-
Notifications
You must be signed in to change notification settings - Fork 2.8k
/
StandaloneModuleGraph.zig
1119 lines (954 loc) · 45.4 KB
/
StandaloneModuleGraph.zig
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
//! Originally, we tried using LIEF to inject the module graph into a MachO segment
//! But this incurred a fixed 350ms overhead on every build, which is unacceptable
//! so we give up on codesigning support on macOS for now until we can find a better solution
const bun = @import("root").bun;
const std = @import("std");
const Schema = bun.Schema.Api;
const strings = bun.strings;
const Output = bun.Output;
const Global = bun.Global;
const Environment = bun.Environment;
const Syscall = bun.sys;
const SourceMap = bun.sourcemap;
const StringPointer = bun.StringPointer;
const w = std.os.windows;
pub const StandaloneModuleGraph = struct {
bytes: []const u8 = "",
files: bun.StringArrayHashMap(File),
entry_point_id: u32 = 0,
// We never want to hit the filesystem for these files
// We use the `/$bunfs/` prefix to indicate that it's a virtual path
// It is `/$bunfs/` because:
//
// - `$` makes it unlikely to collide with a real path
// - `/$bunfs/` is 8 characters which is fast to compare for 64-bit CPUs
pub const base_path = switch (Environment.os) {
else => "/$bunfs/",
// Special case for windows because of file URLs being invalid
// if they do not have a drive letter. B drive because 'bun' but
// also because it's more unlikely to collide with a real path.
.windows => "B:\\~BUN\\",
};
pub const base_public_path = targetBasePublicPath(Environment.os, "");
pub const base_public_path_with_default_suffix = targetBasePublicPath(Environment.os, "root/");
pub fn targetBasePublicPath(target: Environment.OperatingSystem, comptime suffix: [:0]const u8) [:0]const u8 {
return switch (target) {
.windows => "B:/~BUN/" ++ suffix,
else => "/$bunfs/" ++ suffix,
};
}
pub fn isBunStandaloneFilePath(str: []const u8) bool {
return bun.strings.hasPrefixComptime(str, base_path) or
(Environment.isWindows and bun.strings.hasPrefixComptime(str, base_public_path));
}
pub fn entryPoint(this: *const StandaloneModuleGraph) *File {
return &this.files.values()[this.entry_point_id];
}
// by normalized file path
pub fn find(this: *const StandaloneModuleGraph, name: []const u8) ?*File {
if (!isBunStandaloneFilePath(base_path)) {
return null;
}
return this.findAssumeStandalonePath(name);
}
pub fn findAssumeStandalonePath(this: *const StandaloneModuleGraph, name: []const u8) ?*File {
if (Environment.isWindows) {
var normalized_buf: bun.PathBuffer = undefined;
const normalized = bun.path.platformToPosixBuf(u8, name, &normalized_buf);
return this.files.getPtr(normalized);
}
return this.files.getPtr(name);
}
pub const CompiledModuleGraphFile = struct {
name: Schema.StringPointer = .{},
contents: Schema.StringPointer = .{},
sourcemap: Schema.StringPointer = .{},
bytecode: Schema.StringPointer = .{},
encoding: Encoding = .latin1,
loader: bun.options.Loader = .file,
module_format: ModuleFormat = .none,
};
pub const Encoding = enum(u8) {
binary = 0,
latin1 = 1,
// Not used yet.
utf8 = 2,
};
pub const ModuleFormat = enum(u8) {
none = 0,
esm = 1,
cjs = 2,
};
pub const File = struct {
name: []const u8 = "",
loader: bun.options.Loader,
contents: [:0]const u8 = "",
sourcemap: LazySourceMap,
cached_blob: ?*bun.JSC.WebCore.Blob = null,
encoding: Encoding = .binary,
wtf_string: bun.String = bun.String.empty,
bytecode: []u8 = "",
module_format: ModuleFormat = .none,
pub fn lessThanByIndex(ctx: []const File, lhs_i: u32, rhs_i: u32) bool {
const lhs = ctx[lhs_i];
const rhs = ctx[rhs_i];
return bun.strings.cmpStringsAsc({}, lhs.name, rhs.name);
}
pub fn toWTFString(this: *File) bun.String {
if (this.wtf_string.isEmpty()) {
switch (this.encoding) {
.binary, .utf8 => {
this.wtf_string = bun.String.createUTF8(this.contents);
},
.latin1 => {
this.wtf_string = bun.String.createStaticExternal(this.contents, true);
},
}
}
// We don't want this to free.
