add durable jobs and live job monitor

This commit is contained in:
Marco0300
2026-09-02 18:12:21 +02:00
parent bc33b03075
commit cf034288e6
12 changed files with 342 additions and 18 deletions
+111 -2
View File
@@ -1,5 +1,5 @@
from __future__ import annotations
import argparse, hashlib, json, os, re, secrets, sqlite3, sys
import argparse, hashlib, json, os, re, secrets, sqlite3, sys, threading, time
from datetime import datetime, timedelta, timezone
from http.cookies import SimpleCookie
from http.server import BaseHTTPRequestHandler, ThreadingHTTPServer
@@ -15,8 +15,23 @@ SCHEMA = Path(__file__).resolve().parents[1] / "schema.sql"
SESSION_DAYS = 7
PBKDF2_ITERATIONS = 300_000
MUTATING_ROLES = {"owner", "admin", "researcher"}
JOB_TYPES = {"noop", "prospect_recalculate"}
JOB_PAGE_SIZE = 100
SECRET_KEYS = {"password", "passwd", "secret", "token", "api_key", "apikey", "authorization", "credential", "private_key"}
CHILD_TABLES = {"contacts": ("name", "email", "phone", "title", "do_not_contact"), "domains": ("domain", "kind"), "websites": ("url", "website_class"), "evidence": ("kind", "url", "claim"), "notes": ("body",)}
def redact(value):
if isinstance(value, dict): return {k: ("[REDACTED]" if str(k).lower() in SECRET_KEYS or any(s in str(k).lower() for s in ("password", "token", "secret", "api_key")) else redact(v)) for k,v in value.items()}
if isinstance(value, list): return [redact(v) for v in value[:100]]
if isinstance(value, str): return value[:2000]
return value
def job_json(row):
result = row_json(row)
try: result["payload"] = json.loads(result.get("payload") or "{}")
except (ValueError, TypeError): result["payload"] = {}
return result
def hash_password(password: str, salt: bytes | None = None) -> tuple[str, str]:
salt = salt or secrets.token_bytes(16); return hashlib.pbkdf2_hmac("sha256", password.encode(), salt, PBKDF2_ITERATIONS).hex(), salt.hex()
def verify_password(password, encoded_hash, encoded_salt):
@@ -95,6 +110,8 @@ class ApiHandler(BaseHTTPRequestHandler):
if path=="/api/v1/dashboard/summary":
row=db.execute("SELECT COUNT(*) businesses,COALESCE(AVG(score),0) average_score FROM businesses WHERE organization_id=?",(org,)).fetchone(); return self.send_json(200,{"organization_id":org,"businesses":row["businesses"],"average_score":round(row["average_score"],2),"suppressed":db.execute("SELECT COUNT(*) FROM suppressions WHERE organization_id=?",(org,)).fetchone()[0]})
if path=="/api/v1/businesses": return self.list_businesses(db,org,parse_qs(parsed.query))
if path=="/api/v1/jobs": return self.list_jobs(db,org,parse_qs(parsed.query))
if path.startswith("/api/v1/jobs/"): return self.get_job_route(db,org,path,parse_qs(parsed.query))
if path.startswith("/api/v1/businesses/"):
bits=path.split("/"); ident=bits[4] if len(bits)>4 else ""
if not ident.isdigit(): return self.send_json(404,{"error":"not_found"})
@@ -103,6 +120,59 @@ class ApiHandler(BaseHTTPRequestHandler):
payload=row_json(row); payload.update(self.nested(db,int(ident),org)); return self.send_json(200,payload)
return self.send_json(404,{"error":"not_found"})
finally: db.close()
def list_jobs(self, db, org, query):
try: limit=max(1,min(int(query.get("page_size",[50])[0]),JOB_PAGE_SIZE)); offset=max(0,int(query.get("offset",[0])[0]))
except (ValueError, TypeError): return self.send_json(400,{"error":"invalid_pagination"})
rows=db.execute("SELECT * FROM jobs WHERE organization_id=? ORDER BY id DESC LIMIT ? OFFSET ?",(org,limit+1,offset)).fetchall(); more=len(rows)>limit
return self.send_json(200,{"organization_id":org,"items":[job_json(r) for r in rows[:limit]],"limit":limit,"offset":offset,"has_more":more})
def get_job_route(self, db, org, path, query):
bits=path.split("/")
if len(bits)<5 or not bits[4].isdigit(): return self.send_json(404,{"error":"not_found"})
job=db.execute("SELECT * FROM jobs WHERE id=? AND organization_id=?",(int(bits[4]),org)).fetchone()
if not job:return self.send_json(404,{"error":"not_found"})
if len(bits)==5:return self.send_json(200,job_json(job))
if len(bits)==6 and bits[5]=="events":
try: after=max(0,int(query.get("after",[0])[0]))
except (ValueError,TypeError): return self.send_json(400,{"error":"invalid_sequence"})
