Re: [Fwd: Re: [Translate-devel] [diff] Autocorrection]

[Dwayne B. <dw...@tr...>]

Ah now I understand it.  Thanks Friedel.  I wonder if we can simplify it
even further and create the XML file from a simple CSV spreadsheet or
text file that has all the transformations.  It build the XML and then
packages it.



On Wed, 2005-09-14 at 18:57 +0200, F Wolff wrote:
> (Just saw I didn't include the list when replying...)
> 
> On Sun, 2005-09-04 at 18:20 +0200, Alberto Escudero-Pascual wrote:
> > Hi,
> > 
> > This is Alberto. This sounds very interesting as i do not know of many
> > that have play around with the dict module. Can you give me (us) a brief
> > intro of your goal and what the patch is exactly doing. I am not
> > familiar with what you call a "jar?" Java? 
> > 
> > Alberto
> > 
> 
> Hi Alberto
> 
> This is indeed a Java jar file containing 3 XML files. It describes
> functionality for enhancing typing in three ways:
> 
> ** It can automatically correct typing / spelling errors immediately
> after they are typed. In English, for example, typing "teh"
> automatically is corrected to "the", and typing (C) is automatically
> replaced with the copyright sign. This is defined with the
> "wrong"-"correct" pairs in DocumentList.xml
> 
> ** In the case that automatic capitalisation is used (see
> Tools->Autocorrect->Options->Capitalize first letter of every sentence)
> it can suppress OpenOffice.org's automatic capitalisation after a full
> stop for certain defined cases. If I type 
> 	"apples, pears, etc. for sale", 
> I don't want it to automatically change it to 
> 	"apples, pears, etc. For sale" (with capital 'F')
> by assuming that the full stop indicated an end of sentence. This is
> specified in SentenceExceptList.xml
> 
> ** In the case where Tools->Autocorrect->Options->Correct TWo INitial
> CApitals is enabled, the patterns specified in WordExceptList.xml can be
> used to override the behaviour in certain cases. In English it is used
> for keeping the capitalisation of "CDs" as is, for example, without
> changing it to "Cds".
> 
> The functionality is therefore meant to enhance typing speed and
> accuracy in the light of human errors and autocorrection that would
> otherwise cause errors. 
> 
> My patch adds build targets to the language independent makefile. If the
> 3 XML files are present in the languages "autocorr" directory, typing
> "make autocorr" will create the jar file to be used (it defaults to
> building for OO.o 2.0 - for OO.o 1.1, use "make autocorr1"). My previous
> post gives an example of the three XML files, but the English ones
> supplied with OO.o can simply be unzipped to see the English files. No
> java utilities are needed. Only zip.
> 
> The generated jar file is automatically used if it is placed in one (or
> both) of the share/autocorr or user/autocorr directories. For
> OpenOffice.org 2.0 the file is named (for Afrikaans, South Africa)
> acor_af-ZA.dat. For OpenOffice.org 1.1 it is called acor1078.dat (where
> 1078 is Afrikaans' LCID). No "registering" or installation is required
> as is the case with the dictionaries. 
> 
> I don't know how many localisation projects are using this or would be
> interested, though. I see German and French already had autocorrection
> in 1.1.3. I didn't find much documentation on how this works, so I
> thought sharing my finding and integrating in the build process might be
> useful to someone. 
> 
> Friedel
> 
> 
> 
> -------------------------------------------------------
> SF.Net email is sponsored by:
> Tame your development challenges with Apache's Geronimo App Server. 
> Download it for free - -and be entered to win a 42" plasma tv or your very
> own Sony(tm)PSP.  Click here to play: http://sourceforge.net/geronimo.php
> _______________________________________________
> Translate-devel mailing list
> Tra...@li...
> https://lists.sourceforge.net/lists/listinfo/translate-devel
-- 
Dwayne Bailey
Translate.org.za

+27-12-460-1095 (w)
+27-83-443-7114 (cell)