Re: [Cjk] Re: how can I use your Unicode.sfd?
From: "Edward G.J. Lee" <[EMAIL PROTECTED]> > On Mon, Sep 04, 2006, Danai SAE-HAN wrote: > > From: zhaohs <[EMAIL PROTECTED]> > > > > >2. In the pdf files that are generated by dvipdfm(x),I find > > > the Type1 fonts are used,so they are scalable.In the former > > > case--that is--generating thepdf files by dvipdf,the chinese > > > charaters can't be copy and pasted;in the latter,the chinese > > > charaters can be copy and pasted. > > > > Hmmm, dunno about copying and pasting. Do you use evince or acroread? > > AFAIK, only dvipdfmx can embed ToUnicode information into PDF, > of cource your Type 1 must have Unicode(UCS2/4) mapping already. > * FontForge will do this job. > > pdf[la]tex/dvipdfm/dvips/ps2pdf/dvipdf don't have this ability > yet. > > So CJKV people use dvipdfmx is a good choice I think. Thanks for that update! I used dvipdfmx anyway, but never had the need to copy and paste from my own PDF files, since I got the source. =D Best regards Danai SAE-HAN 韓達耐 -- 題目:《秋思》 作者:陸游（１１２５－１２１０） 利欲驅人萬火牛，江湖浪跡一沙鷗。 日長似歲閑方覺，事大如山醉亦休。 衣杵相望深巷月，井桐搖落故園秋。 欲舒老眼無高處，安得元龍百尺樓。 ___ Cjk maillist - Cjk@ffii.org http://lists.ffii.org/mailman/listinfo/cjk
Re: [Cjk] Re: how can I use your Unicode.sfd?
On Mon, Sep 04, 2006 at 03:35:52PM +0200, Danai SAE-HAN wrote:
>
> Aloha!
>
> From: zhaohs <[EMAIL PROTECTED]>
>
> > Now I have put the latest Unicode.sfd into the following place
> > to substitute the original ones:
> > /usr/share/texmf/ttf2pk/Unicode.sfd
> > /usr/share/texmf/ttf2tfm/Unicode.sfd
>
> That's incorrect. teTeX3 and the latest TeXlive now use TeX Directory
> Structure v1.1 (TDS1.1). All .sfd files must be installed in
> $(LOCAL)TEXMF/fonts/sfd/ . So in your case that would be in
> /usr/(local/)share/texmf/fonts/sfd/Unicode.sfd .
>
>
> > But I can't generate pdf files even though I done all the
> > things that you told me.Following are my questions now:
>
> > 1. When I use the command "pdflatex","dvipdf"or"dvips+ps2pdf"
> > to generate the pdf file, I can get the pdf file ,but in it,I can't
> > find the rare-used chinese characters,furthmore,even the chinese
> > glyphs that have appeared in the pdf file are the Type3 fonts.I find
> > them are not scalable glyph--why dose this happen?
>
> Do the rare-used Chinese characters appear in xdvi?
> And I don't have any Type3 fonts on my computer, so I wouldn't know.
> Make sure that, after you get the sunext.map, it is put in
> /usr/local/share/texmf/fonts/map/ and that you run "update-updmap" and
> "upmap-sys" (I suppose Ubuntu uses the same tools as Debian).
>
>
> >2. In the pdf files that are generated by dvipdfm(x),I find
> > the Type1 fonts are used,so they are scalable.In the former
> > case--that is--generating thepdf files by dvipdf,the chinese
> > charaters can't be copy and pasted;in the latter,the chinese
> > charaters can be copy and pasted.
>
> Hmmm, dunno about copying and pasting. Do you use evince or acroread?
>
>
> > 3. How can I get the ps or pdf file with type1 or truetype
> > fonts otherwise with the pk charaters in it,it is not
> > scalable. The following information show the fonts that I
> > have been embedded in my pdf files.
>
> IOW (In other words), dvipdfm and dvipdfmx have the fonts embedded,
> dvipdf and dvips+ps2pdf don't. Dunno why. Make sure you remove any
> leftovers from failed installations. And perhaps remove all the PK
> fonts that MAKETEXPK might have created from sunext.
>
> Strange that pdflatex didn't embed them. Make sure you remove those
> .enc files.
>
>
> > 4. When you generate the subfonts with the sunfonts.pe ,you
> > have said that "The 00 plane created from Sun-ExtB.ttf can
> > +be deleted,why and why it isn't the case for Sun-ExtA.ttf,on second
> > thoughts，can we revise the script avoid this.
>
> The nasty thing is that Sun-Ext uses two fonts instead of one. Both
> have the 00 plane, but since Sun-ExtA.ttf also covers a *huge* amount
> of non-Chinese glyphs (letters with accents, etc.) and Sun-ExtB.ttf
> covers none of them, it is evident that the 00 plane of Sun-ExtB.ttf
> is only there as a placeholder.
>
> So delete the 00 plane of Sun-ExtB.ttf and use the one from
> Sun-ExtA.ttf.
>
> The Fontforge script expects only one TTF. It's not worth the trouble
> just to take account of this one exception. It is much easier and
> more transparent just to delete the files.
>
> Now there might be some problems to get vertical files working
> (sunextv and sunextvr), but I'll probably come to that later.
>
>
> > 5. You have said that:
> [...]
>
> > But you don't have sunexta.map anyway,I think the
> > sunexta.map should be sunext.map,what do you think about it?
>
> Indeed, a typo! Shame on me. ;D
>
>
> > On the other hand, the 10sunext.cfg file can also be named
> > like 10local.cfg or so,and serveral Map entries can be
> > +wrote in it,do you think so?
>
> You can name it anything you like. The number in front "10" will
> determine where exactly the .cfg file will be put in the final
> updmap.cfg file. If you put "99", it will come as last. If you put
> 10, it will be somewhere at the beginning. It doesn't really matter.
> Anything above 00 is okay.
>
>
> > 6.In the instructions that you have give to me,you didn't use
> >the .enc files,why or why can I use them?
>
> Dunno the technical details about that one. Just don't use them.
>
>
> > 7.If I use the subfonts.pe to generate subfonts for other
> >fonts,should I always delete the "The 00 plane created"
> >subfonts or not?
>
> No. The first plane, 00, contains the ASCII characters. If you use
> \usepackage[T1]{CJKutf8} then you can indeed delete the 00 plane, but
> if you use \usepackage{CJKutf8} without any font encoding, then CJK
> will only use the font that you specify in \begin{CJK}{xxx}.
>
> So it's best not to delete the first plane, just in case.
>
>
> Cheers
>
>
>
> Danai SAE-HAN
> 韓達耐
>
> --
> 題目:《偶題》
> 作者:張耒（１０５２－１１１２）
>
> 相逢記得畫橋頭，花似精神柳似柔。
> 莫謂無情即無語，春風傳意水傳愁。
Dear Danai,
Thank you very much!From your and Werner's le
Re: [Cjk] Re: how can I use your Unicode.sfd?
> many slide macro packages just prefer pdftex, and copy & paste also can > be available plus with some diversion of cmap package. Oh yes, CCT's ccmap pcakage can do this job well. > I think dvipdfmx just produce a smaller cid fonts pdf together with copy > & paste, which is the most value. It seems that pdfTeX people do not like compact Type 1 font? :) Edward ___ Cjk maillist - Cjk@ffii.org http://lists.ffii.org/mailman/listinfo/cjk
Re: [Cjk] Re: how can I use your Unicode.sfd?
> Because I have veyr poor knowledge about the glyph or font,so I
> have to ask you for advice again,how can I add a call to the
> script subfonts.pe with name SetFontNames(),can you give me a
> example?
Umm, have you actually looked at the FontForge documentation link I
gave you earlier? The change is trivial. Just insert
SetFontNames("HaiTi")
or something similar somewhere before the Generate(...) line in
subfonts.pe.
Werner
___
Cjk maillist - Cjk@ffii.org
http://lists.ffii.org/mailman/listinfo/cjk
Re: [Cjk] Re: how can I use your Unicode.sfd?
On Tue, Sep 05, 2006 at 07:27:41AM +0200, Werner LEMBERG wrote:
>
> > When I use the subfonts.pe to generate the subfonts form
> > simhei.ttf,though the ultimate files can be generated,but I also see
> > the following error messages:
> >
> > The Postscript font name "黑体" is invalid.
> > It should be printable ASCII,
> > must not contain (){}[]<>%/ or space
> > and must be shorted than 63 characters
>
> This is not an error message (which would cause fontforge to abort)
> but a warning, and you can it ignore, I think. Of course, as Danai
> has written, it would be better to fix it adding a call to
> SetFontNames() (to a local copy of the subfonts.pe script, of course).
>
>
> Werner
Dear Werner,
Because I have veyr poor knowledge about the glyph or font,so I have to ask
you for advice again,how can I add a call to the script subfonts.pe with name
SetFontNames(),can you give me a example?
Best regards.
Hongsheng.
__
�Ͽ�ע���Ż�?
http://cn.mail.yahoo.com
___
Cjk maillist - Cjk@ffii.org
http://lists.ffii.org/mailman/listinfo/cjk
Re: [Cjk] Re: how can I use your Unicode.sfd?
> many slide macro packages just prefer pdftex, and copy & paste also > can be available plus with some diversion of cmap package. Exactly. Nobody has yet `diverted' to do that for the CJK ranges -- I've just done a single CJK cmap file for proving the concept of cmap.sty (since I've added the code to cmap.sty so that it cooperates with the CJK package :-). Volunteers welcome. > I think dvipdfmx just produce a smaller cid fonts pdf together with > copy & paste, which is the most value. Hopefully, pdftex will soon provide similar functionality. Werner ___ Cjk maillist - Cjk@ffii.org http://lists.ffii.org/mailman/listinfo/cjk
Re: [Cjk] Re: how can I use your Unicode.sfd?
