@thinkingai/ae-cli 6.1.16 → 6.1.17

package/README.md

@@ -200,7 +200,10 @@ ae-cli kb +add --name engineering-handbook --files '["./docs/guide.md","https://
 ae-cli kb +schema --name engineering-handbook
 ae-cli kb +compile --name engineering-handbook
 ae-cli kb +status --name engineering-handbook
-ae-cli kb +query -q "How is the sandbox configured?" --top-k 10
+ae-cli kb +ask -q "How is the sandbox configured?"
+# Submit only, poll later:
+ae-cli kb +ask -q "Another question" --no-wait
+ae-cli kb +ask-status --execution-id <id>
 ```
 
 External Agents can use deterministic retrieval without a server-side LLM:
@@ -208,7 +211,7 @@ External Agents can use deterministic retrieval without a server-side LLM:
 ```bash
 ae-cli kb +list
 ae-cli kb +index --sources '[{"scope":"company","name":"engineering-handbook"}]'
-ae-cli kb +grep -q "sandbox config" --sources '[{"scope":"company","name":"engineering-handbook"}]'
+ae-cli kb +grep -q "sandbox config" --sources '[{"scope":"company","name":"engineering-handbook"}]' --paths '["wiki/sandbox.md"]'
 ae-cli kb +read --source '{"scope":"company","name":"engineering-handbook"}' --path "wiki/sandbox.md"
 ```
 

package/README.zh.md

@@ -200,7 +200,10 @@ ae-cli kb +add --name engineering-handbook --files '["./docs/guide.md","https://
 ae-cli kb +schema --name engineering-handbook
 ae-cli kb +compile --name engineering-handbook
 ae-cli kb +status --name engineering-handbook
-ae-cli kb +query -q "如何配置沙盒？" --top-k 10
+ae-cli kb +ask -q "如何配置沙盒？"
+# 仅提交，后续轮询：
+ae-cli kb +ask -q "另一个问题" --no-wait
+ae-cli kb +ask-status --execution-id <id>
 ```
 
 外部 Agent 可以使用不依赖服务端 LLM 的确定性检索：
@@ -208,7 +211,7 @@ ae-cli kb +query -q "如何配置沙盒？" --top-k 10
 ```bash
 ae-cli kb +list
 ae-cli kb +index --sources '[{"scope":"company","name":"engineering-handbook"}]'
-ae-cli kb +grep -q "沙盒配置" --sources '[{"scope":"company","name":"engineering-handbook"}]'
+ae-cli kb +grep -q "沙盒配置" --sources '[{"scope":"company","name":"engineering-handbook"}]' --paths '["wiki/sandbox.md"]'
 ae-cli kb +read --source '{"scope":"company","name":"engineering-handbook"}' --path "wiki/sandbox.md"
 ```
 

package/dist/{data-integration-2MYMANJI.js → data-integration-XQYB4X4F.js}

@@ -7,6 +7,94 @@ import "./chunk-QGM4M3NI.js";
 // src/commands/data-integration/local-data/inspect.ts
 import { basename as basename3 } from "path";
 
