add reflow command (#2128)

* add reflow command

Users need to be able to hard-wrap text for many applications, including
comments in code, git commit messages, plaintext documentation, etc. It
often falls to the user to manually insert line breaks where appropriate
in order to hard-wrap text.

This commit introduces the "reflow" command (both in the TUI and core
library) to automatically hard-wrap selected text to a given number of
characters (defined by Unicode "extended grapheme clusters"). It handles
lines with a repeated prefix, such as comments ("//") and indentation.

* reflow: consider newlines to be word separators

* replace custom reflow impl with textwrap crate

* Sync reflow command docs with book

* reflow: add default max_line_len language setting

Co-authored-by: Vince Mutolo <vince@mutolo.org>

7 files changed, 84 insertions, 0 deletions

diff --git a/Cargo.lock b/Cargo.lock
index 18f50af6..f075249a 100644
--- a/Cargo.lock
+++ b/Cargo.lock
@@ -374,6 +374,7 @@ dependencies = [
  "slotmap",
  "smallvec",
  "smartstring",
+ "textwrap",
  "toml",
  "tree-sitter",
  "unicode-general-category",
@@ -1006,6 +1007,12 @@ dependencies = [
 ]
 
 [[package]]
+name = "smawk"
+version = "0.3.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "f67ad224767faa3c7d8b6d91985b78e70a1324408abcb1cfcc2be4c06bc06043"
+
+[[package]]
 name = "socket2"
 version = "0.4.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
@@ -1045,6 +1052,17 @@ dependencies = [
 ]
 
 [[package]]
+name = "textwrap"
+version = "0.15.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "b1141d4d61095b28419e22cb0bbf02755f5e54e0526f97f1e3d1d160e60885fb"
+dependencies = [
+ "smawk",
+ "unicode-linebreak",
+ "unicode-width",
+]
+
+[[package]]
 name = "thiserror"
 version = "1.0.30"
 source = "registry+https://github.com/rust-lang/crates.io-index"
@@ -1180,6 +1198,15 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "1218098468b8085b19a2824104c70d976491d247ce194bbd9dc77181150cdfd6"
 
 [[package]]
+name = "unicode-linebreak"
+version = "0.1.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "3a52dcaab0c48d931f7cc8ef826fa51690a08e1ea55117ef26f89864f532383f"
+dependencies = [
+ "regex",
+]
+
+[[package]]
 name = "unicode-normalization"
 version = "0.1.19"
 source = "registry+https://github.com/rust-lang/crates.io-index"
diff --git a/book/src/generated/typable-cmd.md b/book/src/generated/typable-cmd.md
index df0841b6..426598e3 100644
--- a/book/src/generated/typable-cmd.md
+++ b/book/src/generated/typable-cmd.md
@@ -59,6 +59,7 @@
 | `:get-option`, `:get` | Get the current value of a config option. |
 | `:sort` | Sort ranges in selection. |
 | `:rsort` | Sort ranges in selection in reverse order. |
+| `:reflow` | Hard-wrap the current selection of lines to a given width. |
 | `:tree-sitter-subtree`, `:ts-subtree` | Display tree sitter subtree under cursor, primarily for debugging queries. |
 | `:config-reload` | Refreshes helix's config. |
 | `:config-open` | Open the helix config.toml file. |
diff --git a/helix-core/Cargo.toml b/helix-core/Cargo.toml
index 6e019a42..ab937f0b 100644
--- a/helix-core/Cargo.toml
+++ b/helix-core/Cargo.toml
@@ -41,6 +41,7 @@ encoding_rs = "0.8"
 chrono = { version = "0.4", default-features = false, features = ["alloc", "std"] }
 
 etcetera = "0.3"
+textwrap = "0.15.0"
 
