Initial commit.

gen-manufacturers.py

@@ -0,0 +1,150 @@
+#!/usr/bin/env python3
+# Regenerate manufacturers.tsv from the Matter/CHIP manufacturer ID map.
+#
+# The TSV has two columns, "<decimal vendor id>\t<name>", sorted by vendor id. It
+# is consumed by matter-decode.sh and is mirrored by the embedded <map> in
+# matter-qr-decoder.html. It is derived from upstream project-chip/connectedhomeip's
+# manufacturers.xml -- refresh from there rather than hand-editing.
+#
+# https://raw.githubusercontent.com/project-chip/connectedhomeip/refs/heads/master/src/app/zap-templates/zcl/data-model/manufacturers.xml
+#
+# By default the script pulls the upstream file over the network (via curl) and
+# writes manufacturers.tsv:
+#
+#     python3 gen-manufacturers.py
+#
+# A GitHub blob/ (HTML) URL is auto-rewritten to its raw.githubusercontent.com
+# form, so the URL from the repo's "From ..." comment works as-is.
+#
+# Override the source with a positional argument that is a URL or a local file:
+#
+#     python3 gen-manufacturers.py <url-or-file>
+#     python3 gen-manufacturers.py matter-qr-decoder.html     # offline: embedded block
+#     python3 gen-manufacturers.py -o out.tsv <source>
+#
+# The default network fetch uses curl (falling back to urllib if curl is
+# missing); a .html source and a plain XML file need no network. Requires only
+# the Python 3 standard library.
+
+import argparse
+import re
+import shutil
+import subprocess
+import sys
+import urllib.request
+import xml.etree.ElementTree as ET
+from pathlib import Path
+
+SCRIPT_DIR = Path(__file__).resolve().parent
+
+# Pinned upstream source -- the same commit the HTML block was generated from, so
+# a fresh pull reproduces the committed TSV byte-for-byte until upstream moves.
+UPSTREAM_RAW = ("https://raw.githubusercontent.com/project-chip/connectedhomeip/"
+                "7986ee3d82b26403fc2d0f70bdfbc9e8ff650df1/"
+                "src/app/zap-templates/zcl/data-model/manufacturers.xml")
+
+
+def by_local_tag(root, local_name):
+    for el in root.iter():
+        tag = el.tag.rsplit("}", 1)[-1] if "}" in el.tag else el.tag
+        if tag == local_name:
+            yield el
+
+
+def mappings_from_xml_text(xml_text):
+    m = re.search(r"<map\b.*?</map>", xml_text, re.DOTALL | re.IGNORECASE)
+    frag = m.group(0) if m else xml_text
+    root = ET.fromstring(frag)
+
+    rows = []
+    for el in by_local_tag(root, "mapping"):
+        code = el.get("code")
+        name = el.get("translation", "")
+        if code is None:
+            continue
+        # Upstream codes are hex with a 0x prefix ("0x1026"); strip it.
+        hexpart = code[2:] if code[:2].lower() == "0x" else code
+        try:
+            dec = int(hexpart, 16)
+        except ValueError:
+            sys.stderr.write(f"warning: skipping unparseable code {code!r}\n")
+            continue
+        rows.append((dec, name))
+    return rows
+
+
+def to_raw_url(url):
+    m = re.match(r"https://github\.com/([^/]+)/([^/]+)/blob/(.+)", url)
+    if m:
+        owner, repo, path = m.groups()
+        return f"https://raw.githubusercontent.com/{owner}/{repo}/{path}"
+    return url
+
+
+def fetch_url(url):
+    url = to_raw_url(url)
+    if shutil.which("curl"):
+        try:
+            out = subprocess.run(["curl", "-fsSL", url],
+                                 capture_output=True, check=True).stdout
+            return out.decode("utf-8")
+        except subprocess.CalledProcessError as e:
+            sys.stderr.write(e.stderr.decode("utf-8", "replace"))
+            sys.exit(f"error: failed to fetch {url}")
+    req = urllib.request.Request(url, headers={"User-Agent": "gen-manufacturers.py"})
+    try:
+        with urllib.request.urlopen(req) as resp:
+            return resp.read().decode("utf-8")
+    except Exception as e:
+        sys.exit(f"error: failed to fetch {url}: {e}")
+
+
+def read_source(source):
+    if source.startswith(("http://", "https://")):
+        return fetch_url(source)
+    path = Path(source)
+    if not path.exists():
+        sys.exit(f"error: no such file: {path}")
+    if path.suffix.lower() == ".html":
+        html = path.read_text(encoding="utf-8")
+        m = re.search(r'<script\s+type="text/xml"[^>]*>(.*?)</script>',
+                      html, re.DOTALL | re.IGNORECASE)
+        if not m:
+            sys.exit('error: no <script type="text/xml"> block in ' + str(path))
+        return m.group(1)
+    return path.read_text(encoding="utf-8")
+
+
+def write_tsv(path, rows):
+    rows.sort(key=lambda r: r[0])
+    with open(path, "w", encoding="utf-8", newline="") as f:
+        for dec, name in rows:
+            f.write(f"{dec}\t{name}\n")
+
+
+def main():
+    ap = argparse.ArgumentParser(
+        description="Regenerate manufacturers.tsv from the Matter/CHIP "
+                    "manufacturer ID map (fetches upstream by default).")
+    ap.add_argument(
+        "source", nargs="?", default=UPSTREAM_RAW,
+        help="a URL or local file (default: pinned upstream manufacturers.xml). "
+             "A GitHub blob/ URL is auto-converted to its raw form; a .html path "
+             'uses its embedded <script type="text/xml"> block.')
+    ap.add_argument(
+        "-o", "--output", default=str(SCRIPT_DIR / "manufacturers.tsv"),
+        help="output path (default: manufacturers.tsv next to this script)")
+    args = ap.parse_args()
+
+    xml_text = read_source(args.source)
+    rows = mappings_from_xml_text(xml_text)
+    if not rows:
+        sys.exit("error: no <mapping> entries found in the source")
+
+    out = Path(args.output)
+    write_tsv(out, rows)
+    sys.stderr.write(f"wrote {len(rows)} mappings to {out} (from {args.source})\n")
+
+
+if __name__ == "__main__":
+    main()