sobota, 12 lutego 2022 o 11:44:49 UTC+1 Richard Harnden napisał(a):
> On 12/02/2022 01:19, fir wrote:
> > piątek, 11 lutego 2022 o 22:32:37 UTC+1 Ed Prochak napisał(a):
> >> On Thursday, February 10, 2022 at 4:30:12 PM UTC-5, Vir Campestris wrote:
> >>> On 09/02/2022 22:24, fir wrote:
> >>>> it still not fully clear to me but it seems i should use this asci one as a minus and find some unicode as a hypen only (it seems (though im not sure) that this asci one they name as hypoen-minus becouse its both hypen and minus...in my context it should be probably only minus)
> >>> While the characters exist I wouldn't recommend using them. Too much
> >>> chance of confusion.
> >>>
> >>> Andy
> >> I agree with Andy.
> >>
> >> In a C like language that supports a hyphen character and a minus character,
> >> there has to be a clear way for the HUMAN programmer to distinguish them.
> >>
> >> c=half-life
> >>
> >> Is that c assigned the value of the variable half minus the value of the variable life?
> >> or
> >> Is that c assigned the value of the variable half-life?
> >>
> >> fir, it is your language so your choice. But you are going to need to control the font as well to make this work.
> >> I suggest the benefit (allowing hyphenated variable names) is not worth the cost of new character assignments.
> >>
> >> Ed
> >
> > in case of zero and O it showed to eb no problem (to have two of them), imo same is with minus and hypen as to distinguishability of the two ...they are differend things... now the situation is just that there is no hypen in c and i think having this would be handy
> >
> > there is also a question what else to add to the basic set... i wouldnt wnat to add very many many new chars probably but i would add like 7 or so ...*maybe i even would and more but probably the first added would be of most importance (most commonly used)
> >
> > some propositions? (which one to add)
> >
> > probably there could be added:
> > - hypen
> > - middle dot, middle cross
> > - 4 arrows
> > - degree sign
> > - maybe some new parentheses like << >>, round one, and brackets , like < > but more vertical..
> > ...
> > in fact more can be gathered and i should maybe chose by need of them to the given language constructions maybe...in fact all unicode may be added as it maybe already added but he wuestion is what should have meaningfull meaning in form of what construction and this is to see
> None of these things exist on your keyboard, so you'll need to reinvent
> trigraphs. Or quadgraphs.
>
> Nobody will thank you for that.

you will just need an editor (and i even have half of my aditor written..hovevver i not coded last year becouse i feel i got limited time and wonted to focus on possibly more importtant thingz)

the more problem if i would go in unicode opennes nothing stops me fropm adding unlimited amount of chars and this is not neccesary good, the closed set of ascii of c seems interesting..but this affair with hypen is kinda proof few news are needed..what else i think is to take a bunch of new and "map" them on unused part of ascii (<32)..at least maping it in my head and in sources encode it in unicode..but maybe use this 7biut encoding for fun and on some old systems for fun

sobota, 12 lutego 2022 o 12:07:25 UTC+1 fir napisał(a):
> sobota, 12 lutego 2022 o 11:44:49 UTC+1 Richard Harnden napisał(a):
> > On 12/02/2022 01:19, fir wrote:
> > > piątek, 11 lutego 2022 o 22:32:37 UTC+1 Ed Prochak napisał(a):
> > >> On Thursday, February 10, 2022 at 4:30:12 PM UTC-5, Vir Campestris wrote:
> > >>> On 09/02/2022 22:24, fir wrote:
> > >>>> it still not fully clear to me but it seems i should use this asci one as a minus and find some unicode as a hypen only (it seems (though im not sure) that this asci one they name as hypoen-minus becouse its both hypen and minus...in my context it should be probably only minus)
> > >>> While the characters exist I wouldn't recommend using them. Too much
> > >>> chance of confusion.
> > >>>
> > >>> Andy
> > >> I agree with Andy.
> > >>
> > >> In a C like language that supports a hyphen character and a minus character,
> > >> there has to be a clear way for the HUMAN programmer to distinguish them.
> > >>
> > >> c=half-life
> > >>
> > >> Is that c assigned the value of the variable half minus the value of the variable life?
> > >> or
> > >> Is that c assigned the value of the variable half-life?
> > >>
> > >> fir, it is your language so your choice. But you are going to need to control the font as well to make this work.
> > >> I suggest the benefit (allowing hyphenated variable names) is not worth the cost of new character assignments.
> > >>
> > >> Ed
> > >
> > > in case of zero and O it showed to eb no problem (to have two of them), imo same is with minus and hypen as to distinguishability of the two ...they are differend things... now the situation is just that there is no hypen in c and i think having this would be handy
> > >
> > > there is also a question what else to add to the basic set... i wouldnt wnat to add very many many new chars probably but i would add like 7 or so ...*maybe i even would and more but probably the first added would be of most importance (most commonly used)
> > >
> > > some propositions? (which one to add)
> > >
> > > probably there could be added:
> > > - hypen
> > > - middle dot, middle cross
> > > - 4 arrows
> > > - degree sign
> > > - maybe some new parentheses like << >>, round one, and brackets , like < > but more vertical..
> > > ...
> > > in fact more can be gathered and i should maybe chose by need of them to the given language constructions maybe...in fact all unicode may be added as it maybe already added but he wuestion is what should have meaningfull meaning in form of what construction and this is to see
> > None of these things exist on your keyboard, so you'll need to reinvent
> > trigraphs. Or quadgraphs.
> >
> > Nobody will thank you for that.
> you will just need an editor (and i even have half of my aditor written..hovevver i not coded last year becouse i feel i got limited time and wonted to focus on possibly more importtant thingz)
>
> the more problem if i would go in unicode opennes nothing stops me fropm adding unlimited amount of chars and this is not neccesary good, the closed set of ascii of c seems interesting..but this affair with hypen is kinda proof few news are needed..what else i think is to take a bunch of new and "map" them on unused part of ascii (<32)..at least maping it in my head and in sources encode it in unicode..but maybe use this 7biut encoding for fun and on some old systems for fun

