developers
Developer GuideTable of contents
Description
So, you’ve decided to use npm to develop (and maybe publish/deploy) your project.
Fantastic!
There are a few things that you need to do above the simple steps that your users will do to install your program.
About These Documents
These are man pages. If you install npm, you should be able to
then do man npm-thing
to get the documentation on a particular
topic, or npm help thing
to see the same information.
What is a Package
A package is:
- a) a folder containing a program described by a package.json file
- b) a gzipped tarball containing (a)
- c) a url that resolves to (b)
- d) a
<name>@<version>
that is published on the registry with (c) - e) a
<name>@<tag>
that points to (d) - f) a
<name>
that has a “latest” tag satisfying (e) - g) a
git
url that, when cloned, results in (a).
Even if you never publish your package, you can still get a lot of benefits of using npm if you just want to write a node program (a), and perhaps if you also want to be able to easily install it elsewhere after packing it up into a tarball (b).
Git urls can be of the form:
git://github.com/user/project.git#commit-ish
git+ssh://user@hostname:project.git#commit-ish
git+http://user@hostname/project/blah.git#commit-ish
git+https://user@hostname/project/blah.git#commit-ish
The commit-ish
can be any tag, sha, or branch which can be supplied as
an argument to git checkout
. The default is whatever the repository uses
as its default branch.
The package.json File
You need to have a package.json
file in the root of your project to do
much of anything with npm. That is basically the whole interface.
See package.json
for details about what
goes in that file. At the very least, you need:
-
name: This should be a string that identifies your project. Please do not use the name to specify that it runs on node, or is in JavaScript. You can use the “engines” field to explicitly state the versions of node (or whatever else) that your program requires, and it’s pretty well assumed that it’s JavaScript.
It does not necessarily need to match your github repository name.
So,
node-foo
andbar-js
are bad names.foo
orbar
are better. -
version: A semver-compatible version.
-
engines: Specify the versions of node (or whatever else) that your program runs on. The node API changes a lot, and there may be bugs or new functionality that you depend on. Be explicit.
-
author: Take some credit.
-
scripts: If you have a special compilation or installation script, then you should put it in the
scripts
object. You should definitely have at least a basic smoke-test command as the “scripts.test” field. See scripts. -
main: If you have a single module that serves as the entry point to your program (like what the “foo” package gives you at require(“foo”)), then you need to specify that in the “main” field.
-
directories: This is an object mapping names to folders. The best ones to include are “lib” and “doc”, but if you use “man” to specify a folder full of man pages, they’ll get installed just like these ones.
You can use npm init
in the root of your package in order to get you
started with a pretty basic package.json file. See npm init
for more info.
Keeping files out of your Package
Use a .npmignore
file to keep stuff out of your package. If there’s no
.npmignore
file, but there is a .gitignore
file, then npm will ignore
the stuff matched by the .gitignore
file. If you want to include
something that is excluded by your .gitignore
file, you can create an
empty .npmignore
file to override it. Like git
, npm
looks for
.npmignore
and .gitignore
files in all subdirectories of your package,
not only the root directory.
.npmignore
files follow the same pattern
rules
as .gitignore
files:
- Blank lines or lines starting with
#
are ignored. - Standard glob patterns work.
- You can end patterns with a forward slash
/
to specify a directory. - You can negate a pattern by starting it with an exclamation point
!
.
By default, the following paths and files are ignored, so there’s no
need to add them to .npmignore
explicitly:
.*.swp
._*
.DS_Store
.git
.hg
.npmrc
.lock-wscript
.svn
.wafpickle-*
config.gypi
CVS
npm-debug.log
Additionally, everything in node_modules
is ignored, except for
bundled dependencies. npm automatically handles this for you, so don’t
bother adding node_modules
to .npmignore
.
The following paths and files are never ignored, so adding them to
.npmignore
is pointless:
package.json
README
(and its variants)CHANGELOG
(and its variants)LICENSE
/LICENCE
If, given the structure of your project, you find .npmignore
to be a
maintenance headache, you might instead try populating the files
property of package.json
, which is an array of file or directory names
that should be included in your package. Sometimes manually picking
which items to allow is easier to manage than building a block list.
Testing whether your .npmignore
or files
config works
If you want to double check that your package will include only the files
you intend it to when published, you can run the npm pack
command locally
which will generate a tarball in the working directory, the same way it
does for publishing.
Link Packages
npm link
is designed to install a development package and see the
changes in real time without having to keep re-installing it. (You do
need to either re-link or npm rebuild -g
to update compiled packages,
of course.)
More info at npm link
.
Before Publishing: Make Sure Your Package Installs and Works
This is important.
If you can not install it locally, you’ll have problems trying to publish it. Or, worse yet, you’ll be able to publish it, but you’ll be publishing a broken or pointless package. So don’t do that.
In the root of your package, do this:
npm install . -g
That’ll show you that it’s working. If you’d rather just create a symlink package that points to your working directory, then do this:
npm link
Use npm ls -g
to see if it’s there.
To test a local install, go into some other folder, and then do:
cd ../some-other-folder
npm install ../my-package
to install it locally into the node_modules folder in that other place.
Then go into the node-repl, and try using require(“my-thing”) to bring in your module’s main module.
Create a User Account
Create a user with the adduser command. It works like this:
npm adduser
and then follow the prompts.
This is documented better in npm adduser.
Publish your Package
This part’s easy. In the root of your folder, do this:
npm publish
You can give publish a url to a tarball, or a filename of a tarball, or a path to a folder.
Note that pretty much everything in that folder will be exposed
by default. So, if you have secret stuff in there, use a
.npmignore
file to list out the globs to ignore, or publish
from a fresh checkout.
Brag about it
Send emails, write blogs, blab in IRC.
Tell the world how easy it is to install your program!