toolLooper/tools/patch_files.js

#!/usr/bin/env node

/**
 * A self-contained JavaScript utility for applying human-readable
 * "pseudo-diff" patch files to a collection of text files.
 * 





*** Begin Patch
[YOUR_PATH]
*** End Patch


Where [YOUR_PATCH] is the actual content of your patch, specified in the following V4A diff format.

*** [ACTION] File: [path/to/file] -> ACTION can be one of Add, Update, or Delete.
For each snippet of code that needs to be changed, repeat the following:
[context_before] -> See below for further instructions on context.
- [old_code] -> Precede the old code with a minus sign.
+ [new_code] -> Precede the new, replacement code with a plus sign.
[context_after] -> See below for further instructions on context.

For instructions on [context_before] and [context_after]:
- By default, show 3 lines of code immediately above and 3 lines immediately below each change. If a change is within 3 lines of a previous change, do NOT duplicate the first change’s [context_after] lines in the second change’s [context_before] lines.
- If 3 lines of context is insufficient to uniquely identify the snippet of code within the file, use the @@ operator to indicate the class or function to which the snippet belongs. For instance, we might have:
@@ class BaseClass
[3 lines of pre-context]
- [old_code]
+ [new_code]
[3 lines of post-context]

- If a code block is repeated so many times in a class or function such that even a single @@ statement and 3 lines of context cannot uniquely identify the snippet of code, you can use multiple `@@` statements to jump to the right context. For instance:

@@ class BaseClass
@@ 	def method():
[3 lines of pre-context]
- [old_code]
+ [new_code]
[3 lines of post-context]

Note, then, that we do not use line numbers in this diff format, as the context is enough to uniquely identify code. An example of a message that you might pass as "input" to this function, in order to apply a patch, is shown below.

 * 
 * 
 * 
 */

// --------------------------------------------------------------------------- //
//  Domain objects
// --------------------------------------------------------------------------- //
class ActionType {
    static ADD = "add";
    static DELETE = "delete";
    static UPDATE = "update";
}

class FileChange {
    constructor(type, old_content = null, new_content = null, move_path = null) {
        this.type = type;
        this.old_content = old_content;
        this.new_content = new_content;
        this.move_path = move_path;
    }
}

class Commit {
    constructor() {
        this.changes = {};
    }
}

// --------------------------------------------------------------------------- //
//  Exceptions
// --------------------------------------------------------------------------- //
class DiffError extends Error {
    /** Any problem detected while parsing or applying a patch. */
    constructor(message) {
        super(message);
        this.name = "DiffError";
    }
}

// --------------------------------------------------------------------------- //
//  Helper classes used while parsing patches
// --------------------------------------------------------------------------- //
class Chunk {
    constructor(orig_index = -1, del_lines = [], ins_lines = []) {
        this.orig_index = orig_index;
        this.del_lines = del_lines;
        this.ins_lines = ins_lines;
    }
}

class PatchAction {
    constructor(type, new_file = null, chunks = [], move_path = null) {
        this.type = type;
        this.new_file = new_file;
        this.chunks = chunks;
        this.move_path = move_path;
    }
}

class Patch {
    constructor() {
        this.actions = {};
    }
}

// --------------------------------------------------------------------------- //
//  Patch text parser
// --------------------------------------------------------------------------- //
class Parser {
    constructor(current_files, lines) {
        this.current_files = current_files;
        this.lines = lines;
        this.index = 0;
        this.patch = new Patch();
        this.fuzz = 0;
    }

    // ------------- low-level helpers -------------------------------------- //
    _cur_line() {
        if (this.index >= this.lines.length) {
            throw new DiffError("Unexpected end of input while parsing patch");
        }
        return this.lines[this.index];
    }

    static _norm(line) {
        /** Strip CR so comparisons work for both LF and CRLF input. */
        return line.replace(/\r$/, "");
    }

