Feb 27, 2012 at 8:05 PM
Edited Feb 27, 2012 at 8:07 PM
I use i-ngma to parse both QrCode, it give me same result as mecard contact. Google decoder can not. By further investment on output, I found that by copy past your content to generate QrCode by QrCode.Net. The result I got is different to the one you post
Now I have use QrCode.Net to create a QrCode, uploaded to imageshack, you can check yourself.
After use ZXing decoder online and use browser's view source function you can find out exactly raw text.
For the one I generate:
Here is the one rendered by Google-API that you provide up there.
And here is the one you create by QrCode.Net
The one you encode actually have some extra string at front. I don't know what's going on there, maybe you can find out.
Note: As you can see, my image is exactly same as Google-API's one. How I generate is just use demo contain inside sourcecode and copy paste your input string. Our library use google-API as test reference, only two part is different.
1. QrCode.Net will sometime generate QrCode one size smaller than Google-API. As they lack on determine algorithm to properly use most efficient size.
2. We use auto detect for eight-bit encode where Google-API ask you to input encoding table(for example: iso-8859-1 ). That will cause Google-API turn some character that not belong to that encoding table to become question
mark. You can easily test out by select character encoding to iso-8859-1 and input text with some japanese, korean or chinese on ZXing generator page. You will find out result during decode is all question mark.
That's only output viable different between two encoder. Other difference will be just algorithm we use, but result will be exactly same.
Hope that helps you.