Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

merge isAsciiWhitespace implementations #2803

Merged
merged 1 commit into from
Sep 26, 2024
Merged

merge isAsciiWhitespace implementations #2803

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
37 changes: 37 additions & 0 deletions src/workerd/api/BUILD.bazel
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,6 +12,8 @@ filegroup(
exclude = [
"html-rewriter.c++",
"data-url.c++",
"encoding.c++",
"util.c++",
"rtti.c++",
"**/*test*.c++",
"pyodide.c++",
Expand All @@ -23,6 +25,8 @@ filegroup(
exclude = [
"html-rewriter.c++",
"data-url.c++",
"encoding.c++",
"util.c++",
"rtti.c++",
"**/*test*.c++",
"pyodide.c++",
Expand All @@ -38,6 +42,8 @@ api_header_exclusions = [
"html-rewriter.h",
"deferred-proxy.h",
"data-url.h",
"encoding.h",
"util.h",
"modules.h",
anonrig marked this conversation as resolved.
Show resolved Hide resolved
"rtti.h",
"**/*test*.h",
Expand Down Expand Up @@ -121,10 +127,40 @@ kj_test(
],
)

wd_cc_library(
name = "util",
srcs = ["util.c++"],
hdrs = ["util.h"],
implementation_deps = ["@simdutf"],
visibility = ["//visibility:public"],
deps = [
"//src/workerd/jsg",
anonrig marked this conversation as resolved.
Show resolved Hide resolved
"//src/workerd/util",
"@capnp-cpp//src/kj",
],
)

wd_cc_library(
name = "encoding",
srcs = ["encoding.c++"],
hdrs = ["encoding.h"],
visibility = ["//visibility:public"],
deps = [
":util",
"//src/workerd/io:compatibility_date_capnp",
"//src/workerd/jsg",
"//src/workerd/util",
"@capnp-cpp//src/kj",
],
)

wd_cc_library(
name = "data-url",
srcs = ["data-url.c++"],
hdrs = ["data-url.h"],
implementation_deps = [
":encoding",
],
visibility = ["//visibility:public"],
deps = [
"//src/workerd/jsg:url",
Expand Down Expand Up @@ -206,6 +242,7 @@ kj_test(
kj_test(
src = "api-rtti-test.c++",
deps = [
":encoding",
":html-rewriter",
"//src/workerd/io",
"//src/workerd/jsg:rtti",
Expand Down
1 change: 1 addition & 0 deletions src/workerd/api/analytics-engine.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7,6 +7,7 @@
#include <workerd/api/analytics-engine-impl.h>
#include <workerd/api/analytics-engine.capnp.h>
#include <workerd/api/util.h>
#include <workerd/io/io-util.h>
#include <workerd/jsg/jsg.h>

namespace workerd::api {
Expand Down
14 changes: 2 additions & 12 deletions src/workerd/api/data-url.c++
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,5 +1,7 @@
#include "data-url.h"

#include <workerd/api/encoding.h>

#include <kj/encoding.h>

namespace workerd::api {
Expand All @@ -11,18 +13,6 @@ kj::Maybe<DataUrl> DataUrl::tryParse(kj::StringPtr url) {
return kj::none;
}

static constexpr kj::FixedArray<uint8_t, 256> ascii_whitespace_table = []() consteval {
kj::FixedArray<uint8_t, 256> result{};
for (uint8_t c: {0x09, 0x0a, 0x0c, 0x0d, 0x20}) {
result[c] = true;
}
return result;
}();

constexpr bool isAsciiWhitespace(uint8_t c) noexcept {
Copy link
Collaborator

@fhanau fhanau Sep 25, 2024 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Conceptually, this could affect performance – isAsciiWhitespace() does a single table lookup so having it defined in a different source file will make using it much slower based on needing a function call now. The compiler might just be able to inline it properly if LTO is enabled (such as with the downstream release build), so this is non-blocking.

