Open DoktorJ opened 2 years ago
You can override built in sequences just fine. e.g. You could define
I'm curious where and how the stock sequences are defined -- if I git clone the repo, where do I find the stock definitions in the source?
What @dkeenan7 offered would be a perfect solution – longer sequences when defined later should override shorter sequences, and the early-exit path to access the default (or earlier-defined user) sequences would be ideal. Just like I use –
versus —
(also, I'm just noticing in the editor here that an em-dash is 38% longer than it should be, at least in this particular font, but that's not a Wincompose issue 🤣)
I'm glad you like it @DoktorJ. I just hope it is easy for our hard-working Sam to implement.
I assume you wrote above: "Just like I use
Also, I think you can change the title of this issue, e.g. to "Unable to override built-in sequence with longer sequence", if you want to, by editing your original post.
Thanks for catching that @dkeenan7 , I didn't realize git had eaten the \
I am trying to define custom sequences for vulgar fractions:
The ⅔ and ¾ fractions work fine, but the first three don't, because there's a built-in:
It's not in any of the res files, and even disabling the other sequence options doesn't get rid of it, making it entirely impossible for me to define these sequences. I can't see myself ever using the fraction numerator character. To get 1/4,1/3,1/2 I'd have to use that and then the sequence for the relevant subscript denominator e.g.
⅟₂ =
<MK><1></><MK><_><2>
⅟₄ =<MK><1></><MK><_><4>
Which just really seems unnecessary to me. It consumes an extra character space on limited-length platforms, and is more prone to editorial munging (like if I'm typing
⅟₂tsp
in a recipe but decide I want a space, accidentally hit backspace 4 times instead of 3 and now I'm left with⅟ tsp
which is a lot less visually distinguishable as an error thantsp
). Also, typographically, ⅟₂ and ½ may look different in different fonts (here, the subscript 2 is set below the baseline as a subscript should be, but the vulgar fraction sits on the baseline), and if a font has the specific vulgar fraction defined I'd much rather stick with that.I'm sure there are other folks who want to define sequences that are "blocked" by built-ins. What would be really lovely is to have a priority list that a user can drag around to specify which set of sequences should be processed first – notably, this needs to include partial sequences, so that my .XCompose definition of
<Multi_key> <1> </> <4>
can override the default Fraction Numerator One sequence. I've seen other bugs submitted where users have problems with a partial sequence (#464 , #323 , etc), and think that being able to prioritize which list takes precedence as well as whether partials should be "held" to check for longer sequences would be nice.Yes I know there are sequences for the vulgar fractions, but my OCD brain always tries to put the slash and I have to go back and correct myself. I'd much rather be able to define it the way I want to use it, and I'm sure others have run into similar conflicts, possibly with other sequences, hence this bug.