some know maybe what dangers should i be carefull if trying to add some graphical chars like this mentioned hypen middle dot, middle cros, 4 arrows and new parenthesis when i will try to put them in ascii <32 chars? and some suggestion what to map on what?
in wiki you get for example old asci table from 1972
https://pl.wikipedia.org/wiki/ASCII

sobota, 12 lutego 2022 o 12:49:03 UTC+1 fir napisał(a):
> >
> > the more problem if i would go in unicode opennes nothing stops me fropm adding unlimited amount of chars and this is not neccesary good, the closed set of ascii of c seems interesting..but this affair with hypen is kinda proof few news are needed..what else i think is to take a bunch of new and "map" them on unused part of ascii (<32)..at least maping it in my head and in sources encode it in unicode..but maybe use this 7biut encoding for fun and on some old systems for fun
> some know maybe what dangers should i be carefull if trying to add some graphical chars like this mentioned hypen middle dot, middle cros, 4 arrows and new parenthesis when i will try to put them in ascii <32 chars? and some suggestion what to map on what?
> in wiki you get for example old asci table from 1972
> https://pl.wikipedia.org/wiki/ASCII

i kinda know i shouldnt use 0, 10, 13 (i mean i may even add graphical representation to them but they are 'used') but can i use anything else?

sobota, 12 lutego 2022 o 13:02:08 UTC+1 fir napisał(a):
> sobota, 12 lutego 2022 o 12:49:03 UTC+1 fir napisał(a):
> > >
> > > the more problem if i would go in unicode opennes nothing stops me fropm adding unlimited amount of chars and this is not neccesary good, the closed set of ascii of c seems interesting..but this affair with hypen is kinda proof few news are needed..what else i think is to take a bunch of new and "map" them on unused part of ascii (<32)..at least maping it in my head and in sources encode it in unicode..but maybe use this 7biut encoding for fun and on some old systems for fun
> > some know maybe what dangers should i be carefull if trying to add some graphical chars like this mentioned hypen middle dot, middle cros, 4 arrows and new parenthesis when i will try to put them in ascii <32 chars? and some suggestion what to map on what?
> > in wiki you get for example old asci table from 1972
> > https://pl.wikipedia.org/wiki/ASCII
> i kinda know i shouldnt use 0, 10, 13 (i mean i may even add graphical representation to them but they are 'used') but can i use anything else?

as to this set i would add some candidates i git are made by need to add pairs to what is already present
i think for example
1) adding mid dot
2) adding up dot (someting like degree sign but just dot)
3) adding mid x (cross product like)
4) ading small mid "Y" three-arms-cross (shuriken)
5/6) possibly adding rectangle (maybe even two, big and small)
7/8 adding << and >>
9/10 ) adding new parenthesis, maybe something like narrow <> or turtle ()
11) adding low "
12) adding upperscore (upper _)
13) adding reversed ^ (thosle last two looks boring but probably better have it)
14/15) having left and right arrow
16/17) and maybe also up and down (not sure)
18( maybe having reversed ? (as in spanish)
19) and reversed !
20) maybe also low double dot ..
21) upper double dot (?)
( 0) forgot mentioned hypen)
most of it seems sane for me

Den 08/02/2022 kl. 11.09 skrev fir:
> i realized totay morning that 'unifying' the text based "-" (so called dash) and math based "-" (minus sign) is big mistake in ascii ...this is something a bit like using "O" sign for both 0 and O
>
> i will maybe not even wider explain why its big mistake as i think its somewhat understood (in c for example it removes a dash sign from normal usage when
> you really would like it)
>
> the question is is it somewhat possible to mend it some way? by mending it i mean introducing usage of dashes in c (by introducing i mean if i would write
> compiler but want base it on ascii can i introduce dashes (like allow them in
> identifiers or some syntax constructions) without reaching for unicode?
> Is there something maybe in thoise ">127" or "<32" fields which i could use for dash?

Actually - there are four types of dashes (se top right side):
https://en.wikipedia.org/wiki/Dash

Three types of hyphens:
https://en.wikipedia.org/wiki/Hyphen

Two types of minus signs (unary and binary):
https://en.wikipedia.org/wiki/Plus_and_minus_signs#Minus_sign

( https://en.wikipedia.org/wiki/Hyphen-minus )