 [dev-dependencies]
 quickcheck = { version = "1", default-features = false }
diff --git a/helix-core/src/lib.rs b/helix-core/src/lib.rs
index 02341265..a022a42a 100644
--- a/helix-core/src/lib.rs
+++ b/helix-core/src/lib.rs
@@ -27,6 +27,7 @@ pub mod syntax;
 pub mod test;
 pub mod textobject;
 mod transaction;
+pub mod wrap;
 
 pub mod unicode {
     pub use unicode_general_category as category;
diff --git a/helix-core/src/syntax.rs b/helix-core/src/syntax.rs
index 3f9e7bcf..eab3ab79 100644
--- a/helix-core/src/syntax.rs
+++ b/helix-core/src/syntax.rs
@@ -67,6 +67,7 @@ pub struct LanguageConfiguration {
     pub shebangs: Vec<String>, // interpreter(s) associated with language
     pub roots: Vec<String>,      // these indicate project roots <.git, Cargo.toml>
     pub comment_token: Option<String>,
+    pub max_line_length: Option<usize>,
 
     #[serde(default, skip_serializing, deserialize_with = "deserialize_lsp_config")]
     pub config: Option<serde_json::Value>,
diff --git a/helix-core/src/wrap.rs b/helix-core/src/wrap.rs
new file mode 100644
index 00000000..eabc47d4
--- /dev/null
+++ b/helix-core/src/wrap.rs
@@ -0,0 +1,7 @@
+use smartstring::{LazyCompact, SmartString};
+
+/// Given a slice of text, return the text re-wrapped to fit it
+/// within the given width.
+pub fn reflow_hard_wrap(text: &str, max_line_len: usize) -> SmartString<LazyCompact> {
+    textwrap::refill(text, max_line_len).into()
+}
diff --git a/helix-term/src/commands/typed.rs b/helix-term/src/commands/typed.rs
index 373c7018..ec86e446 100644
--- a/helix-term/src/commands/typed.rs
+++ b/helix-term/src/commands/typed.rs
@@ -1051,6 +1051,45 @@ fn sort_impl(
     Ok(())
 }
 
+fn reflow(
+    cx: &mut compositor::Context,
+    args: &[Cow<str>],
+    _event: PromptEvent,
+) -> anyhow::Result<()> {
+    let (view, doc) = current!(cx.editor);
+
+    const DEFAULT_MAX_LEN: usize = 79;
+
+    // Find the max line length by checking the following sources in order:
+    //   - The passed argument in `args`
+    //   - The configured max_line_len for this language in languages.toml
+    //   - The const default we set above
+    let max_line_len: usize = args
+        .get(0)
+        .map(|num| num.parse::<usize>())
+        .transpose()?
+        .or_else(|| {
+            doc.language_config()
+                .and_then(|config| config.max_line_length)
+        })
+        .unwrap_or(DEFAULT_MAX_LEN);
+
+    let rope = doc.text();
+
+    let selection = doc.selection(view.id);
+    let transaction = Transaction::change_by_selection(rope, selection, |range| {
+        let fragment = range.fragment(rope.slice(..));
+        let reflowed_text = helix_core::wrap::reflow_hard_wrap(&fragment, max_line_len);
+
+        (range.from(), range.to(), Some(reflowed_text))
+    });
+
+    doc.apply(&transaction, view.id);
+    doc.append_changes_to_history(view.id);
+
+    Ok(())
+}
+
 fn tree_sitter_subtree(
     cx: &mut compositor::Context,
     _args: &[Cow<str>],
@@ -1571,6 +1610,13 @@ pub const TYPABLE_COMMAND_LIST: &[TypableCommand] = &[
             completer: None,
         },
         TypableCommand {
+            name: "reflow",
+            aliases: &[],
+            doc: "Hard-wrap the current selection of lines to a given width.",
+            fun: reflow,
+            completer: None,
+        },
+        TypableCommand {
             name: "tree-sitter-subtree",
             aliases: &["ts-subtree"],
             doc: "Display tree sitter subtree under cursor, primarily for debugging queries.",