test: patch seekSequence matching and Command.hints placeholder extraction

Cover two previously untested code paths: (1) Patch.deriveNewContentsFromChunks
with seekSequence's 4-pass matching (exact, rstrip, trim, Unicode-normalized)
and EOF anchoring/change_context seeking; (2) Command.hints() argument placeholder
extraction that drives TUI argument prompts for slash commands.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

https://claude.ai/code/session_01Nvdry9ChLovXGjHNS8kR7R

Author: claude

packages/opencode/test/command/hints.test.ts

@@ -0,0 +1,63 @@
+import { describe, test, expect } from "bun:test"
+import { Command } from "../../src/command"
+
+/**
+ * Tests for Command.hints() — the pure function that extracts argument
+ * placeholders ($1, $2, ..., $ARGUMENTS) from command templates.
+ *
+ * These hints drive the TUI's argument prompt display. If hints are wrong,
+ * users see incorrect or missing argument suggestions when invoking commands
+ * like /review, /init, or custom commands.
+ */
+
+describe("Command.hints", () => {
+  test("returns empty array for template with no placeholders", () => {
+    expect(Command.hints("Run all tests")).toEqual([])
+  })
+
+  test("extracts single numbered placeholder", () => {
+    expect(Command.hints("Review commit $1")).toEqual(["$1"])
+  })
+
+  test("extracts multiple numbered placeholders in sorted order", () => {
+    expect(Command.hints("Compare $2 against $1")).toEqual(["$1", "$2"])
+  })
+
+  test("deduplicates repeated placeholders", () => {
+    expect(Command.hints("Use $1 then reuse $1 again")).toEqual(["$1"])
+  })
+
+  test("extracts $ARGUMENTS placeholder", () => {
+    expect(Command.hints("Execute with $ARGUMENTS")).toEqual(["$ARGUMENTS"])
+  })
+
+  test("extracts both numbered and $ARGUMENTS, numbered first", () => {
+    expect(Command.hints("Run $1 with $ARGUMENTS")).toEqual(["$1", "$ARGUMENTS"])
+  })
+
+  test("handles multi-digit placeholders like $10 (lexicographic sort)", () => {
+    const result = Command.hints("Lots of args: $1 $2 $10")
+    // JavaScript's default sort is lexicographic: "$10" < "$2" because "1" < "2"
+    expect(result).toEqual(["$1", "$10", "$2"])
+  })
+
+  test("returns empty for empty template string", () => {
+    expect(Command.hints("")).toEqual([])
+  })
+
+  test("does not match $ followed by letters (not ARGUMENTS)", () => {
+    expect(Command.hints("Use $FOO and $BAR")).toEqual([])
+  })
+
+  test("$ARGUMENTS is case-sensitive", () => {
+    expect(Command.hints("Use $arguments")).toEqual([])
+  })
+
+  test("handles template with only whitespace", () => {
+    expect(Command.hints("   \n\t  ")).toEqual([])
+  })
+
+  test("handles $0 as a valid numbered placeholder", () => {
+    expect(Command.hints("$0 is first")).toEqual(["$0"])
+  })
+})

