stephanm87 / biocathubangular Goto Github PK

Currently the app is a mono module app which makes sense at this stage.
But for easier extendability I would already sort stuff in modules. If at some stage someone might add a new (lazy loaded) module, currently everything had to be moved around. I there is a complete refactoring happening anyway, we should immediately move everything in modules. Maybe something like this:

• CodeModule: Contains the landing page (dashboard) and "app frame" (the clarity components for the header etc)
• BioCatModule: Contains the experiment editor (lazy loaded)
• SharedModule: Contains generic code that can be used by all modules, e.g. form validators, some pipes, ...

This would lead to some folder structure like this:

src/
├─ app/
│  ├─ modules/
│  │  ├─ core/
│  │  ├─ shared/
│  │  ├─ biocat/
│  ├─ services/
│  ├─ models/
├─ assets/
├─ environments/

Retrobiohub

      if (Object.keys(params).length === 0){ // TODO

The call should be implemented in:

DashboardComponent --> dataService --> experimentService --> RESTAPI Call

Convention is one class per file. Ergo, split this file into separate files per interface.

The Zenodo adapter needs to be updated for three different features:

1. Enable the export of not completely filled forms
2. Enable the Addition of personal token to access personal access to the DB
3. enbale the edition of existing entries

Hi there,

I entered some dummy values in https://biocathubangular.herokuapp.com/ -- the point plot is looking rather unexpected (because of turned y axis):

Add linter and also configure jsdoc rules. The linter should error when there is missing documentation.

I can provide the linter rules from LabNode.

Best practice is to have only environment as a variable. You can combine

export const environment = {
  production: false,
  backendUrl: '/api/db-query'
};

export const environmentEnzymeML = {
  production: false,
  backendUrl:'http://127.0.0.1:8000/'
};

to

export const environment = {
  production: false,
  backendUrlDBQuery: '/api/db-query',
  backendUrlEnzymeML: '/api/enzyme-ml'
};

The DataService is way too big and has to be refactored. Basically all problems can be solved by making use of Angular Forms instead of custom Validation:

• Create a FormGroup for the Experiment model
• Add "sub form groups" for each stepper page
• Each property of the form group will be a form control (e.g. x_unit, y_unit, notes, ...)
• Make use of built in form validation, e.g. Required Validator, ...
• Write custom form validators for cases where built ins don't apply (e.g. validate string currently uses trim which is not done by the standard required validator)
• Replace all progress functions by a generic progress function that does the following (per form group aka stepper page): Count form controls, check how many form controls are valid, display progress
• To display the error messages we can again just check the form group object and collect invalid controls

src/app/model should be src/app/models. The same applies for other occurrences too.

in the step "experimental data", I have to give "notes" about my measurement. Can you make that an optional field? Or maybe you want to ask for method details, e.g. how s.th. was measured or alike? Thanks for considering!

Those files belong to the app, so they should live in the app directory

No underscore variables in Typescript allowed per convention.
There are two places whre those are used. Marked with TODO in the refactoring branch.

The Linode Deploy branch should be merged to master. IDK if any todos are left before this can go there, so I'm assigning it to you Stephan.

events{}

http {

include /etc/nginx/mime.types;

server {

    listen 80;
    server_name localhost;
    root /usr/share/nginx/html;
    index index.html;

    location / {
            try_files $uri $uri/ /index.html;
    }
}

}

Update Clarity and Angular Versions. In case of a complete rewrite it's actually not an update but setting up a completely new project.

Add the spreadsheet library:
https://ej2.syncfusion.com/angular/documentation/spreadsheet/getting-started/

as in #3 , you spoke of "newer version" -- maybe you can add a link to further information on the currently deployed version in the footer of your heroku app? Also maybe one to this github repo, to make it as easy as possible for people to file issues. :) Best, Robert

after completing the workflow, upload to zenodo actually succeeds, but the link shared above is wrong -- instead of (for my example...): https://sandbox.zenodo.org/deposit/829002 it is actually: https://sandbox.zenodo.org/record/829002

hi @StephanM87 ,

looking at this point in the workflow:

I didn't know intuitively, that the lowest line asked for the buffer concentration... Maybe you can put a label left to it?

Thanks and best,
Robert

... at least on my system: mac OS, Firefox.