Re: [XeTeX] Diameter symbol

[Will Robertson]

On 2010-03-16 00:31:10 +1030, David Cottenden 
 said:



Or are you looking for EMPTY SET, ∅, at U+2205? DIAMETER SIGN, ⌀, is
at U+2300.

The latter is the one that I am after, but I know that my main body font
doesn't include it. Is there a decent font which does? I'm looking to
match with Chaparral Pro, if that's possible.



If you're using XeTeX (and why wouldn't you be asking on this list -- ! 
-- sorry I didn't twig this answer straight away), you can just use a 
text glyph directly:


\documentclass{article}
\usepackage{fontspec,siunitx}
\begin{document}
\newfontface\symbolsfont{TeX Gyre Pagella}
\newcommand\diametersymbol{\mbox{\symbolsfont \char"2300}}
% (Or just type the symbol in there instead of the \char nonsense.)
\newcommand\diam{\diametersymbol\,}
The cylinder has dimensions \diam\SI{2x3}{m}.\par
(Not sure how that spacing should look, exactly; hand-tune it for the font.)
\end{document}

Hope this helps,
Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] Problems with fontspec and mathspec

[Will Robertson]

On 2010-05-13 00:04:29 +0930, Tobias Schoel 
 said:



Removing the option cm-default from mathspec (which passes it to
fontspec) results in no error-messages and the desired effect.

Why is this?



cm-default was originally intended for users of fontspec who didn't 
have the Latin Modern fonts installed; nowadays it serves less purpose.


It's not entirely clear to me what is triggering that error, however. 
(Without having looked into it in much detail.) Is this going to cause 
a problem or is the current behaviour workable?


W




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] unicode-math testing

[Will Robertson]

On 2010-05-27 17:43:15 +0930, David Cottenden 
 said:



Mildly off topic, but does that mean that the siunitx incompatibility is
solved?


Unfortunately not :(
But I'll take a look (and try and develop a workaround for Joseph if 
possible) before I send it to CTAN. It is quite a significant problem.


You don't happen to be using Windows, are you? (Sorry, I can't recall.) 
Can you reproduce the output of that example?


Will





--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

[XeTeX] fontspec v2.0 advanced notice

[Will Robertson]

Dear all,

I have just -- finally -- submitted a new version of fontspec to CTAN.
It should still work with TeX Live 2009 versions of XeTeX, but not much 
testing has been done.


More information here: 
http://latex-alive.tumblr.com/post/637156378/fontspec-in-tex-live-2010


Feedback very much appreciated. While I haven't fixed all the bugs nor 
implemented all the features outstanding, this would be a good time to 
hit me up for any changes you've been expecting or waiting for.


Many thanks,
Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] unicod-math testing / siunitx

[Will Robertson]

On 2010-05-28 08:47:33 +0930, Ross Moore 
 said:




On 28/05/2010, at 7:45 AM, Joseph Wright wrote:


On 27/05/2010 22:32, David Cottenden wrote:


Note that this used the sinuitx from last November, so it might be
out
of date. I hope this is enough to go on - I don't remember much
that far
back, I'm afraid, but if there's anything else I /might/ be able to
dredge the memory!

David


This looks okay to me using siunitx v2 and unicode-math/fontspec
from GitHub.


And it's fine with packages:  siunitx,  expl3,  xpackages



Wow, that's great.
I didn't think the conflict between the old version and unicode-math 
was able to be resolved. So whatever siunitx is now doing (I'll dig 
into it again at some stage) is a great improvement -- for me, at least.


Thanks!

Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] unicod-math testing / siunitx

[Will Robertson]

On 2010-05-28 14:42:24 +0930, Ross Moore 
 said:



I don't have an old  siunitx.sty  to try to reproduce the old bug.
With just the .log file it is impossible to tell what is going on,
since  siunitx.sty  uses \AtBeginDocument to delay a lot of stuff,
and it is amongst those things where the error occurs.



This gist of the problem was that siunitx was using things like

   \setbox...@tempboxa=\hbox{%
 $\mathsf{\global\chardef...@fam@sf=\fam}$}%

and then

 \ifnum\the\fam=...@fam@sf

to detect when maths alphabets were available.

Since unicode-math doesn't use a different fam for each alphabet those 
tests were breaking.


(The current implementation of unicode-math very inefficiently uses 
mathcode remapping to effect the maths font switching. But at least it 
works and is flexible for now. Future plans are to merge breqn's 
"everything active" approach to improve efficiency and compatibility 
with that package.)


I haven't actually checked yet, however, whether siunitx v2 gives 
correct output while inside a unicode-math \mathsf. I'd be surprised if 
it did.


Cheers,
Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] (Off-topic) STIX Fonts

[Will Robertson]

On 2010-05-30 23:20:07 +0930, Michael Lynch 
 said:



Does anyone know any more about the project? Has the math feature
definitely been delayed until a future version?


Yep, the intial release has always been about glyph coverage and 
kerning. As they say, OpenType features will come next.


The fonts will still be very useful for accessing maths glyphs. I'm 
about to release a package "unicode-math" that will allow access to 
these fonts. (Although there will be some limitations due to the lack 
of OpenType features.)


Khaled Hosny has put together an OpenType version of STIX here:
 http://github.com/khaledhosny/xits-math

I'm going to test it out soon.

Cheers,
Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] (Off-topic) STIX Fonts

[Will Robertson]

On 2010-05-31 01:31:40 +0930, Michael Lynch 
 said:



If you were to use the STIX fonts in the meantime, given the lack of
OpenType features in the STIX release, does it make more sense to use
mathspec instead of unicode-math? I was under the impression that
unicode-math was exclusively designed for working with “proper” 
OpenType math fonts.


mathspec does a great job mapping text alphabet fonts into math 
alphabet fonts, but it wasn't designed to be a general solution for 
assigning control sequences or behaviour for the full repertoire of 
unicode math symbols (i.e., the non-alphabetic ones).


I haven't really done enough testing to see how far you get with the 
STIX fonts without OpenType features, but you can at the very least 
*access* all the glyphs. But things like delimiter scaling and big 
operator sizes and square roots won't work; they require a "real" 
OpenType maths font. I probably won't spend too much time trying to 
fake them, however; in the meantime we can use XITS-Math and the other 
real OpenType maths fonts for the parts we need them for.


W




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

[XeTeX] catcodes of superscripts, etc.

[Will Robertson]

(Sorry for the cross-posting, but this seems of very minor possible interest to 
a few different people.)

Hi Jonathan,

I've just noticed that catcodes of superscript i and n have catcode 11, as do 
the latin & greek subscripts. (Code points appended.)

I'm going to change this in unicode-math (so they're all cc12), but do you 
think this is something that should be changed in the ini file?

(I'll check this with LuaTeX, too.) 

The reason it's a problem is that users try to write things like $\Cⁿ$ and the 
expected behaviour, well, isn't.

Many thanks,
-- Will


2071 {i}
207f {n}

2090 {a}
2091 {e}
1d62 {i}
2092 {o}
1d63 {r}
1d64 {u}
1d65 {v}
2093 {x}
1d66 {\beta}
1d67 {\gamma}
1d68 {\rho}
1d69 {\phi}
1d6a {\chi}


--
Subscriptions, Archive, and List information, etc.:
  http://tug.org/mailman/listinfo/xetex

[XeTeX] Fwd: New on CTAN: unicode-math

[Will Robertson]

Hello,

For those who have been interested in my unicode-math package, the first 
released version is now available on CTAN and will be included in TeX Live 
2010. (See below.) It requires a very recent version of fontspec, which itself 
requires a few other updates to other packages; if you're not keen on updating 
packages manually (I'm not) then feel free to wait until it's included in your 
distribution.

This package provides similar functionality to Andrew Moschou's mathspec 
package, but goes further in providing access to symbols in the unicode maths 
range. You'll need a unicode maths font to access them, of course :) Asana Math 
is the best free option, otherwise Microsoft's Cambria Math is currently the 
gold standard.

Feedback most welcome!

Best regards,
Will

From: CTAN Announcements 
Subject: New on CTAN: unicode-math

the daemon perks up with the appearance of the following (which he had
previously imagined to be a myth concocted by people on the net):

The following information was provided by the package's contributor.

Name of contribution: unicode-math
Author's name: Will Robertson
Location on CTAN: macros/latex/contrib/unicode-math
Summary description: Unicode math support for XeTeX and LuaTeX
License type: lppl

Announcement text given by the package's contributor:
--
Well, I guess it's now or never. With an archive of emails about the
thing going back to 2006, I think it's finally time to release this
package. (Sorry about the hold-up, really.)

This is the initial release of the unicode-math package for inclusion
in TeX Live 2010. It is designed to typeset mathematics with unicode
input and using OpenType math fonts. (There is very very little
compatibility with older maths packages.) XeTeX support is fully
tested but LuaTeX not quite so much. (At least the example file
runs...)

You can typeset with the STIX fonts with the present version of the
package, but there is no support yet for its extra alphabets in the
Unicode `private use area'. I hope to add support for them soon.

Thanks to all the people who've provided valuable feedback over the
years, and especially to Barbara Beeton for laying the foundations
with her work with the STIX fonts.

Good luck; I hope it works. Let me know if it doesn't. Finally, this
package is by no means complete (it's called the "initial release" for
a reason). Your feedback and suggestions now are greatly
appreciated. Anyone coming along to the TUG 2010 conference will have
the opportunity to hear me speak about it in a couple of weeks. (But
will I be able to muster up a full 30 minutes? Actually, yeah, I
probably will.)

Time for a beer.

have a good one!

i have installed the package and created an entry in the catalogue
repository.

Users may view the package catalogue entry at
http://www.tex.ac.uk/tex-archive/help/Catalogue/entries/unicode-math.html
or they may browse the package directory at
http://www.tex.ac.uk/tex-archive/macros/latex/contrib/unicode-math

though, of course, the catalogue entry won't appear on the web until
some time tonight.

Robin Fairbairns

For the CTAN team


-- Will




--
Subscriptions, Archive, and List information, etc.:
  http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] Typesetting XeLaTeX and LaTeX logo under MacTeX 2009

[Will Robertson]

On 2010-06-03 23:45:37 +0930, Iwan Setyawan 
 said:

> (/usr/local/texlive/2009/texmf-dist/tex/xelatex/xltxtra/xltxtra.sty
> Package: xltxtra 2009/09/02 v0.5 Improvements for the "XeLaTeX" format
> 
> (/usr/local/texlive/2009/texmf-dist/tex/latex/metalogo/metalogo.sty
> Package: metalogo 2009/09/04 v0.11 Extended TeX logo macros
> 

Version 0.5 of xltxtra had incorrect parameters for metalogo; this was fixed in 
version 0.5a. (And then a bug was fixed in 0.5b.)