    // ------------- scanning convenience ----------------------------------- //
    is_done(prefixes = null) {
        if (this.index >= this.lines.length) {
            return true;
        }
        if (
            prefixes &&
            prefixes.length > 0
        ) {
            const currentLine = Parser._norm(this._cur_line());
            for (const prefix of prefixes) {
                if (currentLine.startsWith(prefix)) {
                    return true;
                }
            }
        }
        return false;
    }

    startswith(prefix) {
        return Parser._norm(this._cur_line()).startsWith(prefix);
    }

    read_str(prefix) {
        /**
         * Consume the current line if it starts with *prefix* and return the text
         * **after** the prefix.  Raises if prefix is empty.
         */
        if (prefix === "") {
            throw new Error("read_str() requires a non-empty prefix");
        }
        if (Parser._norm(this._cur_line()).startsWith(prefix)) {
            const text = this._cur_line().substring(prefix.length);
            this.index += 1;
            return text;
        }
        return "";
    }

    read_line() {
        /** Return the current raw line and advance. */
        const line = this._cur_line();
        this.index += 1;
        return line;
    }

    // ------------- public entry point -------------------------------------- //
    parse() {
        while (!this.is_done(["*** End Patch"])) {
            // ---------- UPDATE ---------- //
            let path = this.read_str("*** Update File: ");
            if (path) {
                if (path in this.patch.actions) {
                    throw new DiffError(`Duplicate update for file: ${path}`);
                }
                const move_to = this.read_str("*** Move to: ");
                if (!(path in this.current_files)) {
                    throw new DiffError(`Update File Error - missing file: ${path}`);
                }
                const text = this.current_files[path];
                const action = this._parse_update_file(text);
                action.move_path = move_to || null;
                this.patch.actions[path] = action;
                continue;
            }

            // ---------- DELETE ---------- //
            path = this.read_str("*** Delete File: ");
            if (path) {
                if (path in this.patch.actions) {
                    throw new DiffError(`Duplicate delete for file: ${path}`);
                }
                if (!(path in this.current_files)) {
                    throw new DiffError(`Delete File Error - missing file: ${path}`);
                }
                this.patch.actions[path] = new PatchAction(ActionType.DELETE);
                continue;
            }

            // ---------- ADD ---------- //
            path = this.read_str("*** Add File: ");
            if (path) {
                if (path in this.patch.actions) {
                    throw new DiffError(`Duplicate add for file: ${path}`);
                }
                if (path in this.current_files) {
                    throw new DiffError(`Add File Error - file already exists: ${path}`);
                }
                this.patch.actions[path] = this._parse_add_file();
                continue;
            }

            throw new DiffError(`Unknown line while parsing: ${this._cur_line()}`);
        }

        if (!this.startswith("*** End Patch")) {
            throw new DiffError("Missing *** End Patch sentinel");
        }
        this.index += 1; // consume sentinel
    }

    // ------------- section parsers ---------------------------------------- //
    _parse_update_file(text) {
        const action = new PatchAction(ActionType.UPDATE);
        const lines = text.split("\n");
        let index = 0;
        while (!this.is_done([
            "*** End Patch",
            "*** Update File:",
            "*** Delete File:",
            "*** Add File:",
            "*** End of File",
        ])) {
            const def_str = this.read_str("@@ ");
            let section_str = "";
            if (!def_str && Parser._norm(this._cur_line()) === "@@") {
                section_str = this.read_line();
            }

            if (!(def_str || section_str || index === 0)) {
                throw new DiffError(`Invalid line in update section:\n${this._cur_line()}`);
            }

            if (def_str && def_str.trim()) {
                let found = false;
                if (!lines.slice(0, index).includes(def_str)) {
                    for (let i = index; i < lines.length; i++) {
                        if (lines[i] === def_str) {
                            index = i + 1;
                            found = true;
                            break;
                        }
                    }
                }
                if (!found && !lines.slice(0, index).map(s => s.trim()).includes(def_str.trim())) {
                    for (let i = index; i < lines.length; i++) {
                        if (lines[i].trim() === def_str.trim()) {
                            index = i + 1;
                            this.fuzz += 1;
                            found = true;
                            break;
                        }
                    }
                }
            }

