31 KiB
Deno Manual
[toc]
Disclaimer
A word of caution: Deno is very much under development. We encourage brave early adopters, but expect bugs large and small. The API is subject to change without notice. Bug reports do help!
Introduction
A secure JavaScript/TypeScript runtime built with V8, Rust, and Tokio
Philosophy
Deno aims to be a productive and secure scripting environment for the modern programmer.
It will always be distributed as a single executable - and that executable will be sufficient software to run any deno program. Given a URL to a deno program, you should be able to execute it with nothing more than the 50 megabyte deno executable.
Deno explicitly takes on the role of both runtime and package manager. It uses a standard browser-compatible protocol for loading modules: URLs.
Deno provides security guarantees about how programs can access your system with the default being the most restrictive secure sandbox.
Deno provides a set of reviewed (audited) standard modules that are guaranteed to work with Deno.
Goals
-
Support TypeScript out of the box.
-
Like the browser, allows imports from URLs:
import * as log from "https://deno.land/std/log/mod.ts";
-
Remote code is fetched and cached on first execution, and never updated until the code is run with the
--reload
flag. (So, this will still work on an airplane. See~/.deno/src
for details on the cache.) -
Uses "ES Modules" and does not support
require()
. -
File system and network access can be controlled in order to run sandboxed code. Access between V8 (unprivileged) and Rust (privileged) is only done via serialized messages defined in this flatbuffer. This makes it easy to audit. For example, to enable write access use the flag
--allow-write
or for network access--allow-net
. -
Only ship a single executable.
-
Always dies on uncaught errors.
-
Browser compatible: The subset of Deno programs which are written completely in JavaScript and do not use the global
Deno
namespace (or feature test for it), ought to also be able to be run in a modern web browser without change. -
Be able to serve HTTP efficiently. (Currently it is relatively slow.)
-
Provide useful tooling out of the box:
Non-goals
-
No
package.json
. -
No npm.
-
Not explicitly compatible with Node.
Setup
Binary Install
Deno works on OSX, Linux, and Windows. Deno is a single binary executable. It has no external dependencies.
deno_install provides convenience scripts to download and install the binary.
Using Shell:
curl -fsSL https://deno.land/x/install/install.sh | sh
Using PowerShell:
iwr https://deno.land/x/install/install.ps1 | iex
Using Scoop (windows):
scoop install deno
Using Homebrew (mac):
brew install deno
Deno can also be installed manually, by downloading a tarball or zip file at github.com/denoland/deno/releases. These packages contain just a single executable file. You will have to set the executable bit on Mac and Linux.
Once it's installed and in your $PATH
, try it:
deno run https://deno.land/welcome.ts
Build from source
Clone on Unix/MacOs
# Fetch deps.
git clone --recurse-submodules https://github.com/denoland/deno.git
Clone on Windows (with Admin privileges)
git clone -c core.symlinks=true --recurse-submodules https://github.com/denoland/deno.git
cd deno
./tools/setup.py
# You may need to ensure that sccache is running.
# (TODO it's unclear if this is necessary or not.)
# prebuilt/mac/sccache --start-server
# Build.
./tools/build.py
# Run.
./target/debug/deno run tests/002_hello.ts
# Test.
./tools/test.py
# Format code.
./tools/format.py
Prerequisites
To ensure reproducible builds, deno has most of its dependencies in a git submodule. However, you need to install separately:
Extra steps for Mac users: install XCode :(
Extra steps for Windows users:
- Add
python.exe
toPATH
(e.g.set PATH=%PATH%;C:\Python27\python.exe
) - Get VS Community 2017 with
"Desktop development with C++" toolkit and make sure to select the following
required tools listed below along with all C++ tools.
- Windows 10 SDK >= 10.0.17134
- Visual C++ ATL for x86 and x64
- Visual C++ MFC for x86 and x64
- C++ profiling tools
- Enable "Debugging Tools for Windows". Go to "Control Panel" → "Programs" → "Programs and Features" → Select "Windows Software Development Kit - Windows 10" → "Change" → "Change" → Check "Debugging Tools For Windows" → "Change" -> "Finish".
- Make sure you are using git version 2.19.2.windows.1 or newer.