The current release on CTAN requires a new version of fontspec, so you probably 
don't want update everything from there. I've attached the .sty file for v0.5b; 
I hope this fixes the problem for you.

Best wishes,
-- Will



xltxtra.sty
Description: Binary data


--
Subscriptions, Archive, and List information, etc.:
  http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] XeTeX and luaTeX

[Will Robertson]

On 2010-06-04 22:36:14 +0930, Arthur Reutenauer 
 said:



Maybe Will wants to say a word about fontspec for LuaTeX at this point :-)


Nothing more right now than to say: When you get your hands on TeX Live 
2010, give it try :)


W




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] XITS Math font - first beata

[Will Robertson]

On 2010-06-05 11:40:28 +0930, 
wodzi...@math.berkeley.edu said:



On the other hand, with

\setmathfont{XITS Math}
\setmainfont[Numbers=OldStyle,Mapping=tex-text]{STIXGeneral}

I get improved STIX in Math, and the regular STIX in text with *no* Old
Style figures, plus unnecesssary duplication of fonts (as XITS Math is
just improved STIXGeneral upright Roman), but with 'fontspec' recognizing
that for italic and bold it has to look up the corresponding 'Italic' and
'Bold' font files.



I have an alternative suggestion :)
Instead of trying to use STIX Math as a text font, let's leave it doing 
what it does best and instead use a matching Times-clone for the text. 
This is what I would recommend:


\setmainfont[
   Extension=.otf,
   UprightFont=*-regular,
   ItalicFont=*-italic,
   BoldFont=*-bold,
   BoldItalicFont=*-bolditalic,
   Numbers=OldStyle,
   Mapping=tex-text
 ]{texgyretermes}
\setmathfont{XITS Math}

(If you install TeX Gyre Terms in your system then you don't need the 
long-winded \setmainfont options.)


When you think about it, it doesn't exactly make sense to try and use 
STIX as a text font -- the bold and italic letters and kerning will 
have been designed for maths use, not text. And while they *can* 
function acceptably in text, TeX Gyre Termes is specifically a text 
font and has excellent glyph coverage.


Hope this helps,
Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] Solved -- Was: Re: Typesetting XeLaTeX and LaTeX logo under MacTeX 2009

[Will Robertson]

On 2010-06-05 00:17:16 +0930, Iwan Setyawan 
 said:



Thanks for the solution! The new file fixed the problem. Everything
seems to be working correctly now.


FWIW, I (attempt to) always tag releases on Github when they're sent to 
CTAN, so you can always get historical versions of packages from, e.g., 
github.com/wspr/xltxra (or whichever package of mine).


Cheers,
Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] XITS Math font - first beata

[Will Robertson]

On 2010-06-05 14:58:22 +0930, 
wodzi...@math.berkeley.edu said:



I read your answer to my post as an admission that STIXGeneral, which
covers actually all the usual 'text' blocks of the Unicode is not really
suitable for text? Is it really so? My naive perhaps perception was that
STIXGeneral was not just a 'Math' font but a complete font that was
expected to be used for both text and Math.


No, sorry, I misspoke. I wouldn't say that it's *not* suitable (in 
fact, it has a very good glyph coverage in certain areas), just that 
using separate -- but similar -- fonts for text and math might end up 
being even better than just one.


In fact, I take it all back. STIX is fine for text, and so is TeX Gyre 
Termes. I guess I just meant to say that there are options available. 
But XTIS math, unless I am mistaken, should purely be used for maths.



Why then bother about designing 'OldStyle' figures in XITS Math? My
understanding is that one uses them exclusively in the text -- not in the
Math mode. Or there is a provision somewhere in 'unicode-math' for them to
be used also in Math? (if there is one, I haven't found it)


If you want them in math, it should be possible to write

\setmathfont[range=\mathup/{num},Numbers=OldStyle]{XITS Math}

after the regular math font declaration, but I haven't tried it. I hope 
it works :)


If you ask me, I can add this as an option somewhere in the package. 
(I've seen old-style figures in maths in examples before, but never "in 
the wild".)


Thanks,
Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] XeTeX and luaTeX

[Will Robertson]

On 2010-06-05 18:42:06 +0930, Pablo Rodríguez 
 said:



Which will be those differences between podTeX and LuaTeX?


I can't remember where it's talked about, but I think the relevant 
details are that there are minor incompatibilities in kerning and 
hyphenation and justification when typesetting paragraphs. Basically, 
on any arbitrarily large pdfTeX document, you can't guarantee you'll 
get the same output when typeset with LuaTeX. (As a change in 
hyphenation can change a single line break, which can change a page 
break, which can change a chapter break...)


Of course, there's also some differences in extended funtionality, such 
as the pdfTeX primitives that are only available with Lua in LuaTeX, 
but Heiko Oberdiek's pdftexcmds provides a unified interface to this 
(and to the relevant XeTeX primitives as well).


Certainly nothing to worry about (especially because hyphenation 
patterns aren't as constant in the TeX world as TeX itelf), but 
necessary to be aware of.


Cheers,
Will 





--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] XITS Math font - first beata

[Will Robertson]

On 2010-06-06 01:28:46 +0930, Eelis van der Weegen 
 said:



On 2010-06-04 20:31, Khaled Hosny wrote:

As promised, the OpenType MATH enriched version of STIX fonts, XITS,
is now available.


XITS looks very neat!

Are the STIX people planning to release /official/ OpenType MATH
enriched versions at some point as well (perhaps as part of STIX version
1.1 or 1.2)?


v1.1 will have OpenType MATH support.
v1.2 will denote (La)TeX support.


 If so, to what extent would this make XITS obsolete?


Depends how good a job they do :)

W




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] XITS Math font - first beata

[Will Robertson]

On 2010-06-06 02:54:13 +0930, Taco Hoekwater 
 said:



Will Robertson wrote:
On 2010-06-06 01:28:46 +0930, Eelis van der Weegen 
 said:



If so, to what extent would this make XITS obsolete?


Depends how good a job they do :)


And whether the current STIX release schedule is reliable. Judging
from past results, it is altogether possible that v1.1 won't be
ready for release for half a decade yet.



I'm a little sensitive to calling people out on missed deadlines... 
anyway, if this ends up being true then they may as well commission 
Khaled to do the job for them :)


W




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] XITS Math font - first beata

[Will Robertson]

On 2010-06-07 01:41:41 +0930, "George N. White III" 
 said:



I replaced the accent macros with unicode glyphs, e.g. Möbius,  Poincar
é, etc.
One AMS text symbol (\S --> §) generated and undefined.


Have you loaded xunicode?


cmsy is still being used somehow, somewhere:


Needs to be fixed. But I wouldn't be surprised for a few edge cases still.


1) existing documents may need editing to replace macros with unicode glyphs


All math symbols should still be defined as macros, although I might 
have missed a few. The idea is to make it easy to use the new fonts, 
hopefully, without (m)any changes to the source document.



2) recommended xetex, etc. versions, e.g,

 XeTeX, Version 3.1415926-2.2-0.9997.1 (TeX Live 2010)


TeXLive 2010 for unicode-math and fontspec v2 and LuaLaTeX (if you want 
it) is a good metric. The actual version of XeTeX isn't too important 
as it's had maths support for years. (Sorry it's taken me so long.)


W




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

[XeTeX] fontspec vs polyglossia

[Will Robertson]

Dear all,

The recently released version 2 of fontspec now clashes with polyglossia, 
because the internals of fontspec have changed.

polyglossia was previously hooking into low-level macros to perform what is now 
provided by a fontspec programming interface. (E.g., \fontspec_if_script:nTF 
{latn}{true}{false} .)

I've emailed François about the changes required but I seem to remember him 
saying that he no longer has time to maintain polyglossia. Is there anyone 
willing to write a patch for polyglossia so that it will support the new 
version of fontspec?

Best regards,
-- Will




--
Subscriptions, Archive, and List information, etc.:
  http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] wish on \showhyphens from xltxtra

[Will Robertson]

On 2010-06-14 05:24:47 +0930, Pablo Rodríguez 
 said:



trying to check whether a rather large passage in ancient Greek was
properly hyphenated, I realized that it would be extremely helpful if
xlxtra could provide also the spaces in \showhpyphens.


I agree xltxtra's \showhyphens could be improved.
See attached proposed modification. How is this? Too verbose?

W

\documentclass{article}
\usepackage{xltxtra}
\makeatletter
\def\showhyphens#1{%
 \typeout{^^J***
  \string\showhyphens:
  ***}%
 \...@loopspace \...@showhyphens #1 \...@nil
 \typeout{^^J***%
  ***%
  ***^^J}}
\def\...@loopspace#1#2 #...@nil{%
 #1{#2}%
 \d...@tempa{#3}%
 \i...@tempa\@empty\else
   \...@loopspace#1#3\@nil
 \fi
}
\makeatother
\begin{document}
\showhyphens{discombobulated hyphenation routine}
\end{document}





--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] internationalization bug in the selection of smart font features

[Will Robertson]

On 2010-06-15 02:21:49 +0930, 
j_mach_w...@shared-files.de said:



In XeTeX, you can select a smart feature by using the package =20
fontspec, for instance like this:

\font\hoeflerengraved=3D"Hoefler Text/AAT:Style Options=3DEngraved Text"


[snip]


I don't know whether this is a bug of XeTeX or of fontspec.


These are plain XeTeX font definitions. The fontspec package defines 
commands such as \fontspec and \setmainfont. Sounds like the problem is 
in XeTeX itself, then.


Cheers,
Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] wish on \showhyphens from xltxtra

[Will Robertson]

On 2010-06-15 05:41:18 +0930, Pablo Rodríguez 
 said:


I'd prefer the standard way with babel and (pdf)latex. But this is not
possible, this is better than the present solution.


The original code is Jonathan's; I've done nothing more to it than add 
a wrapper to handle multiple words. One day I might look into it in 
more detail, but use my previous code for now. I'll update xltxtra to 
use that instead, sooner or later.


Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] Problem: Beamer+xelatex+\XeTeXdefaultencoding

[Will Robertson]

On 2010-06-22 05:32:57 -0700, Jonathan Kew 
 said:


It turns out the problem can also be triggered by setting 
\XeTeXinputnormalization=1 (or 2) in a similar way.


