[chandas] Add basic implementation

This commit adds a basic implementation of `vidyut-chandas` along with
some starter documentation. `vidyut-chandas` is pre-alpha code that I'm
shipping as a scaffold for future work.

Currently, `vidyut-chandas` supports a variety of vrtta meters, but it
has no support for jAti meters, which is the next major feature to
implement. Our documentation is also lacking, but it's a promising
start.

This commit was prepared primarily by me but with major contributions
from Pradyumna Malladi (@MSSRPRAD), who created a [first draft][1] of
the code based on a design document we collaborated on [here][2].

My contributions were:

- some cleanup to the input data format to use simple TSVs instead of
  JSON files with serde. We currently have a serde dependency to support
  our frontend, but this is not a core part of the crate and can be
  removed in a future refactor.

- some simplification to the API, particularly the use of a single
  `name` instead of `names` for each meter and the avoidance of some
  intermediate structs during serde

- the `classify` method and its matching logic

- some tweaks to our logic for calculating meter ganas

- most of the documentation, including the README, docstrings, and
  sample code

- most unit tests, especially for classifying different meters

- our WebAssembly bindings and frontend

Pradyuma's contributions were:

- a partial first draft of an end-to-end solution, including the basic
  data model for meters and aksharas

- data management, including sourcing our original meters file and
  writing the logic to read this file into our data structures

- a first draft of logic to calculate meter ganas

- a partial port of some of Vidyut's utility functions for testing
  Sanskrit sounds

- suggestions to make the returned matching object richer in order to
  support future ranking use-cases

All errors in recollection are my own.

[1]: #82
[2]: https://paper.dropbox.com/doc/Metrical-recognizer-API-v0--CCgFzji5rgBZYEMJkWAt_j5bAg-pZGhBvlQQBFcPx7SeWSE9

Co-authored-by: Pradyumna Malladi <[email protected]>

Cargo.toml

@@ -1,6 +1,7 @@
 [workspace]
 
 members = [
+    "vidyut-chandas",
     "vidyut-cheda",
     "vidyut-kosha",
     "vidyut-lipi",
@@ -21,6 +22,7 @@ license = "MIT"
 edition = "2021"
 
 [dependencies]
+vidyut-chandas = { path = "./vidyut-chandas" }
 vidyut-cheda = { path = "./vidyut-cheda" }
 vidyut-kosha = { path = "./vidyut-kosha" }
 vidyut-lipi = { path = "./vidyut-lipi" }

README.md

@@ -108,6 +108,13 @@ independently depending on your use case.
 In Rust, components of this kind are called *crates*.
 
 
+### [`vidyut-chandas`][vidyut-chandas]
+
+`vidyut-chandas` is an experimental classifier for Sanskrit meters.
+
+For details, see the [vidyut-chandas README][vidyut-chandas].
+
+
 ### [`vidyut-cheda`][vidyut-cheda]
 
 `vidyut-cheda` segments Sanskrit expressions into words then annotates those
@@ -147,6 +154,7 @@ between words. It is fast, simple, and appropriate for most use cases.
 For details, see the [vidyut-sandhi README][vidyut-sandhi].
 
 
+[vidyut-chandas]: vidyut-chandas/README.md
 [vidyut-cheda]: vidyut-cheda/README.md
 [vidyut-kosha]: vidyut-kosha/README.md
 [vidyut-prakriya]: vidyut-prakriya/README.md

vidyut-chandas/.gitignore

@@ -0,0 +1,2 @@
+www/static/wasm
+www/static/data

vidyut-chandas/Cargo.toml

@@ -0,0 +1,22 @@
+[package]
+name = "vidyut-chandas"
+version = "0.1.0"
+authors = ["Arun Prasad <[email protected]>"]
+description = "A Sanskrit metrical classifier"
+homepage = "https://github.com/ambuda-org/vidyut"
+repository = "https://github.com/ambuda-org/vidyut"
+categories = ["text-processing"]
+keywords = ["sanskrit"]
+license = "MIT"
+edition = "2021"
+
+[dependencies]
+console_error_panic_hook = "0.1.7"
+lazy_static = "1.4.0"
+serde = { version = "1.0.150", features = ["derive"] }
+serde-wasm-bindgen = "0.4"
+serde_derive = "1.0.193"
+wasm-bindgen = "0.2"
+
+[lib]
+crate-type = ["cdylib", "rlib"]

