Initial commit: SIAX Scrape Console (migrated from laptop)

2026-11-03 16:09:25 +01:00
commit 78146bb5ab
9 changed files with 2998 additions and 0 deletions
@@ -0,0 +1,4 @@
+__pycache__/
+*.pyc
+*.bak
+*.bak-*
@@ -0,0 +1,7 @@
+FROM python:3.12-slim
+WORKDIR /app
+COPY *.py ./
+ENV PYTHONUNBUFFERED=1
+ENV HOME=/data
+EXPOSE 8899
+CMD ["python3", "server.py"]
@@ -0,0 +1 @@
+{"biluppgifter_records": {"rows": 3592, "bytes": -1, "key": "16:14914251"}, "sf_spool": {"rows": 0, "bytes": -1, "key": "0:0"}, "carinfo_trims": {"rows": 407, "bytes": -1, "key": "1791395435.6649714:141170"}, "vehicles_clean": {"rows": 119714, "bytes": -1, "key": "1786478796.0:57346606"}, "allabolag": {"rows": 204961, "bytes": -1, "key": "1786478742.0:81531189"}, "hitta": {"rows": 88104, "bytes": -1, "key": "1786478749.0:7202748"}, "ratsit": {"rows": 229349, "bytes": -1, "key": "1786478755.0:12276419"}, "bolagsfakta": {"rows": 888661, "bytes": -1, "key": "1786478747.0:18001402"}, "hitta_companies": {"rows": 8, "bytes": -1, "key": "1790959874.31363:2577"}, "queue_biluppgifter": {"rows": 1819062, "bytes": -1, "key": "1791204095.6476479:86491329"}, "queue_biluppgifter_crawled": {"rows": 213065, "bytes": -1, "key": "1791204095.6476479:9993850"}, "queue_multi_fordonssok": {"rows": 0, "bytes": -1, "key": "1791216576.2764611:0"}, "queue_multi_fordonskollen": {"rows": 0, "bytes": -1, "key": "1791216576.2764611:0"}, "queue_multi_carinfo": {"rows": 0, "bytes": -1, "key": "1791216576.2764611:0"}, "reco_companies": {"rows": 4, "bytes": -1, "key": "1790959888.311806:395"}, "hitta_person_reg": {"rows": 2000000, "bytes": -1, "key": "1791210077.6520803:305528424"}, "hitta_company_reg": {"rows": 1588677, "bytes": -1, "key": "1791219600.018239:103172372"}, "ratsit_person_reg": {"rows": 2000000, "bytes": -1, "key": "1791212987.7546084:661407740"}, "ratsit_company_reg": {"rows": 1312997, "bytes": -1, "key": "1791213028.0060341:74810637"}, "bolagsfakta_reg": {"rows": 1538781, "bytes": -1, "key": "1791217621.7285998:91414094"}, "allabolag_reg": {"rows": 1285485, "bytes": -1, "key": "1786478090.0:47562943"}, "relations": {"rows": 0, "bytes": -1, "key": "0:0"}, "queue_hitta_person": {"rows": 2000000, "bytes": -1, "key": "1791376301.2104456:198937064"}, "queue_hitta_company": {"rows": 921216, "bytes": -1, "key": "1791376445.3741066:57927832"}, "queue_ratsit_person": {"rows": 38112, "bytes": -1, "key": "1791233954.4740746:4089353"}, "queue_bolagsfakta_company": {"rows": 1520927, "bytes": -1, "key": "1791292143.7172053:90353412"}, "identified": {"rows": 0, "bytes": -1, "key": "0:0"}, "relations_v2": {"rows": 0, "bytes": -1, "key": "0:0"}}
@@ -0,0 +1,121 @@
+#!/usr/bin/env python3
+"""Anpassningsbar widget-dashboard för Scrape Console (sidan /dash).
+Layouten sparas i webbläsarens localStorage. Widgets läser befintliga endpoints."""
+
+DASH_PAGE = """<!DOCTYPE html><html lang="sv"><head><meta charset="utf-8">
+<meta name="viewport" content="width=device-width,initial-scale=1">
+<title>Scrape Console · Min dashboard</title>
+<style>body{background:#0d1117;color:#e6edf3;font-family:system-ui,sans-serif;margin:0;padding:16px}
+a{color:#58a6ff}h1{font-size:20px;margin:0 0 8px}
+#toolbar{display:flex;gap:8px;flex-wrap:wrap;margin-bottom:12px;align-items:center;background:#161b22;border:1px solid #30363d;border-radius:8px;padding:10px}
+select,input{background:#0d1117;color:#e6edf3;border:1px solid #30363d;border-radius:6px;padding:6px 10px;font-size:13px}
+button{background:#21262d;color:#e6edf3;border:1px solid #30363d;border-radius:6px;padding:6px 12px;cursor:pointer;font-size:13px}
+button:hover{background:#30363d}
+#grid{display:grid;grid-template-columns:repeat(3,1fr);gap:12px}
+.box{background:#161b22;border:1px solid #30363d;border-radius:8px;padding:12px;min-width:0}
+.box.w1{grid-column:span 1}.box.w2{grid-column:span 2}.box.w3{grid-column:span 3}
+.box h3{font-size:14px;margin:0 0 6px}
+.tools{float:right;display:flex;gap:2px}.tools button{padding:1px 7px;font-size:11px;margin:0}
+.bar{background:#30363d;border-radius:4px;height:8px;margin:4px 0 8px}.bar>i{display:block;height:8px;border-radius:4px;background:#1f6feb}
+.src{font-size:12px;margin:2px 0}.src b{color:#e6edf3}.kv{font-size:12px;color:#8b949e}.kv b{color:#e6edf3}
+pre{background:#0d1117;border:1px solid #30363d;border-radius:6px;padding:8px;font-size:11px;overflow:auto;max-height:220px;white-space:pre-wrap}
+table{border-collapse:collapse;width:100%;font-size:11px}th,td{border:1px solid #30363d;padding:4px 6px;text-align:left;max-width:200px;overflow:hidden;text-overflow:ellipsis;white-space:nowrap}
+th{background:#0d1117}svg{width:100%;height:50px}
+textarea{width:100%;background:#0d1117;color:#e6edf3;border:1px solid #30363d;border-radius:6px;font-size:12px;min-height:60px}
+@media(max-width:900px){#grid{grid-template-columns:1fr}.box.w1,.box.w2,.box.w3{grid-column:span 1}}</style></head>
+<body><h1>🕷️ <a href="/">Scrape Console</a> · <a href="/data-page">Alla data</a> · Min dashboard</h1>
+<div id="toolbar"><b>Lägg till box:</b><select id="wtype"></select><select id="wref"></select>
+<button onclick="addBox()">+ Lägg till</button><button onclick="resetDash()">↺ Återställ</button>
+<span id="ts" style="color:#8b949e;font-size:12px"></span></div>
+<div id="grid"></div>
+<script>
+const WDEF = {
+  progress: {t:'Progress per källa', refs:['*']},
+  rate: {t:'Takt-graf', refs:['biluppgifter','fordonskollen','fordonssok','carinfo']},
+  log: {t:'Loggsvans', refs:['multi:SF multi','carinfo:car.info']},
+  dataset: {t:'Datasett-tabell', refs:[]},
+  status: {t:'Scraper-status', refs:['*']},
+  ckpt: {t:'Checkpoint', refs:['*']},
+  note: {t:'Anteckning (egen text)', refs:['*']},
+};
+let L = JSON.parse(localStorage.getItem('scrape-dash') || 'null') || [
+  {id:1,type:'status',ref:'*',w:3},
+  {id:2,type:'progress',ref:'*',w:2},
+  {id:3,type:'rate',ref:'fordonssok',w:1},
+  {id:4,type:'log',ref:'multi',w:2},
+  {id:5,type:'dataset',ref:'vehicles_clean',w:3},
+];
+let DS = [], ST = null, HIST = [];
+function save(){localStorage.setItem('scrape-dash', JSON.stringify(L));}
+function resetDash(){localStorage.removeItem('scrape-dash');location.reload();}
+async function init(){
+  DS = await (await fetch('/datasets')).json();
+  WDEF.dataset.refs = DS.map(d=>d.id+':'+d.label.split('·').pop().trim().slice(0,28));
+  const t = document.getElementById('wtype'), r = document.getElementById('wref');
+  t.innerHTML = Object.entries(WDEF).map(([k,v])=>`<option value="${k}">${v.t}</option>`).join('');
+  t.onchange = ()=>{r.innerHTML = WDEF[t.value].refs.map(x=>{const [v,l]=x.split(/:(.+)/);return `<option value="${v}">${l||v}</option>`;}).join('');};
+  t.onchange(); render();
+}
+function addBox(){
+  const type = document.getElementById('wtype').value, ref = document.getElementById('wref').value;
+  let extra = {};
+  if(type==='note'){const txt = prompt('Text:',''); if(txt===null) return; extra={text:txt};}
+  L.push({id:Date.now(), type, ref, w:type==='status'||type==='dataset'?3:1, ...extra});
+  save(); render();
+}
+function boxOp(id, op){
+  const i = L.findIndex(b=>b.id===id); if(i<0) return;
+  if(op==='del') L.splice(i,1);
+  if(op==='left' && i>0) [L[i-1],L[i]] = [L[i],L[i-1]];
+  if(op==='right' && i<L.length-1) [L[i+1],L[i]] = [L[i],L[i+1]];
+  if(op==='w') L[i].w = L[i].w%3+1;
+  if(op==='edit' && L[i].type==='note'){const t=prompt('Text:',L[i].text||''); if(t!==null)L[i].text=t;}
+  save(); render();
+}
+function tools(b){
+  return `<span class="tools"><button onclick="boxOp(${b.id},'left')" title="flytta vänster">◀</button><button onclick="boxOp(${b.id},'right')" title="flytta höger">▶</button><button onclick="boxOp(${b.id},'w')" title="bredd">↔${b.w}</button>${b.type==='note'?`<button onclick="boxOp(${b.id},'edit')" title="redigera">✎</button>`:''}<button onclick="boxOp(${b.id},'del')" title="ta bort">✕</button></span>`;
+}
+async function render(){
+  ST = await (await fetch('/state')).json();
+  HIST = await (await fetch('/history')).json();
+  document.getElementById('ts').textContent = 'Uppdaterad ' + ST.ts + ' (auto 15 s)';