packages/opencode/test/patch/seek-sequence.test.ts

@@ -0,0 +1,250 @@
+import { describe, test, expect, beforeEach, afterEach } from "bun:test"
+import { Patch } from "../../src/patch"
+import * as fs from "fs/promises"
+import * as path from "path"
+import { tmpdir } from "os"
+
+/**
+ * Tests for Patch.deriveNewContentsFromChunks — the core function that applies
+ * update chunks to file content using seekSequence's multi-pass matching.
+ *
+ * seekSequence tries 4 comparison strategies in order:
+ *   1. Exact match
+ *   2. Trailing whitespace trimmed (rstrip)
+ *   3. Both-end whitespace trimmed (trim)
+ *   4. Unicode-normalized + trimmed
+ *
+ * These tests verify that real-world patch application succeeds even when the
+ * LLM-generated patch text has minor whitespace or Unicode differences from
+ * the actual file content — a common source of "Failed to find expected lines"
+ * errors for users.
+ */
+
+describe("Patch.deriveNewContentsFromChunks — seekSequence matching", () => {
+  let tempDir: string
+
+  beforeEach(async () => {
+    tempDir = await fs.mkdtemp(path.join(tmpdir(), "seek-test-"))
+  })
+
+  afterEach(async () => {
+    await fs.rm(tempDir, { recursive: true, force: true })
+  })
+
+  test("exact match: replaces old_lines with new_lines", () => {
+    const filePath = path.join(tempDir, "exact.txt")
+    const content = "line1\nline2\nline3\n"
+    require("fs").writeFileSync(filePath, content)
+
+    const result = Patch.deriveNewContentsFromChunks(filePath, [
+      {
+        old_lines: ["line2"],
+        new_lines: ["REPLACED"],
+      },
+    ])
+
+    expect(result.content).toBe("line1\nREPLACED\nline3\n")
+    expect(result.unified_diff).toContain("-line2")
+    expect(result.unified_diff).toContain("+REPLACED")
+  })
+
+  test("rstrip pass: matches despite trailing whitespace differences", () => {
+    const filePath = path.join(tempDir, "rstrip.txt")
+    // File has trailing spaces on line2
+    const content = "line1\nline2   \nline3\n"
+    require("fs").writeFileSync(filePath, content)
+
+    // Patch references line2 without trailing spaces
+    const result = Patch.deriveNewContentsFromChunks(filePath, [
+      {
+        old_lines: ["line2"],
+        new_lines: ["REPLACED"],
+      },
+    ])
+
+    expect(result.content).toContain("REPLACED")
+  })
+
+  test("trim pass: matches despite leading whitespace differences", () => {
+    const filePath = path.join(tempDir, "trim.txt")
+    // File has extra leading spaces
+    const content = "  function foo() {\n    return 1\n  }\n"
+    require("fs").writeFileSync(filePath, content)
+
+    // Patch references without leading spaces
+    const result = Patch.deriveNewContentsFromChunks(filePath, [
+      {
+        old_lines: ["return 1"],
+        new_lines: ["return 42"],
+      },
+    ])
+
+    expect(result.content).toContain("return 42")
+  })
+
+  test("unicode pass: matches smart quotes to ASCII quotes", () => {
+    const filePath = path.join(tempDir, "unicode.txt")
+    // File uses Unicode smart quotes (common in copy-pasted code)
+    const content = 'const msg = \u201CHello World\u201D\n'
+    require("fs").writeFileSync(filePath, content)
+
+    // Patch uses ASCII double quotes
+    const result = Patch.deriveNewContentsFromChunks(filePath, [
+      {
+        old_lines: ['const msg = "Hello World"'],
+        new_lines: ['const msg = "Goodbye World"'],
+      },
+    ])
+
+    expect(result.content).toContain("Goodbye World")
+  })
+
+  test("unicode pass: matches em-dash to hyphen", () => {
+    const filePath = path.join(tempDir, "emdash.txt")
+    // File uses Unicode em-dash
+    const content = "value \u2014 description\n"
+    require("fs").writeFileSync(filePath, content)
+
+    // Patch uses ASCII hyphen
+    const result = Patch.deriveNewContentsFromChunks(filePath, [
+      {
+        old_lines: ["value - description"],
+        new_lines: ["value - updated"],
+      },
+    ])
+
+    expect(result.content).toContain("updated")
