[Lcdproc] [PATCH] 2 wrong symbols in HD44780 driver charmap

[David Lopez Perez]

> Hi David,
> 
> Your new mapping is not equal to the original Hitachi HD44780 mapping.
> I think you have used documentation of a different display claiming to
> be HD44780 compliant. There are many different charsets, and there 
> are also multiple charsets of the original HD44780 (extensions A00
> and A02).


My LCDs have Hitachi chips labeled "HD44780" on the back, and the documentation I used is from the Hitachi Semiconductor HD44780 datasheet:

http://pdf1.alldatasheet.com/datasheet-pdf/view/63673/HITACHI/HD44780.html

 > There is an n with ~ in the original char set. I guess the char was
> mapped to a normal n was because there were too many charsets that did
> not support it. Maybe the HD44780 standard could better have been sticked
> to.

Well, I am sticking to the standard: the original charset table on the official datasheet from Hitachi Semiconductor, extension A00 (page 17).
Comparing that table to the HD44780_default charmap, you can see that the greek letter mu is currently being mapped to the glyph for the large 'y', when there is a proper glyph for the mu letter on the same table.
Also, the HD44780_default charmap already maps 'a', 'u' and 'o' with diacritic mark (¨), both upper and lowercase,  to the proper lowercase glyphs - IMHO it should do the same for the 'N/n' + '~' because the glyph is also there on the standard charmap.

> Some years ago we defined to translate from ISO 8859-1. Since -15 is
> practically compatible maybe that one is better. You can always add your
> own translation char set. Maybe more of these should exist to choose
> from for "almost compatible" displays. Why don't these manufacturers
> just stick to the existing standard ?

My original intention was to add my own new charmap, but then I realized
that the HD44780_default one was quite right, only those two characters were wrong, so I decided to change it
instead. I'm not against adding more charmaps to improve compatibility
for other LCDs that do not stick to the standard, but my LCD does, and the default charmap does not.
Or should we have two different charmaps "HD44780_default" and "HD44780_standard"?



That's the beautiful thing about standards: there are so many to choose from...

> Regards,
> Joris
> 
> 
> > The '?' (greek letter mu) was mapped to a different glyph (11111001 in the HD44780 table) when it should be 11100100.
> > 
> > Also, the letter '?' (n with a ~ on top) was mapped to the same glyph as the 'n'. There is a glyph in the table for the '?' (11101110), and it is a completely different letter from the n, so it should be mapped to this glyph. Same goes for the uppercase '?' (just like '?' and '?' are both mapped to the '?' glyph).
> > 
> > Attached is a patch for LCDproc 0.5.2 to fix these issues.
> > 
> > --- hd44780-charmap.h    2009-09-18 01:33:04.000000000 +0200
> > +++ hd44780-charmap.h.original    2009-09-18 01:30:43.000000000 +0200
> > @@ -59,17 +59,17 @@
> >          /* #160 */
> >          160,  33, 236, 237, 164,  92, 124, 167,
> >           34, 169, 170, 171, 172, 173, 174, 175,
> > -        223, 177, 178, 179,  39, 228, 247, 165,
> > +        223, 177, 178, 179,  39, 249, 247, 165,
> >           44, 185, 186, 187, 188, 189, 190,  63,
> >          /* #192 */
> >           65,  65,  65,  65, 225,  65,  65,  67,
> >           69,  69,  69,  69,  73,  73,  73,  73,
> > -         68, 238,  79,  79,  79,  79, 239, 120,
> > +         68,  78,  79,  79,  79,  79, 239, 120,
> >           48,  85,  85,  85, 245,  89, 240, 226,
> >          /* #224 */
> >           97,  97,  97,  97, 225,  97,  97,  99,
> >          101, 101, 101, 101, 105, 105, 105, 105,
> > -        111, 238, 111, 111, 111, 111, 239, 253,
> > +        111, 110, 111, 111, 111, 111, 239, 253,
> >           48, 117, 117, 117, 245, 121, 240, 255
> >  };
> >  
> > 

_________________________________________________________________
Comparte tus mejores momentos del verano ¡Hazlo con Windows Live Fotos!
http://www.vivelive.com/compartirfotos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.omnipotent.net/pipermail/lcdproc/attachments/20090920/92d44c98/attachment.htm>