
13 Mar
2007
13 Mar
'07
8:15 p.m.
Hello Is there any example how to use boost::iostream mapped_file ? I couldnt find any.. Which are the reasons that should make me consider using mapped_file stream? I want to read my text file as fast as possible (not all at once) - it should be seekable.. I know I should open file in binary mode and do read() to fixed size blocks to achieve the best preformance, but then I would have problems separating lines (I was consider using tokenizer but it would require stl data type so I would have to copy data). Is it possible to solve this problem with using boost::iostream filter so that it would be faster than getline()? Any help is highly appreciated! Thank you very much Aljaz