Files
bun.sh/src/fs.zig
Meghan Denny 45760cd53c ci: instrument being able to run leaksanitizer (#21142)
tests not in `test/no-validate-leaksan.txt` will run with leaksanitizer
in CI
leaks documented in `test/leaksan.supp` will not cause a test failure

-- notes about leaksanitizer

- will not catch garbage collected objects accumulated during
long-running processes
- will not catch js objects (eg a strong held to a promise)
- will catch native calls to `malloc` not `free`d
- will catch allocations made in C, Zig, C++, libc, dependencies,
dlopen'd

---------

Co-authored-by: Jarred Sumner <jarred@jarredsumner.com>
2025-09-19 02:06:02 -07:00

1993 lines
74 KiB
Zig

const Fs = @This();
pub const debug = Output.scoped(.fs, .hidden);
// pub const FilesystemImplementation = @import("./fs_impl.zig");
pub const Preallocate = struct {
pub const Counts = struct {
pub const dir_entry: usize = 2048;
pub const files: usize = 4096;
};
};
pub const FileSystem = struct {
top_level_dir: stringZ,
// used on subsequent updates
top_level_dir_buf: bun.PathBuffer = undefined,
fs: Implementation,
dirname_store: *DirnameStore,
filename_store: *FilenameStore,
threadlocal var tmpdir_handle: ?std.fs.Dir = null;
pub fn topLevelDirWithoutTrailingSlash(this: *const FileSystem) []const u8 {
if (this.top_level_dir.len > 1 and this.top_level_dir[this.top_level_dir.len - 1] == std.fs.path.sep) {
return this.top_level_dir[0 .. this.top_level_dir.len - 1];
} else {
return this.top_level_dir;
}
}
pub fn tmpdir(fs: *FileSystem) !std.fs.Dir {
if (tmpdir_handle == null) {
tmpdir_handle = try fs.fs.openTmpDir();
}
return tmpdir_handle.?;
}
pub fn getFdPath(this: *const FileSystem, fd: FileDescriptor) ![]const u8 {
var buf: bun.PathBuffer = undefined;
const dir = try bun.getFdPath(fd, &buf);
return try this.dirname_store.append([]u8, dir);
}
var tmpname_id_number = std.atomic.Value(u32).init(0);
pub fn tmpname(_: *const FileSystem, extname: string, buf: []u8, hash: u64) ![*:0]u8 {
const hex_value = @as(u64, @truncate(@as(u128, @intCast(hash)) | @as(u128, @intCast(std.time.nanoTimestamp()))));
return try std.fmt.bufPrintZ(buf, ".{any}-{any}.{s}", .{
bun.fmt.hexIntLower(hex_value),
bun.fmt.hexIntUpper(tmpname_id_number.fetchAdd(1, .monotonic)),
extname,
});
}
pub var max_fd: std.posix.fd_t = 0;
pub inline fn setMaxFd(fd: std.posix.fd_t) void {
if (Environment.isWindows) {
return;
}
if (!FeatureFlags.store_file_descriptors) {
return;
}
max_fd = @max(fd, max_fd);
}
pub var instance_loaded: bool = false;
pub var instance: FileSystem = undefined;
pub const DirnameStore = allocators.BSSStringList(Preallocate.Counts.dir_entry, 128);
pub const FilenameStore = allocators.BSSStringList(Preallocate.Counts.files, 64);
pub const Error = error{
ENOENT,
EACCESS,
INVALID_NAME,
ENOTDIR,
};
pub fn init(top_level_dir: ?stringZ) !*FileSystem {
return initWithForce(top_level_dir, false);
}
pub fn initWithForce(top_level_dir_: ?stringZ, comptime force: bool) !*FileSystem {
const allocator = bun.default_allocator;
var top_level_dir = top_level_dir_ orelse (if (Environment.isBrowser) "/project/" else try bun.getcwdAlloc(allocator));
_ = &top_level_dir;
if (!instance_loaded or force) {
instance = FileSystem{
.top_level_dir = top_level_dir,
.fs = Implementation.init(top_level_dir),
// must always use default_allocator since the other allocators may not be threadsafe when an element resizes
.dirname_store = DirnameStore.init(bun.default_allocator),
.filename_store = FilenameStore.init(bun.default_allocator),
};
instance_loaded = true;
_ = DirEntry.EntryStore.init(allocator);
}
return &instance;
}
pub fn deinit(this: *const FileSystem) void {
this.dirname_store.deinit();
this.filename_store.deinit();
}
pub const DirEntry = struct {
pub const EntryMap = bun.StringHashMapUnmanaged(*Entry);
pub const EntryStore = allocators.BSSList(Entry, Preallocate.Counts.files);
dir: string,
fd: FD = .invalid,
generation: bun.Generation = 0,
data: EntryMap,
// pub fn removeEntry(dir: *DirEntry, name: string) !void {
// // dir.data.remove(name);
// }
pub fn addEntry(dir: *DirEntry, prev_map: ?*EntryMap, entry: *const bun.DirIterator.IteratorResult, allocator: std.mem.Allocator, comptime Iterator: type, iterator: Iterator) !void {
const name_slice = entry.name.slice();
const found_kind: ?Entry.Kind = switch (entry.kind) {
.directory => .dir,
.file => .file,
// For a symlink, we will need to stat the target later
.sym_link,
// Some filesystems return `.unknown` from getdents() no matter the actual kind of the file
// (often because it would be slow to look up the kind). If we get this, then code that
// needs the kind will have to find it out later by calling stat().
.unknown,
=> null,
.block_device,
.character_device,
.named_pipe,
.unix_domain_socket,
.whiteout,
.door,
.event_port,
=> return,
};
const stored = try brk: {
if (prev_map) |map| {
var stack_fallback = std.heap.stackFallback(512, allocator);
const stack = stack_fallback.get();
const prehashed = bun.StringHashMapContext.PrehashedCaseInsensitive.init(stack, name_slice);
defer prehashed.deinit(stack);
if (map.getAdapted(name_slice, prehashed)) |existing| {
existing.mutex.lock();
defer existing.mutex.unlock();
existing.dir = dir.dir;
existing.need_stat = existing.need_stat or
found_kind == null or
existing.cache.kind != found_kind;
// TODO: is this right?
if (existing.cache.kind != found_kind) {
// if found_kind is null, we have set need_stat above, so we
// store an arbitrary kind
existing.cache.kind = found_kind orelse .file;
existing.cache.symlink = PathString.empty;
}
break :brk existing;
}
}
// name_slice only lives for the duration of the iteration
const name = try strings.StringOrTinyString.initAppendIfNeeded(
name_slice,
*FileSystem.FilenameStore,
FileSystem.FilenameStore.instance,
);
const name_lowercased = try strings.StringOrTinyString.initLowerCaseAppendIfNeeded(
name_slice,
*FileSystem.FilenameStore,
FileSystem.FilenameStore.instance,
);
break :brk EntryStore.instance.append(.{
.base_ = name,
.base_lowercase_ = name_lowercased,
.dir = dir.dir,
.mutex = .{},
// Call "stat" lazily for performance. The "@material-ui/icons" package
// contains a directory with over 11,000 entries in it and running "stat"
// for each entry was a big performance issue for that package.