events=[row_json(r) for r in db.execute("SELECT * FROM job_events WHERE job_id=? AND organization_id=? AND sequence>? ORDER BY sequence",(job["id"],org,after))]
return self.send_json(200,{"items":events,"after":after})
if len(bits)==7 and bits[5]=="events" and bits[6]=="stream":
try: after=max(0,int(query.get("after",[0])[0]))
except (ValueError,TypeError): return self.send_json(400,{"error":"invalid_sequence"})
events=[row_json(r) for r in db.execute("SELECT * FROM job_events WHERE job_id=? AND organization_id=? AND sequence>? ORDER BY sequence",(job["id"],org,after))]
body=b"".join((b"event: "+str(e["event_type"]).encode()+b"\\ndata: "+json.dumps(e,sort_keys=True).encode()+b"\\n\\n") for e in events)
self.send_response(200);self.send_header("Content-Type","text/event-stream");self.send_header("Cache-Control","no-cache");self.send_header("Content-Length",str(len(body)));self.end_headers();self.wfile.write(body);return
return self.send_json(404,{"error":"not_found"})
def create_job(self, payload, db, user):
kind=str(payload.get("type","")).strip(); key=str(payload.get("idempotency_key","")).strip(); data=payload.get("payload",{})
if kind not in JOB_TYPES:return self.send_json(400,{"error":"invalid_job_type"})
if not key or len(key)>200 or not isinstance(data,dict):return self.send_json(400,{"error":"invalid_job_request"})
safe=json.dumps(redact(data),sort_keys=True,separators=(",",":")); max_attempts=max(1,min(int(payload.get("max_attempts",3)),5)) if str(payload.get("max_attempts",3)).isdigit() else 3
try:
cur=db.execute("INSERT INTO jobs(organization_id,idempotency_key,type,payload,max_attempts) VALUES(?,?,?,?,?)",(user["organization_id"],key,kind,safe,max_attempts)); jid=cur.lastrowid
self.add_job_event(db,jid,user["organization_id"],"queued","Job queued",0); self.audit(db,user,"job.created",str(jid)); db.commit(); getattr(self.server,"job_wakeup",threading.Event()).set()
return self.send_json(201,job_json(db.execute("SELECT * FROM jobs WHERE id=?",(jid,)).fetchone()))
except sqlite3.IntegrityError:
row=db.execute("SELECT * FROM jobs WHERE organization_id=? AND idempotency_key=?",(user["organization_id"],key)).fetchone(); return self.send_json(200,job_json(row))
def add_job_event(self, db, jid, org, event_type, message, progress=0, error_code=None):
seq=db.execute("SELECT COALESCE(MAX(sequence),0)+1 FROM job_events WHERE job_id=?",(jid,)).fetchone()[0]
db.execute("INSERT INTO job_events(job_id,organization_id,sequence,event_type,message,progress,error_code) VALUES(?,?,?,?,?,?,?)",(jid,org,seq,event_type,str(message)[:500],progress,error_code)); return seq
def job_action(self, db, user, path):
bits=path.split("/"); jid=int(bits[4]) if len(bits)>4 and bits[4].isdigit() else -1; action=bits[5] if len(bits)>5 else ""
job=db.execute("SELECT * FROM jobs WHERE id=? AND organization_id=?",(jid,user["organization_id"])).fetchone()
if not job:return self.send_json(404,{"error":"not_found"})
if action=="cancel":
if job["status"] in ("queued","running"): db.execute("UPDATE jobs SET status='cancelled',completed_at=CURRENT_TIMESTAMP,updated_at=CURRENT_TIMESTAMP WHERE id=?",(jid,));self.add_job_event(db,jid,user["organization_id"],"cancelled","Job cancelled",job["progress"])
self.audit(db,user,"job.cancelled",str(jid));db.commit();return self.send_json(200,job_json(db.execute("SELECT * FROM jobs WHERE id=?",(jid,)).fetchone()))
if action=="retry":
if job["status"]!="failed":return self.send_json(409,{"error":"job_not_failed"})
db.execute("UPDATE jobs SET status='queued',error_code=NULL,completed_at=NULL,updated_at=CURRENT_TIMESTAMP WHERE id=?",(jid,));self.add_job_event(db,jid,user["organization_id"],"retry","Job retry queued",job["progress"]);self.audit(db,user,"job.retried",str(jid));db.commit();getattr(self.server,"job_wakeup",threading.Event()).set();return self.send_json(200,job_json(db.execute("SELECT * FROM jobs WHERE id=?",(jid,)).fetchone()))
return self.send_json(404,{"error":"not_found"})
def list_businesses(self,db,org,query):
def number(name, default=None):
raw=query.get(name,[None])[0]
@@ -132,8 +202,13 @@ class ApiHandler(BaseHTTPRequestHandler):
c=SimpleCookie();c.load(self.headers.get("Cookie",""));t=c.get("session");
