Hi Hans, Thanks for your explanations. When using constructions such as $f’’_2(x)$, I will adapt my input in order to get what seems correct. Best regards: OK

On 17 May 2017, at 17:17, Hans Hagen wrote:

On 5/17/2017 8:35 AM, Otared Kavian wrote:

...

Hi Hans,

Thanks for having fixed the behavior of the prime in the latest beta release (ConTeXt version 2017.05.15 21:48). However, as you may see in the two attached PDF files, the result is still incorrect when one types for example $f_2’’(x)$, which should give the same result as $f’’_2(x)$. The correct behaviour is that obtained with mkii. Indeed one can circumvent the behaviour of the latest beta release of mkiv by using tricks such as the ones in the example code below, but this should be avoided in my opinion.

well, it depends on how one sees this '' mess

the problem with primes is that in macro packages they are messy entities that should collapse or not (active char mess) and the more clever a macro packages becomes the harder it becomes to makes it robust and/or to fight against such heuristics

add to that the fact that the symbol itself is a funny positioned something already raised in a font or not i.e. sometimes in need of superscript and sometimes explicitly not

for unicode we need to turn two '' into a double and ''' into a and triple such

that said, in context these thingies are at some point intercepted and their current state is looked at: superscript? multipel in a row? one ebing the nucleus of another ... etc .. the current (working again) behavior is what we support and i don't look forward to more heuristics

...

The following PDF have been obtained with mkii and mkiv with the following example code. Best regards: OK

%%% begin derivatives-prime.tex \starttext

\startformula f''_{2}(x) = f_{2}''(x) = f^{''}_{2}(x) = f_{2}^{''}(x) % = f^{{}^{''}}_{2}(x) \stopformula As one may see the result of \type{f_{2}''(x) = } $f_{2}''(x)$ is not correct in \type{mkiv}.

The following positions of the superscripts \type{(3)} are correct both in \type{mkii} and in \type{mkiv}. \startformula f^{(3)}_{2}(x) = f_{2}^{(3)}(x) \stopformula

\stoptext %%% end derivatives-prime.tex

___________________________________________________________________________________ If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://context.aanhet.net archive : https://bitbucket.org/phg/context-mirror/commits/ wiki : http://contextgarden.net ___________________________________________________________________________________

--

----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | www.pragma-ade.nl | www.pragma-pod.nl ----------------------------------------------------------------- ___________________________________________________________________________________ If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://context.aanhet.net archive : https://bitbucket.org/phg/context-mirror/commits/ wiki : http://contextgarden.net ___________________________________________________________________________________

On Thu, May 18, 2017 at 9:50 AM, Mikael P. Sundqvist wrote:

Den 18 maj 2017 10:13 fm skrev "Otared Kavian" :

Hi Hans,

Thanks for your explanations. When using constructions such as $f’’_2(x)$, I will adapt my input in order to get what seems correct.

Best regards: OK

...

On 17 May 2017, at 17:17, Hans Hagen wrote:

On 5/17/2017 8:35 AM, Otared Kavian wrote:

...

Hi Hans,

Thanks for having fixed the behavior of the prime in the latest beta release (ConTeXt version 2017.05.15 21:48). However, as you may see in the two attached PDF files, the result is still incorrect when one types for example $f_2’’(x)$, which should give the same result as $f’’_2(x)$. The correct behaviour is that obtained with mkii. Indeed one can circumvent the behaviour of the latest beta release of mkiv by using tricks such as the ones in the example code below, but this should be avoided in my opinion.

well, it depends on how one sees this '' mess

the problem with primes is that in macro packages they are messy entities that should collapse or not (active char mess) and the more clever a macro packages becomes the harder it becomes to makes it robust and/or to fight against such heuristics

add to that the fact that the symbol itself is a funny positioned something already raised in a font or not i.e. sometimes in need of superscript and sometimes explicitly not

for unicode we need to turn two '' into a double and ''' into a and triple such

that said, in context these thingies are at some point intercepted and their current state is looked at: superscript? multipel in a row? one ebing the nucleus of another ... etc .. the current (working again) behavior is what we support and i don't look forward to more heuristics

...

The following PDF have been obtained with mkii and mkiv with the following example code. Best regards: OK

%%% begin derivatives-prime.tex \starttext

\startformula f''_{2}(x) = f_{2}''(x) = f^{''}_{2}(x) = f_{2}^{''}(x) % = f^{{}^{''}}_{2}(x) \stopformula As one may see the result of \type{f_{2}''(x) = } $f_{2}''(x)$ is not correct in \type{mkiv}.

The following positions of the superscripts \type{(3)} are correct both in \type{mkii} and in \type{mkiv}. \startformula f^{(3)}_{2}(x) = f_{2}^{(3)}(x) \stopformula

\stoptext %%% end derivatives-prime.tex

___________________________________________________________________________________ If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://context.aanhet.net archive : https://bitbucket.org/phg/context-mirror/commits/ wiki : http://contextgarden.net

___________________________________________________________________________________

--

----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | www.pragma-ade.nl | www.pragma-pod.nl -----------------------------------------------------------------

___________________________________________________________________________________ If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://context.aanhet.net archive : https://bitbucket.org/phg/context-mirror/commits/ wiki : http://contextgarden.net

___________________________________________________________________________________

___________________________________________________________________________________ If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://context.aanhet.net archive : https://bitbucket.org/phg/context-mirror/commits/ wiki : http://contextgarden.net ___________________________________________________________________________________

OK, very good, it seems that we have a working solution and one specific way to write it. I have absolutely nothing against that, but I would like us (me?) to document it, since I think future users might have use of it. Is the math page in the wiki the right place?