+  const g = document.getElementById('grid'); g.innerHTML = '';
+  for(const b of L){
+    let h = `<div class="box w${b.w}">${tools(b)}`;
+    if(b.type==='status'){
+      h += `<h3>Scraper-status</h3>` + ST.scrapers.map(c=>`<div class="src"><b>${c.label}</b> ${c.active} · ${c.mem_mb} MB · ${c.log?'':''}</div>`).join('');
+    } else if(b.type==='progress'){
+      const p = ST.scrapers[0].progress || {};
+      h += `<h3>Progress per källa</h3>`;
+      for(const [k,v] of Object.entries(p)){ if(k.startsWith('_')) continue;
+        const pct = 100*v.pos/v.total;
+        h += `<div class="src"><b>${k}</b> ${v.ok} ok/${v.fail} fail · ${(pct).toFixed(2)}%</div><div class="bar"><i style="width:${Math.min(100,pct)}%"></i></div>`; }
+    } else if(b.type==='rate'){
+      h += `<h3>Takt: ${b.ref}</h3>` + spark(b.ref);
+    } else if(b.type==='log'){
+      const [src] = b.ref.split(':');
+      const t = await (await fetch(`/log?src=${src}&n=12`)).text();
+      h += `<h3>Logg: ${b.ref}</h3><pre>${t.replace(/</g,'&lt;')}</pre>`;
+    } else if(b.type==='dataset'){
+      const d = await (await fetch(`/data?ds=${b.ref}&page=1`)).json();
+      const cols = (d.cols||[]).slice(0,5);
+      h += `<h3>${(DS.find(x=>x.id===b.ref)||{}).label||b.ref} (${d.total})</h3><table><tr>${cols.map(c=>`<th>${c}</th>`).join('')}</tr>` +
+        (d.rows||[]).slice(0,8).map(r=>`<tr>${cols.map(c=>`<td>${String(r[c]??'').replace(/</g,'&lt;')}</td>`).join('')}</tr>`).join('') + `</table>`;
+    } else if(b.type==='ckpt'){
+      h += `<h3>car.info checkpoint</h3><div class="src"><b>${Object.entries(ST.scrapers[1].ckpt||{}).map(([k,v])=>k+'='+v).join(' · ')}</b></div>`;
+    } else if(b.type==='note'){
+      h += `<h3>📝 Anteckning</h3><div class="src">${(b.text||'').replace(/</g,'&lt;')}</div>`;
+    }
+    g.innerHTML += h + '</div>';
+  }
+}
+function spark(src){
+  const pts = HIST.filter(h=>h.ok && h.ok[src]!==undefined).slice(-60);
+  if(pts.length < 2) return '<div class="kv">ingen historik ännu</div>';
+  const vals = pts.map(h=>h.ok[src]), mx = Math.max(...vals), mn = Math.min(...vals);
+  const W=300,H=46, path = vals.map((v,i)=>`${(i/(vals.length-1)*W).toFixed(1)},${(H-2-((v-mn)/Math.max(1,mx-mn))*(H-4)).toFixed(1)}`).join(' ');
+  const rate = ((vals[vals.length-1]-vals[0])/Math.max(1,(pts[pts.length-1].t-pts[0].t)/60)).toFixed(0);
+  return `<svg viewBox="0 0 ${W} ${H}"><polyline points="${path}" fill="none" stroke="#1f6feb" stroke-width="1.5"/></svg><div class="kv"><b>${rate}/min</b></div>`;
+}
+init(); setInterval(render, 15000);
+</script></body></html>"""
@@ -0,0 +1,613 @@
+#!/usr/bin/env python3
+"""Data-vyer för Scrape Console: dataset-register, tabell, postdetalj, export.
+Endast stdlib. Anropas från server.py (routes) och serverar /data-sidan."""
+
+import csv, gzip, json, os, time
+
+HOME = os.path.expanduser("~")
+SF = os.path.join(HOME, "Documents/SF_Extractor")
+CI = os.path.join(HOME, ".local/share/carinfo")
+COUNT_CACHE = {"ts": 0, "data": {}}
+
+
+def _files_jsonl_gz(d):
+    try:
+        return sorted(
+            [
+                os.path.join(d, f)
+                for f in os.listdir(d)
+                if f.endswith(".jsonl.gz") and os.path.getsize(os.path.join(d, f)) > 0
+            ]
+        )
+    except Exception:
+        return []
+
+
+def _iter_jsonl_gz(path):
+    try:
+        with gzip.open(path, "rt", errors="replace") as f:
+            for line in f:
+                line = line.strip()
+                if line:
+                    try:
+                        yield json.loads(line)
+                    except Exception:
+                        continue
+    except Exception:
+        return
+
+
+def _iter_jsonl(path):
+    try:
+        with open(path, errors="replace") as f:
+            for line in f:
+                line = line.strip()
+                if line:
+                    try:
+                        yield json.loads(line)
+                    except Exception:
+                        continue
+    except Exception:
+        return
+
+
+def _iter_csv(path):
+    try:
+        with open(path, newline="", errors="replace") as f:
+            for row in csv.DictReader(f):
+                yield row
+    except Exception:
+        return
+
+
+def _count(it, cap=2000000):
+    n = 0
+    for _ in it:
+        n += 1
+        if n >= cap:
+            break
+    return n
+
+
+SCRAPER_OF = {"biluppgifter_records": "sf", "sf_spool": "sf", "carinfo_trims": "carinfo",
+    "vehicles_clean": "sf", "allabolag": "sf", "hitta": "sf", "ratsit": "sf",
+    "bolagsfakta": "sf", "hitta_companies": "sf", "reco_companies": "sf",
+    "regnr": "sf", "queue_biluppgifter": "sf", "queue_biluppgifter_crawled": "sf",
+    "queue_multi_fordonssok": "sf", "queue_multi_fordonskollen": "sf",
+    "queue_multi_carinfo": "sf"}
+DATASETS = [
+    {
+        "id": "biluppgifter_records",
+        "label": "SF · biluppgifter records (extraherat)",
+        "kind": "multi-jsonl-gz",
+        "dir": os.path.join(SF, "full_data/biluppgifter"),
+        "prefix": "records-",
+        "desc": "titel, tabeller, json_ld, text per fordonssida",
+    },
+    {
+        "id": "sf_spool",
+        "label": "SF · spool (råcaptures, live)",
+        "kind": "multi-jsonl-gz",
+        "dir": os.path.join(SF, "spool"),
+        "prefix": "multi-",
+        "desc": "{url, html} per hämtad sida, nyast först",
+    },
+    {
+        "id": "carinfo_trims",
+        "label": "car.info · trims/specs",
+        "kind": "jsonl",
+        "path": os.path.join(CI, "trims.jsonl"),
+        "desc": "{url, make, model, specs{...}}",
+    },
+    {
+        "id": "vehicles_clean",
+        "label": "SF · vehicles_clean.csv",
+        "kind": "csv",
+        "path": os.path.join(SF, "cleaned_allabolag/vehicles_clean.csv"),
+        "desc": "40+ fordonskolumner (regnr, vin, make, model, skatt, besiktning ...)",
+    },
+    {
+        "id": "allabolag",
+        "label": "SF · extracted_allabolag.csv",
+        "kind": "csv",
+        "path": os.path.join(SF, "cleaned_allabolag/extracted_allabolag.csv"),
+        "desc": "",
+    },
+    {
+        "id": "hitta",
+        "label": "SF · extracted_hitta.csv",
+        "kind": "csv",
+        "path": os.path.join(SF, "cleaned_allabolag/extracted_hitta.csv"),
+        "desc": "",
+    },
+    {
+        "id": "ratsit",
+        "label": "SF · extracted_ratsit.csv",
+        "kind": "csv",
+        "path": os.path.join(SF, "cleaned_allabolag/extracted_ratsit.csv"),
+        "desc": "",
+    },
+    {
+        "id": "bolagsfakta",
+        "label": "SF · extracted_bolagsfakta.csv",
+        "kind": "csv",
+        "path": os.path.join(SF, "cleaned_allabolag/extracted_bolagsfakta.csv"),
+        "desc": "",
+    },
+    {
+        "id": "hitta_companies",
+        "label": "SF · hitta companies.csv",
+        "kind": "csv",
+        "path": os.path.join(SF, "multisite_data/hitta/companies.csv"),
+        "desc": "",
+    },
+    {
+        "id": "regnr",
+        "label": "Register · regnr x domän (14,5M, indexerat)",
+        "kind": "sqlite",
+        "desc": "prefixsök på regnr — ms-svar via index",
+    },
+    {
+        "id": "queue_biluppgifter",
+        "label": "SF · kö: biluppgifter frontier (1,8M)",
+        "kind": "lines",
+        "path": os.path.join(SF, "full_data/biluppgifter/frontier.txt"),
+        "desc": "",
+    },
+    {
+        "id": "queue_biluppgifter_crawled",
+        "label": "SF · kö: biluppgifter crawled",
+        "kind": "lines",
+        "path": os.path.join(SF, "full_data/biluppgifter/crawled.txt"),
+        "desc": "",
+    },
+    {
+        "id": "queue_multi_fordonssok",
+        "label": "SF · kö: multi/fordonssok crawled",
+        "kind": "lines",
+        "path": os.path.join(SF, "full_data/multi/fordonssok/crawled.txt"),
+        "desc": "",
+    },
+    {
+        "id": "queue_multi_fordonskollen",
+        "label": "SF · kö: multi/fordonskollen crawled",
+        "kind": "lines",
+        "path": os.path.join(SF, "full_data/multi/fordonskollen/crawled.txt"),
+        "desc": "",
+    },
+    {
