key_exists('dc_type', $this->meta_tags) && $this->meta_tags['dc_type'] == 'journalArticle')
||
(array_key_exists('dcsext_pn-cat', $this->meta_tags) && $this->meta_tags['dcsext_pn-cat'] == 'Article')
||
array_key_exists('citation_journal_title', $this->meta_tags)
||
array_key_exists('prism_publicationname', $this->meta_tags);
}
/**
* Gets relevant metatag
*
* @param array the metatags
* @param string... the list of acceptable metatags
*
* @return string the first metatag value found
*/
static function getMetaTag () {
$tags = func_get_args();
$metatags = array_shift($tags);
foreach ($tags as $tag) {
$tag_lowercase = strtolower($tag);
foreach ($metatags as $key => $value) {
if ($tag_lowercase == strtolower($key)) return $value;
}
}
return '';
}
/**
* Finds a portion of text included between $before and $after strings on the current page
*
* @param string $before The string at the left of the text to be grabbed
* @param string $after The string at the right of the text to be grabbed
*
* @return string The text found between $before and $after
*/
function between ($before, $after) {
return self::grab($this->data, $before, $after);
}
/**
* Finds a portion of text included between $before and $after strings
*
* @param string $text The text where to find the substring
* @param string $before The string at the left of the text to be grabbed
* @param string $after The string at the right of the text to be grabbed [facultative]
*
* @return string The text found between $before and $after
*/
static function grab ($text, $before, $after = null) {
$pos1 = strpos($text, $before);
if ($pos1 === false) {
return false;
} else {
$pos1 += strlen($before);
}
if ($after === null) {
return substr($text, $pos1);
}
$pos2 = strpos($text, $after, $pos1 + 1);
if ($pos2 === false) {
return false;
}
return substr($text, $pos1, $pos2 - $pos1);
}
/**
* Downloads, through CURL library, accepting cookies.
*
* @param $url The URL to fetch
*/
static function curl_download ($url, $agent = '') {
$ch = curl_init();
$timeout = 5;
$cookie_file = tmpfile();
$cookie_file = tempnam(sys_get_temp_dir(), "cookie-sourcesgen-");
curl_setopt($ch, CURLOPT_COOKIESESSION, true);
curl_setopt($ch, CURLOPT_COOKIEFILE, $cookie_file);
curl_setopt($ch, CURLOPT_COOKIEJAR, $cookie_file);
curl_setopt($ch, CURLOPT_URL, $url);
curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1);
curl_setopt($ch, CURLOPT_FOLLOWLOCATION, true);
curl_setopt($ch, CURLOPT_CONNECTTIMEOUT, $timeout);
if ($agent != '') {
curl_setopt($ch, CURLOPT_USERAGENT, $agent);
}
$data = curl_exec($ch);
curl_close($ch);
unlink($cookie_file);
return $data;
}
}
diff --git a/templates/wikipedia-fr/Article.php b/templates/wikipedia-fr/Article.php
index 329c844..68c7816 100644
--- a/templates/wikipedia-fr/Article.php
+++ b/templates/wikipedia-fr/Article.php
@@ -1,204 +1,208 @@
<?php
setlocale(LC_TIME, 'fr_FR.UTF-8');
/**
* Represents an {{Article}} template on the French Wikipedia
*/
class ArticleTemplate extends Template {
public $lang;
public $title;
public $periodique;
public $year;
public $accessdate;
/**
* Initializes a new instance of the ArticleTemplate class
*/
function __construct () {
$this->name = "Article";
$this->accessdate = trim(strftime(LONG_DATE_FORMAT));
}
/**
* Creates a new instance of the ArticleTemplate class from a Page object
*
* @param Page $page The page to fill the template with data from
* @return ArticleTemplate
*/
static function loadFromPage ($page) {
$template = new self();
$t = $page->meta_tags;
//Language
$template->lang = page::getMetaTag($t, 'dc_language', 'citation_language', 'dc.Language');
//Authors
- if ($author = $page->author ?: page::getMetaTag($t, 'author', 'dc_creator', 'citation_authors', 'dc_contributor', 'citation_author', 'dc.Creator', 'DCSext.author')) {
+ if (count($page->coauthors)) {
+ foreach ($page->coauthors as $author) {
+ $template->authors[] = explode(', ', $author, 2);
+ }
+ } elseif ($author = $page->author ?: page::getMetaTag($t, 'author', 'dc_creator', 'citation_authors', 'dc_contributor', 'citation_author', 'dc.Creator', 'DCSext.author')) {
//TODO: handle Alpha Beta syntax instead Beta, Alpha
$author = trim($author, self::getTrimCharacterMask());
$template->authors[] = explode(', ', $author, 2);
}
//Title
if (!$template->title = page::getMetaTag($t, 'dc_title', 'citation_title')) {
$template->title = $page->title;
}
//Journal, publisher
$template->journal = $page->journal ?: page::getMetaTag($t, 'prism_publicationname', 'citation_journal_title', 'og:site_name');
$template->journalLink = $t['dc_source'];
$template->publisher = $page->publisher ?: page::getMetaTag($t, 'dc_publisher', 'citation_publisher');
//Issue name, number and volume
$template->issue = $page->issue ?: page::getMetaTag($t, 'prism_number', 'citation_issue');
$template->volume = $page->volume ?: page::getMetaTag($t, 'citation_volume');
if (
(!$template->issueName = $t['prism_issuename'])
&&
array_key_exists('dc_relation_ispartof', $t)
) {
