Merge pull request #17 from lupodevelop/14-html-escaping-and-unescaping-is-slow-and-incomplete

14 html escaping and unescaping is slow and incomplete

Author: lupodevelop

.github/workflows/ci.yml

@@ -0,0 +1,23 @@
+name: test
+
+on:
+  push:
+    branches:
+      - master
+      - main
+  pull_request:
+
+jobs:
+  test:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+      - uses: erlef/setup-beam@v1
+        with:
+          otp-version: "28"
+          gleam-version: "1.13.0"
+          rebar3-version: "3"
+          # elixir-version: "1"
+      - run: gleam deps download
+      - run: gleam test
+      - run: gleam format --check src test

CHANGELOG.md

@@ -2,6 +2,24 @@
 
 All notable changes to this project are documented in this file.
 
+## [1.2.3] - 2026-01-08
+### Changed
+- Replaced `escape_html` implementation with `houdini.escape` for faster,
+  allocation-friendly HTML escaping.
+- Replaced `unescape_html` with `odysseus.unescape` for comprehensive HTML
+  entity unescaping (named entities, numeric decimal and hex entities).
+- Added dependencies: `houdini`, `odysseus`.
+
+### Tests
+- Added tests for HTML escape/unescape and numeric entities (decimal and hex).
+
+Contributed by: Daniele (`lupodevelop`)
+Suggested by: Louis Pilfold (`@lpil`)
+
+Suggested by: NNB (`@NNBnh`)
+Suggested change: updated README logo pointer to use the raw.githubusercontent URL
+(pointing to the repository commit) so the logo is resolvable on Hexdocs.
+
 ## [1.2.2] - 2026-01-05
 ### Added
 - Added internal helper `grapheme_len/1` (internal) to centralize grapheme cluster length computation and avoid repetitive `string.to_graphemes |> list.length` patterns.

CONTRIBUTING.md

@@ -0,0 +1,34 @@
+# Contributing to str
+
+Thanks for helping! Short, practical guide.
+
+## Quick start
+- Fork, create a branch: `git switch -c feat/your-change`.
+- Run `gleam format` and `gleam test` locally.
+- Open a PR against `main` with a short description and tests.
+
+## Setup
+- Requirements: Gleam (see `gleam.toml`)
+
+Commands:
+```bash
+gleam format
+gleam test
+```
+
+## Commits
+Use brief prefixes: `feat:`, `fix:`, `chore:`, `test:`, `perf:`.
+Example: `feat(display): add truncate_display`
+No strict enforcement, use these prefixes as a guideline, not a hard rule.
+
+## PR checklist
+- [ ] Tests added/updated
+- [ ] `gleam format` & `gleam test` pass
+- [ ] Update `CHANGELOG.md` if behaviour changes
+- [ ] Document noteworthy changes in `README.md` , docs/ or examples/
+
+## Deprecations
+- Report breaking changes in an issue and add migration notes in PRs. See `DEPRECATIONS.md` if present.
+
+## Testing
+- Add unit tests for edge cases (ZWJ, skin tones, combining marks, CJK, ambiguous widths).

README.md

@@ -1,5 +1,5 @@
 <p align="center">
-  <img src="assets/img/logo-str.png" alt="str logo" width="280">
+  <img src="https://raw.githubusercontent.com/lupodevelop/str/c190b21/assets/img/logo-str.png" alt="str logo" width="280">
 </p>
 