return this.wtf_string.dupeRef();
}
pub fn blob(this: *File, globalObject: *bun.JSC.JSGlobalObject) *bun.JSC.WebCore.Blob {
if (this.cached_blob == null) {
const store = bun.JSC.WebCore.Blob.Store.init(@constCast(this.contents), bun.default_allocator);
// make it never free
store.ref();
const b = bun.JSC.WebCore.Blob.initWithStore(store, globalObject).new();
b.allocator = bun.default_allocator;
if (bun.http.MimeType.byExtensionNoDefault(bun.strings.trimLeadingChar(std.fs.path.extension(this.name), '.'))) |mime| {
store.mime_type = mime;
b.content_type = mime.value;
b.content_type_was_set = true;
b.content_type_allocated = false;
}
// The real name goes here:
store.data.bytes.stored_name = bun.PathString.init(this.name);
// The pretty name goes here:
if (strings.hasPrefixComptime(this.name, base_public_path_with_default_suffix)) {
b.name = bun.String.createUTF8(this.name[base_public_path_with_default_suffix.len..]);
} else if (this.name.len > 0) {
b.name = bun.String.createUTF8(this.name);
}
this.cached_blob = b;
}
return this.cached_blob.?;
}
};
pub const LazySourceMap = union(enum) {
serialized: SerializedSourceMap,
parsed: *SourceMap.ParsedSourceMap,
none,
/// It probably is not possible to run two decoding jobs on the same file
var init_lock: bun.Lock = .{};
pub fn load(this: *LazySourceMap) ?*SourceMap.ParsedSourceMap {
init_lock.lock();
defer init_lock.unlock();
return switch (this.*) {
.none => null,
.parsed => |map| map,
.serialized => |serialized| {
var stored = switch (SourceMap.Mapping.parse(
bun.default_allocator,
serialized.mappingVLQ(),
null,
std.math.maxInt(i32),
std.math.maxInt(i32),
)) {
.success => |x| x,
.fail => {
this.* = .none;
return null;
},
};
const source_files = serialized.sourceFileNames();
const slices = bun.default_allocator.alloc(?[]u8, source_files.len * 2) catch bun.outOfMemory();
const file_names: [][]const u8 = @ptrCast(slices[0..source_files.len]);
const decompressed_contents_slice = slices[source_files.len..][0..source_files.len];
for (file_names, source_files) |*dest, src| {
dest.* = src.slice(serialized.bytes);
}
@memset(decompressed_contents_slice, null);
const data = bun.new(SerializedSourceMap.Loaded, .{
.map = serialized,
.decompressed_files = decompressed_contents_slice,
});
stored.external_source_names = file_names;
stored.underlying_provider = .{ .data = @truncate(@intFromPtr(data)) };
stored.is_standalone_module_graph = true;
const parsed = stored.new(); // allocate this on the heap
parsed.ref(); // never free
this.* = .{ .parsed = parsed };
return parsed;
},
};
}
};
pub const Offsets = extern struct {
byte_count: usize = 0,
modules_ptr: bun.StringPointer = .{},
entry_point_id: u32 = 0,
};
const trailer = "\n---- Bun! ----\n";
pub fn fromBytes(allocator: std.mem.Allocator, raw_bytes: []u8, offsets: Offsets) !StandaloneModuleGraph {
if (raw_bytes.len == 0) return StandaloneModuleGraph{
.files = bun.StringArrayHashMap(File).init(allocator),
};
const modules_list_bytes = sliceTo(raw_bytes, offsets.modules_ptr);
const modules_list: []align(1) const CompiledModuleGraphFile = std.mem.bytesAsSlice(CompiledModuleGraphFile, modules_list_bytes);
if (offsets.entry_point_id > modules_list.len) {
return error.@"Corrupted module graph: entry point ID is greater than module list count";
}
var modules = bun.StringArrayHashMap(File).init(allocator);
try modules.ensureTotalCapacity(modules_list.len);
