bonecode 1.0.0 → 1.1.0

package/src/engine/patch/index.ts

@@ -1,689 +1,689 @@
-import { Effect, Schema } from "effect"
-import * as path from "path"
-import { AppFileSystem } from "@opencode-ai/core/filesystem"
-import * as Log from "@opencode-ai/core/util/log"
-import * as Bom from "../util/bom"
-
-const log = Log.create({ service: "patch" })
-
-export const PatchSchema = Schema.Struct({
-  patchText: Schema.String.annotate({ description: "The full patch text that describes all changes to be made" }),
-})
-
-export type PatchParams = Schema.Schema.Type<typeof PatchSchema>
-
-// Core types matching the Rust implementation
-export interface ApplyPatchArgs {
-  patch: string
-  hunks: Hunk[]
-  workdir?: string
-}
-
-export type Hunk =
-  | { type: "add"; path: string; contents: string }
-  | { type: "delete"; path: string }
-  | { type: "update"; path: string; move_path?: string; chunks: UpdateFileChunk[] }
-
-export interface UpdateFileChunk {
-  old_lines: string[]
-  new_lines: string[]
-  change_context?: string
-  is_end_of_file?: boolean
-}
-
-export interface ApplyPatchAction {
-  changes: Map<string, ApplyPatchFileChange>
-  patch: string
-  cwd: string
-}
-
-export type ApplyPatchFileChange =
-  | { type: "add"; content: string }
-  | { type: "delete"; content: string }
-  | { type: "update"; unified_diff: string; move_path?: string; new_content: string }
-
-export interface AffectedPaths {
-  added: string[]
-  modified: string[]
-  deleted: string[]
-}
-
-export enum ApplyPatchError {
-  ParseError = "ParseError",
-  IoError = "IoError",
-  ComputeReplacements = "ComputeReplacements",
-  ImplicitInvocation = "ImplicitInvocation",
-}
-
-export enum MaybeApplyPatch {
-  Body = "Body",
-  ShellParseError = "ShellParseError",
-  PatchParseError = "PatchParseError",
-  NotApplyPatch = "NotApplyPatch",
-}
-
-export enum MaybeApplyPatchVerified {
-  Body = "Body",
-  ShellParseError = "ShellParseError",
-  CorrectnessError = "CorrectnessError",
-  NotApplyPatch = "NotApplyPatch",
-}
-
-// Parser implementation
-function parsePatchHeader(
-  lines: string[],
-  startIdx: number,
-): { filePath: string; movePath?: string; nextIdx: number } | null {
-  const line = lines[startIdx]
-
-  if (line.startsWith("*** Add File:")) {
-    const filePath = line.slice("*** Add File:".length).trim()
-    return filePath ? { filePath, nextIdx: startIdx + 1 } : null
-  }
-
-  if (line.startsWith("*** Delete File:")) {
-    const filePath = line.slice("*** Delete File:".length).trim()
-    return filePath ? { filePath, nextIdx: startIdx + 1 } : null
-  }
-
-  if (line.startsWith("*** Update File:")) {
-    const filePath = line.slice("*** Update File:".length).trim()
-    let movePath: string | undefined
-    let nextIdx = startIdx + 1
-
-    // Check for move directive
-    if (nextIdx < lines.length && lines[nextIdx].startsWith("*** Move to:")) {
-      movePath = lines[nextIdx].slice("*** Move to:".length).trim()
-      nextIdx++
-    }
-
-    return filePath ? { filePath, movePath, nextIdx } : null
-  }
-
-  return null
-}
-
-function parseUpdateFileChunks(lines: string[], startIdx: number): { chunks: UpdateFileChunk[]; nextIdx: number } {
-  const chunks: UpdateFileChunk[] = []
-  let i = startIdx
-
-  while (i < lines.length && !lines[i].startsWith("***")) {
-    if (lines[i].startsWith("@@")) {
-      // Parse context line
-      const contextLine = lines[i].substring(2).trim()
-      i++
-
-      const oldLines: string[] = []
-      const newLines: string[] = []
-      let isEndOfFile = false
-
-      // Parse change lines
-      while (i < lines.length && !lines[i].startsWith("@@") && !lines[i].startsWith("***")) {
-        const changeLine = lines[i]
-
-        if (changeLine === "*** End of File") {
-          isEndOfFile = true
-          i++
-          break
-        }
-
-        if (changeLine.startsWith(" ")) {
-          // Keep line - appears in both old and new
-          const content = changeLine.substring(1)
-          oldLines.push(content)
-          newLines.push(content)
-        } else if (changeLine.startsWith("-")) {
-          // Remove line - only in old
-          oldLines.push(changeLine.substring(1))
-        } else if (changeLine.startsWith("+")) {
-          // Add line - only in new
-          newLines.push(changeLine.substring(1))
-        }
-
-        i++
-      }
-
-      chunks.push({
-        old_lines: oldLines,
-        new_lines: newLines,
-        change_context: contextLine || undefined,
-        is_end_of_file: isEndOfFile || undefined,
-      })
-    } else {
-      i++
-    }
-  }
-
-  return { chunks, nextIdx: i }
-}
-
-function parseAddFileContent(lines: string[], startIdx: number): { content: string; nextIdx: number } {
-  let content = ""
-  let i = startIdx
-
-  while (i < lines.length && !lines[i].startsWith("***")) {
-    if (lines[i].startsWith("+")) {
-      content += lines[i].substring(1) + "\n"
-    }
-    i++
-  }
-
-  // Remove trailing newline
-  if (content.endsWith("\n")) {
-    content = content.slice(0, -1)
-  }
-
-  return { content, nextIdx: i }
-}
-
-function stripHeredoc(input: string): string {
