Hi Jacko, There is some confusion about tcaron. The problem si that when another name is used, am2tfm cannot create proper small caps when there is no matching upper/lowercase slot. What do you advice in this. Should the ec enc file be changed? 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 -----------------------------------------------------------------
Hans Hagen wrote:
well, since mojca want some additional glyphs as well, why not make an ecx.enc file that fixes these things; actually, you only need that file for generating metrics and rename the ecx-* files to ec-* afterwards; we can tweak texfotn to use a different encoding name and output name if needed.
I don't need any additional glyphs, it's only that the stuff that should work, doesn't work.
Should the ec enc file be changed?
Thanks, Patrick for pointing me to this files in CTAN. Now I saw another couple of strange things, that HAVE TO be corrected in order to make things work properly. - In ec encoding, dstroke (dcroat according to the Adobe glyph name) is named dbar. In my opinion, this should be renamed as soon as possible and also synchronized for the fonts present in CTAN. dbar is against any standard naming scheme. - In xl2 encoding, which states to be T1 + ISO Latin 2 in the upper part, someone made a mistake at the glyph F0, which is eth, but should be dcroat !!! The mistake was easy to make since Eth equals Dcroat, but eth does not equal dcroat. The one who wrote the file didn't notice that. (consecutively, D0 is also wrong) The Vietnamese encoding seems to be the only one when things were done OK. Can someone do anything about that? It seems to be a similar "bug" as tcaron. And about the ConTeXt-specific: dcroat and Eth are both already present in ec encoding, so I'm not asking for any additional glyphs. It's not OK to use Eth when someone asks for Dcroat (Dstroke), but it's better than using the improvized (althoug corrected) version, pseudoencodedDJ. My question was: can anything be done, so that dcroat will be chosen instead of \pseudoencodeddj and that Eth will be chosen instead of \pseudoencodedDJ when someone types \dstroke or \Dstroke (when in ec encoding)? Btw: which languages use Eth? Is it more common than Dcroat? Hans, thank you for including the new version of \pseudoencodedDJ in the distribution before I even managed to go online :) Thank you, Mojca
Mojca Miklavec wrote:
And about the ConTeXt-specific: dcroat and Eth are both already present in ec encoding, so I'm not asking for any additional glyphs. It's not OK to use Eth when someone asks for Dcroat (Dstroke), but it's better than using the improvized (althoug corrected) version, pseudoencodedDJ. My question was: can anything be done, so that dcroat will be chosen instead of \pseudoencodeddj and that Eth will be chosen instead of \pseudoencodedDJ when someone types \dstroke or \Dstroke (when in ec encoding)?
just send me the patch because otherwise i will mess things up 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 -----------------------------------------------------------------
Hans Hagen wrote:
Mojca Miklavec wrote:
And about the ConTeXt-specific: dcroat and Eth are both already present in ec encoding, so I'm not asking for any additional glyphs. It's not OK to use Eth when someone asks for Dcroat (Dstroke), but it's better than using the improvized (althoug corrected) version, pseudoencodedDJ. My question was: can anything be done, so that dcroat will be chosen instead of \pseudoencodeddj and that Eth will be chosen instead of \pseudoencodedDJ when someone types \dstroke or \Dstroke (when in ec encoding)?
just send me the patch because otherwise i will mess things up
I would if I knew/understood at which place ConTeXt decides whether to select a glyph from a font or to make a composed character (or use a macro such as \pseudoencodeddj). Mojca
Mojca Miklavec wrote:
Hans Hagen wrote:
Mojca Miklavec wrote:
And about the ConTeXt-specific: dcroat and Eth are both already present in ec encoding, so I'm not asking for any additional glyphs. It's not OK to use Eth when someone asks for Dcroat (Dstroke), but it's better than using the improvized (althoug corrected) version, pseudoencodedDJ. My question was: can anything be done, so that dcroat will be chosen instead of \pseudoencodeddj and that Eth will be chosen instead of \pseudoencodedDJ when someone types \dstroke or \Dstroke (when in ec encoding)?
just send me the patch because otherwise i will mess things up
I would if I knew/understood at which place ConTeXt decides whether to select a glyph from a font or to make a composed character (or use a macro such as \pseudoencodeddj).
