commit a795a38cfe10087df87b8dc5ebbc00726f0241ac Author: sirlilpanda Date: Wed Jun 17 00:43:43 2026 +1200 piecetable impl with undo diff --git a/build.zig b/build.zig new file mode 100644 index 0000000..14f6333 --- /dev/null +++ b/build.zig @@ -0,0 +1,156 @@ +const std = @import("std"); + +// Although this function looks imperative, it does not perform the build +// directly and instead it mutates the build graph (`b`) that will be then +// executed by an external runner. The functions in `std.Build` implement a DSL +// for defining build steps and express dependencies between them, allowing the +// build runner to parallelize the build automatically (and the cache system to +// know when a step doesn't need to be re-run). +pub fn build(b: *std.Build) void { + // Standard target options allow the person running `zig build` to choose + // what target to build for. Here we do not override the defaults, which + // means any target is allowed, and the default is native. Other options + // for restricting supported target set are available. + const target = b.standardTargetOptions(.{}); + // Standard optimization options allow the person running `zig build` to select + // between Debug, ReleaseSafe, ReleaseFast, and ReleaseSmall. Here we do not + // set a preferred release mode, allowing the user to decide how to optimize. + const optimize = b.standardOptimizeOption(.{}); + // It's also possible to define more custom flags to toggle optional features + // of this build script using `b.option()`. All defined flags (including + // target and optimize options) will be listed when running `zig build --help` + // in this directory. + + // This creates a module, which represents a collection of source files alongside + // some compilation options, such as optimization mode and linked system libraries. + // Zig modules are the preferred way of making Zig code available to consumers. + // addModule defines a module that we intend to make available for importing + // to our consumers. We must give it a name because a Zig package can expose + // multiple modules and consumers will need to be able to specify which + // module they want to access. + const mod = b.addModule("zig_piecetable", .{ + // The root source file is the "entry point" of this module. Users of + // this module will only be able to access public declarations contained + // in this file, which means that if you have declarations that you + // intend to expose to consumers that were defined in other files part + // of this module, you will have to make sure to re-export them from + // the root file. + .root_source_file = b.path("src/root.zig"), + // Later on we'll use this module as the root module of a test executable + // which requires us to specify a target. + .target = target, + }); + + // Here we define an executable. An executable needs to have a root module + // which needs to expose a `main` function. While we could add a main function + // to the module defined above, it's sometimes preferable to split business + // logic and the CLI into two separate modules. + // + // If your goal is to create a Zig library for others to use, consider if + // it might benefit from also exposing a CLI tool. A parser library for a + // data serialization format could also bundle a CLI syntax checker, for example. + // + // If instead your goal is to create an executable, consider if users might + // be interested in also being able to embed the core functionality of your + // program in their own executable in order to avoid the overhead involved in + // subprocessing your CLI tool. + // + // If neither case applies to you, feel free to delete the declaration you + // don't need and to put everything under a single module. + const exe = b.addExecutable(.{ + .name = "zig_piecetable", + .root_module = b.createModule(.{ + // b.createModule defines a new module just like b.addModule but, + // unlike b.addModule, it does not expose the module to consumers of + // this package, which is why in this case we don't have to give it a name. + .root_source_file = b.path("src/main.zig"), + // Target and optimization levels must be explicitly wired in when + // defining an executable or library (in the root module), and you + // can also hardcode a specific target for an executable or library + // definition if desireable (e.g. firmware for embedded devices). + .target = target, + .optimize = optimize, + // List of modules available for import in source files part of the + // root module. + .imports = &.{ + // Here "zig_piecetable" is the name you will use in your source code to + // import this module (e.g. `@import("zig_piecetable")`). The name is + // repeated because you are allowed to rename your imports, which + // can be extremely useful in case of collisions (which can happen + // importing modules from different packages). + .