This post is to remind me to add this primitive to xetexref. I don't 
think it's in there. (I'm in SF right now, so I'm not able to do it 
yet.) Is it documented anywhere else?


W




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] Problem with unicode-math and acronym packages

[Will Robertson]

On 2010-06-15 16:23:33 -0700, Michael Lynch 
 said:



I've come across a conflict between the unicode-math and acronym packages.


Hi Michael,

Thanks for reporting this. unicode-math gets its definitions from the 
STIX fonts, and the complete list (some 2400 symbols) has been 
partially invented to cope with the new glyphs that have never had TeX 
names. Some of the new names are not always perfect and we need 
feedback like this to ensure that good names have been chosen.


In this case, \ac is defined as U+223E; we'll change the name to 
something more proper like \invlazys.


For now just comment out or change the line in unicode-math-table.tex 
that deals with this and it'll be fixed in the next update.


Thanks,
Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] XeTeX equivalent to \pdfpagewidth ?

[Will Robertson]

On 2010-07-02 07:22:18 -0700, Ulrike Fischer 
 said:



But if you set \pdfpagewidth you should also set \pdfpageheight. At
least on my system (miktex 2.7) xetex will otherwise \pdfpagewidth
setting.

\documentclass{article}
\pdfpagewidth=20cm
%\pdfpageheight=15cm %needed too.


IMO this should be reported as a bug.
But what do I know... does this the same thing happen in pdftex?

(Sorry, recovering from jetlag so can't test for myself right now.)

W




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

[XeTeX] Detecting glyph change after feature selection

[Will Robertson]

Hi,

I was hoping that something like this would let me detect whether a certain 
font feature affects the glyph obtained by inputting a certain character:

\documentclass{article}
\begin{document}
\font\1="Adobe Jenson Pro"\1 1234567890abcdefg

\number\XeTeXcharglyph`\a

\number\XeTeXcharglyph`\c

\font\1="Adobe Jenson Pro:+sups"\1 1234567890abcdefg

\number\XeTeXcharglyph`\a

\number\XeTeXcharglyph`\c

\end{document}

I'm not sure if that even makes sense, but if it does, is there a way to 
retrieve this information?

E.g., in the example above, Jenson Pro doesn't contain the "c" superscript, and 
I'd like to be able to test for that. But \XeTeXcharglyph only returns the 
regular glyph slot before the font feature is selected.

-- Will




--
Subscriptions, Archive, and List information, etc.:
  http://tug.org/mailman/listinfo/xetex

[XeTeX] [OT] Free fonts for fontspec examples?

[Will Robertson]

Dear all,

I'm trying to improve the examples in the fontspec manual. As well as improving 
the examples already there, I'd like to eliminate the use of any non-free fonts 
so that others can compile the manual without trouble.

I've got a few areas I'm stuck on, since my font repertoire isn't very large.

1.  Does anyone know of any free fonts that have optical size variants?

2.  I'm looking for examples which use (and don't need too many explanations) 
the following features:

- Contextual swashes (cswh)

- Uppercase spacing/kerning (cpsp)

- Historical style (hist)

Does anyone have any good ideas here?

3.  All of my Japanese examples use the Hiragino fonts distributed with Mac OS 
X; are there any free alternatives I can use that supports features such as 
ruby, kana style, annotations, "CJK Shape", and so on?

4.  I'm aware that I'm not fully covering the OpenType feature list in 
fontspec. Has anyone noticed obvious areas that should be included? (And, if 
so, suggestions on how they should be used with examples would be very, very 
helpful.)

Many thanks,
-- Will




--
Subscriptions, Archive, and List information, etc.:
  http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] [OT] Free fonts for fontspec examples?

[Will Robertson]

On 2010-07-12 18:33:54 +0930, Alexey Kryukov 
 said:



On Mon, 12 Jul 2010 16:45:49 +0930
Will Robertson wrote:


- Historical style (hist)


You can try my Theano Old Style font
(http://www.thessalonica.org.ru/en/theano.html), which has some
historical forms implemented both for Greek and Latin.


Thanks, I'll take a look.


 But please note
that this feature is essentially deprecated: it is better for a font
designer to use stylistic sets instead.


Why? If they are "historical forms" isn't it better to give them a 
meaningful OpenType feature name?



4.  I'm aware that I'm not fully covering the OpenType feature list
in fontspec. Has anyone noticed obvious areas that should be
included?


Am I right fontspec still doesn't support stylistic sets
(ss00--ss20), except via RawFeature?


This has always been available under the "Variant=0/1/2/3/..." feature 
but this name wasn't very obvious. In more recent versions of fontspec 
you can use "StylisticSet=0/1/2...".


Please ask if it looks like I've missed adding a feature in fontspec; 
chances are I need to add code or documentation or both to the package.


Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] [OT] Free fonts for fontspec examples?

[Will Robertson]

On 2010-07-12 22:01:42 +0930, Khaled Hosny 
 said:



On Mon, Jul 12, 2010 at 04:45:49PM +0930, Will Robertson wrote:


1.  Does anyone know of any free fonts that have optical size variants?


Latin Modern.


Oh! It's like the purloined letter, hiding in plain sight.

Thanks,
Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] [OT] Free fonts for fontspec examples?

[Will Robertson]

On 2010-07-12 17:07:37 +0930, Yves Codet 
 said:



For the last feature you could use a Greek font which includes it, for
instance "Alexander", that you can download here:

http://users.teilar.gr/~g1951d/

If you typeset the word λόγου with and without 'hist' you will see
a difference.


Thanks for the suggestion; unfortunately I think these fonts only have 
historical ligatures (hlig) rather than historical alternates (hist).


Cheers,
Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] [OT] Free fonts for fontspec examples?

[Will Robertson]

On 2010-07-13 03:37:13 +0930, David Perry 
 said:



This has always been available under the "Variant=0/1/2/3/..." feature
but this name wasn't very obvious. In more recent versions of fontspec
you can use "StylisticSet=0/1/2...".

Thank you, thank you.  I hadn't picked up on this.  The naming
conventions have always confused me, since there is a Fontspec feature
"Alternate" and "alternate" is also an option within the "Style"
feature, while stylistic sets were "Variants."


Style=Alternate and Alternate=0 are synonyms (for historical reasons).

The reason the naming scheme is so chaotic is that it was first written 
to support Mac OS X AAT fonts (this was before XeTeX supported 
OpenType!) and the OpenType support slowly merged with and then 
supplanted the original purpose of the package.


I'm trying to document OpenType font features now with clear reference 
to the actual feature tag, but I'm aware some of my terminology still 
needs to be updated. Please do tell me if you find an unclear paragraph 
(or twenty) in the manual and I'll do what I can to clear things up.




If clearer terminology could be found for all this, it would be a real
improvement.


For the hypothetical fontspec v3.0, I would like to totally revisit all 
of the feature names (while supporting backwards compatibility of 
course) but it's a big job for not much reward :)


It might also be useful for advanced users to be able to type

   \fontspec[ss01,hlig,cwsh]{...}

That's certainly much easier to achieve!



In the last couple of days I have been thinking about using the
Character Variant feature (cv01-cv99) rather the stylistic alternate
feature for accessing glyphs such those I referred above.  I haven't
gotten very far with this, but it might be something to add.  (Does
anyone know of examples of this in use?)


I only just heard about these new features, too. I would also like to 
see some examples -- I assume that a feature name like


CharacterVariant=0/1/2/3/4...

would be fine for fontspec?

Thanks,
Will





--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] [OT] Free fonts for fontspec examples?

[Will Robertson]

On 2010-07-13 21:50:50 +0930, David Perry 
 said:



Sure, or cv01 (like ss01) if you implement the short tags.  Fontspec is
unusual in that it lets (actually, requires) users to request stylistic
alternates by number.  Most programs present the alternates visually in
a character palette.


Well, there's not really an alternative, right? :)


So for XeTeX users, from one point of view,
there's no advantage to having the cvXX option available.  However, I
think having cvXX might help interchange of data with other software.


Hang on, aren't the cv0x "just more" OpenType features in the same 
style as ss01 but targeting individual characters specifically? So if a 
font contains alternates accessed wth cv0x features, then we have to 
offer them for fontspec users to be able to use those glyphs.


But I agree with the general sentiment that numerical features aren't 
very user-friendly for a batch system like LaTeX.


W




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] [OT] Free fonts for fontspec examples?

[Will Robertson]

On 2010-07-14 12:38:56 +0930, Alexey Kryukov 
 said:



On Tue, 13 Jul 2010 00:00:00 +0930
Will Robertson wrote:


Why? If they are "historical forms" isn't it better to give them a
meaningful OpenType feature name?


To my mind, an attempt to provide meaningful feature names for every
possible situation was a key mistake of the OpenType standard. It is
especially clear in case of historical forms, because it is difficult
to describe the huge number of possible glyph/character variations with
just one or two predefined tags.


Okay, I can concede that "hist" is a case with fuzzy edges. Still, 
surely you're not arguing against having features like onum/lnum!


If a certain style of variation recurs consistently in multiple fonts, 
it only seems sensible to me to have a standard OpenType feature to tie 
them all together.



As for meaningful names, it's no longer a problem, as the OT
specification currently allows to assign a friendly name (it even
can be localized) to each stylistic set.


But these aren't useful for the XeTeX user. AAT font features are 
referred to by name, but inconsistencies between fonts meant it was a 
nightmare to use them in XeTeX; one of the original advantages of 
fontspec was to provide consistency here so you didn't need to remember 
multiple ways to write "Numbers=OldStyle".


Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] XeLaTex and Armenian

[Will Robertson]

On 2010-07-14 06:40:52 +0930, Apostolos Syropoulos 
 said:



In different words,
if a font includes glyphs for the Armenian script and the Latin script and the
kern and liga features are defined for both scripts, then there is no reason to
use different fonts.


But don't you still need markup (perhaps implicit using interchartoks) 
to switch hyphenation routines (or different spacings or whatever) for 
the different languages? Then it doesn't matter if you need to switch 
font features or not since other things are happening anyway.


Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] [OT] Free fonts for fontspec examples?

[Will Robertson]

On 2010-07-14 19:46:02 +0930, François Charette 
 said:



[quite off-topic]

In 19th-century Rome, a prince and mathematician named Baldassare 
Boncompagni...



P.S.  I love this list.

Thanks :)
W




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] [OT] Free fonts for fontspec examples?