if you use regimes: input encoding font encoding <active char> => \namedglyph => 8bit char | fallback so, it depends on it the 8 bit char being defined in the font encoding, else there is a fallback onto the default encoding 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 -----------------------------------------------------------------
Hans Hagen wrote:
if you use regimes:
input encoding font encoding
<active char> => \namedglyph => 8bit char | fallback
OK, thanks. I didn't check if this works, but I hope the problem will finally be solved now: In enco-ec.tex change \definecharacter dmacron 158 into \definecharacter dstroke 158 and add \definecharacter Dstroke 208 % perhaps also something like \definecharacter dcroat 158 % but I think that you deal with synonyms at some place Please also take a look at lang-hr.pat (or, better, please grep for all occurencies of dbar and dmacron and see if you can get rid of as many of them as possible). Did I miss any other files? Another remark. I'm not sure which name is better, dstroke (unicode-based) or dcroat (adobe-based). ConTeXt seems to have a strange mixture of unicode- and adobe- based names. "hungarumlaut" is adobe-based, while "diaeresis" is unicode-based for example. I don't want to introduce just another name, incompatible with any standards. This additional patch below is the wrong way of doing things. It gives wrong result in the sense that the copy-paste will result in wrong character codes (and that a Croat won't notice that he's using the wrong encoding, just as I didn't notice anything weird in ccaron for a long time), but the resulting glyphs in PDF will be better and in the printed material this won't be seen, which is the most important part. If you use fallback, the results are worse in both senses anyway (the document is not any more searchable). So, can you please also add \definecharacter Dstroke 208 to enco-qx.tex and enco-ans.tex? And add also a comment that the glyphs are identical, but the charecters differ, just in case anyone wonders a couple of years later what these dstrokes are doing there. Thank you, Mojca PS: So, let me count: dstroke, dmacron, dbar, dcroat, [wrong] eth, ... any other suggestions for standardisation ;) ?
I'm now synchronized with latest distro (2005.08.05) and pdfetex 1.3 Under http://www.logosrl.it/context/ there are *)modules.pdf (~21MByte) *) modules.zip, used to make modules.pdf *)patches.zip, a zip file with following patches: core-rul.patch font-chi.patch spec-fdf.patch supp-lan.patch supp-pdf.patch *) modules.log I found some problem with mill.png: I put it under base/ and it was correctly included. Also, partial indeces are not ok. luigi
I'm now synchronized with latest distro (2005.08.05) and pdfetex 1.3
Under http://www.logosrl.it/context/ there are
*)modules.pdf (~21MByte)
Great! I would suggest to keep the old files and name them like 2005-08-16-modules.pdf (and perhaps to keep all the files (except ConTeXt itself of course) since they have to be updated from time to time as new files are added to the distribution). That way the old content will not misteriously disappear. It would be a nice idea to put that files on the garden anyway (together with museum or on its own). Mojca
Mojca Miklavec wrote:
I would suggest to keep the old files and name them like 2005-08-16-modules.pdf
done
(and perhaps to keep all the files (except ConTeXt itself of course) since they have to be updated from time to time as new files are added to the distribution). That way the old content will not misteriously disappear.
patches.zip contains patches for files with some typo-errors (eg. $x4 instead of $x$) they are (almost) irrelevant at format-generation time, because they are not processed at all . Also, some macros are in dutch (eg, \blank[2*groot]); I excluded some others macros (\startpostponing ), correct some others ( \getvalue{period graybox} with \dotgray) added \loadmap[gbk].... and so on. Partial contents are ugly (or broken ?) ; there are many important modules to document (see Core Commands in modules.pdf; inside modules.tex, 3=>ok, 1=>to be documented, 0=> to be exctract) There are some overfull lines; maybe there is a better sectioning (or multiple pdfs ?). So, I think that this is a good starting point I think that all patches should be approvated before to go into an official distro; after that we can discuss all other stuffs.
It would be a nice idea to put that files on the garden anyway
no problem to me luigi
luigi.scarso wrote:
There are some overfull lines; maybe there is a better sectioning (or multiple pdfs ?).
i'll look into that later; also, it makes sense to used condensed monospaced latin modern fonts (will be available in a couple of weeks) 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 -----------------------------------------------------------------
luigi.scarso wrote:
*)patches.zip, a zip file with following patches: core-rul.patch font-chi.patch spec-fdf.patch supp-lan.patch supp-pdf.patch
patched 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 -----------------------------------------------------------------
Mojca Miklavec wrote:
Another remark. I'm not sure which name is better, dstroke (unicode-based) or dcroat (adobe-based). ConTeXt seems to have a strange mixture of unicode- and adobe- based names. "hungarumlaut" is adobe-based, while "diaeresis" is unicode-based for example. I don't want to introduce just another name, incompatible with any standards.
take the one used in latin modern 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 -----------------------------------------------------------------
Hans Hagen wrote:
Mojca Miklavec wrote:
Another remark. I'm not sure which name is better, dstroke (unicode-based) or dcroat (adobe-based). ConTeXt seems to have a strange mixture of unicode- and adobe- based names. "hungarumlaut" is adobe-based, while "diaeresis" is unicode-based for example. I don't want to introduce just another name, incompatible with any standards.
