116 lines
5.4 KiB
Python
116 lines
5.4 KiB
Python
#!/usr/bin/env python3
|
|
"""P0-таксономия Комбайна: группы, дубли, пробелы, кандидаты SIP/H.323."""
|
|
import json
|
|
import re
|
|
import yaml
|
|
from collections import Counter, defaultdict
|
|
|
|
CAT = "/opt/work-hub/compat/catalog.yaml"
|
|
VAL = "/opt/work-hub/compat/values.yaml"
|
|
OUT_JSON = "/root/buryan-infra/ops/combine/taxonomy-2026-11-18.json"
|
|
OUT_MD = "/root/buryan-infra/ops/combine/TAXONOMY-2026-11-18.md"
|
|
|
|
cat = yaml.safe_load(open(CAT, encoding="utf-8"))
|
|
catalog = cat.get("devices", [])
|
|
values = yaml.safe_load(open(VAL, encoding="utf-8"))
|
|
vlist = values if isinstance(values, list) else values.get("devices", [])
|
|
vdev = [x.get("device") for x in vlist if isinstance(x, dict) and x.get("device")]
|
|
|
|
cnames = {d.get("name", "") for d in catalog}
|
|
|
|
|
|
def group(name, role=""):
|
|
n = (name or "").lower()
|
|
r = (role or "").lower()
|
|
if "meetingbar" in n:
|
|
return "B_video_bars"
|
|
if "meetingboard" in n or "pvb" in n:
|
|
return "B_video_bars_boards"
|
|
if any(x in n for x in ("meetingeye", "mcore", "mvc", "pvt", "teams rooms", "mtr-w")):
|
|
return "A_integration_terminals"
|
|
if any(x in n for x in ("sip-t", "t88", "w70", "wh6", "uh38", "bh72", "ehs", "dect")):
|
|
return "T_telephony"
|
|
if any(x in n for x in ("uvc", "smartvision", "mb-cam", "mb-12x", "cam50", "byod")):
|
|
return "D_usb_byod"
|
|
if any(x in n for x in ("vcm", "cm20", "cm50", "mspeaker", "pa20", "cp50", "cp70", "cp90", "cp96", "cpe", "cpw", "cs10", "ap08", "mspeech", "audio")):
|
|
return "E_audio"
|
|
if any(x in n for x in ("ctp", "mtouch", "touch")):
|
|
return "E_control"
|
|
if any(x in n for x in ("rch", "ax83", "ax86", "switch", "poe")):
|
|
return "E_infra"
|
|
if any(x in n for x in ("display", "etv", "телевизор", "проектор", "tv", "projector")):
|
|
return "E_display"
|
|
if any(x in n for x in ("mount", "mtower", "mb-fs", "кронштейн")):
|
|
return "E_mount"
|
|
if any(x in n for x in ("yps", "power", "розетка", "заряд", "charger")):
|
|
return "E_power"
|
|
if any(x in n for x in ("room designer", "roomconnect", "software", "ymcs")):
|
|
return "E_software"
|
|
return "E_other"
|
|
|
|
|
|
def clean(name):
|
|
n = (name or "").strip()
|
|
n = re.sub(r"\s*\((?:\d+|only.*?|included.*?)\)\s*$", "", n, flags=re.I)
|
|
n = re.sub(r"\s*\[\d+\]\s*$", "", n)
|
|
return n.strip()
|
|
|
|
|
|
# duplicates by cleaned name
|
|
buckets = defaultdict(list)
|
|
for d in catalog:
|
|
buckets[clean(d.get("name", ""))].append(d.get("name", ""))
|
|
dups = {k: v for k, v in buckets.items() if len(v) > 1}
|
|
|
|
# merged/suspicious names
|
|
suspicious = [n for n in cnames if any(x in n for x in ("/", "&", ";", "MeetingBar (", "Meetingboard (", "Yealink ("))]
|
|
|
|
groups = defaultdict(list)
|
|
for d in catalog:
|
|
groups[group(d.get("name"), d.get("role"))].append({"name": d.get("name"), "role": d.get("role"), "source": "catalog"})
|
|
for name in vdev:
|
|
if name in cnames:
|
|
continue
|
|
groups[group(name)].append({"name": name, "role": None, "source": "values-only"})
|
|
|
|
# candidates for SIP/H.323: explicit protocols absent, so only name-based candidates
|
|
candidates = []
|
|
for d in catalog:
|
|
n = (d.get("name") or "").lower()
|
|
if any(x in n for x in ("meetingeye", "meetingbar a40", "meetingbar a50", "pvb", "pvt")):
|
|
candidates.append({"name": d.get("name"), "reason": "по имени/линейке; protocols в данных отсутствует", "verified": False})
|
|
|
|
result = {
|
|
"generated": "2026-11-18",
|
|
"catalog_count": len(catalog),
|
|
"values_count": len(vdev),
|
|
"values_only": sorted([v for v in vdev if v not in cnames]),
|
|
"duplicates": dups,
|
|
"suspicious": sorted(suspicious),
|
|
"group_counts": {k: len(v) for k, v in groups.items()},
|
|
"groups": {k: v for k, v in groups.items()},
|
|
"protocols_field": "absent",
|
|
"sip_h323_candidates": candidates,
|
|
"missing_fields": ["vendor", "domain", "line", "protocols", "form_factor", "integrated", "limits", "source", "version"],
|
|
}
|
|
json.dump(result, open(OUT_JSON, "w", encoding="utf-8"), ensure_ascii=False, indent=2)
|
|
|
|
lines = ["# P0-таксономия «Комбайна» — 18.11.2026", "",
|
|
"Каталог: **%d** записей. Values: **%d** устройств. Только в values: **%d**." % (len(catalog), len(vdev), len(result["values_only"])), "",
|
|
"## Группы", "| Группа | Кол-во |", "|---|---|"]
|
|
for k, n in sorted(result["group_counts"].items(), key=lambda x: -x[1]):
|
|
lines.append("| %s | %d |" % (k, n))
|
|
lines += ["", "## Дубли (по очищенному имени)", "| Имя | Варианты |", "|---|---|"]
|
|
for k, v in sorted(dups.items()):
|
|
lines.append("| %s | %s |" % (k, " · ".join(v)))
|
|
lines += ["", "## Подозрительные слитые записи", *["- " + x for x in sorted(suspicious)]]
|
|
lines += ["", "## SIP/H.323: кандидаты (не подтверждены)", "| Модель | Причина |", "|---|---|"]
|
|
for c in candidates:
|
|
lines.append("| %s | %s |" % (c["name"], c["reason"]))
|
|
lines += ["", "## Устройства только в values.yaml (первые 60)", *["- " + x for x in result["values_only"][:60]]]
|
|
lines += ["", "## Чего не хватает в схеме", *["- " + x for x in result["missing_fields"]]]
|
|
open(OUT_MD, "w", encoding="utf-8").write("\n".join(lines) + "\n")
|
|
print("saved", OUT_JSON, OUT_MD)
|
|
print("groups", result["group_counts"])
|
|
print("dups", len(dups), "values-only", len(result["values_only"]), "sip candidates", len(candidates))
|