+        "id": "queue_multi_carinfo",
+        "label": "SF · kö: multi/carinfo crawled",
+        "kind": "lines",
+        "path": os.path.join(SF, "full_data/multi/carinfo/crawled.txt"),
+        "desc": "",
+    },
+    {
+        "id": "reco_companies",
+        "label": "SF · reco companies.csv",
+        "kind": "csv",
+        "path": os.path.join(SF, "multisite_data/reco/companies.csv"),
+        "desc": "",
+    },
+    {
+        "id": "hitta_person_reg",
+        "label": "Register · hitta personer (6,08M)",
+        "kind": "lines",
+        "path": os.path.join(SF, "hitta_person_urls.txt"),
+        "desc": "person-URL:er från hitta.se sitemaps",
+    },
+    {
+        "id": "hitta_company_reg",
+        "label": "Register · hitta företag (1,59M rena)",
+        "kind": "lines",
+        "path": os.path.join(SF, "hitta_company_clean.txt"),
+        "desc": "malformade sök-URL:er borttagna",
+    },
+    {
+        "id": "ratsit_person_reg",
+        "label": "Register · ratsit personer (6,2M)",
+        "kind": "lines",
+        "path": os.path.join(SF, "ratsit_person_urls.txt"),
+        "desc": "namn+ort+födelsedatum i URL",
+    },
+    {
+        "id": "ratsit_company_reg",
+        "label": "Register · ratsit företag (1,31M)",
+        "kind": "lines",
+        "path": os.path.join(SF, "ratsit_company_urls.txt"),
+        "desc": "",
+    },
+    {
+        "id": "bolagsfakta_reg",
+        "label": "Register · bolagsfakta företag (1,54M)",
+        "kind": "lines",
+        "path": os.path.join(SF, "bolagsfakta_company_urls.txt"),
+        "desc": "",
+    },
+    {
+        "id": "allabolag_reg",
+        "label": "Register · allabolag företag (1,29M)",
+        "kind": "lines",
+        "path": os.path.join(SF, "allabolag_urls_for_sf.txt"),
+        "desc": "",
+    },
+    {
+        "id": "relations",
+        "label": "VEHIQ · relationsgraf fordon→ägare→person→företag→adress",
+        "kind": "multi-jsonl-gz",
+        "dir": os.path.join(HOME, "sf-s10/full_data/relations"),
+        "prefix": "relations-",
+        "desc": "kanter: src_kind/src_id/relation/dst_kind/dst_id/weight",
+    },
+    {"id":"queue_hitta_person","label":"SF · kö: multi/hitta_person crawled","kind":"lines","path":os.path.join(SF,"full_data/multi/hitta_person/crawled.txt"),"desc":""},
+    {"id":"queue_hitta_company","label":"SF · kö: multi/hitta_company crawled","kind":"lines","path":os.path.join(SF,"full_data/multi/hitta_company/crawled.txt"),"desc":""},
+    {"id":"queue_ratsit_person","label":"SF · kö: multi/ratsit_person crawled","kind":"lines","path":os.path.join(SF,"full_data/multi/ratsit_person/crawled.txt"),"desc":""},
+    {"id":"queue_bolagsfakta_company","label":"SF · kö: multi/bolagsfakta_company crawled","kind":"lines","path":os.path.join(SF,"full_data/multi/bolagsfakta_company/crawled.txt"),"desc":""},
+    {
+        "id": "identified",
+        "label": "SF · identifierade records (_id + regnr/orgnr/personnummer)",
+        "kind": "multi-jsonl-gz",
+        "dir": os.path.join(HOME, "sf-s10/full_data/identified"),
+        "prefix": "identified",
+        "desc": "_id (vehicle:/company:/person:), id_type, regnr, orgnr, personnummer",
+    },
+    {
+        "id": "relations_v2",
+        "label": "VEHIQ · relations v2 (företag→adress, företag→person, samma adress)",
+        "kind": "multi-jsonl-gz",
+        "dir": os.path.join(HOME, "sf-s10/full_data/relations"),
+        "prefix": "relations_v2",
+        "desc": "located_at / related_person / related_company / same_address / brand / model",
+    },
+]
+
+
+def _wc(path):
+    try:
+        with open(path, "rb") as f:
+            return sum(1 for _ in f)
+    except Exception:
+        return -1
+
+
+def _ds(dsid):
+    for d in DATASETS:
+        if d["id"] == dsid:
+            return d
+    return None
+
+
+def _iter_ds(d, reverse_spool=True):
+    if d["kind"] == "multi-jsonl-gz":
+        files = [
+            f
+            for f in _files_jsonl_gz(d["dir"])
+            if os.path.basename(f).startswith(d.get("prefix", ""))
+        ]
+        if reverse_spool:
+            files = sorted(files, key=os.path.getmtime, reverse=True)
+        for fp in files:
+            yield from _iter_jsonl_gz(fp)
+    elif d["kind"] == "jsonl":
+        yield from _iter_jsonl(d["path"])
+    elif d["kind"] == "csv":
+        yield from _iter_csv(d["path"])
+    elif d["kind"] == "lines":
+        yield from _iter_lines(d["path"])
+
+
+def counts():
+    if time.time() - COUNT_CACHE["ts"] < 120:
+        return COUNT_CACHE["data"]
+    cfile = os.path.join(HOME, ".local/share/scrape-console/counts.json")
+    try:
+        saved = json.load(open(cfile))
+    except Exception:
+        saved = {}
+
+    def fkey(d):
+        try:
+            if d["kind"] in ("csv", "jsonl", "lines"):
+                st = os.stat(d["path"])
+                return "%s:%s" % (st.st_mtime, st.st_size)
+            fs = sorted(_files_jsonl_gz(d["dir"]))
+            return "%s:%s" % (len(fs), sum(os.path.getsize(f) for f in fs))
+        except Exception:
+            return "?"
+
+    out, dirty = {}, False
+    for d in DATASETS:
+        if d["kind"] == "sqlite":
+            out[d["id"]] = {"rows": sum(DOMAIN_COUNTS.values()), "bytes": 4100000000}
+            continue
+        key = fkey(d)
+        if saved.get(d["id"], {}).get("key") == key:
+            out[d["id"]] = saved[d["id"]]
+            continue
+        try:
+            if d["kind"] == "csv":
+                n = _count(_iter_csv(d["path"]))
+            elif d["kind"] == "jsonl":
+                n = _count(_iter_jsonl(d["path"]))
+            elif d["kind"] == "lines":
+                n = _count(_iter_lines(d["path"]))
+            else:
+                files = [
+                    f
+                    for f in _files_jsonl_gz(d["dir"])
+                    if os.path.basename(f).startswith(d.get("prefix", ""))
+                ]
+                n = 0
+                for fp in files:
+                    n += _count(_iter_jsonl_gz(fp))
+                    if n >= 2000000:
+                        break
+        except Exception as e:
+            out[d["id"]] = {"rows": -1, "bytes": -1, "err": str(e)[:80]}
+            continue
+        out[d["id"]] = {"rows": n, "bytes": -1, "key": key}
+        saved[d["id"]] = out[d["id"]]
+        dirty = True
+    if dirty:
+        try:
+            json.dump(saved, open(cfile, "w"))
+        except Exception:
+            pass
+    COUNT_CACHE.update(ts=time.time(), data=out)
+    return out
+
+
+REGDB = os.path.expanduser("~/.local/share/regnr-register/regnr.db")
+DOMAIN_COUNTS = {"fordonssok": 11524863, "fordonskollen": 13366899,
+                 "carinfo": 11075425, "minbilkoll": 98658}  # snapshot 2026-10-05
+
+def _sqlite_sightings(q="", page=1, per=50):
+    import sqlite3
+    if not os.path.exists(REGDB):
+        raise FileNotFoundError("registret återbyggs — klart inom ~1h")
+    con = sqlite3.connect("file:" + REGDB + "?mode=ro", uri=True)
+    cur = con.cursor()
+    q = (q or "").strip().upper()
+    if q:
+        cur.execute("SELECT regnr, domain, url FROM sightings WHERE regnr LIKE ? "
+                    "ORDER BY regnr, domain LIMIT 500", (q + "%",))
+        rows = [{"regnr": r[0], "domain": r[1], "url": r[2]} for r in cur.fetchall()]
+    else:
+        cur.execute("SELECT regnr, domain, url FROM sightings ORDER BY regnr LIMIT 50")
+        rows = [{"regnr": r[0], "domain": r[1], "url": r[2]} for r in cur.fetchall()]
+    con.close()
+    return rows
+
+
+def _iter_lines(path):
+    try:
+        with open(path, errors="replace") as f:
+            for line in f:
+                line = line.strip()
+                if line:
+                    yield {"rad": line[:300]}
+    except Exception:
+        return
+
+
+def _spool_title(html):
+    import re
+    m = re.search(r"<title>(.*?)</title>", html or "", re.S | re.I)
+    return (m.group(1).strip()[:140] if m else "-")
+
+
+def _flat(rec):
+
+    """Platta till en nivå för tabellvisning (kapslade -> JSON-sträng, trunkerad)."""