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Edward G.J. Lee wrote: > On Mon, Sep 04, 2006, Danai SAE-HAN wrote: >> From: zhaohs <[EMAIL PROTECTED]> >> >>>2. In the pdf files that are generated by dvipdfm(x),I find >>> the Type1 fonts are used,so they are scalable.In the former >>> case--that is--generating thepdf files by dvipdf,the chinese >>> charaters can't be copy and pasted;in the latter,the chinese >>> charaters can be copy and pasted. >> Hmmm, dunno about copying and pasting. Do you use evince or acroread? > > AFAIK, only dvipdfmx can embed ToUnicode information into PDF, > of cource your Type 1 must have Unicode(UCS2/4) mapping already. > * FontForge will do this job. > > pdf[la]tex/dvipdfm/dvips/ps2pdf/dvipdf don't have this ability > yet. > > So CJKV people use dvipdfmx is a good choice I think. > > > > Edward > > ___ > Cjk maillist - Cjk@ffii.org > http://lists.ffii.org/mailman/listinfo/cjk > many slide macro packages just prefer pdftex, and copy & paste also can be available plus with some diversion of cmap package. I think dvipdfmx just produce a smaller cid fonts pdf together with copy & paste, which is the most value. -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.5 (GNU/Linux) iD8DBQFE/Q0jexs2apGZoVgRAvuLAKDb+1mUCv7hkzaIYE4OKU92vCZvMQCgnkKh 6Jxj7LK6ow+wO1aZLmvIrb8= =OmJH -END PGP SIGNATURE- ___ Cjk maillist - Cjk@ffii.org http://lists.ffii.org/mailman/listinfo/cjk
Re: [Cjk] Re: how can I use your Unicode.sfd?
> > I have no idea. Please send a *small* sample document to the list > > which shows your problem (perhaps containing two characters only, one > > which can be displayed and one which is missing), together with the > > PDF output. > > The attachment that I'v give to you can be obtain from: > http://lists.ffii.org/pipermail/cjk/attachments/20060904/bb88c845/jpg.tar-0001.obj > You can download it than rename it with jpg.tar.gz. You completely misunderstood me. What you've sent are images which show some font information. This is quite useless for further testing. I want to see an *input* document, together with the resulting *output*, this is, the PDF you have created. >sunext00.afm,sunext00.enc,sunext00.pfb.sunext00.tfm. >Danai SAE-HAN told me to delete the four files that generated >from Sun-ExtB.ttf, and use the four files that are generated from >Sun-ExtA.ttf,why? I think Danai has answered that sufficiently already. Werner ___ Cjk maillist - Cjk@ffii.org http://lists.ffii.org/mailman/listinfo/cjk
Re: [Cjk] Re: how can I use your Unicode.sfd?
> When I use the subfonts.pe to generate the subfonts form
> simhei.ttf,though the ultimate files can be generated,but I also see
> the following error messages:
>
> The Postscript font name "黑体" is invalid.
> It should be printable ASCII,
> must not contain (){}[]<>%/ or space
> and must be shorted than 63 characters
This is not an error message (which would cause fontforge to abort)
but a warning, and you can it ignore, I think. Of course, as Danai
has written, it would be better to fix it adding a call to
SetFontNames() (to a local copy of the subfonts.pe script, of course).
Werner
___
Cjk maillist - Cjk@ffii.org
http://lists.ffii.org/mailman/listinfo/cjk
Re: [Cjk] Re: how can I use your Unicode.sfd?
On Mon, Sep 04, 2006, Danai SAE-HAN wrote: > From: zhaohs <[EMAIL PROTECTED]> > > >2. In the pdf files that are generated by dvipdfm(x),I find > > the Type1 fonts are used,so they are scalable.In the former > > case--that is--generating thepdf files by dvipdf,the chinese > > charaters can't be copy and pasted;in the latter,the chinese > > charaters can be copy and pasted. > > Hmmm, dunno about copying and pasting. Do you use evince or acroread? AFAIK, only dvipdfmx can embed ToUnicode information into PDF, of cource your Type 1 must have Unicode(UCS2/4) mapping already. * FontForge will do this job. pdf[la]tex/dvipdfm/dvips/ps2pdf/dvipdf don't have this ability yet. So CJKV people use dvipdfmx is a good choice I think. Edward ___ Cjk maillist - Cjk@ffii.org http://lists.ffii.org/mailman/listinfo/cjk
Re: [Cjk] Re: how can I use your Unicode.sfd?
zhaohs schreef:
> When I use the subfonts.pe to generate the subfonts form simhei.ttf,though
> the ultimate files can be generated,but I also see the following error
> messages:
>
> ##
> [EMAIL PROTECTED]:~/Desktop/simhei$ fontforge -script ../subfonts.pe
> ../simhei.ttf hei ../Unicode.sfd
> Copyright (c) 2000-2005 by George Williams.
> Executable based on sources from 12:08 5-Dec-2005.
> Loading ../simhei.ttf...
> The Postscript font name "黑体" is invalid.
> It should be printable ASCII,
> must not contain (){}[]<>%/ or space
> and must be shorted than 63 characters
> ###
> #
>Why should this happen,how I can resolve the problem?
Well, that's pretty obvious, isn't it? Call it "heiti" or "simhei"
and it won't appear anymore. :P
All files in $TEXMF should be written in ASCII. I know that's not
fair, but that's just the way it is.
Technically it's not impossible, but it will give you headaches.
Now on Windows, font files can also be in Chinese, but Windows will
usually just screw up and use a wrong encoding, making the filename
show up as gibberish, "mojibake", 乱码. On an English Win2k setup,
almost all the Chinese fonts that are installed there will be screwed
up, for example the Sun-Ext{A,B}.ttf fonts.
Danai SAE-HAN
韓達耐
--
題目:《春懷示鄰里》
作者:陳師道（１０５３－１１０１）
斷牆著雨蝸成字，老屋無僧燕作家，
剩欲出門追語笑，卻嫌歸鬢著塵沙。
風翻蛛網開三面，雷動蜂窠趁兩衙。
屢失南鄰春事約，只今容有未開花。
___
Cjk maillist - Cjk@ffii.org
http://lists.ffii.org/mailman/listinfo/cjk
Re: [Cjk] Re: how can I use your Unicode.sfd?
Dear Werner,
When I use the subfonts.pe to generate the subfonts form simhei.ttf,though
the ultimate files can be generated,but I also see the following error
messages:
##
[EMAIL PROTECTED]:~/Desktop/simhei$ fontforge -script ../subfonts.pe
../simhei.ttf hei ../Unicode.sfd
Copyright (c) 2000-2005 by George Williams.
Executable based on sources from 12:08 5-Dec-2005.
Loading ../simhei.ttf...
The Postscript font name "黑体" is invalid.
It should be printable ASCII,
must not contain (){}[]<>%/ or space
and must be shorted than 63 characters
###
#
Why should this happen,how I can resolve the problem?
Best regards
Hongsheng.
__
�Ͽ�ע���Ż�?
http://cn.mail.yahoo.com
___
Cjk maillist - Cjk@ffii.org
http://lists.ffii.org/mailman/listinfo/cjk
Re: [Cjk] Re: how can I use your Unicode.sfd?
From: zhaohs <[EMAIL PROTECTED]>
> Now,I also puzzled by the the followinf things:
> When I use the script subfonts.pe within the Cjk-current
> package,which you have told me,to generate the subfonts of
> Sun-ExtA.ttf and Sun-ExtB.ttf,I can get the *.enc,*.afm,*.tmf
> and *.pfb files,but why the Sun-ExtA.ttf and the Sun-ExtB.ttf
> both can generate the following four files with the same name:
>sunext00.afm,sunext00.enc,sunext00.pfb.sunext00.tfm.
>Yor told me to delete the four files that generated from
> Sun-ExtB.ttf, and use the four files that are generated from
> Sun-ExtA.ttf,why?
>I also wonder that though the four files are with the same name，
> they are different subfonts,why can we delete them arbitrarily?
You don't delete them arbitrarily. Have a look at the
Sun-Ext{A,B}.ttf fonts, either with Fontforge itself (requires a *LOT*
of memory) or with gfontview (which is very fast, and shows plane per
plane; switch anti-alias on), and you'll see what I mean.
Now it's pretty simple why they get the same name: you gave "sunext"
as one of the arguments of the fontforge command when processing both
fonts. Usually, when you compile two fonts, each font should have its
own name, obviously. But apparently Sun Haifeng decided to split one
font into two fonts.
Fonts A and B have a 00 plane, but A has a lot of other scripts
besides Chinese. Just have a look at it with gfontview or fontforge
if you don't believe me. The B font only has a handful of ASCII
characters and the rest are Chinese characters from Extension B.
So keep the 00 plane of font A, and remove the one from font B.
>Furthmore,why the subfonts.pe will generate the files with the
>same name on 00 plane,is this a deficiency of the subfonts.pe or
>the original fonts have problems?If I use the script to generate
>subfonts for other coding such as GBK, will the same thing happen
>or not?
LoL! That's certainly *not* a deficiency of subfonts.pe. If there's
one deficiency, then it would be Sun Haifeng's fonts since it is split
into two fonts. Normally TrueType Fonts only have *one* TTF (there
can be more fonts for the different variants). So there's nothing
wrong with the Fontforge script. Why is deleting the 00 plane of the
B font so problematic for you? o_O?
I didn't try to convert it yet to the GBK encoding. You can try to
create subfonts using the GBK encoding yourself. You might also try
to use "virtual fonts": they use the Unicode fonts but use the GBK
codes.
Have a look at the uni2sfd.pl Perl file. For more information, view
the Perl files and have a look at the files in doc/.
Cheers
Danai SAE-HAN
韓達耐
--
題目:《無題》
作者:晏殊（９９１－１０５５）
油壁香車不再逢，峽云無跡任西東。
梨花院落溶溶月，柳絮池塘淡淡風。
几日寂寥傷酒后，一番蕭瑟禁煙中。
魚書欲寄何由達，水遠山長處處同。
___
Cjk maillist - Cjk@ffii.org
http://lists.ffii.org/mailman/listinfo/cjk
Re: [Cjk] Re: how can I use your Unicode.sfd?
Danai SAE-HAN The attachment that I'v give to you can be obtain from: http://lists.ffii.org/pipermail/cjk/attachments/20060904/bb88c845/jpg.tar-0001.obj You can download it than rename it with jpg.tar.gz. Now,I also puzzled by the the followinf things: When I use the script subfonts.pe within the Cjk-current package,which you have told me,to generate the subfonts of Sun-ExtA.ttf and Sun-ExtB.ttf,I can get the *.enc,*.afm,*.tmf and *.pfb files,but why the Sun-ExtA.ttf and the Sun-ExtB.ttf both can generate the following four files with the same name: sunext00.afm,sunext00.enc,sunext00.pfb.sunext00.tfm. Yor told me to delete the four files that generated from Sun-ExtB.ttf, and use the four files that are generated from Sun-ExtA.ttf,why? I also wonder that though the four files are with the same name，they are different subfonts,why can we delete them arbitrarily? Furthmore,why the subfonts.pe will generate the files with the same name on 00 plane,is this a deficiency of the subfonts.pe or the original fonts have problems?If I use the script to generate subfonts for other coding such as GBK, will the same thing happen or not? Best regards Hongsheng. __ �Ͽ�ע���Ż�? http://cn.mail.yahoo.com ___ Cjk maillist - Cjk@ffii.org http://lists.ffii.org/mailman/listinfo/cjk
Re: [Cjk] Re: how can I use your Unicode.sfd?
