Brassica SCA [v0.2.0]

[WarpedWartWars]

What's a parser combinator?

The idea is that you can use combinators (i.e., functions) to build up bigger parsers from smaller parsers: thus, for instance, if you have parsers ‘a’ or ‘b’, you can get things like ‘a then b’, ‘either a or b’, ‘a and b in any order’, ‘a inside b’ and so forth. Brassica uses parser combinators (via the megaparsec library) to parse both sound changes and paradigms.

Interesting!

[Travis B.]

I would gladly learn Haskell, except I am not downloading a >55 MB compiler for it.

Those 55 MB are worth it, seriously. I'm surprised that that much data is an issue in this day and age, unless one's sole Internet connection is a cell network hotspot.

Except my sole Internet connection is a cell network hotspot.

Could you at least find someplace with free public WiFi such as a coffee shop to download GHC at?

[Travis B.]

What's a parser combinator?

The idea is that you can use combinators (i.e., functions) to build up bigger parsers from smaller parsers: thus, for instance, if you have parsers ‘a’ or ‘b’, you can get things like ‘a then b’, ‘either a or b’, ‘a and b in any order’, ‘a inside b’ and so forth. Brassica uses parser combinators (via the megaparsec library) to parse both sound changes and paradigms.

Interesting!

Parser combinators are one of the things that are practically unique to Haskell, that simply do not translate to other programming languages (save possibly Idris and like).

[WarpedWartWars]

Except my sole Internet connection is a cell network hotspot.

Could you at least find someplace with free public WiFi such as a coffee shop to download GHC at?

No coffee shops here, but there is the library and a few grocery stores. I just don't know if my mom'll let me bring the laptop.

[WarpedWartWars]

Interesting!

Parser combinators are one of the things that are practically unique to Haskell, that simply do not translate to other programming languages (save possibly Idris and like).

Huh.

[WarpedWartWars]

Initial numbers in the paradigm builder don't seem to work. I copied this:

Code: Select all

() 2.en
1 wim () soo aa
-2 zhaa woo yaa
-1 zh w y

from the documentation into it, then ran it on the root kood (also from the docs), and I got this:

Code: Select all

2:7:
  |
2 | 1 wim () soo aa
  |       ^
unexpected '('
expecting '>' or alphanumeric character

Edit: 99b20f4 might be the problem.

[WarpedWartWars]

especially since I'm not going to download the 32.7 megabyte installer.

Apparently I ended up downloading it.

[bradrn]

The idea is that you can use combinators (i.e., functions) to build up bigger parsers from smaller parsers: thus, for instance, if you have parsers ‘a’ or ‘b’, you can get things like ‘a then b’, ‘either a or b’, ‘a and b in any order’, ‘a inside b’ and so forth. Brassica uses parser combinators (via the megaparsec library) to parse both sound changes and paradigms.

Interesting!

Parser combinators are one of the things that are practically unique to Haskell, that simply do not translate to other programming languages (save possibly Idris and like).

I’m not sure this is quite correct… any typed functional language can do them (so including OCaml, F# etc.), and I’m pretty sure most untyped FP languages can too. I even made a parser combinator library in C# once — it’s doable, though a little unergonomic. (Thankfully, C# has do-notation, at least.)

Initial numbers in the paradigm builder don't seem to work. I copied this:

Code: Select all

() 2.en
1 wim () soo aa
-2 zhaa woo yaa
-1 zh w y

from the documentation into it, then ran it on the root kood (also from the docs), and I got this:

Code: Select all

2:7:
  |
2 | 1 wim () soo aa
  |       ^
unexpected '('
expecting '>' or alphanumeric character

Edit: 99b20f4 might be the problem.

Oh, interesting… thanks for finding that! I’ll look at it once I get home.

[WarpedWartWars]

[...]

Oh, interesting… thanks for finding that! I’ll look at it once I get home.

Let me see...nope, it doesn't work if I remove the parens:

Code: Select all

2:13:
  |
2 | 1 wim soo aa
  |             ^
unexpected newline
expecting '>' or alphanumeric character

[bradrn]

[...]

Oh, interesting… thanks for finding that! I’ll look at it once I get home.

Let me see...nope, it doesn't work if I remove the parens:

Code: Select all

2:13:
  |
2 | 1 wim soo aa
  |             ^
unexpected newline
expecting '>' or alphanumeric character

Oh, now that is bad. That should work perfectly. (As should the other, of course, but this case is simpler and hence more worrying.)

Out of curiosity, have you found any similar inputs which do work?

[WarpedWartWars]

Oh, now that is bad. That should work perfectly. (As should the other, of course, but this case is simpler and hence more worrying.)

Out of curiosity, have you found any similar inputs which do work?

Prefixing all the affixes with <number>. works.

