wartung: Groesse je Quantisierung, GitHub-Antworten 15 min gemerkt, Hermes-Version auch bei haengendem Abruf
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
Hitonabi
co-authored by
Claude Opus 5.5
Claude Opus 5.5
parent
0e961f112b
commit
ee2bb9d03c
+13
-6
@@ -17,10 +17,12 @@ def normalize_repo(s: str) -> str:
|
||||
return s.strip("/")
|
||||
|
||||
|
||||
def list_quants(repo: str) -> list[str]:
|
||||
"""Verfügbare Quant-Stufen eines Repos (aus den GGUF-Dateinamen, ohne mmproj)."""
|
||||
def list_quants(repo: str) -> list[dict]:
|
||||
"""Verfügbare Quant-Stufen eines Repos mit Downloadgröße (alle Teile + mmproj), ohne mmproj
|
||||
als eigene Stufe. Ein Aufruf der Hugging-Face-API für alle Stufen."""
|
||||
baum = _tree(repo)
|
||||
quants: set[str] = set()
|
||||
for e in _tree(repo):
|
||||
for e in baum:
|
||||
p = str(e.get("path", ""))
|
||||
if p.lower().endswith(".gguf") and "mmproj" not in p.lower():
|
||||
m = re.search(r"(I?Q\d[\w]*|F16|BF16|FP16|F32)", p, re.IGNORECASE)
|
||||
@@ -28,7 +30,8 @@ def list_quants(repo: str) -> list[str]:
|
||||
quants.add(m.group(1).upper())
|
||||
# gängige Reihenfolge zuerst
|
||||
order = {"Q4_K_M": 0, "Q4_K_S": 1, "Q5_K_M": 2, "Q6_K": 3, "Q8_0": 4, "Q3_K_M": 5, "Q2_K": 6}
|
||||
return sorted(quants, key=lambda q: (order.get(q, 99), q))
|
||||
return [{"quant": q, "total_bytes": auswahl(baum, q)["total_bytes"]}
|
||||
for q in sorted(quants, key=lambda q: (order.get(q, 99), q))]
|
||||
|
||||
|
||||
def search(q: str = "", limit: int = 24) -> list[dict]:
|
||||
@@ -68,12 +71,16 @@ def _size(entry: dict) -> int:
|
||||
|
||||
|
||||
def resolve_gguf(repo: str, quant: str = "Q4_K_M") -> dict:
|
||||
"""Beste GGUF-Auswahl eines Repos für einen Quant. Behandelt Split-GGUFs
|
||||
"""Beste GGUF-Auswahl eines Repos für einen Quant (siehe auswahl)."""
|
||||
return auswahl(_tree(repo), quant)
|
||||
|
||||
|
||||
def auswahl(tree: list[dict], quant: str = "Q4_K_M") -> dict:
|
||||
"""GGUF-Auswahl aus dem Dateibaum eines Repos für einen Quant. Behandelt Split-GGUFs
|
||||
(-00001-of-000NN) als Gruppe. Liefert die Datei-/Pattern-Infos für den Download.
|
||||
|
||||
Rückgabe: {files:[paths], first:path, total_bytes:int, mmproj:path|None, split:bool}
|
||||
"""
|
||||
tree = _tree(repo)
|
||||
ggufs = [e for e in tree if str(e.get("path", "")).lower().endswith(".gguf")]
|
||||
q = quant.lower()
|
||||
# mmproj separat (Vision-Projektor)
|
||||
|
||||
@@ -50,10 +50,26 @@ _engine_cache = {"ts": 0.0, "avail": False}
|
||||
_ENGINE_ASSET_RX = re.compile(r"ubuntu-vulkan-x64\.tar\.gz$", re.IGNORECASE)