.need_stat = found_kind == null,
.cache = .{
.symlink = PathString.empty,
// if found_kind is null, we have set need_stat above, so we
// store an arbitrary kind
.kind = found_kind orelse .file,
},
});
};
const stored_name = stored.base();
try dir.data.put(allocator, stored.base_lowercase(), stored);
if (comptime Iterator != void) {
iterator.next(stored, dir.fd);
}
if (comptime FeatureFlags.verbose_fs) {
if (found_kind == .dir) {
Output.prettyln(" + {s}/", .{stored_name});
} else {
Output.prettyln(" + {s}", .{stored_name});
}
}
}
pub fn init(dir: string, generation: bun.Generation) DirEntry {
if (comptime FeatureFlags.verbose_fs) {
Output.prettyln("\n {s}", .{dir});
}
return .{
.dir = dir,
.data = .{},
.generation = generation,
};
}
pub const Err = struct {
original_err: anyerror,
canonical_error: anyerror,
};
pub fn deinit(d: *DirEntry, allocator: std.mem.Allocator) void {
d.data.deinit(allocator);
allocator.free(d.dir);
}
pub fn get(entry: *const DirEntry, _query: string) ?Entry.Lookup {
if (_query.len == 0 or _query.len > bun.MAX_PATH_BYTES) return null;
var scratch_lookup_buffer: bun.PathBuffer = undefined;
const query = strings.copyLowercaseIfNeeded(_query, &scratch_lookup_buffer);
const result = entry.data.get(query) orelse return null;
const basename = result.base();
if (!strings.eqlLong(basename, _query, true)) {
return Entry.Lookup{ .entry = result, .diff_case = Entry.Lookup.DifferentCase{
.dir = entry.dir,
.query = _query,
.actual = basename,
} };
}
return Entry.Lookup{ .entry = result, .diff_case = null };
}
pub fn getComptimeQuery(entry: *const DirEntry, comptime query_str: anytype) ?Entry.Lookup {
comptime var query_var: [query_str.len]u8 = undefined;
comptime for (query_str, 0..) |c, i| {
query_var[i] = std.ascii.toLower(c);
};
const query_hashed = comptime std.hash_map.hashString(&query_var);
const query = query_var[0..query_str.len].*;
const result = entry.data.getAdapted(
@as([]const u8, &query),
struct {
pub fn hash(_: @This(), _: []const u8) @TypeOf(query_hashed) {
return query_hashed;
}
pub fn eql(_: @This(), _: []const u8, b: []const u8) bool {
return strings.eqlComptime(b, query);
}
}{},
) orelse return null;
const basename = result.base();
if (!strings.eqlComptime(basename, comptime query[0..query_str.len])) {
return Entry.Lookup{
.entry = result,
.diff_case = Entry.Lookup.DifferentCase{
.dir = entry.dir,
.query = &query,
.actual = basename,
},
};
}
return Entry.Lookup{ .entry = result, .diff_case = null };
}
pub fn hasComptimeQuery(entry: *const DirEntry, comptime query_str: anytype) bool {
comptime var query_var: [query_str.len]u8 = undefined;
comptime for (query_str, 0..) |c, i| {
query_var[i] = std.ascii.toLower(c);
};
const query = query_var[0..query_str.len].*;
const query_hashed = comptime std.hash_map.hashString(&query);
return entry.data.containsAdapted(
@as([]const u8, &query),
struct {
pub fn hash(_: @This(), _: []const u8) @TypeOf(query_hashed) {
return query_hashed;
}
pub fn eql(_: @This(), _: []const u8, b: []const u8) bool {
return strings.eqlComptime(b, &query);
}
}{},
);
}
};
pub const Entry = struct {
cache: Cache = .{},
dir: string,
base_: strings.StringOrTinyString,
// Necessary because the hash table uses it as a key
base_lowercase_: strings.StringOrTinyString,
mutex: Mutex,
need_stat: bool = true,
abs_path: PathString = PathString.empty,
pub inline fn base(this: *Entry) string {
return this.base_.slice();
}
pub inline fn base_lowercase(this: *Entry) string {
return this.base_lowercase_.slice();
}
pub const Lookup = struct {
entry: *Entry,
diff_case: ?DifferentCase,
pub const DifferentCase = struct {
dir: string,
query: string,
actual: string,
};
};
pub fn deinit(e: *Entry, allocator: std.mem.Allocator) void {
e.base_.deinit(allocator);
allocator.free(e.dir);
allocator.free(e.cache.symlink.slice());
allocator.destroy(e);
}
pub const Cache = struct {
symlink: PathString = PathString.empty,
/// Too much code expects this to be 0
/// don't make it bun.invalid_fd
fd: FD = .invalid,
kind: Kind = .file,
};
pub const Kind = enum {
dir,
file,
};
pub fn kind(entry: *Entry, fs: *Implementation, store_fd: bool) Kind {
if (entry.need_stat) {
entry.need_stat = false;
// This is technically incorrect, but we are choosing not to handle errors here
entry.cache = fs.kind(entry.dir, entry.base(), entry.cache.fd, store_fd) catch return entry.cache.kind;
}
return entry.cache.kind;
}
pub fn symlink(entry: *Entry, fs: *Implementation, store_fd: bool) string {
if (entry.need_stat) {
entry.need_stat = false;
// This is technically incorrect, but we are choosing not to handle errors here
// This error can happen if the file was deleted between the time the directory was scanned and the time it was read
entry.cache = fs.kind(entry.dir, entry.base(), entry.cache.fd, store_fd) catch return "";
}
return entry.cache.symlink.slice();
}
};
// pub fn statBatch(fs: *FileSystemEntry, paths: []string) ![]?Stat {
// }
// pub fn stat(fs: *FileSystemEntry, path: string) !Stat {
// }
// pub fn readFile(fs: *FileSystemEntry, path: string) ?string {
// }
// pub fn readDir(fs: *FileSystemEntry, path: string) ?[]string {
// }
pub fn normalize(_: *@This(), str: string) string {
return @call(bun.callmod_inline, path_handler.normalizeString, .{ str, true, bun.path.Platform.auto });
}
pub fn normalizeBuf(_: *@This(), buf: []u8, str: string) string {
return @call(bun.callmod_inline, path_handler.normalizeStringBuf, .{ str, buf, false, bun.path.Platform.auto, false });
}
pub fn join(_: *@This(), parts: anytype) string {
return @call(bun.callmod_inline, path_handler.joinStringBuf, .{
&join_buf,
parts,
bun.path.Platform.loose,
});
}
pub fn joinBuf(_: *@This(), parts: anytype, buf: []u8) string {
return @call(bun.callmod_inline, path_handler.joinStringBuf, .{
buf,
parts,
bun.path.Platform.loose,
});
}
pub fn relative(_: *@This(), from: string, to: string) string {
return @call(bun.callmod_inline, path_handler.relative, .{
from,
to,
});
}
pub fn relativePlatform(_: *@This(), from: string, to: string, comptime platform: path_handler.Platform) string {
return @call(bun.callmod_inline, path_handler.relativePlatform, .{
from,
to,
platform,
false,
});
}
pub fn relativeTo(f: *@This(), to: string) string {
return @call(bun.callmod_inline, path_handler.relative, .{
f.top_level_dir,
to,
});
}
pub fn relativeFrom(f: *@This(), from: string) string {
return @call(bun.callmod_inline, path_handler.relative, .{
from,
f.top_level_dir,
});
}
pub fn absAlloc(f: *@This(), allocator: std.mem.Allocator, parts: anytype) !string {
const joined = path_handler.joinAbsString(
f.top_level_dir,
parts,
.loose,
);
return try allocator.dupe(u8, joined);
}
pub fn absAllocZ(f: *@This(), allocator: std.mem.Allocator, parts: anytype) ![*:0]const u8 {
const joined = path_handler.joinAbsString(
f.top_level_dir,
parts,
.loose,
);
return try allocator.dupeZ(u8, joined);
}
pub fn abs(f: *@This(), parts: anytype) string {
return path_handler.joinAbsString(
f.top_level_dir,
parts,
.loose,
);
}
pub fn absBuf(f: *@This(), parts: anytype, buf: []u8) string {
return path_handler.joinAbsStringBuf(f.top_level_dir, buf, parts, .loose);
}
pub fn absBufZ(f: *@This(), parts: anytype, buf: []u8) stringZ {
return path_handler.joinAbsStringBufZ(f.top_level_dir, buf, parts, .loose);
}