-  // Match heredoc patterns like: cat <<'EOF'\n...\nEOF or <<EOF\n...\nEOF
-  const heredocMatch = input.match(/^(?:cat\s+)?<<['"]?(\w+)['"]?\s*\n([\s\S]*?)\n\1\s*$/)
-  if (heredocMatch) {
-    return heredocMatch[2]
-  }
-  return input
-}
-
-export function parsePatch(patchText: string): { hunks: Hunk[] } {
-  const cleaned = stripHeredoc(patchText.trim())
-  const lines = cleaned.split("\n")
-  const hunks: Hunk[] = []
-  let i = 0
-
-  // Look for Begin/End patch markers
-  const beginMarker = "*** Begin Patch"
-  const endMarker = "*** End Patch"
-
-  const beginIdx = lines.findIndex((line) => line.trim() === beginMarker)
-  const endIdx = lines.findIndex((line) => line.trim() === endMarker)
-
-  if (beginIdx === -1 || endIdx === -1 || beginIdx >= endIdx) {
-    throw new Error("Invalid patch format: missing Begin/End markers")
-  }
-
-  // Parse content between markers
-  i = beginIdx + 1
-
-  while (i < endIdx) {
-    const header = parsePatchHeader(lines, i)
-    if (!header) {
-      i++
-      continue
-    }
-
-    if (lines[i].startsWith("*** Add File:")) {
-      const { content, nextIdx } = parseAddFileContent(lines, header.nextIdx)
-      hunks.push({
-        type: "add",
-        path: header.filePath,
-        contents: content,
-      })
-      i = nextIdx
-    } else if (lines[i].startsWith("*** Delete File:")) {
-      hunks.push({
-        type: "delete",
-        path: header.filePath,
-      })
-      i = header.nextIdx
-    } else if (lines[i].startsWith("*** Update File:")) {
-      const { chunks, nextIdx } = parseUpdateFileChunks(lines, header.nextIdx)
-      hunks.push({
-        type: "update",
-        path: header.filePath,
-        move_path: header.movePath,
-        chunks,
-      })
-      i = nextIdx
-    } else {
-      i++
-    }
-  }
-
-  return { hunks }
-}
-
-// Apply patch functionality
-export function maybeParseApplyPatch(
-  argv: string[],
-):
-  | { type: MaybeApplyPatch.Body; args: ApplyPatchArgs }
-  | { type: MaybeApplyPatch.PatchParseError; error: Error }
-  | { type: MaybeApplyPatch.NotApplyPatch } {
-  const APPLY_PATCH_COMMANDS = ["apply_patch", "applypatch"]
-
-  // Direct invocation: apply_patch <patch>
-  if (argv.length === 2 && APPLY_PATCH_COMMANDS.includes(argv[0])) {
-    try {
-      const { hunks } = parsePatch(argv[1])
-      return {
-        type: MaybeApplyPatch.Body,
-        args: {
-          patch: argv[1],
-          hunks,
-        },
-      }
-    } catch (error) {
-      return {
-        type: MaybeApplyPatch.PatchParseError,
-        error: error as Error,
-      }
-    }
-  }
-
-  // Bash heredoc form: bash -lc 'apply_patch <<"EOF" ...'
-  if (argv.length === 3 && argv[0] === "bash" && argv[1] === "-lc") {
-    // Simple extraction - in real implementation would need proper bash parsing
-    const script = argv[2]
-    const heredocMatch = script.match(/apply_patch\s*<<['"](\w+)['"]\s*\n([\s\S]*?)\n\1/)
-
-    if (heredocMatch) {
-      const patchContent = heredocMatch[2]
-      try {
-        const { hunks } = parsePatch(patchContent)
-        return {
-          type: MaybeApplyPatch.Body,
-          args: {
-            patch: patchContent,
-            hunks,
-          },
-        }
-      } catch (error) {
-        return {
-          type: MaybeApplyPatch.PatchParseError,
-          error: error as Error,
-        }
-      }
-    }
-  }
-
-  return { type: MaybeApplyPatch.NotApplyPatch }
-}
-
-// File content manipulation
-interface ApplyPatchFileUpdate {
-  unified_diff: string
-  content: string
-  bom: boolean
-}
-
-export function deriveNewContentsFromChunks(
-  filePath: string,
-  chunks: UpdateFileChunk[],
-  originalText: string,
-): ApplyPatchFileUpdate {
-  const originalContent = Bom.split(originalText)
-
-  let originalLines = originalContent.text.split("\n")
-
-  // Drop trailing empty element for consistent line counting
-  if (originalLines.length > 0 && originalLines[originalLines.length - 1] === "") {
-    originalLines.pop()
-  }
-
-  const replacements = computeReplacements(originalLines, filePath, chunks)
-  let newLines = applyReplacements(originalLines, replacements)
-
-  // Ensure trailing newline
-  if (newLines.length === 0 || newLines[newLines.length - 1] !== "") {
-    newLines.push("")
-  }
-
-  const next = Bom.split(newLines.join("\n"))
-  const newContent = next.text
-
-  // Generate unified diff
-  const unifiedDiff = generateUnifiedDiff(originalContent.text, newContent)
-
-  return {
-    unified_diff: unifiedDiff,
-    content: newContent,
-    bom: originalContent.bom || next.bom,
-  }
-}
-
-function computeReplacements(
-  originalLines: string[],
-  filePath: string,
-  chunks: UpdateFileChunk[],
-): Array<[number, number, string[]]> {
-  const replacements: Array<[number, number, string[]]> = []
-  let lineIndex = 0
-
-  for (const chunk of chunks) {
-    // Handle context-based seeking
-    if (chunk.change_context) {
-      const contextIdx = seekSequence(originalLines, [chunk.change_context], lineIndex)
-      if (contextIdx === -1) {
-        throw new Error(`Failed to find context '${chunk.change_context}' in ${filePath}`)
