first commit

scripts/check-wbxml-names.py

@@ -0,0 +1,31 @@
+#!/usr/bin/env python3
+"""Verify that every literal "Namespace:Tag" used in the sources exists in the WBXML code pages.
+
+Element::new() panics on unknown names at runtime; this catches typos at build time.
+"""
+import pathlib
+import re
+import sys
+
+root = pathlib.Path(__file__).resolve().parent.parent
+cp = (root / "src/wbxml/codepages.rs").read_text()
+namespaces = re.findall(r'"([^"]+)"', re.search(r"NAMESPACES: &\[&str\] = &\[(.*?)\];", cp, re.S).group(1))
+body = cp[cp.index("pub const PAGES"):]
+known = set()
+for i, chunk in enumerate(re.split(r"\n\s*// \d+ \w+\n", body)[1:]):
+    for tag in re.findall(r'\(0x[0-9A-Fa-f]+,\s*"([^"]+)"\)', chunk):
+        known.add(f"{namespaces[i]}:{tag}")
+
+pattern = re.compile(r'"((?:%s):[A-Za-z0-9]+)"' % "|".join(re.escape(n) for n in namespaces))
+bad = []
+for f in (root / "src").rglob("*.rs"):
+    if f.name == "codepages.rs":
+        continue
+    for n, line in enumerate(f.read_text().splitlines(), 1):
+        for m in pattern.finditer(line):
+            if m.group(1) not in known:
+                bad.append(f"{f.relative_to(root)}:{n}: {m.group(1)}")
+if bad:
+    print("Unknown WBXML tag names:\n  " + "\n  ".join(bad))
+    sys.exit(1)
+print(f"WBXML names: OK ({len(known)} known tags)")