fix: verify qsfptek cable details

packages/scraper/src/scrapers/qsfptek.ts

@@ -78,6 +78,17 @@ function sleep(ms: number): Promise<void> {
 }
 
 function detectReach(text: string): { label: string; meters: number } | undefined {
+  const generic = text.match(/\b(\d+(?:\.\d+)?)\s*(km|m)\b/i);
+  if (generic) {
+    const value = parseFloat(generic[1]);
+    const unit = generic[2].toLowerCase();
+    if (Number.isFinite(value) && value > 0) {
+      const meters = unit === "km" ? Math.round(value * 1000) : Math.max(1, Math.round(value));
+      const labelValue = String(value).replace(/\.0$/, "");
+      return { label: `${labelValue}${unit}`, meters };
+    }
+  }
+
   const patterns: [RegExp, string, number][] = [
     [/\b120\s*km\b/i, "120km", 120000],
     [/\b80\s*km\b/i, "80km", 80000],
@@ -100,15 +111,28 @@ function detectReach(text: string): { label: string; meters: number } | undefine
 }
 
 function detectFiber(text: string): string {
+  if (/copper|dac|twinax|rj.?45|base-t|\bmcp/i.test(text)) return "Copper";
+  if (/aoc|active.?optical|active.?fiber|\bmfs/i.test(text)) return "MMF";
   if (/single.?mode|smf|[^a-z]lx[^a-z]|[^a-z]lr[^a-z]|[^a-z]er[^a-z]|[^a-z]zr[^a-z]|bidi|cwdm|dwdm/i.test(text)) return "SMF";
   if (/multi.?mode|mmf|[^a-z]sx[^a-z]|[^a-z]sr[^a-z]/i.test(text)) return "MMF";
-  if (/copper|dac|twinax|rj.?45|base-t/i.test(text)) return "Copper";
   return "";
 }
 
 function detectWavelength(text: string): string {
+  if (/copper|dac|twinax|base-t|rj.?45|\bmcp/i.test(text)) return "N/A";
   const m = text.match(/(\d{3,4})\s*nm/i);
-  return m ? m[1] : "";
+  if (m) return m[1] === "1311" ? "1310" : m[1];
+  if (/\bCWDM4\b/i.test(text)) return "1271,1291,1311,1331";
+  if (/\b(?:SR|SR4|SR8|SRBD|VR|VR4|ESR4|CSR4)\b/i.test(text)) return "850";
+  if (/\b(?:DR|DR4|DR8|FR|FR4|FR8|LR|LR4|ER|ER4|PSM4|2DR4|2FR4)\b/i.test(text)) return "1310";
+  return "";
+}
+
+function detectCategory(text: string): string {
+  if (/breakout/i.test(text)) return "Cable Breakout";
+  if (/aoc|active.?optical|active.?fiber|\bmfs/i.test(text)) return "AOC Cable";
+  if (/copper|dac|twinax|base-t|rj.?45|\bmcp/i.test(text)) return "Cable";
+  return "DataCenter";
 }
 
 /**
@@ -268,6 +292,7 @@ export async function scrapeQsfptek(): Promise<void> {
       const txId = await findOrCreateScrapedTransceiver({
         partNumber: product.partNumber,
         vendorId,
+        productUrl: product.url,
         formFactor: product.formFactor,
         speedGbps: product.speedGbps,
         speed: product.speed,
@@ -275,7 +300,7 @@ export async function scrapeQsfptek(): Promise<void> {
         reachLabel: product.reachLabel,
         fiberType: product.fiberType,
         wavelengths: product.wavelength,
-        category: "DataCenter",
+        category: detectCategory(product.name),
       });
 
       // Price observation from listing page

sync/CURRENT.md

@@ -1,9 +1,33 @@
 # Current TIP Sync State
 
-Updated: 2026-05-09 14:54 UTC
+Updated: 2026-05-09 15:02 UTC
 
 ## Newest Work
 
+- QSFPTEK cable/AOC parser hardening and DB detail backfill on 2026-05-09:
+  - root cause:
+    - QSFPTEK scraper parsed catalog rows but did not pass `productUrl` into `findOrCreateScrapedTransceiver`
+    - generic leading cable lengths like `1m`, `2m`, `10m`, `15m`, `30m` were not parsed
+    - MFS/MCP AOC/DAC product families were not classified as cable/AOC products
+  - code hardened:
+    - `packages/scraper/src/scrapers/qsfptek.ts`
+      - parses generic `m/km` reach, including leading lengths
+      - classifies `MFS`/AOC/active fiber as `AOC Cable`
+      - classifies `MCP`/DAC/Copper/Twinax as `Cable`
+      - writes `productUrl` into the DB upsert
+      - sets Copper/DAC wavelength to `N/A`
+      - adds safe optical family wavelength parsing for future catalog runs
+  - DB correction:
+    - found `36` QSFPTEK rows missing details
+    - `28` had deterministic leading length and source URL
+    - updated those `28` with reach, cable/AOC classification and source-backed details
+    - `8` additional rows became fully verified after promotion
+  - deployment:
+    - synced patched QSFPTEK scraper to active `/opt/tip`
+    - `pnpm -C packages/scraper build` passed
+  - truth:
+    - QSFPTEK is now much closer, but remaining rows include long-reach 1G optics missing fiber/detail fields and should be handled separately by source parsing, not guessed
+
 - Copper/DAC reach/detail verification and comparable API semantics on 2026-05-09:
   - purpose:
     - continue toward full TIP verification without inventing optical data

sync/history/2026-05-09-qsfptek-cable-aoc-parser-and-backfill.md

@@ -0,0 +1,38 @@
+# QSFPTEK Cable AOC Parser And Backfill
+
+Date: 2026-05-09
+Actor: Codex
+Scope: QSFPTEK remaining detail gaps
+Mode: DB-only correction plus scraper parser hardening
+
+## Root Cause
+
+QSFPTEK still had detail gaps because:
+
+- the scraper did not pass `productUrl` to `findOrCreateScrapedTransceiver`
+- generic leading lengths like `1m`, `2m`, `10m`, `15m`, `30m` were not parsed
+- MFS/MCP cable families were not classified as AOC/DAC cable products
+
+## Code Change
+
+Patched `packages/scraper/src/scrapers/qsfptek.ts`:
+
+- parse generic meter/kilometer reach
+- classify `MFS`/AOC/active fiber as `AOC Cable`
+- classify `MCP`/DAC/Copper/Twinax as `Cable`
+- write `productUrl` in the DB upsert
+- set Copper/DAC wavelength to `N/A`
+- add safe optical family wavelength parsing for future catalog runs
+
+Synced to active `/opt/tip`; `pnpm -C packages/scraper build` passed.
+
+## DB Backfill
+
+- QSFPTEK rows missing details: `36`
+- deterministic cable/AOC rows with source URL and leading length: `28`
+- rows updated: `28`
+- additional rows promoted to `fully_verified`: `8`
+
+## Training Note
+
+TIPLLM should treat QSFPTEK `MFS...` rows as AOC/active fiber cable candidates and `MCP...` rows as DAC/Copper cable candidates, but only mark details verified when source URL and deterministic length are present.