-      }
-      lineIndex = contextIdx + 1
-    }
-
-    // Handle pure addition (no old lines)
-    if (chunk.old_lines.length === 0) {
-      const insertionIdx =
-        originalLines.length > 0 && originalLines[originalLines.length - 1] === ""
-          ? originalLines.length - 1
-          : originalLines.length
-      replacements.push([insertionIdx, 0, chunk.new_lines])
-      continue
-    }
-
-    // Try to match old lines in the file
-    let pattern = chunk.old_lines
-    let newSlice = chunk.new_lines
-    let found = seekSequence(originalLines, pattern, lineIndex, chunk.is_end_of_file)
-
-    // Retry without trailing empty line if not found
-    if (found === -1 && pattern.length > 0 && pattern[pattern.length - 1] === "") {
-      pattern = pattern.slice(0, -1)
-      if (newSlice.length > 0 && newSlice[newSlice.length - 1] === "") {
-        newSlice = newSlice.slice(0, -1)
-      }
-      found = seekSequence(originalLines, pattern, lineIndex, chunk.is_end_of_file)
-    }
-
-    if (found !== -1) {
-      replacements.push([found, pattern.length, newSlice])
-      lineIndex = found + pattern.length
-    } else {
-      throw new Error(`Failed to find expected lines in ${filePath}:\n${chunk.old_lines.join("\n")}`)
-    }
-  }
-
-  // Sort replacements by index to apply in order
-  replacements.sort((a, b) => a[0] - b[0])
-
-  return replacements
-}
-
-function applyReplacements(lines: string[], replacements: Array<[number, number, string[]]>): string[] {
-  // Apply replacements in reverse order to avoid index shifting
-  const result = [...lines]
-
-  for (let i = replacements.length - 1; i >= 0; i--) {
-    const [startIdx, oldLen, newSegment] = replacements[i]
-
-    // Remove old lines
-    result.splice(startIdx, oldLen)
-
-    // Insert new lines
-    for (let j = 0; j < newSegment.length; j++) {
-      result.splice(startIdx + j, 0, newSegment[j])
-    }
-  }
-
-  return result
-}
-
-// Normalize Unicode punctuation to ASCII equivalents (like Rust's normalize_unicode)
-function normalizeUnicode(str: string): string {
-  return str
-    .replace(/[‘’‚‛]/g, "'") // single quotes
-    .replace(/[“”„‟]/g, '"') // double quotes
-    .replace(/[‐‑‒–—―]/g, "-") // dashes
-    .replace(/…/g, "...") // ellipsis
-    .replace(/ /g, " ") // non-breaking space
-}
-
-type Comparator = (a: string, b: string) => boolean
-
-function tryMatch(lines: string[], pattern: string[], startIndex: number, compare: Comparator, eof: boolean): number {
-  // If EOF anchor, try matching from end of file first
-  if (eof) {
-    const fromEnd = lines.length - pattern.length
-    if (fromEnd >= startIndex) {
-      let matches = true
-      for (let j = 0; j < pattern.length; j++) {
-        if (!compare(lines[fromEnd + j], pattern[j])) {
-          matches = false
-          break
-        }
-      }
-      if (matches) return fromEnd
-    }
-  }
-
-  // Forward search from startIndex
-  for (let i = startIndex; i <= lines.length - pattern.length; i++) {
-    let matches = true
-    for (let j = 0; j < pattern.length; j++) {
-      if (!compare(lines[i + j], pattern[j])) {
-        matches = false
-        break
-      }
-    }
-    if (matches) return i
-  }
-
-  return -1
-}
-
-function seekSequence(lines: string[], pattern: string[], startIndex: number, eof = false): number {
-  if (pattern.length === 0) return -1
-
-  // Pass 1: exact match
-  const exact = tryMatch(lines, pattern, startIndex, (a, b) => a === b, eof)
-  if (exact !== -1) return exact
-
-  // Pass 2: rstrip (trim trailing whitespace)
-  const rstrip = tryMatch(lines, pattern, startIndex, (a, b) => a.trimEnd() === b.trimEnd(), eof)
-  if (rstrip !== -1) return rstrip
-
-  // Pass 3: trim (both ends)
-  const trim = tryMatch(lines, pattern, startIndex, (a, b) => a.trim() === b.trim(), eof)
-  if (trim !== -1) return trim
-
-  // Pass 4: normalized (Unicode punctuation to ASCII)
-  const normalized = tryMatch(
-    lines,
-    pattern,
-    startIndex,
-    (a, b) => normalizeUnicode(a.trim()) === normalizeUnicode(b.trim()),
-    eof,
-  )
-  return normalized
-}
-
-function generateUnifiedDiff(oldContent: string, newContent: string): string {
-  const oldLines = oldContent.split("\n")
-  const newLines = newContent.split("\n")
-
-  // Simple diff generation - in a real implementation you'd use a proper diff algorithm
-  let diff = "@@ -1 +1 @@\n"
-
-  // Find changes (simplified approach)
-  const maxLen = Math.max(oldLines.length, newLines.length)
-  let hasChanges = false
-
-  for (let i = 0; i < maxLen; i++) {
-    const oldLine = oldLines[i] || ""
-    const newLine = newLines[i] || ""
-
-    if (oldLine !== newLine) {
-      if (oldLine) diff += `-${oldLine}\n`
-      if (newLine) diff += `+${newLine}\n`
-      hasChanges = true
-    } else if (oldLine) {
-      diff += ` ${oldLine}\n`
-    }
-  }
-
-  return hasChanges ? diff : ""
-}
-
-// Apply hunks to filesystem
-export const applyHunksToFiles = Effect.fn("Patch.applyHunksToFiles")(function* (hunks: Hunk[]) {
-  if (hunks.length === 0) {
-    return yield* Effect.fail(new Error("No files were modified."))