+    if not isinstance(rec, dict):
+        return {"value": str(rec)[:300]}
+    if set(rec.keys()) == {"u", "h"}:
+        return {"url": rec.get("u", "")[:200], "titel": _spool_title(rec.get("h", "")),
+                "html_kb": round(len(rec.get("h", "")) / 1024, 1)}
+    out = {}
+    for k, v in rec.items():
+        if isinstance(v, (dict, list)):
+            s = json.dumps(v, ensure_ascii=False)
+            out[k] = (s[:220] + "…") if len(s) > 220 else s
+        else:
+            s = str(v)
+            out[k] = (s[:220] + "…") if len(s) > 220 else s
+    return out
+
+
+def query(dsid, q="", page=1, per=50, scan_cap=400000, sort=None, sdir="asc"):
+    d = _ds(dsid)
+    if not d:
+        return {"error": "okänt dataset"}
+    if d["kind"] == "sqlite":
+        try:
+            rows = _sqlite_sightings(q, page, per)
+        except Exception as e:
+            return {"error": str(e)[:120], "total": 0, "page": page, "per": per,
+                    "pages": 1, "cols": [], "rows": [], "scanned": 0, "truncated": False}
+        cols = ["regnr", "domain", "url"]
+        if sort == "domain":
+            rows.sort(key=lambda r: r["domain"], reverse=(sdir == "desc"))
+        start = (page - 1) * per
+        total = len(rows) if q else sum(DOMAIN_COUNTS.values())
+        return {"total": total, "page": page, "per": per,
+                "pages": max(1, (min(total, 500) + per - 1) // per),
+                "cols": cols, "rows": rows[start:start + per],
+                "scanned": len(rows), "truncated": len(rows) >= 500,
+                "sort": sort, "dir": sdir}
+    if d["kind"] == "lines" and not q and not sort:
+        total = _wc(d["path"])
+        start = (page - 1) * per
+        rows, i = [], 0
+        for rec in _iter_lines(d["path"]):
+            if i >= start and len(rows) < per:
+                rows.append(_flat(rec))
+            i += 1
+            if i >= start + per:
+                break
+        return {"total": total if total >= 0 else 0, "page": page, "per": per,
+                "pages": max(1, (max(total, 0) + per - 1) // per),
+                "cols": ["rad"], "rows": rows, "scanned": i,
+                "truncated": False, "sort": sort, "dir": sdir}
+    q = (q or "").lower()
+    matched, scanned = [], 0
+    for rec in _iter_ds(d):
+        scanned += 1
+        if scanned > scan_cap:
+            break
+        if q and q not in json.dumps(rec, ensure_ascii=False).lower():
+            continue
+        matched.append(rec)
+    cols = list(_flat(matched[0]).keys())[:12] if matched else []
+    if sort and sort in cols:
+        try:
+            matched.sort(key=lambda r: str(_flat(r).get(sort, "")),
+                         reverse=(sdir == "desc"))
+        except Exception:
+            pass
+    total = len(matched)
+    start = (page - 1) * per
+    return {
+        "total": total,
+        "page": page,
+        "per": per,
+        "pages": max(1, (total + per - 1) // per),
+        "cols": cols,
+        "rows": [_flat(r) for r in matched[start:start + per]],
+        "scanned": scanned,
+        "truncated": scanned >= scan_cap,
+        "sort": sort,
+        "dir": sdir,
+    }
+
+
+def record_at(dsid, n, q=""):
+    d = _ds(dsid)
+    if not d:
+        return {"error": "okänt dataset"}
+    q = (q or "").lower()
+    i = -1
+    for rec in _iter_ds(d):
+        if q and q not in json.dumps(rec, ensure_ascii=False).lower():
+            continue
+        i += 1
+        if i == n:
+            return {"n": n, "record": rec}
+        if i > n + 5:
+            break
+    return {"error": "finns ej"}
+
+
+def export(dsid, q="", fmt="csv", cap=100000):
+    d = _ds(dsid)
+    q = (q or "").lower()
+    rows, cols, n = [], [], 0
+    for rec in _iter_ds(d):
+        if q and q not in json.dumps(rec, ensure_ascii=False).lower():
+            continue
+        f = _flat(rec)
+        if not cols:
+            cols = list(f.keys())
+        rows.append(f)
+        n += 1
+        if n >= cap:
+            break
+    if fmt == "json":
+        body = json.dumps(rows, ensure_ascii=False, indent=1).encode()
+        return body, "application/json", f"{dsid}.json"
+    if fmt == "html":
+        h = ["<table><tr>" + "".join(f"<th>{c}</th>" for c in cols) + "</tr>"]
+        for r in rows:
+            h.append("<tr>" + "".join(f"<td>{str(r.get(c, ''))}</td>" for c in cols) + "</tr>")
+        h.append("</table>")
+        return ("".join(h)).encode(), "text/html", f"{dsid}.html"
+    if fmt == "xml":
+        import xml.etree.ElementTree as ET
+        root = ET.Element("rows")
+        for r in rows:
+            it = ET.SubElement(root, "row")
+            for c in cols:
+                e = ET.SubElement(it, "f", name=c)
+                e.text = str(r.get(c, ""))
+        return ET.tostring(root, encoding="unicode").encode(), "application/xml", f"{dsid}.xml"
+    import io
+
+    buf = io.StringIO()
+    w = csv.DictWriter(buf, fieldnames=cols, extrasaction="ignore")
+    w.writeheader()
+    w.writerows(rows)
+    return buf.getvalue().encode(), "text/csv", f"{dsid}.csv"
+
+
+PAGE = """<!DOCTYPE html><html lang="sv"><head><meta charset="utf-8">
+<meta name="viewport" content="width=device-width,initial-scale=1">
+<title>Scrape Console · Alla data</title>
+<style>body{background:#0d1117;color:#e6edf3;font-family:system-ui,sans-serif;margin:0;padding:16px}
+a{color:#58a6ff}h1{font-size:20px;margin:0 0 8px}.bar{display:flex;gap:8px;flex-wrap:wrap;margin-bottom:12px;align-items:center}
+select,input{background:#0d1117;color:#e6edf3;border:1px solid #30363d;border-radius:6px;padding:6px 10px;font-size:13px}
+button{background:#21262d;color:#e6edf3;border:1px solid #30363d;border-radius:6px;padding:6px 12px;cursor:pointer;font-size:13px}
+button:hover{background:#30363d}
+table{border-collapse:collapse;width:100%;font-size:12px}th,td{border:1px solid #30363d;padding:5px 8px;text-align:left;vertical-align:top;max-width:280px;overflow:hidden;text-overflow:ellipsis;white-space:nowrap}
+th{background:#161b22;position:sticky;top:0;cursor:pointer}th:first-child,td:first-child{position:sticky;left:0;background:#161b22;z-index:1}
+tr:hover td{background:#1c2530;cursor:pointer}
+#info{color:#8b949e;font-size:12px;margin:8px 0}
+#cols{font-size:11px;color:#8b949e;display:flex;gap:8px;flex-wrap:wrap;margin-bottom:8px}
+#cols label{cursor:pointer}
+#modal{position:fixed;inset:0;background:rgba(0,0,0,.7);display:none;align-items:center;justify-content:center}
+#mbox{background:#161b22;border:1px solid #30363d;border-radius:8px;max-width:90%;max-height:85%;overflow:auto;padding:16px}
+pre{font-size:11px;white-space:pre-wrap}</style></head>
+<body><h1>🕷️ <a href="/">Scrape Console</a> · <a href="/dash">Min dashboard</a> · Alla data</h1>
+<div class="bar"><select id="scr"><option value="">Alla scrapers</option><option value="sf">SF multi-scrape</option><option value="carinfo">car.info</option></select><select id="ds"></select><input id="q" placeholder="sök (delsträng)…" size="28">
+<button onclick="load(1)">Sök</button><button onclick="exp('csv')">CSV</button><button onclick="exp('json')">JSON</button>
+<span id="info"></span></div>
+<div id="cols"></div>
+<div id="pg" class="bar"></div>
+<table><thead id="th"></thead><tbody id="tb"></tbody></table>
+<div id="modal" onclick="this.style.display='none'"><div id="mbox" onclick="event.stopPropagation()"></div></div>
+<script>
+let DS='',Q='',PAGEG=1,COLS=[],HIDE=new Set(),SORT=null,DIR='asc',ALLDS=[];
+async function init(){
+  ALLDS = await (await fetch('/datasets')).json();
+  const prm = new URLSearchParams(location.search);
+  if(prm.get('scraper')) document.getElementById('scr').value = prm.get('scraper');
+  document.getElementById('scr').onchange = fillDS;
+  fillDS();
+  if(prm.get('ds')) document.getElementById('ds').value = prm.get('ds');
+  DS = document.getElementById('ds').value;
+  document.getElementById('q').addEventListener('keydown',e=>{if(e.key==='Enter')load(1);});
+  load(1);
+}
+function fillDS(){
+  const f = document.getElementById('scr').value;
+  const s = document.getElementById('ds');
+  s.innerHTML = ALLDS.filter(d=>!f||d.scraper===f).map(x=>`<option value="${x.id}">${x.label} (${x.rows})</option>`).join('');
+  DS = s.value;
+}
+async function load(p){
+  Q = document.getElementById('q').value; PAGEG = p;
+  DS = document.getElementById('ds').value;
+  let url = `/data?ds=${DS}&q=${encodeURIComponent(Q)}&page=${p}`;
+  if(SORT) url += `&sort=${encodeURIComponent(SORT)}&dir=${DIR}`;
+  const r = await (await fetch(url)).json();
+  if(r.error){document.getElementById('info').textContent=r.error;return;}
+  COLS = r.cols;
+  document.getElementById('info').textContent = `${r.total} träffar · sida ${r.page}/${r.pages}${r.truncated?' · cappad genomsökning':''}${r.sort?` · sort ${r.sort} ${r.dir}`:''}`;
+  document.getElementById('cols').innerHTML = 'Kolumner: ' + COLS.map(c=>`<label><input type="checkbox" ${HIDE.has(c)?'':'checked'} onchange="tglCol('${c.replace(/'/g,'')}')"> ${c}</label>`).join('');
+  const vis = COLS.filter(c=>!HIDE.has(c));
+  document.getElementById('th').innerHTML = '<tr><th>#</th>'+vis.map(c=>`<th onclick="sortBy('${c.replace(/'/g,'')}')">${c}${SORT===c?(DIR==='asc'?' ▲':' ▼'):''}</th>`).join('')+'</tr>';
+  document.getElementById('tb').innerHTML = r.rows.map((row,i)=>`<tr onclick="openRec(${(r.page-1)*r.per+i})"><td>${(r.page-1)*r.per+i}</td>`+vis.map(c=>`<td>${String(row[c]??'').replace(/</g,'&lt;')}</td>`).join('')+'</tr>').join('');
+  let pg = '';
+  for(const pp of [1,r.page-1,r.page,r.page+1,r.pages]) if(pp>=1&&pp<=r.pages) pg += `<button ${pp===r.page?'disabled':''} onclick="load(${pp})">${pp}</button>`;
+  document.getElementById('pg').innerHTML = pg;
+}
+function tglCol(c){HIDE.has(c)?HIDE.delete(c):HIDE.add(c);load(PAGEG);}
+function sortBy(c){if(SORT===c)DIR=DIR==='asc'?'desc':'asc';else{SORT=c;DIR='asc';}load(PAGEG);}
+async function openRec(n){
+  const r = await (await fetch(`/record?ds=${DS}&n=${n}&q=${encodeURIComponent(Q)}`)).json();
+  document.getElementById('mbox').innerHTML = `<h3>Post #${n} (${DS})</h3><pre>${JSON.stringify(r.record,null,1).replace(/</g,'&lt;').slice(0,60000)}</pre>`;
+  document.getElementById('modal').style.display = 'flex';
+}
+function exp(fmt){location.href = `/export?ds=${DS}&q=${encodeURIComponent(Q)}&fmt=${fmt}`;}
+init();
+</script></body></html>
+"""
@@ -0,0 +1,215 @@
+#!/usr/bin/env python3
+"""Ops-vyer för Scrape Console (Octoparse-paritet): runs-historik, scheman,
+proxy-pool, pipeline-steg och alerts. Endast stdlib. Endpoint: /ops."""