vidyut-chandas/Makefile

@@ -0,0 +1,2 @@
+debugger:
+	./scripts/run-debugger.sh

vidyut-chandas/README.md

@@ -0,0 +1,59 @@
+<div align="center">
+<h1><code>vidyut-chandas</code></h1>
+<p><i>A Sanskrit metrical classifier</i></p>
+</div>
+
+`vidyut-chandas` is an experimental classifier for Sanskrit meters.
+
+This [crate][crate] is under active development as part of the [Ambuda][ambuda]
+project. If you enjoy our work and wish to contribute to it, we encourage you
+to [join our Discord server][discord], where you can meet other Sanskrit
+programmers and enthusiasts.
+
+An online demo is available [here][demo].
+
+[crate]: https://doc.rust-lang.org/book/ch07-01-packages-and-crates.html
+[ambuda]: https://ambuda.org
+[discord]: https://discord.gg/7rGdTyWY7Z
+[demo]: https://ambuda-org.github.io/vidyut-lipi/
+
+- [Overview](#overview)
+- [Usage](#usage)
+
+
+Overview
+--------
+
+Sanskrit poetry uses a variety of *meters*, which specify the syllable patterns
+that a verse must follow. `vidyut-chandas` aims to provide a simple API for
+identifying the meter that a given verse uses.
+
+`vidyut-chandas` is experimental code and follows in the footsteps of great
+projects like [sanskritmetres][sm] and [Skrutable][skrutable].
+
+[sm]: https://github.com/shreevatsa/sanskrit
+[skrutable]: https://github.com/tylergneill/skrutable
+
+
+Usage
+-----
+
+*(This API is unstable.)*
+
+We recommend using `vidyut-chandas` through our `Chandas` API:
+
+```rust
+use vidyut_chandas::{Chandas, MatchType, Vrtta};
+
+let vrttas: Vec<Vrtta> = vec![
+    "vasantatilakA\tvrtta\tGGLGLLLGLLGLGG".try_into().unwrap(),
+    "mandAkrAntA\tvrtta\tGGGGLLLLLGGLGGLGG".try_into().unwrap(),
+    "puzpitAgrA\tvrtta\tLLLLLLGLGLGG/LLLLGLLGLGLGG".try_into().unwrap(),
+    "udgatA\tvrtta\tLLGLGLLLGL/LLLLLGLGLG/GLLLLLLGLLG/LLGLGLLLGLGLG".try_into().unwrap()
+];
+let chandas = Chandas::new(vrttas);
+
+let result = chandas.classify("mAtaH samastajagatAM maDukEwaBAreH");
+assert_eq!(result.vrtta().as_ref().unwrap().name(), "vasantatilakA");
+assert_eq!(result.match_type(), MatchType::Pada);
+```

vidyut-chandas/data/README.md

@@ -0,0 +1,21 @@
+Data
+====
+
+We have designed `vidyut-chandas` so that it can run without any side data. But
+in practice, it is useful to have a list of meters available already. For this
+reason, the `vidyut-chandas` crate includes `meters.tsv`, which you can use to
+get started.
+
+
+Creating the data file
+----------------------
+
+`meters.tsv` was sourced from the [Sanskrit metres][s-m] project by Shreevatsa
+Rajagopalan, which itself sources its data from a transcription of the
+*Vṛttaratnākara* prepared by Dr. Dhaval Patel. Our version of this data
+contains only *vṛtta* meters and does not currently support *jāti* meters.
+
+We extracted this data using `extract_meter_data.py`, which you can find in the
+`scripts` directory of this crate.
+
+[s-m]: https://github.com/shreevatsa/sanskrit