(It will take some time before I will be able to act. I'm currently travelling with close to no working connection.)

/Mikael

I'm sorry, but there seem to still be problems with primes. Look at the height of the primes in this document (ctx-listexample30.pdf). %%%% %\definefallbackfamily [lucidaopentype][mm][Xits Math][range={2660,2661,2662,2663,2664,2665,2666,2667}] \definefontfamily[lucidaopentype][rm][Lucida Bright OT] \definefontfamily[lucidaopentype][ss][Lucida Sans OT] \definefontfamily[lucidaopentype][tt][Lucida Sans Typewriter OT] \definefontfamily[lucidaopentype][mm][Lucida Bright Math OT] \setupbodyfont[lucidaopentype,10pt] \starttext Look at $f'$ and $f''$ \stoptext %%%% If I uncomment the \definefallbackfamily, the result is the one given in ctx-listexample30-var.pdf, look at the prime. It is still not correct (and the double prime is wrong as in the first example). /Mikael

On Sat, May 20, 2017 at 7:32 PM, Mikael P. Sundqvist wrote:

On Sat, May 20, 2017 at 3:33 PM, Mikael P. Sundqvist wrote:

...

On Thu, May 18, 2017 at 9:50 AM, Mikael P. Sundqvist wrote:

...

Den 18 maj 2017 10:13 fm skrev "Otared Kavian" :

Hi Hans,

Thanks for your explanations. When using constructions such as $f’’_2(x)$, I will adapt my input in order to get what seems correct.

Best regards: OK

...

On 17 May 2017, at 17:17, Hans Hagen wrote:

On 5/17/2017 8:35 AM, Otared Kavian wrote:

...

Hi Hans,

Thanks for having fixed the behavior of the prime in the latest beta release (ConTeXt version 2017.05.15 21:48). However, as you may see in the two attached PDF files, the result is still incorrect when one types for example $f_2’’(x)$, which should give the same result as $f’’_2(x)$. The correct behaviour is that obtained with mkii. Indeed one can circumvent the behaviour of the latest beta release of mkiv by using tricks such as the ones in the example code below, but this should be avoided in my opinion.

well, it depends on how one sees this '' mess

the problem with primes is that in macro packages they are messy entities that should collapse or not (active char mess) and the more clever a macro packages becomes the harder it becomes to makes it robust and/or to fight against such heuristics

add to that the fact that the symbol itself is a funny positioned something already raised in a font or not i.e. sometimes in need of superscript and sometimes explicitly not

for unicode we need to turn two '' into a double and ''' into a and triple such

that said, in context these thingies are at some point intercepted and their current state is looked at: superscript? multipel in a row? one ebing the nucleus of another ... etc .. the current (working again) behavior is what we support and i don't look forward to more heuristics

...

The following PDF have been obtained with mkii and mkiv with the following example code. Best regards: OK

%%% begin derivatives-prime.tex \starttext

\startformula f''_{2}(x) = f_{2}''(x) = f^{''}_{2}(x) = f_{2}^{''}(x) % = f^{{}^{''}}_{2}(x) \stopformula As one may see the result of \type{f_{2}''(x) = } $f_{2}''(x)$ is not correct in \type{mkiv}.

The following positions of the superscripts \type{(3)} are correct both in \type{mkii} and in \type{mkiv}. \startformula f^{(3)}_{2}(x) = f_{2}^{(3)}(x) \stopformula

\stoptext %%% end derivatives-prime.tex

___________________________________________________________________________________ If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://context.aanhet.net archive : https://bitbucket.org/phg/context-mirror/commits/ wiki : http://contextgarden.net

___________________________________________________________________________________

--

----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | www.pragma-ade.nl | www.pragma-pod.nl -----------------------------------------------------------------

___________________________________________________________________________________ If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://context.aanhet.net archive : https://bitbucket.org/phg/context-mirror/commits/ wiki : http://contextgarden.net

___________________________________________________________________________________

___________________________________________________________________________________ If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://context.aanhet.net archive : https://bitbucket.org/phg/context-mirror/commits/ wiki : http://contextgarden.net ___________________________________________________________________________________

OK, very good, it seems that we have a working solution and one specific way to write it. I have absolutely nothing against that, but I would like us (me?) to document it, since I think future users might have use of it. Is the math page in the wiki the right place?

(It will take some time before I will be able to act. I'm currently travelling with close to no working connection.)

/Mikael

I'm sorry, but there seem to still be problems with primes.

Look at the height of the primes in this document (ctx-listexample30.pdf).

%%%% %\definefallbackfamily [lucidaopentype][mm][Xits Math][range={2660,2661,2662,2663,2664,2665,2666,2667}] \definefontfamily[lucidaopentype][rm][Lucida Bright OT] \definefontfamily[lucidaopentype][ss][Lucida Sans OT] \definefontfamily[lucidaopentype][tt][Lucida Sans Typewriter OT] \definefontfamily[lucidaopentype][mm][Lucida Bright Math OT]

\setupbodyfont[lucidaopentype,10pt]

\starttext Look at $f'$ and $f''$ \stoptext %%%%

If I uncomment the \definefallbackfamily, the result is the one given in ctx-listexample30-var.pdf, look at the prime. It is still not correct (and the double prime is wrong as in the first example).

