I want to write a cross platform file IO utility using POSIX. I was wondering if a file I wrote on Windows could be moved to OS X (via email or a thumbdrive) and still appear to contain the exact same content? And how about if I moved a file from OS X to Windows?
Also, are there any weird UTF-8 anomalies on either platform that I should be aware of when writing strings to this file?
This question is to be answered only for newer Macs, not the PowerPC Macs. I just want to make sure that all the endianness and related byte ordering problems died with PowerPC.
If your POSIX code uses the ‘binary’ flag when opening the file for writing, and both machines have the same endianness (which is true for modern Intel-based Macs and Intel-based Windows PCs), then yes, files should be byte-for-byte identical.
As for UTF-8, the main thing to be aware of is that in general, Windows prefers pre-composed characters (“normalisation form NFC”) while Mac OS prefers de-composed characters (“normalisation form NFD”). That is, Windows will encode “é” as U+00E9 LATIN SMALL LETTER E WITH ACUTE, while Mac OS will encode it as U+0065 LATIN SMALL LETTER E followed by U+0301 COMBINING ACUTE ACCENT. Obviously both OSs will render both sequences the same way, but I believe this affects characters entered by the user, or read from filenames on-disk.
As for endianness-related problems dying with PowerPC, that’s not strictly true. ARM at least can operate in either endianness, and while, say, iOS tends to use little-endian (to match Intel), other OSs might run in big-endian mode, and some (such as Linux) can be compiled either way. Wikipedia doesn’t mention which endianness Android devices typically use.