feat(mods): Add external RST renderer

gitea/mods/rst/files/default/grst

@@ -0,0 +1,128 @@
+#!/usr/bin/python3
+
+import subprocess
+import sys
+
+from docutils import nodes, core, io
+from docutils.parsers.rst import directives, Directive
+
+
+class Chroma(Directive):
+    """Source code syntax highlighting."""
+
+    required_arguments = 1
+    optional_arguments = 0
+    final_argument_whitespace = True
+    option_spec = {}
+    has_content = True
+
+    def run(self):
+        self.assert_has_content()
+        data = "\n".join(self.content)
+        try:
+            parsed = subprocess.run(
+                ["{{ gitea.lookup.mod_rst.chroma_bin }}", "--html", "--html-only", "-l", self.arguments[0]],
+                text=True,
+                input=data,
+                capture_output=True,
+                check=True,
+            )
+        except subprocess.CalledProcessError:
+            parsed = subprocess.run(
+                ["{{ gitea.lookup.mod_rst.chroma_bin }}", "--html", "--html-only"],
+                text=True,
+                input=data,
+                capture_output=True,
+                check=True,
+            )
+
+        return [nodes.raw("", parsed.stdout, format="html")]
+
+
+def html_parts(
+    input_string,
+    source_path=None,
+    destination_path=None,
+    input_encoding="utf8",
+    doctitle=True,
+    initial_header_level=1,
+):
+    """
+    Given an input string, returns a dictionary of HTML document parts.
+
+    Dictionary keys are the names of parts, and values are Unicode strings;
+    encoding is up to the client.
+
+    Parameters:
+
+    - `input_string`: A multi-line text string; required.
+    - `source_path`: Path to the source file or object.  Optional, but useful
+      for diagnostic output (system messages).
+    - `destination_path`: Path to the file or object which will receive the
+      output; optional.  Used for determining relative paths (stylesheets,
+      source links, etc.).
+    - `input_encoding`: The encoding of `input_string`.  If it is an encoded
+      8-bit string, provide the correct encoding.  If it is a Unicode string,
+      use "unicode", the default.
+    - `doctitle`: Disable the promotion of a lone top-level section title to
+      document title (and subsequent section title to document subtitle
+      promotion); enabled by default.
+    - `initial_header_level`: The initial level for header elements (e.g. 1
+      for "<h1>").
+    """
+    overrides = {
+        "input_encoding": input_encoding,
+        "doctitle_xform": doctitle,
+        "initial_header_level": initial_header_level,
+        "report_level": 4,
+    }
+
+    parts = core.publish_parts(
+        source=input_string,
+        source_path=source_path,
+        destination_path=destination_path,
+        writer_name="html",
+        settings_overrides=overrides,
+    )
+    return parts
+
+
+def html_body(
+    input_string,
+    source_path=None,
+    destination_path=None,
+    input_encoding="utf8",
+    output_encoding="unicode",
+    doctitle=True,
+    initial_header_level=1,
+):
+    """
+    Given an input string, returns an HTML fragment as a string.
+
+    The return value is the contents of the <body> element.
+
+    Parameters (see `html_parts()` for the remainder):
+
+    - `output_encoding`: The desired encoding of the output.  If a Unicode
+      string is desired, use the default value of "unicode" .
+    """
+    parts = html_parts(
+        input_string=input_string,
+        source_path=source_path,
+        destination_path=destination_path,
+        input_encoding=input_encoding,
+        doctitle=doctitle,
+        initial_header_level=initial_header_level,
+    )
+    fragment = parts["html_body"]
+    if output_encoding != "unicode":
+        fragment = fragment.encode(output_encoding)
+    return fragment
+
+
+if __name__ == "__main__":
+    directives.register_directive("code-block", Chroma)
+
+    with open(sys.argv[1], "r") as f:
+        source = "".join(f.readlines())
+    print(html_body(source.encode()))