[Will Robertson]

On 2010-07-14 22:24:30 +0930, "Fr. Michael Gilmary" 
 
said:



Oh, good! Then, I humbly request, for the benefit of those you love on
this list, that the next version of fontspec documentation have
hyperlinks for ToC, etc  Maybe you've already done this in more
recent versions that we have here.



The new version does use hyperlinks, but I haven't had time to make the 
output look really nice. Captions for figures/tables and for examples 
aren't consistent, and even the ToC needs adjusting because the section 
numbers are too long. (And the hyperlinking for cross-referencing is a 
bit all over the place because I'm not using refstyle when I should be.)


As always, if you see things in the manual that could be easily fixed, 
I will happy accept contributions. Things like the ToC and caption 
formatting only takes a few minutes, but it's the context switching 
that kills my time so I never get around to looking at such 
important-but-mundane details.


W




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] \ifincsname: unexpected behaviour after nested \csname...\endcsname

[Will Robertson]

On 2010-07-22 09:12:14 +0930, Grzegorz Murzynowski 
 said:


I noted quite unexpectedly that nesting \csname…\endcsname, results in 
turning |\ifincsname| false: the example above typesets


\before nested csname: In csname \ after nested csname: Not in csname

Is it a bug or a feature? (I use nested csnames sometimes).



The same behaviour occurs in pdfTeX, so from that respect at least it's 
consistent. But it definitely looks like a bug to me.


Perhaps a clarifying example:

\def\a{\ifincsname T \else F \fi}
\expandafter\string\csname
   1\a
   \expandafter\string\csname2\a\endcsname
   3\a
\endcsname

#1 and #2 and "true" but #3 is "false". Clearly the \endcsname has 
reset the "am i in a csname" switch rather than popped it back to its 
previous value.


I think it would be worth sending this to the pdfTeX mailing list as well.

Cheers,
Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] Pipes in XeTeX

[Will Robertson]

On 2010-07-27 00:07:02 +0930, Florian Gilcher 
 said:


I really like the following feature of pdftex and others for 
development purposes:


  \input{|"darcs changes -s"} % print a detailed log of changes

and include that into my document as a "running log" for other reviewers.

Sure, it requires --shell-escape and --enable-write18, but thats just 
fine in my environment.


Actually I think this is a MiKTeX-only feature of pdfTeX (or it was 
last time I checked). Is there a significant problem to doing this?:


   \immediate\write18{darcs changes -s > tmp.txt}
   \input{tmp.txt}

Wrap it up in a macro and it's no less convenient than using the pipe, right?

Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] fontspec: loading different shapes/scales of one font as different font instances?

[Will Robertson]

On 2010-07-28 23:32:14 +0930, BPJ 
 said:



As the subject line says I wonder if it's possible
with fontspec to load different shapes of the same
font as different font instances?


I don't understand the question, sorry.
fontspec provides

\newfontfamily\foo[]{}

You can use this as many times as you like with whatever different 
options you like. Is there something missing?


Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] Pipes in XeTeX

[Will Robertson]

On 2010-07-28 22:10:49 +0930, Florian Gilcher 
 said:



Actually, pdftex in both texlive 2009 and 2010 supports it.

The difference between MiKTeX and texlive is that MiKTeX has a special 
CLI flag for it (--enable-pipes) while pdftex on texlive checks for 
(--shell-escape).


Thanks for the info. It's been a couple of years since I used pipes. Is 
this documented *anywhere*? Neither the pdftex nor the texlive 
documentation seem to have anything on this topic.


Well: the problem with \write18: it is not quite the same (for example, 
i cannot write to a pipe [1]), it clutters the file system


Since write18 is enabled anyway you can always "rm ..." or "del..." 
your temporary file.



 and I cannot read from any arbitrary source.


How so? (Not questioning, just curious. I've used write18 a bit and 
haven't run into trouble yet.)


Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] fontspec: loading different shapes/scales of one font as different font instances?

[Will Robertson]

On 2010-08-02 06:39:27 +0930, BPJ 
 said:



I'll still need to stick in commands/define features to
get smallcaps, though...


\newfontface\foosc[Letters=SmallCaps]{TeX Gyre Pagella}

W




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

[XeTeX] fontspec and xunicode

[Will Robertson]

Hey,

Quick message to say that the new version of fontspec (v2.1) on GitHub now 
loads xunicode directly, and references to xunicode been removed entirely from 
euenc. If necessary, Khaled can release fontspec & euenc to CTAN while I'm away 
for the next couple of weeks, but otherwise if all is well I'll send it into 
CTAN when I get back.

fontspec loads xunicode even while running on XeLaTeX; this is slightly 
incompatible with previous behaviour but I don't think it would actually cause 
any problems. This will mean that xltxtra is well and truly deprecated (and 
\usepackage{fontspec} is all that is recommended for new XeLaTeX documents); 
I've just sent "realscripts" into CTAN to replicate its "use OpenType 
superscripts in footnotes" for those who want it in LuaLaTeX or 
XeLaTeX-without-xltxtra.

Justification for the changes above: forking xunicode for LuaLaTeX wasn't the 
best idea for maintenance issues, but then a wrapper had to be written around 
xunicode for LuaLaTeX support (i.e., it couldn't be loaded directly). I didn't 
want different behaviour, if possible, in XeLaTeX and LuaLaTeX, so loading 
xunicode inside fontspec seemed to be my only solution.

Please discuss while I'm away, if necessary. Happy to resolve the situation in 
another way if that would be better.

-- Will




--
Subscriptions, Archive, and List information, etc.:
  http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] [lltx] fontspec and xunicode

[Will Robertson]

On 05/08/2010, at 4:58 PM, Khaled Hosny wrote:

> Based on the previous discussion here, I think there is a general
> agreement that a simple wrapper around xunicode is better than forking
> it.

Where the wrapper == fontspec? (As I just implemented.)
Or do you mean an entirely separate package that is loaded individually?

I'd prefer to keep things simple for the user who might be using both XeLaTeX 
and LuaLaTeX, and I can't think of a good way to provide an interface to the 
latter option (the wrapper package idea) without causing confusion.

-- Will




--
Subscriptions, Archive, and List information, etc.:
  http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] [lltx] fontspec and xunicode

[Will Robertson]

On 05/08/2010, at 5:05 PM, Khaled Hosny wrote:

> Well, I can not actually upload fontspec to CTAN since I am unable to
> build the documentation :) (even if I got all the non-free fonts, I
> still can not generate the AAT examples). I think, adding the PDFs of
> the examples to the repository would help anyone to build the
> documentation without the fonts.

Hmm, good point. I don't really want to add the PDFs to the repository since 
they're effectively derived files, but if necessary you can get the PDFs from 
CTAN.

-- Will




--
Subscriptions, Archive, and List information, etc.:
  http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] [lltx] fontspec and xunicode

[Will Robertson]

On 05/08/2010, at 6:07 PM, Khaled Hosny wrote:

> And even with the "/doc" from ctan, I still can't build the package
> because `make doc` insists on generating the test files which fails
> here.

Bleah. Sounds like I need to revisit the makefile.

You can force the tests to pass with `rm testfiles/*.pdf` then `make initest`, 
but unless you're in a great hurry I don't mind releasing fontspec to CTAN when 
I get back.

Thanks for looking into it,
-- Will




--
Subscriptions, Archive, and List information, etc.:
  http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] unicode-math (I think) redefines \slash

[Will Robertson]

On 2010-08-12 04:54:45 +0930, "Joel C. Salomon" 
 said:



Using all packages as of TL ’10 pretest.

At one point in my document I needed an optional line-break after a
slash, but
blah blah overfull line UNIX\slash Linux
didn’t break.  Inserting the line
\show\slash
showed me that \slash was now simply the character ‘/’.  Even
re-defining it in my preamble didn’t work right away; I needed
\AtBeginDocument{\def\slash{/\penalty \exhyphenpenalty}}
since unicode-math also does its manipulations then.


Hi Joel,

Thanks for reporting this. (And noted in the bug tracker -- thanks for 
that especially so I don't forget!). Definitely a problem with 
unicode-math. I'll take a look when I can.


Cheers,
Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] Checking for existence of font feature

[Will Robertson]

On 2010-08-12 16:16:54 +0930, Khaled Hosny 
 said:



Ah, thank you. Are these features (relatively) new? Rather
embarrassing not to have spotted them, but anyway…


All were introduced in version 2, I think.


Yep, very new.


I've got a problem using these commands. Using XeTeX v0.9995.1
texlive svn 15079 on MiKTeX 2.8, fontspec 2010/08/01, none of these
“lower-level” fontspec features I've tried work.

I've copied the appropriate example from the manual:

\fontspec_if_feature:nTF {smcp} {True} {False}.



Since those command use expl3 syntax, you will need \ExplSyntaxOn before
using them.


Right; something like this:

\ExplSyntaxOn
% note that spaces are completely ignored in here!
\newcommand \testcaps [2] {
 \fontspec_if_feature:nTF {smcp} {#1}{#2}
}
\ExplSyntaxOff
...
\fontspec{Times New Roman}
\testcaps{Caps}{No Caps}

Hope this helps,
Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] OldStyle Numbers not changeable to Lining Numbers

[Will Robertson]

On 2010-08-17 08:23:15 +0930, Tobias Schoel 
 said:



\defaultfeatures{Numbers=OldStyle} gives medieval numbers,
\addfontfeature{Numbers=Lining} afterwards keeps medieval numbers.

Minimal Example:

\documentclass{article}
\usepackage{fontspec}
\defaultfontfeatures{Numbers={OldStyle}}
\setmainfont[Mapping=tex-text]{Linux Libertine O}
\begin{document}
A0123456789

\addfontfeatures{Numbers={Lining}}A0123456789
\end{document}


"Mixing" conflicting font features like this is not very reliable in 
fontspec, because after saying Numbers=Lining, fontspec is not smart 
enough to deactivate Numbers=OldStyle. (Unfortunately. Work is planned, 
theoretically, to fix this problem.)


So you end up with the font having to make the decision about what to 
do with BOTH features active at the same time. Which is all a little 
frustrating, since fontspec is supposed to be making things easier to 
use, but oh well.


On a related note...

On 2010-08-17 16:31:35 +0930, "M. Niedermair" 
 said:



\newfontfamily\libertineX[Mapping=tex-text,
   RawFeature=+liga% ;+pnum
  ]{Linux Libertine O}
...
You can change the rawfeature as you like.
smcp, frac, hlig, dlig, lnum, pnum, zero, ...