On Mon, Sep 04, 2006 at 06:33:06PM +0200, Werner LEMBERG wrote: > > 1. When I use the command "pdflatex","dvipdf"or"dvips+ps2pdf" to > > generate the pdf file, I can get the pdf file, but in it, I can't > > find the rare-used chinese characters,furthmore, even the chinese > > glyphs that have appeared in the pdf file are the Type3 fonts. I > > find them are not scalable glyph--why dose this happen? > > I have no idea. Please send a *small* sample document to the list > which shows your problem (perhaps containing two characters only, one > which can be displayed and one which is missing), together with the > PDF output. > > > 2. In the pdf files that are generated by dvipdfm(x), I find the > > Type1 fonts are used, so they are scalable. In the former > > case--that is--generating the pdf files by dvipdf,the chinese > > charaters can't be copy and pasted; in the latter,the chinese > > charaters can be copy and pasted. > > This is correct. dvipdfm doesn't have support for these things. You > should generally avoid dvipdf completely, using dvipdfmx only. > > > Werner Dear Werner. The attachment that I'v give to you can be obtain from: http://lists.ffii.org/pipermail/cjk/attachments/20060904/bb88c845/jpg.tar-0001.obj You can download it than rename it with jpg.tar.gz. When I use the script subfonts.pe within the Cjk-current package,as Danai SAE-HAN have told me,to generate the subfonts of Sun-ExtA.ttf and Sun-ExtB.ttf,I can get the *.enc,*.afm,*.tmf and *.pfb files,but why the Sun-ExtA.ttf and the Sun-ExtB.ttf both can generate the following four files with the same name: sunext00.afm,sunext00.enc,sunext00.pfb.sunext00.tfm. Danai SAE-HAN told me to delete the four files that generated from Sun-ExtB.ttf, and use the four files that are generated from Sun-ExtA.ttf,why? I also wonder that though the four files are with the same name，they are different subfonts,why can we delete them arbitrarily. Furthmore,why the subfonts.pe will generate the files with the same name on 00 plane,is this a deficiency of the subfonts.pe or the original fonts have problems?If I use the script to generate subfonts for other coding such as GBK, will the same thing happen or not? Best regards Hongsheng. __ �Ͽ�ע���Ż�? http://cn.mail.yahoo.com ___ Cjk maillist - Cjk@ffii.org http://lists.ffii.org/mailman/listinfo/cjk
Re: [Cjk] Re: how can I use your Unicode.sfd?
From: Werner LEMBERG <[EMAIL PROTECTED]> > > This is pdfeTeX, Version 3.141592-1.21a-2.2 (Web2C 7.5.4) > > > Warning: pdflatex (file > > /usr/share/texmf-tetex/fonts/truetype/sunext/Sun-ExtB.t tf): glyph > > `u20053' not found > > Hmm, I'm not sure whether this version of pdfeTeX can handle Unicode > characters > U+. I have the same version, and it works for me. > > pdflatex: writettf.c:702: prepend_subset_tags: Assertion `index > 0 > > && index < name_record_num && fm_cur->subset_tag != ((void *)0)' > > failed. > > This is clearly a bug in pdfeTeX. Look at > http://sarovar.org/projects/pdftex/; there you can see that the > current stable version of pdftex is 1.30.6. I suppose this bug has > been fixed meanwhile. > > Anyway, I really don't recommend pdfTeX currently for documents which > include CJK fonts! As far as I know, dvipdfmx produces documents > which are far smaller (with the additional benefit of proper > cut-and-paste support out of the box). True; dvipdfmx produces files which are abuot 40% smaller. And you're right too about copying and pasting: it works, even in evince. Olé! Danai SAE-HAN 韓達耐 -- 題目:《書湖陰先生壁》 作者:王安石（１０２１－１０８６） 茅檐長掃靜無苔，花木成畦手自栽。 一水護田將綠繞，兩山排闥送青來。 ___ Cjk maillist - Cjk@ffii.org http://lists.ffii.org/mailman/listinfo/cjk
Re: [Cjk] Re: how can I use your Unicode.sfd?
> 1. When I use the command "pdflatex","dvipdf"or"dvips+ps2pdf" to > generate the pdf file, I can get the pdf file, but in it, I can't > find the rare-used chinese characters,furthmore, even the chinese > glyphs that have appeared in the pdf file are the Type3 fonts. I > find them are not scalable glyph--why dose this happen? I have no idea. Please send a *small* sample document to the list which shows your problem (perhaps containing two characters only, one which can be displayed and one which is missing), together with the PDF output. > 2. In the pdf files that are generated by dvipdfm(x), I find the > Type1 fonts are used, so they are scalable. In the former > case--that is--generating the pdf files by dvipdf,the chinese > charaters can't be copy and pasted; in the latter,the chinese > charaters can be copy and pasted. This is correct. dvipdfm doesn't have support for these things. You should generally avoid dvipdf completely, using dvipdfmx only. Werner ___ Cjk maillist - Cjk@ffii.org http://lists.ffii.org/mailman/listinfo/cjk
Re: [Cjk] Re: how can I use your Unicode.sfd?
Hi!
From: zhaohs <[EMAIL PROTECTED]>
> Now when I use the pdflatex,I find the error messages as follows:
>
> ###
> [EMAIL PROTECTED]:~/Desktop/test$ pdflatex pdflatex.tex
> This is pdfeTeX, Version 3.141592-1.21a-2.2 (Web2C 7.5.4)
> entering extended mode
> (./pdflatex.tex
> LaTeX2e <2003/12/01>
> Babel and hyphenation patterns for loaded.
> (/usr/share/texmf-tetex/tex/latex/base/article.cls
> Document Class: article 2004/02/16 v1.4f Standard LaTeX document class
> (/usr/share/texmf-tetex/tex/latex/base/size12.clo))
> (/usr/share/texmf/tex/latex/CJK/CJKutf8.sty
> (/usr/share/texmf-tetex/tex/latex/base/inputenc.sty
> (/usr/share/texmf/tex/latex/ucs/utf8.def))
> (/usr/share/texmf/tex/latex/ucs/ucs.sty
> (/usr/share/texmf/tex/latex/ucs/data/uni-global.def))
Why are you using the UCS package? Remove it from nobmp2.tex.
It will only interfere with CJK.
> (/usr/share/texmf/tex/latex/CJK/CJK.sty
> (/usr/share/texmf/tex/latex/CJK/mule/MULEenc.sty)
> (/usr/share/texmf/tex/latex/CJK/CJK.enc))
> (/usr/share/texmf-tetex/tex/latex/base/fontenc.sty))
> (/usr/share/texmf-tetex/tex/latex/tools/array.sty) (./pdflatex.aux)
> (/usr/share/texmf/tex/latex/ucs/ucsencs.def)
> (/usr/share/texmf/tex/latex/CJK/UTF8/UTF8.bdg)
> (/usr/share/texmf/tex/latex/CJK/UTF8/UTF8.enc)
> (/usr/share/texmf/tex/latex/CJK/UTF8/UTF8.chr)
> (/usr/share/texmf/tex/latex/CJK/UTF8/c70sunext.fd)
> [1{/var/lib/texmf/fonts/map/
> pdftex/updmap/pdftex.map}] (./pdflatex.aux)
> ) 1/bluesky/cm/cmr12.pfb>{/usr/share/texmf-tetex/fonts/enc/dvips/sunext/sunext020
> 0.enc} Warning: pdflatex (file
> /usr/share/texmf-tetex/fonts/truetype/sunext/Sun-ExtB.t
> tf): glyph `u20053' not found
>
> Warning: pdflatex (file
> /usr/share/texmf-tetex/fonts/truetype/sunext/Sun-ExtB.t
> tf): glyph `u20061' not found
> pdflatex: writettf.c:702: prepend_subset_tags: Assertion `index > 0 && index
> < name_record_num && fm_cur->subset_tag != ((void *)0)' failed.
> 已放弃
> #
> What's the reason?How can I resolve the problem?
The problem is that pdflatex uses the TTF, not the Type1 fonts.
And it still uses the .enc files. Make sure you remove *all* of the
.enc files in /usr/share/texmf/ and /usr/local/share/texmf/.
BTW (By the way), you should NEVER install things yourself in
/usr/share/texmf-tetex/! Only in /usr/local/share/texmf/. The
reason is that /usr/share/texmf-tetex/ is reserved only for files
installed by Debian packages. If you install things yourself, put
them in /usr/local/share/texmf/. It makes it a *lot* easier, for
example now, to see what you have installed so you can remove them as
well.
After you have removed the .enc files, run "texhash" as root again to
update the different ls-R files.
If you have changed 10local.cfg, also run "update-updmap" and
"updmap-sys" as root.
Best regards
Danai SAE-HAN
韓達耐
--
題目:《新開棋軒呈元珍表臣》
作者:歐陽修（１００７－１０７２）
竹樹日已滋，軒窗漸幽興。
人間與世遠，鳥語知境靜。
春光靄欲布，山色寒尚映。
獨收萬籟心，于此一枰競。
___
Cjk maillist - Cjk@ffii.org
http://lists.ffii.org/mailman/listinfo/cjk
Re: [Cjk] Re: how can I use your Unicode.sfd?
> This is pdfeTeX, Version 3.141592-1.21a-2.2 (Web2C 7.5.4) > Warning: pdflatex (file > /usr/share/texmf-tetex/fonts/truetype/sunext/Sun-ExtB.t tf): glyph > `u20053' not found Hmm, I'm not sure whether this version of pdfeTeX can handle Unicode characters > U+. > pdflatex: writettf.c:702: prepend_subset_tags: Assertion `index > 0 > && index < name_record_num && fm_cur->subset_tag != ((void *)0)' > failed. This is clearly a bug in pdfeTeX. Look at http://sarovar.org/projects/pdftex/; there you can see that the current stable version of pdftex is 1.30.6. I suppose this bug has been fixed meanwhile. Anyway, I really don't recommend pdfTeX currently for documents which include CJK fonts! As far as I know, dvipdfmx produces documents which are far smaller (with the additional benefit of proper cut-and-paste support out of the box). Werner ___ Cjk maillist - Cjk@ffii.org http://lists.ffii.org/mailman/listinfo/cjk
Re: [Cjk] Re: how can I use your Unicode.sfd?
