From 39115527dc6d2326d6e386c29f70fa6d79ba3d03 Mon Sep 17 00:00:00 2001
From: abhay-codes07 <abhaysingh0293@gmail.com>
Date: Thu, 2 Jul 2026 08:12:51 +0530
Subject: [PATCH] fix(web): stop truncating multi-line plugin transcript
 messages

parseTranscriptMessages builds its regex with the m flag, and the
lookahead that terminates a message body ends with |$. Under m, $
matches at every line end, so the lazy [\s\S]*? body stops at the
first newline: every line of a message after the first was silently
dropped from Codex session and Amp thread documents. Continuation
lines carrying "memory id: ..." were lost too, so extractArtifacts
never surfaced their Memory ID chips, and previews/message counts
were computed from truncated text.

Replace the |$ alternative with (?![\s\S]), a true end-of-input
assertion unaffected by the m flag. The adjacent parseClaudeCodeTurns
regex is fine as-is - it uses $ without the m flag.

Add regression tests through parsePluginDocument covering multi-line
bodies, memory-id artifact extraction from continuation lines,
literal \n normalization, and unchanged single-line behavior. The
multi-line cases fail against the previous regex.

Fixes #1184
---
 apps/web/lib/plugin-document.test.ts | 86 ++++++++++++++++++++++++++++
 apps/web/lib/plugin-document.ts      |  5 +-
 2 files changed, 90 insertions(+), 1 deletion(-)
 create mode 100644 apps/web/lib/plugin-document.test.ts

diff --git a/apps/web/lib/plugin-document.test.ts b/apps/web/lib/plugin-document.test.ts
new file mode 100644
index 000000000..79dc630d2
--- /dev/null
+++ b/apps/web/lib/plugin-document.test.ts
@@ -0,0 +1,86 @@
+import { describe, expect, it } from "bun:test"
+import { parsePluginDocument } from "./plugin-document"
+
+type PluginDocumentInput = Parameters<typeof parsePluginDocument>[0]
+
+function makeCodexSessionDocument(content: string): PluginDocumentInput {
+	return {
+		id: "doc_1",
+		title: "Codex session",
+		content,
+		metadata: { sm_source: "codex" },
+		memoryEntries: [],
+	} as unknown as PluginDocumentInput
+}
+
+describe("parsePluginDocument — session transcripts", () => {
+	it("keeps multi-line message bodies intact", () => {
+		const parsed = parsePluginDocument(
+			makeCodexSessionDocument(
+				[
+					"[Session abc-123]",
+					"1. [user] Hello there",
+					"Here is more context on line two",
+					"2. [assistant] Sure!",
+					"Second line of the reply",
+				].join("\n"),
+			),
+		)
+
+		expect(parsed).not.toBeNull()
+		expect(parsed?.kind).toBe("codex-session")
+		expect(parsed?.messages).toHaveLength(2)
+		expect(parsed?.messages[0]?.text).toBe(
+			"Hello there\nHere is more context on line two",
+		)
+		expect(parsed?.messages[1]?.text).toBe("Sure!\nSecond line of the reply")
+	})
+
+	it("surfaces memory id artifacts from continuation lines", () => {
+		const parsed = parsePluginDocument(
+			makeCodexSessionDocument(
+				[
+					"[Session abc-123]",
+					"1. [user] Remember my editor is Neovim",
+					"memory id: mem_456",
+					"2. [assistant] Saved it.",
+				].join("\n"),
+			),
+		)
+
+		expect(parsed?.messages[0]?.text).toBe("Remember my editor is Neovim")
+		expect(parsed?.artifacts).toContainEqual({
+			label: "Memory ID",
+			value: "mem_456",
+		})
+	})
+
+	it("normalizes literal \\n escapes before splitting messages", () => {
+		const parsed = parsePluginDocument(
+			makeCodexSessionDocument(
+				"[Session abc-123]\\n1. [user] First line\\nSecond line\\n2. [assistant] Reply",
+			),
+		)
+
+		expect(parsed?.messages).toHaveLength(2)
+		expect(parsed?.messages[0]?.text).toBe("First line\nSecond line")
+		expect(parsed?.messages[1]?.text).toBe("Reply")
+	})
+
+	it("parses single-line messages as before", () => {
+		const parsed = parsePluginDocument(
+			makeCodexSessionDocument(
+				["[Session abc-123]", "1. [user] Hi", "2. [assistant] Hello!"].join(
+					"\n",
+				),
+			),
+		)
+
+		expect(parsed?.messages).toHaveLength(2)
+		expect(parsed?.messages[0]?.text).toBe("Hi")
+		expect(parsed?.messages[1]?.text).toBe("Hello!")
+		expect(parsed?.summary).toBe(
+			"1 user message and 1 assistant message captured from Codex.",
+		)
+	})
+})
diff --git a/apps/web/lib/plugin-document.ts b/apps/web/lib/plugin-document.ts
index 3a13ec245..1606311f5 100644
--- a/apps/web/lib/plugin-document.ts
+++ b/apps/web/lib/plugin-document.ts
@@ -227,8 +227,11 @@ function parseTranscriptMessages(content: string): {
 } {
 	const messages: PluginDocumentMessage[] = []
 	const artifacts: PluginArtifact[] = []
+	// The lookahead must end the last message at the true end of input:
+	// with the m flag, a bare $ matches every line end and would cut each
+	// message body off at its first newline.
 	const regex = new RegExp(
-		`^\\s*(\\d+)\\.\\s+\\[(${TRANSCRIPT_ROLE_PATTERN})\\]\\s*([\\s\\S]*?)(?=^\\s*\\d+\\.\\s+\\[(?:${TRANSCRIPT_ROLE_PATTERN})\\]\\s*|$)`,
+		`^\\s*(\\d+)\\.\\s+\\[(${TRANSCRIPT_ROLE_PATTERN})\\]\\s*([\\s\\S]*?)(?=^\\s*\\d+\\.\\s+\\[(?:${TRANSCRIPT_ROLE_PATTERN})\\]\\s*|(?![\\s\\S]))`,
 		"gm",
 	)