aboutsummaryrefslogtreecommitdiffhomepage
path: root/index.php
diff options
context:
space:
mode:
authorArthurHoaro <arthur@hoa.ro>2016-01-04 10:45:54 +0100
committerArthurHoaro <arthur@hoa.ro>2016-01-11 21:19:31 +0100
commit1557cefbd76257ceb830f65806831b490faf0acc (patch)
tree787f6d8fdabe8ea2fc0c37b61d616e667cdfbda5 /index.php
parentc0a50f3663e207d5df007e0fa321219c1b32d6ea (diff)
downloadShaarli-1557cefbd76257ceb830f65806831b490faf0acc.tar.gz
Shaarli-1557cefbd76257ceb830f65806831b490faf0acc.tar.zst
Shaarli-1557cefbd76257ceb830f65806831b490faf0acc.zip
Fixes #410 - Retrieve title fails in multiple cases
* `get_http_url()` renamed to `get_http_response()`. * Use the same HTTP context to retrieve response headers and content. * Follow HTTP 301 and 302 redirections to retrieve the title (default max 3 redirections). * Add `LinkUtils` to extract titles and charset. * Try to retrieve charset from HTTP headers first (new), then HTML content. * Use mb_string to re-encode title if necessary.
Diffstat (limited to 'index.php')
-rw-r--r--index.php87
1 files changed, 37 insertions, 50 deletions
diff --git a/index.php b/index.php
index cd83600b..600b2f55 100644
--- a/index.php
+++ b/index.php
@@ -152,6 +152,7 @@ require_once 'application/FileUtils.php';
152require_once 'application/HttpUtils.php'; 152require_once 'application/HttpUtils.php';
153require_once 'application/LinkDB.php'; 153require_once 'application/LinkDB.php';
154require_once 'application/LinkFilter.php'; 154require_once 'application/LinkFilter.php';
155require_once 'application/LinkUtils.php';
155require_once 'application/TimeZone.php'; 156require_once 'application/TimeZone.php';
156require_once 'application/Url.php'; 157require_once 'application/Url.php';
157require_once 'application/Utils.php'; 158require_once 'application/Utils.php';
@@ -578,13 +579,6 @@ function linkdate2iso8601($linkdate)
578 return date('c',linkdate2timestamp($linkdate)); // 'c' is for ISO 8601 date format. 579 return date('c',linkdate2timestamp($linkdate)); // 'c' is for ISO 8601 date format.
579} 580}
580 581
581// Extract title from an HTML document.
582// (Returns an empty string if not found.)
583function html_extract_title($html)
584{
585 return preg_match('!<title>(.*?)</title>!is', $html, $matches) ? trim(str_replace("\n",' ', $matches[1])) : '' ;
586}
587
588// ------------------------------------------------------------------------------------------ 582// ------------------------------------------------------------------------------------------
589// Token management for XSRF protection 583// Token management for XSRF protection
590// Token should be used in any form which acts on data (create,update,delete,import...). 584// Token should be used in any form which acts on data (create,update,delete,import...).
@@ -1642,7 +1636,7 @@ function renderPage()
1642 1636
1643 // -------- User want to post a new link: Display link edit form. 1637 // -------- User want to post a new link: Display link edit form.
1644 if (isset($_GET['post'])) { 1638 if (isset($_GET['post'])) {
1645 $url = cleanup_url($_GET['post']); 1639 $url = cleanup_url(escape($_GET['post']));
1646 1640
1647 $link_is_new = false; 1641 $link_is_new = false;
1648 // Check if URL is not already in database (in this case, we will edit the existing link) 1642 // Check if URL is not already in database (in this case, we will edit the existing link)
@@ -1660,35 +1654,24 @@ function renderPage()
1660 // If this is an HTTP(S) link, we try go get the page to extract the title (otherwise we will to straight to the edit form.) 1654 // If this is an HTTP(S) link, we try go get the page to extract the title (otherwise we will to straight to the edit form.)
1661 if (empty($title) && strpos(get_url_scheme($url), 'http') !== false) { 1655 if (empty($title) && strpos(get_url_scheme($url), 'http') !== false) {
1662 // Short timeout to keep the application responsive 1656 // Short timeout to keep the application responsive
1663 list($headers, $data) = get_http_url($url, 4); 1657 list($headers, $content) = get_http_response($url, 4);
1664 // FIXME: Decode charset according to specified in either 1) HTTP response headers or 2) <head> in html
1665 if (strpos($headers[0], '200 OK') !== false) { 1658 if (strpos($headers[0], '200 OK') !== false) {
1666 // Look for charset in html header. 1659 // Retrieve charset.