While there's nothing wrong with this, I'll just point out that these 
are all equivalent to fontspec features such as Ligatures=Common, 
Numbers=Lining, and so on.


There's no error checking on the raw features (yet), which is the main 
reason I recommend using the fontspec features, but I also find the 
OpenType features harder to remember. Whichever you're more comfortable 
with, though.


W




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] OldStyle Numbers not changeable to Lining Numbers

[Will Robertson]

On 2010-08-18 19:24:01 +0930, "M. Niedermair" 
 said:



Is there a way to display with fontspec all supported features of a
special font?


No (eventually planned), but you can use the opentype-info.tex 
"document" to do this.



Are all standard otf feature mapped to fontspec-parameters?


If not, they should be. The manual for v2 tries to be more explicit 
about how fontspec features relate to OpenType features.


W






--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] XeTeX documentation? (from "Checking if a font exists")

[Will Robertson]

On 2010-08-31 00:02:43 +0930, Heiko Oberdiek 
 said:



However it's quite incomplete, two examples:
* \suppressfontnotfounderror is missing
* For the correct syntax of \XeTeXpdffile I had to look into
  the sources of XeTeX.


I've added \suppressfontnotfounderror to the source:
   http://github.com/wspr/xetexref

Will update to CTAN sooner or later. (I have trouble doing so from 
behind the proxy at uni.)


I don't know the syntax of \XeTeXpdffile either, so contributions 
accepted. I figure this is one of those primitives that will rarely 
come up since it's well abstracted already by \includegraphics.



But better a good start than none at all.


Thanks :)

I just add primitives when I notice they're missing. This resulted in a 
couple of primitives being left out of LuaTeX because of omissions of 
mine here...oops.


Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] XeTeX documentation? (from "Checking if a font exists")

[Will Robertson]

On 2010-08-31 21:22:31 +0930, Shrisha Rao 
 said:



There is an interesting paper titled "If writers can't program and programm
ers can't write, who's writing user documentation?" -- see http://portal.ac
m.org/citation.cfm?id=10563.10574.



Great link, thanks!

Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] XeTeX documentation? (from "Checking if a font exists")

[Will Robertson]

On 2010-09-01 18:34:30 +0930, Heiko Oberdiek 
 said:



The rotation is specified in degree (360 is full circle)
and the rotation is counterclockwise as usual.


Thanks for the info. I've added it to the project, and I'll try and 
write it up sooner rather than later:


http://github.com/wspr/xetexref/issues/issue/3

W




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] Invalid fontname

[Will Robertson]

Philip Taylor (Webmaster, Ret'd  writes:

> I get a different error, using TL 2010 pre-release :
> 
> (e:/TeX/Live/2010/texmf-dist/tex/latex/fontspec/fontspec.cfg))
> ! Undefined control sequence.
>  \newfontinstance

This command is deprecated; please use \newfontfamily or \newfontface instead. 

Cheers,
Will



--
Subscriptions, Archive, and List information, etc.:
  http://tug.org/mailman/listinfo/xetex

[XeTeX] XeTeXpdffile media box etc

[Will Robertson]

Hi,

I'm writing up the syntax of \XeTeXpdffile and \XeTeXpicfile and I'd 
like someone who knows better than I do to explain the optional 
argument to \XeTeXpdffile to control the bounding box of the graphic:


[ crop | media | bleed | trim | art ]

Any volunteers? (If these are standard "industry" terms, is there a 
good reference page to link to, perhaps?)


Also, is there a way to access this functionality through \includegraphics?

Thanks,
Will





--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] XeTeXpdffile media box etc

[Will Robertson]

Hi,

Thanks Paul and William for the useful info. Another clarification:

Crop, Bleed and Trim are standard printing terms and the usage reflects 
that. Media is used to describe the underlying page size to which one 
would likely be printing the file and art is what one wants people to 
see.


And how do these behave? In all cases, is the PDF cropped to whichever 
bounding box is specified? Or do they just relate to the alignment of 
the inserted box? Or does it vary between them?


(Sorry for the continuing questions, without appropriate images I can't 
test for myself!)


Cheers,
Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] XeTeXpdffile media box etc

[Will Robertson]

On 2010-09-10 01:14:34 +0930, Heiko Oberdiek 
 said:


Who are the maintainers of xetex.def? Can they add the option `pagebox' 
as well?


Not me, I'm afraid. The copyright lists Jonathan (Kew) and Ross Moore.

W




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] XeTeXpdffile media box etc

[Will Robertson]

On 2010-09-09 16:33:55 +0930, Paul Isambert 
 said:


The pdfTeX manual states:

"one may also decide in the pdf image case, which page box of the image 
is to be

treated as a final bounding box."

I guess XeTeX works in the same way.


Thanks.

For what it's worth, the development version of xetexref with the 
additional documentation for \XeTeXpicfile and \XeTeXpdffile can be 
found here:


http://github.com/wspr/xetexref/blob/master/XeTeX-reference.ltx

The descriptions are a bit terse, but they're about all I've got time 
for. Any corrections/improvements happily incorporated; I'll be 
uploading it probably late weekend.


Cheers,
Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] XeTeX documentation "initiative"

[Will Robertson]

On 2010-09-12 05:11:54 +0930, Philipp Stephani 
 said:


OpenType Math is still in a very early stage in XeTeX and has so many 
bugs that it is not ready for production use.


Thanks for the vote of confidence.
And where are all these OpenType math fonts you'd like to use?

W




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] XeTeX documentation initiative - a website

[Will Robertson]

On 2010-09-12 19:29:31 +0930, Michiel Kamermans 
 said:



I've put up a wiki with a highly tentative structure on
http://wiki.xelatex.org/doku.php - sadly xetex.org is taken by some
domain hiking service so there's no obvious xetex.org counterpart to be
had...


Looks great so far. If I could tentatively suggest a guideline for 
adding more "general LaTeX"-related material:


- We already have very good documentation for lots of packages
- So let's not to avoid replicating as much detail as possible from 
those packages with good manuals
- A description of "why" you would like to use such-and-such a package, 
and let the package speak for itself in regards to the "how"

- In other words, stick to illustrative examples
- Where possible, recommend packages that have "best of breed" 
solutions and try not to overlap functionality: choice can be evil (but 
isn't always, obviously)


I don't have time to write content, at the moment, but I'll keep and 
eye on what's there and edit where I can see improvements to be made. 
E.g., there's no need to recommend xltxtra nowadays; since TL2010 
fontspec does everything itself, and xltxtra just contains mostly 
unused "extras".


Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] Fontspec: Historic not Historical Ligatures?

[Will Robertson]

On 2010-09-13 00:04:53 +0930, David Perry 
 said:



On 9/12/2010 10:05 AM, Kevin Klement wrote:


Wouldn't it be possible, however, to
alias one to the other so that both worked?

That would make sense, unless something else is going on here that we
don't know about; Will can perhaps speak to this.


Sorry about all this mess.

I did change the syntax to Ligatures=Historic to be consistent with 
Style=Historic but I forgot to keep the old name around—oops!


Please add

   \aliasfontfeatureoption{Ligatures}{Historic}{Historical}

to your preamble or fontspec.cfg, and I'll add the same line in the 
fontspec code for the next update.


Thanks,
Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] wspr's realscripts & updated xelatex templates

[Will Robertson]

On 2010-09-15 04:19:40 +0930, Khaled Hosny 
 said:



On Tue, Sep 14, 2010 at 10:44:06AM -0400, Joel C. Salomon wrote:

I’ve not been following the recent back-and-forth regarding which
XɘLaTeX packages are now obsolete, and which are compatible with LuaLaTeX.

Right now my personal style files files have lines like these:
\ifxetex
\RequirePackage{fontspec, xunicode, xltxtra}
\fi
\ifluatex
\RequirePackage{fontspec}
\fi
Am I missing something here?  Where does realscripts fit in?


The xltxtra documentation isn't that complex, is it? :) This is what 
the readme says:


- Loads fontspec and Ross Moore's xunicode automatically.

- Loads Andrew Moschou's metalogo package for \XeTeX and \XeLaTeX logos.

- Loads the fixltx2e package and patches other LaTeX commands:
 \textsuperscript & \textsubscript:
now use fontspec to access
real superior/inferior characters,
 \showhyphens: now works,

- Defines the dubiously useful commands
\vfrac - for vulgar fractions with fontspec
\namedglyph - to access font glyphs by name

Of these, #1 is covered just by \usepackage{fontspec}, #2 you can do on 
your own if you need it, #3(a) is what realscripts does (for XeLaTeX or 
LuaLaTeX), #3(b) isn't used often, and #4(a,b) are probably never used.


When I have a moment I'll remove the code from xltxtra that does #3(a) 
and just load the realscripts package instead.


Long story short, I'd recommend these days just loading fontspec and 
seeing if that works for you :)




Remove xltxtra, add realscripts (if you need its functionality, but
usually the fonts are broken so it does not work the way it should), now
if you need the extra logos, load one of the packages that provide
extra tex logos, hologo for example.


I wouldn't say the fonts are "usually" broken, but they are broken 
often enough that doing it automatically in fontspec would be a bad 
idea.


Cheers,
Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] wspr's realscripts & updated xelatex templates

[Will Robertson]

On 2010-09-16 00:01:32 +0930, Ulrike Fischer 
 said:



Am Tue, 14 Sep 2010 21:49:40 +0300 schrieb Khaled Hosny:


Fontspec loads xunicode for luatex


Only the version from github. The "official" version on CTAN
2010/08/01 v2.0c  doesn't load xunicode yet.


Oops, thanks. Need to get that updated ASAP.
I was meaning to do so before I went travelling and it then slipped my mind.

W




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] wspr's realscripts & updated xelatex templates

[Will Robertson]

On 2010-09-15 17:50:42 +0930, David Cottenden 
 said:



On 15/09/10 09:08, Will Robertson wrote:


- Defines the dubiously useful commands
\vfrac - for vulgar fractions with fontspec
\namedglyph - to access font glyphs by name



#4(a,b) are probably never used.


No! Don't remove these, please! They have at least /one/ user...
\vfrac is especially useful.

Am I alone in this, or does anyone else use them? Speak now, or I'll
shut up, go away, and copy the bits of code out of xltxtra before their
demise...


Oh, don't worry, xltxtra will never go away -- but at this point in 
time I'm no longer recommending that it needs to be used "by default" 
by most users.


