Generate barcode in angular 2 nativescript - angular2-nativescript

I need to generate barcode based on some string data in angular2 nativescript.
Please help me to achieve this.

Font wasn't a good option for my purposes because I can't control it's width/height.
nativescript-zxing is not compatible with nativescript-barcodescanner.
So, I've done next thing
import { DOMImplementation, XMLSerializer } from 'xmldom';
import * as JsBarcode from 'jsbarcode';
const xmlSerializer = new XMLSerializer();
const document = new DOMImplementation().createDocument('http://www.w3.org/1999/xhtml', 'html', null);
const svgNode = document.createElementNS('http://www.w3.org/2000/svg', 'svg');
JsBarcode(svgNode, 'test', { xmlDocument: document });
const svgText = xmlSerializer.serializeToString(svgNode);
const buffer = new Buffer(svgText);
const src = `data:image/svg+xml;base64,${buffer.toString('base64')}`;
And now src can be passed to SVGImage from plugin nativescript-svg

You should probably try this one nativescript-zxing. This plugin allows you to read barcodes, QR-codes, etc. and create them as well based on provided string. Check the documentation for more information. Have a nice day!

If you only need to generate a barcode and not scan it, you can just use a barcode font and set it to your label, following these steps:
Find a barcode font that suits your needs and copy the font file to the /font folder in your app. This font is a good option.
Add create a new class in your app.css that sets the font-family to the barcode font:
.barcode {
font-family: free3of9 /* Replace with the actual name of your font, free3of9 is the name of the font above */
}
Add Label with text set to the value that you want the barcode to represent and class set to your css class. In order to work correctly you need to prefix and suffix the text with *'s. For example, if you want your barcode to read 12316516132, your Label should look like:
&ltLabel text="*12316516132*" class="barcode"></Label>
This is more efficient way than installing a plugin as it has very little footprint on your application resources.

Thanks to #rsboarder I was able to provide a final solution.
Below the steps should be done to make it work
Install (if any) the following plugins:
npm i jsbarcode
npm i xmldom
npm i #sergeymell/nativescript-svg
don't forget to add this to the module yourname.module.ts
import { NativeScriptSvgModule } from '#sergeymell/nativescript-svg/angular'
...
imports [
...
NativeScriptSvgModule
...
]
...
My config it looks like this:
Create this method on your component yourname.component.ts
import * as JsBarcode from 'jsbarcode'
import { DOMImplementation, XMLSerializer } from 'xmldom'
import { Buffer } from 'buffer'
import { ImageSource } from '#nativescript/core'
...
getBarcodeSVGSrc(barcodeNumber, barcodeFormat: string): string {
const xmlSerializer = new XMLSerializer()
const document = new DOMImplementation().createDocument('http://www.w3.org/1999/xhtml', 'html', null)
const svgNode = document.createElementNS('http://www.w3.org/2000/svg', 'svg')
let data: string = barcodeNumber
const result: ImageSource = new ImageSource()
JsBarcode(svgNode, data, {
xmlDocument: document,
format: barcodeFormat // here: https://github.com/lindell/JsBarcode/wiki/ the various barcode formats EAN13, CODE38, etc.
})
const svgText = xmlSerializer.serializeToString(svgNode)
const buffer = Buffer.from(svgText)
return `data:image/svg+xml;base64,${buffer.toString('base64')}`
}
Complete the HTML file yourname.component.html
< StackLayout >
<SVGImage [src]="getBarcodeSVGSrc(item.barcodenr, item.barcodetype)">
< /StackLayout >

Related

how to add and select color for nodes/tree view items in explorer view in my vscode extension

