/**
* Grabs the failed tries to get to the comics, and if this Slide is active
* as a load balancer, it will fetch data from the master Slide, copy and
* serve it.
*
* @author Woxxy
*/
public function comics()
{
// grab the urls
$this->comic_dir = $this->uri->segment(3);
$this->chapter_dir = $this->uri->segment(4);
$this->filename = $this->uri->segment(5);
// check that $comic is actually the interesting kind of foldername
// get the divider index
if (($comic_split = $this->_lastIndexOf($this->comic_dir, '_')) == -1) {
show_404();
}
// check that $chapter is actually the interesting kind of foldername
// get the divider index
if (($chapter_split = $this->_lastIndexOf($this->chapter_dir, '_')) == -1) {
show_404();
}
if (!$this->filename) {
show_404();
}
// separate stub and uniqid from both folders
$this->comic_stub = substr($this->comic_dir, 0, $comic_split);
$this->comic_uniqid = substr($this->comic_dir, $comic_split + 1);
$this->chapter_stub = substr($this->chapter_dir, 0, $chapter_split);
$this->chapter_uniqid = substr($this->chapter_dir, $chapter_split + 1);
// flag that forces updating the data for this chapter
$this->update = FALSE;
// flat that allows us not to send an image, used if we're at least cleaning up
$this->give_404 = FALSE;
// check that the comic exists in the database
$this->comic = new Comic();
$this->comic->where('stub', $this->comic_stub)->where('uniqid', $this->comic_uniqid)->limit(1)->get();
// if there's a result, let's check if there's the chapter available
if ($this->comic->result_count() == 1) {
// we got the comic! let's see if we got the chapter
$this->chapter = new Chapter();
$this->chapter->where('stub', $this->chapter_stub)->where('uniqid', $this->chapter_uniqid)->limit(1)->get();
if ($this->chapter->result_count() == 1) {
// we got the chapter! let's see if we're lucky and we already have its page data
$this->page = new Page();
$this->page->where('chapter_id', $this->chapter->id)->where('filename', $this->filename)->get();
if ($this->page->result_count() == 1) {
// we got its pagedata! let's grab the image
if ($this->_grab_page()) {
$this->output->set_content_type($this->page->mime)->set_output($this->file);
// good end
return TRUE;
}
}
}
}
// we will need the url to the master Slide anyway
$this->url = get_setting('fs_balancer_master_url');
// we want it always with a trailing slash
if (substr($this->url, -1, 0) != '/') {
$this->url = $this->url . '/';
}
$this->load->library('curl');
// first of all, does the image even exist? Since we're going to grab
// the image anyway if it exists, lets get ahead and grab it first
// uri_string starts with a slash, so we have to remove it
$this->file = $this->curl->simple_get($this->url . 'content/comics/' . $this->comic_stub . '_' . $this->comic_uniqid . '/' . $this->chapter_stub . '_' . $this->chapter_uniqid . '/' . $this->filename);
$this->load->helper('file');
/**
* @todo this still doesn't work in chrome at first load, even with echo
*/
$this->output->set_content_type(get_mime_by_extension($this->filename))->_display($this->file);
// if the file doesn't exist, let's not go through the rest of the mess
if (!$this->file) {
show_404();
}
// oh no, this chapter might not be up to date! let's grab the comic data
// /api/reader/comic gives us the comic data and all its chapters!
// form the get request and decode the result
// if the master server works it should be trustable
$request_url = $this->url . 'api/reader/comic/stub/' . $this->comic_stub . '/uniqid/' . $this->comic_uniqid . '/chapter_stub/' . $this->chapter_stub . '/chapter_uniqid/' . $this->chapter_uniqid . '/format/json';
$result = $this->curl->simple_get($request_url);
$result = json_decode($result, TRUE);
// if there's PHP errors in the $result, the json_decode might fail
// and return NULL, show a 404.
if (is_null($result)) {
log_message('error', 'content:comics() json_decode failed');
show_404();
}
// just show 404 if the API gives a formal error
if (isset($result["error"])) {
log_message('error', 'content:comics() json had an error: ' . $result["error"]);
show_404();
}
// search for the value of the chapter, so we don't bother anymore
// if it doesn't exist in database - though it should
$found = FALSE;
foreach ($result["chapters"] as $key => $item) {
if ($item["chapter"]["stub"] == $this->chapter_stub && $item["chapter"]["uniqid"] == $this->chapter_uniqid) {
// update the comic in the database
$comic = new Comic($result["comic"]["id"]);
// the comic array fits just right, an update costs nearly nothing
$comic->from_array($result["comic"]);
if ($comic->result_count() == 0) {
$comic->save_as_new();
} else {
$comic->save();
}
// remove remainants of deleted or updated chapters
// we need an array of chapters
$chapter_objects = array();
foreach ($result["chapters"] as $k => $i) {
$chapter_objects[] = $i["chapter"];
}
$this->_clean_comic($result["comic"]["id"], $chapter_objects);
$this->_clean_chapter($item["chapter"]["id"], $item["chapter"]["pages"]);
$found = TRUE;
break;
}
}
if (!$found) {
log_message('error', 'content:comics() chapter was not in the json array');
show_404();
}
$this->_grab_page();
}
