This is the same as Issue 26158. Truncating text files is not clearly documented for a start, and truncating after reading doesn’t seem to be considered much in the implementations.

Your question is answered at <https://bugs.python.org/issue26158#msg258619>. Your code calls the C implementation of “io.TextIOWrapper.truncate”. This implementation does not consider that there are 8 unread bytes in a buffer, so it truncates the file at the end of the buffer (5 read + 8 unread = 13 bytes).