On Fri, Jan 07, 2005 at 12:39:28AM -0800, Andrew Sly wrote:
I have tried to download one, but have had trouble installing it.
I reply on the list to ask at the same time: is there any effort to centralize such tools, lists of words (for different centuries and languages), people competent in such or such language, etc.? I usually use Paul Zimmermann's epelle. Sometimes I have to tweak a program to clean up the source and avoid useless warnings, especially when the work is in several files (I include special words in a "local dictionary" file but as I didn't understand how to tell epelle to use it, I programmed a filter upstream)
An html version can be found here: http://www.victoria.tc.ca/~sly/RETOUR2.HTM.zip
The HTML version "shouldn't" have em-dashes as they are, especially in French (see my previous message where I back this opinion up a little).
Alternatively, I could provide a plain text version.
w3m -dump did it for me. w3m -dump Retour2.htm | perl -pe 's/([^-])--([^-])/$1 $2/g' > Retour2.txt $ epelle -8bit Retour2.txt | wc -l 339 There are scannos all right. (I developed a scanno-finder program as well as other help tools; see http://www.eleves.ens.fr/home/blondeel/PGDP/AutoCorrect/ about that) The problem of epelle is sometimes it muches stuff: tata' could become tatá and you are lost to find the problem in the file... this is where I use other programs around it. The best is to do that interactively, with context, using ispell with a French dictionary for example. If mdashes are a problem add spaces around them, ispell the resulting file, and switch mdashes back like they were. (File sent in private).