Dear Werner. Thank you for your advice. Now I have put the latest Unicode.sfd to the followint place to substitute the original ones: /usr/share/texmf/ttf2pk/Unicode.sfd /usr/share/texmf/ttf2tfm/Unicode.sfd But I can't generate pdf files even though I done all the things that you told me.Following are my questions now: 1. When I use the command "pdflatex","dvipdf"or"dvips+ps2pdf" to generate the pdf file, I can get the pdf file ,but in it,I can't find the rare-used chinese characters,furthmore,even the chinese glyphs that have appeared in the pdf file are the Type3 fonts.I find them are not scalable glyph--why dose this happen? 2. In the pdf files that are generated by dvipdfm(x),I find the Type1 fonts are used,so they are scalable.In the former case--that is--generating thepdf files by dvipdf,the chinese charaters can't be copy and pasted;in the latter,the chinese charaters can be copy and pasted. 3. How can I get the ps or pdf file with type1 or truetype fonts otherwise with the pk charaters in it,it is not scalable. The attachments show the fonts that have been embedded in my pdf files. So you can see it more clearly.The file names are accorrding to the pdf's generating ways. 4. > Do you have fonts for those Unicode ranges (in the above it is > U+21000-U+210FF)? And have you used the right Unicode.sfd file posted > recently to this list (and which is part of the CVS repository of the > ttf2pk package)? Older versions lack definitions for Unicode values > greater than U+. Yes,I have,you can alse download the font from the following URL: http://okuc.net/software/UniFonts.exe, Best regards Hongsheng. jpg.tar.gz Description: Binary data ___ Cjk maillist - Cjk@ffii.org http://lists.ffii.org/mailman/listinfo/cjk
Re: [Cjk] Re: how can I use your Unicode.sfd?
Aloha!
From: zhaohs <[EMAIL PROTECTED]>
> Now I have put the latest Unicode.sfd into the following place
> to substitute the original ones:
> /usr/share/texmf/ttf2pk/Unicode.sfd
> /usr/share/texmf/ttf2tfm/Unicode.sfd
That's incorrect. teTeX3 and the latest TeXlive now use TeX Directory
Structure v1.1 (TDS1.1). All .sfd files must be installed in
$(LOCAL)TEXMF/fonts/sfd/ . So in your case that would be in
/usr/(local/)share/texmf/fonts/sfd/Unicode.sfd .
> But I can't generate pdf files even though I done all the
> things that you told me.Following are my questions now:
> 1. When I use the command "pdflatex","dvipdf"or"dvips+ps2pdf"
> to generate the pdf file, I can get the pdf file ,but in it,I can't
> find the rare-used chinese characters,furthmore,even the chinese
> glyphs that have appeared in the pdf file are the Type3 fonts.I find
> them are not scalable glyph--why dose this happen?
Do the rare-used Chinese characters appear in xdvi?
And I don't have any Type3 fonts on my computer, so I wouldn't know.
Make sure that, after you get the sunext.map, it is put in
/usr/local/share/texmf/fonts/map/ and that you run "update-updmap" and
"upmap-sys" (I suppose Ubuntu uses the same tools as Debian).
>2. In the pdf files that are generated by dvipdfm(x),I find
> the Type1 fonts are used,so they are scalable.In the former
> case--that is--generating thepdf files by dvipdf,the chinese
> charaters can't be copy and pasted;in the latter,the chinese
> charaters can be copy and pasted.
Hmmm, dunno about copying and pasting. Do you use evince or acroread?
> 3. How can I get the ps or pdf file with type1 or truetype
> fonts otherwise with the pk charaters in it,it is not
> scalable. The following information show the fonts that I
> have been embedded in my pdf files.
IOW (In other words), dvipdfm and dvipdfmx have the fonts embedded,
dvipdf and dvips+ps2pdf don't. Dunno why. Make sure you remove any
leftovers from failed installations. And perhaps remove all the PK
fonts that MAKETEXPK might have created from sunext.
Strange that pdflatex didn't embed them. Make sure you remove those
.enc files.
> 4. When you generate the subfonts with the sunfonts.pe ,you
> have said that "The 00 plane created from Sun-ExtB.ttf can
> +be deleted,why and why it isn't the case for Sun-ExtA.ttf,on second
> thoughts，can we revise the script avoid this.
The nasty thing is that Sun-Ext uses two fonts instead of one. Both
have the 00 plane, but since Sun-ExtA.ttf also covers a *huge* amount
of non-Chinese glyphs (letters with accents, etc.) and Sun-ExtB.ttf
covers none of them, it is evident that the 00 plane of Sun-ExtB.ttf
is only there as a placeholder.
So delete the 00 plane of Sun-ExtB.ttf and use the one from
Sun-ExtA.ttf.
The Fontforge script expects only one TTF. It's not worth the trouble
just to take account of this one exception. It is much easier and
more transparent just to delete the files.
Now there might be some problems to get vertical files working
(sunextv and sunextvr), but I'll probably come to that later.
> 5. You have said that:
[...]
> But you don't have sunexta.map anyway,I think the
> sunexta.map should be sunext.map,what do you think about it?
Indeed, a typo! Shame on me. ;D
> On the other hand, the 10sunext.cfg file can also be named
> like 10local.cfg or so,and serveral Map entries can be
> +wrote in it,do you think so?
You can name it anything you like. The number in front "10" will
determine where exactly the .cfg file will be put in the final
updmap.cfg file. If you put "99", it will come as last. If you put
10, it will be somewhere at the beginning. It doesn't really matter.
Anything above 00 is okay.
> 6.In the instructions that you have give to me,you didn't use
>the .enc files,why or why can I use them?
Dunno the technical details about that one. Just don't use them.
> 7.If I use the subfonts.pe to generate subfonts for other
>fonts,should I always delete the "The 00 plane created"
>subfonts or not?
No. The first plane, 00, contains the ASCII characters. If you use
\usepackage[T1]{CJKutf8} then you can indeed delete the 00 plane, but
if you use \usepackage{CJKutf8} without any font encoding, then CJK
will only use the font that you specify in \begin{CJK}{xxx}.
So it's best not to delete the first plane, just in case.
Cheers
Danai SAE-HAN
韓達耐
--
題目:《偶題》
作者:張耒（１０５２－１１１２）
相逢記得畫橋頭，花似精神柳似柔。
莫謂無情即無語，春風傳意水傳愁。
___
Cjk maillist - Cjk@ffii.org
http://lists.ffii.org/mailman/listinfo/cjk
Re: [Cjk] Re: how can I use your Unicode.sfd?
Hi!
From: zhaohs <[EMAIL PROTECTED]>
> (/usr/share/texmf/tex/latex/CJK/UTF8/c70sunext.fd)
> [1{/var/lib/texmf/fonts/map/
> pdftex/updmap/pdftex.map}] (./pdflatex.aux)
> ) 1/bluesky/cm/cmr12.pfb>{/usr/share/texmf-tetex/fonts/enc/dvips/sunext/sunext020
> 0.enc} Warning: pdflatex (file
> /usr/share/texmf-tetex/fonts/truetype/sunext/Sun-ExtB.t
> tf): glyph `u20053' not found
>
> Warning: pdflatex (file
> /usr/share/texmf-tetex/fonts/truetype/sunext/Sun-ExtB.t
> tf): glyph `u20061' not found
> pdflatex: writettf.c:702: prepend_subset_tags: Assertion `index > 0 && index
> < name_record_num && fm_cur->subset_tag != ((void *)0)' failed.
> 已放弃
I haven't had a look at your prior email, but you must also delete
*all* the .enc files. Don't install them.
Yes, sunexta.map must be sunext.map. It's a typo.
And yes, you can name sunext.cfg anything you like. You can name it
local.cfg, or something else. That's totally up to you.
Cheers
Danai SAE-HAN
韓達耐
--
題目:《春日》
作者:秦觀（１０４９－１１００）
一夕輕雷落萬絲，霽光浮瓦碧參差。
有情芍藥含春淚，無力薔薇臥曉枝。
___
Cjk maillist - Cjk@ffii.org
http://lists.ffii.org/mailman/listinfo/cjk
Re: [Cjk] Re: how can I use your Unicode.sfd?
Dear Danai SAE-HAN,
Now when I use the pdflatex,I find the error messages as follows:
###
###
[EMAIL PROTECTED]:~/Desktop/test$ pdflatex pdflatex.tex
This is pdfeTeX, Version 3.141592-1.21a-2.2 (Web2C 7.5.4)
entering extended mode
(./pdflatex.tex
LaTeX2e <2003/12/01>
Babel and hyphenation patterns for loaded.
(/usr/share/texmf-tetex/tex/latex/base/article.cls
Document Class: article 2004/02/16 v1.4f Standard LaTeX document class
(/usr/share/texmf-tetex/tex/latex/base/size12.clo))
(/usr/share/texmf/tex/latex/CJK/CJKutf8.sty
(/usr/share/texmf-tetex/tex/latex/base/inputenc.sty
(/usr/share/texmf/tex/latex/ucs/utf8.def))
(/usr/share/texmf/tex/latex/ucs/ucs.sty
(/usr/share/texmf/tex/latex/ucs/data/uni-global.def))
(/usr/share/texmf/tex/latex/CJK/CJK.sty
(/usr/share/texmf/tex/latex/CJK/mule/MULEenc.sty)
(/usr/share/texmf/tex/latex/CJK/CJK.enc))
(/usr/share/texmf-tetex/tex/latex/base/fontenc.sty))
(/usr/share/texmf-tetex/tex/latex/tools/array.sty) (./pdflatex.aux)
(/usr/share/texmf/tex/latex/ucs/ucsencs.def)
(/usr/share/texmf/tex/latex/CJK/UTF8/UTF8.bdg)
(/usr/share/texmf/tex/latex/CJK/UTF8/UTF8.enc)
(/usr/share/texmf/tex/latex/CJK/UTF8/UTF8.chr)
(/usr/share/texmf/tex/latex/CJK/UTF8/c70sunext.fd) [1{/var/lib/texmf/fonts/map/
pdftex/updmap/pdftex.map}] (./pdflatex.aux) ){/usr/share/texmf-tetex/fonts/enc/dvips/sunext/sunext020
0.enc} 0 && index <
name_record_num && fm_cur->subset_tag != ((void *)0)' failed.
已放弃
#
#
What's the reason?How can I resolve the problem?
Best regards!
Hongsheng
__
�Ͽ�ע���Ż�?
http://cn.mail.yahoo.com
___
Cjk maillist - Cjk@ffii.org
http://lists.ffii.org/mailman/listinfo/cjk
Re: [Cjk] Re: how can I use your Unicode.sfd?