+
+import json, os, re, shutil, subprocess, time
+from collections import Counter
+
+HOME = os.path.expanduser("~")
+SF = os.path.join(HOME, "Documents/SF_Extractor")
+CI = os.path.join(HOME, ".local/share/carinfo")
+CACHE = {"ts": 0, "data": None}
+
+UNITS = [
+    "sf-multi-scrape.service",
+    "carinfo-crawl.service",
+    "siax-enrich.service",
+    "siax-laptop-scanner.service",
+]
+
+
+def _runs(unit, days=3):
+    try:
+        out = subprocess.run(
+            [
+                "journalctl",
+                "--user",
+                "-u",
+                unit,
+                "--since",
+                f"{days} days ago",
+                "-o",
+                "short-iso",
+                "--no-pager",
+            ],
+            capture_output=True,
+            text=True,
+            timeout=30,
+        ).stdout.splitlines()
+    except Exception:
+        return []
+    runs, cur = [], None
+    for line in out:
+        m = re.match(
+            r"(\S+) \S+ systemd\[\d+\]: (Started|Stopped|.*Failed with result '([^']+)')",
+            line,
+        )
+        if not m:
+            continue
+        ts, ev = m.group(1), m.group(2)
+        if ev == "Started":
+            cur = {"start": ts[11:16], "day": ts[:10]}
+        elif cur is not None and (ev == "Stopped" or "Failed" in ev):
+            cur["end"] = ts[11:16]
+            cur["result"] = m.group(3) if "Failed" in ev else "ok"
+            runs.append(cur)
+            cur = None
+    if cur is not None:
+        cur["end"] = "pågår"
+        cur["result"] = "running"
+        runs.append(cur)
+    return runs[-12:]
+
+
+def _schedules():
+    out = []
+    try:
+        txt = subprocess.run(
+            ["systemctl", "--user", "list-timers", "--no-pager"],
+            capture_output=True,
+            text=True,
+            timeout=15,
+        ).stdout
+        for line in txt.splitlines():
+            for key in ("p0lytics", "sf-scrape", "proton", "opencode-db", "sloso"):
+                if key in line:
+                    parts = line.split()
+                    out.append(
+                        {
+                            "next": parts[0] + " " + parts[1]
+                            if len(parts) > 2
+                            else parts[0],
+                            "unit": next((p for p in parts if ".timer" in p), "?"),
+                        }
+                    )
+                    break
+    except Exception:
+        pass
+    try:
+        cron = subprocess.run(
+            ["crontab", "-l"], capture_output=True, text=True, timeout=10
+        ).stdout
+        for line in cron.splitlines():
+            if (
+                line.strip()
+                and not line.startswith("#")
+                and any(k in line for k in ("p0lytics", "plane-sync", "valkompass"))
+            ):
+                out.append({"next": "cron", "unit": line[:110]})
+    except Exception:
+        pass
+    return out
+
+
+def _proxy():
+    try:
+        d = json.load(open(os.path.join(SF, "proxies/pool.json")))
+        recs = d.get("records", [])
+        now = time.time()
+        fresh = sum(
+            1
+            for r in recs
+            if r.get("lastChecked") and now - _ts(r["lastChecked"]) < 3600
+        )
+        healthy = sum(1 for r in recs if (r.get("consecutiveFailures") or 0) == 0)
+        lat = [r["latencyMsEwma"] for r in recs if r.get("latencyMsEwma")]
+        lat.sort()
+        countries = Counter(r.get("country", "?") for r in recs).most_common(5)
+        return {
+            "total": len(recs),
+            "fresh_1h": fresh,
+            "healthy": healthy,
+            "lat_p50": round(lat[len(lat) // 2]) if lat else -1,
+            "top_countries": countries,
+        }
+    except Exception as e:
+        return {"error": str(e)[:100]}
+
+
+def _ts(iso):
+    try:
+        return (
+            time.mktime(time.strptime(iso[:19], "%Y-%m-%dT%H:%M:%S"))
+            - time.altzone * 0
+            + (7200 if iso.endswith("Z") else 0)
+        )
+    except Exception:
+        return 0
+
+
+def _wc(path):
+    try:
+        with open(path, "rb") as f:
+            return sum(1 for _ in f)
+    except Exception:
+        return -1
+
+
+def _pipeline():
+    stages = {}
+    b = os.path.join(SF, "full_data/biluppgifter")
+    stages["biluppgifter frontier (väntar)"] = _wc(os.path.join(b, "frontier.txt"))
+    stages["biluppgifter crawled"] = _wc(os.path.join(b, "crawled.txt"))
+    stages["biluppgifter prio-kö"] = _wc(os.path.join(b, "pending_priority.txt"))
+    m = os.path.join(SF, "full_data/multi")
+    try:
+        for src in sorted(os.listdir(m)):
+            cp = os.path.join(m, src, "crawled.txt")
+            if os.path.exists(cp):
+                stages[f"multi/{src} crawled"] = _wc(cp)
+    except Exception:
+        pass
+    return stages
+
+
+def ops():
+    if time.time() - CACHE["ts"] < 300:
+        return CACHE["data"]
+    runs = {u: _runs(u) for u in UNITS}
+    data = {
+        "runs": runs,
+        "schedules": _schedules(),
+        "proxy": _proxy(),
+        "pipeline": _pipeline(),
+        "alerts": [],
+        "disk_free_gb": 0,
+    }
+    try:
+        data["disk_free_gb"] = round(shutil.disk_usage("/").free / 1e9, 1)
+    except Exception:
+        pass
+    A = data["alerts"]
+    if data["disk_free_gb"] and data["disk_free_gb"] < 6:
+        A.append(
+            {
+                "level": "crit",
+                "text": f"Disken nästan full: {data['disk_free_gb']} GB fritt",
+            }
+        )
+    try:
+        clog = (
+            open(os.path.join(CI, "crawl.log"), errors="replace")
+            .read()
+            .splitlines()[-3:]
+        )
+        if clog and all("cooldown" in l for l in clog):
+            A.append(
+                {
+                    "level": "warn",
+                    "text": "car.info stampar i cooldown-loop (tunna modeller)",
+                }
+            )
+    except Exception:
+        pass
+    try:
+        if "not found" in open("/tmp/p0lytics-health.log", errors="replace").read():
+            A.append(
+                {
+                    "level": "err",
+                    "text": "p0lytics-cron trasig (.venv saknas) — ingen health-data",
+                }
+            )
+    except Exception:
+        pass
+    CACHE.update(ts=time.time(), data=data)
+    return data
@@ -0,0 +1,131 @@
+#!/usr/bin/env python3
+"""Scraper-detaljsidor: /scraper/<id> - spec + live-data per scraper."""
