feat: DTD checker script

dtd/check_dtd.py

@@ -0,0 +1,102 @@
+#!/usr/bin/env python
+"""Check mews-0.1.dtd against the XHTML Mobile 1.2 driver and the spec.
+
+The Mews DTD is maintained by hand. It cannot be generated by stripping
+xhtml-mobile12.dtd: that file is a driver which declares no elements (the
+declarations live in the W3C and OMA module files it points to), DTD
+attribute lists only ever add, a re-declared element is an error rather
+than an override, and the rules that define Mews (the head model of
+SPEC.md 3.3, the body variant classes of 5.3, the ban on nested tables
+of 4.2, and lang and dir) do not exist in XHTML-MP to be kept.
+
+This script verifies what the driver and the spec can check:
+
+- the driver is the expected XHTML Mobile 1.2 driver, with event
+  attributes included and bidi ignored, so lang and dir are Mews
+  extensions (SPEC.md 3);
+- the DTD declares exactly the elements SPEC.md 4.1 permits;
+- the DTD keeps the driver's pre content model verbatim;
+- the DTD keeps the li value and ol start attributes the driver adds.
+
+Run with: uv run check_dtd.py
+"""
+
+import re
+import sys
+from pathlib import Path
+
+ROOT = Path(__file__).resolve().parent
+DRIVER_PATH = ROOT / "xhtml-mobile12.dtd"
+DTD_PATH = ROOT / "mews-0.1.dtd"
+
+DRIVER_FPI = "-//OMA//DTD XHTML Mobile 1.2//EN"
+
+# The Mews element set (SPEC.md 4.1).
+KEPT_ELEMENTS = (
+    "html head title body "
+    "h1 h2 h3 h4 h5 h6 "
+    "p blockquote pre address hr "
+    "em strong b i code kbd samp var q cite abbr dfn br "
+    "a "
+    "ul ol li dl dt dd "
+    "table caption tr th td "
+    "img "
+    "form fieldset label input select option optgroup textarea "
+    "meta link"
+).split()
+
+
+def entity(text, name):
+    match = re.search(r"<!ENTITY\s+%\s*" + re.escape(name) + r'\s+"([^"]*)"', text)
+    return match.group(1).strip() if match else None
+
+
+def pre_content(text):
+    match = re.search(r'<!ENTITY\s+%\s*pre\.content\s+"(.*?)"\s*>', text, re.DOTALL)
+    return re.sub(r"\s+", " ", match.group(1)) if match else None
+
+
+def attlist(text, element):
+    match = re.search(r"<!ATTLIST\s+" + element + r"\s+(.*?)\n>", text, re.DOTALL)
+    return match.group(1) if match else None
+
+
+def main():
+    driver = DRIVER_PATH.read_text(encoding="utf-8")
+    dtd = DTD_PATH.read_text(encoding="utf-8")
+    errors = []
+
+    def check(condition, message):
+        if not condition:
+            errors.append(message)
+
+    check(entity(driver, "XHTML.version") == DRIVER_FPI, "driver is not XHTML Mobile 1.2")
+    check(
+        entity(driver, "xhtml-events.module") == "INCLUDE",
+        "driver must include event attributes; Mews drops them",
+    )
+    check(
+        entity(driver, "XHTML.bidi") == "IGNORE",
+        "driver must ignore bidi, so dir is a Mews extension",
+    )
+
+    declared = set(re.findall(r"<!ELEMENT\s+([\w:-]+)", dtd))
+    expected = set(KEPT_ELEMENTS)
+    check(
+        declared == expected,
+        f"elements differ from SPEC.md 4.1: {sorted(declared ^ expected)}",
+    )
+
+    check(pre_content(dtd) == pre_content(driver), "pre content model differs from the driver's")
+    check("value" in (attlist(dtd, "li") or ""), "li is missing the driver's value attribute")
+    check("start" in (attlist(dtd, "ol") or ""), "ol is missing the driver's start attribute")
+
+    if errors:
+        for message in errors:
+            print(f"error: {message}", file=sys.stderr)
+        sys.exit(1)
+    print(f"{DTD_PATH.name} matches the driver and SPEC.md 4.1 ({len(expected)} elements)")
+
+
+if __name__ == "__main__":
+    main()