+// src/commands/data-integration/local-data/estimate.ts
+var XLS_SIZE_WARN_BYTES = 100 * 1024 * 1024;
+var LARGE_FILE_WARN_BYTES = 1024 * 1024 * 1024;
+var XLS_HARD_LIMIT_BYTES = 1024 * 1024 * 1024;
+var THROUGHPUT_BYTES_PER_SECOND = {
+  jsonl: 10 * 1024 * 1024,
+  json: 8 * 1024 * 1024,
+  csv: 4 * 1024 * 1024,
+  tsv: 4 * 1024 * 1024,
+  xlsx: 2 * 1024 * 1024
+};
+var SLOW_ENCODINGS = /* @__PURE__ */ new Set(["gbk", "gb2312", "gb18030", "big5", "shift_jis", "euc-jp", "euc-kr"]);
+function estimateProcessingSeconds(format, sizeBytes, encoding) {
+  if (format === "xls") return 0;
+  let throughput = THROUGHPUT_BYTES_PER_SECOND[format];
+  if (encoding && SLOW_ENCODINGS.has(encoding.toLowerCase())) throughput /= 2;
+  return sizeBytes / throughput;
+}
+function formatDuration(seconds) {
+  if (seconds < 60) {
+    const low2 = Math.max(1, Math.round(seconds * 0.5));
+    const high2 = Math.max(low2 + 1, Math.round(seconds * 1.5));
+    return `roughly ${low2} to ${high2} seconds`;
+  }
+  const low = Math.max(1, Math.round(seconds / 60 * 0.5));
+  const high = Math.max(low + 1, Math.round(seconds / 60 * 1.5));
+  return `roughly ${low} to ${high} minutes`;
+}
+function humanSize(bytes) {
+  if (bytes >= 1024 * 1024 * 1024) return `${(bytes / (1024 * 1024 * 1024)).toFixed(1)} GB`;
+  return `${Math.round(bytes / (1024 * 1024))} MB`;
+}
+function xlsMemoryWarning(fileName, sizeBytes) {
+  return `Warning: ${fileName} is ${humanSize(sizeBytes)} (XLS). The legacy XLS parser loads the entire workbook into memory, roughly 5-10x the file size. Prefer converting to XLSX or splitting the workbook first.`;
+}
+function largeFileTimeWarning(fileName, sizeBytes, format, encoding) {
+  const duration = formatDuration(estimateProcessingSeconds(format, sizeBytes, encoding));
+  return `Warning: ${fileName} is ${humanSize(sizeBytes)}; processing is estimated to take ${duration}. Consider splitting the file if that is too long.`;
+}
+function largeFileMemoryWarning(format) {
+  if (format === "xlsx") {
+    return "Memory risk: XLSX processing keeps the workbook's shared string table in memory, so peak memory can substantially exceed the file size.";
+  }
+  if (format === "json") {
+    return "Memory risk: a JSON root object or a very large record may be materialized in memory during inspection, so peak memory can substantially exceed the file size.";
+  }
+  return null;
+}
+function assessFileSize(fileName, format, sizeBytes, encoding) {
+  const baseAssessment = {
+    size: humanSize(sizeBytes),
+    estimatedDuration: null,
+    warning: null,
+    reason: null,
+    memoryRisk: false,
+    rejected: false
+  };
+  if (format === "xls") {
+    if (sizeBytes > XLS_HARD_LIMIT_BYTES) {
+      return {
+        ...baseAssessment,
+        reason: "XLS files larger than 1 GB are not supported; convert the workbook to XLSX or split it first.",
+        memoryRisk: true,
+        rejected: true
+      };
+    }
+    if (sizeBytes > XLS_SIZE_WARN_BYTES) {
+      return {
+        ...baseAssessment,
+        warning: xlsMemoryWarning(fileName, sizeBytes),
+        memoryRisk: true
+      };
+    }
+    return baseAssessment;
+  }
+  if (sizeBytes > LARGE_FILE_WARN_BYTES) {
+    const memoryWarning = largeFileMemoryWarning(format);
+    const timeWarning = largeFileTimeWarning(fileName, sizeBytes, format, encoding);
+    return {
+      ...baseAssessment,
+      estimatedDuration: formatDuration(estimateProcessingSeconds(format, sizeBytes, encoding)),
+      warning: memoryWarning ? `${timeWarning} ${memoryWarning}` : timeWarning,
+      memoryRisk: Boolean(memoryWarning)
+    };
+  }
+  return baseAssessment;
+}
+
 // src/commands/data-integration/local-data/input.ts
 import { createHash } from "crypto";
 import { createReadStream as createReadStream2, statSync } from "fs";
@@ -486,12 +574,10 @@ function isStrongDateTime(value) {
 }
 