{ .name = "zig_piecetable", .module = mod }, + }, + }), + }); + + // This declares intent for the executable to be installed into the + // install prefix when running `zig build` (i.e. when executing the default + // step). By default the install prefix is `zig-out/` but can be overridden + // by passing `--prefix` or `-p`. + b.installArtifact(exe); + + // This creates a top level step. Top level steps have a name and can be + // invoked by name when running `zig build` (e.g. `zig build run`). + // This will evaluate the `run` step rather than the default step. + // For a top level step to actually do something, it must depend on other + // steps (e.g. a Run step, as we will see in a moment). + const run_step = b.step("run", "Run the app"); + + // This creates a RunArtifact step in the build graph. A RunArtifact step + // invokes an executable compiled by Zig. Steps will only be executed by the + // runner if invoked directly by the user (in the case of top level steps) + // or if another step depends on it, so it's up to you to define when and + // how this Run step will be executed. In our case we want to run it when + // the user runs `zig build run`, so we create a dependency link. + const run_cmd = b.addRunArtifact(exe); + run_step.dependOn(&run_cmd.step); + + // By making the run step depend on the default step, it will be run from the + // installation directory rather than directly from within the cache directory. + run_cmd.step.dependOn(b.getInstallStep()); + + // This allows the user to pass arguments to the application in the build + // command itself, like this: `zig build run -- arg1 arg2 etc` + if (b.args) |args| { + run_cmd.addArgs(args); + } + + // Creates an executable that will run `test` blocks from the provided module. + // Here `mod` needs to define a target, which is why earlier we made sure to + // set the releative field. + const mod_tests = b.addTest(.{ + .root_module = mod, + }); + + // A run step that will run the test executable. + const run_mod_tests = b.addRunArtifact(mod_tests); + + // Creates an executable that will run `test` blocks from the executable's + // root module. Note that test executables only test one module at a time, + // hence why we have to create two separate ones. + const exe_tests = b.addTest(.{ + .root_module = exe.root_module, + }); + + // A run step that will run the second test executable. + const run_exe_tests = b.addRunArtifact(exe_tests); + + // A top level step for running all tests. dependOn can be called multiple + // times and since the two run steps do not depend on one another, this will + // make the two of them run in parallel. + const test_step = b.step("test", "Run tests"); + test_step.dependOn(&run_mod_tests.step); + test_step.dependOn(&run_exe_tests.step); + + // Just like flags, top level steps are also listed in the `--help` menu. + // + // The Zig build system is entirely implemented in userland, which means + // that it cannot hook into private compiler APIs. All compilation work + // orchestrated by the build system will result in other Zig compiler + // subcommands being invoked with the right flags defined. You can observe + // these invocations when one fails (or you pass a flag to increase + // verbosity) to validate assumptions and diagnose problems. + // + // Lastly, the Zig build system is relatively simple and self-contained, + // and reading its source code will allow you to master it. +} diff --git a/build.zig.zon b/build.zig.zon new file mode 100644 index 0000000..a4d6bad --- /dev/null +++ b/build.zig.zon @@ -0,0 +1,81 @@ +.{ + // This is the default name used by packages depending on this one. For + // example, when a user runs `zig fetch --save `, this field is used + // as the key in the `dependencies` table. Although the user can choose a + // different name, most users will stick with this provided value. + // + // It is redundant to include "zig" in this name because it is already + // within the Zig package namespace. + .name = .zig_piecetable, + // This is a [Semantic Version](https://semver.org/). + // In a future version of Zig it will be used for package deduplication. + .version = "0.0.0", + // Together with name, this represents a globally unique package + // identifier. This field is generated by the Zig toolchain when the + // package is first created, and then *never changes*. This allows + // unambiguous detection of one package being an updated version of + // another. + // + // When forking a Zig project, this id should be regenerated (delete the + // field and run `zig build`) if the upstream project is still maintained. + // Otherwise, the fork is *hostile*, attempting to take control over the + // original project's identity. Thus it is recommended to leave the comment + // on the following line intact, so that it shows up in code reviews that + // modify the field. + .fingerprint = 0x9be7c0b888bfd1e9, // Changing this has security and trust implications. + // Tracks the earliest Zig version that the package considers to be a + // supported use case. + .minimum_zig_version = "0.16.0", + // This field is optional. + // Each dependency must either provide a `url` and `hash`, or a `path`. + // `zig build --fetch` can be used to fetch all dependencies of a package, recursively. + // Once all dependencies are fetched, `zig build` no longer requires + // internet connectivity. + .dependencies = .