Euro-letter encoding (was: Dense Encoding)
Hans Hagen said this at Sun, 21 Aug 2005 18:36:21 +0200:
so the advised font encodings will be: texnansi, qx, t5, dense
Okay, thinking about it, dense isn't a great name, either. Just about all the 8-bit encodings are dense--they're just dense with different stuff. So with that in mind, I present euro-letters: http://homepage.mac.com/atl/tex/euro-letter.pdf http://homepage.mac.com/atl/tex/euro-letter.zip In the course of things, I found that for whatever reason, the unic-030 (VN) vector was off by two slots. I added [ST]commaaccent to enco-uc (should be in enco-def, too) and unic-002. The key files, however, are enco-el.tex and eurolett.enc. The tester file was brought to life simply by installing: texfont --ma --in --ve=public --co=lm --so=auto --en=eurolett \ --pat="lmvtt*" Please run with normal text--anything you can get your hands on. I am nearly a monoglot, and don't have a lot of stuff lying around that would actually put this stuff to the test. -- =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Adam T. Lindsay, Computing Dept. atl@comp.lancs.ac.uk Lancaster University, InfoLab21 +44(0)1524/510.514 Lancaster, LA1 4WA, UK Fax:+44(0)1524/510.492 -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Adam Lindsay wrote:
So with that in mind, I present euro-letters: http://homepage.mac.com/atl/tex/euro-letter.pdf http://homepage.mac.com/atl/tex/euro-letter.zip
Nice! I have promised myself not to spend too much time on encodings and regimes, so I doubt I will have time to do testing (besides, I do not have any input in european-but- not-latin1 encodings anyway), but nice work! Taco
Adam Lindsay said this at Mon, 22 Aug 2005 00:59:30 +0100:
So with that in mind, I present euro-letters: http://homepage.mac.com/atl/tex/euro-letter.pdf http://homepage.mac.com/atl/tex/euro-letter.zip
Okay, I did some further tests, and noticed that the following should be added to enco-cas: \defineULcharacter Ccedilla ccedilla \defineULcharacter Gcedilla gcedilla \defineULcharacter Scommaaccent scommaaccent \defineULcharacter Tcommaaccent tcommaaccent \defineULcharacter Thorn thorn \defineULcharacter Eth eth \defineULcharacter Lslash lslash \defineULcharacter Ntilde ntilde \defineULcharacter Yacute yacute \defineULcharacter Ygrave ygrave Here's one for the TeXperts: do I need to move things around so that there's no letter in slot 0? I just got reminded of the \lccode rules thanks to Lars's posting on the .etx format. Also, is an endash more desirable in slot 32 than the current quotesinglebase? adam (example test) \WORD{\Abreve \Gbreve \Aogonek \Eogonek \Iogonek \Uogonek \Ccedilla \Scedilla \Gcedilla \Kcedilla \Lcedilla \Ncedilla \Rcedilla \Scommaaccent \Tcommaaccent \Idotaccent} -- =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Adam T. Lindsay, Computing Dept. atl@comp.lancs.ac.uk Lancaster University, InfoLab21 +44(0)1524/510.514 Lancaster, LA1 4WA, UK Fax:+44(0)1524/510.492 -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Adam Lindsay wrote:
Here's one for the TeXperts: do I need to move things around so that there's no letter in slot 0? I just got reminded of the \lccode rules thanks to Lars's posting on the .etx format.
Yes, true. I think one of the quote markers is best, with the matching one in 0x10, or perhaps a non-letter ligature (endash or emdash).
Also, is an endash more desirable in slot 32 than the current quotesinglebase?
possibly. The slot should probably contain a non-letter, in any case. Taco
Adam Lindsay wrote:
Here's one for the TeXperts: do I need to move things around so that there's no letter in slot 0? I just got reminded of the \lccode rules thanks to Lars's posting on the .etx format.
just put a cha there that plays no role in hyphenation or casing Hans ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | fax: 038 477 53 74 | www.pragma-ade.com | www.pragma-pod.nl -----------------------------------------------------------------
Hey, revised version. Some glyphs shuffled around. Adam Lindsay said this at Mon, 22 Aug 2005 17:32:14 +0100:
So with that in mind, I present euro-letters:
http://homepage.mac.com/atl/tex/euro-letter.v2.pdf http://homepage.mac.com/atl/tex/euro-letter.v2.zip and the tester.pdf (with notes on what it needs in terms of macro support, Hans) from inside the zip: http://homepage.mac.com/atl/tex/tester.pdf adam -- =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Adam T. Lindsay, Computing Dept. atl@comp.lancs.ac.uk Lancaster University, InfoLab21 +44(0)1524/510.514 Lancaster, LA1 4WA, UK Fax:+44(0)1524/510.492 -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Adam, Do you remember why you added /equals yet? (perhaps space or cwm would be more useful) Taco Adam Lindsay wrote:
Hey, revised version. Some glyphs shuffled around.