Other useful commands
# Call ninja manually.
./third_party/depot_tools/ninja -C target/debug
# Build a release binary.
./tools/build.py --release deno
# List executable targets.
./third_party/depot_tools/gn ls target/debug //:* --as=output --type=executable
# List build configuration.
./third_party/depot_tools/gn args target/debug/ --list
# Edit build configuration.
./third_party/depot_tools/gn args target/debug/
# Describe a target.
./third_party/depot_tools/gn desc target/debug/ :deno
./third_party/depot_tools/gn help
# Update third_party modules
git submodule update
# Skip downloading binary build tools and point the build
# to the system provided ones (for packagers of deno ...).
./tools/setup.py --no-binary-download
export DENO_BUILD_ARGS="clang_base_path=/usr clang_use_chrome_plugins=false"
DENO_GN_PATH=/usr/bin/gn DENO_NINJA_PATH=/usr/bin/ninja ./tools/build.py
Environment variables: DENO_BUILD_MODE
, DENO_BUILD_PATH
, DENO_BUILD_ARGS
,
DENO_DIR
, DENO_GN_PATH
, DENO_NINJA_PATH
.
API reference
deno types
To get an exact reference of deno's runtime API, run the following in the command line:
$ deno types
This is what the output looks like.
Reference websites
If you are embedding deno in a Rust program, see Rust Deno API.
The Deno crate is hosted on crates.io.
Examples
An implementation of the unix "cat" program
In this program each command-line argument is assumed to be a filename, the file is opened, and printed to stdout.
(async () => {
for (let i = 1; i < Deno.args.length; i++) {
let filename = Deno.args[i];
let file = await Deno.open(filename);
await Deno.copy(Deno.stdout, file);
file.close();
}
})();
The copy()
function here actually makes no more than the necessary kernel ->
userspace -> kernel copies. That is, the same memory from which data is read
from the file, is written to stdout. This illustrates a general design goal for
I/O streams in Deno.
Try the program:
$ deno run --allow-read https://deno.land/std/examples/cat.ts /etc/passwd
TCP echo server
This is an example of a simple server which accepts connections on port 8080, and returns to the client anything it sends.
const { listen, copy } = Deno;
(async () => {
const addr = "0.0.0.0:8080";
const listener = listen("tcp", addr);
console.log("listening on", addr);
while (true) {
const conn = await listener.accept();
copy(conn, conn);
}
})();
When this program is started, the user is prompted for permission to listen on the network:
$ deno run https://deno.land/std/examples/echo_server.ts
⚠️ Deno requests network access to "listen". Grant? [a/y/n/d (a = allow always, y = allow once, n = deny once, d = deny always)]
For security reasons, deno does not allow programs to access the network without explicit permission. To avoid the console prompt, use a command-line flag:
$ deno run --allow-net https://deno.land/std/examples/echo_server.ts
To test it, try sending a HTTP request to it by using curl. The request gets written directly back to the client.
$ curl http://localhost:8080/
GET / HTTP/1.1
Host: localhost:8080
User-Agent: curl/7.54.0
Accept: */*
It's worth noting that like the cat.ts
example, the copy()
function here
also does not make unnecessary memory copies. It receives a packet from the
kernel and sends back, without further complexity.
Inspecting and revoking permissions
Sometimes a program may want to revoke previously granted permissions. When a program, at a later stage, needs those permissions, a new prompt will be presented to the user.
const { permissions, revokePermission, open, remove } = Deno;
(async () => {
// lookup a permission
if (!permissions().write) {
throw new Error("need write permission");
}
const log = await open("request.log", "a+");
// revoke some permissions
revokePermission("read");
revokePermission("write");
// use the log file
const encoder = new TextEncoder();
await log.write(encoder.encode("hello\n"));
// this will prompt for the write permission or fail.
await remove("request.log");
})();
File server
This one serves a local directory in HTTP.
alias file_server="deno run --allow-net --allow-read \
https://deno.land/std/http/file_server.ts"
Run it:
$ file_server .
Downloading https://deno.land/std/http/file_server.ts...
[...]
HTTP server listening on http://0.0.0.0:4500/
And if you ever want to upgrade to the latest published version:
$ file_server --reload
Permissions whitelist
deno also provides permissions whitelist.
This is an example to restrict File system access by whitelist.