In time I'll merge \vfrac's functionality into the xfrac package, I 
suspect. Until then continue to use xltxtra :)


W




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] \...@nameuse question

[Will Robertson]

On 2010-09-17 19:44:54 +0930, Michiel Kamermans 
 said:



 \if\value{iclass}<\...@nameuse{#1class}


These lines should be \ifnum, at the very least. You might also want to 
terminate them with \relax:



  \if\value{iclass}<\...@nameuse{#1class}\relax


I haven't checked the rest of the code for correctness, however :)

W




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] Separate fonts for Latin and Greek with fontspec?

[Will Robertson]

On 2010-09-18 03:17:21 +0930, "Fr. Michael Gilmary" 
 
said:



Gareth Hughes wrote:


I've been using that version of kashida.sty with Syriac definitions for
some time now. Are there any plans to make it an official CTAN package, 
or part of one?


Well, I haven't a clue how to do that.


Simply create a README and a PDF documentation file for the package, 
and send it to . You'll need to assign a 
license to the code, but when in doubt the LPPL is what is usually used 
for LaTeX packages.


Doing so will ensure that the package is included in TeX Live and 
(assumedly) MiKTeX.


W





--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] Let me get it straight...

[Will Robertson]

On 2010-09-19 00:27:40 +0930, Herbert Schulz 
 said:


Right now I'm loading xltxtra alone (since it loads xunicode and 
fontspec). In some of my older documents I am loading all three. With 
an updated fontspec (as of version ?) under xelatex which of the three 
packages do I need to be loading. Just trying to streamline my xelatex 
templates.


Right, sorry for all the confusion around this. I think I spoke too 
soon about what people should do before I actually updated the code!


You should now only really need to load fontspec only:
- xunicode is loaded internally by fontspec.
- all the *necessary* parts of xltxtra are now inside fontspec

(You'll need very recent versions of all the bits'n'pieces for this to 
be correct.)


Edge cases (all the "unnecessary" parts of xltxtra):

- if you want OpenType subscripts/superscripts, load the realscripts 
package separately or load xltxtra again

- if you want the XeTeX/XeLaTeX logos, load metalogo or load xltxtra again
- if you want any of the following, load xltxtra again:
 \showhyphens that works
 \vfrac
 \namedglyph

So if you don't want to have to think about these extra things, there's 
no problem with loading xltxtra always and by default. But I'd 
recommend new users just load fontspec -- and this is the only option 
on LuaLaTeX, as xltxtra doesn't exist for that engine.


Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] Let me get it straight...

[Will Robertson]

On 2010-09-19 21:02:27 +0930, Herbert Schulz 
 said:


Thanks! That's the information I was wanting. Do you have version 
numbers after which this is all true? Are the bits and pieces released 
to CTAN and now in TeX Live?


Everything is now on CTAN but not quite yet in TeX Live I think. But 
since my last fontspec update there's another bug been reported 
(involving Colour) so I need to fix that first :)


The current versions on CTAN are

fontspec v2.1 (soon to be v2.1a)
euenc 0.1h
xltxtra 0.5e

Cheers,
Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] If font under fontspec

[Will Robertson]

On 2010-09-21 09:39:51 +0930, Gareth Hughes 
 said:



Peter Dyballa wrote:


Am 20.09.2010 um 20:44 schrieb Gareth Hughes:


I'm asking for a more useful implementation that checks whether a
certain font is in use now



Wouldn't it be sufficient to check what \{rm,sf,tt}default are set to?


Perhaps, but a more general, fontspec solution is needed because the
fonts I want to work with are Syriac fonts. Also, I need to activate
kerning corrections only when that font is in use, i.e. it would be no
good to see if the font is being used in a document and then apply the
kerning corrections for all fonts in the document.


It sort of depends how you want to write the code for the test. Here's 
a proof of concept that might do what you need:


\documentclass{article}
\usepackage{fontspec}

\makeatletter
\newcommand\IsFont[4][]{%
 \begingroup
   \let\curr...@family\f@family
   \fontspec[#1]{#2}
   \ifx\curr...@family\zf@family
 \global\l...@tempa\@firstoftwo
   \else
 \global\l...@tempa\@secondoftwo
   \fi
 \endgroup
 \...@tempa{#3}{#4}
}
\makeatother

\begin{document}
\newfontfamily\foo{Times New Roman}
\setmainfont{Helvetica}
test: \IsFont{Times New Roman}{Times}{not Times}
\par\foo
test: \IsFont{Times New Roman}{Times}{not Times}
\end{document}


I've considered adding such code to fontspec before. But this interface 
is not very reliable -- font options must be exactly the same for two 
fonts to be "equal". See how you go, though.


Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] Redefining \em and \emph

[Will Robertson]

On 2010-09-23 09:07:39 +0930, Gareth Hughes 
 said:



-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

As I understand it, the line about redefining emphasis and inner
emphasis at the end of section 6.3 of the fontspec documentation

XETEX users will need to load the xltxtra package before the advice 
above works.


is no longer true seeing as fontspec-patches now does this, and fontspec
loads it automatically.


You're right, thanks. I've fixed it up in the source.


I'm looking at ways to redefine emphasis within Syriac text (seeing as
slanted text is not traditional or pretty). I've got most of the
redefinitions working, but I'd also like to redefine \emph as an alias
for \aemph in Syriac text. What's the best way of doing that?


Good question.
I'm not very familiar with multilingual typesetting -- how do you 
switch fonts? Is there a switch that you can query in this case?


Cheers,
Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] Extra trailing space in IPA font with xltxtra

[Will Robertson]

On 2010-09-23 20:38:14 +0930, Paolo Matteucci 
 said:


Therefore *maybe", whatever approach is used to address this bug, the 
same "fix" should be applied to fontspec, too...?


The \scantokens is somewhere in tipa.sty or its accompanying packages. 
But the problem only occurs if fontspec is loaded. I don't know why 
that would be, though, off the top of my head.


Sorry, I'm out of time tonight to look any further.

W




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] Redefining \em and \emph

[Will Robertson]

On 2010-09-23 22:20:05 +0930, Gareth Hughes 
 said:



Will Robertson wrote:
On 2010-09-23 09:07:39 +0930, Gareth Hughes 
 said:



I'm looking at ways to redefine emphasis within Syriac text (seeing as
slanted text is not traditional or pretty). I've got most of the
redefinitions working, but I'd also like to redefine \emph as an alias
for \aemph in Syriac text. What's the best way of doing that?


Good question.
I'm not very familiar with multilingual typesetting -- how do you switch
fonts? Is there a switch that you can query in this case?


I think the most obvious query would be directionality, so \...@rtl from
Vafa's bidi package might be the appropriate test: it would be true for
Syriac (and Persian etc.) and false for English (and Korean etc.). I
don't know if polyglossia has any language conditionals, but that would
give finer control.


It doesn't seem to. In that case I'd write something like

\let\oldemph\emph
\renewcommand\emph{%
 \...@rtl \expandafter\aemph \else \expandafter\oldelse \fi
}

Hope that helps,
Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] Redefining \em and \emph

[Will Robertson]

On 2010-09-25 09:33:20 +0930, Gareth Hughes 
 said:



Will Robertson wrote:

It doesn't seem to. In that case I'd write something like

\let\oldemph\emph
\renewcommand\emph{%
\...@rtl \expandafter\aemph \else \expandafter\oldelse \fi
}


I still can't figure out how to do this for \eminnershape as defined in
fixltx2e. Any suggestions how to do that?


Do you mean so that you can write

   \aemph{ ... \aemph{ ...} ... }

and have the inner \aemph not put the overline on top? In this case, 
you can write (and this might be a reasonable addition to polyglossia)


\documentclass{article}
\makeatletter
\def\aemph#1{%
 \begingroup
   \let\aemph\inneraemph
   $\overline{\hbox{#1}}$%
 \endgroup
}
\def\inneraemph#1{%
 \egroup% close \hbox
 \egroup% close \overline
 $% close math
 {\eminnershape #1}% inner emphasis!
 $% "reopen" math
 \expandafter\overline\bgroup % "reopen" \overline
 \hbox\bgroup % "reopen" \hbox
}
\makeatother
\begin{document}
hello there \aemph{foo \aemph{oof rab} bar} baz black
\end{document}

Modulo checks for bidi and using \hboxR instead of \hbox and a 
"correct" definition for \eminnershape in this context, and so on. But 
this has nothing to do with the idea of \eminnershape: if you want to 
tie the two together, fontspec actually overrides fixltx2e's 
emph/eminnershape code:


\ExplSynaxOn
\DeclareRobustCommand \em {
 \...@nomath\em
 \tl_if_eq:xxTF \...@shape \itdefault \eminnershape
 {
   \tl_if_eq:xxTF \...@shape \sldefault \eminnershape \emshape
 }
}
\DeclareTextFontCommand{\emph}{\em}
\cs_set_eq:NN \emshape \itshape
\cs_set_eq:NN \eminnershape \upshape
\ExplSyntaxOff

Not sure if that helps directly...

Will





--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] Problem changing default fonts

[Will Robertson]

On 2010-09-27 02:14:03 +0930, Drébon 
 said:



\usepackage{pxfonts}
\usepackage{txfonts}
\usepackage[osf,sc]{mathpazo}


You just changed the math/text font three times...


\usepackage[scaled=0.85]{beramono}
\usepackage[euler-digits]{eulervm}
\usepackage{helvet}


...and this is the second change of the sans serif font.
I suggest culling your preamble a little.


The first group of packages are used to change the fonts. But if I use
the xltxtra package, they have no effect.

On the other hand, if I do not use xltxtra package, the œ character is
not accepted and thus 'cœu' becomes 'ceur'...


If you're only using UTF8 for Latin input text, you can use pdfLaTeX with

   \usepackage[utf8]{inputenc}

But if you insist on using XeLaTeX, you should probably go the whole 
hog and load a Unicode font. Try TeX Gyre Pagella as a replacement for 
mathpazo as your text font. Something like


\setmainfont[
 Extension=.otf,
 UprightFont=*-regular,
 ItalicFont=*-italic,
 BoldFont=*-bold,
 BoldItalicFont=*-bolditalic,
]{texgyrepagella}

Cheers,
Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] Small caps bug in fontspec

[Will Robertson]

On 2010-09-29 15:50:29 +0930, Andrew Moschou 
 said:



It seems setting Numbers=OldStyle breaks \scshape. This example uses Sabon,
but it happens with other fonts too. My fontspec version is 2.1a.


