From ee99f119f055aba4263310955e5f9d290e4e3acd Mon Sep 17 00:00:00 2001
From: Julia Lange <public@julialange.org>
Date: Wed, 2 Jul 2025 10:48:44 -0700
Subject: [PATCH] Atproto, add bound string

---
 atproto/Cargo.toml                |  1 +
 atproto/src/error.rs              |  2 ++
 atproto/src/types.rs              |  2 ++
 atproto/src/types/bound_string.rs | 57 +++++++++++++++++++++++++++++++
 4 files changed, 62 insertions(+)
 create mode 100644 atproto/src/types/bound_string.rs

diff --git a/atproto/Cargo.toml b/atproto/Cargo.toml
index 3f5445b..0881493 100644
--- a/atproto/Cargo.toml
+++ b/atproto/Cargo.toml
@@ -13,6 +13,7 @@ time = { version = "0.3.41", features = ["parsing", "formatting"] }
 tracing-subscriber.workspace = true
 tracing.workspace = true
 thiserror.workspace = true
+unicode-segmentation = "1.9.0"
 
 [features]
 default = []
diff --git a/atproto/src/error.rs b/atproto/src/error.rs
index 5b52dfb..e320a61 100644
--- a/atproto/src/error.rs
+++ b/atproto/src/error.rs
@@ -24,6 +24,8 @@ pub enum ParseError {
   Serde(#[from] serde_json::error::Error),
   #[error("Length of parsed object too long, max: {max:?}, got: {got:?}.")]
   Length { max: usize, got: usize },
+  #[error("Length of parsed object too short, min: {min:?}, got: {got:?}.")]
+  MinLength { min: usize, got: usize },
   #[error("Currently Did is enforced, cannot use handle, {handle:?}")]
   ForceDid { handle: String },
   #[error("Incorrectly formatted")]
diff --git a/atproto/src/types.rs b/atproto/src/types.rs
index 5baa02d..2e1c387 100644
--- a/atproto/src/types.rs
+++ b/atproto/src/types.rs
@@ -56,6 +56,7 @@ macro_rules! basic_string_type {
 }
 
 mod authority;
+mod bound_string;
 mod cid;
 mod datetime;
 mod did;
@@ -63,6 +64,7 @@ mod record_key;
 mod strong_ref;
 mod uri;
 pub use authority::Authority;
+pub use bound_string::BoundString;
 pub use cid::Cid;
 pub use datetime::Datetime;
 pub use did::Did;
diff --git a/atproto/src/types/bound_string.rs b/atproto/src/types/bound_string.rs
new file mode 100644
index 0000000..db81bcf
--- /dev/null
+++ b/atproto/src/types/bound_string.rs
@@ -0,0 +1,57 @@
+use unicode_segmentation::UnicodeSegmentation;
+use crate::error::{Error, ParseError};
+use std::{
+  fmt::{Display, Formatter, Result as FmtResult},
+  str::FromStr,
+};
+
+pub struct BoundString<
+  const MIN: usize, const MAX: usize>
+{
+  value: String,
+}
+
+impl<const MIN: usize, const MAX: usize> BoundString<MIN, MAX> {
+  fn check_length(s: &str) -> Result<(), Error> {
+    let grapheme_count: usize = s.graphemes(true).take(MAX + 1).count();
+    if grapheme_count > MAX {
+      return Err(Error::Parse {
+        err: ParseError::Length { max: MAX, got: grapheme_count },
+        object: "String".to_string(),
+      });
+    }
+    if grapheme_count < MIN {
+      return Err(Error::Parse {
+        err: ParseError::MinLength { min: MIN, got: grapheme_count },
+        object: "String".to_string(),
+      });
+    }
+    Ok(())
+  }
+}
+
+impl<const MIN: usize, const MAX: usize> Display for BoundString<MIN, MAX> {
+  fn fmt(&self, f: &mut Formatter<'_>) -> FmtResult {
+    write!(f, "{}", self.value)
+  }
+}
+
+impl<const MIN: usize, const MAX: usize> FromStr for BoundString<MIN, MAX> {
+  type Err = Error;
+  fn from_str(s: &str) -> Result<Self, Self::Err> {
+    Self::check_length(s)?;
+
+    Ok(BoundString { value: s.to_string() })
+  }
+}
+
+impl<'de, const MIN: usize, const MAX: usize> serde::de::Deserialize<'de> 
+    for BoundString<MIN, MAX> {
+  fn deserialize<D>(deserializer: D) -> Result<Self, D::Error> 
+  where
+    D: serde::de::Deserializer<'de>,
+  {
+    let value: String = serde::de::Deserialize::deserialize(deserializer)?;
+    value.parse::<BoundString<MIN,MAX>>().map_err(<D::Error as serde::de::Error>::custom)
+  }
+}