/Mikael

Hi,

I am now more confused. I tried in different fonts, and the result is attached. The difference in height of the prime and the double prime is confusing me. In some fonts the prime is above the double prime, in some the other way around (the worst is lucida, which perhaps is why I saw it in the first place). Does someone understand why this is so? Is it the fonts that are constructed like this?

/Mikael

PS The test file I used was the following, with the natural changes in fonts each time.

\definefontfamily[myface][rm][TeX Gyre Pagella] \definefontfamily[myface][mm][TeX Gyre Pagella Math] \setupbodyfont[myface,10pt] \starttext \startTEXpage[offset=2pt] $f'f''$ \stopTEXpage \stoptext

And just for reference, the document below gives the attached pdf, compiled with lualatex. I think the location of the primes there look better. Sorry for spamming about this issue, but I think it is important for all people writing math with ConTeXt. /Mikael \RequirePackage{luatex85} \documentclass[border=2pt]{standalone} \usepackage{unicode-math} \setmainfont{LucidaBrightOT} \setmathfont{LucidaBrightMathOT} \begin{document} $f'f''$ \end{document}

On Sun, May 21, 2017 at 10:29 PM, Otared Kavian wrote:

Hi Mikael,

I agree with what you are reporting: the current behavior of the \prime operator is not completely satisfying when using ConTeXt mkiv, while in mkii it is the most satisfying (in my opinion…).

The turn around consisting in using $f^{‘’}$ or $f^{\prime\prime}$, or even $f^{{}^{\prime\prime}}$, is not really satisfying either.

But I think (or I hope that) at some point Hans will fix it in mkiv.

Best regards: OK

...

On 21 May 2017, at 08:14, Mikael P. Sundqvist wrote: […]

And just for reference, the document below gives the attached pdf, compiled with lualatex. I think the location of the primes there look better. Sorry for spamming about this issue, but I think it is important for all people writing math with ConTeXt.

/Mikael

\RequirePackage{luatex85} \documentclass[border=2pt]{standalone}

\usepackage{unicode-math} \setmainfont{LucidaBrightOT} \setmathfont{LucidaBrightMathOT}

\begin{document} $f'f''$ \end{document} ___________________________________________________________________________________ If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://context.aanhet.net archive : https://bitbucket.org/phg/context-mirror/commits/ wiki : http://contextgarden.net ___________________________________________________________________________________

___________________________________________________________________________________ If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://context.aanhet.net archive : https://bitbucket.org/phg/context-mirror/commits/ wiki : http://contextgarden.net ___________________________________________________________________________________

Otared, thank you for confirmation. I also hope there will be a fix soon. In the meantime, I put yet another test file. The location of the primes depends on the next character (I use, at the moment current version: 2017.05.15 21:48) /Mikael \definefontfamily[myface][rm][Lucida Bright OT] \definefontfamily[myface][mm][Lucida Bright Math OT] \setupbodyfont[myface,10pt] \starttext \startTEXpage[offset=2pt] $f'(x)f''(x)f'f''$ \stopTEXpage \stoptext

Hi Mikael and Otared, I can confirm issues regarding primes, too. If you are using the STIX 2 fonts (http://www.stixfonts.org http://www.stixfonts.org/) and my example below you get a somewhat strange behaviour of primes that is pretty obvious. I think the issues should be fixed. By the way the issue seems to depend to some extend on the fonts. Using Lucida by \setupbodyfont[lucidaot] it is not that obvious but still present. Using Latin Modern it seems that the spacing between the primes changes between f'' and f^{''}. Cambria seems to show that behaviour (like Latin Modern), too, but less obvious. Best regards Mathias %% example \unprotect \definefontfeature [stixtwo-math-feature] [kern=yes, ss01=no, % calligraphic letters ss02=yes, % lowercase italic variants (e.g. open g) ss03=no, % hbar und lambdabar variants ss04=yes, % larger and lower primes, minute etc. ss05=no, % shorter arrows ss06=no, % narrower and shorter element and similar symbols ss07=no, % some smaller operators and emptyset variant ss08=yes, % upright integrals etc. ss09=no, % variant negated symbols ss10=yes, % variants of greater and lesser and similar symbols ss11=yes, % variant of negated set (inclusion etc.) symbols ss14=yes, % ss16=yes] % variants of circled operators \starttypescriptcollection[Stix2] \starttypescript [\s!serif] [Stix2] [\s!all] \definefontsynonym [Stix2Regular] [file:STIX2Text-Regular.otf] \definefontsynonym [Stix2Bold] [file:STIX2Text-Bold.otf] \definefontsynonym [Stix2Italic] [file:STIX2Text-Italic.otf] \definefontsynonym [Stix2BoldItalic] [file:STIX2Text-BoldItalic.otf] \definefontsynonym[\s!Serif] [Stix2Regular] [\s!features=\s!default] \definefontsynonym[\s!SerifBold] [Stix2Bold] [\s!features=\s!default] \definefontsynonym[\s!SerifItalic] [Stix2Italic] [\s!features=\s!default] \definefontsynonym[\s!SerifBoldItalic] [Stix2BoldItalic] [\s!features=\s!default] \stoptypescript \starttypescript [\s!math] [Stix2] \definefontsynonym [Stix2Math] [file:STIX2Math.otf] [features=math\mathsizesuffix] \definefontsynonym [\s!MathRoman] [Stix2Math] [\s!features={\s!math\mathsizesuffix,mathextra,stixtwo-math-feature},\s!designsize=\s!auto] \stoptypescript \starttypescript [Stix2] \definetypeface [\typescriptone] [\s!rm] [\s!serif] [Stix2] [\s!default] \definetypeface [\typescriptone] [\s!mm] [\s!math] [Stix2] [\s!default] \stoptypescript \stoptypescriptcollection \protect \setupbodyfont[Stix2] \starttext % Compare $f'$ to $(f')$ and to $f''$, $(f'')$, $f^{''}$ and $(f^{''})$. % \stoptext %% end of example