+
+SCRAPERS = {
+    "sf": {
+        "label": "SF multi-scrape", "unit": "sf-multi-scrape.service",
+        "desc": "Multi-source vehicle scraper: biluppgifter/fordonskollen via pool, fordonssok/car.info direkt",
+        "cmd": "Documents/SF_Extractor/.venv/bin/python scrape_sources.py",
+        "sources": ["biluppgifter", "fordonskollen", "fordonssok", "carinfo"],
+        "spec": [
+            ("Källor", "biluppgifter.se, fordonskollen.se, fordonssok.se, car.info (sv-se)"),
+            ("Indata", "scrape_urls/*.txt (14,5M regnr, sitemap-enumererade)"),
+            ("Utdata", "spool/*.jsonl.gz (rå HTML) + full_data/*/records-*.jsonl.gz (extraherat)"),
+            ("Takt", "~4 parallella per domän; biluppgifter via pool (CF IP-block)"),
+            ("Drift", "systemd always-restart (startar om efter stop-timeout, återupptar)"),
+            ("Logg", "Documents/SF_Extractor/logs/multi_scrape.log — [källa] n= ok= fail= @pos/total"),
+        ],
+    },
+    "carinfo": {
+        "label": "car.info specs", "unit": "carinfo-crawl.service",
+        "desc": "car.info spec crawler, resumable (vehiq VEH-HIST-030)",
+        "cmd": "/usr/bin/python3 ~/.local/share/carinfo/crawl.py",
+        "sources": ["car.info"],
+        "spec": [
+            ("Källa", "car.info modell/generation/motorvarianter (en-se)"),
+            ("Utdata", "~/.local/share/carinfo/trims.jsonl {url, make, model, specs}"),
+            ("Checkpoint", "ckpt.json (makes_done, models_done, gens_done) — återupptar"),
+            ("Drift", "systemd; cooldown-loop vid tunna modeller"),
+            ("Logg", "~/.local/share/carinfo/crawl.log"),
+        ],
+    },
+    "enrich": {
+        "label": "SIAX enrich", "unit": "siax-enrich.service",
+        "desc": "Job Description Enricher (LinkedIn jobbsidor)",
+        "cmd": "job-automation/deploy/scanner/enrich-descriptions-laptop.py",
+        "sources": ["linkedin"],
+        "spec": [
+            ("Källa", "LinkedIn jobbsidor"),
+            ("Utdata", "ingen lokal resultatfil (fjeder/nätverk) — endast status synlig"),
+            ("Drift", "systemd"),
+        ],
+    },
+    "scanner": {
+        "label": "SIAX laptop-scanner", "unit": "siax-laptop-scanner.service",
+        "desc": "Laptop Job Scanner (residential IP, pushar till server6)",
+        "cmd": "job-automation/deploy/scanner/laptop-scanner.py",
+        "sources": ["jobbsajter"],
+        "spec": [
+            ("Källa", "jobbsajter via residential IP"),
+            ("Utdata", "pushas till server6 — ingen lokal resultatfil"),
+            ("Drift", "systemd"),
+        ],
+    },
+}
+
+DETAIL = """<!DOCTYPE html><html lang="sv"><head><meta charset="utf-8">
+<meta name="viewport" content="width=device-width,initial-scale=1">
+<title>Scrape Console · __LABEL__</title>
+<style>body{background:#0d1117;color:#e6edf3;font-family:system-ui,sans-serif;margin:0;padding:16px}
+a{color:#58a6ff}h1{font-size:20px;margin:0 0 4px}h2{font-size:15px;margin:16px 0 6px}
+.card{background:#161b22;border:1px solid #30363d;border-radius:8px;padding:12px;margin-bottom:12px}
+table{border-collapse:collapse;width:100%;font-size:12px}th,td{border:1px solid #30363d;padding:5px 8px;text-align:left;vertical-align:top}
+th{background:#0d1117;width:140px;color:#8b949e}
+.badge{display:inline-block;font-size:11px;padding:2px 8px;border-radius:10px}
+.on{background:#1a7f37}.off{background:#8c1515}
+.kv{font-size:12px;color:#8b949e;display:flex;gap:12px;flex-wrap:wrap}.kv b{color:#e6edf3}
+.bar{background:#30363d;border-radius:4px;height:8px;margin:4px 0 8px}.bar>i{display:block;height:8px;border-radius:4px;background:#1f6feb}
+.src{font-size:12px;margin:2px 0}.src b{color:#e6edf3}
+pre{background:#0d1117;border:1px solid #30363d;border-radius:6px;padding:8px;font-size:11px;overflow:auto;max-height:200px;white-space:pre-wrap}
+button{background:#21262d;color:#e6edf3;border:1px solid #30363d;border-radius:6px;padding:4px 10px;margin-right:4px;cursor:pointer;font-size:12px}</style></head>
+<body><h1>🕷️ <a href="/">Scrape Console</a> · <span id="t"></span></h1>
+<div class="card"><span id="badge"></span> <span id="ctl"></span><div class="kv" id="kv"></div><div id="prog"></div></div>
+<div class="card"><h2>Specifikation</h2><table id="spec"></table><div class="kv" id="cmd" style="margin-top:8px"></div></div>
+<div class="card"><h2>Dataset</h2><div id="ds"></div></div>
+<div class="card"><h2>Logg</h2><pre id="log"></pre></div>
+<div class="card"><h2>Körningar (runs-historik, 3 dygn)</h2><div id="runs"></div></div>
+<div class="card" id="pipeCard" style="display:none"><h2>Pipeline-steg</h2><div id="pipe"></div></div>
+<div class="card" id="proxyCard" style="display:none"><h2>Proxy-pool (IP-rotation)</h2><div id="proxy"></div></div>
+<div class="card"><h2>Schema</h2><div id="sched"></div></div>
+<script>
+const ID = "__ID__", LOCAL = ['localhost','127.0.0.1'].includes(location.hostname);
+async function tick(){
+  const s = await (await fetch('/state')).json();
+  const c = s.scrapers.find(x=>x.unit=== "__UNIT__");
+  const meta = await (await fetch('/scraper-meta?id='+ID)).json();
+  document.title = 'Scrape Console · ' + meta.label;
+  document.getElementById('t').textContent = meta.label;
+  const on = c && c.active==='active';
+  document.getElementById('badge').innerHTML = `<span class="badge ${on?'on':'off'}">${c?c.active+' · pid '+c.pid:'okänd'}</span>`;
+  document.getElementById('ctl').innerHTML = LOCAL
+    ? `<button onclick="ctl('restart')">↻ Omstart</button><button onclick="ctl('stop')">⏹ Stopp</button><button onclick="ctl('start')">▶ Start</button><a href="/data-page?scraper=${meta.key}"><button>Alla data →</button></a>`
+    : `<a href="/data-page?scraper=${meta.key}"><button>Alla data →</button></a>`;
+  document.getElementById('kv').innerHTML = c ? `<span>minne <b>${c.mem_mb} MB</b></span><span>cpu <b>${c.cpu_s}s</b></span><span>start <b>${c.started||'-'}</b></span>` : '';
+  let ph = '';
+  if (c && c.progress) for (const [k,v] of Object.entries(c.progress)) {
+    if (k.startsWith('_')) continue;
+    const pct = 100*v.pos/v.total, okr = (100*v.ok/Math.max(1,v.ok+v.fail)).toFixed(1);
+    ph += `<div class="src"><b>${k}</b> ${v.ok} ok / ${v.fail} fail (ok-rate ${okr}%) · ${v.pos}/${v.total} (${pct.toFixed(2)}%)</div><div class="bar"><i style="width:${Math.min(100,pct)}%"></i></div>`;
+  }
+  if (c && c.ckpt) ph += `<div class="src">checkpoint: <b>${Object.entries(c.ckpt).map(([k,v])=>k+'='+v).join(' · ')}</b></div>`;
+  document.getElementById('prog').innerHTML = ph;
+  document.getElementById('spec').innerHTML = meta.spec.map(([k,v])=>`<tr><th>${k}</th><td>${v}</td></tr>`).join('');
+  document.getElementById('cmd').innerHTML = `<span>kommando: <b>${meta.cmd}</b></span><span>${meta.desc}</span>`;
+  const ds = await (await fetch('/datasets')).json();
+  document.getElementById('ds').innerHTML = ds.filter(d=>d.scraper===meta.key).map(d=>`<div class="src"><b>${d.label}</b> — ${d.rows} rader · <a href="/data-page?ds=${d.id}">öppna tabell →</a></div>`).join('') || '<div class="kv">inga lokala dataset (statusläge)</div>';
+  document.getElementById('log').textContent = (c && c.log) || '(ingen logg)';
+  const ops = await (await fetch('/ops')).json();
+  const runs = (ops.runs && ops.runs["__UNIT__"]) || [];
+  document.getElementById('runs').innerHTML = runs.length ? runs.slice().reverse().map(r =>
+    `<div class="src"><b>${r.day||''} ${r.start} → ${r.end}</b> · ${r.result}</div>`).join('')
+    : '<div class="kv">inga journalförda starter senaste 3 dygnen</div>';
+  document.getElementById('sched').innerHTML = (ops.schedules||[]).map(s =>
+    `<div class="src">nästa <b>${s.next}</b> · ${String(s.unit).slice(0,80)}</div>`).join('') || '<div class="kv">inga scheman</div>';
+  if (ID === 'sf') {
+    document.getElementById('pipeCard').style.display = 'block';
+    document.getElementById('proxyCard').style.display = 'block';
+    document.getElementById('pipe').innerHTML = Object.entries(ops.pipeline||{}).map(([k,v]) =>
+      `<div class="src">${k}: <b>${v}</b></div>`).join('');
+    const px = ops.proxy||{};
+    document.getElementById('proxy').innerHTML = px.error ? px.error :
+      `<div class="kv"><span>proxies <b>${px.total}</b></span><span>fräscha &lt;1h <b>${px.fresh_1h}</b></span><span>friska <b>${px.healthy}</b></span><span>latens p50 <b>${px.lat_p50} ms</b></span></div>
+       <div class="src">topp-länder: <b>${(px.top_countries||[]).map(([c,n])=>c+' '+n).join(' · ')}</b></div>`;
+  }
+}
+async function ctl(action){
+  if(!confirm(action+'?')) return;
+  const r = await fetch('/control',{method:'POST',headers:{'Content-Type':'application/json'},body:JSON.stringify({unit:"__UNIT__",action})});
+  alert(await r.text()); tick();
+}
+tick(); setInterval(tick, 10000);
+</script></body></html>"""
@@ -0,0 +1,406 @@
+#!/usr/bin/env python3
+"""Scrape Console v2: live-översikt + historik + kontroller (localhost-only).
+GET / /state /history /log?src=multi&n=50 · POST /control (127.0.0.1 only)."""