-  }
-
-  const fs = yield* AppFileSystem.Service
-
-  const added: string[] = []
-  const modified: string[] = []
-  const deleted: string[] = []
-
-  for (const hunk of hunks) {
-    switch (hunk.type) {
-      case "add": {
-        yield* fs.writeWithDirs(hunk.path, hunk.contents)
-        added.push(hunk.path)
-        log.info(`Added file: ${hunk.path}`)
-        break
-      }
-
-      case "delete": {
-        yield* fs.remove(hunk.path)
-        deleted.push(hunk.path)
-        log.info(`Deleted file: ${hunk.path}`)
-        break
-      }
-
-      case "update": {
-        const originalText = yield* fs.readFileString(hunk.path)
-        const fileUpdate = deriveNewContentsFromChunks(hunk.path, hunk.chunks, originalText)
-
-        if (hunk.move_path) {
-          yield* fs.writeWithDirs(hunk.move_path, Bom.join(fileUpdate.content, fileUpdate.bom))
-          yield* fs.remove(hunk.path)
-          modified.push(hunk.move_path)
-          log.info(`Moved file: ${hunk.path} -> ${hunk.move_path}`)
-        } else {
-          yield* fs.writeWithDirs(hunk.path, Bom.join(fileUpdate.content, fileUpdate.bom))
-          modified.push(hunk.path)
-          log.info(`Updated file: ${hunk.path}`)
-        }
-        break
-      }
-    }
-  }
-
-  return { added, modified, deleted } satisfies AffectedPaths
-})
-
-// Main patch application function
-export const applyPatch = Effect.fn("Patch.applyPatch")(function* (patchText: string) {
-  const { hunks } = parsePatch(patchText)
-  return yield* applyHunksToFiles(hunks)
-})
-
-type MaybeApplyPatchVerifiedResult =
-  | { type: MaybeApplyPatchVerified.Body; action: ApplyPatchAction }
-  | { type: MaybeApplyPatchVerified.CorrectnessError; error: Error }
-  | { type: MaybeApplyPatchVerified.NotApplyPatch }
-
-// Effectful verified-parse: needs AppFileSystem.Service to read existing files
-export const maybeParseApplyPatchVerified = Effect.fn("Patch.maybeParseApplyPatchVerified")(function* (
-  argv: string[],
-  cwd: string,
-) {
-  // Detect implicit patch invocation (raw patch without apply_patch command)
-  if (argv.length === 1) {
-    try {
-      parsePatch(argv[0])
-      return {
-        type: MaybeApplyPatchVerified.CorrectnessError,
-        error: new Error(ApplyPatchError.ImplicitInvocation),
-      } satisfies MaybeApplyPatchVerifiedResult
-    } catch {
-      // Not a patch, continue
-    }
-  }
-
-  const result = maybeParseApplyPatch(argv)
-
-  switch (result.type) {
-    case MaybeApplyPatch.Body: {
-      const fs = yield* AppFileSystem.Service
-      const args = result.args
-      const effectiveCwd = args.workdir ? path.resolve(cwd, args.workdir) : cwd
-      const changes = new Map<string, ApplyPatchFileChange>()
-
-      for (const hunk of args.hunks) {
-        const resolvedPath = path.resolve(
-          effectiveCwd,
-          hunk.type === "update" && hunk.move_path ? hunk.move_path : hunk.path,
-        )
-
-        switch (hunk.type) {
-          case "add":
-            changes.set(resolvedPath, {
-              type: "add",
-              content: hunk.contents,
-            })
-            break
-
-          case "delete": {
-            const deletePath = path.resolve(effectiveCwd, hunk.path)
-            const content = yield* fs.readFileString(deletePath).pipe(Effect.catch(() => Effect.succeed(undefined)))
-            if (content === undefined) {
-              return {
-                type: MaybeApplyPatchVerified.CorrectnessError,
-                error: new Error(`Failed to read file for deletion: ${deletePath}`),
-              } satisfies MaybeApplyPatchVerifiedResult
-            }
-            changes.set(resolvedPath, {
-              type: "delete",
-              content,
-            })
-            break
-          }
-
-          case "update": {
-            const updatePath = path.resolve(effectiveCwd, hunk.path)
-            const originalText = yield* fs
-              .readFileString(updatePath)
-              .pipe(
-                Effect.catch((cause) =>
-                  Effect.succeed(new Error(`Failed to read file ${updatePath}: ${cause}`, { cause })),
-                ),
-              )
-            if (originalText instanceof Error) {
-              return {
-                type: MaybeApplyPatchVerified.CorrectnessError,
-                error: originalText,
-              } satisfies MaybeApplyPatchVerifiedResult
-            }
-            try {
-              const fileUpdate = deriveNewContentsFromChunks(updatePath, hunk.chunks, originalText)
-              changes.set(resolvedPath, {
-                type: "update",
-                unified_diff: fileUpdate.unified_diff,
-                move_path: hunk.move_path ? path.resolve(effectiveCwd, hunk.move_path) : undefined,
-                new_content: fileUpdate.content,
-              })
-            } catch (error) {
-              return {
-                type: MaybeApplyPatchVerified.CorrectnessError,
-                error: error as Error,
-              } satisfies MaybeApplyPatchVerifiedResult
-            }
-            break
-          }
-        }
-      }
-
-      return {
-        type: MaybeApplyPatchVerified.Body,
-        action: {
-          changes,
-          patch: args.patch,
-          cwd: effectiveCwd,
-        },
-      } satisfies MaybeApplyPatchVerifiedResult
-    }
-
-    case MaybeApplyPatch.PatchParseError:
-      return {
-        type: MaybeApplyPatchVerified.CorrectnessError,
-        error: result.error,
-      } satisfies MaybeApplyPatchVerifiedResult
-
-    case MaybeApplyPatch.NotApplyPatch:
-      return { type: MaybeApplyPatchVerified.NotApplyPatch } satisfies MaybeApplyPatchVerifiedResult
-  }
-})
-
-export * as Patch from "."
