As Sebastian suggested below, the issue may be related to nxml-mode,
which validates immediately upon opening the file and continually
throughout the editing process. This can be a little slow with large
documents, at least on the first pass through a file. If the extra
overhead of continual validation is the problem, you can turn off
validation (with C-c C-v) to speed things up, and turn validation back
on (again, with C-c C-v) as needed.
| John A. Walsh, Manager, Electronic Text Technologies
| Digital Library Program / University Information Technology Services
| Indiana University, 1320 East Tenth Street, Bloomington, IN 47405
| Voice:812-855-8758 Fax:812-856-2062 <mailto:[log in to unmask]>
On Feb 11, 2004, at 6:00 AM, Lou Burnard wrote:
> Performance of tei-emacs seems to depend on
> - how much memory the PC has available, real or virtual.
> - how many tags there are in the document
> 100Kb seems absurdly small to me, though. I routinely use emacs on a
> for editing files up to 5 times as big without any problem.
> Sebastian Rahtz wrote:
>> On Mon, Feb 09, 2004 at 12:51:58PM -0500, David Sewell wrote:
>>> Also, for what it's worth, I'm using tei-emacs on a PC (windows)
>>> system, and it gets unwieldy with file sizes over 100k. I haven't
>>> figured out how to increase it's limit, or if that's even possible,
>>> it's easier for me to work with smaller files.
>>> Obviously that makes it impossible to put a single volume into a
>>> TEI.2 element as desired. I don't use Emacs (and if I did it
>>> wouldn't be
>>> under Windows anyway); can anyone familiar with the Windows-based
>>> version give me any advice to pass along?
>> there are no limits in Emacs per se, and its often used with
>> multi-megabyte files. I find the report very surprising indeed - this
>> is one of the last criticisms I'd expect to hear about Emacs.
>> I'd want to hear more about what the person is doing,
>> and how he measures "unwieldy".
>> Ah. Maybe the person is using Emacs in nxml-mode with large
>> unstructured files? that can be cumbersome, depending on the
>> degree of unstructuredness. I suggest they switch to the old emacs
>> mode, if so, while doing basic markup (M-x xml-mode).