pub fn joinAlloc(f: *@This(), allocator: std.mem.Allocator, parts: anytype) !string {
const joined = f.join(parts);
return try allocator.dupe(u8, joined);
}
pub fn printLimits() void {
const LIMITS = [_]std.posix.rlimit_resource{ std.posix.rlimit_resource.STACK, std.posix.rlimit_resource.NOFILE };
Output.print("{{\n", .{});
inline for (LIMITS, 0..) |limit_type, i| {
const limit = std.posix.getrlimit(limit_type) catch return;
if (i == 0) {
Output.print(" \"stack\": [{d}, {d}],\n", .{ limit.cur, limit.max });
} else if (i == 1) {
Output.print(" \"files\": [{d}, {d}]\n", .{ limit.cur, limit.max });
}
}
Output.print("}}\n", .{});
Output.flush();
}
pub const RealFS = struct {
entries_mutex: Mutex = .{},
entries: *EntriesOption.Map,
cwd: string,
file_limit: usize = 32,
file_quota: usize = 32,
pub var win_tempdir_cache: ?[]const u8 = undefined;
pub fn platformTempDir() []const u8 {
return switch (Environment.os) {
// https://learn.microsoft.com/en-us/windows/win32/api/fileapi/nf-fileapi-gettemppathw#remarks
.windows => win_tempdir_cache orelse {
const value = bun.getenvZ("TEMP") orelse bun.getenvZ("TMP") orelse brk: {
if (bun.getenvZ("SystemRoot") orelse bun.getenvZ("windir")) |windir| {
break :brk std.fmt.allocPrint(
bun.default_allocator,
"{s}\\Temp",
.{strings.withoutTrailingSlash(windir)},
) catch |err| bun.handleOom(err);
}
if (bun.getenvZ("USERPROFILE")) |profile| {
var buf: bun.PathBuffer = undefined;
var parts = [_]string{"AppData\\Local\\Temp"};
const out = bun.path.joinAbsStringBuf(profile, &buf, &parts, .loose);
break :brk bun.handleOom(bun.default_allocator.dupe(u8, out));
}
var tmp_buf: bun.PathBuffer = undefined;
const cwd = std.posix.getcwd(&tmp_buf) catch @panic("Failed to get cwd for platformTempDir");
const root = bun.path.windowsFilesystemRoot(cwd);
break :brk std.fmt.allocPrint(
bun.default_allocator,
"{s}\\Windows\\Temp",
.{strings.withoutTrailingSlash(root)},
) catch |err| bun.handleOom(err);
};
win_tempdir_cache = value;
return value;
},
.mac => "/private/tmp",
else => "/tmp",
};
}
pub const Tmpfile = switch (Environment.os) {
.windows => TmpfileWindows,
else => TmpfilePosix,
};
pub var tmpdir_path: []const u8 = undefined;
pub var tmpdir_path_set = false;
pub fn tmpdirPath(_: *const @This()) []const u8 {
if (!tmpdir_path_set) {
tmpdir_path = bun.getenvZ("BUN_TMPDIR") orelse bun.getenvZ("TMPDIR") orelse platformTempDir();
tmpdir_path_set = true;
}
return tmpdir_path;
}
pub fn openTmpDir(_: *const RealFS) !std.fs.Dir {
if (!tmpdir_path_set) {
tmpdir_path = bun.getenvZ("BUN_TMPDIR") orelse bun.getenvZ("TMPDIR") orelse platformTempDir();
tmpdir_path_set = true;
}
if (comptime Environment.isWindows) {
return (try bun.sys.openDirAtWindowsA(bun.invalid_fd, tmpdir_path, .{
.iterable = true,
// we will not delete the temp directory
.can_rename_or_delete = false,
.read_only = true,
}).unwrap()).stdDir();
}
return try bun.openDirAbsolute(tmpdir_path);
}
pub fn entriesAt(this: *RealFS, index: allocators.IndexType, generation: bun.Generation) ?*EntriesOption {
var existing = this.entries.atIndex(index) orelse return null;
if (existing.* == .entries) {
if (existing.entries.generation < generation) {
var handle = bun.openDirForIteration(FD.cwd(), existing.entries.dir).unwrap() catch |err| {
existing.entries.data.clearAndFree(bun.default_allocator);
return this.readDirectoryError(existing.entries.dir, err) catch unreachable;
};
defer handle.close();
const new_entry = this.readdir(
false,
&existing.entries.data,
existing.entries.dir,
generation,
handle.stdDir(),
void,
void{},
) catch |err| {
existing.entries.data.clearAndFree(bun.default_allocator);
return this.readDirectoryError(existing.entries.dir, err) catch unreachable;
};
existing.entries.data.clearAndFree(bun.default_allocator);
existing.entries.* = new_entry;
}
}
return existing;
}
pub fn getDefaultTempDir() string {
return bun.getenvZ("BUN_TMPDIR") orelse bun.getenvZ("TMPDIR") orelse platformTempDir();
}
pub fn setTempdir(path: ?string) void {
tmpdir_path = path orelse getDefaultTempDir();
tmpdir_path_set = true;
}
pub const TmpfilePosix = struct {
fd: bun.FileDescriptor = bun.invalid_fd,
dir_fd: bun.FileDescriptor = bun.invalid_fd,
pub inline fn dir(this: *TmpfilePosix) std.fs.Dir {
return this.dir_fd.stdDir();
}
pub inline fn file(this: *TmpfilePosix) std.fs.File {
return this.fd.stdFile();
}
pub fn close(this: *TmpfilePosix) void {
if (this.fd.isValid()) this.fd.close();
}
pub fn create(this: *TmpfilePosix, _: *RealFS, name: [:0]const u8) !void {
// We originally used a temporary directory, but it caused EXDEV.
const dir_fd = bun.FD.cwd();
this.dir_fd = dir_fd;
const flags = bun.O.CREAT | bun.O.RDWR | bun.O.CLOEXEC;
this.fd = try bun.sys.openat(dir_fd, name, flags, std.posix.S.IRWXU).unwrap();
}
pub fn promoteToCWD(this: *TmpfilePosix, from_name: [*:0]const u8, name: [*:0]const u8) !void {
bun.assert(this.fd != bun.invalid_fd);
bun.assert(this.dir_fd != bun.invalid_fd);
try bun.sys.moveFileZWithHandle(this.fd, this.dir_fd, bun.sliceTo(from_name, 0), bun.FD.cwd(), bun.sliceTo(name, 0));
this.close();
}
pub fn closeAndDelete(this: *TmpfilePosix, name: [*:0]const u8) void {
this.close();
if (comptime !Environment.isLinux) {
if (this.dir_fd == bun.invalid_fd) return;
this.dir().deleteFileZ(name) catch {};
}
}
};
pub const TmpfileWindows = struct {
fd: bun.FileDescriptor = bun.invalid_fd,
existing_path: []const u8 = "",
pub inline fn dir(_: *TmpfileWindows) std.fs.Dir {
return Fs.FileSystem.instance.tmpdir();
}
pub inline fn file(this: *TmpfileWindows) std.fs.File {
return this.fd.stdFile();
}
pub fn close(this: *TmpfileWindows) void {
if (this.fd.isValid()) this.fd.close();
}
pub fn create(this: *TmpfileWindows, rfs: *RealFS, name: [:0]const u8) !void {
const tmp_dir = try rfs.openTmpDir();
const flags = bun.O.CREAT | bun.O.WRONLY | bun.O.CLOEXEC;
this.fd = try bun.sys.openat(.fromStdDir(tmp_dir), name, flags, 0).unwrap();
var buf: bun.PathBuffer = undefined;
const existing_path = try bun.getFdPath(this.fd, &buf);
this.existing_path = try bun.default_allocator.dupe(u8, existing_path);
}
pub fn promoteToCWD(this: *TmpfileWindows, from_name: [*:0]const u8, name: [:0]const u8) !void {
_ = from_name;
var existing_buf: bun.WPathBuffer = undefined;
var new_buf: bun.WPathBuffer = undefined;
this.close();
const existing = bun.strings.toExtendedPathNormalized(&new_buf, this.existing_path);
const new = if (std.fs.path.isAbsoluteWindows(name))
bun.strings.toExtendedPathNormalized(&existing_buf, name)
else
bun.strings.toWPathNormalized(&existing_buf, name);
if (comptime Environment.allow_assert) {
debug("moveFileExW({s}, {s})", .{ bun.fmt.utf16(existing), bun.fmt.utf16(new) });
}
if (bun.windows.kernel32.MoveFileExW(existing.ptr, new.ptr, bun.windows.MOVEFILE_COPY_ALLOWED | bun.windows.MOVEFILE_REPLACE_EXISTING | bun.windows.MOVEFILE_WRITE_THROUGH) == bun.windows.FALSE) {
try bun.windows.Win32Error.get().unwrap();
}
}
pub fn closeAndDelete(this: *TmpfileWindows, name: [*:0]const u8) void {
_ = name;
this.close();
}
};
pub fn needToCloseFiles(rfs: *const RealFS) bool {
if (!FeatureFlags.store_file_descriptors) {
return true;
}
if (Environment.isWindows) {
// 'false' is okay here because windows gives you a seemingly unlimited number of open
// file handles, while posix has a lower limit.
