pretty versions of some units in math mode

[Alex-Jordan]

This sort of addresses #503, except by way of the newer contextUnits.pl. Each of the supported units from Units.pm that has a special unicode character will now print that way, if the unit is being used in math mode. Test with the problem below. It all looks fine in HTML. In hardcopy, we must be careful that the default font we use has glyphs for these characters. This is why two characters are used for degrees Fahrenheit, but a single character can be used for degrees Celsius. Also if someone switched away from xelatex over to pdflatex, these won't work. I'm unsure if we support not using xelatex.

Test with:

DOCUMENT();
loadMacros(qw(PGstandard.pl PGML.pl contextUnits.pl));

Context("Units")->withUnitsFor('length', 'temperature', 'angles', 'time', 'force', 'electricity', 'radiation');

$thickness = Compute('70 um');
$boils = Compute('100 degC');
$freezes = Compute('32 degF');
$zero = Compute('0 degK');
$pi = Compute('180 deg');
$us = Compute('us');
$angstrom = Compute('angstrom');
$bacterium = Compute('0.01 uN');
$charge = Compute('1 uC');
$farad = Compute('1 uF');
$ohm = Compute('ohm');
$kohm = Compute('kohm');
$Mohm = Compute('Mohm');
$uSv = Compute('uSv');

BEGIN_PGML
The thickness of a human hair is about [$thickness], or [`[$thickness]`].  
Water boils at [$boils], or [`[$boils]`].  
Water freezes at [$freezes], or [`[$freezes]`].  
Absolute zero is [$zero], or [`[$zero]`].  
[`\pi`] is [$pi], or [`[$pi]`].  
A very short unit of time is a [$us], or [`[$us]`].  
A very short distance is an [$angstrom], or [`[$angstrom]`].  
A bacterium has weight [$bacterium], or [`[$bacterium]`].  
A very small charge is a [$charge], or [`[$charge]`].  
A very small unit of capacitance is a [$farad], or [`[$farad]`].  
[$ohm], [$kohm], [$Mohm], or [`[$ohm]`], [`[$kohm]`], [`[$Mohm]`]  
[$uSv] or [`[$uSv]`]
END_PGML

ENDDOCUMENT();

[Alex-Jordan]

Now this also prints pretty versions of the same units outside of math mode.

[somiaj]

Small thing, but the space before the degree symbol seems a little out of place to me.

[Alex-Jordan]

Small thing, but the space before the degree symbol seems a little out of place to me.

That is not a change, in the sense that currently you have 180 deg. This changes the deg to °. But the space in between the quantity and the unit is a global separator. Something could be done wherever that separator is used, to omit that for special units where spaces are not appropriate.

[somiaj]

Looking at the mathjax speech text, some seems good, 0.01 micronewtons, and others not so much, micro s or cap A ring. Also I noticed that degrees F vs degrees C get different speech text. degree C just gets ℃ but degree F gets degrees cap F. Though this might be something MathJax should deal with better, just some things I noticed in my testing.

[somiaj]

Small thing, but the space before the degree symbol seems a little out of place to me.

That is not a change, in the sense that currently you have 180 deg. This changes the deg to °. But the space in between the quantity and the unit is a global separator. Something could be done wherever that separator is used, to omit that for special units where spaces are not appropriate.

I wasn't staying it was a change, but to me 180 deg seems proper, but 180 ° doesn't, I would always write it as 180°. I'm okay with it as is, just some things I've noticed and unsure if we want to address or not.

[Alex-Jordan]

Surely, even these speech strings are improvements over what currently would be there. But I think some of this is controlled by which speech rule engine the user chooses (or defaults to) in the MJ menu. Alex Jordan Mathematics Instructor Portland Community College