{ + // See `zig fetch --save ` for a command-line interface for adding dependencies. + //.example = .{ + // // When updating this field to a new URL, be sure to delete the corresponding + // // `hash`, otherwise you are communicating that you expect to find the old hash at + // // the new URL. If the contents of a URL change this will result in a hash mismatch + // // which will prevent zig from using it. + // .url = "https://example.com/foo.tar.gz", + // + // // This is computed from the file contents of the directory of files that is + // // obtained after fetching `url` and applying the inclusion rules given by + // // `paths`. + // // + // // This field is the source of truth; packages do not come from a `url`; they + // // come from a `hash`. `url` is just one of many possible mirrors for how to + // // obtain a package matching this `hash`. + // // + // // Uses the [multihash](https://multiformats.io/multihash/) format. + // .hash = "...", + // + // // When this is provided, the package is found in a directory relative to the + // // build root. In this case the package's hash is irrelevant and therefore not + // // computed. This field and `url` are mutually exclusive. + // .path = "foo", + // + // // When this is set to `true`, a package is declared to be lazily + // // fetched. This makes the dependency only get fetched if it is + // // actually used. + // .lazy = false, + //}, + }, + // Specifies the set of files and directories that are included in this package. + // Only files and directories listed here are included in the `hash` that + // is computed for this package. Only files listed here will remain on disk + // when using the zig package manager. As a rule of thumb, one should list + // files required for compilation plus any license(s). + // Paths are relative to the build root. Use the empty string (`""`) to refer to + // the build root itself. + // A directory listed here means that all files within, recursively, are included. + .paths = .{ + "build.zig", + "build.zig.zon", + "src", + // For example... + //"LICENSE", + //"README.md", + }, +} diff --git a/src/root.zig b/src/root.zig new file mode 100644 index 0000000..d468cda --- /dev/null +++ b/src/root.zig @@ -0,0 +1,319 @@ +//! By convention, root.zig is the root source file when making a package. +const std = @import("std"); +const Io = std.Io; + +const Piece = struct { + // each piece could just be a slice however this make it easier to write + start: usize, + length: usize, + + text: *const []const u8, + + const nothing: []const u8 = ""; + + pub const none = Piece{ + .start = 0, + .length = 0, + .text = ¬hing, + }; + + pub fn format(self: Piece, writer: *std.Io.Writer) !void { + try writer.print( + "piece {{src[{},{}] : \"{s}\"}}", + .{ self.start, self.length, self.text.* }, + ); + } +}; + +const ActionType = enum { + append, + insert, + insert_splits_piece, + delete, +}; + +const Action = struct { + piece_table_index: usize, + old_start: usize, + old_length: usize, +}; + +const ActionGroup = union(ActionType) { + /// this will only occur when the table was already empty so boof it + append: void, + + /// happens when a piece is cleanly inserted in to the table + insert: usize, + + /// happens when the piece requires to be split before indexing + insert_splits_piece: Action, + + /// happens when a piece is deletes (length set to zero) + /// appends all modifed pieces to be restored + delete: std.ArrayList(Action), +}; + +/// abstracted so i can implement redo +const ActionList = struct { + const Self = @This(); + + actions: std.ArrayList(ActionGroup) = .empty, + + pub const empty = ActionList{}; + + pub fn append(self: *Self, alloc: std.mem.Allocator, action_group: ActionGroup) !void { + try self.actions.append(alloc, action_group); + } + + pub fn pop(self: *Self) ?ActionGroup { + return self.actions.pop(); + } + + pub fn deinit(self: *Self, alloc: std.mem.Allocator) void { + for (self.actions.items) |*act| { + switch (act.*) { + .delete => act.delete.deinit(alloc), + else => continue, + } + } + self.actions.deinit(alloc); + } +}; + +/// piece table with built in undo +/// since the piece table is going in a text editor +/// if you dont have undo is not going to be a very +/// good text editor we aint trying to be ed here +const PieceTable = struct { + const Self = @This(); + const Alloc = std.mem.Allocator; + + pieces: std.ArrayList(Piece) = .empty, + actions: ActionList = .empty, + + // i have this here because its easy + total_len: usize = 0, + + pub fn init() Self { + return Self{}; + } + + /// self : piece table that is being operated on + /// alloc : allocator used for adding to the arraylists + /// piece : the piece to add + /// index : the index in the text where the new piece will be added + pub fn insert(self: *Self, alloc: Alloc, piece: Piece, index: usize) !void { + var currnet_start: usize = 0; + + for (self.pieces.items, 0..) |p, i| { + // find the index where the split occurs + + if (currnet_start + p.length > index) { + // we gotta split the piece + + const pieces: [3]Piece = [3]Piece{ + Piece{ + .text = p.text, + .start = p.start, + .length = index - currnet_start, + }, + piece, + Piece{ + .text = p.text, + .start = p.start + (index - currnet_start), + .length = p.length - (index - currnet_start), + }, + }; + + try self.pieces.replaceRange(alloc, i, 1, &pieces); + try self.actions.append(alloc, .