//
// This limit does not extend to the C-Runtime which is only 512 to 8196 or so,
// but we know that all resolver-related handles are not C-Runtime handles because
// `setMaxFd` on Windows (besides being a no-op) only takes in `HANDLE`.
//
// Handles are automatically closed when the process exits as stated here:
// https://learn.microsoft.com/en-us/windows/win32/procthread/terminating-a-process
// But in a crazy experiment to find the upper-bound of the number of open handles,
// I found that opening upwards of 500k to a million handles in a single process
// would cause the process to hang while closing. This might just be Windows slowly
// closing the handles, not sure. This is likely not something to worry about.
//
// If it is decided that not closing files ever is a bad idea. This should be
// replaced with some form of intelligent count of how many files we opened.
// On POSIX we can get away with measuring how high `fd` gets because it typically
// assigns these descriptors in ascending order (1 2 3 ...). Windows does not
// guarantee this.
return false;
}
// If we're not near the max amount of open files, don't worry about it.
return !(rfs.file_limit > 254 and rfs.file_limit > (FileSystem.max_fd + 1) * 2);
}
/// Returns `true` if an entry was removed
pub fn bustEntriesCache(rfs: *RealFS, file_path: string) bool {
return rfs.entries.remove(file_path);
}
pub const Limit = struct {
pub var handles: usize = 0;
pub var handles_before = std.mem.zeroes(if (Environment.isPosix) std.posix.rlimit else struct {});
};
// Always try to max out how many files we can keep open
pub fn adjustUlimit() !usize {
if (comptime !Environment.isPosix) {
return std.math.maxInt(usize);
}
var file_limit: usize = 0;
blk: {
const resource = std.posix.rlimit_resource.NOFILE;
const limit = try std.posix.getrlimit(resource);
Limit.handles_before = limit;
file_limit = limit.max;
Limit.handles = file_limit;
const max_to_use: @TypeOf(limit.max) = if (Environment.isMusl)
// musl has extremely low defaults here, so we really want
// to enable this on musl or tests will start failing.
@max(limit.max, 163840)
else
// apparently, requesting too high of a number can cause other processes to not start.
// https://discord.com/channels/876711213126520882/1316342194176790609/1318175562367242271
// https://github.com/postgres/postgres/blob/fee2b3ea2ecd0da0c88832b37ac0d9f6b3bfb9a9/src/backend/storage/file/fd.c#L1072
limit.max;
if (limit.cur < max_to_use) {
var new_limit = std.mem.zeroes(std.posix.rlimit);
new_limit.cur = max_to_use;
new_limit.max = max_to_use;
std.posix.setrlimit(resource, new_limit) catch break :blk;
file_limit = new_limit.max;
Limit.handles = file_limit;
}
}
return file_limit;
}
var _entries_option_map: *EntriesOption.Map = undefined;
var _entries_option_map_loaded: bool = false;
pub fn init(
cwd: string,
) RealFS {
const file_limit = adjustUlimit() catch unreachable;
if (!_entries_option_map_loaded) {
_entries_option_map = EntriesOption.Map.init(bun.default_allocator);
_entries_option_map_loaded = true;
}
return RealFS{
.entries = _entries_option_map,
.cwd = cwd,
.file_limit = file_limit,
.file_quota = file_limit,
};
}
pub const ModKeyError = error{
Unusable,
};
pub const ModKey = struct {
inode: std.fs.File.INode = 0,
size: u64 = 0,
mtime: i128 = 0,
mode: std.fs.File.Mode = 0,
threadlocal var hash_name_buf: [1024]u8 = undefined;
pub fn hashName(
this: *const ModKey,
basename: string,
) !string {
const hex_int = this.hash();
return try std.fmt.bufPrint(
&hash_name_buf,
"{s}-{any}",
.{
basename,
bun.fmt.hexIntLower(hex_int),
},
);
}
pub fn hash(
this: *const ModKey,
) u64 {
var hash_bytes: [32]u8 = undefined;
// We shouldn't just read the contents of the ModKey into memory
// The hash should be deterministic across computers and operating systems.
// inode is non-deterministic across volumes within the same compuiter
// so if we're not going to do a full content hash, we should use mtime and size.
// even mtime is debatable.
var hash_bytes_remain: []u8 = hash_bytes[0..];
std.mem.writeInt(@TypeOf(this.size), hash_bytes_remain[0..@sizeOf(@TypeOf(this.size))], this.size, .little);
hash_bytes_remain = hash_bytes_remain[@sizeOf(@TypeOf(this.size))..];
std.mem.writeInt(@TypeOf(this.mtime), hash_bytes_remain[0..@sizeOf(@TypeOf(this.mtime))], this.mtime, .little);
hash_bytes_remain = hash_bytes_remain[@sizeOf(@TypeOf(this.mtime))..];
bun.assert(hash_bytes_remain.len == 8);
hash_bytes_remain[0..8].* = @as([8]u8, @bitCast(@as(u64, 0)));
return bun.hash(&hash_bytes);
}
pub fn generate(_: *RealFS, _: string, file: std.fs.File) anyerror!ModKey {
const stat = try file.stat();
const seconds = @divTrunc(stat.mtime, @as(@TypeOf(stat.mtime), std.time.ns_per_s));
// We can't detect changes if the file system zeros out the modification time
if (seconds == 0 and std.time.ns_per_s == 0) {
return error.Unusable;
}
// Don't generate a modification key if the file is too new
const now = std.time.nanoTimestamp();
const now_seconds = @divTrunc(now, std.time.ns_per_s);
if (seconds > seconds or (seconds == now_seconds and stat.mtime > now)) {
return error.Unusable;
}
return ModKey{
.inode = stat.inode,
.size = stat.size,
.mtime = stat.mtime,
.mode = stat.mode,
// .uid = stat.
};
}
pub const SafetyGap = 3;
};
pub fn modKeyWithFile(fs: *RealFS, path: string, file: anytype) anyerror!ModKey {
return try ModKey.generate(fs, path, file);
}
pub fn modKey(fs: *RealFS, path: string) anyerror!ModKey {
var file = try std.fs.openFileAbsolute(path, std.fs.File.OpenFlags{ .mode = .read_only });
defer {
if (fs.needToCloseFiles()) {
file.close();
}
}
return try fs.modKeyWithFile(path, file);
}
pub const EntriesOption = union(Tag) {
entries: *DirEntry,
err: DirEntry.Err,
pub const Tag = enum {
entries,
err,
};
// This custom map implementation:
// - Preallocates a fixed amount of directory name space
// - Doesn't store directory names which don't exist.
