Update for Gleam 0.33, split codebase

.tool-versions

@@ -1,3 +1,3 @@
-gleam 0.30.2
-erlang 25.2.2
+gleam 0.33.0
+erlang 26.1.2
 nodejs 18.14.0

CHANGELOG.txt

@@ -1,3 +1,10 @@
+6.0.0
+-----
+
+* Update for Gleam 0.32+.
+* Split encoders and the decoder into separate files so that they can be separately imported.
+  This helps keep the file size down.
+
 5.0.0
 -----
 

README.md

@@ -20,6 +20,10 @@ glentities.decode("&#x3C;&#x2F;html&#x3E;")       // "</html>"
 glentities.decode("&lt;&sol;html&gt;")            // "</html>"
 ```
 
+### Code size
+
+If you import `glentities`, it will pull in all the encoders and the entire named decoder. Since there are so many different entity names, this will result in a hefty JS payload. If you need to minimise the payload size, prefer importing `glentities/decoder` or one of the `glentities/*_encoder` modules directly.
+
 ## Development
 
 ```sh

gleam.toml

@@ -1,5 +1,5 @@
 name = "glentities"
-version = "5.0.0"
+version = "6.0.0"
 description = "HTML entity encoder/decoder for Gleam"
 
 # Fill out these fields if you intend to generate HTML documentation or publish
@@ -10,9 +10,10 @@ repository = { type = "gitlab", user = "Nicd", repo = "glentities" }
 links = []
 
 internal_modules = ["glentities/internal/*"]
+gleam = ">= 0.32.0"
 
 [dependencies]
-gleam_stdlib = "~> 0.29"
+gleam_stdlib = "~> 0.32"
 
 [dev-dependencies]
-gleeunit = "~> 0.10"
+gleeunit = "~> 1.0"

manifest.toml

@@ -2,10 +2,10 @@
 # You typically do not need to edit this file
 
 packages = [
-  { name = "gleam_stdlib", version = "0.30.0", build_tools = ["gleam"], requirements = [], otp_app = "gleam_stdlib", source = "hex", outer_checksum = "03710B3DA047A3683117591707FCA19D32B980229DD8CE8B0603EB5B5144F6C3" },
-  { name = "gleeunit", version = "0.10.1", build_tools = ["gleam"], requirements = ["gleam_stdlib"], otp_app = "gleeunit", source = "hex", outer_checksum = "ECEA2DE4BE6528D36AFE74F42A21CDF99966EC36D7F25DEB34D47DD0F7977BAF" },
+  { name = "gleam_stdlib", version = "0.34.0", build_tools = ["gleam"], requirements = [], otp_app = "gleam_stdlib", source = "hex", outer_checksum = "1FB8454D2991E9B4C0C804544D8A9AD0F6184725E20D63C3155F0AEB4230B016" },
+  { name = "gleeunit", version = "1.0.1", build_tools = ["gleam"], requirements = ["gleam_stdlib"], otp_app = "gleeunit", source = "hex", outer_checksum = "4E75DCF846D653848094169304743DFFB386E3AECCCF611F99ADB735FF4D4DD9" },
 ]
 
 [requirements]
-gleam_stdlib = { version = "~> 0.29" }
-gleeunit = { version = "~> 0.10" }
+gleam_stdlib = { version = "~> 0.32" }
+gleeunit = { version = "~> 1.0" }

src/glentities/hex_encoder.gleam

@@ -0,0 +1,24 @@
+import gleam/list
+import gleam/string
+import gleam/int
+import glentities/internal/string_utils
+
+/// Encode text using HTML hex entities, except a-z, A-Z, 0-9, newline, tab, carriage return, and space.
+pub fn encode(text: String) {
+  text
+  |> string_utils.normalise()
+  |> do_encode()
+}
+
+fn do_encode(text: String) {
+  text
+  |> string.to_utf_codepoints()
+  |> list.map(fn(codepoint) {
+    case string.utf_codepoint_to_int(codepoint) {
+      c if c == 9 || c == 10 || c == 13 || c == 32 || c >= 48 && c <= 57 || c >= 65 && c <= 90 || c >= 97 && c <= 122 ->
+        string.from_utf_codepoints([codepoint])
+      encodable -> "&#x" <> int.to_base16(encodable) <> ";"
+    }
+  })
+  |> string.join("")
+}

src/glentities/html_encoder.gleam

@@ -0,0 +1,33 @@
+import gleam/string
+import gleam/string_builder.{type StringBuilder}
+import glentities/internal/string_utils
+
+/// Encode text to be safe in the HTML body, inside element or attribute content.
+///
+/// `&`, `<`, `>`, `'`, and `"` are encoded.
+///
+/// Note! Not suitable for outputting inside `<style>`, `<script>` elements.
+pub fn encode(text: String) -> String {
+  text
+  |> string_utils.normalise()
+  |> do_encode(string_builder.new())
+}
+
+fn do_encode(text: String, acc: StringBuilder) {
+  case text {
+    "" -> string_builder.to_string(acc)
+    "&" <> rest -> do_encode(rest, string_builder.append(acc, "&amp;"))
+    "<" <> rest -> do_encode(rest, string_builder.append(acc, "&lt;"))
+    ">" <> rest -> do_encode(rest, string_builder.append(acc, "&gt;"))
+    "\"" <> rest -> do_encode(rest, string_builder.append(acc, "&quot;"))
+    "'" <> rest -> do_encode(rest, string_builder.append(acc, "&#39;"))
+    other -> {
+      let maybe_grapheme = string.pop_grapheme(other)
+      case maybe_grapheme {
+        Ok(#(grapheme, rest)) ->
+          do_encode(rest, string_builder.append(acc, grapheme))
+        Error(Nil) -> string_builder.to_string(acc)
+      }
+    }
+  }
+}