I have added my own explorer view in my extension.
Here I added nodes/tree view items however I am not finding any way to customize and choose color my tree view items in explorer view.
Any idea how to achieve this?
There should be some way because when some file has error then its color is set to different compared to other open file.
[I assume this is your github issue: Not able to use FileDecorationProvider for tree view item.]
Here is my attempt at using a FileDecorationProvider for a custom TreeView. With the caveat that I am new to typescript and FileDecorations.
If you have seen Support proposed DecorationProvider api on custom views you know there are limitations on using a FileDecorationProvider for coloring TreeItem's - primarily that the decoration/coloration cannot be limited to your treeView - wherever that resourceUri apeears, like in the Explorer, your fileDecoration will be applied. That is very unfortunate but I don't believe there is any way to avoid that for now.
First, in your TreeItem class you will have to give whichever items you want decorated a resourceUri. Like this:
export class TreeTab extends vscode.TreeItem {
constructor( public readonly tab: vscode.Tab, public index: number = 0 ) {
super(tab.label, vscode.TreeItemCollapsibleState.None);
this.tab = tab;
if (tab.input instanceof vscode.TabInputText) {
this.resourceUri = tab.input.uri;
}
}
Ignore the specifics of the code for my extension, the point is:
this.resourceUri = <some vscode.Uri>;
Secondly, this is how I set up my FileDecoration class:
import {window, Tab, TabInputText, Uri, Disposable, Event, EventEmitter, FileDecoration, FileDecorationProvider, ThemeColor} from 'vscode';
export class TreeFileDecorationProvider implements FileDecorationProvider {
private disposables: Array<Disposable> = [];
private readonly _onDidChangeFileDecorations: EventEmitter<Uri | Uri[]> = new EventEmitter< Uri | Uri[]>();
readonly onDidChangeFileDecorations: Event<Uri | Uri[]> = this._onDidChangeFileDecorations.event;
constructor() {
this.disposables = [];
this.disposables.push(window.registerFileDecorationProvider(this));
}
async updateActiveEditor(activeTab: Tab): Promise<void> {
if (activeTab.input instanceof TabInputText)
this._onDidChangeFileDecorations.fire(activeTab.input.uri);
// filter to get only non-activeTabs
activeTab.group.tabs.map( tab => {
if (!tab.isActive && tab.input instanceof TabInputText)
this._onDidChangeFileDecorations.fire(tab.input.uri);
});
}
async provideFileDecoration(uri: Uri): Promise<FileDecoration | undefined> {
const activeEditor = window.activeTextEditor.document.uri;
if (uri.fsPath === activeEditor.fsPath) {
return {
badge: "⇐",
color: new ThemeColor("charts.red"),
// color: new vscode.ThemeColor("tab.activeBackground"),
// tooltip: ""
};
}
else return null; // to get rid of the custom fileDecoration
}
dispose() {
this.disposables.forEach((d) => d.dispose());
}
}
provideFileDecoration(uri: Uri) does the actual decorating. It finds only certain files and decorates them, and by returning null resets that previously decorated uri (as supplied by the uri argument).
updateActiveEditor() is an exported method that I call in other parts of the extension when I want to change a file decoration. So elsewhere I have this in another file:
import { TreeFileDecorationProvider } from './fileDecorator';
export class EditorManager {
public TreeItemDecorator: TreeFileDecorationProvider;
// and then on a listener that gets triggered when I need to make a change to some things including the FileDecoration for a uri
this.TreeItemDecorator.updateActiveEditor(activeTab);
this.TreeItemDecorator.updateActiveEditor(activeTab); that calls the updateActiveEditor method in the TreeFileDecorationProvider class which calls the this._onDidChangeFileDecorations.fire(<some uri>); method for uri's that need to have the decoration applied and also for uri's that need to have the decoration removed.
this._onDidChangeFileDecorations.fire(<some uri>); will call provideFileDecoration(uri: Uri) where the actual decoration will be applied or removed depending on some state of that uri.
I am sure there is a way to call onDidChangeFileDecorations() directly from another file in your project (if you don't need to do any pre-processing of the uri like I have to do. I just haven't figured out how to construct the argument for that function yet. Perhaps someone will help on that point.
You can see here:
color: new ThemeColor("charts.red"),
// color: new vscode.ThemeColor("tab.activeBackground"),
how a color is chosen - it must be some ThemeColor. The charts theme colors has a few basic colors that are handy to refer to. See theme color references, Charts therein.
The badge option can take up to 2 characters, but as you see I copied/pasted a unicode character for mine and that works.
As I mentioned my FileDecorationProvider is called from an eventListener, but you may not need that for your use case - if decorations do not have to added and removed based on user actions like in my case. So you may be able to call your FileDecorationProvider right from your extension.ts activate() like so:
import * as vscode from 'vscode';
import { TreeFileDecorationProvider } from './fileDecorator';
export async function activate(context: vscode.ExtensionContext) {
new TreeFileDecorationProvider();
}
Other references:
a treeDecorationProvider.ts example
part of the git extension that does file decorations
Custom view decorations in VSCode extension

Can I show an image in a HoverProvider?