$ deno run --allow-read=/usr https://deno.land/std/examples/cat.ts /etc/passwd
⚠️ Deno requests read access to "/etc/passwd". Grant? [a/y/n/d (a = allow always, y = allow once, n = deny once, d = deny always)]
You can grant read permission under /etc
dir
$ deno run --allow-read=/etc https://deno.land/std/examples/cat.ts /etc/passwd
--allow-write
works same as --allow-read
.
This is an example to restrict host.
(async () => {
const result = await fetch("https://deno.land/std/examples/echo_server.ts");
})();
$ deno run --allow-net=deno.land allow-net-whitelist-example.ts
Run subprocess
Example:
async function main() {
// create subprocess
const p = Deno.run({
args: ["echo", "hello"]
});
// await its completion
await p.status();
}
main();
Run it:
$ deno run --allow-run ./subprocess_simple.ts
hello
By default when you use Deno.run()
subprocess inherits stdin
, stdout
and
stderr
of parent process. If you want to communicate with started subprocess
you can use "piped"
option.
async function main() {
const decoder = new TextDecoder();
const fileNames = Deno.args.slice(1);
const p = Deno.run({
args: [
"deno",
"run",
"--allow-read",
"https://deno.land/std/examples/cat.ts",
...fileNames
],
stdout: "piped",
stderr: "piped"
});
const { code } = await p.status();
if (code === 0) {
const rawOutput = await p.output();
await Deno.stdout.write(rawOutput);
} else {
const rawError = await p.stderrOutput();
const errorString = decoder.decode(rawError);
console.log(errorString);
}
Deno.exit(code);
}
main();
When you run it:
$ deno run --allow-run ./subprocess.ts <somefile>
[file content]
$ deno run --allow-run ./subprocess.ts non_existent_file.md
Uncaught NotFound: No such file or directory (os error 2)
at DenoError (deno/js/errors.ts:22:5)
at maybeError (deno/js/errors.ts:41:12)
at handleAsyncMsgFromRust (deno/js/dispatch.ts:27:17)
Linking to third party code
In the above examples, we saw that Deno could execute scripts from URLs. Like browser JavaScript, Deno can import libraries directly from URLs. This example uses a URL to import a test runner library:
import { test, runIfMain } from "https://deno.land/std/testing/mod.ts";
import { assertEquals } from "https://deno.land/std/testing/asserts.ts";
test(function t1() {
assertEquals("hello", "hello");
});
test(function t2() {
assertEquals("world", "world");
});
runIfMain(import.meta);
Try running this:
$ deno run test.ts
running 2 tests
test t1 ... ok
test t2 ... ok
test result: ok. 2 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out
Note that we did not have to provide the --allow-net
flag for this program,
and yet it accessed the network. The runtime has special access to download
imports and cache them to disk.
Deno caches remote imports in a special directory specified by the $DENO_DIR
environmental variable. It defaults to the system's cache directory if
$DENO_DIR
is not specified. The next time you run the program, no downloads
will be made. If the program hasn't changed, it won't be recompiled either. The
default directory is:
- On Linux/Redox:
$XDG_CACHE_HOME/deno
or$HOME/.cache/deno
- On Windows:
%LOCALAPPDATA%/deno
(%LOCALAPPDATA%
=FOLDERID_LocalAppData
) - On macOS:
$HOME/Library/Caches/deno
- If something fails, it falls back to
$HOME/.deno
But what if https://deno.land/
goes down? Relying on external servers is
convenient for development but brittle in production. Production software should
always bundle its dependencies. In Deno this is done by checking the $DENO_DIR
into your source control system, and specifying that path as the $DENO_DIR
environmental variable at runtime.
How do you import to a specific version? Simply specify the version in the
URL. For example, this URL fully specifies the code being run:
https://unpkg.com/liltest@0.0.5/dist/liltest.js
. Combined with the
aforementioned technique of setting $DENO_DIR
in production to stored code,
one can fully specify the exact code being run, and execute the code without
network access.
