fix(formatter): fixed speed on formatter

closes #601

.github/workflows/demo.yml

@@ -11,24 +11,24 @@ jobs:
       matrix:
         tool: [django, wagtail, pallets, apostrophe, bolt]
         include:
-          - tool: django
-            repo: https://github.com/django/django
-            profile: django
-          - tool: wagtail
-            repo: https://github.com/wagtail/wagtail
-            profile: django
-          - tool: pallets
-            repo: https://github.com/pallets/website
-            profile: jinja
-          - tool: apostrophe
-            repo: https://github.com/apostrophecms/apostrophe
-            profile: nunjucks
-          - tool: bolt
-            repo: https://github.com/bolt/core
-            profile: nunjucks
-          - tool: sylius
-            repo: https://github.com/Sylius/Sylius
-            profile: nunjucks
+        - tool: django
+          repo: https://github.com/django/django
+          profile: django
+        - tool: wagtail
+          repo: https://github.com/wagtail/wagtail
+          profile: django
+        - tool: pallets
+          repo: https://github.com/pallets/website
+          profile: jinja
+        - tool: apostrophe
+          repo: https://github.com/apostrophecms/apostrophe
+          profile: nunjucks
+        - tool: bolt
+          repo: https://github.com/bolt/core
+          profile: nunjucks
+        - tool: sylius
+          repo: https://github.com/Sylius/Sylius
+          profile: nunjucks
     name: format
     runs-on: ubuntu-latest
     steps:
@@ -52,23 +52,23 @@ jobs:
         mkdir "${{ matrix.tool }}_raw"; mkdir "${{ matrix.tool }}_source"
         cd "${{ matrix.tool }}_raw"; git clone --depth 1 "${{ matrix.repo }}" .; cd ..
         find "${{ matrix.tool }}_raw" -iname '*.html*' -exec mv '{}' "${{ matrix.tool }}_source/" \;
-  
+
     - name: publish source
       uses: peaceiris/actions-gh-pages@v3
       with:
         github_token: ${{ secrets.GITHUB_TOKEN }}
-        publish_dir: "./${{ matrix.tool }}_source"
-        publish_branch: "${{ matrix.tool }}-source"
+        publish_dir: ./${{ matrix.tool }}_source
+        publish_branch: ${{ matrix.tool }}-source
         enable_jekyll: true
         force_orphan: true
-    
+
     - name: cleanup source
       run: rm -r "${{ matrix.tool }}_source"; rm -r "${{ matrix.tool }}_raw"
 
     - uses: actions/checkout@v3
       with:
-        path: "${{ matrix.tool }}-source"
-        ref: "${{ matrix.tool }}-source"
+        path: ${{ matrix.tool }}-source
+        ref: ${{ matrix.tool }}-source
 
     - name: setup branch
       run: |
@@ -77,17 +77,16 @@ jobs:
         git config user.email "$(git log -n 1 --pretty=format:%ae)"
         git branch --unset-upstream
         git push --force origin "${{ matrix.tool }}-source":"${{ matrix.tool }}-djlint"
-        
+
     - name: format
       run: |
         cd src
         poetry run python -m djlint "../${{ matrix.tool }}-source" --profile=${{ matrix.profile }} --reformat --warn --quiet
-        
+
     - name: publish foramtted
       run: |
         cd "${{ matrix.tool }}-source"
         export DJ_VERSION=$(poetry version --short)
         git add .
         git commit -m "formatted with djLint $DJ_VERSION"
         git push origin "${{ matrix.tool }}-source":"${{ matrix.tool }}-djlint"
- 

src/djlint/formatter/compress.py

@@ -36,13 +36,14 @@ def _clean_tag(match: re.Match) -> str:
 
         open_braket = match.group(1)
         tag = _fix_case(match.group(2))
+
         attributes = (
             (" " + " ".join(x.strip() for x in match.group(3).strip().splitlines()))
             if match.group(3)
             else ""
         )
         close_braket = (
-            match.group(7) if "/" not in match.group(7) else f" {match.group(7)}"
+            match.group(4) if "/" not in match.group(4) else f" {match.group(4)}"
         )
 
         return f"{open_braket}{tag}{attributes}{close_braket}"

src/djlint/settings.py

@@ -560,18 +560,13 @@ def __init__(
         """
         )
 
-        self.html_tag_regex = (
-            r"""
+        self.html_tag_regex = r"""
             (</?(?:!(?!--))?) # an opening bracket (< or </ or <!), but not a comment
-            ([^\s>!]+\b) # a tag name
-            # ((?:\s*?(?:\"[^\"]*\"|'[^']*'|{{(?:(?!}}).)*}}|{%(?:(?!%}).)*%}|[^'\">{}/\s]|/(?!>)))+)? # any attributes
-            ((?:\s*?"""
-            + self.attribute_pattern
-            + r""")+)?
+            ([^\s>!\[]+\b) # a tag name
+            ((?:\s*?(?:\"[^\"]*\"|'[^']*'|{{(?:(?!}}).)*}}|{%(?:(?!%}).)*%}|[^'\">{}/\s]|/(?!>)))+)? # any attributes
             \s*? # potentially some whitespace
             (/?>) # a closing braket (/> or >)
         """
-        )
 
         self.attribute_style_pattern: str = r"^(.*?)(style=)([\"|'])(([^\"']+?;)+?)\3"
 

tests/test_html/test_doctype_declarations.py

@@ -33,8 +33,7 @@
             "</html>\n"
         ),
         (
-            '<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Frameset//EN"\n'
-            '"http://www.w3.org/TR/html4/frameset.dtd">\n'
+            '<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Frameset//EN" "http://www.w3.org/TR/html4/frameset.dtd">\n'
             "<html>\n"
             "    <head>\n"
             "        <title>An HTML standard template</title>\n"
@@ -62,8 +61,7 @@
             "</html>\n"
         ),
         (
-            '<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN"\n'
-            '"http://www.w3.org/TR/html4/strict.dtd">\n'
+            '<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd">\n'
             "<html>\n"
             "    <head>\n"
             "        <title>An HTML standard template</title>\n"
@@ -91,8 +89,7 @@
             "</html>\n"
         ),
         (
-            '<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"\n'
-            '"http://www.w3.org/TR/html4/loose.dtd">\n'
+            '<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">\n'
             "<html>\n"
             "    <head>\n"
             "        <title>An HTML standard template</title>\n"