@ashx-j/lunr 0.2.27 → 0.2.28

package/dist/modes/interactive/interactive-mode.js

@@ -1,3 +1,4 @@
+import { collectUsageRequests, usageTokens } from "../../core/usage-accounting.js";
 /**
  * Interactive mode for the coding agent.
  * Handles TUI rendering and user interaction, delegating business logic to AgentSession.
@@ -3385,7 +3386,7 @@ export class InteractiveMode {
     }
     /**
      * Show a transcript notice when a completed assistant message paid for a
-     * significant cache miss. Only states observable facts: the miss itself,
+     * significant cache miss. The overlap estimate cannot establish identical prompt content. It also shows
      * a model switch, or an idle gap past the cache TTL.
      */
     maybeShowCacheMissNotice(message) {
@@ -3400,10 +3401,10 @@ export class InteractiveMode {
         if (miss.missedTokens < 20_000 && miss.missedCost < 0.1)
             return;
         const cost = miss.missedCost >= 0.01 ? ` (~$${miss.missedCost.toFixed(2)})` : "";
-        const reBilled = `${formatTokens(miss.missedTokens)} tokens re-billed${cost}`;
-        let label = "Cache miss";
+        const reBilled = `~${formatTokens(miss.missedTokens)} tokens may have missed cache${cost}`;
+        let label = "Estimated cache miss";
         if (miss.modelChanged) {
-            label = "Cache miss after model switch";
+            label = "Estimated cache miss after model switch";
         }
         else if (miss.idleMs >= CACHE_TTL_MS) {
             label = `Cache miss after ${Math.round(miss.idleMs / 60_000)}m idle`;
@@ -6279,9 +6280,24 @@ export class InteractiveMode {
             const prompt = "Generate a title of at most 6 words summarizing this request. " +
                 "Reply with the title only — no quotes, no trailing punctuation, no prefix.\n\n" +
                 `Request:\n${firstUserText.slice(0, 2000)}`;
+            const requestSessionId = this.sessionManager.getSessionId();
+            const requestSessionFile = this.sessionManager.getSessionFile();
             const response = await this.session.modelRuntime.complete(model, {
                 messages: [{ role: "user", content: [{ type: "text", text: prompt }], timestamp: Date.now() }],
             });
+            if (this.sessionManager.getSessionId() !== requestSessionId) {
+                if (requestSessionFile) {
+                    const target = SessionManager.open(requestSessionFile);
+                    try {
+                        target.appendRequestUsage("title", response);
+                    }
+                    finally {
+                        target.dispose();
+                    }
+                }
+                return;
+            }
+            this.sessionManager.appendRequestUsage("title", response);
             let title = response.content
                 .filter((c) => c.type === "text")
                 .map((c) => c.text)
@@ -6307,19 +6323,12 @@ export class InteractiveMode {
         // Cost/token totals per provider/model actually used (e.g. OpenRouter `auto`
         // resolves to a concrete responseModel), sorted by cost descending.
         const perModelMap = new Map();
-        for (const entry of entries) {
-            if (entry.type !== "message" || entry.message.role !== "assistant")
-                continue;
-            const message = entry.message;
-            const usage = message.usage;
-            const key = `${message.provider}/${message.responseModel ?? message.model}`;
-            let bucket = perModelMap.get(key);
-            if (!bucket) {
-                bucket = { key, cost: 0, tokens: 0 };
-                perModelMap.set(key, bucket);
-            }
-            bucket.cost += usage.cost.total;
-            bucket.tokens += usage.input + usage.output + usage.cacheRead + usage.cacheWrite;
+        for (const request of collectUsageRequests(entries)) {
+            const key = `${request.provider}/${request.model}`;
+            const bucket = perModelMap.get(key) ?? { key, cost: 0, tokens: 0 };
+            bucket.cost += request.usage.cost.total;
+            bucket.tokens += usageTokens(request.usage);
+            perModelMap.set(key, bucket);
         }
         const perModel = Array.from(perModelMap.values()).sort((a, b) => b.cost - a.cost);
         let info = `${theme.bold("Session Info")}\n\n`;
@@ -6333,7 +6342,7 @@ export class InteractiveMode {
