dlmw/swisssign-challenge-ui
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Write readme

Browse Source
This commit is contained in:
dylan
2025-01-31 15:50:39 +01:00
parent a82ccdae91
commit dfcffc7bd3
1 changed files with 33 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

33
README.md
View File

@ -12,6 +12,10 @@ npm i --force
ng serve
```
The default login is **john@example.com** for the username and **password** for the password.
No document is selected when you first arrive on the page. **You must select a document first**.
## Generate code after API changes
When the API is changed, the client source code must be regenerated.
@ -19,3 +23,32 @@ When the API is changed, the client source code must be regenerated.
1. Install the CLI `npm install -g @openapitools/openapi-generator-cli`
1. Download the latest OpenAPI spec file from the API repository. **Please don't commit it!**
1. Run `openapi-generator-cli generate -i [path_to_openapi.yml] -g typescript-angular -o src/gen/`. **Commit those changes**
## Choices
### Tailwind
I've first heard of Tailwind a few months ago when setting up my blog with Astro. I've been bitten a few times by component libraries
that provided components not easy to customize. That's especially true when the end-user wants something that is not idiomatically
feasible.
It basically gives you CSS classes that you can use in your HTML, which was exactly what I needed for that project. LLMs use
Tailwind very well as it is fairly popular and well-documented. No more writing CSS files with "!important" everywhere!
### open-api-generator-cli
I discovered OpenAPI generators about a year ago. Turns out [there's one for NPM](https://openapi-generator.tech/docs/installation),
and it worked great to generate the services to interact with the API. When you make an API change, you're pretty much sure that your
client is going to work - as long as the API respects the specification file.
### ng2-pdf-viewer
I've never interacted with files programmatically. This library made it a bit easier, but I find that there's a lack of a button to print,
zoom, download, ...
I think it's good enough and allowed me to quickly see if my API was returning a correctly-formatted PDF.
## Potential improvements
The first thing I would do is use some global variable to store button styling. For example, it should be red when enabled and gray when
disabled. It should also be with rounded borders. You get the idea.