1667 preg_match('#<meta .*charset=.*>#Usi', $data, $meta); 1660 $charset = get_charset($headers, $content);
1668 1661 // Extract title.
1669 // If found, extract encoding. 1662 $title = html_extract_title($content);
1670 if (!empty($meta[0])) { 1663 // Re-encode title in utf-8 if necessary.
1671 // Get encoding specified in header. 1664 if (! empty($title) && $charset != 'utf-8') {
1672 preg_match('#charset="?(.*)"#si', $meta[0], $enc); 1665 $title = mb_convert_encoding($title, $charset, 'utf-8');
1673 // If charset not found, use utf-8.
1674 $html_charset = (!empty($enc[1])) ? strtolower($enc[1]) : 'utf-8';
1675 }
1676 else {
1677 $html_charset = 'utf-8';
1678 }
1679
1680 // Extract title
1681 $title = html_extract_title($data);
1682 if (!empty($title)) {
1683 // Re-encode title in utf-8 if necessary.
1684 $title = ($html_charset == 'iso-8859-1') ? utf8_encode($title) : $title;
1685 } 1666 }
1686 } 1667 }
1687 } 1668 }
1669
1688 if ($url == '') { 1670 if ($url == '') {
1689 $url = '?' . smallHash($linkdate); 1671 $url = '?' . smallHash($linkdate);
1690 $title = 'Note: '; 1672 $title = 'Note: ';
1691 } 1673 }
1674
1692 $link = array( 1675 $link = array(
1693 'linkdate' => $linkdate, 1676 'linkdate' => $linkdate,
1694 'title' => $title, 1677 'title' => $title,
@@ -2314,11 +2297,11 @@ function genThumbnail()
2314 else // This is a flickr page (html) 2297 else // This is a flickr page (html)
2315 { 2298 {
2316 // Get the flickr html page. 2299 // Get the flickr html page.
2317 list($headers, $data) = get_http_url($url, 20); 2300 list($headers, $content) = get_http_response($url, 20);
2318 if (strpos($headers[0], '200 OK') !== false) 2301 if (strpos($headers[0], '200 OK') !== false)
2319 { 2302 {
2320 // flickr now nicely provides the URL of the thumbnail in each flickr page. 2303 // flickr now nicely provides the URL of the thumbnail in each flickr page.
2321 preg_match('!<link rel=\"image_src\" href=\"(.+?)\"!',$data,$matches); 2304 preg_match('!<link rel=\"image_src\" href=\"(.+?)\"!', $content, $matches);
2322 if (!empty($matches[1])) $imageurl=$matches[1]; 2305 if (!empty($matches[1])) $imageurl=$matches[1];
2323 2306
2324 // In albums (and some other pages), the link rel="image_src" is not provided, 2307 // In albums (and some other pages), the link rel="image_src" is not provided,
@@ -2326,7 +2309,7 @@ function genThumbnail()
2326 // <meta property="og:image" content="http://farm4.staticflickr.com/3398/3239339068_25d13535ff_z.jpg" /> 2309 // <meta property="og:image" content="http://farm4.staticflickr.com/3398/3239339068_25d13535ff_z.jpg" />
2327 if ($imageurl=='') 2310 if ($imageurl=='')
2328 { 2311 {
2329 preg_match('!<meta property=\"og:image\" content=\"(.+?)\"!',$data,$matches); 2312 preg_match('!<meta property=\"og:image\" content=\"(.+?)\"!', $content, $matches);
2330 if (!empty($matches[1])) $imageurl=$matches[1]; 2313 if (!empty($matches[1])) $imageurl=$matches[1];
2331 } 2314 }
2332 } 2315 }
@@ -2335,11 +2318,12 @@ function genThumbnail()
2335 if ($imageurl!='') 2318 if ($imageurl!='')
2336 { // Let's download the image. 2319 { // Let's download the image.
2337 // Image is 240x120, so 10 seconds to download should be enough. 2320 // Image is 240x120, so 10 seconds to download should be enough.
2338 list($headers, $data) = get_http_url($imageurl, 10); 2321 list($headers, $content) = get_http_response($imageurl, 10);
2339 if (strpos($headers[0], '200 OK') !== false) { 2322 if (strpos($headers[0], '200 OK') !== false) {
2340 file_put_contents($GLOBALS['config']['CACHEDIR'].'/'.$thumbname,$data); // Save image to cache. 2323 // Save image to cache.