 <h1 align="center">str</h1>
@@ -327,6 +327,8 @@ gleam test
 python3 scripts/generate_character_tables.py
 ```
 
+Note: as of **1.2.3**, `escape_html` now uses the `houdini` library for fast, allocation‑friendly escaping, and `unescape_html` uses `odysseus` for comprehensive entity support (named, decimal and hex numeric entities). See [CHANGELOG.md](CHANGELOG.md) for details.
+
 ---
 
 ## 📊 Test Coverage

gleam.toml

@@ -1,17 +1,18 @@
 name = "str"
-version = "1.2.2"
+version = "1.2.3"
 
 # Project metadata (fill or replace placeholders before publishing)
 description = "Unicode-aware string utilities for Gleam: grapheme-safe operations, pragmatic ASCII transliteration, and slug generation."
 licenses = ["MIT"]
 repository = { type = "github", user = "lupodevelop", repo = "str" }
-links = [{ title = "Repository", href = "https://github.com/lupodevelop/str" }]
 
 # For a full reference of all the available options, see:
 # https://gleam.run/writing-gleam/gleam-toml/
 
 [dependencies]
 gleam_stdlib = ">= 0.44.0 and < 2.0.0"
+houdini = ">= 1.0.0 and < 2.0.0"
+odysseus = ">= 1.0.0 and < 2.0.0"
 
 [dev-dependencies]
 gleeunit = ">= 1.0.0 and < 2.0.0"

manifest.toml

@@ -4,8 +4,12 @@
 packages = [
   { name = "gleam_stdlib", version = "0.65.0", build_tools = ["gleam"], requirements = [], otp_app = "gleam_stdlib", source = "hex", outer_checksum = "7C69C71D8C493AE11A5184828A77110EB05A7786EBF8B25B36A72F879C3EE107" },
   { name = "gleeunit", version = "1.9.0", build_tools = ["gleam"], requirements = ["gleam_stdlib"], otp_app = "gleeunit", source = "hex", outer_checksum = "DA9553CE58B67924B3C631F96FE3370C49EB6D6DC6B384EC4862CC4AAA718F3C" },
+  { name = "houdini", version = "1.2.0", build_tools = ["gleam"], requirements = [], otp_app = "houdini", source = "hex", outer_checksum = "5DB1053F1AF828049C2B206D4403C18970ABEF5C18671CA3C2D2ED0DD64F6385" },
+  { name = "odysseus", version = "1.0.0", build_tools = ["gleam"], requirements = [], otp_app = "odysseus", source = "hex", outer_checksum = "6A97DA1075BDDEA8B60F47B1DFFAD49309FA27E73843F13A0AF32EA7087BA11C" },
 ]
 
 [requirements]
 gleam_stdlib = { version = ">= 0.44.0 and < 2.0.0" }
 gleeunit = { version = ">= 1.0.0 and < 2.0.0" }
+houdini = { version = ">= 1.0.0 and < 2.0.0" }
+odysseus = { version = ">= 1.0.0 and < 2.0.0" }

src/str/core.gleam

@@ -13,6 +13,8 @@ import gleam/dict
 import gleam/int
 import gleam/list
 import gleam/string
+import houdini
+import odysseus
 import str/config
 
 /// Detects if a grapheme cluster likely contains emoji components.
@@ -1766,12 +1768,7 @@ pub fn is_hex(text: String) -> Bool {
 ///   escape_html("Say \"hello\"") -> "Say "hello""
 ///
 pub fn escape_html(text: String) -> String {
-  text
-  |> string.replace("&", "&")
-  |> string.replace("<", "&lt;")
-  |> string.replace(">", "&gt;")
-  |> string.replace("\"", "&quot;")
-  |> string.replace("'", "&#39;")
+  houdini.escape(text)
 }
 
 /// Unescapes HTML entities to their character equivalents.
@@ -1781,12 +1778,7 @@ pub fn escape_html(text: String) -> String {
 ///   unescape_html("Tom &amp; Jerry") -> "Tom & Jerry"
 ///
 pub fn unescape_html(text: String) -> String {
-  text
-  |> string.replace("&#39;", "'")
-  |> string.replace("&quot;", "\"")
-  |> string.replace("&gt;", ">")
-  |> string.replace("&lt;", "<")
-  |> string.replace("&amp;", "&")
+  odysseus.unescape(text)
 }
 
 /// Escapes regex metacharacters so the string can be used as a literal pattern.

test/str_html_escape_extended_test.gleam

@@ -0,0 +1,64 @@
+import gleam/list
+import gleeunit
+import str
+
+pub fn main() -> Nil {
+  gleeunit.main()
+}
+
+pub fn roundtrip_basic_entities_test() {
+  let cases = [
+    "<div>Hello</div>",
+    "Tom & Jerry",
+    "Say \"hello\"",
+    "It's me",
+    "5 < 10 && 10 > 5",
+    "Ampersand: &",
+  ]
+
+  list.fold(cases, True, fn(_, s) {
+    let escaped = str.escape_html(s)
+    let unescaped = str.unescape_html(escaped)
+    assert unescaped == s
+    True
+  })
+}
+
+pub fn numeric_and_named_entities_test() {
+  assert str.unescape_html("&lt;&gt;&amp;&#39;&#x27;&#34;") == "<>&''\""
+  assert str.unescape_html("&quot; and &#34; and &#x22;") == "\" and \" and \""
+  assert str.unescape_html("I like &#39;quotes&#39;") == "I like 'quotes'"