Adam Lindsay said this at Mon, 22 Aug 2005 17:32:14 +0100:
So with that in mind, I present euro-letters:
http://homepage.mac.com/atl/tex/euro-letter.v2.pdf http://homepage.mac.com/atl/tex/euro-letter.v2.zip and the tester.pdf (with notes on what it needs in terms of macro support, Hans) from inside the zip: http://homepage.mac.com/atl/tex/tester.pdf
adam
Adam Lindsay wrote:
Hey, revised version. Some glyphs shuffled around.
http://homepage.mac.com/atl/tex/euro-letter.v2.pdf http://homepage.mac.com/atl/tex/euro-letter.v2.zip and the tester.pdf (with notes on what it needs in terms of macro support, Hans) from inside the zip: http://homepage.mac.com/atl/tex/tester.pdf
adam --
What about the attached patches? See also my posting "[st]cedilla vs. [st]commaaccent". It's mainly about the naming scheme: it's probably unfair to mix commaaccent and cedilla. I didn't try the patch out, there are probably a couple of things missing, but I would like comments about the naming scheme principle, to consistently write commaaccent instead of cedilla for the cedillas on g, k, l, r, n. At least internally, \rcedilla could still be mapped to \rcommaaccent. Adam had some comments in enco-el about that. enco-def would also have to be fixed in this case and some other files (regi-*) as well. What are the definitions like this one doing in enco-el? %definecharacter Ccircumflex {\buildtextaccent\textcircumflex C} Aren't they already (or at least: shouldn't they be) in enco-def.tex? Is that because those characters now have to be composed from characters from two different encodings? Or perhaps just a reminder of what is still missing in the encoding? Would anyone object if the enco-uc file (and all the other enco-[nameofregime] files) would be sorted in the order of unicode insted of grouping the characters? Currently it's difficult to sort out which definitions could still be missing. Mojca
Mojca Miklavec said this at Wed, 24 Aug 2005 00:27:00 +0200:
What about the attached patches? See also my posting "[st]cedilla vs. [st]commaaccent".
I like them. I try(!) to be minimal and conservative in the changes I propose, but I'm always in favour of being "correct"... especially if ConTeXt is going to be described in a "definitive" reference...
It's mainly about the naming scheme: it's probably unfair to mix commaaccent and cedilla.
Agreed.
I didn't try the patch out, there are probably a couple of things missing, but I would like comments about the naming scheme principle, to consistently write commaaccent instead of cedilla for the cedillas on g, k, l, r, n. At least internally, \rcedilla could still be mapped to \rcommaaccent. Adam had some comments in enco-el about that.
enco-def would also have to be fixed in this case and some other files (regi-*) as well.
Yes.
What are the definitions like this one doing in enco-el? %definecharacter Ccircumflex {\buildtextaccent\textcircumflex C} Aren't they already (or at least: shouldn't they be) in enco-def.tex?
They were just a convenient marker (with these blocks copied directly from enco-def) to note what *isn't* in the encoding, but are among ConTeXt's characters. Note the comment (%) at the start of the line. :)
Is that because those characters now have to be composed from characters from two different encodings? Or perhaps just a reminder of what is still missing in the encoding?
The reminder. Note that the composed character might (should?) be entirely from the companion font. All we need is a full alphabet and all accents. :)
Would anyone object if the enco-uc file (and all the other enco-[nameofregime] files) would be sorted in the order of unicode insted of grouping the characters? Currently it's difficult to sort out which definitions could still be missing.
I have no objections! Sounds like a useful way to clean things up... -- =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Adam T. Lindsay, Computing Dept. atl@comp.lancs.ac.uk Lancaster University, InfoLab21 +44(0)1524/510.514 Lancaster, LA1 4WA, UK Fax:+44(0)1524/510.492 -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
participants (4)
-
Adam Lindsay -
Hans Hagen Outside -
Mojca Miklavec -
Taco Hoekwater