I am new to vscode extension development so not sure best places to look for help. I want to hover over a word, and if that word matches the name in our icon directory, show a hover with that image. My first try was to use markdown to show the svg image from the web, but I believe that is not allowed for security that code could be executed. Can I show an svg in the markdown in a vscode.Hover? I can either include the svgs in the extension, or expect that the developer has already installed the npm package with the files into the workspace.
There is actually quite rich support for content and styling in MarkdownString in hovers.
You can use svg's directly this way:
const icon = new vscode.MarkdownString('<img src="icon.svg"/>');
icon.baseUri = vscode.Uri.file(path.join(context.extensionPath, 'images', path.sep));
The baseUri property is crucial. Here it gets the svgs from an images folder.
In the below I added the png images to an extension images folder. You can detect what word is hovered and load the corresponding icon.
Below is testing in a plaintext file:
// with this require/import
const path = require('path');
// ...
let disposable4 = vscode.languages.registerHoverProvider('plaintext', {
provideHover(document, position) {
// one way to find the current hovered word
// note that what is a 'word' is languaged-defined and typically does not include hyphens
// const word = document.getText(document.getWordRangeAtPosition(position));
// first check if there is an icon for the hovered word, if not return undefined
// glob the images folder and find 'icon'.svg
// const icon = new vscode.MarkdownString(`<img src="${word}.svg"/>`);
// dimensions not necessary if you aren't changing them
const content = new vscode.MarkdownString(`<img src="favicon144.png" width=144 height=144/>`);
content.appendMarkdown(`$(zap)`); // notice the little "zap" icon in the hover
content.supportHtml = true;
content.isTrusted = true;
content.supportThemeIcons = true; // to supports codicons
// baseUri was necessary, full path in the img src did not work
// with your icons stroed in the 'images' directory
content.baseUri = vscode.Uri.file(path.join(context.extensionPath, 'images', path.sep));
return new vscode.Hover(content, new vscode.Range(position, position));
}
});
There's only very limited Markdown support in hover items and images are not supported there at all, as far as I know.

Express [413 too large] with QuillJS image

I am trying to use QuillJS to let the user write a rich text, and then store it as JSON to display later on. There are 2 of these rich text areas in a single form, and may include images. QuillJS encodes images as base64 strings, and my POST request results in 413 by Express.
I have tried to change the limits by adding express json parameters, even trying extreme numbers.
// app.js
//----------------------------------------------------
// Middlewares
//----------------------------------------------------
app.use(express.json({limit: '2000mb'}));
app.use(express.urlencoded({extended: true, limit:'2000mb'}));
Even this did not help and I think it is not logical to let these parameters with such values.
I tried with json and urlencoded enctypes. When I tried to post with multipart/form, req.body was empty.
// My html page (pugJS)
form(enctype='application/x-www-form-urlencoded', action='/editor/page',
method='POST', onsubmit='return addContent()')
.form-control
label Content-1
div#toolbar
div#editor
input#content(name='content', type='text', hidden)
addContent() function that runs before form submit simply changes input#content's value with JSON.stringify(#editor.getContents())
I want to be able to store two quill content in a single database row, to display later.
A better approach to this would be to overwrite the image upload function and then save the image in Amazon S3 or some cloud server. Then you paste it inside the editor as <img src="http://uploaded-image-url"> This would solve your problem of maximum memory issue.
I fixed my problem few hours before #argo mentioned and I did it that way. So I wanted to post little bit of detail to the solution. I have been also guided by a github issue but can't seem to find the link again, in case I find it I will edit the post and add it.
// Quill - EN content
var quillEn = new Quill('#editor-en', {
modules: {
toolbar: toolbarOptions
},
theme: 'snow'
});
// set custom image handler
quillEn.getModule('toolbar').addHandler('image', () => {
selectLocalImage(quillEn);
});
// create fake input to upload image to quill
function selectLocalImage(editor) {
const input = document.createElement('input');
input.setAttribute('type', 'file');
input.setAttribute('accept', 'image/png, image/jpeg')
input.click();
// Listen upload local image and save to server
input.onchange = () => {
const file = input.files[0];
saveImageToServer(editor, file);
};
}
// upload image to server
function saveImageToServer(editor, file) {
const fd = new FormData();
fd.append('image', file);
const xhr = new XMLHttpRequest();
xhr.open('POST', '/api/page/upload_image', true);
xhr.onload = () => {
if (xhr.status === 200) {
// this is callback data: url
const url = JSON.parse(xhr.responseText).data;
insertToEditor(editor, url);
}
};
xhr.send(fd);
}
// manipulate quill to replace b64 image with uploaded image
function insertToEditor(editor, url) {
// push image url to rich editor.
const range = editor.getSelection();
editor.insertEmbed(range.index, 'image', url.toString());
}
In the backend where you POST image, you must return json as { data: FullUrlToImg } with 200 response, if you want to change your status to 201 or something else, don't forget to update it in saveImageToServer function.
So to summarize, you set custom image handler for your quill editor, you post the image to server as soon as user chooses to insert, then you replace the URL with your uploaded image in the editor.
Thanks.