Dear Danai SAE-HAN, Thank you for your and Werner's advice. Now I have put the latest Unicode.sfd into the following place to substitute the original ones: /usr/share/texmf/ttf2pk/Unicode.sfd /usr/share/texmf/ttf2tfm/Unicode.sfd But I can't generate pdf files even though I done all the things that you told me.Following are my questions now: 1. When I use the command "pdflatex","dvipdf"or"dvips+ps2pdf" to generate the pdf file, I can get the pdf file ,but in +it,I can't find the rare-used chinese characters,furthmore,even the chinese glyphs that have appeared in the pdf file are +the Type3 fonts.I find them are not scalable glyph--why dose this happen? 2. In the pdf files that are generated by dvipdfm(x),I find the Type1 fonts are used,so they are scalable.In the +former case--that is--generating thepdf files by dvipdf,the chinese charaters can't be copy and pasted;in the latter,the +chinese charaters can be copy and pasted. 3. How can I get the ps or pdf file with type1 or truetype fonts otherwise with the pk charaters in it,it is not +scalable. The following information show the fonts that I have been embedded in my pdf files. So you can see it more clearly: (I):when I use the latex to generate dvi,then use dvipdf or dvips+ps2pdf to generate pdf files,the ultimate pdf files have the same fonts informations,they are as follows: English fonts have the information such as: CMTT12(已嵌入子集) 类型：Type1 编码：Ansi Chinese fonts have the information such as: T3Font_192 类型：Type3 编码：自定义 实际字体：T3Font_192 实际字型：Type3 (II):when I use the pdflatex to generate pdf files,the ultimate pdf files have the same fonts informations,they are as follows: English fonts have the information such as: CMTT12(已嵌入子集) 类型：Type1 编码：自定义 Chinese fonts have the information such as: F30 类型：Type3 编码：自定义 实际字体：F30 实际字型：Type3 (III):when I use the latex to generate dvi,then use dvipdfm to generate pdf files,the ultimate pdf files have the informations as follows: English fonts have the information such as: CMTT12(已嵌入子集) 类型：Type1 编码：自定义 Chinese fonts have the information such as: sunext0200(已嵌入子集) 类型：Type1 编码：内建 (IV):when I use the latex to generate dvi,then use dvipdfmx to generate pdf files,the ultimate pdf files have the informations as follows: English fonts have the information such as: CMTT12(已嵌入子集) 类型：Type1 编码：内建 Chinese fonts have the information such as: sunext0200(已嵌入子集) 类型：Type1 编码：内建 4. When you generate the subfonts with the sunfonts.pe ,you have said that "The 00 plane created from Sun-ExtB.ttf can +be deleted,why and why it isn't the case for Sun-ExtA.ttf,on second thoughts，can we revise the script avoid this. 5. You have said that: $ for filename in *.pfb; do echo "$(basename $filename .pfb) $(basename $filename .pfb) <$filename" \ >> ../../map/dvips/sunext/sunext.map; done # # But in the same letter, you also said: ## ## On Debian you would have to create a file called /etc/texmf/updmap.d/10sunext.cfg with the following lines: ## # 10sunext.cfg Map sunexta.map ## ## ## But you don't have sunexta.map anyway,I think the sunexta.map should be sunext.map,what do you think about it? On the other hand, the 10sunext.cfg file can also be named like 10local.cfg or so,and serveral Map entries can be +wrote in it,do you think so? 6.In the instructions that you have give to me,you didn't use the .enc files,why or why can I use them? 7.If I use the subfonts.pe to generate subfonts for other fonts,should I always delete the "The 00 plane created" subfonts or not? Best regards Hongsheng. __ �Ͽ�ע���Ż�? http://cn.mail.yahoo.com ___ Cjk ma
Re: [Cjk] Re: how can I use your Unicode.sfd?
Dear Werner, Thank you for your advice. Now I have put the latest Unicode.sfd into the following place to substitute the original ones: /usr/share/texmf/ttf2pk/Unicode.sfd /usr/share/texmf/ttf2tfm/Unicode.sfd But I can't generate pdf files even though I done all the things that you told me.Following are my questions now: 1. When I use the command "pdflatex","dvipdf"or"dvips+ps2pdf" to generate the pdf file, I can get the pdf file ,but in +it,I can't find the rare-used chinese characters,furthmore,even the chinese glyphs that have appeared in the pdf file are +the Type3 fonts.I find them are not scalable glyph--why dose this happen? 2. In the pdf files that are generated by dvipdfm(x),I find the Type1 fonts are used,so they are scalable.In the +former case--that is--generating thepdf files by dvipdf,the chinese charaters can't be copy and pasted;in the latter,the +chinese charaters can be copy and pasted. 3. How can I get the ps or pdf file with type1 or truetype fonts otherwise with the pk charaters in it,it is not +scalable. The following information show the fonts that I have been embedded in my pdf files. So you can see it more clearly: (I):when I use the latex to generate dvi,then use dvipdf or dvips+ps2pdf to generate pdf files,the ultimate pdf files have the same fonts informations,they are as follows: English fonts have the information such as: CMTT12(已嵌入子集) 类型：Type1 编码：Ansi Chinese fonts have the information such as: T3Font_192 类型：Type3 编码：自定义 实际字体：T3Font_192 实际字型：Type3 (II):when I use the pdflatex to generate pdf files,the ultimate pdf files have the same fonts informations,they are as follows: English fonts have the information such as: CMTT12(已嵌入子集) 类型：Type1 编码：自定义 Chinese fonts have the information such as: F30 类型：Type3 编码：自定义 实际字体：F30 实际字型：Type3 (III):when I use the latex to generate dvi,then use dvipdfm to generate pdf files,the ultimate pdf files have the informations as follows: English fonts have the information such as: CMTT12(已嵌入子集) 类型：Type1 编码：自定义 Chinese fonts have the information such as: sunext0200(已嵌入子集) 类型：Type1 编码：内建 (IV):when I use the latex to generate dvi,then use dvipdfmx to generate pdf files,the ultimate pdf files have the informations as follows: English fonts have the information such as: CMTT12(已嵌入子集) 类型：Type1 编码：内建 Chinese fonts have the information such as: sunext0200(已嵌入子集) 类型：Type1 编码：内建 4. When you generate the subfonts with the sunfonts.pe ,you have said that "The 00 plane created from Sun-ExtB.ttf can +be deleted,why and why it isn't the case for Sun-ExtA.ttf,on second thoughts，can we revise the script avoid this. 5. You have said that: $ for filename in *.pfb; do echo "$(basename $filename .pfb) $(basename $filename .pfb) <$filename" \ >> ../../map/dvips/sunext/sunext.map; done # # But in the same letter, you also said: ## ## On Debian you would have to create a file called /etc/texmf/updmap.d/10sunext.cfg with the following lines: ## # 10sunext.cfg Map sunexta.map ## ## ## But you don't have sunexta.map anyway,I think the sunexta.map should be sunext.map,what do you think about it? On the other hand, the 10sunext.cfg file can also be named like 10local.cfg or so,and serveral Map entries can be +wrote in it,do you think so? 6.In the instructions that you have give to me,you didn't use the .enc files,why or why can I use them? 7.If I use the subfonts.pe to generate subfonts for other fonts,should I always delete the "The 00 plane created" subfonts or not? Best regards Hongsheng. __ �Ͽ�ע���Ż�? http://cn.mail.yahoo.com ___ Cjk maillist - Cjk@ffii.
Re: [Cjk] Re: how can I use your Unicode.sfd?
From: Werner LEMBERG <[EMAIL PROTECTED]> > >1. How can I use the subfonts.pe to generate the slant characters > > and boldface characters of the original font,just as I can do it > > with the ttf2tfm by using the "-s" parameter;furthmore,how I can > > use the commands,such as "\textit" or "\textbf",to generate the > > boldface and italic of the original chinese font. > > Do you really need slanted Chinese characters? This is quite ugly > IMHO... To emphasize something in Chinese it is better practice to > use a different font, or to use underlining in various forms. I agree. For example, you could just switch from MingTi to KaiTi. I myself am not a big fan of underlining, but since Chinese characters all use the same baseline beneath (unlike Latin characters like "g" or "ç") I guess you could also use underlining. > Creating bold face is far more difficult -- I say it is impossible for > Chinese due to the many strokes per glyph. This means that you have > to find a real bold font if you aren't satisfied with the boldface > emulation. I agree. Danai SAE-HAN 韓達耐 -- 題目:《有美堂暴雨》 作者:蘇軾（１０３６－１１０１） 游人腳底一聲雷，滿座頑云撥不開， 天外黑風吹海立，浙東飛雨過江來。 十分瀲灩金樽凸，千杖敲鏗羯鼓催。 喚起謫仙泉酒面，倒傾鮫室瀉瓊瑰。 ___ Cjk maillist - Cjk@ffii.org http://lists.ffii.org/mailman/listinfo/cjk
Re: [Cjk] Re: how can I use your Unicode.sfd?
From: zhaohs <[EMAIL PROTECTED]>
> Thank you for your detailed advices,now I can accomplish the compiling of
> tex file with latex to get the dvi file,than get the pdf file with
> dvipdfm or dvipdfmx,in the case of use the latex and dvipdfmx,the
> characters are embeded in the pdf,and I can copy and paster the
> chinese characters.Thank you very much!
No problem.
> But now I have the following questions:
> 1. How can I use the subfonts.pe to generate the slant characters of the
> original font,just as I can do it with the ttf2tfm by using the "-s"
> parameter;furthmore,how I can use the commands,such as "\textit" or
> "\textbf",to generate the boldface and italic of the original
> chinese font.
That is an excellent question. Not that slanted or italic Chinese
characters are very beautiful, but
I myself wondered as well how to get slanted version automatically. I
remember the option you could give in ttfonts.map. I'm afraid you
will have to read the site that Werner gave you and start scripting
yourself. :/
To get bold fonts the easy way, you need \CJKbold to make "poor man's
bold" fonts. It will essentially copy the glyph several times over
each other, each time putting it one dot on the right, one to the
left, etc., to make the glyph bold. As you see, it works for most
cases, but not if you enlarge it to huge sizes. I'm afraid you'll
have to use another font to get this right.
In c70sunext.fd, you should have the following lines:
\DeclareFontShape{C70}{sunext}{m}{n}{<-> CJK * sunext}{}
\DeclareFontShape{C70}{sunext}{bx}{n}{<-> CJKb * sunext}{\CJKbold}
{m} is medium weight, {bx} is bold extended. {b} is bold, {bc} is
bold condensed, etc.