for (modules_list) |module| {
modules.putAssumeCapacity(
sliceToZ(raw_bytes, module.name),
File{
.name = sliceToZ(raw_bytes, module.name),
.loader = module.loader,
.contents = sliceToZ(raw_bytes, module.contents),
.sourcemap = if (module.sourcemap.length > 0)
.{ .serialized = .{
.bytes = @alignCast(sliceTo(raw_bytes, module.sourcemap)),
} }
else
.none,
.bytecode = if (module.bytecode.length > 0) @constCast(sliceTo(raw_bytes, module.bytecode)) else &.{},
.module_format = module.module_format,
},
);
}
modules.lockPointers(); // make the pointers stable forever
return StandaloneModuleGraph{
.bytes = raw_bytes[0..offsets.byte_count],
.files = modules,
.entry_point_id = offsets.entry_point_id,
};
}
fn sliceTo(bytes: []const u8, ptr: bun.StringPointer) []const u8 {
if (ptr.length == 0) return "";
return bytes[ptr.offset..][0..ptr.length];
}
fn sliceToZ(bytes: []const u8, ptr: bun.StringPointer) [:0]const u8 {
if (ptr.length == 0) return "";
return bytes[ptr.offset..][0..ptr.length :0];
}
pub fn toBytes(allocator: std.mem.Allocator, prefix: []const u8, output_files: []const bun.options.OutputFile, output_format: bun.options.Format) ![]u8 {
var serialize_trace = bun.tracy.traceNamed(@src(), "StandaloneModuleGraph.serialize");
defer serialize_trace.end();
var entry_point_id: ?usize = null;
var string_builder = bun.StringBuilder{};
var module_count: usize = 0;
for (output_files) |output_file| {
string_builder.countZ(output_file.dest_path);
string_builder.countZ(prefix);
if (output_file.value == .buffer) {
if (output_file.output_kind == .sourcemap) {
// This is an over-estimation to ensure that we allocate
// enough memory for the source-map contents. Calculating
// the exact amount is not possible without allocating as it
// involves a JSON parser.
string_builder.cap += output_file.value.buffer.bytes.len * 2;
} else if (output_file.output_kind == .bytecode) {
// Allocate up to 256 byte alignment for bytecode
string_builder.cap += (output_file.value.buffer.bytes.len + 255) / 256 * 256 + 256;
} else {
if (entry_point_id == null) {
if (output_file.output_kind == .@"entry-point") {
entry_point_id = module_count;
}
}
string_builder.countZ(output_file.value.buffer.bytes);
module_count += 1;
}
}
}
if (module_count == 0 or entry_point_id == null) return &[_]u8{};
string_builder.cap += @sizeOf(CompiledModuleGraphFile) * output_files.len;
string_builder.cap += trailer.len;
string_builder.cap += 16;
string_builder.cap += @sizeOf(Offsets);
try string_builder.allocate(allocator);
var modules = try std.ArrayList(CompiledModuleGraphFile).initCapacity(allocator, module_count);
var source_map_header_list = std.ArrayList(u8).init(allocator);
defer source_map_header_list.deinit();
var source_map_string_list = std.ArrayList(u8).init(allocator);
defer source_map_string_list.deinit();
var source_map_arena = bun.ArenaAllocator.init(allocator);
defer source_map_arena.deinit();
for (output_files) |output_file| {
if (!output_file.output_kind.isFileInStandaloneMode()) {
continue;
}
if (output_file.value != .buffer) {
continue;
}
const dest_path = bun.strings.removeLeadingDotSlash(output_file.dest_path);
const bytecode: StringPointer = brk: {
if (output_file.bytecode_index != std.math.maxInt(u32)) {
// Use up to 256 byte alignment for bytecode
// Not aligning it correctly will cause a runtime assertion error, or a segfault.