            const [next_ctx, chunks, end_idx, eof] = peek_next_section(this.lines, this.index);
            const [new_index, fuzz] = find_context(lines, next_ctx, index, eof);
            if (new_index === -1) {
                const ctx_txt = next_ctx.join("\n");
                throw new DiffError(
                    `Invalid ${eof ? 'EOF ' : ''}context at ${index}:\n${ctx_txt}`
                );
            }
            this.fuzz += fuzz;
            for (const ch of chunks) {
                ch.orig_index += new_index;
                action.chunks.push(ch);
            }
            index = new_index + next_ctx.length;
            this.index = end_idx;
        }
        return action;
    }

    _parse_add_file() {
        const lines = [];
        while (!this.is_done([
            "*** End Patch",
            "*** Update File:",
            "*** Delete File:",
            "*** Add File:"
        ])) {
            const s = this.read_line();
            if (!s.startsWith("+")) {
                throw new DiffError(`Invalid Add File line (missing '+'): ${s}`);
            }
            lines.push(s.substring(1)); // strip leading '+'
        }
        return new PatchAction(ActionType.ADD, lines.join("\n"));
    }
}

// --------------------------------------------------------------------------- //
//  Helper functions
// --------------------------------------------------------------------------- //
function find_context_core(lines, context, start) {
    if (!context || context.length === 0) {
        return [start, 0];
    }

    // Exact match
    for (let i = start; i <= lines.length - context.length; i++) {
        let match = true;
        for (let j = 0; j < context.length; j++) {
            if (lines[i + j] !== context[j]) {
                match = false;
                break;
            }
        }
        if (match) {
            return [i, 0];
        }
    }
    
    // Rstrip match
    for (let i = start; i <= lines.length - context.length; i++) {
        let match = true;
        for (let j = 0; j < context.length; j++) {
            if (lines[i + j].replace(/\s+$/, "") !== context[j].replace(/\s+$/, "")) {
                match = false;
                break;
            }
        }
        if (match) {
            return [i, 1];
        }
    }
    
    // Trim match
    for (let i = start; i <= lines.length - context.length; i++) {
        let match = true;
        for (let j = 0; j < context.length; j++) {
            if (lines[i + j].trim() !== context[j].trim()) {
                match = false;
                break;
            }
        }
        if (match) {
            return [i, 100];
        }
    }
    
    return [-1, 0];
}

function find_context(lines, context, start, eof) {
    if (eof) {
        let [new_index, fuzz] = find_context_core(lines, context, Math.max(0, lines.length - context.length));
        if (new_index !== -1) {
            return [new_index, fuzz];
        }
        [new_index, fuzz] = find_context_core(lines, context, start);
        return [new_index, fuzz + 10000];
    }
    return find_context_core(lines, context, start);
}

function peek_next_section(lines, index) {
    const old = [];
    let del_lines = [];
    let ins_lines = [];
    const chunks = [];
    let mode = "keep";
    const orig_index = index;

    while (index < lines.length) {
        let s = lines[index];
        if (s.startsWith("@@") || 
            s === "*** End Patch" || 
            s.startsWith("*** Update File:") || 
            s.startsWith("*** Delete File:") || 
            s.startsWith("*** Add File:") || 
            s === "*** End of File") {
            break;
        }
        if (s === "***") {
            break;
        }
        if (s.startsWith("***")) {
            throw new DiffError(`Invalid Line: ${s}`);
        }
        index += 1;

        const last_mode = mode;
        if (s === "") {
            s = " ";
        }
        if (s[0] === "+") {
            mode = "add";
        } else if (s[0] === "-") {
            mode = "delete";
        } else if (s[0] === " ") {
            mode = "keep";
        } else {
            throw new DiffError(`Invalid Line: ${s}`);
        }
        s = s.substring(1);

        if (mode === "keep" && last_mode !== mode) {
            if (ins_lines.length > 0 || del_lines.length > 0) {
                chunks.push(
                    new Chunk(
                        old.length - del_lines.length,
                        [...del_lines],
                        [...ins_lines]
                    )
                );
            }
            del_lines = [];
            ins_lines = [];
        }