2324 file_put_contents($GLOBALS['config']['CACHEDIR'].'/' . $thumbname, $content);
2341 header('Content-Type: image/jpeg'); 2325 header('Content-Type: image/jpeg');
2342 echo $data; 2326 echo $content;
2343 return; 2327 return;
2344 } 2328 }
2345 } 2329 }
@@ -2350,16 +2334,17 @@ function genThumbnail()
2350 // This is more complex: we have to perform a HTTP request, then parse the result. 2334 // This is more complex: we have to perform a HTTP request, then parse the result.
2351 // Maybe we should deport this to JavaScript ? Example: http://stackoverflow.com/questions/1361149/get-img-thumbnails-from-vimeo/4285098#4285098 2335 // Maybe we should deport this to JavaScript ? Example: http://stackoverflow.com/questions/1361149/get-img-thumbnails-from-vimeo/4285098#4285098
2352 $vid = substr(parse_url($url,PHP_URL_PATH),1); 2336 $vid = substr(parse_url($url,PHP_URL_PATH),1);
2353 list($headers, $data) = get_http_url('https://vimeo.com/api/v2/video/'.escape($vid).'.php', 5); 2337 list($headers, $content) = get_http_response('https://vimeo.com/api/v2/video/'.escape($vid).'.php', 5);
2354 if (strpos($headers[0], '200 OK') !== false) { 2338 if (strpos($headers[0], '200 OK') !== false) {
2355 $t = unserialize($data); 2339 $t = unserialize($content);
2356 $imageurl = $t[0]['thumbnail_medium']; 2340 $imageurl = $t[0]['thumbnail_medium'];
2357 // Then we download the image and serve it to our client. 2341 // Then we download the image and serve it to our client.
2358 list($headers, $data) = get_http_url($imageurl, 10); 2342 list($headers, $content) = get_http_response($imageurl, 10);
2359 if (strpos($headers[0], '200 OK') !== false) { 2343 if (strpos($headers[0], '200 OK') !== false) {
2360 file_put_contents($GLOBALS['config']['CACHEDIR'].'/'.$thumbname,$data); // Save image to cache. 2344 // Save image to cache.
2345 file_put_contents($GLOBALS['config']['CACHEDIR'] . '/' . $thumbname, $content);
2361 header('Content-Type: image/jpeg'); 2346 header('Content-Type: image/jpeg');
2362 echo $data; 2347 echo $content;
2363 return; 2348 return;
2364 } 2349 }
2365 } 2350 }
@@ -2370,18 +2355,18 @@ function genThumbnail()
2370 // The thumbnail for TED talks is located in the <link rel="image_src" [...]> tag on that page 2355 // The thumbnail for TED talks is located in the <link rel="image_src" [...]> tag on that page
2371 // http://www.ted.com/talks/mikko_hypponen_fighting_viruses_defending_the_net.html 2356 // http://www.ted.com/talks/mikko_hypponen_fighting_viruses_defending_the_net.html
2372 // <link rel="image_src" href="http://images.ted.com/images/ted/28bced335898ba54d4441809c5b1112ffaf36781_389x292.jpg" /> 2357 // <link rel="image_src" href="http://images.ted.com/images/ted/28bced335898ba54d4441809c5b1112ffaf36781_389x292.jpg" />
2373 list($headers, $data) = get_http_url($url, 5); 2358 list($headers, $content) = get_http_response($url, 5);
2374 if (strpos($headers[0], '200 OK') !== false) { 2359 if (strpos($headers[0], '200 OK') !== false) {
2375 // Extract the link to the thumbnail 2360 // Extract the link to the thumbnail
2376 preg_match('!link rel="image_src" href="(http://images.ted.com/images/ted/.+_\d+x\d+\.jpg)"!',$data,$matches); 2361 preg_match('!link rel="image_src" href="(http://images.ted.com/images/ted/.+_\d+x\d+\.jpg)"!', $content, $matches);
2377 if (!empty($matches[1])) 2362 if (!empty($matches[1]))
2378 { // Let's download the image. 2363 { // Let's download the image.