|
||||
|
||||
|
||||
# GitHub erlaubt ohne Anmeldung 60 Anfragen pro Stunde. Jeder Blick auf die Update-Details fragte
|
||||
# bisher neu; jetzt gilt eine Antwort 15 Minuten. Fehler (auch das Anfragelimit) werden nicht gemerkt.
|
||||
GITHUB_CACHE_S = int(os.environ.get("MC_GITHUB_CACHE_S", "900"))
|
||||
_github_cache: dict[str, tuple[float, object]] = {}
|
||||
|
||||
|
||||
def _github_json(pfad: str, timeout: float = 8) -> object:
|
||||
jetzt = time.time()
|
||||
if (eintrag := _github_cache.get(pfad)) and jetzt - eintrag[0] < GITHUB_CACHE_S:
|
||||
return eintrag[1]
|
||||
r = httpx.get(f"https://api.github.com/{pfad}", timeout=timeout, headers={"User-Agent": "MissionControl2"})
|
||||
r.raise_for_status()
|
||||
daten = r.json()
|
||||
_github_cache[pfad] = (jetzt, daten)
|
||||
return daten
|
||||
|
||||
|
||||
def _latest_engine_asset_release() -> dict | None:
|
||||
try:
|
||||
rels = httpx.get(f"https://api.github.com/repos/{ENGINE_REPO}/releases?per_page=15",
|
||||
timeout=8, headers={"User-Agent": "MissionControl2"}).json()
|
||||
rels = _github_json(f"repos/{ENGINE_REPO}/releases?per_page=15")
|
||||
for rel in (rels if isinstance(rels, list) else []):
|
||||
if any(_ENGINE_ASSET_RX.search(a.get("name", "")) for a in (rel.get("assets") or [])):
|
||||
return rel
|
||||
@@ -160,9 +176,8 @@ def _swap_update_available() -> bool:
|
||||
avail = False
|
||||
fehler = None
|
||||
try:
|
||||
rel = httpx.get(f"https://api.github.com/repos/{SWAP_REPO}/releases/latest",
|
||||
timeout=6, headers={"User-Agent": "MissionControl2"}).json()
|
||||
tag = str(rel.get("tag_name", ""))
|
||||
rel = _github_json(f"repos/{SWAP_REPO}/releases/latest", timeout=6)
|
||||
tag = str(rel.get("tag_name", "")) if isinstance(rel, dict) else ""
|
||||
latest = int(m.group(1)) if (m := re.search(r"(\d{2,})", tag)) else None
|
||||
installed = _installed_swap_version()
|
||||
if latest is not None and installed is not None:
|
||||
@@ -346,8 +361,7 @@ def swap_update_details() -> dict:
|
||||
"latest_build": None, "latest_tag": None, "name": None,
|
||||
"url": None, "body": None}
|
||||
try:
|
||||
rel = httpx.get(f"https://api.github.com/repos/{SWAP_REPO}/releases/latest",
|
||||
timeout=8, headers={"User-Agent": "MissionControl2"}).json()
|
||||
rel = _github_json(f"repos/{SWAP_REPO}/releases/latest")
|
||||
tag = str(rel.get("tag_name", ""))
|
||||
info["latest_tag"] = tag
|
||||
info["latest_build"] = int(m.group(1)) if (m := re.search(r"(\d{2,})", tag)) else None
|
||||
@@ -446,8 +460,8 @@ def _summarize_release(kind: str, key: str, context: str, changes: str) -> dict:
|
||||
def _summarize_hermes_commits(commits: list[dict]) -> dict:
|
||||
subjects = "\n".join(f"- {c['subject']}" for c in commits[:100])
|
||||
context = ("Es geht um ein Update des Hermes-Agenten (das Gehirn/Werkzeug-System der Box auf "
|
||||
"Strix Halo; genutzt werden: api_server/Gateway, memory-provider-Plugin 'mc2-memory', "
|
||||
"terminal-/web-Tools, approvals, cron).")
|
||||
"Strix Halo; genutzt werden: api_server/Gateway, Telegram, eigenes Gedächtnis, "
|
||||
"terminal-/web-Tools, Plugins, cron).")
|
||||
return _summarize_release("hermes", commits[0]["hash"] if commits else "", context, subjects)
|
||||
|
||||
|
||||
@@ -469,6 +483,8 @@ def hermes_update_details() -> dict:
|
||||
info["error"] = "Hermes-Agent-Repo nicht gefunden."
|
||||
return info
|
||||
path = git["path"]
|
||||
# Vor dem Abruf: scheitert der (Zeitlimit, Netz), zeigt die Seite trotzdem die laufende Version.
|
||||
info["installed_version"] = _hermes_version(path)
|
||||
try:
|
||||
branch = (subprocess.run(["git", "-C", path, "rev-parse", "--abbrev-ref", "HEAD"],
|
||||
capture_output=True, text=True, timeout=8).stdout.strip() or "main")
|
||||
@@ -484,7 +500,6 @@ def hermes_update_details() -> dict:
|
||||
commits.append({"hash": parts[0], "subject": parts[1], "when": parts[2]})
|
||||
info["commits"] = commits
|
||||
info["behind"] = len(commits)
|
||||
info["installed_version"] = _hermes_version(path)
|
||||
if commits:
|
||||
info.update(_summarize_hermes_commits(commits))
|
||||
except Exception as exc:
|
||||
|
||||
Reference in New Issue
Block a user