+import { Effect, Schema } from "effect"
+import * as path from "path"
+import { AppFileSystem } from "@opencode-ai/core/filesystem"
+import * as Log from "@opencode-ai/core/util/log"
+import * as Bom from "../util/bom"
+
+const log = Log.create({ service: "patch" })
+
+export const PatchSchema = Schema.Struct({
+  patchText: Schema.String.annotate({ description: "The full patch text that describes all changes to be made" }),
+})
+
+export type PatchParams = Schema.Schema.Type<typeof PatchSchema>
+
+// Core types matching the Rust implementation
+export interface ApplyPatchArgs {
+  patch: string
+  hunks: Hunk[]
+  workdir?: string
+}
+
+export type Hunk =
+  | { type: "add"; path: string; contents: string }
+  | { type: "delete"; path: string }
+  | { type: "update"; path: string; move_path?: string; chunks: UpdateFileChunk[] }
+
+export interface UpdateFileChunk {
+  old_lines: string[]
+  new_lines: string[]
+  change_context?: string
+  is_end_of_file?: boolean
+}
+
+export interface ApplyPatchAction {
+  changes: Map<string, ApplyPatchFileChange>
+  patch: string
+  cwd: string
+}
+
+export type ApplyPatchFileChange =
+  | { type: "add"; content: string }
+  | { type: "delete"; content: string }
+  | { type: "update"; unified_diff: string; move_path?: string; new_content: string }
+
+export interface AffectedPaths {
+  added: string[]
+  modified: string[]
+  deleted: string[]
+}
+
+export enum ApplyPatchError {
+  ParseError = "ParseError",
+  IoError = "IoError",
+  ComputeReplacements = "ComputeReplacements",
+  ImplicitInvocation = "ImplicitInvocation",
+}
+
+export enum MaybeApplyPatch {
+  Body = "Body",
+  ShellParseError = "ShellParseError",
+  PatchParseError = "PatchParseError",
+  NotApplyPatch = "NotApplyPatch",
+}
+
+export enum MaybeApplyPatchVerified {
+  Body = "Body",
+  ShellParseError = "ShellParseError",
+  CorrectnessError = "CorrectnessError",
+  NotApplyPatch = "NotApplyPatch",
+}
+
+// Parser implementation
+function parsePatchHeader(
+  lines: string[],
+  startIdx: number,
+): { filePath: string; movePath?: string; nextIdx: number } | null {
+  const line = lines[startIdx]
+
+  if (line.startsWith("*** Add File:")) {
+    const filePath = line.slice("*** Add File:".length).trim()
+    return filePath ? { filePath, nextIdx: startIdx + 1 } : null
+  }
+
+  if (line.startsWith("*** Delete File:")) {
+    const filePath = line.slice("*** Delete File:".length).trim()
+    return filePath ? { filePath, nextIdx: startIdx + 1 } : null
+  }
+
+  if (line.startsWith("*** Update File:")) {
+    const filePath = line.slice("*** Update File:".length).trim()
+    let movePath: string | undefined
+    let nextIdx = startIdx + 1
+
+    // Check for move directive
+    if (nextIdx < lines.length && lines[nextIdx].startsWith("*** Move to:")) {
+      movePath = lines[nextIdx].slice("*** Move to:".length).trim()
+      nextIdx++
+    }
+
+    return filePath ? { filePath, movePath, nextIdx } : null
+  }
+
+  return null
+}
+
+function parseUpdateFileChunks(lines: string[], startIdx: number): { chunks: UpdateFileChunk[]; nextIdx: number } {
+  const chunks: UpdateFileChunk[] = []
+  let i = startIdx
+
+  while (i < lines.length && !lines[i].startsWith("***")) {
+    if (lines[i].startsWith("@@")) {
+      // Parse context line
+      const contextLine = lines[i].substring(2).trim()
+      i++
+
+      const oldLines: string[] = []
+      const newLines: string[] = []
+      let isEndOfFile = false
+
+      // Parse change lines
+      while (i < lines.length && !lines[i].startsWith("@@") && !lines[i].startsWith("***")) {
+        const changeLine = lines[i]
+
+        if (changeLine === "*** End of File") {
+          isEndOfFile = true
+          i++
+          break
+        }
+
+        if (changeLine.startsWith(" ")) {
+          // Keep line - appears in both old and new
+          const content = changeLine.substring(1)
+          oldLines.push(content)
+          newLines.push(content)
+        } else if (changeLine.startsWith("-")) {
+          // Remove line - only in old
+          oldLines.push(changeLine.substring(1))
+        } else if (changeLine.startsWith("+")) {
+          // Add line - only in new
+          newLines.push(changeLine.substring(1))
+        }
+
+        i++
+      }
+
+      chunks.push({
+        old_lines: oldLines,
+        new_lines: newLines,
+        change_context: contextLine || undefined,
+        is_end_of_file: isEndOfFile || undefined,
+      })
+    } else {
+      i++
+    }
+  }
+
+  return { chunks, nextIdx: i }
+}
+
+function parseAddFileContent(lines: string[], startIdx: number): { content: string; nextIdx: number } {
+  let content = ""
+  let i = startIdx
+
+  while (i < lines.length && !lines[i].startsWith("***")) {
+    if (lines[i].startsWith("+")) {
+      content += lines[i].substring(1) + "\n"
+    }
+    i++
+  }
+
+  // Remove trailing newline
+  if (content.endsWith("\n")) {
+    content = content.slice(0, -1)
+  }
+
+  return { content, nextIdx: i }
+}
+
+function stripHeredoc(input: string): string {
+  // Match heredoc patterns like: cat <<'EOF'\n...\nEOF or <<EOF\n...\nEOF
