C# Tricks for NPVR

[imilne]

I know that at least a few of us don't have much experience with C#, so I thought it might be nice to have a thread with short tips and tricks, but obviously focused around NPVR and its API.

Here's one to kick things off...

This demonstrates two things: how to let a plugin continue to work if the user's skin doesn't support it, and how to merge the contents of a "global.xml" skin file with another file. I needed the latter to define some extra fonts that aren't in the main set without having to redefine them in every skin file that my plugin uses (popups, different views, and so on).

Code:

internal SkinHelper GetSkinHelper(string name)
{
    SkinHelper shMain = null;
    SkinHelper shGlbl = null;

    // Load the required files, either from the current skin's folder...
    if (Directory.Exists(SettingsHelper.GetInstance().GetSkinDirectory() + "VideosPlus"))
    {
        shMain = new SkinHelper("VideosPlus\\" + name);
        shGlbl = new SkinHelper("VideosPlus\\global.xml");
    }
    // ...or from the Default skin if they couldn't be found
    else
    {
        shMain = new SkinHelper("..\\Default\\VideosPlus\\" + name);
        shGlbl = new SkinHelper("..\\Default\\VideosPlus\\global.xml");
    }

    // Now merge in everything found in global.xml
    DataSet dsMain = new DataSet();
    DataSet dsGlbl = new DataSet();
    dsMain.ReadXml(new XmlNodeReader(shMain.GetBackingDocument()));            
    dsGlbl.ReadXml(new XmlNodeReader(shGlbl.GetBackingDocument()));
    dsMain.Merge(dsGlbl);

    shMain.GetBackingDocument().LoadXml(dsMain.GetXml());

    return shMain;
}

Feel free to use, abuse and critique, especially if it turns out that there's a better way of doing it

Iain

[whurlston]

Don't use WebClient.DownloadString or WebClient.DownloadData to download the xml before deserializing.

WebClient.DownloadString will apply UTF-16 encoding and non-English characters will not be decoded properly. WebClient.DownloadData will allow some non-English characters to be decoded properly but not others; Cyrillic characters are not decoded properly.

Instead, use System.Xml.XmlTextReader which can take a url as the source parameter and only requires two lines of code:

Code:

System.Xml.XmlReader reader = new System.Xml.XmlTextReader(someUrl);
MyXMLObject myXmlObj = (MyXMLObject)serializer.Deserialize(reader);

[mvallevand]

Since this is tips when reading strings that have utf-8 (which most files I read online do) I use

StreamReader sr = new StreamReader(filename, System.Text.Encoding.UTF8);

This get rid of many problems with 8859-1 conversions, and odd gibberish on the screen.

Martin

[whurlston]

StreamReader didn't work for me with the YouTube feeds, as the video you posted will attest. I tried it with both DownloadString and DownloadData.

[mvallevand]

For fetching "text" from utf-8 sites with WebRadio I use

Code:

WebRequest objRequest = WebRequest.Create(streamURL);
                WebResponse objResponse = objRequest.GetResponse();
                HttpWebResponse resp = (HttpWebResponse)objResponse;
                try
                {
                    Encoding enc = Encoding.GetEncoding(resp.CharacterSet.ToString());
                    Logger.Debug("Encoding " + enc.ToString());
                }
                catch { }
                StreamReader sr = new StreamReader(objResponse.GetResponseStream(), System.Text.Encoding.UTF8);
                string strContent = sr.ReadToEnd();

I log it in case I ever encounter something other than utf-8 and so far so good.

Martin

[bgowland]

For fetching "text" from utf-8 sites with WebRadio I use

I hope you actually do something in that 'catch' block.

Also, rather than declare 'enc' within your 'try' block, declare it prior and pass it to the constructor of StreamReader and process the text based on the encoding.

Admittedly, UTF-8 is probably the most common but always worth making provision for other encoding.

[mvallevand]

I don't catch that one, I have another try/catch around the whole download, I don't need to know that the encoding wasn't in the header. I can't use other encodings because these populate into NextPVR list boxes and I wasn't sure that they would work if I did.

Martin

[bgowland]

I don't catch that one, I have another try/catch around the whole download,

Hmm, it's time I retired so I won't go into details about why having empty catch blocks or a 'catch all' block are really bad ideas. Not good practices to teach people based on Iain's opening statement and the spirit of this thread.

Cheers,
Brian

[mvallevand]

As some of you know some of the fields in the NPVR database are serialized XML and they are a challenge to parse with normal sql. I finally had a reason to use a snippet of code that I found (I added multiline and ignore case)

Code:

// taken from http://sqlite.phxsoftware.com/forums/p/348/1457.aspx#1457
    [SQLiteFunction(Name = "REGEXP", Arguments = 2, FuncType = FunctionType.Scalar)]
    class MyRegEx : SQLiteFunction
    {
        public override object Invoke(object[] args)
        {
            return System.Text.RegularExpressions.Regex.IsMatch(Convert.ToString(args[1]), Convert.ToString(args[0]),RegexOptions.IgnoreCase|RegexOptions.Multiline);
        }
    }

using this you do things like

Code:

select * from scheduled_recording
where event_details REGEXP ('('Title>*.+vampire*.+</Title>')";

Martin