pub const Map = allocators.BSSMap(EntriesOption, Preallocate.Counts.dir_entry, false, 256, true);
};
pub fn openDir(_: *RealFS, unsafe_dir_string: string) !std.fs.Dir {
const dirfd = if (Environment.isWindows)
bun.sys.openDirAtWindowsA(bun.invalid_fd, unsafe_dir_string, .{ .iterable = true, .no_follow = false, .read_only = true })
else
bun.sys.openA(
unsafe_dir_string,
bun.O.DIRECTORY,
0,
);
const fd = try dirfd.unwrap();
return fd.stdDir();
}
fn readdir(
fs: *RealFS,
store_fd: bool,
prev_map: ?*DirEntry.EntryMap,
_dir: string,
generation: bun.Generation,
handle: std.fs.Dir,
comptime Iterator: type,
iterator: Iterator,
) !DirEntry {
_ = fs;
var iter = bun.iterateDir(.fromStdDir(handle));
var dir = DirEntry.init(_dir, generation);
const allocator = bun.default_allocator;
errdefer dir.deinit(allocator);
if (store_fd) {
FileSystem.setMaxFd(handle.fd);
dir.fd = .fromStdDir(handle);
}
while (try iter.next().unwrap()) |*_entry| {
debug("readdir entry {s}", .{_entry.name.slice()});
try dir.addEntry(prev_map, _entry, allocator, Iterator, iterator);
}
debug("readdir({d}, {s}) = {d}", .{ handle.fd, _dir, dir.data.count() });
return dir;
}
fn readDirectoryError(fs: *RealFS, dir: string, err: anyerror) OOM!*EntriesOption {
if (comptime FeatureFlags.enable_entry_cache) {
var get_or_put_result = try fs.entries.getOrPut(dir);
const opt = try fs.entries.put(&get_or_put_result, EntriesOption{
.err = DirEntry.Err{ .original_err = err, .canonical_error = err },
});
return opt;
}
temp_entries_option = EntriesOption{
.err = DirEntry.Err{ .original_err = err, .canonical_error = err },
};
return &temp_entries_option;
}
threadlocal var temp_entries_option: EntriesOption = undefined;
pub fn readDirectory(
fs: *RealFS,
_dir: string,
_handle: ?std.fs.Dir,
generation: bun.Generation,
store_fd: bool,
) !*EntriesOption {
return fs.readDirectoryWithIterator(_dir, _handle, generation, store_fd, void, {});
}
// One of the learnings here
//
// Closing file descriptors yields significant performance benefits on Linux
//
// It was literally a 300% performance improvement to bundling.
// https://twitter.com/jarredsumner/status/1655787337027309568
// https://twitter.com/jarredsumner/status/1655714084569120770
// https://twitter.com/jarredsumner/status/1655464485245845506
/// Caller borrows the returned EntriesOption. When `FeatureFlags.enable_entry_cache` is `false`,
/// it is not safe to store this pointer past the current function call.
pub fn readDirectoryWithIterator(
fs: *RealFS,
dir_maybe_trail_slash: string,
maybe_handle: ?std.fs.Dir,
generation: bun.Generation,
store_fd: bool,
comptime Iterator: type,
iterator: Iterator,
) !*EntriesOption {
var dir = bun.strings.withoutTrailingSlashWindowsPath(dir_maybe_trail_slash);
bun.resolver.Resolver.assertValidCacheKey(dir);
var cache_result: ?allocators.Result = null;
if (comptime FeatureFlags.enable_entry_cache) {
fs.entries_mutex.lock();
}
defer {
if (comptime FeatureFlags.enable_entry_cache) {
fs.entries_mutex.unlock();
}
}
var in_place: ?*DirEntry = null;
if (comptime FeatureFlags.enable_entry_cache) {
cache_result = try fs.entries.getOrPut(dir);
if (cache_result.?.hasCheckedIfExists()) {
if (fs.entries.atIndex(cache_result.?.index)) |cached_result| {
if (cached_result.* != .entries or (cached_result.* == .entries and cached_result.entries.generation >= generation)) {
return cached_result;
}
in_place = cached_result.entries;
}
}
}
var handle = maybe_handle orelse try fs.openDir(dir);
defer {
if (maybe_handle == null and (!store_fd or fs.needToCloseFiles())) {
handle.close();
}
}
// if we get this far, it's a real directory, so we can just store the dir name.
if (maybe_handle == null) {
dir = try if (in_place) |existing|
existing.dir
else
DirnameStore.instance.append(string, dir_maybe_trail_slash);
}
// Cache miss: read the directory entries
var entries = fs.readdir(
store_fd,
if (in_place) |existing| &existing.data else null,
dir,
generation,
handle,
Iterator,
iterator,
) catch |err| {
if (in_place) |existing| existing.data.clearAndFree(bun.default_allocator);
return try fs.readDirectoryError(dir, err);
};
if (comptime FeatureFlags.enable_entry_cache) {
const entries_ptr = in_place orelse bun.handleOom(bun.default_allocator.create(DirEntry));
if (in_place) |original| {
original.data.clearAndFree(bun.default_allocator);
}
if (store_fd and !entries.fd.isValid())
entries.fd = .fromStdDir(handle);
entries_ptr.* = entries;
const result = EntriesOption{
.entries = entries_ptr,
};
const out = try fs.entries.put(&cache_result.?, result);
return out;
}
temp_entries_option = EntriesOption{ .entries = entries };
return &temp_entries_option;
}
fn readFileError(_: *RealFS, _: string, _: anyerror) void {}
pub fn readFileWithHandle(
fs: *RealFS,
path: string,
_size: ?usize,
file: std.fs.File,
comptime use_shared_buffer: bool,
shared_buffer: *MutableString,
comptime stream: bool,
) !PathContentsPair {
return readFileWithHandleAndAllocator(
fs,
bun.default_allocator,
path,
_size,
file,
use_shared_buffer,
shared_buffer,
stream,
);
}
pub fn readFileWithHandleAndAllocator(
fs: *RealFS,
allocator: std.mem.Allocator,
path: string,
size_hint: ?usize,
std_file: std.fs.File,
comptime use_shared_buffer: bool,
shared_buffer: *MutableString,
comptime stream: bool,
) !PathContentsPair {
FileSystem.setMaxFd(std_file.handle);
const file = bun.sys.File.from(std_file);
var file_contents: []u8 = "";
// When we're serving a JavaScript-like file over HTTP, we do not want to cache the contents in memory
// This imposes a performance hit because not reading from disk is faster than reading from disk
// Part of that hit is allocating a temporary buffer to store the file contents in
// As a mitigation, we can just keep one buffer forever and re-use it for the parsed files
if (use_shared_buffer) {
shared_buffer.reset();
// Skip the extra file.stat() call when possible
var size = size_hint orelse (file.getEndPos() catch |err| {
fs.readFileError(path, err);
return err;
});
debug("stat({d}) = {d}", .{ file.handle, size });
// Skip the pread call for empty files
// Otherwise will get out of bounds errors
// plus it's an unnecessary syscall
if (size == 0) {
if (comptime use_shared_buffer) {
shared_buffer.reset();
return PathContentsPair{ .path = Path.init(path), .contents = shared_buffer.list.items };
} else {
return PathContentsPair{ .path = Path.init(path), .contents = "" };
}
}
var bytes_read: u64 = 0;
try shared_buffer.growBy(size + 1);
shared_buffer.list.expandToCapacity();
// if you press save on a large file we might not read all the
// bytes in the first few pread() calls. we only handle this on
// stream because we assume that this only realistically happens
// during HMR
while (true) {
// We use pread to ensure if the file handle was open, it doesn't seek from the last position
const read_count = file.readAll(shared_buffer.list.items[bytes_read..]) catch |err| {
fs.readFileError(path, err);
return err;
};
shared_buffer.list.items = shared_buffer.list.items[0 .. read_count + bytes_read];
file_contents = shared_buffer.list.items;
debug("read({d}, {d}) = {d}", .{ file.handle, size, read_count });
if (comptime stream) {
// check again that stat() didn't change the file size
// another reason to only do this when stream
const new_size = file.getEndPos() catch |err| {
fs.readFileError(path, err);
return err;
};
bytes_read += read_count;
// don't infinite loop is we're still not reading more
if (read_count == 0) break;
if (bytes_read < new_size) {
try shared_buffer.growBy(new_size - size);
shared_buffer.list.expandToCapacity();
size = new_size;
continue;
}
}
break;
}
if (shared_buffer.list.capacity > file_contents.len) {
file_contents.ptr[file_contents.len] = 0;
}
if (strings.BOM.detect(file_contents)) |bom| {
debug("Convert {s} BOM", .{@tagName(bom)});
file_contents = try bom.removeAndConvertToUTF8WithoutDealloc(allocator, &shared_buffer.list);
}
} else {
var initial_buf: [16384]u8 = undefined;
// Optimization: don't call stat() unless the file is big enough
// that we need to dynamically allocate memory to read it.