const bytecode = output_files[output_file.bytecode_index].value.buffer.bytes;
const aligned = std.mem.alignInSlice(string_builder.writable(), 128).?;
@memcpy(aligned[0..bytecode.len], bytecode[0..bytecode.len]);
const unaligned_space = aligned[bytecode.len..];
const offset = @intFromPtr(aligned.ptr) - @intFromPtr(string_builder.ptr.?);
const len = bytecode.len + @min(unaligned_space.len, 128);
string_builder.len += len;
break :brk StringPointer{ .offset = @truncate(offset), .length = @truncate(len) };
} else {
break :brk .{};
}
};
var module = CompiledModuleGraphFile{
.name = string_builder.fmtAppendCountZ("{s}{s}", .{
prefix,
dest_path,
}),
.loader = output_file.loader,
.contents = string_builder.appendCountZ(output_file.value.buffer.bytes),
.encoding = switch (output_file.loader) {
.js, .jsx, .ts, .tsx => .latin1,
else => .binary,
},
.module_format = if (output_file.loader.isJavaScriptLike()) switch (output_format) {
.cjs => .cjs,
.esm => .esm,
else => .none,
} else .none,
.bytecode = bytecode,
};
if (output_file.source_map_index != std.math.maxInt(u32)) {
defer source_map_header_list.clearRetainingCapacity();
defer source_map_string_list.clearRetainingCapacity();
_ = source_map_arena.reset(.retain_capacity);
try serializeJsonSourceMapForStandalone(
&source_map_header_list,
&source_map_string_list,
source_map_arena.allocator(),
output_files[output_file.source_map_index].value.buffer.bytes,
);
module.sourcemap = string_builder.addConcat(&.{
source_map_header_list.items,
source_map_string_list.items,
});
}
modules.appendAssumeCapacity(module);
}
const offsets = Offsets{
.entry_point_id = @as(u32, @truncate(entry_point_id.?)),
.modules_ptr = string_builder.appendCount(std.mem.sliceAsBytes(modules.items)),
.byte_count = string_builder.len,
};
_ = string_builder.append(std.mem.asBytes(&offsets));
_ = string_builder.append(trailer);
const output_bytes = string_builder.ptr.?[0..string_builder.len];
if (comptime Environment.isDebug) {
// An expensive sanity check:
var graph = try fromBytes(allocator, @alignCast(output_bytes), offsets);
defer {
graph.files.unlockPointers();
graph.files.deinit();
}
bun.assert_eql(graph.files.count(), modules.items.len);
}
return output_bytes;
}
const page_size = if (Environment.isLinux and Environment.isAarch64)
// some linux distros do 64 KB pages on aarch64
64 * 1024
else
std.mem.page_size;
pub fn inject(bytes: []const u8, self_exe: [:0]const u8) bun.FileDescriptor {
var buf: bun.PathBuffer = undefined;
var zname: [:0]const u8 = bun.span(bun.fs.FileSystem.instance.tmpname("bun-build", &buf, @as(u64, @bitCast(std.time.milliTimestamp()))) catch |err| {
Output.prettyErrorln("<r><red>error<r><d>:<r> failed to get temporary file name: {s}", .{@errorName(err)});
Global.exit(1);
});
const cleanup = struct {
pub fn toClean(name: [:0]const u8, fd: bun.FileDescriptor) void {
_ = Syscall.close(fd);
_ = Syscall.unlink(name);
}
}.toClean;
const cloned_executable_fd: bun.FileDescriptor = brk: {
if (comptime Environment.isWindows) {
// copy self and then open it for writing
var in_buf: bun.WPathBuffer = undefined;
strings.copyU8IntoU16(&in_buf, self_exe);
in_buf[self_exe.len] = 0;
const in = in_buf[0..self_exe.len :0];
var out_buf: bun.WPathBuffer = undefined;
strings.copyU8IntoU16(&out_buf, zname);
out_buf[zname.len] = 0;
const out = out_buf[0..zname.len :0];
bun.copyFile(in, out).unwrap() catch |err| {
Output.prettyErrorln("<r><red>error<r><d>:<r> failed to copy bun executable into temporary file: {s}", .{@errorName(err)});
Global.exit(1);
};
const file = bun.sys.openFileAtWindows(
bun.invalid_fd,
out,
// access_mask
w.SYNCHRONIZE | w.GENERIC_WRITE | w.DELETE,
// create disposition
w.FILE_OPEN,
// create options
w.FILE_SYNCHRONOUS_IO_NONALERT | w.FILE_OPEN_REPARSE_POINT,
).unwrap() catch |e| {
Output.prettyErrorln("<r><red>error<r><d>:<r> failed to open temporary file to copy bun into\n{}", .{e});
Global.exit(1);
};
break :brk file;
}
if (comptime Environment.isMac) {
// if we're on a mac, use clonefile() if we can
// failure is okay, clonefile is just a fast path.
