diff --git a/.gitignore b/.gitignore index 0c05f18..2359a6d 100644 --- a/.gitignore +++ b/.gitignore @@ -166,6 +166,11 @@ UpgradeLog*.htm App_Data/*.mdf App_Data/*.ldf +################# +## Other ide stuff +################# +.idea/* + ############# ## Windows detritus ############# diff --git a/bridges/GooglePlusPostBridge.php b/bridges/GooglePlusPostBridge.php new file mode 100644 index 0000000..05cffec --- /dev/null +++ b/bridges/GooglePlusPostBridge.php @@ -0,0 +1,125 @@ +returnError('You must specify a username (?username=...).', 400); + } + + $this->request = $param['username']; + // get content parsed +// $html = file_get_html(__DIR__ . '/../posts2.html' + $html = file_get_html(self::GOOGLE_PLUS_BASE_URL . urlencode($this->request) . '/posts' + // force language + , false, stream_context_create(array('http'=> array( + 'header' => 'Accept-Language: fr,fr-be,fr-fr;q=0.8,en;q=0.4,en-us;q=0.2;*' . "\r\n" + ))) + ) OR $this->returnError('No results for this query.', 404); + + // get title, url, ... there is a lot of intresting stuff in meta + $this->_title = $html->find('meta[property]', 0)->getAttribute('content'); + $this->_url = $html->find('meta[itemprop=url]', 0)->getAttribute('content'); + +// foreach ($html->find('meta') as $e) +// { +// $item = new \Item(); +// $item->content = var_export($e->attr, true); +// $this->items[] = $item; +// } + + // div[jsmodel=XNmfOc] + foreach($html->find('div.yt') as $post) + { + $item = new \Item(); +// $item->content = $post->find('div.Al', 0)->innertext; + $item->username = $item->fullname = $post->find('header.lea h3 a', 0)->innertext; + $item->id = $post->getAttribute('id'); +// $item->title = $item->fullname = $post->find('header.lea', 0)->plaintext; + $item->avatar = $post->find('div.ys img', 0)->src; +// var_dump((($post->find('a.o-U-s', 0)->getAllAttributes()))); + $item->uri = $post->find('a.o-U-s', 0)->href; + $item->timestamp = strtotime($post->find('a.o-U-s', 0)->plaintext); + $this->items[] = $item; + + // hashtag to treat : https://plus.google.com/explore/tag + $hashtags = array(); + foreach($post->find('a.d-s') as $hashtag) + { + $hashtags[ trim($hashtag->plaintext) ] = self::GOOGLE_PLUS_BASE_URL . $hashtag->href; + } + + $item->content = ''; + + // avatar display + $item->content .= '
'; + + $content = $post->find('div.Al', 0); + + // alter link +// $content = $content->innertext; +// $content = str_replace('href="./', 'href="' . self::GOOGLE_PLUS_BASE_URL, $content); +// $content = str_replace('href="photos', 'href="' . self::GOOGLE_PLUS_BASE_URL . 'photos', $content); + // XXX ugly but I don't have any idea how to do a better stuff, str_replace on link doesn't work as expected and ask too many checks + foreach($content->find('a') as $link) + { + $hasHttp = strpos($link->href, 'http'); + $hasDoubleSlash = strpos($link->href, '//'); + + if ((!$hasHttp && !$hasDoubleSlash) + || (false !== $hasHttp && strpos($link->href, 'http') != 0) + || (false === $hasHttp && false !== $hasDoubleSlash && $hasDoubleSlash != 0)) + { + // skipp bad link, for some hashtag or other stuff + if (strpos($link->href, '/') == 0) + { + $link->href = substr($link->href, 1); + } + $link->href = self::GOOGLE_PLUS_BASE_URL . $link->href; + } + } + $content = $content->innertext; + + $item->content .= '