Mail Archives: djgpp-workers/2001/09/30/12:17:38
> > findfirst does not find volume labels. It returns all files (but not
> > volume labels...). Checking the returned attributes and ignoring it
> > if the label bit is not set is probably the best fix here (then we
> > show "Drive X:" syntax. This fix is probably best done in findfirst.
>
> I don't think you can do that in findfirst: returning a failure on the
> first file it returns is not right, since W2K returns all files; and
> looping until you exhaust all files, looking for a real label, is also
> something we don't want, I think.
I hadn't looked at it yet. I'll do the local test (one additional test
in the if statement).
> I'd say let's not look for volume labels at all if DOS version is
> 0x532 and LFN is ON. I.e. let's make a local change in getmntent.
There is currently a test in the code to make sure hidden and system bits
are not set in the returned entry. I added a check to see if the volume
bit was set. This fixed the problem and did not hurt other systems I
tested it on. No need to make it specific for Win2K (which would hurt
NT + LFN which seems to work).
> > My CDROM drive is mounted and ready but it does not show up at all
> > (not debugged yet).
>
> The CD-specific calls in cdrom_drive_ready probably don't work as
> expected...
I added some printf statements and it seemed to return the expected
values. If I get more time I'll look at it further.
I did find one thing in the current code, a comparison to 0x332 as the
dos version, with a comment saying NT. Should this be 0x532 ?
> > Drives > 26 (which is also cds_drives) succeed in truename for some
> > unknown reason. I just disabled returns for drives > cds_drives if
> > version = 0x532 and I get a more normal display.
>
> A good move, I think.
It was a simple fix.
- Raw text -