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I moved the implementation to encoding.h and made it constexpr again. Since it is an extremely small fixed array, I don't think it would have any negative impact, but let me know if you disagree. I can change it back without any hassle.

return ascii_whitespace_table[c];
}

kj::Maybe<DataUrl> DataUrl::from(const jsg::Url& url) {
if (url.getProtocol() != "data:"_kj) return kj::none;
auto clone = url.clone(jsg::Url::EquivalenceOption::IGNORE_FRAGMENTS);
Expand Down
4 changes: 0 additions & 4 deletions src/workerd/api/encoding.c++
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -265,10 +265,6 @@ kj::StringPtr getEncodingId(Encoding encoding) {
Encoding getEncodingForLabel(kj::StringPtr label) {
kj::String labelInsensitive = toLower(label);
const auto trim = [](kj::StringPtr label) {
const auto isAsciiWhitespace = [](auto c) {
return c == 0x09 /* tab */ || c == 0x0a /* lf */ || c == 0x0c /* ff */ ||
c == 0x0d /* cr */ || c == 0x20 /* sp */;
};
size_t start = 0;
auto end = label.size();
while (start < end && isAsciiWhitespace(label[start])) {
Expand Down
12 changes: 12 additions & 0 deletions src/workerd/api/encoding.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -11,6 +11,18 @@

namespace workerd::api {

constexpr kj::FixedArray<uint8_t, 256> ascii_whitespace_table = []() consteval {
kj::FixedArray<uint8_t, 256> result{};
for (uint8_t c: {0x09, 0x0a, 0x0c, 0x0d, 0x20}) {
result[c] = true;
}
return result;
}();

constexpr bool isAsciiWhitespace(uint8_t c) noexcept {
return ascii_whitespace_table[c];
}

// The encodings listed here are defined as required by the Encoding spec.
// The first label is enum we use to identify the encoding in code, while
// the second label is the public identifier.
Expand Down
2 changes: 1 addition & 1 deletion src/workerd/api/form-data.c++
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,10 +6,10 @@

#include "util.h"

#include <workerd/io/io-util.h>
#include <workerd/util/mimetype.h>

#include <kj/compat/http.h>
#include <kj/encoding.h>
#include <kj/parse/char.h>
#include <kj/vector.h>

Expand Down
2 changes: 0 additions & 2 deletions src/workerd/api/gpu/gpu.c++
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,8 +4,6 @@

#include "gpu.h"

#include <workerd/jsg/exception.h>

#include <dawn/dawn_proc.h>

namespace workerd::api::gpu {
Expand Down
3 changes: 2 additions & 1 deletion src/workerd/api/memory-cache.c++
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,6 +2,7 @@

#include <workerd/api/util.h>
#include <workerd/io/io-context.h>
#include <workerd/io/io-util.h>
#include <workerd/jsg/jsg.h>
#include <workerd/jsg/ser.h>
#include <workerd/util/weak-refs.h>
Expand Down Expand Up @@ -29,7 +30,7 @@ static bool hasExpired(const kj::Maybe<double>& expiration, bool allowOutsideIoC
KJ_IF_SOME(e, expiration) {
double now = (allowOutsideIoContext && !IoContext::hasCurrent())
? getCurrentTimeOutsideIoContext()
: api::dateNow();
: dateNow();
return e < now;
}
return false;
Expand Down
2 changes: 0 additions & 2 deletions src/workerd/api/node/async-hooks.c++
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,8 +3,6 @@
// https://opensource.org/licenses/Apache-2.0
#include "async-hooks.h"

#include <kj/vector.h>

namespace workerd::api::node {

jsg::Ref<AsyncLocalStorage> AsyncLocalStorage::constructor(jsg::Lock& js) {
Expand Down
3 changes: 2 additions & 1 deletion src/workerd/api/streams/encoding.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,11 +4,12 @@

#pragma once

#include "../encoding.h"
#include "readable.h"
#include "transform.h"
#include "writable.h"

#include <workerd/api/encoding.h>

namespace workerd::api {

class TextEncoderStream: public TransformStream {
Expand Down
11 changes: 0 additions & 11 deletions src/workerd/api/util.c++
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,9 +6,7 @@

#include "simdutf.h"

#include <workerd/io/io-context.h>
#include <workerd/util/mimetype.h>
#include <workerd/util/thread-scopes.h>