+
+import json, os, re, shutil, subprocess, time
+from http.server import BaseHTTPRequestHandler, HTTPServer
+from urllib.parse import urlparse, parse_qs
+import data_views
+import dash_views
+import scraper_views
+import ops_views
+
+HOME = os.path.expanduser("~")
+SF = os.path.join(HOME, "Documents/SF_Extractor")
+CI = os.path.join(HOME, ".local/share/carinfo")
+HIST = os.path.join(HOME, ".local/share/scrape-console/history.jsonl")
+CACHE = {"ts": 0, "data": None}
+
+
+def sys_show(unit, *props):
+    try:
+        out = subprocess.run(
+            ["systemctl", "--user", "show", unit, "-p", ",".join(props)],
+            capture_output=True,
+            text=True,
+            timeout=10,
+        ).stdout
+        d = {}
+        for line in out.splitlines():
+            if "=" in line:
+                k, v = line.split("=", 1)
+                d[k] = v
+        return d
+    except Exception:
+        return {}
+
+
+def svc(unit, label, desc):
+    p = sys_show(
+        unit,
+        "ActiveState",
+        "MainPID",
+        "MemoryCurrent",
+        "CPUUsageNSec",
+        "ExecMainStartTimestamp",
+    )
+    mem = int(p.get("MemoryCurrent") or 0)
+    return {
+        "unit": unit,
+        "label": label,
+        "desc": desc,
+        "active": p.get("ActiveState", "?"),
+        "pid": p.get("MainPID", "?"),
+        "mem_mb": round(mem / 1024 / 1024, 1),
+        "cpu_s": round(int(p.get("CPUUsageNSec") or 0) / 1e9, 1),
+        "started": (p.get("ExecMainStartTimestamp") or "")[4:16],
+    }
+
+
+def tail(path, n=8):
+    try:
+        with open(path, errors="replace") as f:
+            return "".join(f.readlines()[-n:])
+    except Exception as e:
+        return f"(kan inte läsa {path}: {e})"
+
+
+def sf_progress():
+    out = {}
+    try:
+        with open(os.path.join(SF, "logs/multi_scrape.log"), errors="replace") as f:
+            for line in f:
+                m = re.match(
+                    r"\[(.+?)\]\s+n=(\d+)\s+ok=(\d+)\s+fail=(\d+)\s+@(\d+)/(\d+)",
+                    line.strip(),
+                )
+                if m:
+                    src, n, ok, fail, pos, total = m.groups()
+                    out[src] = {
+                        "n": int(n),
+                        "ok": int(ok),
+                        "fail": int(fail),
+                        "pos": int(pos),
+                        "total": int(total),
+                    }
+    except Exception as e:
+        out["_error"] = str(e)
+    return out
+
+
+def du(path):
+    try:
+        return round(
+            sum(
+                os.path.getsize(os.path.join(dp, f))
+                for dp, _, fns in os.walk(path)
+                for f in fns
+            )
+            / 1024
+            / 1024,
+            1,
+        )
+    except Exception:
+        return -1
+
+
+def collect():
+    try:
+        d = json.load(open(os.path.join(CI, "ckpt.json")))
+        ci_ckpt = {k: (len(v) if isinstance(v, list) else v) for k, v in d.items()}
+    except Exception as e:
+        ci_ckpt = {"_error": str(e)[:100]}
+    try:
+        spool = len(os.listdir(os.path.join(SF, "spool")))
+    except Exception:
+        spool = -1
+    prog = sf_progress()
+    mem = {
+        s["unit"]: s["mem_mb"]
+        for s in [
+            svc("sf-multi-scrape.service", "", ""),
+            svc("carinfo-crawl.service", "", ""),
+        ]
+    }
+    data = {
+        "ts": time.strftime("%H:%M:%S"),
+        "scrapers": [
+            dict(
+                svc(
+                    "sf-multi-scrape.service",
+                    "SF multi-scrape",
+                    "fordon: biluppgifter/fordonskollen/fordonssok/car.info",
+                ),
+                progress=prog,
+                spool_files=spool,
+                data_mb={"full_data": du(os.path.join(SF, "full_data"))},
+                log=tail(os.path.join(SF, "logs/multi_scrape.log"), 6),
+            ),
+            dict(
+                svc(
+                    "carinfo-crawl.service", "car.info specs", "VEH-HIST-030, resumable"
+                ),
+                ckpt=ci_ckpt,
+                log=tail(os.path.join(CI, "crawl.log"), 6),
+            ),
+            dict(svc("siax-enrich.service", "SIAX enrich", "LinkedIn jobbsidor")),
+            dict(
+                svc(
+                    "siax-laptop-scanner.service",
+                    "SIAX laptop-scanner",
+                    "jobb, residential IP",
+                )
+            ),
+        ],
+        "cron": {
+            "p0lytics_health": tail("/tmp/p0lytics-health.log", 2).strip(),
+            "p0lytics_rate": tail("/tmp/p0lytics-rate.log", 2).strip(),
+        },
+    }
+    # historik (1 prov/min)
+    try:
+        last = 0
+        if os.path.exists(HIST):
+            with open(HIST, errors="replace") as f:
+                for line in f:
+                    pass
+                try:
+                    last = json.loads(line).get("t", 0)
+                except Exception:
+                    pass
+        now = int(time.time())
+        if now - last >= 55:
+            snap = {
+                "t": now,
+                "ok": {k: v["ok"] for k, v in prog.items() if not k.startswith("_")},
+                "mem": mem,
+            }
+            with open(HIST, "a") as f:
+                f.write(json.dumps(snap) + "\n")
+            # kapa till ~1500 rader
+            with open(HIST, errors="replace") as f:
+                lines = f.readlines()
+            if len(lines) > 1500:
+                open(HIST, "w").writelines(lines[-1500:])
+    except Exception:
+        pass
+    return data
+
+
+HTML = """<!DOCTYPE html><html lang="sv"><head><meta charset="utf-8">
+<meta name="viewport" content="width=device-width,initial-scale=1">
+<title>Scrape Console live</title>
+<style>body{background:#0d1117;color:#e6edf3;font-family:system-ui,sans-serif;margin:0;padding:16px}
+h1{font-size:20px;margin:0 0 4px}#ts{color:#8b949e;font-size:12px;margin-bottom:12px}
+.grid{display:grid;grid-template-columns:repeat(auto-fit,minmax(360px,1fr));gap:12px}
+.card{background:#161b22;border:1px solid #30363d;border-radius:8px;padding:12px}
+.card h2{font-size:15px;margin:0 0 2px}.desc{color:#8b949e;font-size:12px;margin-bottom:8px}
+.badge{display:inline-block;font-size:11px;padding:2px 8px;border-radius:10px;margin:0 4px 8px 0}
+.on{background:#1a7f37}.off{background:#8c1515}.qq{background:#9e6a03}
+.kv{font-size:12px;color:#8b949e;display:flex;gap:12px;flex-wrap:wrap;margin-bottom:6px}
+.kv b{color:#e6edf3}.bar{background:#30363d;border-radius:4px;height:8px;margin:4px 0 8px}.bar>i{display:block;height:8px;border-radius:4px;background:#1f6feb}
+.src{font-size:12px;margin:2px 0}.src b{color:#e6edf3}
+pre{background:#0d1117;border:1px solid #30363d;border-radius:6px;padding:8px;font-size:11px;overflow:auto;max-height:150px;white-space:pre-wrap}
+button{background:#21262d;color:#e6edf3;border:1px solid #30363d;border-radius:6px;padding:4px 10px;margin:0 4px 4px 0;cursor:pointer;font-size:12px}
+button:hover{background:#30363d}svg{width:100%;height:44px;margin:2px 0 6px}</style></head>
+<body><h1>🕷️ Scrape Console <span style="font-size:12px;color:#8b949e">live</span> <a href="/data-page" style="font-size:13px">Alla data →</a> <a href="/dash" style="font-size:13px">Min dashboard →</a></h1>
+<div id="ts"></div><div id="alerts" style="margin-bottom:12px"></div><div class="grid" id="g"></div>
+<script>
+const LOCAL = ['localhost','127.0.0.1'].includes(location.hostname);
+let hist = [];
+let HISTG = [];
+async function tick(){
+  const s = await (await fetch('/state')).json();
+  try { HISTG = await (await fetch('/history')).json(); } catch(e) {}
+  try {
+    const ops = await (await fetch('/ops')).json();
+    const col = {crit:'#8c1515', err:'#8c1515', warn:'#9e6a03'};
+    document.getElementById('alerts').innerHTML = (ops.alerts||[]).map(x =>
+      `<div class="card" style="border-left:4px solid ${col[x.level]||'#666'};margin-bottom:8px"><b>${x.level.toUpperCase()}</b> · ${x.text}</div>`).join('');
+  } catch(e) {}
+  document.getElementById('ts').textContent = 'Uppdaterad ' + s.ts + ' (auto 10 s)' + (LOCAL ? '' : ' · skrivskyddad vy');
+  hist = await (await fetch('/history')).json();
+  const g = document.getElementById('g'); g.innerHTML = '';
+  for (const c of s.scrapers) {
+    const on = c.active === 'active';
+    let h = `<div class="card"><span class="badge ${on?'on':'off'}">${c.active} · pid ${c.pid}</span>`;
+    if (LOCAL) h += `<button onclick="ctl('${c.unit}','restart')">↻</button><button onclick="ctl('${c.unit}','stop')">⏹</button><button onclick="ctl('${c.unit}','start')">▶</button>`;
+    const SCR0 = {'SF multi-scrape':'sf','car.info specs':'carinfo','SIAX enrich':'enrich','SIAX laptop-scanner':'scanner'};
+    const sid0 = SCR0[c.label] || '';
+    h += `<h2><a href="/scraper/${sid0}" style="color:#e6edf3">${c.label}</a></h2><div class="desc">${c.desc}<br>${c.unit}</div><div class="kv"><span>minne <b>${c.mem_mb} MB</b></span><span>cpu <b>${c.cpu_s}s</b></span><span>start <b>${c.started||'-'}</b></span></div>`;
+    if (sid0) h += `<div class="src"><a href="/data-page?scraper=${sid0}">Visa all data →</a></div>`;
+    if (c.progress) for (const [k,v] of Object.entries(c.progress)) {
+      if (k.startsWith('_')) continue;
+      const pct = (100*v.pos/v.total);
+      const okr = (100*v.ok/Math.max(1,v.ok+v.fail)).toFixed(1);
+      const eta = etaStr(k, v);
+      h += `<div class="src"><b>${k}</b> ${v.ok} ok / ${v.fail} fail (ok-rate ${okr}%) · ${v.pos}/${v.total} (${pct.toFixed(2)}%)${eta}</div><div class="bar"><i style="width:${Math.min(100,pct)}%"></i></div>`;
+      h += spark(k);
+    }
+    if (c.spool_files !== undefined) h += `<div class="src">spool-filer: <b>${c.spool_files}</b> · full_data: <b>${c.data_mb.full_data} MB</b></div>`;
