1*0e16a31dSJim Jagielski#************************************************************** 2*0e16a31dSJim Jagielski# 3*0e16a31dSJim Jagielski# Licensed to the Apache Software Foundation (ASF) under one 4*0e16a31dSJim Jagielski# or more contributor license agreements. See the NOTICE file 5*0e16a31dSJim Jagielski# distributed with this work for additional information 6*0e16a31dSJim Jagielski# regarding copyright ownership. The ASF licenses this file 7*0e16a31dSJim Jagielski# to you under the Apache License, Version 2.0 (the 8*0e16a31dSJim Jagielski# "License"); you may not use this file except in compliance 9*0e16a31dSJim Jagielski# with the License. You may obtain a copy of the License at 10*0e16a31dSJim Jagielski# 11*0e16a31dSJim Jagielski# http://www.apache.org/licenses/LICENSE-2.0 12*0e16a31dSJim Jagielski# 13*0e16a31dSJim Jagielski# Unless required by applicable law or agreed to in writing, 14*0e16a31dSJim Jagielski# software distributed under the License is distributed on an 15*0e16a31dSJim Jagielski# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY 16*0e16a31dSJim Jagielski# KIND, either express or implied. See the License for the 17*0e16a31dSJim Jagielski# specific language governing permissions and limitations 18*0e16a31dSJim Jagielski# under the License. 19*0e16a31dSJim Jagielski# 20*0e16a31dSJim Jagielski#************************************************************** 21*0e16a31dSJim Jagielski 22*0e16a31dSJim Jagielski# usage: report.py <val.json> <kinds.json> [12|14|12x] -- groups errors by normalized message 23*0e16a31dSJim Jagielskiimport sys, re, json, collections 24*0e16a31dSJim Jagielski 25*0e16a31dSJim Jagielski 26*0e16a31dSJim Jagielskidef norm(m): 27*0e16a31dSJim Jagielski m = re.sub(r'; expected .*', '', m) 28*0e16a31dSJim Jagielski m = re.sub(r' \(with xmlns.*', '', m) 29*0e16a31dSJim Jagielski return re.sub(r'value of attribute "([^"]+)" is invalid.*', r'value of attribute "\1" is invalid', m) 30*0e16a31dSJim Jagielski 31*0e16a31dSJim Jagielski 32*0e16a31dSJim Jagielskival, out = sys.argv[1], sys.argv[2] 33*0e16a31dSJim Jagielskiver = sys.argv[3] if len(sys.argv) > 3 else "12" 34*0e16a31dSJim Jagielskid = json.load(open(val)) 35*0e16a31dSJim Jagielskikinds = collections.defaultdict(lambda: {"errors": 0, "packages": set()}) 36*0e16a31dSJim Jagielskifor f, line, col, msg in d["errors"][ver]: 37*0e16a31dSJim Jagielski k = kinds[norm(msg) if f != "?" else "UNPARSED: " + msg] 38*0e16a31dSJim Jagielski k["errors"] += 1; k["packages"].add(f.split("/")[0]) 39*0e16a31dSJim Jagielskires = {k: {"errors": v["errors"], "packages": sorted(v["packages"])} for k, v in kinds.items()} 40*0e16a31dSJim Jagielskijson.dump(res, open(out, "w"), indent=1, sort_keys=True) 41*0e16a31dSJim Jagielskitot = sum(v["errors"] for v in res.values()) 42*0e16a31dSJim Jagielskiprint(f"ODF {ver[0]}.{ver[1]}{' extended, foreign content removed' if ver.endswith('x') else ''}: {len(res)} kinds, {tot} errors, {len({p for v in res.values() for p in v['packages']})} packages with errors") 43*0e16a31dSJim Jagielskifor k, v in sorted(res.items(), key=lambda kv: -kv[1]["errors"]): 44*0e16a31dSJim Jagielski print(f" {v['errors']:5d} {len(v['packages']):4d} {k[:120]}") 45