Am 22.05.2017 um 09:50 schrieb Otared Kavian :

Hi Mikael,

You are correct, the position of \prime depends on whether there is a next character or not after it, and it seems that this has an effect on its position for characters having a subscript as in $f_{1}''(x) \neq f''_{1}(x)$ $f'f''f'’'$ Best regards: OK

...

On 22 May 2017, at 08:51, Mikael P. Sundqvist wrote:

On Sun, May 21, 2017 at 10:29 PM, Otared Kavian wrote:

...

Hi Mikael,

I agree with what you are reporting: the current behavior of the \prime operator is not completely satisfying when using ConTeXt mkiv, while in mkii it is the most satisfying (in my opinion…).

The turn around consisting in using $f^{‘’}$ or $f^{\prime\prime}$, or even $f^{{}^{\prime\prime}}$, is not really satisfying either.

But I think (or I hope that) at some point Hans will fix it in mkiv.

Best regards: OK

...

On 21 May 2017, at 08:14, Mikael P. Sundqvist wrote: […]

And just for reference, the document below gives the attached pdf, compiled with lualatex. I think the location of the primes there look better. Sorry for spamming about this issue, but I think it is important for all people writing math with ConTeXt.

/Mikael

\RequirePackage{luatex85} \documentclass[border=2pt]{standalone}

\usepackage{unicode-math} \setmainfont{LucidaBrightOT} \setmathfont{LucidaBrightMathOT}

\begin{document} $f'f''$ \end{document} ___________________________________________________________________________________ If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://context.aanhet.net archive : https://bitbucket.org/phg/context-mirror/commits/ wiki : http://contextgarden.net ___________________________________________________________________________________

___________________________________________________________________________________ If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://context.aanhet.net archive : https://bitbucket.org/phg/context-mirror/commits/ wiki : http://contextgarden.net ___________________________________________________________________________________

Otared, thank you for confirmation. I also hope there will be a fix soon.

In the meantime, I put yet another test file. The location of the primes depends on the next character (I use, at the moment current version: 2017.05.15 21:48)

/Mikael

\definefontfamily[myface][rm][Lucida Bright OT] \definefontfamily[myface][mm][Lucida Bright Math OT] \setupbodyfont[myface,10pt] \starttext \startTEXpage[offset=2pt] $f'(x)f''(x)f'f''$ \stopTEXpage \stoptext ___________________________________________________________________________________ If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://context.aanhet.net archive : https://bitbucket.org/phg/context-mirror/commits/ wiki : http://contextgarden.net ___________________________________________________________________________________

___________________________________________________________________________________ If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://context.aanhet.net archive : https://bitbucket.org/phg/context-mirror/commits/ wiki : http://contextgarden.net ___________________________________________________________________________________

On Tue, May 23, 2017 at 5:02 PM, Mikael P. Sundqvist wrote:

On Tue, May 23, 2017 at 11:58 AM, Mathias Schickel wrote:

...

Hi Mikael and Otared,

I can confirm issues regarding primes, too. If you are using the STIX 2 fonts (http://www.stixfonts.org) and my example below you get a somewhat strange behaviour of primes that is pretty obvious. I think the issues should be fixed.

By the way the issue seems to depend to some extend on the fonts. Using Lucida by \setupbodyfont[lucidaot] it is not that obvious but still present. Using Latin Modern it seems that the spacing between the primes changes between f'' and f^{''}. Cambria seems to show that behaviour (like Latin Modern), too, but less obvious.

Best regards Mathias

%% example \unprotect

\definefontfeature [stixtwo-math-feature] [kern=yes, ss01=no, % calligraphic letters ss02=yes, % lowercase italic variants (e.g. open g) ss03=no, % hbar und lambdabar variants ss04=yes, % larger and lower primes, minute etc. ss05=no, % shorter arrows ss06=no, % narrower and shorter element and similar symbols ss07=no, % some smaller operators and emptyset variant ss08=yes, % upright integrals etc. ss09=no, % variant negated symbols ss10=yes, % variants of greater and lesser and similar symbols ss11=yes, % variant of negated set (inclusion etc.) symbols ss14=yes, % ss16=yes] % variants of circled operators