{n} is normal. {it} is italics, {sl} is slanted. Check the documents
in texdoc for more information.
You can make additional lines for other forms. Note that since you
don't have the forms, just copying the above lines when using other
forms will merely suppress the warnings.
$(virtual_namestem) medium normal
$(virtual_namestem)omedium slanted ("oblique")
$(virtual_namestem)ccondensed normal
$(virtual_namestem)co condensed slanted
$(virtual_namestem)uultracompressed normal
$(virtual_namestem)uo ultracompressed slanted
$(virtual_namestem)xextended normal
$(virtual_namestem)xo extended slanted
$(virtual_namestem)bbold medium normal
Because slant and italics are the same for Chinese characters, both
will point to *o versions.
But all this is rather pointless if you don't have the glyphs anyway.
If you want just to get rid of the warnings, you can substitute all
forms this way.
So either produce bold fonts (see Werner's email) or get yourself a
professional font.
> 2. Now when I use the "\textbf" generate the boldface of the original
> chinese font,when I amplified the corresponding chinese characters by 64
> times,I can see the edge of these "saw-tooth-like" charaters.
That's because of \CJKbold.
> 3. When I use the pdflatex to generate the pdf,I get more error messages and
> can't generate the pdf file which I give you in the attachments--the file
> "the-error-message-of-pdflalex".
Hmm, did you also produce the fonts from Sun-ExtB.ttf?
> 4. Last, when I generate the pdf file with dvips and ps2pdf,I find that the
> chinese characters can't be appear in the pdf file,and I get error
> codes in the pdf file,which I give you in the attachments--the file
> "generate-by-dvips+ps2pdf.pdf" show this case.
Like Werner said, ditch UTF8enc. And make sure you also have those
.pfb files mentioned in sunext.map.
Cheers
Danai SAE-HAN
韓達耐
--
題目:《初入淮河》（選一）
作者:楊萬里（１１２７－１２０６）
船離洪澤岸頭沙，人到淮河意不佳。
何必桑乾方是遠，中流以北即天涯！
___
Cjk maillist - Cjk@ffii.org
http://lists.ffii.org/mailman/listinfo/cjk
Re: [Cjk] Re: how can I use your Unicode.sfd?
>1. How can I use the subfonts.pe to generate the slant characters > and boldface characters of the original font,just as I can do it > with the ttf2tfm by using the "-s" parameter;furthmore,how I can > use the commands,such as "\textit" or "\textbf",to generate the > boldface and italic of the original chinese font. Do you really need slanted Chinese characters? This is quite ugly IMHO... To emphasize something in Chinese it is better practice to use a different font, or to use underlining in various forms. Anyway, slanting can be done by adding some transformation commands to the FontForge script. Please consult its manual for more details: http://fontforge.sourceforge.net/scripting.html and look for the `Transform' command. Creating bold face is far more difficult -- I say it is impossible for Chinese due to the many strokes per glyph. This means that you have to find a real bold font if you aren't satisfied with the boldface emulation. >2. Now,I can use the "\textbf" generate the boldface of the > original chinese font,but when I amplified the corresponding > chinese characters by 64 times,I can see that the echaraters have > "saw-tooth-like" dges. Exactly. Without a real boldface this can't be improved. However, you can control the `boldness' with the \CJKboldshift macro (as documented in CJK.txt). > kpathsea: Running mktexpk --mfmode / --bdpi 600 --mag 1+120/600 --dpi 720 > sunext0210 Do you have fonts for those Unicode ranges (in the above it is U+21000-U+210FF)? And have you used the right Unicode.sfd file posted recently to this list (and which is part of the CVS repository of the ttf2pk package)? Older versions lack definitions for Unicode values greater than U+. > 4. Last, when I generate the pdf file with dvips and ps2pdf,I find > that the chinese characters can't be appear in the pdf file,and I > get mess codes of chinese characters in the pdf file, following the > log file: > (/usr/share/texmf/tex/latex/ucs/utf8x.def > File: utf8x.def 2004/10/17 UCS: Input encoding UTF-8 You shouldn't use the CJK package together with the UCS package except for very special reasons. Don't load `utf8x' -- using CJKutf8.sty should be fully sufficient for Chinese fonts. Werner ___ Cjk maillist - Cjk@ffii.org http://lists.ffii.org/mailman/listinfo/cjk
Re: [Cjk] Re: how can I use your Unicode.sfd?
Dear Danai SAE-HAN,
Thank you for your detailed advices,now I can accomplish the compiling of
tex file with latex to get the dvi file,than get the pdf file with dvipdfm or
dvipdfmx,in the case of use the latex and dvipdfmx,the characters are embeded
in the pdf,and I can copy and paster the chinese characters.Thank you very much!
But now I have the following questions:
1. How can I use the subfonts.pe to generate the slant characters and
boldface characters of the
original font,just as I can do it with the ttf2tfm by using the "-s"
parameter;furthmore,how I can use the commands,such as "\textit" or
"\textbf",to generate the boldface and italic of the original chinese font.
2. Now,I can use the "\textbf" generate the boldface of the original
chinese font,but when I amplified the corresponding chinese characters by 64
times,I can see that the echaraters have "saw-tooth-like" dges.
3. When I use the pdflatex to generate the pdf,I get more error messages and
can't generate the pdf file,the followint are the errot messages:
[EMAIL PROTECTED]:~/Desktop/test$ pdflatex nobmp.tex
This is pdfeTeX, Version 3.141592-1.21a-2.2 (Web2C 7.5.4)
entering extended mode
(./nobmp.tex
LaTeX2e <2003/12/01>
Babel and hyphenation patterns for loaded.
(/usr/share/texmf-tetex/tex/latex/base/article.cls
Document Class: article 2004/02/16 v1.4f Standard LaTeX document class
(/usr/share/texmf-tetex/tex/latex/base/size12.clo))
(/usr/share/texmf/tex/latex/CJK/CJK.sty
(/usr/share/texmf/tex/latex/CJK/mule/MULEenc.sty)
(/usr/share/texmf/tex/latex/CJK/CJK.enc))
(/usr/share/texmf-tetex/tex/latex/tools/array.sty)
(/usr/share/texmf/tex/latex/ucs/ucs.sty
(/usr/share/texmf/tex/latex/ucs/data/uni-global.def))
(/usr/share/texmf-tetex/tex/latex/base/inputenc.sty
(/usr/share/texmf/tex/latex/ucs/utf8x.def)) (./nobmp.aux)
(/usr/share/texmf/tex/latex/ucs/ucsencs.def)
(/usr/share/texmf/tex/latex/CJK/UTF8/UTF8.bdg)
(/usr/share/texmf/tex/latex/CJK/UTF8/UTF8.enc)
(/usr/share/texmf/tex/latex/CJK/UTF8/UTF8.chr)
(/usr/share/texmf/tex/latex/CJK/UTF8/c70sunext.fd)
Overfull \hbox (0.3055pt too wide) in paragraph at lines 32--35
[]\C70/sunext/m/n/12/90 ^^Y \C70/sunext/m/n/12/66 / \C70/sunext/m/n/12/95 � \C7
0/sunext/m/n/12/65 �\OT1/cmr/m/n/12 L[]T[]X CJK \C70/sunext/m/n/12/64 � \C70/su
next/m/n/12/58 �\OT1/cmr/m/n/12 Unicode \C70/sunext/m/n/12/78 � \C70/sunext/m/n
/12/4f M \C70/sunext/m/n/12/81 �\OT1/cmtt/m/n/12 U+10 \C70/sunext/m/n/12/76
� \C70/sunext/m/n/12/6e , \C70/sunext/m/n/12/8a f \C70/sunext/m/n/12/ff ^^L \C
70/sunext/m/n/12/90 ^^Y \C70/sunext/m/n/12/4e � \C70/sunext/m/n/12/53 ^^E \C70/
sunext/m/n/12/62 � \C70/sunext/m/n/12/4e �\OT1/cmr/m/n/12 CJK
[1{/var/lib/texmf/fonts/map/pdftex/updmap/pdftex.map}] (./nobmp.aux) )
kpathsea: Running mktexpk --mfmode / --bdpi 600 --mag 1+120/600 --dpi 720
sunext0210
mktexpk: don't know how to create bitmap font for sunext0210.
kpathsea: Appending font creation commands to missfont.log.
(see the transcript file for additional information)
Warning: pdflatex (file sunext0210): Font sunext0210 at 720 not found
kpathsea: Running mktexpk --mfmode / --bdpi 600 --mag 1+120/600 --dpi 720
sunext020f
mktexpk: don't know how to create bitmap font for sunext020f.
Warning: pdflatex (file sunext020f): Font sunext020f at 720 not found
kpathsea: Running mktexpk --mfmode / --bdpi 600 --mag 1+120/600 --dpi 720
sunext020e
mktexpk: don't know how to create bitmap font for sunext020e.
Warning: pdflatex (file sunext020e): Font sunext020e at 720 not found
kpathsea: Running mktexpk --mfmode / --bdpi 600 --mag 1+120/600 --dpi 720
sunext0270
mktexpk: don't know how to create bitmap font for sunext0270.
Warning: pdflatex (file sunext0270): Font sunext0270 at 720 not found
kpathsea: Running mktexpk --mfmode / --bdpi 600 --mag 1+120/600 --dpi 720
sunext0277
mktexpk: don't know how to create bitmap font for sunext0277.
Warning: pdflatex (file sunext0277): Font sunext0277 at 720 not found
kpathsea: Running mktexpk --mfmode / --bdpi 600 --mag 1+120/600 --dpi 720
sunext02a3
mktexpk: don't know how to create bitmap font for sunext02a3.
Warning: pdflatex (file sunext02a3): Font sunext02a3 at 720 not found
Fatal error occurred, the output PDF file is not finished!
##
##
4. Last, when I generate the pdf file with dvips and ps2pdf,I find that the
chinese characters can't be appear in the pdf file,and I get mess codes of
chinese characters in the pdf file, following the log file:
*
*
This is pdfeTeX, Version 3.141592-1.21a-2.2 (Web2C 7.5.4) (format=latex
2006.8.23) 3 SEP 2006 22:48
entering extended mode
**nobmp.tex
(./nobmp.tex
LaTeX2e <2003/12/01>
Babel and hyphenation patterns for loaded.
(/usr/share/texmf-tetex/tex/latex/base/article.cls
Document Class: article 200
Re: [Cjk] Re: how can I use your Unicode.sfd?
