[syslinux] Syslinux 4.06-pre12

[Matt Fleming]

On Tue, 2012-10-09 at 16:50 +0200, Ady wrote:
> I think display.c32 is not working (at all) since 4.00.
> 
> There are partial alternatives to display.c32, but there is one 
> functionality that isn't fully replicated in those alternatives. Let 
> me explain.
> 
>  _ "F1...F12": can only display one last page. If the content of the 
> file is longer than one page, the first lines just "run through the 
> screen" and the user can effectively read the last page only.
> 
>  _ "display <filename>": same as above.
> 
>  _ "MENU HELP": same as above.
> 
>  _ "cat.c32": same as above.
> 
>  _ "rosh.c32": Although capable of displaying one page at a time 
> (with some quirks), it is not _that_ easy to manage for a simple, 
> non-technical oriented, sporadic user.
> 
> If you don't want to touch display.c32, then at least add the "MORE" 
> functionality to cat.c32. A "simple" user knows how to use the 
> "arrow" keys in the keyboard to go up and down one line, and the 
> "PageUp" and "PageDown" keys to scroll one page at a time. Keys like 
> "Home" and "End" are probably self-explanatory at this point. 
> Pressing [ESC] could just finish the process and would go back to the 
> previous state (the boot prompt, or menu.c32 or whatever was that 
> previous state), which is the same behavior as in the other "simple" 
> alternatives (but not in rosh).

That would indeed be a nice feature, but definitely something that
should be looked at once 4.06 is out the door.

> > > 5_ What about
> > >  http://www.syslinux.org/archives/2012-September/018039.html ?
> > 
> > This is a missing feature, rather than any kind of a bug. I don't see a
> > reason to block 4.06 for this. If someone writes a patch it can be
> > applied to a future release.
> > 
> 
> Probably the most relevant new feature for final users is the NTFS 
> support. Isn't that report about it? Is Paulo checking it?

As I understood it, the problem is that the NTFS code doesn't handle a
fragmented MFT. Paulo, do you think this is important enough to fix
before the final 4.06 release? Or is the use case special enough that we
can fix it later?

> > > 6_ What about
> > >   http://www.syslinux.org/archives/2012-September/018075.html ?
> > 
> > Again, this looks more involved than a simple one-line bug fix. If
> > someone wants to send patches then I'd be happy to review them.
> > 
> 
> About #6...
> 
> I know it is not a critical feature. It is not a bug in the sense 
> that it is not going to break anything. But, currently, 2 out of the 
> 8 possible cases are completely useless. My suggestion (request) adds 
> actual usefulness to those 2 cases, and corrects the current 
> inaccurate documentation.
> 
> Since the current documentation about it is incomplete (and even 
> wrong), it took me several months of testing to understand the real 
> current behavior (I thought I was doing something wrong) and thus 
> come up with the suggestion. My first email about it was sent close 
> to 3 months ago (after I had been already testing it for some long 
> time before that first email).
> 
> Quoting HPA:
>  "if the current behavior is useless to people AND the corrected 
> behavior would be useful to people, then it is worth fixing."
> 
> I wish I could fix it myself, but I'm not a developer. So, after so 
> much testing and emails, my only hope is that someone else would take 
> the challenge and fix it.

Unfortunately, with things of this nature you're at the mercy of someone
else having the time to write a patch. Though you have made this much
easier by doing so much testing and analysis.

-- 
Matt Fleming, Intel Open Source Technology Center