Merge branch 'main' of github.com:roc-lang/unicode

.github/workflows/tests-nix.yaml

@@ -20,11 +20,4 @@ jobs:
 
         # Run all tests 
         - name: Run all tests
-          run: nix develop -c sh -c 'export ROC=roc && ./ci/all_tests.sh'
-
-        - name: Run all tests and capture output to check for crash later
-          run: nix develop -c sh -c 'export ROC=roc && ./ci/all_tests.sh 2>&1 | tee all_tests_output.log'
-
-        # Workaround for https://github.com/roc-lang/roc/issues/6688
-        - name: Check if crash occurred
-          run: grep -qv "crashed" all_tests_output.log
+          run: nix develop -c sh -c 'export ROC=roc && ./ci/all_tests.sh'

.gitignore

@@ -1,8 +1,8 @@
 
 # Ignore the example binaries
-examples/nrOfCodePoints
-examples/getVisualWidth
-examples/splitGraphemes
+examples/get-visual-width
+examples/nr-of-code-points
+examples/split-graphemes
 
 # Ignore the generated files
 generated-docs

examples/get-visual-width.roc

@@ -0,0 +1,25 @@
+app [main!] {
+    pf: platform "https://github.com/roc-lang/basic-cli/releases/download/0.19.0/bi5zubJ-_Hva9vxxPq4kNx4WHX6oFs8OP6Ad0tCYlrY.tar.br",
+    unicode: "../package/main.roc", # use release URL (ends in tar.br) for local example, see github.com/roc/unicode/releases
+}
+
+import pf.Stdout
+import unicode.CodePoint
+
+word = "世界"
+
+## Get the display width (in amount of characters) of a Str
+get_visual_width : Str -> Result U32 CodePoint.Utf8ParseErr
+get_visual_width = |str|
+    str
+    |> Str.to_utf8
+    |> CodePoint.parse_utf8
+    |> Result.map_ok(|lst| List.map(lst, CodePoint.visual_width))
+    |> Result.map_ok(List.sum)
+
+main! = |_args|
+    when get_visual_width(word) is
+        Ok(width) -> Stdout.line!("\n\nThe word ${word} will be displayed with the width of ${Num.to_str(width)} characters on most UIs.\n\n")
+        Err(_) -> crash("ERROR: Unable to parse ${word}!")
+
+expect (get_visual_width(word)) == Ok(4)

examples/nr-of-code-points.roc

@@ -0,0 +1,22 @@
+app [main!] {
+    pf: platform "https://github.com/roc-lang/basic-cli/releases/download/0.19.0/bi5zubJ-_Hva9vxxPq4kNx4WHX6oFs8OP6Ad0tCYlrY.tar.br",
+    unicode: "../package/main.roc", # use release URL (ends in tar.br) for local example, see github.com/roc/unicode/releases
+}
+
+import pf.Stdout
+import unicode.CodePoint exposing [Utf8ParseErr]
+
+## Get the number of code points for a given Str
+nr_of_code_points : Str -> Result U64 Utf8ParseErr
+nr_of_code_points = |str|
+    str |> Str.to_utf8 |> CodePoint.parse_utf8 |> Result.map_ok(List.len)
+
+main! = |_args|
+    word = "ẇ͓̞͒͟͡ǫ̠̠̉̏͠͡ͅr̬̺͚̍͛̔͒͢d̠͎̗̳͇͆̋̊͂͐"
+
+    when nr_of_code_points(word) is
+        Ok(nr) ->
+            Stdout.line!("String \"${word}\" consists of ${Num.to_str(nr)} code points.")
+
+        Err(_) ->
+            Err(Exit(1, "Failed to parse string ${word} as Utf8."))

examples/split-graphemes.roc

@@ -0,0 +1,19 @@
+app [main!] {
+    pf: platform "https://github.com/roc-lang/basic-cli/releases/download/0.19.0/bi5zubJ-_Hva9vxxPq4kNx4WHX6oFs8OP6Ad0tCYlrY.tar.br",
+    unicode: "../package/main.roc", # use release URL (ends in tar.br) for local example, see github.com/roc/unicode/releases
+}
+
+import pf.Stdout
+import unicode.Grapheme
+
+string = "🇦🇺🦘🪃"
+
+expect Grapheme.split(string) == Ok(["🇦🇺", "🦘", "🪃"])
+
+main! = |_args|
+    string
+    |> Grapheme.split
+    |> Inspect.to_str
+    |> |splitted|
+        Stdout.line!("\n\nThe string \"${string}\" has following graphemes:")?
+        Stdout.line!(splitted)