\starttypescriptcollection[Stix2] \starttypescript [\s!serif] [Stix2] [\s!all] \definefontsynonym [Stix2Regular] [file:STIX2Text-Regular.otf] \definefontsynonym [Stix2Bold] [file:STIX2Text-Bold.otf] \definefontsynonym [Stix2Italic] [file:STIX2Text-Italic.otf] \definefontsynonym [Stix2BoldItalic] [file:STIX2Text-BoldItalic.otf] \definefontsynonym[\s!Serif] [Stix2Regular] [\s!features=\s!default] \definefontsynonym[\s!SerifBold] [Stix2Bold] [\s!features=\s!default] \definefontsynonym[\s!SerifItalic] [Stix2Italic] [\s!features=\s!default] \definefontsynonym[\s!SerifBoldItalic] [Stix2BoldItalic] [\s!features=\s!default] \stoptypescript

\starttypescript [\s!math] [Stix2] \definefontsynonym [Stix2Math] [file:STIX2Math.otf] [features=math\mathsizesuffix] \definefontsynonym [\s!MathRoman] [Stix2Math] [\s!features={\s!math\mathsizesuffix,mathextra,stixtwo-math-feature},\s!designsize=\s!auto] \stoptypescript

\starttypescript [Stix2] \definetypeface [\typescriptone] [\s!rm] [\s!serif] [Stix2] [\s!default] \definetypeface [\typescriptone] [\s!mm] [\s!math] [Stix2] [\s!default] \stoptypescript

\stoptypescriptcollection

\protect

\setupbodyfont[Stix2]

\starttext % Compare $f'$ to $(f')$ and to $f''$, $(f'')$, $f^{''}$ and $(f^{''})$. % \stoptext %% end of example

Am 22.05.2017 um 09:50 schrieb Otared Kavian :

Hi Mikael,

You are correct, the position of \prime depends on whether there is a next character or not after it, and it seems that this has an effect on its position for characters having a subscript as in $f_{1}''(x) \neq f''_{1}(x)$ $f'f''f'’'$ Best regards: OK

On 22 May 2017, at 08:51, Mikael P. Sundqvist wrote:

On Sun, May 21, 2017 at 10:29 PM, Otared Kavian wrote:

Hi Mikael,

I agree with what you are reporting: the current behavior of the \prime operator is not completely satisfying when using ConTeXt mkiv, while in mkii it is the most satisfying (in my opinion…).

The turn around consisting in using $f^{‘’}$ or $f^{\prime\prime}$, or even $f^{{}^{\prime\prime}}$, is not really satisfying either.

But I think (or I hope that) at some point Hans will fix it in mkiv.

Best regards: OK

On 21 May 2017, at 08:14, Mikael P. Sundqvist wrote: […]

And just for reference, the document below gives the attached pdf, compiled with lualatex. I think the location of the primes there look better. Sorry for spamming about this issue, but I think it is important for all people writing math with ConTeXt.

/Mikael

\RequirePackage{luatex85} \documentclass[border=2pt]{standalone}

\usepackage{unicode-math} \setmainfont{LucidaBrightOT} \setmathfont{LucidaBrightMathOT}

\begin{document} $f'f''$ \end{document} ___________________________________________________________________________________ If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://context.aanhet.net archive : https://bitbucket.org/phg/context-mirror/commits/ wiki : http://contextgarden.net ___________________________________________________________________________________

___________________________________________________________________________________ If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://context.aanhet.net archive : https://bitbucket.org/phg/context-mirror/commits/ wiki : http://contextgarden.net ___________________________________________________________________________________

Otared, thank you for confirmation. I also hope there will be a fix soon.

In the meantime, I put yet another test file. The location of the primes depends on the next character (I use, at the moment current version: 2017.05.15 21:48)

/Mikael

\definefontfamily[myface][rm][Lucida Bright OT] \definefontfamily[myface][mm][Lucida Bright Math OT] \setupbodyfont[myface,10pt] \starttext \startTEXpage[offset=2pt] $f'(x)f''(x)f'f''$ \stopTEXpage \stoptext ___________________________________________________________________________________ If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://context.aanhet.net archive : https://bitbucket.org/phg/context-mirror/commits/ wiki : http://contextgarden.net ___________________________________________________________________________________

___________________________________________________________________________________ If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://context.aanhet.net archive : https://bitbucket.org/phg/context-mirror/commits/ wiki : http://contextgarden.net ___________________________________________________________________________________

___________________________________________________________________________________ If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://context.aanhet.net archive : https://bitbucket.org/phg/context-mirror/commits/ wiki : http://contextgarden.net ___________________________________________________________________________________

Thanks Otared and Mathias!

Hans, just to know, are you working on a fix? Is there anything more we can provide?

Best regards, Mikael

In current version: 2017.05.25 12:50 all problems I've noticed with primes seem to be gone! Now we can even write $u''_p$ or $u_p''$. Hans, your work and quick fix of this problem is very much appreciated! Best regards, Mikael

On Thu, May 25, 2017 at 8:05 PM, Mathias Schickel wrote:

...

Am 25.05.2017 um 18:59 schrieb Mikael P. Sundqvist :

On Tue, May 23, 2017 at 5:02 PM, Mikael P. Sundqvist wrote:

...

On Tue, May 23, 2017 at 11:58 AM, Mathias Schickel wrote:

...

Hi Mikael and Otared,

I can confirm issues regarding primes, too. If you are using the STIX 2 fonts (http://www.stixfonts.org) and my example below you get a somewhat strange behaviour of primes that is pretty obvious. I think the issues should be fixed.