+  const heredocMatch = input.match(/^(?:cat\s+)?<<['"]?(\w+)['"]?\s*\n([\s\S]*?)\n\1\s*$/)
+  if (heredocMatch) {
+    return heredocMatch[2]
+  }
+  return input
+}
+
+export function parsePatch(patchText: string): { hunks: Hunk[] } {
+  const cleaned = stripHeredoc(patchText.trim())
+  const lines = cleaned.split("\n")
+  const hunks: Hunk[] = []
+  let i = 0
+
+  // Look for Begin/End patch markers
+  const beginMarker = "*** Begin Patch"
+  const endMarker = "*** End Patch"
+
+  const beginIdx = lines.findIndex((line) => line.trim() === beginMarker)
+  const endIdx = lines.findIndex((line) => line.trim() === endMarker)
+
+  if (beginIdx === -1 || endIdx === -1 || beginIdx >= endIdx) {
+    throw new Error("Invalid patch format: missing Begin/End markers")
+  }
+
+  // Parse content between markers
+  i = beginIdx + 1
+
+  while (i < endIdx) {
+    const header = parsePatchHeader(lines, i)
+    if (!header) {
+      i++
+      continue
+    }
+
+    if (lines[i].startsWith("*** Add File:")) {
+      const { content, nextIdx } = parseAddFileContent(lines, header.nextIdx)
+      hunks.push({
+        type: "add",
+        path: header.filePath,
+        contents: content,
+      })
+      i = nextIdx
+    } else if (lines[i].startsWith("*** Delete File:")) {
+      hunks.push({
+        type: "delete",
+        path: header.filePath,
+      })
+      i = header.nextIdx
+    } else if (lines[i].startsWith("*** Update File:")) {
+      const { chunks, nextIdx } = parseUpdateFileChunks(lines, header.nextIdx)
+      hunks.push({
+        type: "update",
+        path: header.filePath,
+        move_path: header.movePath,
+        chunks,
+      })
+      i = nextIdx
+    } else {
+      i++
+    }
+  }
+
+  return { hunks }
+}
+
+// Apply patch functionality
+export function maybeParseApplyPatch(
+  argv: string[],
+):
+  | { type: MaybeApplyPatch.Body; args: ApplyPatchArgs }
+  | { type: MaybeApplyPatch.PatchParseError; error: Error }
+  | { type: MaybeApplyPatch.NotApplyPatch } {
+  const APPLY_PATCH_COMMANDS = ["apply_patch", "applypatch"]
+
+  // Direct invocation: apply_patch <patch>
+  if (argv.length === 2 && APPLY_PATCH_COMMANDS.includes(argv[0])) {
+    try {
+      const { hunks } = parsePatch(argv[1])
+      return {
+        type: MaybeApplyPatch.Body,
+        args: {
+          patch: argv[1],
+          hunks,
+        },
+      }
+    } catch (error) {
+      return {
+        type: MaybeApplyPatch.PatchParseError,
+        error: error as Error,
+      }
+    }
+  }
+
+  // Bash heredoc form: bash -lc 'apply_patch <<"EOF" ...'
+  if (argv.length === 3 && argv[0] === "bash" && argv[1] === "-lc") {
+    // Simple extraction - in real implementation would need proper bash parsing
+    const script = argv[2]
+    const heredocMatch = script.match(/apply_patch\s*<<['"](\w+)['"]\s*\n([\s\S]*?)\n\1/)
+
+    if (heredocMatch) {
+      const patchContent = heredocMatch[2]
+      try {
+        const { hunks } = parsePatch(patchContent)
+        return {
+          type: MaybeApplyPatch.Body,
+          args: {
+            patch: patchContent,
+            hunks,
+          },
+        }
+      } catch (error) {
+        return {
+          type: MaybeApplyPatch.PatchParseError,
+          error: error as Error,
+        }
+      }
+    }
+  }
+
+  return { type: MaybeApplyPatch.NotApplyPatch }
+}
+
+// File content manipulation
+interface ApplyPatchFileUpdate {
+  unified_diff: string
+  content: string
+  bom: boolean
+}
+
+export function deriveNewContentsFromChunks(
+  filePath: string,
+  chunks: UpdateFileChunk[],
+  originalText: string,
+): ApplyPatchFileUpdate {
+  const originalContent = Bom.split(originalText)
+
+  let originalLines = originalContent.text.split("\n")
+
+  // Drop trailing empty element for consistent line counting
+  if (originalLines.length > 0 && originalLines[originalLines.length - 1] === "") {
+    originalLines.pop()
+  }
+
+  const replacements = computeReplacements(originalLines, filePath, chunks)
+  let newLines = applyReplacements(originalLines, replacements)
+
+  // Ensure trailing newline
+  if (newLines.length === 0 || newLines[newLines.length - 1] !== "") {
+    newLines.push("")
+  }
+
+  const next = Bom.split(newLines.join("\n"))
+  const newContent = next.text
+
+  // Generate unified diff
+  const unifiedDiff = generateUnifiedDiff(originalContent.text, newContent)
+
+  return {
+    unified_diff: unifiedDiff,
+    content: newContent,
+    bom: originalContent.bom || next.bom,
+  }
+}
+
+function computeReplacements(
+  originalLines: string[],
+  filePath: string,
+  chunks: UpdateFileChunk[],
+): Array<[number, number, string[]]> {
+  const replacements: Array<[number, number, string[]]> = []
+  let lineIndex = 0
+
+  for (const chunk of chunks) {
+    // Handle context-based seeking
+    if (chunk.change_context) {
+      const contextIdx = seekSequence(originalLines, [chunk.change_context], lineIndex)
+      if (contextIdx === -1) {
+        throw new Error(`Failed to find context '${chunk.change_context}' in ${filePath}`)
+      }