if (Syscall.clonefile(self_exe, zname) == .result) {
switch (Syscall.open(zname, bun.O.RDWR | bun.O.CLOEXEC, 0)) {
.result => |res| break :brk res,
.err => {},
}
}
}
// otherwise, just copy the file
const fd = brk2: {
var tried_changing_abs_dir = false;
for (0..3) |retry| {
switch (Syscall.open(zname, bun.O.CLOEXEC | bun.O.RDWR | bun.O.CREAT, 0)) {
.result => |res| break :brk2 res,
.err => |err| {
if (retry < 2) {
// they may not have write access to the present working directory
//
// but we want to default to it since it's the
// least likely to need to be copied due to
// renameat() across filesystems
//
// so in the event of a failure, we try to
// we retry using the tmp dir
//
// but we only do that once because otherwise it's just silly
if (!tried_changing_abs_dir) {
tried_changing_abs_dir = true;
const zname_z = bun.strings.concat(bun.default_allocator, &.{
bun.fs.FileSystem.instance.fs.tmpdirPath(),
std.fs.path.sep_str,
zname,
&.{0},
}) catch bun.outOfMemory();
zname = zname_z[0..zname_z.len -| 1 :0];
continue;
}
switch (err.getErrno()) {
// try again
.PERM, .AGAIN, .BUSY => continue,
else => break,
}
Output.prettyErrorln("<r><red>error<r><d>:<r> failed to open temporary file to copy bun into\n{}", .{err});
Global.exit(1);
}
},
}
}
unreachable;
};
const self_fd = brk2: {
for (0..3) |retry| {
switch (Syscall.open(self_exe, bun.O.CLOEXEC | bun.O.RDONLY, 0)) {
.result => |res| break :brk2 res,
.err => |err| {
if (retry < 2) {
switch (err.getErrno()) {
// try again
.PERM, .AGAIN, .BUSY => continue,
else => {},
}
}
Output.prettyErrorln("<r><red>error<r><d>:<r> failed to open bun executable to copy from as read-only\n{}", .{err});
cleanup(zname, fd);
Global.exit(1);
},
}
}
unreachable;
};
defer _ = Syscall.close(self_fd);
bun.copyFile(self_fd.cast(), fd.cast()).unwrap() catch |err| {
Output.prettyErrorln("<r><red>error<r><d>:<r> failed to copy bun executable into temporary file: {s}", .{@errorName(err)});
cleanup(zname, fd);
Global.exit(1);
};
break :brk fd;
};
var total_byte_count: usize = undefined;
if (Environment.isWindows) {
total_byte_count = bytes.len + 8 + (Syscall.setFileOffsetToEndWindows(cloned_executable_fd).unwrap() catch |err| {
Output.prettyErrorln("<r><red>error<r><d>:<r> failed to seek to end of temporary file\n{}", .{err});
cleanup(zname, cloned_executable_fd);
Global.exit(1);
});
} else {
const seek_position = @as(u64, @intCast(brk: {
const fstat = switch (Syscall.fstat(cloned_executable_fd)) {
.result => |res| res,
.err => |err| {
Output.prettyErrorln("{}", .{err});
cleanup(zname, cloned_executable_fd);
Global.exit(1);
},
};
break :brk @max(fstat.size, 0);
}));
total_byte_count = seek_position + bytes.len + 8;
// From https://man7.org/linux/man-pages/man2/lseek.2.html
//
// lseek() allows the file offset to be set beyond the end of the
// file (but this does not change the size of the file). If data is
// later written at this point, subsequent reads of the data in the
// gap (a "hole") return null bytes ('\0') until data is actually
// written into the gap.