take the one used in latin modern
Hmmm ... which one? dbar or dcroat? ;) (/fonts/lm/doc/fonts/lm/0inf0983.txt tells that both of them are present) Well, I didn't mean that seriously, I know that dbar is there just because of compatibility with the "old" ec.enc files, but ... both in adobe glyph list and unicode the situation is just horrible. "stroke" in unicode is represented by 3 different names (croat, bar, slash): - dcroat - hbar - lslash - oslash - tbar Guess why dstroke/dcroat is named dbar in the ec encoding ... And I forgot yet another possibility in my list of names for that single stupid letter that seems to be used nowhere except in Croatian and Vietnamese ... \dj. Plus the one which should be used as a "fallback" in Slovenian: \definecharacter dstroke {d\zcaron} Out of the seven possibilities I vote for \dstroke and leave the strange name (dcroat) to be handled by *.enc files only. Mojca
Mojca Miklavec wrote:
Out of the seven possibilities I vote for \dstroke and leave the strange name (dcroat) to be handled by *.enc files only.
i think that your ultimate satisfaction can come from this private regime: \starttext \startregime[mojka] \dostepwiserecurse{127}{255}{1} {\expanded{\defineactivetoken\recurselevel\space}{\dstroke}} \stopregime \enableregime[mojka] ¡¢£ ¤¥¦§¨© ª«¬®¯°± ²³´µ¶·¸ ¹º»¼½¾ ¿ÀÁÂà ÄÅ ÆÇÈÉ ÊË ÌÍÎÏÐ ÑÒÓÔ ÕÖ×ØÙÚ Û ÜÝÞßàáâ ãäåæçè éêëì íîï ðñòóôõ ö÷øùúûüýþÿ ¡¢£ ¤¥¦§¨© ª«¬®¯°± ²³´µ¶·¸ ¹º»¼½¾ ¿ÀÁÂà ÄÅ ÆÇÈ ÉÊË ÌÍÎÏÐ ÑÒÓÔ ÕÖ×ØÙÚ Û ÜÝÞ ßàáâ ãäåæçè éêë ìíîï ðñòóôõ ö÷øùúûüýþÿ ¡¢£ ¤¥¦§¨© ª«¬®¯°± ²³´µ¶·¸ ¹º»¼½¾ ¿ÀÁÂà ÄÅ ÆÇÈÉ ÊË ÌÍÎÏÐ ÑÒÓÔ ÕÖ×ØÙÚ Û ÜÝÞßàáâ ãäåæçè éêëìí îï ðñòóôõ ö÷øùúûüýþÿ ¡¢£ ¤¥¦§¨© ª«¬®¯°± ²³´µ¶·¸ ¹º»¼½¾ ¿ÀÁÂà ÄÅ ÆÇÈ ÉÊË ÌÍÎÏÐ ÑÒÓÔ ÕÖ×ØÙÚ Û ÜÝÞßàáâ ãäåæçè éêëì íîï ðñòóôõ ö÷øùúûüýþÿ \stoptext 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 -----------------------------------------------------------------
Mojca Miklavec wrote:
Guess why dstroke/dcroat is named dbar in the ec encoding ...
guess: because no one was involved who knew better than that
And I forgot yet another possibility in my list of names for that single stupid letter that seems to be used nowhere except in Croatian and Vietnamese ... \dj. Plus the one which should be used as a "fallback" in Slovenian: \definecharacter dstroke {d\zcaron}
that's an easy one ... and could be within a boring background run ... see attachment: \useencoding[fsl] \starttext \dstroke \language[sl] \dstroke \stoptext of course only visible when the fall bakc is in use (so for testing, replace default by ec)
Out of the seven possibilities I vote for \dstroke and leave the strange name (dcroat) to be handled by *.enc files only.
dcroat is only in enco-def as fallback (does not hurt) 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 -----------------------------------------------------------------
Mojca Miklavec wrote:
Another remark. I'm not sure which name is better, dstroke (unicode-based) or dcroat (adobe-based). ConTeXt seems to have a strange mixture of unicode- and adobe- based names. "hungarumlaut" is adobe-based, while "diaeresis" is unicode-based for example. I don't want to introduce just another name, incompatible with any standards.
on a rainy day we can sort it out and go unicode and use the adobe ones as synoyms 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 -----------------------------------------------------------------
participants (3)
-
Hans Hagen -
luigi.scarso -
Mojca Miklavec