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Danai SAE-HAN (韓達耐) wrote:
> Indeed you should use CJKutf8 instead of CJK. The reason is the
> following.
>
> Once upon a time, each script had its own encodings. This made it
> easy to use different fonts: each encoding would use its own font.
>
> With UTF-8 this is slightly more problematic. And let's face it: the
> Latin script of most Asian fonts just sucks (just look at Arphic). I
> didn't expect anything else, anyway.
>
> If there is no "fontenc" option, all glyphs would use the CJK font.
> To get nice Latin or Cyrillic glyphs, add [T1] or another font
> encoding as an option, and CJK will first use fonts from the selected
> font encoding.
>
> \usepackage{CJKutf8} without any options will always use the CJK
> font.
>
> See doc/CJKutf8.txt for more information.
>
>
> Best regards
>
>
>
> Danai SAE-HAN
> 韓達耐
>
got it, thanks a lot again!
Best wishes
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.5 (GNU/Linux)
iD8DBQFE+tEbexs2apGZoVgRAkCDAKD43tX2uQlcgOqgMZZPSFEh0GWNRACeLxWv
meovN3cqgQsqSo029ZrcxZc=
=qOfW
-END PGP SIGNATURE-
___
Cjk maillist - Cjk@ffii.org
http://lists.ffii.org/mailman/listinfo/cjk
Re: [Cjk] Re: how can I use your Unicode.sfd?
From: Jeff Zhang <[EMAIL PROTECTED]>
> Danai SAE-HAN (韓達耐) wrote:
> > From: Jeff Zhang <[EMAIL PROTECTED]>
> >
> >> where is dvipdfmx?
> >
> > Well, it's certainly not in my pants, last time I checked... <_<
> >
> > dvipdfmx is an update to dvipdfm, and grew from dvipdfm-cjk
> > (dvipdfm-jpn and dvipdfm-kor).
> > On Debian, you have a package called "dvipdfmx".
> >
> > Try Google or Alibaba and see if you can find anything for your
> > distribution (which one should always do in the first place).
> >
> > The latest upstream version can be downloaded at
> > http://project.ktug.or.kr/dvipdfmx/snapshot/ .
>
> thanks so much for the comprehensive explanation! :)
> I am using cjk-current together with utf8 fonts created by subfonts.pe
> and Unicode.sfd.
>
> Just using `\usepackage{CJK}` in the example will give the error message:
>
> ...
> loading : Context Support Macros / PDF (2004.03.26)
> ) (/usr/share/texmf/tex/latex/CJK/UTF8/UTF8.bdg)
> ! Missing control sequence inserted.
>
> \inaccessible
> l.22 \begin{CJK}{UTF8}{nsung}
>
> ?
>
> However, `\usepackage{CJKutf8}` works well. Should I chose
> `\usepackage{CJKutf8}` to work with unicode fonts not `\usepackage{CJK}`?
>
> P.S.
>
> \documentclass[12pt]{article}
> \usepackage{CJKutf8}
> %%%\usepackage{CJK}
>
> \begin{document}
> \begin{CJK}{UTF8}{nsung}
>
> \centering
> 《宿桐庐江寄广陵旧游》\\
> 作者：孟浩然 \\
> 山暝听猿愁，沧江急夜流\\
> 风鸣两岸叶，月照一孤舟\\
> 建德非吾土，维扬忆旧游\\
> 还将两行泪，遥寄海西头
>
> \end{CJK}
> \end{document}
Indeed you should use CJKutf8 instead of CJK. The reason is the
following.
Once upon a time, each script had its own encodings. This made it
easy to use different fonts: each encoding would use its own font.
With UTF-8 this is slightly more problematic. And let's face it: the
Latin script of most Asian fonts just sucks (just look at Arphic). I
didn't expect anything else, anyway.
If there is no "fontenc" option, all glyphs would use the CJK font.
To get nice Latin or Cyrillic glyphs, add [T1] or another font
encoding as an option, and CJK will first use fonts from the selected
font encoding.
\usepackage{CJKutf8} without any options will always use the CJK
font.
See doc/CJKutf8.txt for more information.
Best regards
Danai SAE-HAN
韓達耐
--
題目:《沈園》
作者:陸游（１１２５－１２１０）
城上斜陽畫角哀，沈園非復舊池台，
傷心橋下春波綠，曾是驚鴻照影來。
夢斷香消四十年，沈園柳老不吹綿。
此身行作稽山土，猶吊遺蹤一泫然。
___
Cjk maillist - Cjk@ffii.org
http://lists.ffii.org/mailman/listinfo/cjk
Re: [Cjk] Re: how can I use your Unicode.sfd?
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Danai SAE-HAN (韓達耐) wrote:
> From: Jeff Zhang <[EMAIL PROTECTED]>
>
>> where is dvipdfmx?
>
> Well, it's certainly not in my pants, last time I checked... <_<
>
> dvipdfmx is an update to dvipdfm, and grew from dvipdfm-cjk
> (dvipdfm-jpn and dvipdfm-kor).
> On Debian, you have a package called "dvipdfmx".
>
> Try Google or Alibaba and see if you can find anything for your
> distribution (which one should always do in the first place).
>
> The latest upstream version can be downloaded at
> http://project.ktug.or.kr/dvipdfmx/snapshot/ .
>
>
> Kind regards
>
>
>
> Danai SAE-HAN
> 韓達耐
>
thanks so much for the comprehensive explanation! :)
I am using cjk-current together with utf8 fonts created by subfonts.pe
and Unicode.sfd.
Just using `\usepackage{CJK}` in the example will give the error message:
...
loading : Context Support Macros / PDF (2004.03.26)
) (/usr/share/texmf/tex/latex/CJK/UTF8/UTF8.bdg)
! Missing control sequence inserted.
\inaccessible
l.22 \begin{CJK}{UTF8}{nsung}
?
However, `\usepackage{CJKutf8}` works well. Should I chose
`\usepackage{CJKutf8}` to work with unicode fonts not `\usepackage{CJK}`?
P.S.
\documentclass[12pt]{article}
\usepackage{CJKutf8}
%%%\usepackage{CJK}
\begin{document}
\begin{CJK}{UTF8}{nsung}
\centering
《宿桐庐江寄广陵旧游》\\
作者：孟浩然 \\
山暝听猿愁，沧江急夜流\\
风鸣两岸叶，月照一孤舟\\
建德非吾土，维扬忆旧游\\
还将两行泪，遥寄海西头
\end{CJK}
\end{document}
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.5 (GNU/Linux)
iD8DBQFE+ovaexs2apGZoVgRAobtAKCaW4KBCvfhVLX8VR5ikqSRzqx1XACgzSDP
tZh3oRYxc7MvWz07JhgJiCE=
=+go9
-END PGP SIGNATURE-
___
Cjk maillist - Cjk@ffii.org
http://lists.ffii.org/mailman/listinfo/cjk
Re: [Cjk] Re: how can I use your Unicode.sfd?
From: Jeff Zhang <[EMAIL PROTECTED]> > where is dvipdfmx? Well, it's certainly not in my pants, last time I checked... <_< dvipdfmx is an update to dvipdfm, and grew from dvipdfm-cjk (dvipdfm-jpn and dvipdfm-kor). On Debian, you have a package called "dvipdfmx". Try Google or Alibaba and see if you can find anything for your distribution (which one should always do in the first place). The latest upstream version can be downloaded at http://project.ktug.or.kr/dvipdfmx/snapshot/ . Kind regards Danai SAE-HAN 韓達耐 -- 題目:《晚泊岳陽》 作者:歐陽修（１００７－１０７２） 臥聞岳陽城里鐘，系舟岳陽城下樹。 正見空江明月來，云水蒼茫失江路。 夜深江月弄清輝，水上人歌月下歸； 一闋聲長聽不盡，輕舟短楫去如飛。 ___ Cjk maillist - Cjk@ffii.org http://lists.ffii.org/mailman/listinfo/cjk
Re: [Cjk] Re: how can I use your Unicode.sfd?
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Danai SAE-HAN (韓達耐) wrote:
> (Please also CC to cjk@ffii.org when replying.)
>
> From: zhaohs <[EMAIL PROTECTED]>
>
>> Dear Danai SAE-HAN,
>> Thank you for your test,can you tell me the details，how can I use the
>> subfonts.pe to generate the high quality characters wich can be used by
>> dvipdfmx,I have been puzzling about this parameters in using fontforge.
>
> I don't know how to name the font, so I'll just call it "sunext".
> I'm not going to explain how to get vertical fonts or how to use
> uni2sfd, because I haven't tried myself yet. Check out the CJK
> documentation.
>
> Put Sun-Ext{A,B}.ttf in /usr/local/share/fonts/truetype/sunext/ and
> put symlinks in /usr/local/share/texmf/fonts/truetype/sunext/ .
>
> $ mkdir -p /usr/src/sunext{a,b}
>
> $ cd /usr/src/sunexta
> $ fontforge -script /usr/src/cjk-current/utils/subfonts/subfonts.pe \
> /usr/local/share/fonts/truetype/sunext/Sun-ExtA.ttf sunext \
> /usr/local/share/texmf/fonts/sfd/Unicode.sfd
>
> $ cd ../sunextb
> $ fontforge -script /usr/src/cjk-current/utils/subfonts/subfonts.pe \
> /usr/local/share/fonts/truetype/sunext/Sun-ExtB.ttf sunext \
> /usr/local/share/texmf/fonts/sfd/Unicode.sfd
> $ rm -f sunext00.*
>
> The Fontforge process will consume hundreds of MB of memory and take
> hours to compile.
>
> As root, do the following:
>
> $ mkdir -p /usr/local/share/texmf/fonts/{afm,tfm,type1}/sunext
> $ mkdir -p /usr/local/share/texmf/fonts/map/dvips/sunext/
> $ cp sunext?/*.afm /usr/local/share/texmf/fonts/afm/sunext/
> $ cp sunext?/*.tfm /usr/local/share/texmf/fonts/tfm/sunext/
> $ cp sunext?/*.pfb /usr/local/share/texmf/fonts/type1/sunext/
>
> $ cd /usr/local/share/texmf/fonts/type1/sunext/
> $ for filename in *.pfb;
> do echo "$(basename $filename .pfb) $(basename $filename .pfb)
> <$filename" \
> >> ../../map/dvips/sunext/sunext.map;
> done
>
> $ texhash
>
>
> Now you have to update "updmap" to get the Type1 fonts working
> properly. This depends heavily on your distribution.