2379 $imageurl=$matches[1]; 2364 $imageurl=$matches[1];
2380 // No control on image size, so wait long enough 2365 // No control on image size, so wait long enough
2381 list($headers, $data) = get_http_url($imageurl, 20); 2366 list($headers, $content) = get_http_response($imageurl, 20);
2382 if (strpos($headers[0], '200 OK') !== false) { 2367 if (strpos($headers[0], '200 OK') !== false) {
2383 $filepath=$GLOBALS['config']['CACHEDIR'].'/'.$thumbname; 2368 $filepath=$GLOBALS['config']['CACHEDIR'].'/'.$thumbname;
2384 file_put_contents($filepath,$data); // Save image to cache. 2369 file_put_contents($filepath, $content); // Save image to cache.
2385 if (resizeImage($filepath)) 2370 if (resizeImage($filepath))
2386 { 2371 {
2387 header('Content-Type: image/jpeg'); 2372 header('Content-Type: image/jpeg');
@@ -2398,18 +2383,19 @@ function genThumbnail()
2398 // There is no thumbnail available for xkcd comics, so download the whole image and resize it. 2383 // There is no thumbnail available for xkcd comics, so download the whole image and resize it.
2399 // http://xkcd.com/327/ 2384 // http://xkcd.com/327/
2400 // <img src="http://imgs.xkcd.com/comics/exploits_of_a_mom.png" title="<BLABLA>" alt="<BLABLA>" /> 2385 // <img src="http://imgs.xkcd.com/comics/exploits_of_a_mom.png" title="<BLABLA>" alt="<BLABLA>" />
2401 list($headers, $data) = get_http_url($url, 5); 2386 list($headers, $content) = get_http_response($url, 5);
2402 if (strpos($headers[0], '200 OK') !== false) { 2387 if (strpos($headers[0], '200 OK') !== false) {
2403 // Extract the link to the thumbnail 2388 // Extract the link to the thumbnail
2404 preg_match('!<img src="(http://imgs.xkcd.com/comics/.*)" title="[^s]!',$data,$matches); 2389 preg_match('!<img src="(http://imgs.xkcd.com/comics/.*)" title="[^s]!', $content, $matches);
2405 if (!empty($matches[1])) 2390 if (!empty($matches[1]))
2406 { // Let's download the image. 2391 { // Let's download the image.
2407 $imageurl=$matches[1]; 2392 $imageurl=$matches[1];
2408 // No control on image size, so wait long enough 2393 // No control on image size, so wait long enough
2409 list($headers, $data) = get_http_url($imageurl, 20); 2394 list($headers, $content) = get_http_response($imageurl, 20);
2410 if (strpos($headers[0], '200 OK') !== false) { 2395 if (strpos($headers[0], '200 OK') !== false) {
2411 $filepath=$GLOBALS['config']['CACHEDIR'].'/'.$thumbname; 2396 $filepath=$GLOBALS['config']['CACHEDIR'].'/'.$thumbname;
2412 file_put_contents($filepath,$data); // Save image to cache. 2397 // Save image to cache.
2398 file_put_contents($filepath, $content);
2413 if (resizeImage($filepath)) 2399 if (resizeImage($filepath))
2414 { 2400 {
2415 header('Content-Type: image/jpeg'); 2401 header('Content-Type: image/jpeg');
@@ -2425,10 +2411,11 @@ function genThumbnail()
2425 { 2411 {
2426 // For all other domains, we try to download the image and make a thumbnail. 2412 // For all other domains, we try to download the image and make a thumbnail.
2427 // We allow 30 seconds max to download (and downloads are limited to 4 Mb) 2413 // We allow 30 seconds max to download (and downloads are limited to 4 Mb)
2428 list($headers, $data) = get_http_url($url, 30); 2414 list($headers, $content) = get_http_response($url, 30);
2429 if (strpos($headers[0], '200 OK') !== false) { 2415 if (strpos($headers[0], '200 OK') !== false) {
2430 $filepath=$GLOBALS['config']['CACHEDIR'].'/'.$thumbname; 2416 $filepath=$GLOBALS['config']['CACHEDIR'].'/'.$thumbname;
2431 file_put_contents($filepath,$data); // Save image to cache. 2417 // Save image to cache.
2418 file_put_contents($filepath, $content);
2432 if (resizeImage($filepath)) 2419 if (resizeImage($filepath))
2433 { 2420 {
2434 header('Content-Type: image/jpeg'); 2421 header('Content-Type: image/jpeg');
generated by cgit v1.2.3 (git 2.25.1) at 2024-06-02 14:34:08 +0000