//
switch (Syscall.setFileOffset(cloned_executable_fd, seek_position)) {
.err => |err| {
Output.prettyErrorln(
"{}\nwhile seeking to end of temporary file (pos: {d})",
.{
err,
seek_position,
},
);
cleanup(zname, cloned_executable_fd);
Global.exit(1);
},
else => {},
}
}
var remain = bytes;
while (remain.len > 0) {
switch (Syscall.write(cloned_executable_fd, bytes)) {
.result => |written| remain = remain[written..],
.err => |err| {
Output.prettyErrorln("<r><red>error<r><d>:<r> failed to write to temporary file\n{}", .{err});
cleanup(zname, cloned_executable_fd);
Global.exit(1);
},
}
}
// the final 8 bytes in the file are the length of the module graph with padding, excluding the trailer and offsets
_ = Syscall.write(cloned_executable_fd, std.mem.asBytes(&total_byte_count));
if (comptime !Environment.isWindows) {
_ = bun.C.fchmod(cloned_executable_fd.int(), 0o777);
}
return cloned_executable_fd;
}
pub const CompileTarget = @import("./compile_target.zig");
pub fn download(allocator: std.mem.Allocator, target: *const CompileTarget, env: *bun.DotEnv.Loader) ![:0]const u8 {
var exe_path_buf: bun.PathBuffer = undefined;
var version_str_buf: [1024]u8 = undefined;
const version_str = try std.fmt.bufPrintZ(&version_str_buf, "{}", .{target});
var needs_download: bool = true;
const dest_z = target.exePath(&exe_path_buf, version_str, env, &needs_download);
if (needs_download) {
try target.downloadToPath(env, allocator, dest_z);
}
return try allocator.dupeZ(u8, dest_z);
}
pub fn toExecutable(
target: *const CompileTarget,
allocator: std.mem.Allocator,
output_files: []const bun.options.OutputFile,
root_dir: std.fs.Dir,
module_prefix: []const u8,
outfile: []const u8,
env: *bun.DotEnv.Loader,
output_format: bun.options.Format,
) !void {
const bytes = try toBytes(allocator, module_prefix, output_files, output_format);
if (bytes.len == 0) return;
const fd = inject(
bytes,
if (target.isDefault())
bun.selfExePath() catch |err| {
Output.err(err, "failed to get self executable path", .{});
Global.exit(1);
}
else
download(allocator, target, env) catch |err| {
Output.err(err, "failed to download cross-compiled bun executable", .{});
Global.exit(1);
},
);
fd.assertKind(.system);
if (Environment.isWindows) {
var outfile_buf: bun.OSPathBuffer = undefined;
const outfile_slice = brk: {
const outfile_w = bun.strings.toWPathNormalized(&outfile_buf, std.fs.path.basenameWindows(outfile));
bun.assert(outfile_w.ptr == &outfile_buf);
const outfile_buf_u16 = bun.reinterpretSlice(u16, &outfile_buf);
outfile_buf_u16[outfile_w.len] = 0;
break :brk outfile_buf_u16[0..outfile_w.len :0];
};
bun.C.moveOpenedFileAtLoose(fd, bun.toFD(root_dir.fd), outfile_slice, true).unwrap() catch |err| {
if (err == error.EISDIR) {
Output.errGeneric("{} is a directory. Please choose a different --outfile or delete the directory", .{bun.fmt.utf16(outfile_slice)});
} else {
Output.err(err, "failed to move executable to result path", .{});
}
_ = bun.C.deleteOpenedFile(fd);
Global.exit(1);
};
return;
}
var buf: bun.PathBuffer = undefined;
const temp_location = bun.getFdPath(fd, &buf) catch |err| {
Output.prettyErrorln("<r><red>error<r><d>:<r> failed to get path for fd: {s}", .{@errorName(err)});
Global.exit(1);
};
if (comptime Environment.isMac) {
if (target.os == .mac) {
var signer = std.process.Child.init(
&.{
"codesign",
"--remove-signature",
temp_location,
},
bun.default_allocator,
);
if (bun.logger.Log.default_log_level.atLeast(.verbose)) {
signer.stdout_behavior = .Inherit;
signer.stderr_behavior = .Inherit;
signer.stdin_behavior = .Inherit;
} else {
signer.stdout_behavior = .Ignore;
signer.stderr_behavior = .Ignore;
signer.stdin_behavior = .Ignore;
}
_ = signer.spawnAndWait() catch {};
}
}
bun.C.moveFileZWithHandle(
fd,
bun.FD.cwd(),
bun.sliceTo(&(try std.posix.toPosixPath(temp_location)), 0),
bun.toFD(root_dir.fd),
bun.sliceTo(&(try std.posix.toPosixPath(std.fs.path.basename(outfile))), 0),
) catch |err| {
if (err == error.IsDir) {
Output.prettyErrorln("<r><red>error<r><d>:<r> {} is a directory. Please choose a different --outfile or delete the directory", .{bun.fmt.quote(outfile)});
} else {
Output.prettyErrorln("<r><red>error<r><d>:<r> failed to rename {s} to {s}: {s}", .{ temp_location, outfile, @errorName(err) });
}
_ = Syscall.unlink(
&(try std.posix.toPosixPath(temp_location)),
);
Global.exit(1);
};
}
pub fn fromExecutable(allocator: std.mem.Allocator) !?StandaloneModuleGraph {
// Do not invoke libuv here.