…

On Sat, Jun 13, 2026, 12:04 PM Jaimos Skriletz ***@***.***> wrote: *somiaj* left a comment (openwebwork/pg#1437) <https://urldefense.com/v3/__https://github.com/openwebwork/pg/pull/1437*issuecomment-4699492373__;Iw!!Ka_JY85zDv0FFw!nzzv3f2VIqj_OjXauNgh1k7yEMyKZrAfMxlGeLLeo1-nrw2PEKvdZIMElNFe-RbM-aqrKq-jLtxD__zn2Wt6QO4BWWA$> Looking at the mathjax speech text, some seems good, 0.01 micronewtons, and others not so much, micro s or cap A ring. Also I noticed that degrees F vs degrees C get different speech text. degree C just gets ℃ but degree F gets degrees cap F. Though this might be something MathJax should deal with better, just some things I noticed in my testing. — Reply to this email directly, view it on GitHub <https://urldefense.com/v3/__https://github.com/openwebwork/pg/pull/1437?email_source=notifications&email_token=ABEDOABPVA2WORWGGPJ4AET47WQVFA5CNFSNUABFM5UWIORPF5TWS5BNNB2WEL2JONZXKZKDN5WW2ZLOOQXTINRZHE2DSMRTG4Z2M4TFMFZW63VGMF2XI2DPOKSWK5TFNZ2KYZTPN52GK4S7MNWGSY3L*issuecomment-4699492373__;Iw!!Ka_JY85zDv0FFw!nzzv3f2VIqj_OjXauNgh1k7yEMyKZrAfMxlGeLLeo1-nrw2PEKvdZIMElNFe-RbM-aqrKq-jLtxD__zn2Wt6OuLUx88$>, or unsubscribe <https://urldefense.com/v3/__https://github.com/notifications/unsubscribe-auth/ABEDOAC5Q36V7WEY2M5OBUT47WQVFAVCNFSNUABDKJSXA33TNF2G64TZHMZTANRUGYYDCO2JONZXKZJ3GQ3DINZUHAZDENBTUF3AE__;!!Ka_JY85zDv0FFw!nzzv3f2VIqj_OjXauNgh1k7yEMyKZrAfMxlGeLLeo1-nrw2PEKvdZIMElNFe-RbM-aqrKq-jLtxD__zn2Wt6AVZU2p8$> . Triage notifications, keep track of coding agent tasks and review pull requests on the go with GitHub Mobile for iOS <https://urldefense.com/v3/__https://github.com/notifications/mobile/ios/ABEDOABKSPQ2FWQF24U5ZU347WQVFA5CNFSNUABFM5UWIORPF5TWS5BNNB2WEL2JONZXKZKDN5WW2ZLOOQXTINRZHE2DSMRTG4Z2M4TFMFZW63VGMF2XI2DPOKSWK5TFNZ2KUZTPN52GK4S7NFXXG__;!!Ka_JY85zDv0FFw!nzzv3f2VIqj_OjXauNgh1k7yEMyKZrAfMxlGeLLeo1-nrw2PEKvdZIMElNFe-RbM-aqrKq-jLtxD__zn2Wt6HnFSzTI$> and Android <https://urldefense.com/v3/__https://github.com/notifications/mobile/android/ABEDOAB264NOBWFR6VMH6MD47WQVFA5CNFSNUABFM5UWIORPF5TWS5BNNB2WEL2JONZXKZKDN5WW2ZLOOQXTINRZHE2DSMRTG4Z2M4TFMFZW63VGMF2XI2DPOKSWK5TFNZ2K4ZTPN52GK4S7MFXGI4TPNFSA__;!!Ka_JY85zDv0FFw!nzzv3f2VIqj_OjXauNgh1k7yEMyKZrAfMxlGeLLeo1-nrw2PEKvdZIMElNFe-RbM-aqrKq-jLtxD__zn2Wt6P6XYjvk$>. Download it today! You are receiving this because you authored the thread.Message ID: ***@***.***>

[Alex-Jordan]

I have a question. If the unit is initially defined like Compute('10 microns') should that display as "10 microns" or "10 µm"? There are a few other units with word versions that we may or may not want to respect.

[somiaj]

Agreed, I wasn't sure if there was a solution or something we could do to improve it, I was just pointing it out.

[somiaj]

I have a question. If the unit is initially defined like Compute('10 microns') should that display as "10 microns" or "10 µm"? There are a few other units with word versions that we may or may not want to respect.

Would having some flag that this can be configured be useful to authors? I sometimes like to write units as 10 seconds and other times as 10 s.

[Alex-Jordan]

I added a commit that essentially creates a list of unit symbols for which there will be no separator. Not sure I implemented it in the best way, but see what you think.

10 seconds will be 10 seconds. That is not the kind of thing happening here. What is happening here is that certain units like 10 angstrom will now be 10 Å. But at the same time, this does not change 10 angstroms or 10 Angstrom. Those still comes out as 10 angstroms or 10 Angstrom. My thinking was to only do this kind of prettying on the one lowercase singular-case unit where there is conventionally a special character. Not to just always print the unit the pretty way. The case of micron is especially worth thinking about, where it's actually a new word for micrometer.

[Alex-Jordan]

There are also weird things like Compute('10 ㏀'), which will be recognized. And will display using that one weird ㏀ character. But won't display at all in print because our default font has no glyph for that. If an author used ㏀, I would be inclined to let that stay as ㏀, not try to change it to the two-character kΩ. But this PR does change kohm to kΩ.

My point is, I am wary about just deciding at the distribution level, how to display all equivalent units (like how s is equivalent to seconds). I think it is safer to just do things like specifically display um as µm.

[Alex-Jordan]

I have a feeling I'm not doing one or both of these things here (pretty printing certain units, eliminating certain space separators) the right way. If @dpvc has a chance, it would be good to see what you think, as the designer of this context.

[somiaj]

@Alex-Jordan I'll leave it to you to decide what is best, I haven't used units yet, so don't have any experience, just pointing out things I saw. I didn't realize that this only affected a specific case of the units.

[drgrice1]

Small thing, but the space before the degree symbol seems a little out of place to me.

I have also been bothered by the space between the number and the degree symbol. It is something I have thought about fixing at some point.