Hi Andrew,

Sorry for this, and thanks for reporting it. The bug fix (2.1b) is on 
its way to CTAN.


Cheers,
Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] unicode-math problem

[Will Robertson]

On 2010-09-29 21:29:22 +0930, Herbert Schulz 
 said:


Had another update of unicode-math this morning and have no problems 
compiling the file here.


Right, the two packages were out of sync for a couple of days. It 
should all be working now. Sorry for the inconvenience.


(In the future I need to decouple these packages a little more.)

Cheers,
Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] XeTeX in lshort

[Will Robertson]

On 2010-09-30 07:10:07 +0930, Philipp Stephani 
 said:


[Beginners] don't know what a text file or a text editor is, they have 
never heard the word "Unicode", and they have never used a programming 
language before. What they need are step-
 by-step instructions that tell them, in simple words, how to create 
TeX documents. If TeXworks/TeXshop is included in TeX Live and is good 
enough, then there is no reason to mention other e!



This is the important point, I think. TeXworks is *included* in MiKTeX 
and TeX Live for all platforms. IMHO, I think any comments that you're 
going to make along these lines should be


"MiKTeX and TeX Live both include the TeXworks program for writing your 
XeTeX and LaTeX documents and viewing their typeset output. You are not 
restricted to using TeXworks, however; a list of known editors that 
supports XeTeX are ..."


I agree with others who said this isn't the place to go into much 
detail about which editor they're using. Just point them in the right 
direction, and TeXworks is the easiest to get started with *because 
they don't have to install anything else*.


W




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] Problem with footnotes and xltxtra

[Will Robertson]

On 2010-09-30 18:13:50 +0930, Dominik Wujastyk 
 said:



A quick trial of this shows that the problem goes away when the "no-sscript"
argument is removed from xltxtra (and the "minimal" class is replaced by
"article" to get the \footnotesize command).

Next thing to do (not by me) is debug [no-sscript].


Strange.
I suspect that if you're loading xltxtra with the [no-script] option, 
then you no longer need xltxtra. I recommend, now, just


   \usepackage{fontspec}

instead -- it will automatically load xunicode if you have the fontspec 
v2.1 or later floating around (2.1b is new to CTAN). You can load the 
realscripts package if you want the features provided by xltxtra for 
OpenType superscript symbols in footnotes.


In terms of debugging the problem, I haven't worked out what causes the 
error but I know how to fix it. There seems to be a strange interaction 
with \DeclareRobustCommand and the xparse package which we'll need to 
sort out.


Cheers,
Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] Problem with footnotes and xltxtra

[Will Robertson]

On 2010-09-30 19:25:45 +0930, Will Robertson 
 said:



In terms of debugging the problem, I haven't worked out what causes the
error but I know how to fix it. There seems to be a strange interaction
with \DeclareRobustCommand and the xparse package which we'll need to
sort out.


Scratch that. Just author stupidity in realscripts.sty.

W




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] Undefined control sequence

[Will Robertson]

On 2010-10-03 00:08:20 +0930, Cesar Romani 
 said:



I'm using xelatex from TexLive 2008 on Win XP.

\usepackage{fontspec}
\usepackage{xunicode}
\usepackage{xltxtra}
\usepackage{polyglossia}



I'm afraid your example works correctly for me. There has been some 
development on these packages since 2008; I recommend upgrading your 
distribution if possible. Otherwise, I suspect grabbing the latest 
copies from CTAN and installing them manually will be your best option.


Cheers,
Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] hyperref broken??

[Will Robertson]

Am 03.10.2010 um 15:52 schrieb Will Robertson:

> On 03/10/2010, at 11:41 PM, Philipp Stephani  wrote:
> 
>> Am 03.10.2010 um 14:02 schrieb David Perry:
>> 
>>> If I omit the option, or use 'xetex' in place of 'hypertex' or use the 
>>> driverfallback option, I get:
>>> 
>>> ! Undefined control sequence.
>>> \str_if_eq_p:nn #1#2->\tex_ifnum:D \pdf_strcmp:D
>>>   {\etex_unexpanded:D {#1}}{\...
>>> l.479 ...p:w  \cs_set_nopar:cpn  \cs_set_nopar:Npn
>>> 
>>> I fixed the missing commas but that made no difference.
>> 
>> Looks like an expl3 problem: \pdf_strcmp:D is pdfTeX's \pdfstrcmp, which 
>> might be nonexistent or available under a different name in XeTeX. This 
>> *might* get fixed by updating expl3; otherwise write to the LaTeX list 
>> (late...@listserv.uni-heidelberg.de, I've put it on CC).
> 
> I think I fixed this coincidentally earlier today. The update will roll out 
> in a timely fashion.  



Now that I sit down with the code in front of me, I can't comprehend where this 
problem is coming from.

Oh, maybe the problem is that some dastardly package is setting \XeTeXversion 
to \relax? But then, no, this would only cause problems in non-XeTeX use, from 
what I can see. I can't work this out without stepping through the actual 
problem.

Can someone post a minimal example that fails under MiKTeX? (Including 
\listfiles in the preamble.)

-- Will




--
Subscriptions, Archive, and List information, etc.:
  http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] hyperref broken??

[Will Robertson]

On 2010-10-04 11:57:27 +1030, "David J. Perry" 
 said:



Ross and Will,


So the first question should be what is the version of XeTeX that MikTeX
is providing.
And is David actually using that, or still an earlier version, as I was on
a Mac.
If the primitive is not there, then the latest versions of expl3 and
fontspec just will not work, and may not even succeed in loading
successfully. Is that not correct?


As the person  who raised the issue, I would like to help solve it.
Unfortunately, the problem developed on my machine upstate, where I was for
the weekend.  I'm now back in the city to teach tomorrow morning, and my
machine here has not had the MiKTeX update that I applied this weekend.  If
no one else sends in an example, I will certainly do so when I get back
upstate on Friday.


On further reflection, even confirming that the "not-updated" case 
compiles correctly would be helpful. (As it would indicate 
approximately when the error appeared.) Can you prepare the same 
minimal example on your current setup?


Thanks,
Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] XeTeX in lshort

[Will Robertson]

On 2010-10-05 05:21:21 +1030, Apostolos Syropoulos 
 said:



written in  WEB, which is a Pascal-based system + documentation; it is often
converted to  C for compilation.  And TeX itself doesn't look anything like
LISP to  me, but maybe I'm missing s.t.?  (Like a CAR and a CDR and...)


Several years ago I read somewhere, but I do not remember where, that TeX was
inspired by LISP. It is obvious that TeX processes lists and list form the core
of LISP,
thus, I suppose this is the connection between TeX and LISP


Not really having done any programming in Lisp, as far as I'm aware the 
connection is tenuous at best. Sure, there's this loose analogy with 
lists, but Lisp is a real programming language. TeX is not; it simply 
has enough power to masquerade as one. You can express some Lisp-like 
concepts in TeX, but that's as far as it goes. E.g.,


"While learning TeX I became a bit frustrated that TeX macros were so 
cumbersome as a programming language. I had been learning Lisp, so I 
decided to use it as an extension language for TeX..."  



W




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] Prime + Unicode-math

[Will Robertson]

On 2010-10-05 09:07:07 +1030, Kamal Abdali 
 said:



The math symbols in XITS seem rather small. When trying to increase XITS
character size with the Scale option, I noticed that math accents get
amplified too much, and get placed too far from the subject symbols. The
example below shows that by comparing XITS and Asana math fonts.

\documentclass{article}
\usepackage{unicode-math}
\begin{document}
\setmathfont[Scale=2.5]{XITS}


You need to use the XITS Math font instead.

I can see differences in the height of the accent placement (compared 
to Scale=1), but it doesn't necessary look "wrong" to me. This 
behaviour is not seen in LuaLaTeX, so this could be a variation in how 
XeTeX chooses to place these symbols.


Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] Math italics with unicode-math.

[Will Robertson]

On 2010-10-10 08:08:20 +1030, Khaled Hosny 
 said:



fontspec Warning: Font 'Linux Libertine O' does not contain script 'Math'.


I guess Libertine isn't a math font.

Right, and this warning should be harmful as regard of this context (it
basically means you want get proper radicals or large operators and
stuff like that, but you are using XITS Math for that anyway).


I should probably address this, since it's a frequency complaint. I 
might only send these warnings to the log file.



Nevertheless, similar markup seemed to
yield the desired result in June at http://www.charlietanksley.net/
philtex/the-unicode-math-package-for-xelatex-and-the-stix-fonts/">this blog. Has something changed dramatically since then?


Their example does not have spaces, so I'm not sure of this is a new
regression or it was like that from the start.


Hmmm, not sure either. I've added this as a bug to the issue tracker on GitHub.

Thanks for the help,
Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] Primes and unicode-math.

[Will Robertson]

On 2010-10-12 06:31:36 +1030, Khaled Hosny 
 said:



Put the line
\setmathfont{xits-math.otf} AFTER the line
\setmathfont[range=\mathit/{latin,Latin,greek,Greek}]{Linux Libertine O 
 Italic}


I'm not sure this would be side effect free, at least in XeTeX since
many math typesetting parameters are derived from the first font. What
do you think, Will?


Right; unless I'm missing something here the XITS fonts will override 
the Libertine ones entirely.


Sorry about the prime problem; I'll try to look into it. (Khaled's 
suggestion the other day is a good work-around for now.)


W




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] New environment fails in float

[Will Robertson]

On 2010-10-12 01:20:09 +1030, Gareth Hughes 
 said:



I'm having problems with an environment definition. Using the xparse and
etoolbox packages, I have created the environment syriacpoem in my .sty
file.


Hi Gareth,

This appears to be a bug with nested environments in xparse. I've 
reported the bug and hopefully a fix will be forthcoming.


I'm not aware of an immediate workaround, I'm afraid. Perhaps for now 
you could define the example environment as a command and that might 
work instead? E.g., (untested)


\newsavebox\mysavebox
\newcommand\example[2]{%
 \begin{figure}[h]\centering
 \savebox\mysavebox{\parbox{#1}{#2}}%
 \ovalbox{\usebox\mysavebox}%
 \end{figure}
}
...
\example{\linewidth}{%
 \begin{syriacpoem}
 ...
}

Hope this helps,
Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] Error with latest expl3?

[Will Robertson]

On 2010-10-17 02:17:37 +1030, Herbert Schulz 
 said:




On Oct 16, 2010, at 10:31 AM, 
enrico.grego...@univr.it wrote:



\expandafter\let\csname intexpr_if_even:nTF\expandafter\endcsname
\csname int_if_even:nTF\endcsname



That certainly fixes things here. Kind of ugly though. :-)


Oops, my fault, sorry.
I missed a few lines in the compatibility code.

