On Sat, 2002-10-26 at 20:23, Steve Strong wrote: > > JGrasp does use gcc (the kids are writing in C++, so they're really > using g++). I shouldn't have used the word crash. To be more precise, > they are doing a merge sort of two sorted files into a third, sorted > file. Each file contains records of financial transactions. They have > written overloaded stream insertion, stream extraction and comparison > operators to complete the task. When they read in the records they are > read correctly,
How can you tell? > but when written out to the file, the text is corrupted. What functions are you using to write the text? > Also, what is env? Not all shells understand the syntax: LANG=C /path/to/jGRASP For those shells you need to (and in the rest you can) use "env" to set the variables for the command you intent to launch. I've also seen env used as the shebang for programs whose path varies from system to system (env should not vary, it should always be in /usr/bin): #! /usr/bin/env python > I looked at the man pages and the info pages as > well. I figured out that using env to invoke jgrasp with LANG=C is > changing one of the environment variables, but I don't understand what C > means in this case It means that every character is exactly one byte, and it's value is the same as its numeric byte value. In locale's such as en_US, the rules are very different. Each character is still one byte, but 'A' has the same value (when sorting, for instance) as 'a' although their byte values are very different. In a utf-8 locale, a character may be represented by more than one byte, and locale specific character comparisons are used. > and I don't under stand what 'en_US.UTF-8' means > either. Does that imply that the 'default character encoding' is > english and uni-code, two byte representation? UTF-8 uses exactly as many bytes as it needs to to represent the character. ASCII characters are one byte long, other characters will be longer. > Where can I go to learn more about all of this? One site: http://www.cl.cam.ac.uk/~mgk25/unicode.html Otherwise, google it. -- redhat-list mailing list unsubscribe mailto:redhat-list-request@;redhat.com?subject=unsubscribe https://listman.redhat.com/mailman/listinfo/redhat-list