By the way the issue seems to depend to some extend on the fonts. Using Lucida by \setupbodyfont[lucidaot] it is not that obvious but still present. Using Latin Modern it seems that the spacing between the primes changes between f'' and f^{''}. Cambria seems to show that behaviour (like Latin Modern), too, but less obvious.

Best regards Mathias

%% example \unprotect

\definefontfeature [stixtwo-math-feature] [kern=yes, ss01=no, % calligraphic letters ss02=yes, % lowercase italic variants (e.g. open g) ss03=no, % hbar und lambdabar variants ss04=yes, % larger and lower primes, minute etc. ss05=no, % shorter arrows ss06=no, % narrower and shorter element and similar symbols ss07=no, % some smaller operators and emptyset variant ss08=yes, % upright integrals etc. ss09=no, % variant negated symbols ss10=yes, % variants of greater and lesser and similar symbols ss11=yes, % variant of negated set (inclusion etc.) symbols ss14=yes, % ss16=yes] % variants of circled operators

\starttypescriptcollection[Stix2] \starttypescript [\s!serif] [Stix2] [\s!all] \definefontsynonym [Stix2Regular] [file:STIX2Text-Regular.otf] \definefontsynonym [Stix2Bold] [file:STIX2Text-Bold.otf] \definefontsynonym [Stix2Italic] [file:STIX2Text-Italic.otf] \definefontsynonym [Stix2BoldItalic] [file:STIX2Text-BoldItalic.otf] \definefontsynonym[\s!Serif] [Stix2Regular] [\s!features=\s!default] \definefontsynonym[\s!SerifBold] [Stix2Bold] [\s!features=\s!default] \definefontsynonym[\s!SerifItalic] [Stix2Italic] [\s!features=\s!default] \definefontsynonym[\s!SerifBoldItalic] [Stix2BoldItalic] [\s!features=\s!default] \stoptypescript

\starttypescript [\s!math] [Stix2] \definefontsynonym [Stix2Math] [file:STIX2Math.otf] [features=math\mathsizesuffix] \definefontsynonym [\s!MathRoman] [Stix2Math] [\s!features={\s!math\mathsizesuffix,mathextra,stixtwo-math-feature},\s!designsize=\s!auto] \stoptypescript

\starttypescript [Stix2] \definetypeface [\typescriptone] [\s!rm] [\s!serif] [Stix2] [\s!default] \definetypeface [\typescriptone] [\s!mm] [\s!math] [Stix2] [\s!default] \stoptypescript

\stoptypescriptcollection

\protect

\setupbodyfont[Stix2]

\starttext % Compare $f'$ to $(f')$ and to $f''$, $(f'')$, $f^{''}$ and $(f^{''})$. % \stoptext %% end of example

Am 22.05.2017 um 09:50 schrieb Otared Kavian :

Hi Mikael,

You are correct, the position of \prime depends on whether there is a next character or not after it, and it seems that this has an effect on its position for characters having a subscript as in $f_{1}''(x) \neq f''_{1}(x)$ $f'f''f'’'$ Best regards: OK

On 22 May 2017, at 08:51, Mikael P. Sundqvist wrote:

On Sun, May 21, 2017 at 10:29 PM, Otared Kavian wrote:

Hi Mikael,

I agree with what you are reporting: the current behavior of the \prime operator is not completely satisfying when using ConTeXt mkiv, while in mkii it is the most satisfying (in my opinion…).

The turn around consisting in using $f^{‘’}$ or $f^{\prime\prime}$, or even $f^{{}^{\prime\prime}}$, is not really satisfying either.

But I think (or I hope that) at some point Hans will fix it in mkiv.

Best regards: OK

On 21 May 2017, at 08:14, Mikael P. Sundqvist wrote: […]

And just for reference, the document below gives the attached pdf, compiled with lualatex. I think the location of the primes there look better. Sorry for spamming about this issue, but I think it is important for all people writing math with ConTeXt.

/Mikael

\RequirePackage{luatex85} \documentclass[border=2pt]{standalone}

\usepackage{unicode-math} \setmainfont{LucidaBrightOT} \setmathfont{LucidaBrightMathOT}

\begin{document} $f'f''$ \end{document} ___________________________________________________________________________________ If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://context.aanhet.net archive : https://bitbucket.org/phg/context-mirror/commits/ wiki : http://contextgarden.net ___________________________________________________________________________________

___________________________________________________________________________________ If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://context.aanhet.net archive : https://bitbucket.org/phg/context-mirror/commits/ wiki : http://contextgarden.net ___________________________________________________________________________________

Otared, thank you for confirmation. I also hope there will be a fix soon.

In the meantime, I put yet another test file. The location of the primes depends on the next character (I use, at the moment current version: 2017.05.15 21:48)

/Mikael

\definefontfamily[myface][rm][Lucida Bright OT] \definefontfamily[myface][mm][Lucida Bright Math OT] \setupbodyfont[myface,10pt] \starttext \startTEXpage[offset=2pt] $f'(x)f''(x)f'f''$ \stopTEXpage \stoptext ___________________________________________________________________________________ If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://context.aanhet.net archive : https://bitbucket.org/phg/context-mirror/commits/ wiki : http://contextgarden.net ___________________________________________________________________________________

___________________________________________________________________________________ If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://context.aanhet.net archive : https://bitbucket.org/phg/context-mirror/commits/ wiki : http://contextgarden.net ___________________________________________________________________________________

___________________________________________________________________________________ If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://context.aanhet.net archive : https://bitbucket.org/phg/context-mirror/commits/ wiki : http://contextgarden.net ___________________________________________________________________________________

Thanks Otared and Mathias!

