Regex scan

4 files changed, 64 insertions, 56 deletions

diff --git a/CHANGELOG.md b/CHANGELOG.md
index d8ec189..f2e0b73 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -24,6 +24,7 @@
   string error description.
 - The `string` module gains the `to_option` function.
 - Fixed a bug where `io.print` could crash when printing special characters.
+- The `regex.Match` record no longer has the `byte_index` field any more.
 
 ## v0.16.0 - 2021-06-17
 
diff --git a/src/gleam/regex.gleam b/src/gleam/regex.gleam
index 8eddabe..cd717d9 100644
--- a/src/gleam/regex.gleam
+++ b/src/gleam/regex.gleam
@@ -13,8 +13,6 @@ pub type Match {
   Match(
     /// The full string of the match.
     content: String,
-    /// The byte index of the match in the original string.
-    byte_index: Int,
     /// A Regex can have subpatterns, sup-parts that are in parentheses.
     submatches: List(Option(String)),
   )
@@ -141,26 +139,33 @@ if javascript {
     "../gleam_stdlib.js" "split"
 }
 
+/// Collects all matches of the regular expression.
+///
+/// ## Examples
+///
+///    > assert Ok(re) = regex.from_string("[oi]n a (\\w+)")
+///    > regex.scan(with: re, content: "I am on a boat in a lake.")
+///    [
+///      Match(
+///        content: "on a boat",
+///        submatches: [Some("boat")]
+///      ),
+///      Match(
+///        content: "in a lake",
+///        submatches: [Some("lake")]
+///      )
+///    ]
+///
+pub fn scan(with regex: Regex, content string: String) -> List(Match) {
+  do_scan(regex, string)
+}
+
 if erlang {
-  /// Collects all matches of the regular expression.
-  ///
-  /// ## Examples
-  ///
-  ///    > assert Ok(re) = regex.from_string("[oi]n a (\\w+)")
-  ///    > regex.scan(with: re, content: "I am on a boat in a lake.")
-  ///    [
-  ///      Match(
-  ///        content: "on a boat",
-  ///        byte_index: 5,
-  ///        submatches: [Some("boat")]
-  ///      ),
-  ///      Match(
-  ///        content: "in a lake",
-  ///        byte_index: 15,
-  ///        submatches: [Some("lake")]
-  ///      )
-  ///    ]
-  ///
-  pub external fn scan(with: Regex, content: String) -> List(Match) =
+  external fn do_scan(Regex, String) -> List(Match) =
     "gleam_stdlib" "regex_scan"
 }
+
+if javascript {
+  external fn do_scan(Regex, String) -> List(Match) =
+    "../gleam_stdlib.js" "regex_scan"
+}
diff --git a/src/gleam_stdlib.js b/src/gleam_stdlib.js
index d2454ba..652b7b5 100644
--- a/src/gleam_stdlib.js
+++ b/src/gleam_stdlib.js
@@ -7,7 +7,11 @@ import {
   toBitString,
   stringBits,
 } from "./gleam.js";
-import { CompileError as RegexCompileError } from "./gleam/regex.js";
+import {
+  CompileError as RegexCompileError,
+  Match as RegexMatch,
+} from "./gleam/regex.js";
+import { Some, None } from "./gleam/option.js";
 
 const Nil = undefined;
 
@@ -236,13 +240,21 @@ export function regex_check(regex, string) {
 
 export function compile_regex(pattern, options) {
   try {
-    let flags = "";
+    let flags = "gu";
     if (options.case_insensitive) flags += "i";
     if (options.multi_line) flags += "m";
     return new Ok(new RegExp(pattern, flags));
   } catch (error) {
-    return new Error(
-      new RegexCompileError(error.message, error.columnNumber || 0)
-    );
+    let number = (error.columnNumber || 0) | 0;
+    return new Error(new RegexCompileError(error.message, number));
   }
 }
+
+export function regex_scan(regex, string) {
+  let matches = Array.from(string.matchAll(regex)).map((match) => {
+    let content = match.shift();
+    let submatches = match.map((x) => (x ? new Some(x) : new None()));
+    return new RegexMatch(content, List.fromArray(submatches));
+  });
+  return List.fromArray(matches);
+}
diff --git a/test/gleam/regex_test.gleam b/test/gleam/regex_test.gleam
index 87535a0..5916b63 100644
--- a/test/gleam/regex_test.gleam
+++ b/test/gleam/regex_test.gleam
@@ -46,33 +46,23 @@ pub fn split_test() {
   |> should.equal(["foo", "32", "4", "9", "0"])
 }
 
-if erlang {
-  pub fn scan_test() {
-    assert Ok(re) = regex.from_string("Gl\\w+")
-
-    regex.scan(re, "!Gleam")
-    |> should.equal([Match(content: "Gleam", byte_index: 1, submatches: [])])
-
-    regex.scan(re, "हGleam")
-    |> should.equal([Match(content: "Gleam", byte_index: 3, submatches: [])])
-
-    regex.scan(re, "𐍈Gleam")
-    |> should.equal([Match(content: "Gleam", byte_index: 4, submatches: [])])
-
-    assert Ok(re) = regex.from_string("[oi]n a(.?) (\\w+)")
-
-    regex.scan(re, "I am on a boat in a lake.")
-    |> should.equal([
-      Match(
-        content: "on a boat",
-        byte_index: 5,
-        submatches: [None, Some("boat")],
-      ),
-      Match(
-        content: "in a lake",
-        byte_index: 15,
-        submatches: [None, Some("lake")],
-      ),
-    ])
-  }
+pub fn scan_test() {
+  assert Ok(re) = regex.from_string("Gl\\w+")
+
+  regex.scan(re, "!Gleam")
+  |> should.equal([Match(content: "Gleam", submatches: [])])
+
+  regex.scan(re, "हGleam")
+  |> should.equal([Match(content: "Gleam", submatches: [])])
+
+  regex.scan(re, "𐍈Gleam")
+  |> should.equal([Match(content: "Gleam", submatches: [])])
+
+  assert Ok(re) = regex.from_string("[oi]n a(.?) (\\w+)")
+
+  regex.scan(re, "I am on a boat in a lake.")
+  |> should.equal([
+    Match(content: "on a boat", submatches: [None, Some("boat")]),
+    Match(content: "in a lake", submatches: [None, Some("lake")]),
+  ])
 }