        if (mode === "delete") {
            del_lines.push(s);
            old.push(s);
        } else if (mode === "add") {
            ins_lines.push(s);
        } else if (mode === "keep") {
            old.push(s);
        }
    }

    if (ins_lines.length > 0 || del_lines.length > 0) {
        chunks.push(
            new Chunk(
                old.length - del_lines.length,
                [...del_lines],
                [...ins_lines]
            )
        );
    }

    if (index < lines.length && lines[index] === "*** End of File") {
        index += 1;
        return [old, chunks, index, true];
    }

    if (index === orig_index) {
        throw new DiffError("Nothing in this section");
    }
    return [old, chunks, index, false];
}

// --------------------------------------------------------------------------- //
//  Patch → Commit and Commit application
// --------------------------------------------------------------------------- //
function _get_updated_file(text, action, path) {
    if (action.type !== ActionType.UPDATE) {
        throw new DiffError("_get_updated_file called with non-update action");
    }
    const orig_lines = text.split("\n");
    const dest_lines = [];
    let orig_index = 0;

    for (const chunk of action.chunks) {
        if (chunk.orig_index > orig_lines.length) {
            throw new DiffError(
                `${path}: chunk.orig_index ${chunk.orig_index} exceeds file length`
            );
        }
        if (orig_index > chunk.orig_index) {
            throw new DiffError(
                `${path}: overlapping chunks at ${orig_index} > ${chunk.orig_index}`
            );
        }

        // Add lines before the chunk
        for (let i = orig_index; i < chunk.orig_index; i++) {
            dest_lines.push(orig_lines[i]);
        }
        orig_index = chunk.orig_index;

        // Add inserted lines
        dest_lines.push(...chunk.ins_lines);
        
        // Skip deleted lines
        orig_index += chunk.del_lines.length;
    }

    // Add remaining lines
    for (let i = orig_index; i < orig_lines.length; i++) {
        dest_lines.push(orig_lines[i]);
    }
    
    return dest_lines.join("\n");
}

function patch_to_commit(patch, orig) {
    const commit = new Commit();
    for (const [path, action] of Object.entries(patch.actions)) {
        if (action.type === ActionType.DELETE) {
            commit.changes[path] = new FileChange(
                ActionType.DELETE,
                orig[path],
                null,
                null
            );
        } else if (action.type === ActionType.ADD) {
            if (action.new_file === null) {
                throw new DiffError("ADD action without file content");
            }
            commit.changes[path] = new FileChange(
                ActionType.ADD,
                null,
                action.new_file,
                null
            );
        } else if (action.type === ActionType.UPDATE) {
            const new_content = _get_updated_file(orig[path], action, path);
            commit.changes[path] = new FileChange(
                ActionType.UPDATE,
                orig[path],
                new_content,
                action.move_path
            );
        }
    }
    return commit;
}

// --------------------------------------------------------------------------- //
//  User-facing helpers
// --------------------------------------------------------------------------- //
function text_to_patch(text, orig) {
    // Handle different line ending scenarios
    let lines = text.split(/\r?\n/);
    
    // Remove trailing empty line if it exists (from trailing newline)
    if (lines.length > 0 && lines[lines.length - 1] === "") {
        lines = lines.slice(0, -1);
    }
    
    // Debug logging
    console.log("Lines count:", lines.length);
    console.log("First line:", JSON.stringify(lines[0]));
    if (lines.length > 0) {
        console.log("Last line:", JSON.stringify(lines[lines.length - 1]));
        console.log("First line normalized:", JSON.stringify(Parser._norm(lines[0])));
        console.log("Last line normalized:", JSON.stringify(Parser._norm(lines[lines.length - 1])));
    }
    
    if (
        lines.length < 2 ||
        !Parser._norm(lines[0]).startsWith("*** Begin Patch") ||
        (lines.length > 0 && Parser._norm(lines[lines.length - 1]) !== "*** End Patch")
    ) {
        throw new DiffError("Invalid patch text - missing sentinels");
    }