+  assert str.unescape_html("Hex: &#x27;") == "Hex: '"
+}
+
+pub fn malformed_and_unknown_entity_test() {
+  // Missing semicolon should remain unchanged
+  assert str.unescape_html("This &amp is broken") == "This &amp is broken"
+
+  // Unknown entity should remain unchanged
+  assert str.unescape_html("This &notanentity; remains")
+    == "This &notanentity; remains"
+}
+
+pub fn combined_and_adjacent_entities_test() {
+  assert str.unescape_html("&lt;&lt; &gt;&gt;") == "<< >>"
+  assert str.unescape_html("&amp;&amp;&amp;") == "&&&"
+}
+
+pub fn unicode_and_emoji_roundtrip_test() {
+  let s = "Café — ️👩‍👩‍👧‍👦 \u{00A0}"
+  let escaped = str.escape_html(s)
+  // Expect unescape to restore the original (escape may not change emoji/nbspace)
+  assert str.unescape_html(escaped) == s
+}
+
+pub fn idempotence_and_double_escape_test() {
+  let s = "&"
+  let once = str.escape_html(s)
+  let twice = str.escape_html(once)
+  assert once == "&amp;"
+  assert twice == "&amp;amp;"
+  // unescape decodes one level: "&amp;amp;" -> "&amp;"; double unescape restores original
+  assert str.unescape_html(twice) == "&amp;"
+  assert str.unescape_html(str.unescape_html(twice)) == s
+}

test/str_html_escape_fuzz_test.gleam

@@ -0,0 +1,70 @@
+import gleeunit
+import str
+import gleam/list
+import gleam/string
+
+pub fn main() -> Nil {
+  gleeunit.main()
+}
+
+// Deterministic, simple generator over a token pool.
+fn gen_token_pool() -> List(String) {
+  [
+    "a","b","c","1","2","3"," ","\n","<",">","&","\"","'",
+    "&amp;","&lt;","&gt;","&quot;","&#39;","&#x27;","&#x22;","&notanentity;",
+    "&","&amp","&#", "&#x",
+    "\u{00A0}", // NBSP
+    "Café","naïve","ø","漢","字",
+    "👩‍👩‍👧‍👦","👨‍👩‍👧","️","✈️","🏳️‍🌈",
+    "\u{0301}", // combining acute
+    "&alpha;","&beta;","&gamma;"
+  ]
+}
+
+// Deterministic pseudo-random index using seed and i
+fn idx_for(seed: Int, i: Int, len: Int) -> Int {
+  // simple LCG-ish formula; keep small to avoid large-int overhead
+  let v = seed * 1103515245 + 12345 + i
+  let v_pos = case v < 0 { True -> -v False -> v }
+  v_pos % len
+}
+
+fn gen_string(seed: Int, tokens: List(String), n: Int) -> String {
+  let len = list.length(tokens)
+  let seq = list.range(0, n - 1)
+  seq
+  |> list.map(fn(i) {
+    let j = idx_for(seed, i, len)
+    case list.drop(tokens, j) {
+      [first, ..] -> first
+      [] -> ""
+    }
+  })
+  |> list.fold("", fn(acc, s) { acc <> s })
+}
+
+fn run_cfg(seed: Int, n: Int, tokens: List(String)) -> Bool {
+  let s = gen_string(seed, tokens, n)
+  // Roundtrip: unescape(escape(s)) == s
+  let escaped = str.escape_html(s)
+  let unescaped = str.unescape_html(escaped)
+  assert unescaped == s
+
+  // Escaped string must not contain raw angle brackets or quotes
+  assert string.contains(escaped, "<") == False
+  assert string.contains(escaped, ">") == False
+  assert string.contains(escaped, "\"") == False
+  assert string.contains(escaped, "'") == False
+
+  True
+}
+
+pub fn fuzz_roundtrip_test() {
+  let tokens = gen_token_pool()
+
+  run_cfg(1, 20, tokens)
+  run_cfg(42, 50, tokens)
+  run_cfg(123, 200, tokens)
+
+  True
+}

test/str_html_escape_test.gleam

@@ -0,0 +1,31 @@
+import str
+
+pub fn escape_basic_test() {
+  assert str.escape_html("<div>Hello</div>") == "&lt;div&gt;Hello&lt;/div&gt;"
+  assert str.escape_html("Tom & Jerry") == "Tom &amp; Jerry"
+  assert str.escape_html("Say \"hello\"") == "Say &quot;hello&quot;"
+}
+
+pub fn unescape_basic_test() {
+  assert str.unescape_html("&lt;div&gt;") == "<div>"
+  assert str.unescape_html("Tom &amp; Jerry") == "Tom & Jerry"
+  assert str.unescape_html("Say &quot;hello&quot;") == "Say \"hello\""
+  assert str.unescape_html("It&#39;s me") == "It's me"
+}
+
+pub fn roundtrip_test() {
+  let s = "Hello & < > \""
+  let escaped = str.escape_html(s)
+  assert str.unescape_html(escaped) == s
+}
+
+pub fn numeric_entities_test() {
+  // Decimal numeric entity
+  assert str.unescape_html("I like &#39;quotes&#39;") == "I like 'quotes'"
+
+  // Hex numeric entity
+  assert str.unescape_html("Hex: &#x27;") == "Hex: '"
+
+  // Double quote numeric and hex
+  assert str.unescape_html("&quot; and &#34; and &#x22;") == "\" and \" and \""
+}