{ + .insert_splits_piece = Action{ + .piece_table_index = i, + .old_length = p.length, + .old_start = p.start, + }, + }); + return; + } + if (currnet_start + p.length == index) { + try self.pieces.insert(alloc, i + 1, piece); + try self.actions.append(alloc, .{ .insert = i + 1 }); + return; + } // just append after + if (currnet_start + p.length < index) currnet_start += p.length; + } + + // if there are no items this runs + try self.pieces.append(alloc, piece); + try self.actions.append(alloc, .append); + } + + /// will delete all from (index) to (index) + /// self : piece table that is being operated on + /// alloc : allocator used for adding the actions to the action list. + /// from : the index to be deleted from + /// to : the index the deletion will stop + pub fn delete(self: *Self, alloc: std.mem.Allocator, from: usize, to: usize) !void { + var currnet_index: usize = 0; + var edited_piece_index: usize = 0; + var delete_actions: std.ArrayList(Action) = .empty; + + // from set + for (self.pieces.items, 0..) |p, i| { + // find the index where the split occurs + if (currnet_index + p.length > from) { + // we gotta split the piece + var current_piece = &self.pieces.items[i]; + + try delete_actions.append( + alloc, + Action{ + .piece_table_index = i, + .old_start = current_piece.start, + .old_length = current_piece.length, + }, + ); + + current_piece.length -= from - currnet_index; + edited_piece_index = i; + + break; + } + if (currnet_index + p.length <= from) currnet_index += p.length; + } + + // to set + + for (self.pieces.items, edited_piece_index..) |p, i| { + if (currnet_index + p.length < to) { + var current_piece = &self.pieces.items[i]; + + try delete_actions.append( + alloc, + Action{ + .piece_table_index = i, + .old_start = current_piece.start, + .old_length = current_piece.length, + }, + ); + + current_piece.length = 0; + currnet_index += p.length; + } + + if (currnet_index + p.length >= to) { + var current_piece = &self.pieces.items[i]; + + try delete_actions.append( + alloc, + Action{ + .piece_table_index = i, + .old_start = current_piece.start, + .old_length = current_piece.length, + }, + ); + + // shrink it + current_piece.start += to - currnet_index; + current_piece.length -= to - currnet_index; + break; + } + } + + try self.actions.append(alloc, .{ .delete = delete_actions }); + } + + pub fn undo(self: *Self, alloc: std.mem.Allocator) !void { + var action_to_undo = self.actions.pop() orelse return; + + switch (action_to_undo) { + .append => { + self.pieces.deinit(alloc); + self.pieces = .empty; + }, + .insert => |insert_index| { + + // for the small price of one alloc you too can just swap nothing in + // might be faster, dont know should test + // [TODO] test if this is faster + try self.pieces.append(alloc, .none); + _ = self.pieces.swapRemove(insert_index); + }, + .insert_splits_piece => |action| { + + // remove the 2 addtional pieces + self.pieces.orderedRemoveMany(&.{ + action.piece_table_index + 1, + action.piece_table_index + 2, + }); + + const first_half_of_split_piece = &self.pieces.items[action.piece_table_index]; + first_half_of_split_piece.length = action.old_length; + first_half_of_split_piece.start = action.old_start; + }, + .delete => |deletes| { + for (deletes.items) |action| { + const piece = &self.pieces.items[action.piece_table_index]; + piece.length = action.old_length; + piece.start = action.old_start; + } + + action_to_undo.delete.deinit(alloc); + }, + } + } + + pub fn format(self: Self, writer: *std.Io.Writer) !void { + _ = try writer.write("table : \""); + for (self.pieces.items) |p| { + _ = try writer.write(p.text.*[p.start .. p.start + p.length]); + } + _ = try writer.write("\""); + } + + // pub fn undo(self: Self) !void {} + + pub fn deinit(self: *Self, alloc: std.mem.Allocator) void { + self.actions.deinit(alloc); + self.pieces.deinit(alloc); + } +}; + +test "PieceTable insert" { + const orignal_text: []const u8 = "damn these are some cool toads"; + const new_text: []const u8 = "frogs and "; + const even_newer_text: []const u8 = "cats but not "; + + var piece_table = PieceTable.init(); + + try piece_table.insert(std.testing.allocator, .{ + .start = 0, + .length = orignal_text.len, + .text = &orignal_text, + }, 0); + std.debug.print("{f}\n", .{piece_table}); + + try piece_table.insert(std.testing.allocator, .{ + .text = &new_text, + .start = 0, + .length = new_text.len, + }, 20); + std.debug.print("{f}\n", .{piece_table}); + + try piece_table.delete(std.testing.allocator, 20, 20 + new_text.len); + + std.debug.print("{f}\n", .{piece_table}); + + try piece_table.insert(std.testing.allocator, .{ + .text = &even_newer_text, + .start = 0, + .length = even_newer_text.len, + }, 20); + std.debug.print("{f}\n", .{piece_table}); + + try piece_table.undo(std.testing.allocator); + + std.debug.print("{f}\n", .{piece_table}); + + piece_table.deinit(std.testing.allocator); +}