Access to Items of data of XML

[alirezaok]

hi

I use of simplexml_load_string() function to extract data of: http://weather.yahooapis.com/forecastrss?w=2255054&u=c
ofter this job, i can not give some items of this page. for example: Forecast: Items there are not in output of this function.

any one can help me?

thanks

[bradgrafelman]

What does your code look like?

[alirezaok]

$pre="http://weather.yahooapis.com/forecastrss?w=2255054&u=c";
	$curl = curl_init();
    curl_setopt ($curl, CURLOPT_URL, $pre);
    curl_setopt($curl, CURLOPT_USERAGENT, "Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1)");
	curl_setopt($curl, CURLOPT_HEADER, 0);
    curl_setopt($curl, CURLOPT_RETURNTRANSFER, 1);
	curl_setopt($curl, CURLOPT_TIMEOUT, 50); 
    $html = curl_exec ($curl);
    curl_close ($curl);

$doc = simplexml_load_string($html);
print_r ($doc);

[dalecosp]

A brief investigation would seem to indicate that simplexml_load_string isn't going to do here; you may have to try DOMDocument.

Alternatively, you might just load the data with file_get_contents (assuming your server allows this) and parse it yourself.

[sneakyimp]

In my experience, SimpleXML kinda sucks. You might get some reasonably good XML and load it, but then it gets really hard to actually access the data in the resulting object.

What is the output of your script? Does it actually result in a SimpleXML object or does it fail?

[dalecosp]

I tested it. It returns and prints (print_r) an XML object, but without the pertinent data, namely the forecast info.

I wondered if it had anything to do with location, but the Yahoo page clearly shows the weather for his location, and the opposite test, for which I used Yahoo's address in Sunnyvale, showed correspondingly identical output (without the forecast).

OTOH, calling file_get_contents() on the URI returned XML that included forecast data. So, my conclusion was that simplexml somehow wasn't up to the task of parsing the file. I didn't go so far as to run the XML through a syntax checker.

[sneakyimp]

Looks like DOMDocument is a lot more effective (and powerful). I had some luck with this script:

<?php

$pre="http://weather.yahooapis.com/forecastrss?w=2255054&u=c";
    $curl = curl_init();
    curl_setopt ($curl, CURLOPT_URL, $pre);
    curl_setopt($curl, CURLOPT_USERAGENT, "Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1)");
    curl_setopt($curl, CURLOPT_HEADER, 0);
    curl_setopt($curl, CURLOPT_RETURNTRANSFER, 1);
    curl_setopt($curl, CURLOPT_TIMEOUT, 50); 
    $html = curl_exec ($curl);
    curl_close ($curl);

//echo "<pre>";
//echo htmlspecialchars($html);
//echo "</pre>";

$doc = new DOMDocument();
if (!$doc->loadXML($html)) {
        throw new Exception("xml load failed.");
}

//echo htmlspecialchars($doc->saveXML());
//die();

// <yweather:forecast day="Wed" date="4 Dec 2013" low="0" high="3" text="Cloudy" code="26"/>
$attribute_names = array(
        "day",
        "date",
        "low",
        "high",
        "text",
        "code"
);

$forecasts = $doc->getElementsByTagNameNS("http://xml.weather.yahoo.com/ns/rss/1.0", "forecast");
foreach ($forecasts as $fc) {
        foreach($attribute_names as $name) {
                echo "&nbsp;&nbsp;" . $name . "=" . $fc->getAttribute($name) . "<br>";
        }
}


?>

[johanafm]

+1 for DOM.

I'd advice also using [man]libxml_use_internal_errors[/man] to turn off automatic error output so that you may instead check for errors yourself.