const initial_read = if (size_hint == null) brk: {
const buf: []u8 = &initial_buf;
const read_count = file.readAll(buf).unwrap() catch |err| {
fs.readFileError(path, err);
return err;
};
if (read_count + 1 < buf.len) {
const allocation = try allocator.dupeZ(u8, buf[0..read_count]);
file_contents = allocation[0..read_count];
if (strings.BOM.detect(file_contents)) |bom| {
debug("Convert {s} BOM", .{@tagName(bom)});
file_contents = try bom.removeAndConvertToUTF8AndFree(allocator, file_contents);
}
return PathContentsPair{ .path = Path.init(path), .contents = file_contents };
}
break :brk buf[0..read_count];
} else initial_buf[0..0];
// Skip the extra file.stat() call when possible
const size = size_hint orelse (file.getEndPos().unwrap() catch |err| {
fs.readFileError(path, err);
return err;
});
debug("stat({}) = {d}", .{ file.handle, size });
var buf = try allocator.alloc(u8, size + 1);
@memcpy(buf[0..initial_read.len], initial_read);
if (size == 0) {
return PathContentsPair{ .path = Path.init(path), .contents = "" };
}
// stick a zero at the end
buf[size] = 0;
const read_count = file.readAll(buf[initial_read.len..]).unwrap() catch |err| {
fs.readFileError(path, err);
return err;
};
file_contents = buf[0 .. read_count + initial_read.len];
debug("read({}, {d}) = {d}", .{ file.handle, size, read_count });
if (strings.BOM.detect(file_contents)) |bom| {
debug("Convert {s} BOM", .{@tagName(bom)});
file_contents = try bom.removeAndConvertToUTF8AndFree(allocator, file_contents);
}
}
return PathContentsPair{ .path = Path.init(path), .contents = file_contents };
}
pub fn kindFromAbsolute(
fs: *RealFS,
absolute_path: [:0]const u8,
existing_fd: StoredFileDescriptorType,
store_fd: bool,
) !Entry.Cache {
var outpath: bun.PathBuffer = undefined;
const stat = try bun.sys.lstat_absolute(absolute_path);
const is_symlink = stat.kind == std.fs.File.Kind.SymLink;
var _kind = stat.kind;
var cache = Entry.Cache{
.kind = Entry.Kind.file,
.symlink = PathString.empty,
};
var symlink: []const u8 = "";
if (is_symlink) {
var file = try if (existing_fd != 0)
std.fs.File{ .handle = existing_fd }
else if (store_fd)
std.fs.openFileAbsoluteZ(absolute_path, .{ .mode = .read_only })
else
bun.openFileForPath(absolute_path);
setMaxFd(file.handle);
defer {
if ((!store_fd or fs.needToCloseFiles()) and existing_fd == 0) {
file.close();
} else if (comptime FeatureFlags.store_file_descriptors) {
cache.fd = file.handle;
}
}
const _stat = try file.stat();
symlink = try bun.getFdPath(file.handle, &outpath);
_kind = _stat.kind;
}
bun.assert(_kind != .SymLink);
if (_kind == .Directory) {
cache.kind = .dir;
} else {
cache.kind = .file;
}
if (symlink.len > 0) {
cache.symlink = PathString.init(try FilenameStore.instance.append([]const u8, symlink));
}
return cache;
}
pub fn kind(
fs: *RealFS,
_dir: string,
base: string,
existing_fd: StoredFileDescriptorType,
store_fd: bool,
) !Entry.Cache {
var cache = Entry.Cache{
.kind = Entry.Kind.file,
.symlink = PathString.empty,
};
const dir = _dir;
var combo = [2]string{ dir, base };
var outpath: bun.PathBuffer = undefined;
const entry_path = path_handler.joinAbsStringBuf(fs.cwd, &outpath, &combo, .auto);
outpath[entry_path.len + 1] = 0;
outpath[entry_path.len] = 0;
var absolute_path_c: [:0]const u8 = outpath[0..entry_path.len :0];
if (comptime bun.Environment.isWindows) {
var file = bun.sys.getFileAttributes(absolute_path_c) orelse return error.FileNotFound;
var depth: usize = 0;
const buf2: *bun.PathBuffer = bun.path_buffer_pool.get();
defer bun.path_buffer_pool.put(buf2);
const buf3: *bun.PathBuffer = bun.path_buffer_pool.get();
defer bun.path_buffer_pool.put(buf3);
var current_buf: *bun.PathBuffer = buf2;
var other_buf: *bun.PathBuffer = &outpath;
var joining_buf: *bun.PathBuffer = buf3;
while (file.is_reparse_point) : (depth += 1) {
var read: [:0]const u8 = try bun.sys.readlink(absolute_path_c, current_buf).unwrap();
if (std.fs.path.isAbsolute(read)) {
std.mem.swap(*bun.PathBuffer, &current_buf, &other_buf);
} else {
read = bun.path.joinAbsStringBufZ(std.fs.path.dirname(absolute_path_c) orelse absolute_path_c, joining_buf, &.{read}, .windows);
std.mem.swap(*bun.PathBuffer, &joining_buf, &other_buf);
}
file = bun.sys.getFileAttributes(read) orelse return error.FileNotFound;
absolute_path_c = read;
if (depth > 20) {
return error.TooManySymlinks;
}
}
if (depth > 0) {
cache.symlink = PathString.init(try FilenameStore.instance.append([]const u8, absolute_path_c));
}
if (file.is_directory) {
cache.kind = .dir;
} else {
cache.kind = .file;
}
return cache;
}
const stat = try bun.sys.lstat_absolute(absolute_path_c);
const is_symlink = stat.kind == std.fs.File.Kind.sym_link;
var file_kind = stat.kind;
var symlink: []const u8 = "";
if (is_symlink) {
var file: bun.FD = if (existing_fd.unwrapValid()) |valid|
valid
else if (store_fd)
.fromStdFile(try std.fs.openFileAbsoluteZ(absolute_path_c, .{ .mode = .read_only }))
else
.fromStdFile(try bun.openFileForPath(absolute_path_c));
setMaxFd(file.native());
defer {
if ((!store_fd or fs.needToCloseFiles()) and !existing_fd.isValid()) {
file.close();
} else if (comptime FeatureFlags.store_file_descriptors) {
cache.fd = file;
}
}
const file_stat = try file.stdFile().stat();
symlink = try file.getFdPath(&outpath);
file_kind = file_stat.kind;
}
bun.assert(file_kind != .sym_link);
if (file_kind == .directory) {
cache.kind = .dir;
} else {
cache.kind = .file;
}
if (symlink.len > 0) {
cache.symlink = PathString.init(try FilenameStore.instance.append([]const u8, symlink));
}
return cache;
}
// // Stores the file entries for directories we've listed before
// entries_mutex: std.Mutex
// entries map[string]entriesOrErr
// // If true, do not use the "entries" cache
// doNotCacheEntries bool
};
pub const Implementation = RealFS;
// pub const Implementation = switch (build_target) {
// .wasi, .native => RealFS,
// .wasm => WasmFS,
// };
};
pub const PathContentsPair = struct { path: Path, contents: string };
pub const NodeJSPathName = struct {
base: string,
dir: string,
/// includes the leading .
ext: string,
filename: string,
pub fn init(_path: string, comptime isWindows: bool) NodeJSPathName {
const platform: path_handler.Platform = if (isWindows) .windows else .posix;
const getLastSep = comptime platform.getLastSeparatorFunc();
var path = _path;
var base = path;
// ext must be empty if not detected
var ext: string = "";
var dir = path;
var is_absolute = true;
var _i = getLastSep(path);
var first = true;
while (_i) |i| {
// Stop if we found a non-trailing slash
if (i + 1 != path.len and path.len >= i + 1) {
base = path[i + 1 ..];
dir = path[0..i];
is_absolute = false;
break;
}
// If the path starts with a slash and it's the only slash, it's absolute
if (i == 0 and first) {
base = path[1..];
dir = &([_]u8{});
break;
}
first = false;
// Ignore trailing slashes
path = path[0..i];
_i = getLastSep(path);
}
// clean trailing slashs
if (base.len > 1 and platform.isSeparator(base[base.len - 1])) {
base = base[0 .. base.len - 1];
}
// filename is base without extension
var filename = base;
// if only one character ext = "" even if filename it's "."
if (filename.len > 1) {
// Strip off the extension
if (strings.lastIndexOfChar(filename, '.')) |dot| {
if (dot > 0) {
filename = filename[0..dot];
ext = base[dot..];
}
}
}
if (is_absolute) {
dir = &([_]u8{});
}
return NodeJSPathName{
.dir = dir,
.base = base,
.ext = ext,
.filename = filename,
};
}
};
pub const PathName = struct {
base: string,
dir: string,
/// includes the leading .