const self_exe = openSelf() catch return null;
defer _ = Syscall.close(self_exe);
var trailer_bytes: [4096]u8 = undefined;
std.posix.lseek_END(self_exe.cast(), -4096) catch return null;
var read_amount: usize = 0;
while (read_amount < trailer_bytes.len) {
switch (Syscall.read(self_exe, trailer_bytes[read_amount..])) {
.result => |read| {
if (read == 0) return null;
read_amount += read;
},
.err => {
return null;
},
}
}
if (read_amount < trailer.len + @sizeOf(usize) + 32)
// definitely missing data
return null;
var end = @as([]u8, &trailer_bytes).ptr + read_amount - @sizeOf(usize);
const total_byte_count: usize = @as(usize, @bitCast(end[0..8].*));
if (total_byte_count > std.math.maxInt(u32) or total_byte_count < 4096) {
// sanity check: the total byte count should never be more than 4 GB
// bun is at least like 30 MB so if it reports a size less than 4096 bytes then something is wrong
return null;
}
end -= trailer.len;
if (!bun.strings.hasPrefixComptime(end[0..trailer.len], trailer)) {
// invalid trailer
return null;
}
end -= @sizeOf(Offsets);
const offsets: Offsets = std.mem.bytesAsValue(Offsets, end[0..@sizeOf(Offsets)]).*;
if (offsets.byte_count >= total_byte_count) {
// if we hit this branch then the file is corrupted and we should just give up
return null;
}
var to_read = try bun.default_allocator.alloc(u8, offsets.byte_count);
var to_read_from = to_read;
// Reading the data and making sure it's page-aligned + won't crash due
// to out of bounds using mmap() is very complicated.
// we just read the whole thing into memory for now.
// at the very least
// if you have not a ton of code, we only do a single read() call
if (Environment.allow_assert or offsets.byte_count > 1024 * 3) {
const offset_from_end = trailer_bytes.len - (@intFromPtr(end) - @intFromPtr(@as([]u8, &trailer_bytes).ptr));
std.posix.lseek_END(self_exe.cast(), -@as(i64, @intCast(offset_from_end + offsets.byte_count))) catch return null;
if (comptime Environment.allow_assert) {
// actually we just want to verify this logic is correct in development
if (offsets.byte_count <= 1024 * 3) {
to_read_from = try bun.default_allocator.alloc(u8, offsets.byte_count);
}
}
var remain = to_read_from;
while (remain.len > 0) {
switch (Syscall.read(self_exe, remain)) {
.result => |read| {
if (read == 0) return null;
remain = remain[read..];
},
.err => {
bun.default_allocator.free(to_read);
return null;
},
}
}
}
if (offsets.byte_count <= 1024 * 3) {
// we already have the bytes
end -= offsets.byte_count;
@memcpy(to_read[0..offsets.byte_count], end[0..offsets.byte_count]);
if (comptime Environment.allow_assert) {
bun.assert(bun.strings.eqlLong(to_read, end[0..offsets.byte_count], true));
}
}
return try StandaloneModuleGraph.fromBytes(allocator, to_read, offsets);
}
/// heuristic: `bun build --compile` won't be supported if the name is "bun", "bunx", or "node".
/// this is a cheap way to avoid the extra overhead of opening the executable, and also just makes sense.
fn isBuiltInExe(comptime T: type, argv0: []const T) bool {
if (argv0.len == 0) return false;
if (argv0.len == 3) {
if (bun.strings.eqlComptimeCheckLenWithType(T, argv0, bun.strings.literal(T, "bun"), false)) {
return true;
}
}
if (argv0.len == 4) {
if (bun.strings.eqlComptimeCheckLenWithType(T, argv0, bun.strings.literal(T, "bunx"), false)) {
return true;
}
if (bun.strings.eqlComptimeCheckLenWithType(T, argv0, bun.strings.literal(T, "node"), false)) {
return true;
}
}
if (comptime Environment.isDebug) {
if (bun.strings.eqlComptimeCheckLenWithType(T, argv0, bun.strings.literal(T, "bun-debug"), true)) {
return true;
}
if (bun.strings.eqlComptimeCheckLenWithType(T, argv0, bun.strings.literal(T, "bun-debugx"), true)) {
return true;
}
}
return false;
}
fn openSelf() std.fs.OpenSelfExeError!bun.FileDescriptor {
if (!Environment.isWindows) {
const argv = bun.argv;
if (argv.len > 0) {
if (isBuiltInExe(u8, argv[0])) {
return error.FileNotFound;
}
}
}
switch (Environment.os) {
.linux => {
if (std.fs.openFileAbsoluteZ("/proc/self/exe", .{})) |easymode| {
return bun.toFD(easymode.handle);
} else |_| {
if (bun.argv.len > 0) {
// The user doesn't have /proc/ mounted, so now we just guess and hope for the best.
