auctex
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [AUCTeX] Text corruption


From: David Kastrup
Subject: Re: [AUCTeX] Text corruption
Date: Wed, 19 Dec 2012 10:02:45 +0100
User-agent: Gnus/5.13 (Gnus v5.13) Emacs/24.3.50 (gnu/linux)

"Axel E. Retif" <address@hidden> writes:

> On 12/15/2012 05:39 PM, Axel E. Retif wrote:
>
>> Dear list,
>>
>> I posted this in comp.emacs:
>>
>> http://goo.gl/jvtvZ
>>
>> Please be so kind to read it.
>
> I should have checked my notes to be more precise. Instead of
>
>> Emacs [snapshot] was inserting \$ in the middle of an environment
>
> what happened is that when I inserted a `\' in the middle of a theorem
> environment (corollary), it changed \end{corollary} to
> \end^?corollary};

{ is 0x7b, ^? is 0x7f.  Again, just bit 2 is flipped.  Do you see a
pattern?


> Also, I said in this thread that I hadn't seen text corruption in org
> buffers, but in my notes I see that I had an `$' where there was a
> space.

0x20 is space, 0x24 is $.  They differ by a single bit, bit 2.

> And this made me recall that in the first few weeks after installing
> Ubuntu 12.04, I had a recurring and very annoying bug of `$'s getting
> inserted instead of spaces in /var/lib/dpkg/available and
> /var/lib/dpkg/status, which caused updates and installations to halt
> with errors, and I had to ` sudo gedit ' those files to retry the
> updates or installations. Most of the times the offending `$'s
> *recurred in the same places*.

Sure, if the program accessing them used the same physical memory
locations.

> Fortunately, I found in the web the solution: sudo dpkg
> --clear-avail. After that, I haven't had any more that problem.

Probably because then the program arrangement differed and the faulty
memory was mapped elsewhere.

> Before Ubuntu 12.04, I hadn't had that problem at all.

Well, faulty memory or a virus or kernel damage or buggy driver focusing
on bit 2.  A bad peripheral is less likely: those would corrupt whole
bytes in unforeseen places.

-- 
David Kastrup



reply via email to

[Prev in Thread] Current Thread [Next in Thread]