What you have is perfectly adequate for things like the above. I used to use 3.1 a bit before going back to M$. Visual Studio 2013 Express is free of any licen(c|s)e hassles (although M$ probably make you jump through hoops to download it now), and works on 32 bit machines.
I think I looked for an alternative command-line C compiler once. I already had an ancient version of Turbo C, but that only produced 16-bit code. I seem to recall it could only cope with 1MB of data (sounds low, but it was for DOS rather than Windows) and it required you to go through hoops to access all the memory. Anyway, couldn't find one. I think the version of Borland C++ Builder I used comes with a command-line compiler. It's possible that the free compiler Martin referenced would do as well - (Turbo C, Borland C++ are from the same supplier and became branded Embarcadero c2009)
Probably. Anything without an IDE or that can be used without the IDE could be 16-bit, maybe 32-bit but unlikely to be 64-bit.
Never used it. But gather it was developed by a well known quiche eater Niklaus Wirth. ( https://www.ee.ryerson.ca/~elf/hack/realmen.html )
@EEPhil
. But why is your .exe 1.7MB when EEPhil's is a mere 60K (that it doesn't work notwithstanding)??? There must be a load of unnecessary garbage rolled in by the compiler.
Just for my benefit can you let me have some idea as to what the difference is between those that work and those that don’t. I’ve obviously boobed somewhere and I’d like to understand why. (Are your files big? Would you be prepared to somehow send them to me if they are not too big?)
I didn't realise you were against a re-format (eg after backup and restore). I thought doing that will produce a drive that you can use between your iPad and other device you used to write to the NTFS drive.
Non-destructively?
It happens to be a big archival desktop drive with stuff already on it.
Where did I say I had? I'll fish through my USB drives to see whether I have one that is FAT or can be transmogrified to FAT, just to establish the truth of the matter, but I suspect this is an NTFS compatibility issue. Network access by SMB is format-agnostic (the remote SMB server handles all that, SMB operates at the file rather than the file system level).
No problem, you only need to parse the ID3 part of the file, so I could crop out the MP3 payload and PM them.
My mistake, I confused /r with /R. You are correct. This works:
echo off
for /R %%g in (*.mp3) do (
echo %%~nxg
id3genre "%%~fg"
)Definitely not possible at the moment, id3genre outputs to the console but apparently not by stout or sterror. I've tried redirecting both to file, but the output remains resolutely to console!
I've replaced the payload with 10 seconds of silence... but in the process something has broken (not with the mp3 files, but the system no longer behaves properly even with the original files). No time to find out why right now.
Okay. You have a solution and explanation for the size difference. We could leave it at that but my curiosity has got the better of me. Any help understanding where I’ve gone wrong would be appreciated, but I’m not going to lose sleep over it. One unmolested failing file would help. Otherwise I’ll have to file it back in the pending bin.
Ah! If you didn't re-implement the proposed logic, there might be an error in that rather than EEPhil's implementation.
Yes, it will be the proposed logic or the implementation. EEPhil needs your fail case data to determine which.
I used the 32 bit version of this about 15 years ago to build the Windows version of humaxrw.
www.mingw-w64.org
If that's me again - the only things that should be outputted are printable characters (in the program isprint(ch)) and cr/lf (maybe also null). Very weird!
Yes it was me. There is something different about the failing files - they have large frames associated with one of the ID3 tags (COMM mostly). Having ballsed up the way I calculated the size the larger frame broke the calculation. Mea culpa.The most likely problem is I’ve made a dog’s breakfast out of it. There could be something different about the failing files which messes up BH’s algorithm - but that’s wishful thinking on my part.