It seems unwieldy to import URLs everywhere. What if one of the URLs links to
a subtly different version of a library? Isn't it error prone to maintain URLs
everywhere in a large project? The solution is to import and re-export your
external libraries in a central deps.ts
file (which serves the same purpose as
Node's package.json
file). For example, let's say you were using the above
testing library across a large project. Rather than importing
"https://deno.land/std/testing/mod.ts"
everywhere, you could create a
deps.ts
file the exports the third-party code:
export { test, assertEquals } from "https://deno.land/std/testing/mod.ts";
And throughout project one can import from the deps.ts
and avoid having many
references to the same URL:
import { test, assertEquals } from "./deps.ts";
This design circumvents a plethora of complexity spawned by package management software, centralized code repositories, and superfluous file formats.
Testing if current file is the main program
To test if the current script has been executed as the main input to the program
check import.meta.main
.
if (import.meta.main) {
console.log("main");
}
Command line interface
Flags
Use deno help
to see the help text.
USAGE:
deno [FLAGS] [OPTIONS] [SUBCOMMAND]
FLAGS:
-A, --allow-all Allow all permissions
--allow-env Allow environment access
--allow-hrtime Allow high resolution time measurement
--allow-run Allow running subprocesses
-h, --help Prints help information
--no-prompt Do not use prompts
-r, --reload Reload source code cache (recompile TypeScript)
--v8-options Print V8 command line options
OPTIONS:
--allow-net=<allow-net> Allow network access
--allow-read=<allow-read> Allow file system read access
--allow-write=<allow-write> Allow file system write access
-c, --config <FILE> Load compiler configuration file
--importmap <FILE> Load import map file
-L, --log-level <log-level> Set log level [possible values: debug, info]
--seed <NUMBER> Seed Math.random()
--v8-flags=<v8-flags> Set V8 command line options
SUBCOMMANDS:
[SCRIPT] Script to run
bundle Bundle module and dependencies into single file
completions Generate shell completions
eval Eval script
fetch Fetch the dependencies
fmt Format files
help Prints this message or the help of the given subcommand(s)
info Show source file related info
install Install script as executable
run Run a program given a filename or url to the source code
types Print runtime TypeScript declarations
version Print the version
xeval Eval a script on text segments from stdin
ENVIRONMENT VARIABLES:
DENO_DIR Set deno's base directory
NO_COLOR Set to disable color
Environmental variables
There are several env vars that control how Deno behaves:
DENO_DIR
defaults to $HOME/.deno
but can be set to any path to control where
generated and cached source code is written and read to.
NO_COLOR
will turn off color output if set. See https://no-color.org/. User
code can test if NO_COLOR
was set without having --allow-env
by using the
boolean constant Deno.noColor
.
Shell completion
You can generate completion script for your shell using the
deno completions <shell>
command. The command outputs to stdout so you should
redirect it to an appropriate file.
The supported shells are:
- zsh
- bash
- fish
- powershell
- elvish
Example:
deno completions bash > /usr/local/etc/bash_completion.d/deno.bash
source /usr/local/etc/bash_completion.d/deno.bash
V8 flags
V8 has many many internal command-line flags, that you can see with
--v8-options
.
It looks like this.
Particularly useful ones:
--async-stack-trace
Bundling
deno bundle [URL]
will output a single JavaScript file, using
AMD, which
includes all dependencies of the specified input.
> deno bundle https://deno.land/std/examples/colors.ts
Bundling "colors.bundle.js"
Emitting bundle to "colors.bundle.js"
9.2 kB emitted.
To run then bundle in Deno use
deno https://deno.land/std/bundle/run.ts colors.bundle.js
Bundles can also be loaded in the web browser with the assistance of
RequireJS. Suppose we have a bundle called
website.bundle.js
, then the following HTML should be able to load it:
<script src="https://cdnjs.cloudflare.com/ajax/libs/require.js/2.3.6/require.min.js"></script>
<script src="website.bundle.js"></script>
<script>
requirejs(["website"], website => website.main());
</script>
Here we assume there's an exported function main()
from website.ts
.
// website.ts
export main() {
console.log("hello from the web browser");
}
Installing executable scripts
Deno provides ability to easily install and distribute executable code via
deno install
command.
deno install [EXE_NAME] [URL] [FLAGS...]
will install script available at
URL
with name EXE_NAME
.
This command is a thin wrapper that creates executable shell scripts which
invoke deno
with specified permissions and CLI flags.