[bradrn]

The bug should now be fixed (98d9f8a67575a6f3a526c33b168a4b999de730a3). I’ll make a new release after I update the documentation and fix some more bugs I’m aware of.

[WarpedWartWars]

How do I make a sound change to remove spaces?

Edit: And how do I match one of any letter without needing to make a category for all letters used?

[bradrn]

How do I make a sound change to remove spaces?

Um… something is very wrong if you need a sound change to remove spaces, because all the spaces should be stripped out in the first place. Why do you need this?

Edit: And how do I match one of any letter without needing to make a category for all letters used?

You can’t, at the moment. (I should probably add this.) But then again, why do you want to do this in the first place? ‘All letters’ would mean the sound change applies in exactly the same way to both consonants and vowels, which seems unusual enough to me that it deserves being spelt out.

[dhok]

I'm popping by after an absence of lord knows how long to say that I'm overjoyed that this exists. Is there a reason that whitespaces seem to break the desktop version? It took me quite a while to work out that that was going on given the readme.

E.g., if the Categories box has

K=kg
Č=čĵ


and the Sound Changes box has K/Č/_[ie], then things run as expected: *kito gives čito. However, if I add whitespaces as K / Č / _[ie] (or indeed add whitespace anywhere in the rule), it breaks and the sound change doesn't apply.

One of the things I really appreciated about GoatFace's Haedus SCA was the ability to reserve strings like <kʷ> or <p'> so that they'd act as single tokens. Haedus also used whitespace to demarcate tokens, which I found really intuitive and useful. You might want to consider adding a box for "Reserved sequences" where the user can add digraphs and trigraphs like <kʰ>, then use whitespace syntax in the categories so that a category is written K = k g kʰ or thereabouts. Lack of whitespace can make things crowded. (I also liked the use of <0> for null inputs and outputs in Haedus, although the readme used a font that made it look like the symbol for bilabial click. This would also fit better with the option to use arrow syntax, which I think is also broken on the desktop--h → 0 / _C is intuitive while h→/_C just looks weird...)

This would probably make rules longer than they are now, so maybe instead of portrait-aligned boxes, you could perhaps go for landscape-aligned ones. This could look something like this (ugly Paint diagram incoming):



One further comment: semicolon comments don't seem to completely comment out their line. If I try to comment out a rule, things get screwed up. (Maybe use <%> to mark comments and give the whole line its own color? Semicolons are weird and feel unintuitive.)

But this is all very much nitpicking and I'm really enthusiastic about this. It's the sort of software package the community's been waiting for for about two decades.

[bradrn]

I'm popping by after an absence of lord knows how long to say that I'm overjoyed that this exists. Is there a reason that whitespaces seem to break the desktop version? It took me quite a while to work out that that was going on given the readme.

E.g., if the Categories box has

K=kg
Č=čĵ


and the Sound Changes box has K/Č/_[ie], then things run as expected: *kito gives čito. However, if I add whitespaces as K / Č / _[ie] (or indeed add whitespace anywhere in the rule), it breaks and the sound change doesn't apply.

Wait… what SCA are you using? It can’t be Brassica, because it doesn’t have separate boxes for categories and sound changes; instead, Brassica has a single box for both. And it should support multigraphs just fine too, without even needing to reserve strings. Here’s the correct link: http://bradrn.com/brassica/index.html

[dhok]

I'm popping by after an absence of lord knows how long to say that I'm overjoyed that this exists. Is there a reason that whitespaces seem to break the desktop version? It took me quite a while to work out that that was going on given the readme.

E.g., if the Categories box has

K=kg
Č=čĵ


and the Sound Changes box has K/Č/_[ie], then things run as expected: *kito gives čito. However, if I add whitespaces as K / Č / _[ie] (or indeed add whitespace anywhere in the rule), it breaks and the sound change doesn't apply.

Wait… what SCA are you using? It can’t be Brassica, because it doesn’t have separate boxes for categories and sound changes; instead, Brassica has a single box for both. And it should support multigraphs just fine too, without even needing to reserve strings. Here’s the correct link: http://bradrn.com/brassica/index.html

The readme says "exSCA version 2.1.1"...ah, wait,hmm. You are quite sure I want the online version that says version 0.0.2, not the desktop version linked therein?

[bradrn]

The readme says "exSCA version 2.1.1"...ah, wait,hmm. You are quite sure I want the online version that says version 0.0.2, not the desktop version linked therein?

No, I’m quite sure you don’t want exSCA at all, because that’s a different SCA. In fact, it’s my old SCA which is now long-deprecated in favour of Brassica. How come you ended up there?

[Rounin Ryuuji]

Can the sound change applier be made to apply the steps in a specific order, rather than apparently all at once?

[bradrn]

Can the sound change applier be made to apply the steps in a specific order, rather than apparently all at once?

It does apply them in order. Could you supply more details as to your specific problem please?