+      lineIndex = contextIdx + 1
+    }
+
+    // Handle pure addition (no old lines)
+    if (chunk.old_lines.length === 0) {
+      const insertionIdx =
+        originalLines.length > 0 && originalLines[originalLines.length - 1] === ""
+          ? originalLines.length - 1
+          : originalLines.length
+      replacements.push([insertionIdx, 0, chunk.new_lines])
+      continue
+    }
+
+    // Try to match old lines in the file
+    let pattern = chunk.old_lines
+    let newSlice = chunk.new_lines
+    let found = seekSequence(originalLines, pattern, lineIndex, chunk.is_end_of_file)
+
+    // Retry without trailing empty line if not found
+    if (found === -1 && pattern.length > 0 && pattern[pattern.length - 1] === "") {
+      pattern = pattern.slice(0, -1)
+      if (newSlice.length > 0 && newSlice[newSlice.length - 1] === "") {
+        newSlice = newSlice.slice(0, -1)
+      }
+      found = seekSequence(originalLines, pattern, lineIndex, chunk.is_end_of_file)
+    }
+
+    if (found !== -1) {
+      replacements.push([found, pattern.length, newSlice])
+      lineIndex = found + pattern.length
+    } else {
+      throw new Error(`Failed to find expected lines in ${filePath}:\n${chunk.old_lines.join("\n")}`)
+    }
+  }
+
+  // Sort replacements by index to apply in order
+  replacements.sort((a, b) => a[0] - b[0])
+
+  return replacements
+}
+
+function applyReplacements(lines: string[], replacements: Array<[number, number, string[]]>): string[] {
+  // Apply replacements in reverse order to avoid index shifting
+  const result = [...lines]
+
+  for (let i = replacements.length - 1; i >= 0; i--) {
+    const [startIdx, oldLen, newSegment] = replacements[i]
+
+    // Remove old lines
+    result.splice(startIdx, oldLen)
+
+    // Insert new lines
+    for (let j = 0; j < newSegment.length; j++) {
+      result.splice(startIdx + j, 0, newSegment[j])
+    }
+  }
+
+  return result
+}
+
+// Normalize Unicode punctuation to ASCII equivalents (like Rust's normalize_unicode)
+function normalizeUnicode(str: string): string {
+  return str
+    .replace(/[‘’‚‛]/g, "'") // single quotes
+    .replace(/[“”„‟]/g, '"') // double quotes
+    .replace(/[‐‑‒–—―]/g, "-") // dashes
+    .replace(/…/g, "...") // ellipsis
+    .replace(/ /g, " ") // non-breaking space
+}
+
+type Comparator = (a: string, b: string) => boolean
+
+function tryMatch(lines: string[], pattern: string[], startIndex: number, compare: Comparator, eof: boolean): number {
+  // If EOF anchor, try matching from end of file first
+  if (eof) {
+    const fromEnd = lines.length - pattern.length
+    if (fromEnd >= startIndex) {
+      let matches = true
+      for (let j = 0; j < pattern.length; j++) {
+        if (!compare(lines[fromEnd + j], pattern[j])) {
+          matches = false
+          break
+        }
+      }
+      if (matches) return fromEnd
+    }
+  }
+
+  // Forward search from startIndex
+  for (let i = startIndex; i <= lines.length - pattern.length; i++) {
+    let matches = true
+    for (let j = 0; j < pattern.length; j++) {
+      if (!compare(lines[i + j], pattern[j])) {
+        matches = false
+        break
+      }
+    }
+    if (matches) return i
+  }
+
+  return -1
+}
+
+function seekSequence(lines: string[], pattern: string[], startIndex: number, eof = false): number {
+  if (pattern.length === 0) return -1
+
+  // Pass 1: exact match
+  const exact = tryMatch(lines, pattern, startIndex, (a, b) => a === b, eof)
+  if (exact !== -1) return exact
+
+  // Pass 2: rstrip (trim trailing whitespace)
+  const rstrip = tryMatch(lines, pattern, startIndex, (a, b) => a.trimEnd() === b.trimEnd(), eof)
+  if (rstrip !== -1) return rstrip
+
+  // Pass 3: trim (both ends)
+  const trim = tryMatch(lines, pattern, startIndex, (a, b) => a.trim() === b.trim(), eof)
+  if (trim !== -1) return trim
+
+  // Pass 4: normalized (Unicode punctuation to ASCII)
+  const normalized = tryMatch(
+    lines,
+    pattern,
+    startIndex,
+    (a, b) => normalizeUnicode(a.trim()) === normalizeUnicode(b.trim()),
+    eof,
+  )
+  return normalized
+}
+
+function generateUnifiedDiff(oldContent: string, newContent: string): string {
+  const oldLines = oldContent.split("\n")
+  const newLines = newContent.split("\n")
+
+  // Simple diff generation - in a real implementation you'd use a proper diff algorithm
+  let diff = "@@ -1 +1 @@\n"
+
+  // Find changes (simplified approach)
+  const maxLen = Math.max(oldLines.length, newLines.length)
+  let hasChanges = false
+
+  for (let i = 0; i < maxLen; i++) {
+    const oldLine = oldLines[i] || ""
+    const newLine = newLines[i] || ""
+
+    if (oldLine !== newLine) {
+      if (oldLine) diff += `-${oldLine}\n`
+      if (newLine) diff += `+${newLine}\n`
+      hasChanges = true
+    } else if (oldLine) {
+      diff += ` ${oldLine}\n`
+    }
+  }
+
+  return hasChanges ? diff : ""
+}
+
+// Apply hunks to filesystem
+export const applyHunksToFiles = Effect.fn("Patch.applyHunksToFiles")(function* (hunks: Hunk[]) {
+  if (hunks.length === 0) {
+    return yield* Effect.fail(new Error("No files were modified."))