Hans, just to know, are you working on a fix? Is there anything more we can provide?

Best regards, Mikael

In

current version: 2017.05.25 12:50

all problems I've noticed with primes seem to be gone! Now we can even write $u''_p$ or $u_p''$.

Hans, your work and quick fix of this problem is very much appreciated!

Best regards, Mikael

Hi Mikael and Hans,

I can confirm that, too: Everything is working very well in the current ConTeXt version. Thank you very much, Hans, for quietly solving all reported issues regarding primes!

Best regards Mathias

...

___________________________________________________________________________________ If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://context.aanhet.net archive : https://bitbucket.org/phg/context-mirror/commits/ wiki : http://contextgarden.net ___________________________________________________________________________________

___________________________________________________________________________________ If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://context.aanhet.net archive : https://bitbucket.org/phg/context-mirror/commits/ wiki : http://contextgarden.net ___________________________________________________________________________________

I hate to bring this up again, but look at the attached output of %%% %\setupbodyfont[lucidaot] \starttext \startformula f'(0)=2\frac{f'(0)}{2} \stopformula \stoptext %%% The prime in the fraction is in my opinion too low (it is more clear in the example with lucida). /Mikael

On 6/4/2017 11:06 PM, Otared Kavian wrote:

Hi Mikael,

Indeed your observation is correct, but using \over instead of \frac yields the correct behavior. Please try the following sample:

%%% begin derivatives-prime-4.tex \setupbodyfont[lucidaot]

\starttext

Compare the position of the prime sign in $f'/g$ in these two expressions: the first one using \type{{... \over ...}} \startformula \left({f \over g}\right)' = {f' \over g} - {f \over g^2}g' \stopformula (which yields a correct position of the prime for $f'$ in $f'/g$), and the second using \type{\frac{...}{...}} \startformula \left(\frac{f}{g} \right)' = \frac{f'}{g} - \frac{f}{g^2} g' \stopformula which yields a prime sign for $f'$ a little bit lower.

\stoptext %%% end derivatives-prime-4.tex The problem with primes is that it's a real mess, conceptually as well as in fonts.

- we need to deal with not only one prime but also with double, tripple and quadruple ones and they need to become proper unicode symbols too - fonts have one, two or three sizes, often inconsistent and with funny dimensions, onlyt the text size can be trusted - we cannot use prime as nuclues as it can be combined with a subscript - but it isn't designed as a script symbol and is already raised (somewhat) - so we have to deal with a good old tex / fonts / plain inheritance, unicode but also want to avoid active character and parsing mess etc. I tried several approaches and it's hard to come up with a general solution that suits all fonts. Now, the bad is that when we use a superscript with a smashed virtual characters we end up in script choice mess with primes being one step too small. So, I decided to follow a different route and extend luatex with a sizeless super/subscript feature so that we can handle such weird cases (there might be more). This also means that the primes get larger. I only use the text size prime and scale that down for the other sizes i.e. no stty for primes as they are such a mess) but we can live with that. The good news is that it gets better but the bad news is that you have to wait till we release luatex 1.0.5 (planned for the ctx meeting but we might do an intermediate garden release anyway). btw, with \frac you get cramped styles and these have different positioning. Such is life. Hans ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | www.pragma-ade.nl | www.pragma-pod.nl -----------------------------------------------------------------

On Wed, Jun 7, 2017 at 9:57 PM, Mikael P. Sundqvist wrote:

On Mon, Jun 5, 2017 at 11:00 PM, Hans Hagen wrote:

...

On 6/4/2017 11:06 PM, Otared Kavian wrote:

...

Hi Mikael,

Indeed your observation is correct, but using \over instead of \frac yields the correct behavior. Please try the following sample:

%%% begin derivatives-prime-4.tex \setupbodyfont[lucidaot]

\starttext

Compare the position of the prime sign in $f'/g$ in these two expressions: the first one using \type{{... \over ...}} \startformula \left({f \over g}\right)' = {f' \over g} - {f \over g^2}g' \stopformula (which yields a correct position of the prime for $f'$ in $f'/g$), and the second using \type{\frac{...}{...}} \startformula \left(\frac{f}{g} \right)' = \frac{f'}{g} - \frac{f}{g^2} g' \stopformula which yields a prime sign for $f'$ a little bit lower.

\stoptext %%% end derivatives-prime-4.tex

The problem with primes is that it's a real mess, conceptually as well as in fonts.

- we need to deal with not only one prime but also with double, tripple and quadruple ones and they need to become proper unicode symbols too

- fonts have one, two or three sizes, often inconsistent and with funny dimensions, onlyt the text size can be trusted

- we cannot use prime as nuclues as it can be combined with a subscript

- but it isn't designed as a script symbol and is already raised (somewhat)

- so we have to deal with a good old tex / fonts / plain inheritance, unicode but also want to avoid active character and parsing mess etc.

I tried several approaches and it's hard to come up with a general solution that suits all fonts.