if t:db.execute("DELETE FROM sessions WHERE token_hash=?",(hashlib.sha256(t.value.encode()).hexdigest(),))
self.audit(db,user,"logout");db.commit();return self.send_json(200,{"ok":True},{"Set-Cookie":self.auth_cookie("",0)})
if path=="/api/v1/jobs":
if user["role"] not in MUTATING_ROLES:return self.send_json(403,{"error":"forbidden"})
return self.create_job(self.read_json(),db,user)
if user["role"] not in MUTATING_ROLES:return self.send_json(403,{"error":"forbidden"})
payload=self.read_json(); org=user["organization_id"]
if path.startswith("/api/v1/jobs/"):
return self.job_action(db,user,path)
if path=="/api/v1/businesses":return self.create_business(payload,db,user)
if path=="/api/v1/suppressions":return self.create_suppression(payload,db,user)
if path=="/api/v1/imports/preview":return self.preview_import(payload,db,org)
@@ -213,8 +288,42 @@ class ApiHandler(BaseHTTPRequestHandler):
return self.send_json(200,{"accepted":len(accepted),"duplicates":len(rows)-len(normalized)+len(normalized)-len(accepted)-suppressed,"suppressed":suppressed,"rows":accepted})
def log_message(self,*_):pass
def _job_worker(server):
while not server.job_stop.is_set():
db=connect(server.db_path)
try:
job=db.execute("SELECT * FROM jobs WHERE status='queued' ORDER BY id LIMIT 1").fetchone()
if not job:
db.close(); server.job_wakeup.wait(.1); server.job_wakeup.clear(); continue
changed=db.execute("UPDATE jobs SET status='running',attempts=attempts+1,started_at=COALESCE(started_at,CURRENT_TIMESTAMP),updated_at=CURRENT_TIMESTAMP WHERE id=? AND status='queued'",(job["id"],)).rowcount
if not changed: db.close(); continue
db.commit(); org=job["organization_id"]; jid=job["id"]; server_handler=object.__new__(ApiHandler)
server_handler.add_job_event(db,jid,org,"started","Job started",0); db.commit()
try: payload=json.loads(job["payload"] or "{}")
except ValueError: payload={}
try: steps=1 if job["type"]=="noop" else max(1,min(int(payload.get("steps",5)),20))
except (ValueError,TypeError): steps=5
cancelled=False
for i in range(steps):
time.sleep(.01)
fresh=db.execute("SELECT status FROM jobs WHERE id=?",(jid,)).fetchone()
if not fresh or fresh["status"]=="cancelled": cancelled=True; break
progress=int((i+1)*100/steps); db.execute("UPDATE jobs SET progress=?,updated_at=CURRENT_TIMESTAMP WHERE id=? AND status='running'",(progress,jid)); server_handler.add_job_event(db,jid,org,"progress",f"Job progress {progress}%",progress); db.commit()
if cancelled: continue
if payload.get("force_fail") or (payload.get("fail_once") and job["attempts"] == 0):
db.execute("UPDATE jobs SET status='failed',error_code='DEMO_FAILURE',completed_at=CURRENT_TIMESTAMP,updated_at=CURRENT_TIMESTAMP WHERE id=?",(jid,));server_handler.add_job_event(db,jid,org,"failed","Job failed",job["progress"],"DEMO_FAILURE")
else:
db.execute("UPDATE jobs SET status='succeeded',progress=100,completed_at=CURRENT_TIMESTAMP,updated_at=CURRENT_TIMESTAMP WHERE id=?",(jid,));server_handler.add_job_event(db,jid,org,"succeeded","Job completed",100)
db.commit()
finally: db.close()
def create_server(host="127.0.0.1",port=8000,db_path="prospects.db"):
server=ThreadingHTTPServer((host,port),ApiHandler);server.db_path=db_path;connect(db_path).close();return server
server=ThreadingHTTPServer((host,port),ApiHandler);server.db_path=db_path;connect(db_path).close();server.job_stop=threading.Event();server.job_wakeup=threading.Event();server.job_thread=threading.Thread(target=_job_worker,args=(server,),daemon=True);server.job_thread.start()
original_close=server.server_close
def close():
server.job_stop.set();server.job_wakeup.set();server.job_thread.join(timeout=2);original_close()
server.server_close=close
return server
if __name__=="__main__":
parser=argparse.ArgumentParser();parser.add_argument("--host",default="127.0.0.1");parser.add_argument("--port",type=int,default=int(os.environ.get("PROSPECT_API_PORT","8000")));parser.add_argument("--db",default=os.environ.get("PROSPECT_API_DB","prospects.db"));args=parser.parse_args();server=create_server(args.host,args.port,args.db);print(f"Prospect API listening on http://{args.host}:{args.port}",flush=True)
try:server.serve_forever()