This is off-topic? perhaps.

[Tim Cross <tcross@xxxxxxxxxxx>]

Kalyan Mukherjea writes:
 > Hi Raman,
 > T. V. Raman writes:
 >  > Run pdftotext when the --raw option
 > 
 > I guess you meant `with' the -raw option. Yes the result is much
 > better. No more lines with >1000 characters. But the disparity between
 > the answers provided by `C-x w' and `C-x l' is quite startling. But
 > since I can listen to the files with comfort and ease, I will not
 > pursue this point.
 >

What is bound to C-x w? I have no binding for that key sequence. 


 > I went through the folder where I store your mails to the list but
 > could not find any clues to how to get a "snapshot" from the S E N
 > repository. If it is on your blog don't bother to reply: I will track
 > it down.
 > 
 > Thanks for the help and lots beside!
 > 

You need to have subversion installed. The command is 

svn co http://emacspeak.googlecode.com/svn/trunk dest-dir

where dest-dir is the destination directory you want the snapshot placed in. I
tend to run two snapshots and alternate between them. Then, if I get a snapshot
which is broken/has problems, I can easily revert back to the previous one I
had. So, I have a directory called svn and to get a snapshot, I cd into that
directory and issue the command

svn co http://emacspeak.googlecode.com/svn/trunk emacspeak-new

then I cd into emacspeak-new and do the normal make config, make emacspeak and
a make install. 

When working with the most recent snapshot, its always very wise to make a
backup of your previous version. The snapshots, by definition are a work in
progress and are not guaranteed to be error free. However, my experience has
been that I've only ever encountered very minor problems when using the
snapshots. Raman is pretty careful about when he checks in updates and this
tends to keep the snapshots pretty stable. Still, there are no guarantees. When
I have had problems, the hardest thing has generally been tracking down the
problem. The fixes have typically been quite simple.

Another alternative which I'll take the liberty of plugging is my txutils.el
file (which I believe is now in the contrib dir of emacspeak. This file makes
it easy to view PDFs and uses a customize interface where you can easily set
command line arguments, such as -raw. Once you have this package installed,
doing a 'v' (view) from within dired on a PDF file will automatically convert
the file to text and open it in a buffer for reading. It also works for
postscript and word files (providing you have the necessary converters
installed). You can also add additional converters for new/other file types
quite easily. It also sets things up so that if you do a view on an html file,
it will render that file with w3m. 

I think the main difference between my approach and the wizzards approach used
by Raman is that my solution is more invasive in that it directly modifies the
view-file function (via advice). I've found no problem with this, but some
people prefer additional functionality to be less invaisive/invisible and allow
the end user more control over when it is used. Its just a matter of taste. I
like the consistency of being able to get text that can be read with the 'v'
key from within dired regardless of the file type. (essentially I'm lazy and
will tend to hit v without really thinking rather than running a wizzard.

Tim


-- 
Tim Cross
tcross@xxxxxxxxxxx

There are two types of people in IT - those who do not manage what they 
understand and those who do not understand what they manage.

-----------------------------------------------------------------------------
To unsubscribe from the emacspeak list or change your address on the
emacspeak list send mail to "emacspeak-request@xxxxxxxxxxx" with a
subject of "unsubscribe" or "help"