 // src/commands/data-integration/local-data/input.ts
-var MAX_STREAMING_FILE_BYTES = 200 * 1024 * 1024;
-var MAX_XLS_FILE_BYTES = 50 * 1024 * 1024;
 var XLSX = XLSXMod.default ?? XLSXMod;
 var require3 = createRequire2(import.meta.url);
 var ExcelWorksheetReader = require3("exceljs/lib/stream/xlsx/worksheet-reader");
-async function inspectLocalDataInput(filePath) {
+function resolveLocalDataInputMeta(filePath) {
   let format = resolveFormat(filePath);
   let delimiter;
   let encoding;
@@ -512,33 +598,43 @@ async function inspectLocalDataInput(filePath) {
       location: { field: "input-file" }
     });
   }
-  const maxBytes = format === "xls" ? MAX_XLS_FILE_BYTES : MAX_STREAMING_FILE_BYTES;
-  if (sizeBytes > maxBytes) {
-    throw new CliValidationError(
-      `${format.toUpperCase()} input exceeds the supported file size limit.`,
-      {
-        code: "LOCAL_DATA_FILE_TOO_LARGE",
-        hint: `Split the file below ${format === "xls" ? "50" : "200"} MB and retry.`,
-        location: { field: "input-file" }
-      }
-    );
-  }
   if (format !== "xls" && format !== "xlsx" && !encoding) {
     encoding = detectEncoding(filePath);
   }
+  return {
+    filePath,
+    format,
+    sizeBytes,
+    ...delimiter ? { delimiter } : {},
+    ...encoding ? { encoding } : {}
+  };
+}
+async function inspectLocalDataInput(filePath) {
+  const meta = resolveLocalDataInputMeta(filePath);
+  emitSizeWarning(meta.filePath, meta.format, meta.sizeBytes, meta.encoding);
   try {
     return {
-      filePath,
-      format,
-      sizeBytes,
+      ...meta,
       sha256: await sha256File(filePath),
-      dataSets: await discoverDataSets(filePath, format),
-      ...delimiter ? { delimiter } : {},
-      ...encoding ? { encoding } : {}
+      dataSets: await discoverDataSets(filePath, meta.format)
     };
   } catch (error) {
     if (error instanceof CliValidationError) throw error;
-    throw localDataParseError(format);
+    throw localDataParseError(meta.format);
+  }
+}
+function emitSizeWarning(filePath, format, sizeBytes, encoding) {
+  const assessment = assessFileSize(basename(filePath), format, sizeBytes, encoding);
+  if (assessment.rejected) {
+    throw new CliValidationError("XLS input exceeds the supported file size limit.", {
+      code: "LOCAL_DATA_FILE_TOO_LARGE",
+      hint: "XLS files larger than 1 GB are not supported; convert the workbook to XLSX or split it first.",
+      location: { field: "input-file" }
+    });
+  }
+  if (assessment.warning) {
+    process.stderr.write(`${assessment.warning}
+`);
   }
 }
 function selectDataSet(input, requested) {
@@ -1847,6 +1943,30 @@ var dataIntegrationInspect = {
     { name: "headerless", type: "boolean", default: false, desc: "Treat the first row as data and auto-generate col_1..col_N names." }
   ],
   risk: "read",
+  // Fast size/time pre-check (stat + format/encoding sniff only — no sha256, no profile).
+  // Lets agents surface the estimate before committing to a multi-minute full inspection.
+  dryRun: async (ctx) => {
+    const files = ctx.list("input-file").map((filePath) => {
+      const meta = resolveLocalDataInputMeta(filePath);
+      const assessment = assessFileSize(basename3(filePath), meta.format, meta.sizeBytes, meta.encoding);
+      return {
+        file: basename3(filePath),
+        format: meta.format,
+        size_bytes: meta.sizeBytes,
+        size: assessment.size,
+        ...assessment.estimatedDuration ? { estimated_duration: assessment.estimatedDuration } : {},
+        ...assessment.warning ? { warning: assessment.warning } : {},
+        ...assessment.reason ? { reason: assessment.reason } : {},
+        ...assessment.memoryRisk ? { memory_risk: true } : {},
+        ...assessment.rejected ? { rejected: true } : {}
+      };
+    });
+    return {
+      version: "ae-local-data-estimate/v1",
+      files,
+      has_large_file: files.some((file) => Boolean(file.warning) || file.rejected)
+    };
+  },
   execute: async (ctx) => {
     const inputFiles = ctx.list("input-file");
     const headerNames = splitHeaders(ctx.str("headers"));

package/dist/index.js

@@ -3496,7 +3496,7 @@ async function loadCommands() {
   } catch {
   }
   try {
-    const teKb = await import("./te-kb-APXBWBDY.js");
+    const teKb = await import("./te-kb-SQCLHG6X.js");
     commands.push(...teKb.default);
   } catch {
   }
@@ -3526,7 +3526,7 @@ async function loadCommands() {
   } catch {
   }
   try {
-    const dataIntegration = await import("./data-integration-2MYMANJI.js");
+    const dataIntegration = await import("./data-integration-XQYB4X4F.js");
     commands.push(...dataIntegration.default);
   } catch {
   }