CSV reading

Cargo.lock

@@ -83,6 +83,18 @@ version = "1.3.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "bef38d45163c2f1dde094a7dfd33ccf595c92905c8f8f4fdc18d06fb1037718a"
 
+[[package]]
+name = "bstr"
+version = "0.2.17"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "ba3569f383e8f1598449f1a423e72e99569137b47740b1da11ef19af3d5c3223"
+dependencies = [
+ "lazy_static",
+ "memchr",
+ "regex-automata",
+ "serde",
+]
+
 [[package]]
 name = "bytemuck"
 version = "1.9.1"
@@ -126,6 +138,28 @@ dependencies = [
  "cfg-if",
 ]
 
+[[package]]
+name = "csv"
+version = "1.1.6"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "22813a6dc45b335f9bade10bf7271dc477e81113e89eb251a0bc2a8a81c536e1"
+dependencies = [
+ "bstr",
+ "csv-core",
+ "itoa 0.4.8",
+ "ryu",
+ "serde",
+]
+
+[[package]]
+name = "csv-core"
+version = "0.1.10"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "2b2466559f260f48ad25fe6317b3c8dac77b5bdb5763ac7d9d6103530663bc90"
+dependencies = [
+ "memchr",
+]
+
 [[package]]
 name = "data-url"
 version = "0.1.1"
@@ -272,6 +306,12 @@ dependencies = [
  "either",
 ]
 
+[[package]]
+name = "itoa"
+version = "0.4.8"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "b71991ff56294aa922b450139ee08b3bfc70982c6b2c7562771375cf73542dd4"
+
 [[package]]
 name = "itoa"
 version = "1.0.2"
@@ -431,7 +471,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "249f9b33a3192626f2cd9f4b0cd66c1ec32d65968d58cf4d8239977feddddead"
 dependencies = [
  "bitflags",
- "itoa",
+ "itoa 1.0.2",
  "ryu",
 ]
 
@@ -547,6 +587,12 @@ dependencies = [
  "regex-syntax",
 ]
 
+[[package]]
+name = "regex-automata"
+version = "0.1.10"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "6c230d73fb8d8c1b9c0b3135c5142a8acee3a0558fb8db5cf1cb65f8d7862132"
+
 [[package]]
 name = "regex-syntax"
 version = "0.6.26"
@@ -665,7 +711,7 @@ version = "1.0.81"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "9b7ce2b32a1aed03c558dc61a5cd328f15aff2dbc17daad8fb8af04d2100e15c"
 dependencies = [
- "itoa",
+ "itoa 1.0.2",
  "ryu",
  "serde",
 ]
@@ -816,6 +862,7 @@ version = "0.1.0"
 dependencies = [
  "bytemuck",
  "codespan-reporting",
+ "csv",
  "dirs",
  "flate2",
  "fxhash",

Cargo.toml

@@ -41,6 +41,7 @@ usvg = { version = "0.22", default-features = false }
 # External implementation of user-facing features
 syntect = { version = "5", default-features = false, features = ["default-syntaxes", "regex-fancy"] }
 rex = { git = "https://github.com/laurmaedje/ReX" }
+csv = "1"
 
 # PDF export
 miniz_oxide = "0.5"

src/lib.rs

@@ -79,6 +79,8 @@ pub fn typeset(ctx: &mut Context, id: SourceId) -> TypResult<Vec<Frame>> {
 
 /// The core context which holds the configuration and stores.
 pub struct Context {
+    /// The loader for fonts and files.
+    pub loader: Arc<dyn Loader>,
     /// Stores loaded source files.
     pub sources: SourceStore,
     /// Stores parsed font faces.
@@ -97,6 +99,7 @@ impl Context {
     /// Create a new context.
     pub fn new(loader: Arc<dyn Loader>, config: Config) -> Self {
         Self {
+            loader: Arc::clone(&loader),
             sources: SourceStore::new(Arc::clone(&loader)),
             fonts: FontStore::new(Arc::clone(&loader)),
             images: ImageStore::new(loader),

src/library/mod.rs

@@ -97,6 +97,7 @@ pub fn new() -> Scope {
     std.def_fn("roman", utility::roman);
     std.def_fn("symbol", utility::symbol);
     std.def_fn("lorem", utility::lorem);
+    std.def_fn("csv", utility::csv);
 
     // Predefined colors.
     std.define("black", Color::BLACK);

src/library/prelude.rs

@@ -2,12 +2,15 @@
 
 pub use std::fmt::{self, Debug, Formatter};
 pub use std::hash::Hash;
+pub use std::io;
 pub use std::num::NonZeroUsize;
 pub use std::sync::Arc;
 
 pub use typst_macros::node;
 
-pub use crate::diag::{with_alternative, At, Error, StrResult, TypError, TypResult};
+pub use crate::diag::{
+    failed_to_load, with_alternative, At, Error, StrResult, TypError, TypResult,
+};
 pub use crate::eval::{
     Arg, Args, Array, Cast, Dict, Dynamic, Func, Machine, Node, RawAlign, RawLength,
     RawStroke, Scope, Smart, Value,

src/library/utility/data.rs

@@ -0,0 +1,30 @@
+use crate::library::prelude::*;
+
+/// Read structured data from a CSV file.
+pub fn csv(vm: &mut Machine, args: &mut Args) -> TypResult<Value> {
+    let Spanned { v: path, span } =
+        args.expect::<Spanned<EcoString>>("path to csv file")?;
+
+    let path = vm.locate(&path).at(span)?;
+    let try_load = || -> io::Result<Value> {
+        let data = vm.ctx.loader.load(&path)?;
+
+        let mut builder = csv::ReaderBuilder::new();
+        builder.has_headers(false);
+
+        let mut reader = builder.from_reader(data.as_slice());
+        let mut vec = vec![];
+
+        for result in reader.records() {
+            vec.push(Value::Array(
+                result?.iter().map(|field| Value::Str(field.into())).collect(),
+            ))
+        }
+
+        Ok(Value::Array(Array::from_vec(vec)))
+    };
+
+    try_load()
+        .map_err(|err| failed_to_load("csv file", &path, err))
+        .at(span)
+}

src/library/utility/mod.rs

@@ -1,10 +1,12 @@
 //! Computational utility functions.
 
 mod color;
+mod data;
 mod math;
 mod string;
 
 pub use color::*;
+pub use data::*;
 pub use math::*;
 pub use string::*;
 

tests/res/bad.csv

@@ -0,0 +1,4 @@
+A,B
+1,2
+3,4,5
+6,7

tests/res/zoo.csv

@@ -0,0 +1,4 @@
+Name,Species,Weight,Length
+Debby,Rhinoceros,1900kg,390cm
+Fluffy,Tiger,115kg,310cm
+Sleepy,Dolphin,150kg,180cm

tests/typ/utility/csv.typ

@@ -0,0 +1,15 @@
+// Test reading structured CSV data.
+
+---
+#set page(width: auto)
+#let data = csv("/res/zoo.csv")
+#let cells = data(0).map(strong) + data.slice(1).flatten()
+#table(columns: data(0).len(), ..cells)
+
+---
+// Error: 6-16 file not found (searched at typ/utility/nope.csv)
+#csv("nope.csv")
+
+---
+// Error: 6-20 failed to load csv file (CSV error: record 2 (line: 3, byte: 8): found record with 3 fields, but the previous record has 2 fields)
+#csv("/res/bad.csv")