Skip to main content

Creating Angular libraries

When your app grows and you make components that could be reused in other apps it's a good thing to split out that code in a library. Also if your app is big already and someone else is going to add a new feature to it, it's much easier if they can work on that feature in a small isolated project and when ready import it into the main app as a library.

The architecture of Angular with modules help us organize the application code into blocks of functionality. The step of splitting into libraries should build on that, but add the benefits of not having to see code not relevant to the modules we're currently working on and to easily install and reuse in several applications without having to duplicate code across projects.

Creating an Angular library is basically about packaging compiled NgModules into a format that can be installed into another application using a package manager like npm. It could be done as easily adding the typescript sources and html templates to a tar.gz with a package.json entry point. However we'd like to save typescript compilation time in the main app by having a pre-compiled library, and also we'd like to make the package as small as possible by flattening and minifying. Also if we're going to use with module loaders like SystemJS we don't want trouble with component relative paths for the templates ( which in some cases requires a moduleId parameter in the component annotation), so to avoid this we'd inline the HTML/CSS templates instead of having them as separate files. So then there's actually quite a bit of work to be done for bundling a library.

But it's possible to automate this, and it doesn't have to be complicated.

The simplest approach would be a command line tool that takes the typescript entry-module source file path and library name as parameter, and emitted a tar.gz that could be installed to another project with npm.

So we did this.

It's sufficient for library re-use across Angular CLI projects, and also into SystemJS setups. It also bundles well with AOT both in the Angular CLI and rollup.

A simple video demo is here: https://youtu.be/6yu-YCqESZ8

And the tool can be found here: https://github.com/fintechneo/ngmakelib

Comments

Popular posts from this blog

My VNC based development environment with Visual Studio Code running on Ubuntu

I've used this setup for my development environment for several years - giving me a developer desktop I can access anywhere. Even from my mobile phone. I've been fixing bugs, writing code and deployed emergency fixes from the bus, train and mountain tops. The setup is based on a lightweight desktop environment. There are lot of alternatives, but I've chosen fluxbox. From a plain AWS Amazon Ubuntu 16.04 ec2 instance I've started like this: sudo apt-get update sudo apt-get install fluxbox Download and extract tigervnc from https://github.com/TigerVNC/tigervnc/releases (I downloaded the binary file from https://bintray.com/tigervnc/stable/tigervnc/1.7.0 named tigervnc-1.7.0.x86_64.tar.gz ) then extract: tar -xvzf tigervnc-1.7.0.x86_64.tar.gz You need to install: sudo apt-get install x11-xkb-utils You need to edit .vnc/xstartup: nano .vnc/xstartup last line replace "twm &" with: fluxbox & Then you can start the

Angular components not reloading on route change

Spent a long time wondering why route changes caused strange effects on my component, and found out that it was because my component wasn't reloading at all. I had this assumption that when a route parameter changed (e.g. /projects/1 changed to /projects/2 ) the component for the route would be reloaded. But this is not the default behaviour of the Angular router. The default behaviour of the Angular router is to reuse the route if the configuration is the same (and not reload the component). But we can override this by providing a RouteReuseStrategy to our @NgModule: providers: [ { provide: RouteReuseStrategy, useClass: AARouteReuseStrategy } ] The full custom implementation of the RouteReuseStrategy will then be like this (and it's the shouldReuseRoute method that changes the behaviour so that the component is reloaded on route parameter change): export class AARouteReuseStrategy extends RouteReuseStrategy { shouldDetach(route: ActivatedRou

Intercepting and adjusting SQL generated by Eclipselink JPA

In some cases it might be useful to intercept and adjust the SQL generated by EclipseLink JPA. E.g. if you want to force an index in mysql you need to append force index (myindex) to the table name. If you create a query on the entitymanager: Query q = em.createQuery("select e from MyEntity"); you can cast it to the EclipseLink JpaQuery: JpaQuery jq = (JpaQuery)q; and the JpaQuery gives you access to the EclipseLink DataBaseQuery where you can prepare it before executing it: DatabaseQuery dbQuery = q.getDatabaseQuery(); dbQuery.prepareCall(((org.eclipse.persistence.jpa.JpaEntityManager)em).getActiveSession(),new DatabaseRecord()); You can now get the sql string and add the forced index: dbQuery.setSQLString(dbQuery.getSQLString()+" force index (myindex)"); and finally you can get the resultset using q.getResultList(). If you have parameters in the sql it's a bit more to it. Currently I've only found one option - probably not optimal since the query is transl