Example:
$ deno install file_server https://deno.land/std/http/file_server.ts --allow-net --allow-read
[1/1] Compiling https://deno.land/std/http/file_server.ts
✅ Successfully installed file_server.
/Users/deno/.deno/bin/file_server
By default scripts are installed at $HOME/.deno/bin
and that directory must be
added to the path manually.
$ echo 'export PATH="$HOME/.deno/bin:$PATH"' >> ~/.bashrc
Installation directory can be changed using -d/--dir
flag:
$ deno install --dir /usr/local/bin prettier https://deno.land/std/prettier/main.ts --allow-write --allow-read
When installing a script you can specify permissions that will be used to run the script. They are placed after the script URL and can be mixed with any additional CLI flags you want to pass to the script.
Example:
$ deno install format_check https://deno.land/std/prettier/main.ts --allow-write --allow-read --check --print-width 88 --tab-width 2
Above command creates an executable called format_check
that runs prettier
with write and read permissions. When you run format_check
deno will run
prettier in check
mode and configured to use 88
column width with 2
column
tabs.
It is a good practice to use import.meta.main
idiom for an entry point for
executable file. See
Testing if current file is the main program
section.
Example:
// https://example.com/awesome/cli.ts
async function myAwesomeCli(): Promise<void> {
-- snip --
}
if (import.meta.main) {
myAwesomeCli();
}
When you create executable script make sure to let users know by adding example installation command to your repository:
# Install using deno install
$ deno install awesome_cli https://example.com/awesome/cli.ts
Import maps
Deno supports import maps.
One can use import map with --importmap=<FILE>
CLI flag.
Current limitations:
- single import map
- no fallback URLs
- Deno does not support
std:
namespace - Does supports only
file:
,http:
andhttps:
schemes
Example:
// import_map.json
{
"imports": {
"http/": "https://deno.land/std/http/"
}
}
// hello_server.ts
import { serve } from "http/server.ts";
async function main() {
const body = new TextEncoder().encode("Hello World\n");
for await (const req of serve(":8000")) {
req.respond({ body });
}
}
main();
$ deno run --importmap=import_map.json hello_server.ts
Internal details
Deno and Linux analogy
Linux | Deno |
---|---|
Processes | Web Workers |
Syscalls | Ops |
File descriptors (fd) | Resource ids (rid) |
Scheduler | Tokio |
Userland: libc++ / glib / boost | deno_std |
/proc/$$/stat | Deno.metrics() |
man pages | deno types |
Resources
Resources (AKA rid
) are Deno's version of file descriptors. They are integer
values used to refer to open files, sockets, and other concepts. For testing it
would be good to be able to query the system for how many open resources there
are.
const { resources, close } = Deno;
console.log(resources());
// output like: { 0: "stdin", 1: "stdout", 2: "stderr", 3: "repl" }
// close resource by rid
close(3);
Metrics
Metrics is deno's internal counters for various statics.
> console.table(Deno.metrics())
┌──────────────────┬────────┐
│ (index) │ Values │
├──────────────────┼────────┤
│ opsDispatched │ 9 │
│ opsCompleted │ 9 │
│ bytesSentControl │ 504 │
│ bytesSentData │ 0 │
│ bytesReceived │ 856 │
└──────────────────┴────────┘
Schematic diagram
Profiling
To start profiling,
# Make sure we're only building release.
# Build deno and V8's d8.
./tools/build.py --release d8 deno
# Start the program we want to benchmark with --prof
./target/release/deno tests/http_bench.ts --allow-net --prof &
# Exercise it.
third_party/wrk/linux/wrk http://localhost:4500/
kill `pgrep deno`
V8 will write a file in the current directory that looks like this:
isolate-0x7fad98242400-v8.log
. To examine this file:
D8_PATH=target/release/ ./third_party/v8/tools/linux-tick-processor
isolate-0x7fad98242400-v8.log > prof.log
# on macOS, use ./third_party/v8/tools/mac-tick-processor instead
prof.log
will contain information about tick distribution of different calls.
To view the log with Web UI, generate JSON file of the log:
D8_PATH=target/release/ ./third_party/v8/tools/linux-tick-processor
isolate-0x7fad98242400-v8.log --preprocess > prof.json
Open third_party/v8/tools/profview/index.html
in your browser, and select
prof.json
to view the distribution graphically.
