<?php

namespace PicoFeed;

require_once __DIR__.'/Logging.php';
require_once __DIR__.'/Filter.php';
require_once __DIR__.'/Encoding.php';
require_once __DIR__.'/Grabber.php';

abstract class Parser
{
    protected $content = '';

    public $id = '';
    public $url = '';
    public $title = '';
    public $updated = '';
    public $items = array();
    public $grabber = false;
    public $grabber_ignore_urls = array();
    public $grabber_timeout = null;
    public $grabber_user_agent = null;


    abstract public function execute();


    public function __construct($content)
    {
        // Strip XML tag to avoid multiple encoding/decoding in next XML processing
        $this->content = Filter::stripXmlTag($content);

        // Encode everything in UTF-8
        $this->content = Encoding::toUTF8($this->content);

        // Workarounds
        $this->content = $this->normalizeData($this->content);
    }


    public function filterHtml($item_content, $item_url)
    {
        $content = '';

        if ($this->grabber && ! in_array($item_url, $this->grabber_ignore_urls)) {
            $grabber = new Grabber($item_url);
            $grabber->download($this->grabber_timeout, $this->grabber_user_agent);
            if ($grabber->parse()) $item_content = $grabber->content;
        }

        if ($item_content) {
            $filter = new Filter($item_content, $item_url);
            $content = $filter->execute();
        }

        return $content;
    }


    public function getXmlErrors()
    {
        $errors = array();

        foreach(\libxml_get_errors() as $error) {

            $errors[] = sprintf('XML error: %s (Line: %d - Column: %d - Code: %d)',
                $error->message,
                $error->line,
                $error->column,
                $error->code
            );
        }

        return implode(', ', $errors);
    }


    // Dirty quickfix before XML parsing
    public function normalizeData($data)
    {
        return str_replace("\xc3\x20", '', $data);
    }


    // Trim whitespace from the begining, the end and inside a string and don't break utf-8 string
    public function stripWhiteSpace($value)
    {
        $value = str_replace("\r", "", $value);
        $value = str_replace("\t", "", $value);
        $value = str_replace("\n", "", $value);
        return trim($value);
    }


    public function generateId()
    {
        // crc32b seems to be faster and shorter than other hash algorithms
        return hash('crc32b', implode(func_get_args()));
    }
}