I wish I could get my head around the expl3 stuff. Right now it only 
gives me headaches. It also still seems to be quite dynamic so we 
continuously end up with problems in packages that are used quite 
extensively. Eventually it will all stabilize.


Well, we try to avoid these problems but we've recently had a bad run. 
For me, I'm trying to do too much so I'm making careless errors to 
often for my liking. Sorry for the inconvenience,


Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] bug in fontspec?

[Will Robertson]

On 2010-10-19 03:30:24 +1030, Pablo Rodríguez 
 said:



as Ulrike Fischer has noticed
(http://tug.org/pipermail/xetex/2010-October/018895.html), fontspec
enters in a loop italics are defined as slanted:

\def\itdefault{sl}


This problem, funnily enough, has existed pretty much forever. I guess 
you're the first person to use the slides class with fontspec :)


This is now fixed in the nightly build of fontspec, which you can 
install iff you have TeX Live 2010 with the following command:


tlmgr --repository http://tlcontrib.metatex.org/2010 update fontspec

(more details here: 
)


The 

package source on GitHub (http://github.com/wspr/fontspec) contains the 
latest version as well, of course.




(I thought I had already reported this issue
[http://tug.org/pipermail/xetex/2010-October/018895.html], but due to
the lack of proper description, it hasn't been taken as a proper bug
report.]


Sorry, I'd missed this. I don't necessarily follow this list exhaustively.

Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] bug in fontspec?

[Will Robertson]

On 2010-10-19 22:55:23 +1030, Herbert Schulz 
 said:


How long in testing before you submit the update to CTAN/TeX Live?


Who knows :)
I'm not sure what precendent to set here. I don't exactly want to 
bother the CTAN people for every minor bug fix, but I don't want to 
wait too long before releasing new versions officially.


I'll tentatively say that I'll release a new version to CTAN on the 
second weekend after the last nightly build on TLContrib. That should 
give me enough time to catch any accidental slip ups and chain any 
minor bug fixes together.



Is the fix for the expl3 problem local to fontspec too (i.e., in this updat
e) or will that need a change to expl3?


The updated version of fontspec should fix this problem too.

Cheers,
Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] Problems with spacing in math formulae

[Will Robertson]

On 2010-10-20 07:52:09 +1030, Ross Moore 
 said:



On 20/10/2010, at 6:14 AM, Christian Pech wrote:


I already reported this problem a week ago, but maybe the message got lost.
The following latex-file produces wrong spacing when compiled with xelatex,


I confirm the bad layout, as reported below.
This is using freshly download versions of  unicode-math
and  euler.otf .


Strangely enough, I *cannot* reproduce this.
(I've had a similar bug report before.)
I suspect it's an engine bug. I'm using TL2010 on Mac OS X 10.6.

W




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] Problems with spacing in math formulae

[Will Robertson]

On 2010-10-20 09:13:36 +1030, Khaled Hosny 
 said:



I can reproduce it and while ago have seen it in a posted sample
somewhere on the web. I don't know what is special about Euler's
display summation to trigger such a bug (if it is actually a bug),



Strange, with a little more effort I can now reproduce the problem, 
too. Apparently Font Book wasn't installing the latest version over the 
top of the old one correctly. I must remember to remove old fonts 
before I try and install new versions.


So I can revert to the correct behaviour by installing an older version 
of the font with Unique Name: "FontForge 2.0 : Neo Euler : 18-9-2009" 
(quite a lot older, I know).


Does this help track down the problem, Khaled?

Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] Problems with spacing in math formulae

[Will Robertson]

On 2010-10-20 19:09:52 +1030, Taco Hoekwater 
 said:



On 10/20/2010 10:35 AM, Khaled Hosny wrote:


Very useful! Knowing that there were a working version I was able to
track this down (with some git bisect magic), the culprit is the
DisplayOperatorMinHeight math parameter, while ago I increased it from
1400 to 2180 and XeTeX seems not to like that. Now I need to remember
why this was done in the first place (never use terse commit messages,
one day you will regret it).


That was because otherwise the display style operator size in luatex
was too small (essentially outputting a text style operator in all
cases).


Is it possible to set a happy medium here or is there an inconsistency 
(or bug) in XeTeX in how this value is being used? If necessary, 
unicode-math can manually override whichever fontdimen setting this is 
in XeTeX...


W




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] WordSpace in SizeFeatures

[Will Robertson]

On 2010-10-27 22:11:47 +1030, Vadim Radionov 
 said:



Is it possible to set different interword space for different font
sizes?
SizeFeatures={
{Size=-10,LetterSpace=-10,WordSpace={20,1,1}},
{Size=10-,WordSpace={1.1,1,1}}}


I am afraid that this is a limitation in LaTeX's font selection 
mechanism (although not in TeX itself).


If you need to change the inter-word space locally, you can use the 
\spaceskip and \xspaceskip commands (see TeX by Topic §20.3).


I should document all this in fontspec itself. I've added a note to do 
so for the next release.


Cheers,
Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] Font variant selection in Fontspec/Mathspec options incomplete

[Will Robertson]

On 2010-11-10 19:09:33 +1030, Christian Schmidt 
 said:



Selecting a font via the extension option/feature from Truetype or Opentype
 font files and, at the same time, specifying the corresponding font varian
ts through the other options, fontspec/mathspec is not able to set the vari
ant "Bold Italic". It simply seems to ignore the "BoldItalicFont" option th
at I have set.


I think this is now fixed in the most recent release (v2.1d).

Sorry for the inconvenience,
Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] problem with package versions?

[Will Robertson]

On 2010-11-15 10:18:56 +1030, Adam McCollum 
 said:



When I run XeLaTeX on a file (I've tried several), I get this error
message.
[snip]
By the way, I've run TeX Live and updated everything.


You must be using TeX Live 2009. The current version is TeX Live 2010, 
and you can't upgrade from the former to the latter -- it must be 
installed fresh (multiple distributions happily live in parallel, 
though).


Another alternative is that you have an old version of fontspec 
installed in your local texmf tree.


Cheers,
Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] Getting vinculum in fractions with Asana Math and unicode-math when compiled with xelatex

[Will Robertson]

On 2011-01-05 11:50:55 +1030, Paul Vojta 
 said:



I've attached a patch that fixes the problem, seems not to break anything
else, and seems reasonable.

It also fixes:

problems with segfaults with delimiters (xe-test-delim.tex)

problems with incorrect rendering of radicals (xe-test-sqrt.tex)

rendering problems with \overleftarrow in the expression
  \[\alpha+\varpi\Longrightarrow\uplus\overleftarrow{abc}\]


Fantastic, great work!

This has been a common problem over the last year or so. Are you able 
to get it into the main source repository or do you need someone else 
to apply it? (I don't *think* I have access to the XeTeX repository any 
more...)


Thanks,
Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] xdv2pdf vs xdvipdfmx

[Will Robertson]

On 2011-01-21 18:57:06 +1030, Oleg Parashchenko 
 said:



what is the state of xdv2pdf? Is it ok to use this driver in production
environment, or are there any issues? What xdvipdfmx provides, what
xdv2pdf can not do?


It's probably safe to say that xdvipdfmx is the more supported option 
these days. In terms of functionality, xdvipdfmx can do things like 
pstricks support and better support for various PDF related functions, 
since it uses (an extended version of) dvipdfmx.


On the other hand, xdv2pdf can do font transparency and multiple 
master/AAT fonts and it accepts a greater range of image formats. I've 
long in the past had some issues printing from xdv2pdf-created PDFs, 
but nothing too serious.



(Starting with 10.5, Mac OS X provides a font-from-file function.
A test code does work under my 10.6.)


How does this work? Can you show an example?

Best wishes,
Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex

[XeTeX] Testing unicode-math

[Will Robertson]

Dear all,

After many months of work largely by Philipp Stephani, I've finally got around 
to fixing a couple of niggling bugs and packaging up a new release for 
unicode-math. In general there are no large changes, but I'd still like to make 
a test release of the package available in case I've accidentally messed 
anything up :)

The new version of unicode-math is now available from the TLContrib server; in 
short, you can obtain it by running 

tlmgr --repository http://tlcontrib.metatex.org/2010 update unicode-math

on the command line. If you haven't updated your TeX Live packages for a while, 
then you'll probably want to run

tlmgr update --all

before installing unicode-math just to make sure you're up to date.

See here for some more details on how to use TLContrib:
 



Any comments are much appreciated -- especially anything along the lines of "it 
works!" :) If no serious problems are found I'll send this version to CTAN in 
the coming week.

Best wishes,
-- Will




--
Subscriptions, Archive, and List information, etc.:
  http://tug.org/mailman/listinfo/xetex

Re: [XeTeX] \XeTeXglyphbounds question

[Will Robertson]

On 2011-02-04 23:57:38 +1030, Jonathan Kew 
 said:



On 4 Feb 2011, at 05:41, Adam Twardoch (List) wrote:


I could use:
\XeTeXcharglyph`f

but this only gives me the glyph ID of the *default* glyph for the "f"
character. Yet since the font uses contextual alternates, I may end up
having any alternate "f" there, depending on the contents of \sampletext

So: how do I find out which glyph ID is the first (or 2nd, or last, for
that matter) in my box?


Unfortunately, I don't think there's currently any way to do that.



This reminds me that it would also be very useful to be able to 
differentiate glyphs when OT features are applied; e.g., if 
\XeTeXcharglyph expanded to different values according to the currently 
enabled font features (and contrarywise allow you to detect when a 
feature doesn't affect a certain glyph).


Is this a fundamental limitation with the means by which XeTeX handles 
\XeTeXcharglyph? In a way, it's surprising that


\font\1="[texgyrepagella-regular.otf]" at 12pt
\font\2="[texgyrepagella-regular.otf]:+smcp" at 12pt

\1 a: \the\XeTeXcharglyph`\a /
\2 a: \the\XeTeXcharglyph`\a

would result in the same glyph slot. Or have I fundamentally 
misunderstood the concept of "glyph slots"? (I *thought* they were a 
unique index to each glyph in a font, but now I'm considering they 
could still be affected by substitution features. And in fact that 
would make quite a bit of sense.)


Best,
Will




--
Subscriptions, Archive, and List information, etc.:
 http://tug.org/mailman/listinfo/xetex