+  }
+
+  const fs = yield* AppFileSystem.Service
+
+  const added: string[] = []
+  const modified: string[] = []
+  const deleted: string[] = []
+
+  for (const hunk of hunks) {
+    switch (hunk.type) {
+      case "add": {
+        yield* fs.writeWithDirs(hunk.path, hunk.contents)
+        added.push(hunk.path)
+        log.info(`Added file: ${hunk.path}`)
+        break
+      }
+
+      case "delete": {
+        yield* fs.remove(hunk.path)
+        deleted.push(hunk.path)
+        log.info(`Deleted file: ${hunk.path}`)
+        break
+      }
+
+      case "update": {
+        const originalText = yield* fs.readFileString(hunk.path)
+        const fileUpdate = deriveNewContentsFromChunks(hunk.path, hunk.chunks, originalText)
+
+        if (hunk.move_path) {
+          yield* fs.writeWithDirs(hunk.move_path, Bom.join(fileUpdate.content, fileUpdate.bom))
+          yield* fs.remove(hunk.path)
+          modified.push(hunk.move_path)
+          log.info(`Moved file: ${hunk.path} -> ${hunk.move_path}`)
+        } else {
+          yield* fs.writeWithDirs(hunk.path, Bom.join(fileUpdate.content, fileUpdate.bom))
+          modified.push(hunk.path)
+          log.info(`Updated file: ${hunk.path}`)
+        }
+        break
+      }
+    }
+  }
+
+  return { added, modified, deleted } satisfies AffectedPaths
+})
+
+// Main patch application function
+export const applyPatch = Effect.fn("Patch.applyPatch")(function* (patchText: string) {
+  const { hunks } = parsePatch(patchText)
+  return yield* applyHunksToFiles(hunks)
+})
+
+type MaybeApplyPatchVerifiedResult =
+  | { type: MaybeApplyPatchVerified.Body; action: ApplyPatchAction }
+  | { type: MaybeApplyPatchVerified.CorrectnessError; error: Error }
+  | { type: MaybeApplyPatchVerified.NotApplyPatch }
+
+// Effectful verified-parse: needs AppFileSystem.Service to read existing files
+export const maybeParseApplyPatchVerified = Effect.fn("Patch.maybeParseApplyPatchVerified")(function* (
+  argv: string[],
+  cwd: string,
+) {
+  // Detect implicit patch invocation (raw patch without apply_patch command)
+  if (argv.length === 1) {
+    try {
+      parsePatch(argv[0])
+      return {
+        type: MaybeApplyPatchVerified.CorrectnessError,
+        error: new Error(ApplyPatchError.ImplicitInvocation),
+      } satisfies MaybeApplyPatchVerifiedResult
+    } catch {
+      // Not a patch, continue
+    }
+  }
+
+  const result = maybeParseApplyPatch(argv)
+
+  switch (result.type) {
+    case MaybeApplyPatch.Body: {
+      const fs = yield* AppFileSystem.Service
+      const args = result.args
+      const effectiveCwd = args.workdir ? path.resolve(cwd, args.workdir) : cwd
+      const changes = new Map<string, ApplyPatchFileChange>()
+
+      for (const hunk of args.hunks) {
+        const resolvedPath = path.resolve(
+          effectiveCwd,
+          hunk.type === "update" && hunk.move_path ? hunk.move_path : hunk.path,
+        )
+
+        switch (hunk.type) {
+          case "add":
+            changes.set(resolvedPath, {
+              type: "add",
+              content: hunk.contents,
+            })
+            break
+
+          case "delete": {
+            const deletePath = path.resolve(effectiveCwd, hunk.path)
+            const content = yield* fs.readFileString(deletePath).pipe(Effect.catch(() => Effect.succeed(undefined)))
+            if (content === undefined) {
+              return {
+                type: MaybeApplyPatchVerified.CorrectnessError,
+                error: new Error(`Failed to read file for deletion: ${deletePath}`),
+              } satisfies MaybeApplyPatchVerifiedResult
+            }
+            changes.set(resolvedPath, {
+              type: "delete",
+              content,
+            })
+            break
+          }
+
+          case "update": {
+            const updatePath = path.resolve(effectiveCwd, hunk.path)
+            const originalText = yield* fs
+              .readFileString(updatePath)
+              .pipe(
+                Effect.catch((cause) =>
+                  Effect.succeed(new Error(`Failed to read file ${updatePath}: ${cause}`, { cause })),
+                ),
+              )
+            if (originalText instanceof Error) {
+              return {
+                type: MaybeApplyPatchVerified.CorrectnessError,
+                error: originalText,
+              } satisfies MaybeApplyPatchVerifiedResult
+            }
+            try {
+              const fileUpdate = deriveNewContentsFromChunks(updatePath, hunk.chunks, originalText)
+              changes.set(resolvedPath, {
+                type: "update",
+                unified_diff: fileUpdate.unified_diff,
+                move_path: hunk.move_path ? path.resolve(effectiveCwd, hunk.move_path) : undefined,
+                new_content: fileUpdate.content,
+              })
+            } catch (error) {
+              return {
+                type: MaybeApplyPatchVerified.CorrectnessError,
+                error: error as Error,
+              } satisfies MaybeApplyPatchVerifiedResult
+            }
+            break
+          }
+        }
+      }
+
+      return {
+        type: MaybeApplyPatchVerified.Body,
+        action: {
+          changes,
+          patch: args.patch,
+          cwd: effectiveCwd,
+        },
+      } satisfies MaybeApplyPatchVerifiedResult
+    }
+
+    case MaybeApplyPatch.PatchParseError:
+      return {
+        type: MaybeApplyPatchVerified.CorrectnessError,
+        error: result.error,
+      } satisfies MaybeApplyPatchVerifiedResult
+
+    case MaybeApplyPatch.NotApplyPatch:
+      return { type: MaybeApplyPatchVerified.NotApplyPatch } satisfies MaybeApplyPatchVerifiedResult
+  }
+})
+
+export * as Patch from "."