+  })
+
+  test("is_end_of_file: anchors match to end of file", () => {
+    const filePath = path.join(tempDir, "eof.txt")
+    const content = "line1\nline2\nline3\nline2\n"
+    require("fs").writeFileSync(filePath, content)
+
+    // Should match the LAST occurrence of "line2" because is_end_of_file is true
+    const result = Patch.deriveNewContentsFromChunks(filePath, [
+      {
+        old_lines: ["line2"],
+        new_lines: ["LAST"],
+        is_end_of_file: true,
+      },
+    ])
+
+    // Exact expected output: first "line2" untouched, last "line2" replaced
+    expect(result.content).toBe("line1\nline2\nline3\nLAST\n")
+  })
+
+  test("change_context: seeks to context line before matching old_lines", () => {
+    const filePath = path.join(tempDir, "context.txt")
+    const content = "function foo() {\n  return 1\n}\nfunction bar() {\n  return 1\n}\n"
+    require("fs").writeFileSync(filePath, content)
+
+    // Use change_context to target the return inside bar(), not foo()
+    const result = Patch.deriveNewContentsFromChunks(filePath, [
+      {
+        old_lines: ["  return 1"],
+        new_lines: ["  return 99"],
+        change_context: "function bar() {",
+      },
+    ])
+
+    expect(result.content).toContain("function foo() {\n  return 1")
+    expect(result.content).toContain("function bar() {\n  return 99")
+  })
+
+  test("throws when old_lines cannot be found", () => {
+    const filePath = path.join(tempDir, "missing.txt")
+    require("fs").writeFileSync(filePath, "hello\nworld\n")
+
+    expect(() =>
+      Patch.deriveNewContentsFromChunks(filePath, [
+        {
+          old_lines: ["nonexistent line"],
+          new_lines: ["replacement"],
+        },
+      ]),
+    ).toThrow("Failed to find expected lines")
+  })
+
+  test("throws when file does not exist", () => {
+    expect(() =>
+      Patch.deriveNewContentsFromChunks("/tmp/nonexistent-file-12345.txt", [
+        {
+          old_lines: ["x"],
+          new_lines: ["y"],
+        },
+      ]),
+    ).toThrow("Failed to read file")
+  })
+
+  test("multiple chunks applied in sequence", () => {
+    const filePath = path.join(tempDir, "multi.txt")
+    const content = "alpha\nbeta\ngamma\ndelta\n"
+    require("fs").writeFileSync(filePath, content)
+
+    const result = Patch.deriveNewContentsFromChunks(filePath, [
+      {
+        old_lines: ["beta"],
+        new_lines: ["BETA"],
+      },
+      {
+        old_lines: ["delta"],
+        new_lines: ["DELTA"],
+      },
+    ])
+
+    expect(result.content).toBe("alpha\nBETA\ngamma\nDELTA\n")
+  })
+
+  test("pure addition chunk (empty old_lines) appends content", () => {
+    const filePath = path.join(tempDir, "append.txt")
+    require("fs").writeFileSync(filePath, "existing\n")
+
+    const result = Patch.deriveNewContentsFromChunks(filePath, [
+      {
+        old_lines: [],
+        new_lines: ["new_line"],
+      },
+    ])
+
+    expect(result.content).toContain("existing")
+    expect(result.content).toContain("new_line")
+  })
+})
+
+describe("Patch.parsePatch — stripHeredoc handling", () => {
+  test("parses patch wrapped in heredoc with cat <<'EOF'", () => {
+    const input = `cat <<'EOF'
+*** Begin Patch
+*** Add File: hello.txt
++Hello
+*** End Patch
+EOF`
+
+    const result = Patch.parsePatch(input)
+    expect(result.hunks).toHaveLength(1)
+    expect(result.hunks[0].type).toBe("add")
+  })
+
+  test("parses patch wrapped in heredoc with <<HEREDOC", () => {
+    const input = `<<HEREDOC
+*** Begin Patch
+*** Delete File: old.txt
+*** End Patch
+HEREDOC`
+
+    const result = Patch.parsePatch(input)
+    expect(result.hunks).toHaveLength(1)
+    expect(result.hunks[0].type).toBe("delete")
+  })
+
+  test("parses unwrapped patch text unchanged", () => {
+    const input = `*** Begin Patch
+*** Add File: test.txt
++content
+*** End Patch`
+
+    const result = Patch.parsePatch(input)
+    expect(result.hunks).toHaveLength(1)
+  })
+})