    const parser = new Parser(orig, lines);
    parser.index = 1;
    parser.parse();
    return [parser.patch, parser.fuzz];
}

function identify_files_needed(text) {
    // Handle line splitting consistently
    let lines = text.split(/\r?\n/);
    if (lines.length > 0 && lines[lines.length - 1] === "") {
        lines = lines.slice(0, -1);
    }
    
    const updateFiles = lines
        .filter(line => line.startsWith("*** Update File: "))
        .map(line => line.substring("*** Update File: ".length));
    
    const deleteFiles = lines
        .filter(line => line.startsWith("*** Delete File: "))
        .map(line => line.substring("*** Delete File: ".length));
    
    return [...updateFiles, ...deleteFiles];
}

function identify_files_added(text) {
    // Handle line splitting consistently
    let lines = text.split(/\r?\n/);
    if (lines.length > 0 && lines[lines.length - 1] === "") {
        lines = lines.slice(0, -1);
    }
    
    return lines
        .filter(line => line.startsWith("*** Add File: "))
        .map(line => line.substring("*** Add File: ".length));
}

// --------------------------------------------------------------------------- //
//  File-system helpers
// --------------------------------------------------------------------------- //
function load_files(paths, open_fn) {
    const result = {};
    for (const path of paths) {
        result[path] = open_fn(path);
    }
    return result;
}

function apply_commit(commit, write_fn, remove_fn) {
    for (const [path, change] of Object.entries(commit.changes)) {
        if (change.type === ActionType.DELETE) {
            remove_fn(path);
        } else if (change.type === ActionType.ADD) {
            if (change.new_content === null) {
                throw new DiffError(`ADD change for ${path} has no content`);
            }
            write_fn(path, change.new_content);
        } else if (change.type === ActionType.UPDATE) {
            if (change.new_content === null) {
                throw new DiffError(`UPDATE change for ${path} has no new content`);
            }
            const target = change.move_path || path;
            write_fn(target, change.new_content);
            if (change.move_path) {
                remove_fn(path);
            }
        }
    }
}

function process_patch(text, open_fn, write_fn, remove_fn) {
    if (!text.startsWith("*** Begin Patch")) {
        throw new DiffError("Patch text must start with *** Begin Patch");
    }
    const paths = identify_files_needed(text);
    const orig = load_files(paths, open_fn);
    const [patch, _fuzz] = text_to_patch(text, orig);
    const commit = patch_to_commit(patch, orig);
    apply_commit(commit, write_fn, remove_fn);
    return "Done!";
}

// --------------------------------------------------------------------------- //
//  Default FS helpers
// --------------------------------------------------------------------------- //
import fs from 'fs';
import path from 'path';

function open_file(filepath) {
    return fs.readFileSync(filepath, 'utf8');
}

function write_file(filepath, content) {
    const dir = path.dirname(filepath);
    if (!fs.existsSync(dir)) {
        fs.mkdirSync(dir, { recursive: true });
    }
    fs.writeFileSync(filepath, content, 'utf8');
}

function remove_file(filepath) {
    if (fs.existsSync(filepath)) {
        fs.unlinkSync(filepath);
    }
}

// Export for the new interface
export default {
    type: "function",
    name: "patch_files",
    description: "Apply a unified diff patch to files within a chroot directory, with option to reverse the patch",
    parameters: {
        type: "object",
        properties: {
            patch: {
                type: "string",
                description: "The unidiff patch string to apply.",
            }
        },
        required: ["patch"],
        additionalProperties: false,
    },
    strict: true,
};

export async function run(args) {
    console.log('patch_files:', args);
    
    try {
        const result = process_patch(
            args.patch, 
            open_file, 
            write_file, 
            remove_file
        );
        return result;
    } catch (error) {
        if (error instanceof DiffError) {
            throw new Error(`Patch error: ${error.message}`);
        }
        throw error;
    }
}