>
> On Debian you would have to create a file called
> /etc/texmf/updmap.d/10sunext.cfg with the following lines:
>
> ##
> # 10sunext.cfg
> Map sunexta.map
> ##
>
> and then as root run "update-updmap" and "updmap-sys".
>
>
> In the nobmp2.tex file, change a few lines:
>
> [...]
>
> \usepackage{array}
> \usepackage[T1]{CJKutf8}
>
> [...]
>
> \begin{CJK}{UTF8}{sunext}
>
> [...]
>
>
>
> That's it!
>
>
>
> Danai SAE-HAN
> 韓達耐
>
where is dvipdfmx?
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.5 (GNU/Linux)
iD8DBQFE+h+6exs2apGZoVgRApLMAKDA1H2zkyOggB4GMlCwQwDGjP5tjQCg+OHa
HFFzifZbCP4x3iV0LW4HrKM=
=zYBm
-END PGP SIGNATURE-
___
Cjk maillist - Cjk@ffii.org
http://lists.ffii.org/mailman/listinfo/cjk
Re: [Cjk] Re: how can I use your Unicode.sfd?
(Please also CC to cjk@ffii.org when replying.)
From: zhaohs <[EMAIL PROTECTED]>
> Dear Danai SAE-HAN,
> Thank you for your test,can you tell me the details，how can I use the
> subfonts.pe to generate the high quality characters wich can be used by
> dvipdfmx,I have been puzzling about this parameters in using fontforge.
I don't know how to name the font, so I'll just call it "sunext".
I'm not going to explain how to get vertical fonts or how to use
uni2sfd, because I haven't tried myself yet. Check out the CJK
documentation.
Put Sun-Ext{A,B}.ttf in /usr/local/share/fonts/truetype/sunext/ and
put symlinks in /usr/local/share/texmf/fonts/truetype/sunext/ .
$ mkdir -p /usr/src/sunext{a,b}
$ cd /usr/src/sunexta
$ fontforge -script /usr/src/cjk-current/utils/subfonts/subfonts.pe \
/usr/local/share/fonts/truetype/sunext/Sun-ExtA.ttf sunext \
/usr/local/share/texmf/fonts/sfd/Unicode.sfd
$ cd ../sunextb
$ fontforge -script /usr/src/cjk-current/utils/subfonts/subfonts.pe \
/usr/local/share/fonts/truetype/sunext/Sun-ExtB.ttf sunext \
/usr/local/share/texmf/fonts/sfd/Unicode.sfd
$ rm -f sunext00.*
The Fontforge process will consume hundreds of MB of memory and take
hours to compile.
As root, do the following:
$ mkdir -p /usr/local/share/texmf/fonts/{afm,tfm,type1}/sunext
$ mkdir -p /usr/local/share/texmf/fonts/map/dvips/sunext/
$ cp sunext?/*.afm /usr/local/share/texmf/fonts/afm/sunext/
$ cp sunext?/*.tfm /usr/local/share/texmf/fonts/tfm/sunext/
$ cp sunext?/*.pfb /usr/local/share/texmf/fonts/type1/sunext/
$ cd /usr/local/share/texmf/fonts/type1/sunext/
$ for filename in *.pfb;
do echo "$(basename $filename .pfb) $(basename $filename .pfb) <$filename" \
>> ../../map/dvips/sunext/sunext.map;
done
$ texhash
Now you have to update "updmap" to get the Type1 fonts working
properly. This depends heavily on your distribution.
On Debian you would have to create a file called
/etc/texmf/updmap.d/10sunext.cfg with the following lines:
##
# 10sunext.cfg
Map sunexta.map
##
and then as root run "update-updmap" and "updmap-sys".
In the nobmp2.tex file, change a few lines:
[...]
\usepackage{array}
\usepackage[T1]{CJKutf8}
[...]
\begin{CJK}{UTF8}{sunext}
[...]
That's it!
Danai SAE-HAN
韓達耐
--
題目:《夏日懷友》
作者:徐璣（１１６２－１２１４）
流水階除靜，孤眠得自由。
月生林欲曉，雨過夜如秋。
遠憶荷花浦，誰吟杜若洲？
良宵恐無夢，有夢即俱游。
___
Cjk maillist - Cjk@ffii.org
http://lists.ffii.org/mailman/listinfo/cjk
Re: [Cjk] Re: how can I use your Unicode.sfd?
From: Danai SAE-HAN (韓達耐) <[EMAIL PROTECTED]> > From: zhaohs <[EMAIL PROTECTED]> > > > Now I give you some attachments,which was wrote by edward_lee,In it have > > some > > characters of "CJK Unified Ideographs Extension B",I downloaded them from: > > http://edt1023.sayya.org/tex/tmp/nobmp2.pdf > > http://edt1023.sayya.org/tex/tmp/nobmp2.tex > > You can alse in the CTEX Forum for more detailed discusses about them from > > the URL: > > http://bbs.ctex.org/forums/index.php?showtopic=31578 > > > > Furthmore,you can see the EdwardLee's blog for this thing, the URL is: > > http://blog.bs2.to/post/EdwardLee/4684 > > > > can you test it for me? > > I haven't installed some of the MS fonts, and none of my current fonts > cover the complete Unicode set, so I can't test it. > > But I'll certainly put it in my bookmarks. :) Well, being the slave driver that I am, I let my computer work all night to process Sun-ExtA.ttf and Sun-ExtB.ttf with Fontforge (only subfonts.pe; the rest, such as vertical versions, not yet). It works like a charm! I've seen characters that I thought would never exist. I should've encountered this fonts much, much earlier. "You see, teacher, that character *does* exist!" Heheh. The 00 plane created from Sun-ExtB.ttf can be deleted. You can get both fonts from http://okuc.net/software/UniFonts.exe, but it is a Win32 executable. Poo! If you want, Werner, I can upload the fonts somewhere if you wish to try these. Best regards Danai SAE-HAN 韓達耐 -- 題目:《春懷示鄰里》 作者:陳師道（１０５３－１１０１） 斷牆著雨蝸成字，老屋無僧燕作家， 剩欲出門追語笑，卻嫌歸鬢著塵沙。 風翻蛛網開三面，雷動蜂窠趁兩衙。 屢失南鄰春事約，只今容有未開花。 ___ Cjk maillist - Cjk@ffii.org http://lists.ffii.org/mailman/listinfo/cjk
[Cjk] Re: how can I use your Unicode.sfd?
From: zhaohs <[EMAIL PROTECTED]>
> Dear Werner LEMBERG,
Sorry to disappoint you, chap, my name is Danai SAE-HAN. ;D
Please CC to cjk@ffii.org next time you send an email, and please
don't "top-post": put your comments always *below* the original text.
It makes it easier for us to read. =)
> Thank you for your reply,I have done all the things according to your
> advice,but I can't do it sucessfully yet!Following are the error messages:
>
> ***
> ***
> [EMAIL PROTECTED]:~/Desktop/test$ latex UTF8.tex
> This is pdfeTeX, Version 3.141592-1.21a-2.2 (Web2C 7.5.4)
> entering extended mode
> (./UTF8.tex
> LaTeX2e <2003/12/01>
> Babel and hyphenation patterns for loaded.
> (/usr/share/texmf-tetex/tex/latex/base/article.cls
> Document Class: article 2004/02/16 v1.4f Standard LaTeX document class
> (/usr/share/texmf-tetex/tex/latex/base/size12.clo))
> (/usr/share/texmf-tetex/tex/latex/CJK/CJK.sty
> (/usr/share/texmf-tetex/tex/latex/CJK/mule/MULEenc.sty)
> (/usr/share/texmf-tetex/tex/latex/CJK/CJK.enc))
> (/usr/share/texmf/tex/latex/ucs/ucs.sty
> (/usr/share/texmf/tex/latex/ucs/data/uni-global.def))
> (/usr/share/texmf-tetex/tex/latex/base/inputenc.sty
> (/usr/share/texmf/tex/latex/ucs/utf8x.def))
> (/usr/share/texmf-tetex/tex/latex/base/fontenc.sty
> (/usr/share/texmf-tetex/tex/latex/base/t1enc.def)) (./UTF8.aux)
> (/usr/share/texmf/tex/latex/ucs/ucsencs.def)
> (/usr/share/texmf-tetex/tex/latex/CJK/UTF8/UTF8.bdg)
> (/usr/share/texmf-tetex/tex/latex/CJK/UTF8/UTF8.enc)
> (/usr/share/texmf-tetex/tex/latex/CJK/UTF8/UTF8.chr)
> (/home/zhaohs/texmf/tex/latex/CJK/UTF8/c70song.fd) [1] (./UTF8.aux) )
> Output written on UTF8.dvi (1 page, 472 bytes).
> Transcript written on UTF8.log.
> [EMAIL PROTECTED]:~/Desktop/test$ dvipdfmx UTF8.dvi UTF8.dvi -> UTF8.pdf
> [1
> ** ERROR ** Failed to read UCS2/UCS4 TrueType cmap...
>
> Output file removed.
>
>
Apparently it can't find the correct CMAP.
You might have to install the packages:
cmap-adobe-{cns1,gb1,japan1,japan2,korea1}.
Or install dvipdfmx.
> In addition，I'm not use the CJK 4.5.1 shipped with the Ubuntu 6.06,but
> use the latest cjk which I downloaded from
> "http://cjk.ffii.org/cjk-current.tar.gz";.
Excellent. Then that means that you can also use Fontforge to create
fonts that are better than those created with ttf2tfm. But I'll show
you later on how to use Fontforge.
> Now I give you some attachments,which was wrote by edward_lee,In it have
> some
> characters of "CJK Unified Ideographs Extension B",I downloaded them from:
> http://edt1023.sayya.org/tex/tmp/nobmp2.pdf
> http://edt1023.sayya.org/tex/tmp/nobmp2.tex
> You can alse in the CTEX Forum for more detailed discusses about them from
> the URL:
> http://bbs.ctex.org/forums/index.php?showtopic=31578
>
> Furthmore,you can see the EdwardLee's blog for this thing, the URL is:
> http://blog.bs2.to/post/EdwardLee/4684
>
> can you test it for me?
I haven't installed some of the MS fonts, and none of my current fonts
cover the complete Unicode set, so I can't test it.
But I'll certainly put it in my bookmarks. :)
Friendly regards
Danai SAE-HAN
韓達耐
--
題目:《讀嚴子陵傳》
作者:楊萬里（１１２７－１２０６）
客星何補漢中興，空有清風冷似冰。
早遣阿瞞移漢鼎，人間何處有嚴陵！
___
Cjk maillist - Cjk@ffii.org
http://lists.ffii.org/mailman/listinfo/cjk