/// extensionless files report ""
ext: string,
filename: string,
pub fn extWithoutLeadingDot(self: *const PathName) string {
return if (self.ext.len > 0 and self.ext[0] == '.') self.ext[1..] else self.ext;
}
pub fn nonUniqueNameStringBase(self: *const PathName) string {
// /bar/foo/index.js -> foo
if (self.dir.len > 0 and strings.eqlComptime(self.base, "index")) {
// "/index" -> "index"
return Fs.PathName.init(self.dir).base;
}
if (comptime Environment.allow_assert) {
bun.assert(!strings.includes(self.base, "/"));
}
// /bar/foo.js -> foo
return self.base;
}
pub fn dirOrDot(this: *const PathName) string {
if (this.dir.len == 0) {
return ".";
}
return this.dir;
}
pub fn fmtIdentifier(self: *const PathName) bun.fmt.FormatValidIdentifier {
return bun.fmt.fmtIdentifier(self.nonUniqueNameStringBase());
}
// For readability, the names of certain automatically-generated symbols are
// derived from the file name. For example, instead of the CommonJS wrapper for
// a file being called something like "require273" it can be called something
// like "require_react" instead. This function generates the part of these
// identifiers that's specific to the file path. It can take both an absolute
// path (OS-specific) and a path in the source code (OS-independent).
//
// Note that these generated names do not at all relate to the correctness of
// the code as far as avoiding symbol name collisions. These names still go
// through the renaming logic that all other symbols go through to avoid name
// collisions.
pub fn nonUniqueNameString(self: *const PathName, allocator: std.mem.Allocator) !string {
return MutableString.ensureValidIdentifier(self.nonUniqueNameStringBase(), allocator);
}
pub inline fn dirWithTrailingSlash(this: *const PathName) string {
// The three strings basically always point to the same underlying ptr
// so if dir does not have a trailing slash, but is spaced one apart from the basename
// we can assume there is a trailing slash there
// so we extend the original slice's length by one
return if (this.dir.len == 0) "./" else this.dir.ptr[0 .. this.dir.len + @as(
usize,
@intCast(@intFromBool(
!bun.path.isSepAny(this.dir[this.dir.len - 1]) and (@intFromPtr(this.dir.ptr) + this.dir.len + 1) == @intFromPtr(this.base.ptr),
)),
)];
}
pub fn init(_path: string) PathName {
if (comptime Environment.isWindows and Environment.isDebug) {
// This path is likely incorrect. I think it may be *possible*
// but it is almost entirely certainly a bug.
bun.assert(!strings.startsWith(_path, "/:/"));
bun.assert(!strings.startsWith(_path, "\\:\\"));
}
var path = _path;
var base = path;
var ext: []const u8 = undefined;
var dir = path;
var is_absolute = true;
const has_disk_designator = path.len > 2 and path[1] == ':' and switch (path[0]) {
'a'...'z', 'A'...'Z' => true,
else => false,
} and bun.path.isSepAny(path[2]);
if (has_disk_designator) {
path = path[2..];
}
while (bun.path.lastIndexOfSep(path)) |i| {
// Stop if we found a non-trailing slash
if (i + 1 != path.len and path.len > i + 1) {
base = path[i + 1 ..];
dir = path[0..i];
is_absolute = false;
break;
}
// Ignore trailing slashes
path = path[0..i];
}
// Strip off the extension
if (strings.lastIndexOfChar(base, '.')) |dot| {
ext = base[dot..];
base = base[0..dot];
} else {
ext = "";
}
if (is_absolute) {
dir = &([_]u8{});
}
if (base.len > 1 and bun.path.isSepAny(base[base.len - 1])) {
base = base[0 .. base.len - 1];
}
if (!is_absolute and has_disk_designator) {
dir = _path[0 .. dir.len + 2];
}
return .{
.dir = dir,
.base = base,
.ext = ext,
.filename = if (dir.len > 0) _path[dir.len + 1 ..] else _path,
};
}
};
threadlocal var normalize_buf: [1024]u8 = undefined;
threadlocal var join_buf: [1024]u8 = undefined;
pub const Path = struct {
/// The display path. In the bundler, this is relative to the current
/// working directory. Since it can be emitted in bundles (and used
/// for content hashes), this should contain forward slashes on Windows.
pretty: string,
/// The location of this resource. For the `file` namespace, this is
/// usually an absolute path with native slashes or an empty string.
text: string,
namespace: string,
// TODO(@paperclover): investigate removing or simplifying this property (it's 64 bytes)
name: PathName,
is_disabled: bool = false,
is_symlink: bool = false,
const ns_blob = "blob";
const ns_bun = "bun";
const ns_dataurl = "dataurl";
const ns_file = "file";
const ns_macro = "macro";
pub fn isFile(this: *const Path) bool {
return this.namespace.len == 0 or strings.eqlComptime(this.namespace, "file");
}
pub fn hashKey(this: *const Path) u64 {
if (this.isFile()) {
return bun.hash(this.text);
}
var hasher = std.hash.Wyhash.init(0);
hasher.update(this.namespace);
hasher.update("::::::::");
hasher.update(this.text);
return hasher.final();
}
/// This hash is used by the hot-module-reloading client in order to
/// identify modules. Since that code is JavaScript, the hash must remain in
/// range [-MAX_SAFE_INTEGER, MAX_SAFE_INTEGER] or else information is lost
/// due to floating-point precision.
pub fn hashForKit(path: Path) u52 {
return @truncate(path.hashKey());
}
pub fn packageName(this: *const Path) ?string {
var name_to_use = this.pretty;
if (strings.lastIndexOf(this.text, std.fs.path.sep_str ++ "node_modules" ++ std.fs.path.sep_str)) |node_modules| {
name_to_use = this.text[node_modules + 14 ..];
}
const pkgname = bun.options.JSX.Pragma.parsePackageName(name_to_use);
if (pkgname.len == 0 or !std.ascii.isAlphanumeric(pkgname[0]))
return null;
return pkgname;
}
pub fn loader(this: *const Path, loaders: *const bun.options.Loader.HashTable) ?bun.options.Loader {
if (this.isDataURL()) {
return bun.options.Loader.dataurl;
}
const ext = this.name.ext;
const result = loaders.get(ext) orelse bun.options.Loader.fromString(ext);
if (result == null or result == .json) {
const str = this.name.filename;
if (strings.eqlComptime(str, "package.json") or strings.eqlComptime(str, "bun.lock")) {
return .jsonc;
}
if (strings.hasSuffixComptime(str, ".jsonc")) {
return .jsonc;
}
if (strings.hasPrefixComptime(str, "tsconfig.") or strings.hasPrefixComptime(str, "jsconfig.")) {
if (strings.hasSuffixComptime(str, ".json")) {
return .jsonc;
}
}
}
return result;
}
pub fn isDataURL(this: *const Path) bool {
return strings.eqlComptime(this.namespace, ns_dataurl);
}
pub fn isBun(this: *const Path) bool {
return strings.eqlComptime(this.namespace, ns_bun);
}
pub fn isMacro(this: *const Path) bool {
return strings.eqlComptime(this.namespace, ns_macro);
}
pub const PackageRelative = struct {
path: string,
name: string,
is_parent_package: bool = false,
};
pub inline fn sourceDir(this: *const Path) string {
return this.name.dirWithTrailingSlash();
}
pub inline fn prettyDir(this: *const Path) string {
return this.name.dirWithTrailingSlash();
}
/// The bundler will hash path.pretty, so it needs to be consistent across platforms.
