Create link-checker.roc (#7747)

* Create link-checker.roc

Create markdown link checker for glossary.md

Signed-off-by: Magdielian-code <oziomaagaecheta81@gmail.com>

* CI check + finishing touches

* improvements + test failure

* test succeeded

---------

Signed-off-by: Magdielian-code <oziomaagaecheta81@gmail.com>
Co-authored-by: Anton-4 <17049058+Anton-4@users.noreply.github.com>

.github/workflows/alwayscheck.yml

@@ -0,0 +1,42 @@
+on:
+  pull_request:
+
+name: SpellCheck+LinkCheck
+
+concurrency:
+  group: ${{ github.workflow }}-${{ github.ref }}
+  cancel-in-progress: true
+
+# Do not add permissions here! Configure them at the job level!
+permissions: {}
+
+env:
+  RUST_BACKTRACE: 1
+
+jobs:
+  spell-link-check:
+    name: spell and md link check
+    runs-on: [ubuntu-24.04]
+    timeout-minutes: 20
+    env:
+      FORCE_COLOR: 1
+    steps:
+      - uses: actions/checkout@v4
+
+      - run: cargo install typos-cli --version 1.0.11
+
+      - name: do spell check with typos-cli 1.0.11
+        run: typos
+
+      - name: setup Roc
+        run: |
+          curl -s -OL https://github.com/roc-lang/roc/releases/download/alpha3-rolling/roc-linux_x86_64-alpha3-rolling.tar.gz
+          tar -xf roc-linux_x86_64-alpha3-rolling.tar.gz
+          rm roc-linux_x86_64-alpha3-rolling.tar.gz
+          cd roc_nightly-*
+          # make roc binary available
+          echo "$(pwd)" >> $GITHUB_PATH
+
+      - run: roc version
+
+      - run: roc ./ci/glossary-link-checker.roc

.github/workflows/spellcheck.yml

@@ -1,29 +0,0 @@
-on:
-  pull_request:
-
-name: SpellCheck
-
-concurrency:
-  group: ${{ github.workflow }}-${{ github.ref }}
-  cancel-in-progress: true
-
-# Do not add permissions here! Configure them at the job level!
-permissions: {}
-
-env:
-  RUST_BACKTRACE: 1
-
-jobs:
-  spell-check:
-    name: spell check
-    runs-on: [ubuntu-22.04]
-    timeout-minutes: 10
-    env:
-      FORCE_COLOR: 1
-    steps:
-      - uses: actions/checkout@v4
-
-      - run: cargo install typos-cli --version 1.0.11
-
-      - name: do spell check with typos-cli 1.0.11
-        run: typos

ci/glossary-link-checker.roc

@@ -0,0 +1,92 @@
+app [main!] { cli: platform "https://github.com/roc-lang/basic-cli/releases/download/0.19.0/Hj-J_zxz7V9YurCSTFcFdu6cQJie4guzsPMUi5kBYUk.tar.br" }
+
+import cli.Stdout
+import cli.Stderr
+import cli.Path
+import "../Glossary.md" as glossary_as_str : Str
+
+# This script checks if all markdown links that point to files or dirs are valid for the file Glossary.md
+# TODO check links to other markdown headers as well, e.g. #tokenization
+
+main! = |_args|
+    md_links =
+        glossary_as_str
+        |> extract_md_links
+
+    full_check_res =
+        md_links
+        |> List.for_each_try!(
+            |link_str|
+                check_res = check_link!(link_str)
+
+                when check_res is
+                    Ok(_) ->
+                        # Stdout.line!("Check successful for $(link_str)")?
+                        Ok({})
+                    Err(BadLink(bad_link_str)) ->
+                        Stderr.line!("I could not find the path: '$(bad_link_str)'. If you recently modified this path, please update '$(bad_link_str)' in Glossary.md at the root of the repo.\n")?
+                        Err(LinkCheckFailed)
+        )
+
+    when full_check_res is
+            Ok(_) ->
+                Stdout.line!("Successfuly checked $(List.len(md_links) |> Num.to_str) links.")
+            Err(_) ->
+                Err(Exit(1, "Link check failed. Please check the above errors."))
+
+expect
+    md_text = "foo [bar](src/baz.md)..."
+    links = extract_md_links(md_text)
+
+    links == ["src/baz.md"]
+
+extract_md_links : Str -> List Str
+extract_md_links = |md_content|
+    split_list = Str.split_on(md_content, "](")
+
+    links =
+        split_list
+        |> List.drop_first(1)
+        |> List.map(
+            |link|
+                res = Str.split_on(link, ")") |> List.first()
+                when res is
+                    Ok(str) -> str
+                    Err(_) -> ""
+        )
+
+    links
+
+expect
+    md_text = "foo [bar](src/baz.md)...[zod](boz.md)"
+    links = extract_md_links(md_text)
+
+    links == ["src/baz.md", "boz.md"]
+
+expect
+    md_text =
+        """
+        foo [bar](src/baz.md)...
+        hey asjkl [zod](boz.md) asmd
+        [bam](crates/check/can_solo/src/desugar.rs)
+        """
+    links = extract_md_links(md_text)
+
+    links == ["src/baz.md", "boz.md", "crates/check/can_solo/src/desugar.rs"]
+
+
+check_link! : Str => Result {} [BadLink Str]
+check_link! = |link_str|
+    if Str.starts_with(link_str, "http") || Str.starts_with(link_str, "#") then
+        # TODO check links to other markdown headers as well, e.g. #tokenization
+        Ok({})
+    else
+        path = Path.from_str(link_str)
+
+        when Path.type!(path) is
+            Ok(_) ->
+                Ok({})
+            Err(_) ->
+                Err(BadLink(link_str))
+        
+