Now, the bad is that when we use a superscript with a smashed virtual characters we end up in script choice mess with primes being one step too small.

So, I decided to follow a different route and extend luatex with a sizeless super/subscript feature so that we can handle such weird cases (there might be more).

This also means that the primes get larger. I only use the text size prime and scale that down for the other sizes i.e. no stty for primes as they are such a mess) but we can live with that.

The good news is that it gets better but the bad news is that you have to wait till we release luatex 1.0.5 (planned for the ctx meeting but we might do an intermediate garden release anyway).

btw, with \frac you get cramped styles and these have different positioning. Such is life.

Hans

----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | www.pragma-ade.nl | www.pragma-pod.nl ----------------------------------------------------------------- ___________________________________________________________________________________ If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://context.aanhet.net archive : https://bitbucket.org/phg/context-mirror/commits/ wiki : http://contextgarden.net ___________________________________________________________________________________

Again, thanks.

In 2017.06.06 13:22 something is still not correct.

The input file

\startTEXpage[offset=3pt] $f'(x)\hat{f}'(x)$ \stopTEXpage

gives the attached output. The prime of f'(x) looks wrong, the one of \hat{f}'(x) looks correct.

/Mikael

I was too quick. The example \setupbodyfont[lucidaot,10pt] \setupmathematics[stylealternative={calligraphic}]% to get the normal math lower case italic \startTEXpage[offset=3pt] $f'(x)\hat{f}'(x)$ \stopTEXpage gives bad result in both cases. The first one with wrong size and wrong vertical placement, the second one only with wrong vertical placement (that was not obvious to me for latin modern). /Mikael

On Wed, Jun 7, 2017 at 11:03 PM, Hans Hagen wrote:

On 6/7/2017 9:57 PM, Mikael P. Sundqvist wrote:

...

In 2017.06.06 13:22 something is still not correct.

i mentioned that you need a new engine (the garden has 1.0.5 binafies for windows but other platforms need to be generated)

Hans

----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | www.pragma-ade.nl | www.pragma-pod.nl ----------------------------------------------------------------- ___________________________________________________________________________________ If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://context.aanhet.net archive : https://bitbucket.org/phg/context-mirror/commits/ wiki : http://contextgarden.net ___________________________________________________________________________________

Ah, sorry, I did not read your mail carefully enough. I tried to compile 1.05 experimental from the svn, but I got some error, so I will probably wait for the 1.05 release (or hope a bit for the intermediate garden release, then). /Mikael PS: the ./build.sh (Checked out revision 6287) runs for a while, and then stops with an error like this (it is probably my fault): Makefile:368: recipe for target '../../../../source/texk/kpathsea/doc/kpathsea.info' failed make[2]: *** [../../../../source/texk/kpathsea/doc/kpathsea.info] Error 127 make[2]: Leaving directory '/opt/luatex-experimental/experimental/build/texk/kpathsea/doc' Makefile:1653: recipe for target 'all-recursive' failed make[1]: *** [all-recursive] Error 1 make[1]: Leaving directory '/opt/luatex-experimental/experimental/build/texk/kpathsea' Makefile:814: recipe for target 'all' failed make: *** [all] Error 2 ./build.sh: line 291: cd: texk/web2c: No such file or directory make: *** No rule to make target 'luatex'. Stop. strip: 'build/texk/web2c/luatex': No such file ls: cannot access 'build/texk/web2c/luatex': No such file or directory

On Thu, Jun 8, 2017 at 10:17 AM, luigi scarso wrote:

On Thu, Jun 8, 2017 at 9:52 AM, Mikael P. Sundqvist wrote:

...

PS: the ./build.sh (Checked out revision 6287) runs for a while, and then stops with an error like this (it is probably my fault):

Makefile:368: recipe for target '../../../../source/texk/kpathsea/doc/kpathsea.info' failed make[2]: *** [../../../../source/texk/kpathsea/doc/kpathsea.info] Error 127

weird, the file is there. Just to be sure, which repo are you using ?

-- luigi ___________________________________________________________________________________ If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://context.aanhet.net archive : https://bitbucket.org/phg/context-mirror/commits/ wiki : http://contextgarden.net ___________________________________________________________________________________

I ran svn checkout --username anonsvn --password anonsvn https://foundry.supelec.fr/svn/luatex/branches/experimental/ I will try to wipe and redo. Maybe something went wrong. /Mikael

On Thu, Jun 8, 2017 at 10:39 AM, Mikael P. Sundqvist wrote:

I get the same error with revision 6349 from https://serveur-svn.lri.fr/svn/modhel/luatex/branches/experimental/ , see below. But maybe we are OT on this list now, and should either go to luatex list or private? not OT, I am checking.

-- luigi

On Thu, 8 Jun 2017, Mikael P. Sundqvist wrote:

And indeed, in luatex 1.05 the primes look fine again. The prime in the fraction is too low, but I take Hans' "Such is life" and use \over for the rare cases where I have primes inside fractions.

I haven't followed the thread carefully, but there is nothing special in \frac. The likely cause is prime is misplaced in cramped style or similar and similar misplacement will happen whenever that style is used. I would be weary of using {... \over ...}. At the very least, use \Ustack{... \over ...}, else any code that uses \mathstyle (instead of \mathchoice) will fail. For an explaination, see Taco's answer here: https://tex.stackexchange.com/a/1261/323 Aditya