I want to archive object to xml file with boost xml archiver. The object can be serialized without problem. The output file can be even shown in Firefox. The whole file is pasted [here](http://codepad.org/GODZ1YvA) But when I load the object from the file, there always comes an input stream error exception. I stepped into the code and found the xml archiver always fails at reading the first <history> tag. I made a short version manually which only contain the <history> tag and all its children tags and I only load the object corresponding to the <history> tag and succeeded with no problem. I think it is weird that the history tag can't be processed with other ones. Could someone help me out. Thx!
You file ends abruptly with so many tags not closed at the end. Maybe you just copied a section of your file. I am not sure if this could be causing the issue, but something to watch out for..
the codepad.org cut the xml short. I reproduce the problem through a few codes. I pasted the code [here]( http://www.rapidshare.com.cn/6RF1IsA ). The origin.dat is generated by xml archiver. It has no problem. Opened in firefox, its tree structure can be shown without problem. The short.dat is the history tag and all its children tags cropped from the origin.dat. Loading origin.dat causes an input stream error when load the history tag. While, Loading history tag alone has no problem at all. It is weird.