$template->issueName = $t['dc_relation_ispartof']
. " <!-- !!! paramètre à nettoyer !!! -->";
}
//Date
if ($page->yyyy && $page->mm && $page->dd) {
$template->yyyy = $page->yyyy;
$template->mm = $page->mm;
$template->dd = $page->dd;
} elseif ($page->unixtime) {
$template->yyyy = date('Y', $page->unixtime);
$template->mm = date('m', $page->unixtime);
$template->dd = date('j', $page->unixtime);
} elseif ($date = page::getMetaTag($t, 'prism_publicationdate', 'dc_date', 'citation_date', 'datePublished', 'DC.date.issued')) {
if ($date[4] == '/' || $date[4] == '-' || $date[4] == '.') {
$template->yyyy = substr($date, 0, 4);
$template->mm = substr($date, 5, 2);
$template->dd = substr($date, 8, 2);
} else {
$template->yyyy = substr($date, 6, 4);
$template->mm = substr($date, 0, 2);
$template->dd = substr($date, 3, 2);
}
} else {
$template->yyyy = $page->yyyy ?: page::getMetaTag($t, 'citation_year', 'citation_publication_date');
}
//Pages
if ($page->pages) {
$template->pageStart = $page->pages;
} else {
$template->pageStart = page::getMetaTag($t, 'prism_startingpage', 'citation_firstpage', 'citation_first_page');
$template->pageEnd = page::getMetaTag($t, 'prism_endingpage', 'citation_lastpage', 'citation_last_page');
}
//ISBN, ISSN, URLs
$template->issn = $page->issn ?: page::getMetaTag($t, 'prism_issn', 'citation_issn');
$template->isbn = page::getMetaTag($t, 'citation_isbn');
$template->doi = $page->doi ?: page::getMetaTag($t, 'citation_doi');
$template->summary = page::getMetaTag($t, 'citation_abstract_html_url');
$template->url = self::getTextURL($page->url, $t);
return $template;
}
function __toString () {
//Langue
$this->params['langue'] = $this->lang;
//Auteur
if (!count($this->authors)) {
//Per http://fr.wikipedia.org/w/?&diff=93455862, print
//one blank set of lines for author when the article
//metadata doesn't offer author information.
$this->authors = [['', '']];
}
$k = 1;
foreach ($this->authors as $author) {
$this->params["prénom$k"] = $author[1];
$this->params["nom$k"] = $author[0];
$this->params["lien auteur$k"] = '';
$k++;
}
//Titre, périodique, éditeur, volume, etc.
$this->params['titre'] = $this->title;
$this->params['périodique'] = $this->journal;
//TODO: vérifier si l'aticle existe sur fr.wikip et contient l'infobox Presse ou est rattaché à une catégorie fille de [[Catégorie:Revue scientifique]]
//$this->params['lien périodique'] = $this->journal;
$this->params['éditeur'] = $this->publisher;
if ($this->volume) $this->params['volume'] = $this->volume;
$this->params['numéro'] = $this->issue;
if ($this->issueName) $this->params['titre numéro'] = $this->issueName;
//Date
if ($this->mm && $this->dd) {
$date = mktime(12, 0, 0, (int)$this->mm, (int)$this->dd, (int)$this->yyyy);
$this->params['jour'] = trim(strftime('%e', $date));
$this->params['mois'] = strftime('%B', $date);
}
$this->params['année'] = $this->yyyy;
//Pages, ISSN, ISBN, DOI, URL, consulté le
$this->params['pages'] = $this->pageEnd ? ($this->pageStart . '-' . $this->pageEnd) : $this->pageStart;
if ($this->issn) $this->params['ISSN'] = $this->issn;
if ($this->isbn) $this->params['ISBN'] = $this->isbn;
if ($this->doi) $this->params['doi'] = $this->doi;
$this->params['url texte'] = $this->url;
if (self::isSummaryPertinent($this->url, $this->summary)) {
$this->params['résumé'] = $this->summary;
}
$this->params['consulté le'] = trim(strftime(LONG_DATE_FORMAT));
return parent::__toString();
}
/**
* Gets article full text URL
*
* @param string $url the article current URL
*
* @return string the article fulltext URL
*/
static function getTextURL ($url, $metatags) {
if (strpos($url, '.revues.org/') > 0) {
//revues.org PDF generation is broken
return $url;
}
if ($text_url = page::getMetaTag($metatags, 'citation_pdf_url', 'citation_fulltext_html_url')) {
return $text_url;
}
return $url;
}
/**
* Determines if a summary is pertinent to include in parameters
*
* @param string $url_article Article URL
* @param string $url_summary Summary URL
*
* @return bool true if the summary URL should be included in templat ; otherwise, false
*/
static function isSummaryPertinent ($url_article, $url_summary) {
//Empty summary or identical to URL are rejected
if ($url_summary == '' || $url_summary == $url_article) return false;
//This site is indexed through /resume.php but gives /article.php as summary URL in metadata
if (substr($url_article, 0, 32) == "http://www.cairn.info/resume.php") return false;
return true;
}
/**
* Gets a trim character mask to clean pure alphanumeric data fields from extraneous ponctuation
*
* @return a character mask to use for trim() avoiding whitespaces and some other ponctuation
*/
static function getTrimCharacterMask() {
return "\t\n\r\0\x0B,; ";
}
}

File Metadata

Mime Type
text/x-diff
Expires
Wed, Mar 18, 14:27 (7 h, 12 m)
Storage Engine
blob
Storage Format
Raw Data
Storage Handle
3538925
Default Alt Text
(19 KB)

Event Timeline

Nasqueron DevCentral · If it had been much bigger the moon would have had a core of ice. · Powered by Phabricator