#include <kj/encoding.h>

Expand Down Expand Up @@ -259,15 +257,6 @@ kj::String redactUrl(kj::StringPtr url) {
return kj::String(redacted.releaseAsArray());
}

double dateNow() {
jasnell marked this conversation as resolved.
Show resolved Hide resolved
if (IoContext::hasCurrent()) {
auto& ioContext = IoContext::current();
return (ioContext.now() - kj::UNIX_EPOCH) / kj::MILLISECONDS;
}

return 0.0;
}

kj::Maybe<jsg::V8Ref<v8::Object>> cloneRequestCf(
jsg::Lock& js, kj::Maybe<jsg::V8Ref<v8::Object>> maybeCf) {
KJ_IF_SOME(cf, maybeCf) {
Expand Down
5 changes: 0 additions & 5 deletions src/workerd/api/util.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -95,11 +95,6 @@ kj::String redactUrl(kj::StringPtr url);

// =======================================================================================

// Returns exactly what Date.now() would return.
double dateNow();

// =======================================================================================

void maybeWarnIfNotText(jsg::Lock& js, kj::StringPtr str);

kj::String fastEncodeBase64Url(kj::ArrayPtr<const byte> bytes);
Expand Down
3 changes: 3 additions & 0 deletions src/workerd/io/BUILD.bazel
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -49,6 +49,7 @@ wd_cc_library(
"hibernation-manager.c++",
"io-context.c++",
"io-own.c++",
"io-util.c++",
"worker.c++",
] + ["//src/workerd/api:srcs"],
hdrs = [
Expand All @@ -57,6 +58,7 @@ wd_cc_library(
"hibernation-manager.h",
"io-context.h",
"io-own.h",
"io-util.h",
"promise-wrapper.h",
"worker.h",
] + ["//src/workerd/api:hdrs"],
Expand Down Expand Up @@ -89,6 +91,7 @@ wd_cc_library(
"//src/workerd/api:analytics-engine_capnp",
"//src/workerd/api:data-url",
"//src/workerd/api:deferred-proxy",
"//src/workerd/api:encoding",
"//src/workerd/api:r2-api_capnp",
"//src/workerd/jsg",
"//src/workerd/util:autogate",
Expand Down
4 changes: 2 additions & 2 deletions src/workerd/io/io-channels.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,8 +4,8 @@

#pragma once

#include <workerd/api/util.h>
#include <workerd/io/actor-id.h>
#include <workerd/io/io-util.h>
#include <workerd/io/trace.h>

#include <kj/debug.h>
Expand Down Expand Up @@ -101,7 +101,7 @@ class IoChannelFactory {
kj::Maybe<kj::StringPtr> featureFlagsForFl;

// Timestamp for when a subrequest is started. (ms since the Unix Epoch)
double startTime = api::dateNow();
double startTime = dateNow();
};

virtual kj::Own<WorkerInterface> startSubrequest(uint channel, SubrequestMetadata metadata) = 0;
Expand Down
22 changes: 22 additions & 0 deletions src/workerd/io/io-util.c++
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,22 @@
// Copyright (c) 2017-2022 Cloudflare, Inc.
// Licensed under the Apache 2.0 license found in the LICENSE file or at:
// https://opensource.org/licenses/Apache-2.0

#include "io-util.h"

#include "io-context.h"

#include <kj/time.h>

namespace workerd {

double dateNow() {
if (IoContext::hasCurrent()) {
auto& ioContext = IoContext::current();
return (ioContext.now() - kj::UNIX_EPOCH) / kj::MILLISECONDS;
}

return 0.0;
}

} // namespace workerd
12 changes: 12 additions & 0 deletions src/workerd/io/io-util.h
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,12 @@
// Copyright (c) 2017-2022 Cloudflare, Inc.
// Licensed under the Apache 2.0 license found in the LICENSE file or at:
// https://opensource.org/licenses/Apache-2.0

#pragma once

namespace workerd {

// Returns exactly what Date.now() would return.
double dateNow();

} // namespace workerd
2 changes: 1 addition & 1 deletion src/workerd/jsg/BUILD.bazel
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -52,6 +52,7 @@ wd_cc_library(
"web-idl.h",
"wrappable.h",
],
implementation_deps = ["@simdutf"],
visibility = ["//visibility:public"],
deps = [
":exception",
Expand All @@ -64,7 +65,6 @@ wd_cc_library(
"//src/workerd/util:thread-scopes",
"//src/workerd/util:uuid",
"@capnp-cpp//src/kj",
"@simdutf",
"@workerd-v8//:v8",
],
)
Expand Down
Loading