         info += `${theme.fg("dim", "User:")} ${stats.userMessages}\n`;
         info += `${theme.fg("dim", "Assistant:")} ${stats.assistantMessages}\n`;
         info += `${theme.fg("dim", "Tools:")} ${stats.toolCalls} calls, ${stats.toolResults} results\n\n`;
-        info += `${theme.bold("Tokens")}\n`;
+        info += `${theme.bold("Parent tokens including auxiliary requests")}\n`;
         // "Input" is the full prompt volume. With cache activity, split it into
         // cached (served from cache) vs uncached (everything else) - the only
         // provider-independent split. Cache writes, where reported, are a detail
@@ -6349,6 +6358,17 @@ export class InteractiveMode {
         }
         info += `${theme.fg("dim", "Output:")} ${stats.tokens.output.toLocaleString()}\n`;
         info += `${theme.fg("dim", "Total:")} ${stats.tokens.total.toLocaleString()}\n`;
+        if (stats.childUsage?.total) {
+            info += `Children: ${stats.childUsage.total.toLocaleString()} tokens, $${stats.childUsage.cost.toFixed(3)}\n`;
+            info += `Combined: ${stats.combinedUsage?.total.toLocaleString()} tokens, $${stats.combinedUsage?.cost.toFixed(3)}\n`;
+        }
+        if (stats.parentUsage?.unknownRequests)
+            info += `Usage unavailable for ${stats.parentUsage.unknownRequests} request(s).\n`;
+        if (stats.parentUsage?.partialRequests)
+            info += `Usage incomplete for ${stats.parentUsage.partialRequests} request(s).\n`;
+        if (stats.parentUsage?.unknownCosts)
+            info += `Pricing unavailable for ${stats.parentUsage.unknownCosts} request(s).\n`;
+        info += "Costs use catalog estimates unless the provider reports its account charge.\n";
         if (stats.cost > 0 || cacheWaste.missedTokens > 0) {
             info += `\n${theme.bold("Cost")}\n`;
             info += `${theme.fg("dim", "Total:")} $${stats.cost.toFixed(3)}`;
@@ -6362,8 +6382,8 @@ export class InteractiveMode {
                 const detail = `${cacheWaste.missedTokens.toLocaleString()} tokens, ${missLabel}`;
                 info +=
                     cacheWaste.missedCost >= 0.0001
-                        ? `\n${theme.fg("dim", "Cache Re-billed:")} $${cacheWaste.missedCost.toFixed(3)} ${theme.fg("dim", `(${detail})`)}`
-                        : `\n${theme.fg("dim", "Cache Re-billed:")} ${detail}`;
+                        ? `\n${theme.fg("dim", "Estimated cache misses:")} $${cacheWaste.missedCost.toFixed(3)} ${theme.fg("dim", `(${detail})`)}`
+                        : `\n${theme.fg("dim", "Estimated cache misses:")} ${detail}`;
             }
         }
         this.chatContainer.addChild(new Spacer(1));
@@ -6411,20 +6431,33 @@ export class InteractiveMode {
     // context-window bar, and subscription plan usage.
     async handleUsageCommand() {
         const stats = this.session.getSessionStats();
-        const sessionTotals = stats.tokens.total > 0
+        const sessionTotals = stats.tokens.total > 0 ||
+            stats.parentUsage?.requests ||
+            stats.parentUsage?.unknownRequests ||
+            stats.parentUsage?.partialRequests
             ? {
                 input: stats.tokens.input,
                 output: stats.tokens.output,
                 cacheRead: stats.tokens.cacheRead,
                 cacheWrite: stats.tokens.cacheWrite,
                 total: stats.tokens.total,
+                requests: stats.parentUsage?.requests,
+                unknownRequests: stats.parentUsage?.unknownRequests,
+                partialRequests: stats.parentUsage?.partialRequests,
             }
             : undefined;
         const context = this.session.getContextUsage();
         const provider = this.session.model?.provider;
         const planResult = await getAllPlanUsageResults(provider, this.session.modelRuntime);
         const breakdown = this.collectUsageBreakdown();
-        const lines = renderUsageBox({ sessionTotals, context, plan: planResult.usages, breakdown }, this.ui.terminal.columns);
+        const lines = renderUsageBox({
+            sessionTotals,
+            childTotals: stats.childUsage,
+            combinedTotals: stats.combinedUsage,
+            context,
+            plan: planResult.usages,
+            breakdown,
+        }, this.ui.terminal.columns);
         this.chatContainer.addChild(new Spacer(1));
         this.chatContainer.addChild(new Text(lines.join("\n"), 1, 0));
         if (planResult.errors.length > 0)