feat: implement db layout caching for parsed ocr & reprocess batch test with corrected parser alignment

backend/pfm-web-app/src/app/api/parse/route.ts

@@ -68,75 +68,98 @@ export async function POST(req: NextRequest) {
 
     const b64 = fileBuffer.toString("base64");
 
-    // Form payload
-    const payload = {
-      file: b64,
-      matchHistoryJob: false,
-      useLayoutDetection: true,
-      fileType: 1,
-      useDocUnwarping: false,
-      useDocOrientationClassify: true
-    };
+    // Check if we already have a parsed document in the database with the exact filename (and has valid layout_parsing_result)
+    const cachedDoc = await query(
+      "SELECT layout_parsing_result, processing_logs FROM documents WHERE filename = $1 AND parsed = true",
+      [safeFile]
+    );
 
-    // Post to Pipeline API
-    const pipelineUrl = process.env.PIPELINE_URL || "http://localhost:7871/layout-parsing";
-    
-    console.log(`Forwarding request to pipeline API: ${pipelineUrl}`);
-    const response = await fetch(pipelineUrl, {
-      method: "POST",
-      headers: {
-        "Content-Type": "application/json"
-      },
-      body: JSON.stringify(payload)
-    });
-
-    if (!response.ok) {
-      const errText = await response.text();
-      console.warn(`Pipeline API error for ${safeFile}: ${errText}. Marking document as parsed with default metadata.`);
-      try {
-        await query(
-          "UPDATE documents SET parsed = true, metadata = $1 WHERE filename = $2",
-          [JSON.stringify({
-            tanggal: "Not Found",
-            noPO: "Not Found",
-            noSO: "Not Found",
-            noDO: "Not Found",
-            vendorInfo: "Not Found",
-            customerInfo: "Not Found",
-            items: []
-          }), safeFile]
-        );
-      } catch (dbErr) {
-        console.error("Failed to mark document as parsed on pipeline error:", dbErr);
-      }
-      return NextResponse.json({ error: `Pipeline API error: ${errText}` }, { status: response.status });
-    }
-
-    let data = await response.json();
-
-    // Check if the image is not straight (tilt > 1.0 degree)
-    const tilt = calculateAverageTilt(data);
+    let data: any = null;
+    let tilt = 0;
     let unwarped = false;
-    if (tilt > 1.0) {
-      console.log(`Parsed document ${safeFile} is not straight (average tilt: ${tilt.toFixed(2)} deg). Re-running with unwarping and orientation classification enabled...`);
-      const unwarpPayload = {
-        ...payload,
-        useDocUnwarping: true,
+
+    if (cachedDoc.rows.length > 0 && cachedDoc.rows[0].layout_parsing_result) {
+      console.log(`[Cache Hit] Found cached layout result for ${safeFile}. Skipping Pipeline API.`);
+      const rawData = cachedDoc.rows[0].layout_parsing_result;
+      const parsedData = typeof rawData === "string" ? JSON.parse(rawData) : rawData;
+      data = parsedData.result || parsedData;
+      
+      const procLogs = cachedDoc.rows[0].processing_logs;
+      if (procLogs) {
+        const parsedLogs = typeof procLogs === "string" ? JSON.parse(procLogs) : procLogs;
+        tilt = parsedLogs?.pipeline_info?.tilt !== undefined ? Number(parsedLogs.pipeline_info.tilt) : 0;
+        unwarped = parsedLogs?.pipeline_info?.unwarped ?? false;
+      }
+    } else {
+      // Form payload
+      const payload = {
+        file: b64,
+        matchHistoryJob: false,
+        useLayoutDetection: true,
+        fileType: 1,
+        useDocUnwarping: false,
         useDocOrientationClassify: true
       };
-      const unwarpResponse = await fetch(pipelineUrl, {
+
+      // Post to Pipeline API
+      const pipelineUrl = process.env.PIPELINE_URL || "http://localhost:7871/layout-parsing";
+      
+      console.log(`Forwarding request to pipeline API: ${pipelineUrl}`);
+      const response = await fetch(pipelineUrl, {
         method: "POST",
         headers: {
           "Content-Type": "application/json"
         },
-        body: JSON.stringify(unwarpPayload)
+        body: JSON.stringify(payload)
       });
-      if (unwarpResponse.ok) {
-        data = await unwarpResponse.json();
-        console.log(`Document unwarped successfully.`);
-        unwarped = true;
-      } else {
-        console.error(`Unwarping failed with status ${unwarpResponse.status}`);
+
+      if (!response.ok) {
+        const errText = await response.text();
+        console.warn(`Pipeline API error for ${safeFile}: ${errText}. Marking document as parsed with default metadata.`);
+        try {
+          await query(
+            "UPDATE documents SET parsed = true, metadata = $1 WHERE filename = $2",
+            [JSON.stringify({
+              tanggal: "Not Found",
+              noPO: "Not Found",
+              noSO: "Not Found",
+              noDO: "Not Found",
+              vendorInfo: "Not Found",
+              customerInfo: "Not Found",
+              items: []
+            }), safeFile]
+          );
+        } catch (dbErr) {
+          console.error("Failed to mark document as parsed on pipeline error:", dbErr);
+        }
+        return NextResponse.json({ error: `Pipeline API error: ${errText}` }, { status: response.status });
+      }
+
+      data = await response.json();
+
+      // Check if the image is not straight (tilt > 1.0 degree)
+      tilt = calculateAverageTilt(data);
+      if (tilt > 1.0) {
+        console.log(`Parsed document ${safeFile} is not straight (average tilt: ${tilt.toFixed(2)} deg). Re-running with unwarping and orientation classification enabled...`);
+        const unwarpPayload = {
+          ...payload,
+          useDocUnwarping: true,
+          useDocOrientationClassify: true
+        };
+        const unwarpResponse = await fetch(pipelineUrl, {
+          method: "POST",
+          headers: {
+            "Content-Type": "application/json"
+          },
+          body: JSON.stringify(unwarpPayload)
+        });
+        if (unwarpResponse.ok) {
+          data = await unwarpResponse.json();
+          console.log(`Document unwarped successfully.`);
+          unwarped = true;
+        } else {
+          console.error(`Unwarping failed with status ${unwarpResponse.status}`);
+        }
       }
     }