Re: New release: "tohgr" image converter

[theycallmeloopy@gmail.com]

Mike,

I'm not sure where you're getting your info, but the tohgr.exe.gz file I downloaded has neither an MZ header nor a PE header anywhere in the file.

Windows executables start with bytes 4D 5A (ASCII "MZ") and have the byte sequence 50 45 00 00 (ASCII "PE" followed by 2 nulls) somewhere in them.  The gz archive in question has neither of these.

In fact, the GZIP specification says that gz files start with the fields ID1 and ID2, with values 1F and 8B, respectively.  The first two bytes in the tohgr archive are 1F and 8B.  I don't think that's a coincidence.

You can find a copy of the spec here: http://www.gzip.org/zlib/rfc-gzip.html#file-format

And what code are you talking about?  The binary file?  What do you see that makes you think mingw was used, and why is that relevant?

Sorry, but you're not making a lot of sense.

On Thursday, June 7, 2012 4:35:40 PM UTC-5, BluPhoenyx wrote:
> On 06/04/2012 06:26 PM, wrote:
> > Mike,
> >
> > No, that doesn't work.
> >
> > Here are some of the things I've already tried:
> >
> > Decompressing with 7-Zip for Windows
> > Decompressing with gzip for Windows
> > Renaming to tohgr.exe and trying to execute
> > Renaming to tohgr.exe and trying to execute in DOSBox
> > Renaming to tohgr.tar.gz and trying to decompress as a tar archive
> > Renaming to tohgr.exe.zip and trying to decompress as a zip archive with both 7-Zip and Windows
> > Poking around in a hex editor to check for obvious issues
> > Scouring the Internet looking for a good copy, or the source so I can build it myself
> >
> > I've also tested the other gz archives on the site and they do not appear to be corrupt.
> >
> > I have emailed the author about this problem, so hopefully we will hear something soon.
> 
> Glad to read that you worked out an alternative method for your needs. 
> My suggestion was essentially correct as the file has the proper windows 
> PE header but there appears to be garbage right after the header so 
> essentially the program is still corrupt. I suspect it might be the 
> cross compiler used (mingw if I read the code right.)
> 
> Cheers,
> Mike T