To learn more about d8
and profiling, check out the following links:
Debugging with LLDB
We can use LLDB to debug deno.
$ lldb -- target/debug/deno run tests/worker.js
> run
> bt
> up
> up
> l
To debug Rust code, we can use rust-lldb
. It should come with rustc
and is a
wrapper around LLDB.
$ rust-lldb -- ./target/debug/deno run --allow-net tests/http_bench.ts
# On macOS, you might get warnings like
# `ImportError: cannot import name _remove_dead_weakref`
# In that case, use system python by setting PATH, e.g.
# PATH=/System/Library/Frameworks/Python.framework/Versions/2.7/bin:$PATH
(lldb) command script import "/Users/kevinqian/.rustup/toolchains/1.34.1-x86_64-apple-darwin/lib/rustlib/etc/lldb_rust_formatters.py"
(lldb) type summary add --no-value --python-function lldb_rust_formatters.print_val -x ".*" --category Rust
(lldb) type category enable Rust
(lldb) target create "../deno/target/debug/deno"
Current executable set to '../deno/target/debug/deno' (x86_64).
(lldb) settings set -- target.run-args "tests/http_bench.ts" "--allow-net"
(lldb) b op_start
(lldb) r
Deno Core
The core binding layer for Deno. It is released as a standalone crate. Inside of core is V8 itself, with a binding API called "libdeno". See the crate documentation for more details.
Flatbuffers
We use Flatbuffers to define common structs and enums between TypeScript and Rust. These common data structures are defined in msg.fbs
Updating prebuilt binaries
$ ./third_party/depot_tools/upload_to_google_storage.py -b denoland \
-e ~/.config/gcloud/legacy_credentials/ry@tinyclouds.org/.boto `which sccache`
$ mv `which sccache`.sha1 prebuilt/linux64/
$ gsutil acl ch -u AllUsers:R gs://denoland/608be47bf01004aa11d4ed06955414e93934516e
Continuous Benchmarks
See our benchmarks over here
The benchmark chart supposes //website/data.json
has the type
BenchmarkData[]
where BenchmarkData
is defined like the below:
interface ExecTimeData {
mean: number;
stddev: number;
user: number;
system: number;
min: number;
max: number;
}
interface BenchmarkData {
created_at: string;
sha1: string;
benchmark: {
[key: string]: ExecTimeData;
};
binarySizeData: {
[key: string]: number;
};
threadCountData: {
[key: string]: number;
};
syscallCountData: {
[key: string]: number;
};
}
Logos
These Deno logos, like the Deno software, are distributed under the MIT license (public domain and free for use)
Contributing
Progress towards future releases is tracked here.
Please don't make the benchmarks worse.
Ask for help in the community chat room.
If you are going to work on an issue, mention so in the issue comments before you start working on the issue.
Submitting a pull request
Before submitting, please make sure the following is done:
- That there is a related issue and it is referenced in the PR text.
- There are tests that cover the changes.
- Ensure
./tools/test.py
passes. - Format your code with
tools/format.py
- Make sure
./tools/lint.py
passes.
Changes to third_party
deno_third_party
contains most
of the external code that Deno depends on, so that we know exactly what we are
executing at any given time. It is carefully maintained with a mixture of manual
labor and private scripts. It's likely you will need help from @ry or
@piscisaureus to make changes.
Adding Ops (aka bindings)
We are very concerned about making mistakes when adding new APIs. When adding an Op to Deno, the counterpart interfaces on other platforms should be researched. Please list how this functionality is done in Go, Node, Rust, and Python.
As an example, see how Deno.rename()
was proposed and added in
PR #671.
Documenting APIs
It is important to document public APIs and we want to do that inline with the code. This helps ensure that code and documentation are tightly coupled together.
Utilize JSDoc
All publicly exposed APIs and types, both via the deno
module as well as the
global/window
namespace should have JSDoc documentation. This documentation is
parsed and available to the TypeScript compiler, and therefore easy to provide
further downstream. JSDoc blocks come just prior to the statement they apply to
and are denoted by a leading /**
before terminating with a */
. For example:
/** A simple JSDoc comment */
export const FOO = "foo";