Zulip Chat Archive

Stream: iris-lean

Topic: Unicode symbol for `-*`


Shreyas Srinivas (Jul 17 2025 at 15:32):

Related. We should have a unicode symbol for the wand. The current ascii rendering stands out distinctly in lean code.

Mario Carneiro (Jul 17 2025 at 15:34):

is there a unicode wand symbol?

Notification Bot (Jul 17 2025 at 15:35):

2 messages were moved here from #iris-lean > Minor UX quibble by Mario Carneiro.

Mario Carneiro (Jul 17 2025 at 15:36):

the wand symbol is -∗ which is not actually ascii (it is -\ast)

suhr (Jul 17 2025 at 15:38):

P ∗ Q 🪄 Q ∗ R

Shreyas Srinivas (Jul 17 2025 at 15:43):

Fair point. This is the vscode symbol list for iris and I can't see one for wand in here : https://gitlab.mpi-sws.org/iris/iris/-/blob/master/docs/vscode

Mario Carneiro (Jul 17 2025 at 15:43):

I don't think it looks that bad. I find <pers> and other modalities to be rather more distracting:
image.png

Shreyas Srinivas (Jul 17 2025 at 15:45):

Right. I thought was persistent. But apparently is persistent affine

Shreyas Srinivas (Jul 17 2025 at 15:51):

Digging into the syntax and notation setup, I wonder if we can't just use notation all the way

Shreyas Srinivas (Jul 17 2025 at 15:51):

We never leave the syntax category of terms anyway

Mario Carneiro (Jul 17 2025 at 15:52):

that's not true, iris terms are in their own syntax category

Mario Carneiro (Jul 17 2025 at 15:53):

No we can't use notation because iris terms are given meaning only inside iprop(_)

Mario Carneiro (Jul 17 2025 at 15:54):

the issue is that et al need to be interpreted differently inside iprop(_)

Shreyas Srinivas (Jul 17 2025 at 15:55):

Ah I just got to that point. It feels like a nice trick to never have to write a separate syntax category

Mario Carneiro (Jul 17 2025 at 15:57):

although now I'm wondering whether making the syntax into macros or notations will fix the issue where go to definition never works properly on iprop expressions

Mario Carneiro (Jul 17 2025 at 15:59):

hm, no luck


Last updated: Dec 20 2025 at 21:32 UTC