var whichbuf: bun.PathBuffer = undefined;
if (bun.which(
&whichbuf,
bun.getenvZ("PATH") orelse return error.FileNotFound,
"",
bun.argv[0],
)) |path| {
return bun.toFD((try std.fs.cwd().openFileZ(path, .{})).handle);
}
}
return error.FileNotFound;
}
},
.mac => {
// Use of MAX_PATH_BYTES here is valid as the resulting path is immediately
// opened with no modification.
const self_exe_path = try bun.selfExePath();
const file = try std.fs.openFileAbsoluteZ(self_exe_path.ptr, .{});
return bun.toFD(file.handle);
},
.windows => {
const image_path_unicode_string = std.os.windows.peb().ProcessParameters.ImagePathName;
const image_path = image_path_unicode_string.Buffer.?[0 .. image_path_unicode_string.Length / 2];
var nt_path_buf: bun.WPathBuffer = undefined;
const nt_path = bun.strings.addNTPathPrefix(&nt_path_buf, image_path);
const basename_start = std.mem.lastIndexOfScalar(u16, nt_path, '\\') orelse
return error.FileNotFound;
const basename = nt_path[basename_start + 1 .. nt_path.len - ".exe".len];
if (isBuiltInExe(u16, basename)) {
return error.FileNotFound;
}
return bun.sys.openFileAtWindows(
bun.FileDescriptor.cwd(),
nt_path,
// access_mask
w.SYNCHRONIZE | w.GENERIC_READ,
// create disposition
w.FILE_OPEN,
// create options
w.FILE_SYNCHRONOUS_IO_NONALERT | w.FILE_OPEN_REPARSE_POINT,
).unwrap() catch {
return error.FileNotFound;
};
},
else => @compileError("TODO"),
}
}
/// Source map serialization in the bundler is specially designed to be
/// loaded in memory as is. Source contents are compressed with ZSTD to
/// reduce the file size, and mappings are stored as uncompressed VLQ.
pub const SerializedSourceMap = struct {
bytes: []const u8,
/// Following the header bytes:
/// - source_files_count number of StringPointer, file names
/// - source_files_count number of StringPointer, zstd compressed contents
/// - the mapping data, `map_vlq_length` bytes
/// - all the StringPointer contents
pub const Header = extern struct {
source_files_count: u32,
map_bytes_length: u32,
};
pub fn header(map: SerializedSourceMap) *align(1) const Header {
return @ptrCast(map.bytes.ptr);
}
pub fn mappingVLQ(map: SerializedSourceMap) []const u8 {
const head = map.header();
const start = @sizeOf(Header) + head.source_files_count * @sizeOf(StringPointer) * 2;
return map.bytes[start..][0..head.map_bytes_length];
}
pub fn sourceFileNames(map: SerializedSourceMap) []align(1) const StringPointer {
const head = map.header();
return @as([*]align(1) const StringPointer, @ptrCast(map.bytes[@sizeOf(Header)..]))[0..head.source_files_count];
}
fn compressedSourceFiles(map: SerializedSourceMap) []align(1) const StringPointer {
const head = map.header();
return @as([*]align(1) const StringPointer, @ptrCast(map.bytes[@sizeOf(Header)..]))[head.source_files_count..][0..head.source_files_count];
}
/// Once loaded, this map stores additional data for keeping track of source code.
pub const Loaded = struct {
map: SerializedSourceMap,
/// Only decompress source code once! Once a file is decompressed,
/// it is stored here. Decompression failures are stored as an empty
/// string, which will be treated as "no contents".
decompressed_files: []?[]u8,
pub fn sourceFileContents(this: Loaded, index: usize) ?[]const u8 {
if (this.decompressed_files[index]) |decompressed| {
return if (decompressed.len == 0) null else decompressed;
}