/// This assertion might be a bit too forceful though.
pub fn assertPrettyIsValid(path: *const Path) void {
if (Environment.isWindows and Environment.allow_assert) {
if (bun.strings.indexOfChar(path.pretty, '\\') != null) {
std.debug.panic("Expected pretty file path to have only forward slashes, got '{s}'", .{path.pretty});
}
}
}
pub inline fn assertFilePathIsAbsolute(path: *const Path) void {
if (bun.Environment.ci_assert) {
if (path.isFile()) {
bun.assert(std.fs.path.isAbsolute(path.text));
}
}
}
pub inline fn isPrettyPathPosix(path: *const Path) bool {
if (!Environment.isWindows) return true;
return bun.strings.indexOfChar(path.pretty, '\\') == null;
}
// This duplicates but only when strictly necessary
// This will skip allocating if it's already in FilenameStore or DirnameStore
pub fn dupeAlloc(this: *const Path, allocator: std.mem.Allocator) !Fs.Path {
if (this.text.ptr == this.pretty.ptr and this.text.len == this.pretty.len) {
if (FileSystem.FilenameStore.instance.exists(this.text) or FileSystem.DirnameStore.instance.exists(this.text)) {
return this.*;
}
var new_path = Fs.Path.init(try FileSystem.FilenameStore.instance.append([]const u8, this.text));
new_path.pretty = this.text;
new_path.namespace = this.namespace;
new_path.is_symlink = this.is_symlink;
return new_path;
} else if (this.pretty.len == 0) {
if (FileSystem.FilenameStore.instance.exists(this.text) or FileSystem.DirnameStore.instance.exists(this.text)) {
return this.*;
}
var new_path = Fs.Path.init(try FileSystem.FilenameStore.instance.append([]const u8, this.text));
new_path.pretty = "";
new_path.namespace = this.namespace;
new_path.is_symlink = this.is_symlink;
return new_path;
} else if (allocators.sliceRange(this.pretty, this.text)) |start_len| {
if (FileSystem.FilenameStore.instance.exists(this.text) or FileSystem.DirnameStore.instance.exists(this.text)) {
return this.*;
}
var new_path = Fs.Path.init(try FileSystem.FilenameStore.instance.append([]const u8, this.text));
new_path.pretty = this.text[start_len[0]..][0..start_len[1]];
new_path.namespace = this.namespace;
new_path.is_symlink = this.is_symlink;
return new_path;
} else {
if ((FileSystem.FilenameStore.instance.exists(this.text) or
FileSystem.DirnameStore.instance.exists(this.text)) and
(FileSystem.FilenameStore.instance.exists(this.pretty) or
FileSystem.DirnameStore.instance.exists(this.pretty)))
{
return this.*;
}
if (strings.indexOf(this.text, this.pretty)) |offset| {
var text = try FileSystem.FilenameStore.instance.append([]const u8, this.text);
var new_path = Fs.Path.init(text);
new_path.pretty = text[offset..][0..this.pretty.len];
new_path.namespace = this.namespace;
new_path.is_symlink = this.is_symlink;
return new_path;
} else {
var buf = try allocator.alloc(u8, this.text.len + this.pretty.len + 2);
bun.copy(u8, buf, this.text);
buf.ptr[this.text.len] = 0;
const new_pretty = buf[this.text.len + 1 ..][0..this.pretty.len];
bun.copy(u8, buf[this.text.len + 1 ..], this.pretty);
var new_path = Fs.Path.init(buf[0..this.text.len]);
buf.ptr[buf.len - 1] = 0;
new_path.pretty = new_pretty;
new_path.namespace = this.namespace;
new_path.is_symlink = this.is_symlink;
return new_path;
}
}
}
pub fn dupeAllocFixPretty(this: *const Path, allocator: std.mem.Allocator) !Fs.Path {
if (this.isPrettyPathPosix()) return this.dupeAlloc(allocator);
comptime bun.assert(bun.Environment.isWindows);
var new = this.*;
new.pretty = "";
new = try new.dupeAlloc(allocator);
const pretty = try allocator.dupe(u8, this.pretty);
bun.path.platformToPosixInPlace(u8, pretty);
new.pretty = pretty;
new.assertPrettyIsValid();
return new;
}
pub const empty = Fs.Path.init("");
pub fn setRealpath(this: *Path, to: string) void {
const old_path = this.text;
this.text = to;
this.name = PathName.init(to);
this.pretty = old_path;
this.is_symlink = true;
}
pub fn jsonStringify(self: *const @This(), writer: anytype) !void {
return try writer.write(self.text);
}
pub fn generateKey(p: *Path, allocator: std.mem.Allocator) !string {
return try std.fmt.allocPrint(allocator, "{s}://{s}", .{ p.namespace, p.text });
}
pub fn init(text: string) Path {
return Path{
.pretty = text,
.text = text,
.namespace = "file",
.name = PathName.init(text),
};
}
pub fn initWithPretty(text: string, pretty: string) Path {
return Path{
.pretty = pretty,
.text = text,
.namespace = "file",
.name = PathName.init(text),
};
}
pub fn initWithNamespace(text: string, namespace: string) Path {
return Path{
.pretty = text,
.text = text,
.namespace = namespace,
.name = PathName.init(text),
};
}
pub inline fn initWithNamespaceVirtual(comptime text: string, comptime namespace: string, comptime package: string) Path {
return comptime Path{
.pretty = namespace ++ ":" ++ package,
.is_symlink = true,
.text = text,
.namespace = namespace,
.name = PathName.init(text),
};
}
pub inline fn initForKitBuiltIn(comptime namespace: string, comptime package: string) Path {
return comptime Path{
.pretty = namespace ++ ":" ++ package,
.is_symlink = true,
.text = "_bun/" ++ package,
.namespace = namespace,
.name = PathName.init(package),
};
}
pub fn isNodeModule(this: *const Path) bool {
return strings.lastIndexOf(this.name.dir, std.fs.path.sep_str ++ "node_modules" ++ std.fs.path.sep_str) != null;
}
pub fn isJSXFile(this: *const Path) bool {
return strings.hasSuffixComptime(this.name.filename, ".jsx") or strings.hasSuffixComptime(this.name.filename, ".tsx");
}
pub fn keyForIncrementalGraph(path: *const Path) []const u8 {
return if (path.isFile())
path.text
else
path.pretty;
}
};
// pub fn customRealpath(allocator: std.mem.Allocator, path: string) !string {
// var opened = try std.posix.open(path, if (Environment.isLinux) bun.O.PATH else bun.O.RDONLY, 0);
// defer std.posix.close(opened);
// }
pub const StatHash = @import("./fs/stat_hash.zig");
const string = []const u8;
const stringZ = [:0]const u8;
const path_handler = @import("./resolver/resolve_path.zig");
const std = @import("std");
const bun = @import("bun");
const Environment = bun.Environment;
const FD = bun.FD;
const FeatureFlags = bun.FeatureFlags;
const FileDescriptor = bun.FileDescriptor;
const MAX_PATH_BYTES = bun.MAX_PATH_BYTES;
const MutableString = bun.MutableString;
const Mutex = bun.Mutex;
const OOM = bun.OOM;
const Output = bun.Output;
const PathBuffer = bun.PathBuffer;
const PathString = bun.PathString;
const StoredFileDescriptorType = bun.StoredFileDescriptorType;
const WPathBuffer = bun.WPathBuffer;
const allocators = bun.allocators;
const default_allocator = bun.default_allocator;
const strings = bun.strings;