+    if (c.ckpt) h += `<div class="src">checkpoint: <b>${Object.entries(c.ckpt).map(([k,v])=>k+'='+v).join(' · ')}</b></div>`;
+    if (c.log) h += `<pre>${c.log.replace(/</g,'&lt;')}</pre>`;
+    g.innerHTML += h + '</div>';
+  }
+  g.innerHTML += `<div class="card"><span class="badge qq">cron</span><h2>p0lytics monitors</h2><div class="desc">health var 15:e min · rate var 30:e</div><pre>health: ${s.cron.p0lytics_health.replace(/</g,'&lt;')}\nrate: ${s.cron.p0lytics_rate.replace(/</g,'&lt;')}</pre></div>`;
+}
+function etaStr(src, v){
+  const pts = HISTG.filter(h=>h.ok && h.ok[src]!==undefined).slice(-12);
+  if(pts.length < 2) return '';
+  const dv = pts[pts.length-1].ok[src]-pts[0].ok[src], dt = (pts[pts.length-1].t-pts[0].t)/60;
+  if(dv <= 0 || dt <= 0) return '';
+  const left = (v.total-v.pos)/(dv/dt), h = Math.floor(left/60), m = Math.round(left%60);
+  return ` · ETA ${h>0?h+'h ':''}${m}m`;
+}
+function spark(src){
+  const pts = hist.filter(h=>h.ok && h.ok[src]!==undefined).slice(-60);
+  if (pts.length < 2) return '';
+  const vals = pts.map(h=>h.ok[src]); const max = Math.max(...vals), min = Math.min(...vals);
+  const W=300,H=40, path = vals.map((v,i)=>`${(i/(vals.length-1)*W).toFixed(1)},${(H-2-((v-min)/Math.max(1,max-min))*(H-4)).toFixed(1)}`).join(' ');
+  const rate = ((vals[vals.length-1]-vals[0])/Math.max(1,(pts[pts.length-1].t-pts[0].t)/60)).toFixed(0);
+  return `<svg viewBox="0 0 ${W} ${H}"><polyline points="${path}" fill="none" stroke="#1f6feb" stroke-width="1.5"/></svg><div class="src">takt: <b>${rate}/min</b> senaste timmen</div>`;
+}
+async function ctl(unit,action){
+  if(!confirm(action+' '+unit+'?')) return;
+  const r = await fetch('/control',{method:'POST',headers:{'Content-Type':'application/json'},body:JSON.stringify({unit,action})});
+  alert(await r.text()); tick();
+}
+tick(); setInterval(tick, 10000);
+</script></body></html>"""
+
+
+class H(BaseHTTPRequestHandler):
+    def log_message(self, *a):
+        pass
+
+    def _send(self, body, ctype):
+        self.send_response(200)
+        self.send_header("Content-Type", ctype)
+        self.send_header("Content-Length", str(len(body)))
+        self.end_headers()
+        self.wfile.write(body)
+
+    def do_GET(self):
+        u = urlparse(self.path)
+        if u.path == "/state":
+            if time.time() - CACHE["ts"] > 10:
+                CACHE["data"] = collect()
+                CACHE["ts"] = time.time()
+            self._send(json.dumps(CACHE["data"]).encode(), "application/json")
+        elif u.path == "/history":
+            try:
+                with open(HIST, errors="replace") as f:
+                    lines = f.readlines()
+                data = [json.loads(l) for l in lines[-90:] if l.strip()]
+            except Exception:
+                data = []
+            self._send(json.dumps(data).encode(), "application/json")
+        elif u.path == "/log":
+            q = parse_qs(u.query)
+            src = q.get("src", ["multi"])[0]
+            n = min(int(q.get("n", ["50"])[0]), 200)
+            path = {
+                "multi": os.path.join(SF, "logs/multi_scrape.log"),
+                "carinfo": os.path.join(CI, "crawl.log"),
+            }.get(src, "")
+            self._send(tail(path, n).encode(), "text/plain; charset=utf-8")
+        elif u.path == "/data-page":
+            self._send(data_views.PAGE.encode(), "text/html; charset=utf-8")
+        elif u.path == "/dash":
+            self._send(dash_views.DASH_PAGE.encode(), "text/html; charset=utf-8")
+        elif u.path == "/datasets":
+            counts = data_views.counts()
+            out = [
+                {
+                    "id": d["id"],
+                    "label": d["label"],
+                    "scraper": data_views.SCRAPER_OF.get(d["id"], ""),
+                    "rows": counts.get(d["id"], {}).get("rows", -1),
+                }
+                for d in data_views.DATASETS
+            ]
+            self._send(json.dumps(out).encode(), "application/json")
+        elif u.path == "/data":
+            q = parse_qs(u.query)
+            res = data_views.query(
+                q.get("ds", [""])[0],
+                q.get("q", [""])[0],
+                int(q.get("page", ["1"])[0]),
+                sort=q.get("sort", [None])[0],
+                sdir=q.get("dir", ["asc"])[0],
+            )
+            self._send(json.dumps(res, ensure_ascii=False).encode(), "application/json")
+        elif u.path == "/record":
+            q = parse_qs(u.query)
+            res = data_views.record_at(
+                q.get("ds", [""])[0], int(q.get("n", ["0"])[0]), q.get("q", [""])[0]
+            )
+            self._send(json.dumps(res, ensure_ascii=False).encode(), "application/json")
+        elif u.path == "/export":
+            q = parse_qs(u.query)
+            body, ctype, fname = data_views.export(
+                q.get("ds", [""])[0], q.get("q", [""])[0], q.get("fmt", ["csv"])[0]
+            )
+            self.send_response(200)
+            self.send_header("Content-Type", ctype)
+            self.send_header("Content-Disposition", "attachment; filename=" + fname)
+            self.send_header("Content-Length", str(len(body)))
+            self.end_headers()
+            self.wfile.write(body)
+        elif u.path.startswith("/scraper/"):
+            sid = u.path.split("/scraper/")[-1].split("?")[0]
+            sc = scraper_views.SCRAPERS.get(sid)
+            if not sc:
+                self.send_response(404); self.end_headers(); return
+            html = scraper_views.DETAIL.replace("__ID__", sid).replace(
+                "__UNIT__", sc["unit"]).replace("__LABEL__", sc["label"])
+            self._send(html.encode(), "text/html; charset=utf-8")
+        elif u.path == "/scraper-meta":
+            q = parse_qs(u.query)
+            sid = q.get("id", [""])[0]
+            sc = scraper_views.SCRAPERS.get(sid, {})
+            self._send(json.dumps({"key": sid, "label": sc.get("label", sid),
+                                   "desc": sc.get("desc", ""), "cmd": sc.get("cmd", ""),
+                                   "spec": sc.get("spec", [])}).encode(),
+                       "application/json")
+        elif u.path == "/ops":
+            self._send(json.dumps(ops_views.ops()).encode(), "application/json")
+        else:
+            self._send(HTML.encode(), "text/html; charset=utf-8")
+
+    def do_POST(self):
+        if self.path != "/control":
+            self.send_response(404)
+            self.end_headers()
+            return
+        if self.client_address[0] not in ("127.0.0.1", "::1"):
+            self.send_response(403)
+            self.end_headers()
+            self.wfile.write(b"localhost only")
+            return
+        try:
+            d = json.loads(self.rfile.read(int(self.headers["Content-Length"] or 0)))
+            unit, action = d["unit"], d["action"]
+            assert re.match(r"^[a-z0-9@._-]+\.service$", unit) and action in (
+                "start",
+                "stop",
+                "restart",
+            )
+            r = subprocess.run(
+                ["systemctl", "--user", action, unit],
+                capture_output=True,
+                text=True,
+                timeout=60,
+            )
+            msg = f"{action} {unit}: {'OK' if r.returncode == 0 else r.stderr[:200]}"
+        except Exception as e:
+            msg = f"fel: {e}"
+        body = msg.encode()
+        self.send_response(200)
+        self.send_header("Content-Length", str(len(body)))
+        self.end_headers()
+        self.wfile.write(body)
+
+
+if __name__ == "__main__":
+    HTTPServer(("127.0.0.1", 8899), H).serve_forever()
				`@@ -0,0 +1 @@`
				{"biluppgifter_records": {"rows": 3592, "bytes": -1, "key": "16:14914251"}, "sf_spool": {"rows": 0, "bytes": -1, "key": "0:0"}, "carinfo_trims": {"rows": 407, "bytes": -1, "key": "1791395435.6649714:141170"}, "vehicles_clean": {"rows": 119714, "bytes": -1, "key": "1786478796.0:57346606"}, "allabolag": {"rows": 204961, "bytes": -1, "key": "1786478742.0:81531189"}, "hitta": {"rows": 88104, "bytes": -1, "key": "1786478749.0:7202748"}, "ratsit": {"rows": 229349, "bytes": -1, "key": "1786478755.0:12276419"}, "bolagsfakta": {"rows": 888661, "bytes": -1, "key": "1786478747.0:18001402"}, "hitta_companies": {"rows": 8, "bytes": -1, "key": "1790959874.31363:2577"}, "queue_biluppgifter": {"rows": 1819062, "bytes": -1, "key": "1791204095.6476479:86491329"}, "queue_biluppgifter_crawled": {"rows": 213065, "bytes": -1, "key": "1791204095.6476479:9993850"}, "queue_multi_fordonssok": {"rows": 0, "bytes": -1, "key": "1791216576.2764611:0"}, "queue_multi_fordonskollen": {"rows": 0, "bytes": -1, "key": "1791216576.2764611:0"}, "queue_multi_carinfo": {"rows": 0, "bytes": -1, "key": "1791216576.2764611:0"}, "reco_companies": {"rows": 4, "bytes": -1, "key": "1790959888.311806:395"}, "hitta_person_reg": {"rows": 2000000, "bytes": -1, "key": "1791210077.6520803:305528424"}, "hitta_company_reg": {"rows": 1588677, "bytes": -1, "key": "1791219600.018239:103172372"}, "ratsit_person_reg": {"rows": 2000000, "bytes": -1, "key": "1791212987.7546084:661407740"}, "ratsit_company_reg": {"rows": 1312997, "bytes": -1, "key": "1791213028.0060341:74810637"}, "bolagsfakta_reg": {"rows": 1538781, "bytes": -1, "key": "1791217621.7285998:91414094"}, "allabolag_reg": {"rows": 1285485, "bytes": -1, "key": "1786478090.0:47562943"}, "relations": {"rows": 0, "bytes": -1, "key": "0:0"}, "queue_hitta_person": {"rows": 2000000, "bytes": -1, "key": "1791376301.2104456:198937064"}, "queue_hitta_company": {"rows": 921216, "bytes": -1, "key": "1791376445.3741066:57927832"}, "queue_ratsit_person": {"rows": 38112, "bytes": -1, "key": "1791233954.4740746:4089353"}, "queue_bolagsfakta_company": {"rows": 1520927, "bytes": -1, "key": "1791292143.7172053:90353412"}, "identified": {"rows": 0, "bytes": -1, "key": "0:0"}, "relations_v2": {"rows": 0, "bytes": -1, "key": "0:0"}}