Editing an entity's decorated text

We have a Figure decorator which allows us to insert a link which you can hover over to get a preview of an image. We insert this image along with some metadata (caption, etc.) using a modal form. This all works great. However we also want the ability to click the link and pop up the modal to edit it.
Entity.replaceData() works great for updating the metadata, the only problem that remains is the decorated text which comes from the modal too. It appears the Entity knows little to nothing about the content it's decorating.
How can we find and replace the text? Is there a way around this?
(I've tried setting the content in Draft to an arbitrary single character and making the decorator show the content/title (which would be fine), however when trying to delete the figure, Draft seems to jump over the content and delete something before it. I guess it's due to different text lengths. I thought setting it as 'IMMUTABLE' would solve this but that didn't help.)
EDIT:
Here's my decorator:
function LinkedImageDecorator(props: Props) {
Entity.mergeData(props.entityKey, { caption: "hello world" });
const entity = Entity.get(props.entityKey);
const data = entity.getData();
return <LinkedImage data={data} text={props.children} offsetKey={props.offsetKey} />
}
function findLinkedImageEntities(contentBlock: ContentBlock, callback: EntityRangeCallback) {
contentBlock.findEntityRanges((character) => {
const entityKey = character.getEntity();
return (
entityKey != null &&
Entity.get(entityKey).getType() === ENTITY_TYPE.IMAGE
);
}, callback);
}
export default {
strategy: findLinkedImageEntities,
component: LinkedImageDecorator,
editable: false,
};
As you can see, I'm testing out Entity.mergeData which will eventually be the callback of my LinkedImage component (which would open a modal onClick.) So the metadata is easy to update, I just need to be able to update the decorated text which is passed in as props.children.
So I finally solved this with the help of Jiang YD and tobiasandersen. Here goes...
First I inject my decorator with a reference to my editor (which keeps track of EditorState):
const decorators = new CompositeDecorator([{
strategy: findLinkedImageEntities,
component: LinkedImageDecorator,
props: { editor: this }
}];
this.editorState = EditorState.set(this.editorState, { decorator });
From there I can do this in my LinkedImageDecorator:
const { decoratedText, children, offsetKey, entityKey, editor } = this.props;
// This looks messy but seems to work fine
const { startOffset, blockKey } = children['0'].props;
const selectionState = SelectionState.createEmpty(blockKey).merge({
anchorOffset: startOffset,
focusOffset: startOffset + decoratedText.length,
});
const editorState = editor.getEditorState();
let newState = Modifier.replaceText(
editorState.getCurrentContent(),
selectionState,
"my new text",
null,
entityKey,
);
editor.editorState = EditorState.push(editorState, newState, 'insert-fragment');
Not sure if this is the cleanest way of doing this but it seems to work well!
just put the global component instance reference in the decorator props.
const compositeDecorator = new CompositeDecorator([
{
strategy: handleStrategy,
component: HandleSpan,
props: {parent:refToYourComponentWhichContainsTheEditorState}
}
props.parent.state.editorState.getCurrentContent()

Create an instance of a React class from a string

I have a string which contains a name of the Class (this is coming from a json file). This string tells my Template Class which layout / template to use for the data (also in json). The issue is my layout is not displaying.
Home.jsx:
//a template or layout.
var Home = React.createClass({
render () {
return (
<div>Home layout</div>
)
}
});
Template.jsx:
var Template = React.createClass({
render: function() {
var Tag = this.props.template; //this is the name of the class eg. 'Home'
return (
<Tag />
);
}
});
I don't get any errors but I also don't see the layout / Home Class. I've checked the props.template and this logs the correct info. Also, I can see the home element in the DOM. However it looks like this:
<div id='template-holder>
<home></home>
</div>
If I change following line to:
var Tag = Home;
//this works but it's not dynamic!
Any ideas, how I can fix this? I'm sure it's either simple fix or I'm doing something stupid. Help would be appreciated. Apologies if this has already been asked (I couldn't find it).
Thanks,
Ewan
This will not work:
var Home = React.createClass({ ... });
var Component = "Home";
React.render(<Component />, ...);
However, this will:
var Home = React.createClass({ ... });
var Component = Home;
React.render(<Component />, ...);
So you simply need to find a way to map between the string "Home" and the component class Home. A simple object will work as a basic registry, and you can build from there if you need more features.
var components = {
"Home": Home,
"Other": OtherComponent
};
var Component = components[this.props.template];
No need to manually map your classes to a dictionary, or "registry", as in Michelle's answer. A wildcard import statement is already a dictionary!
import * as widgets from 'widgets';
const Type = widgets[this.props.template];
...
<Type />
You can make it work with multiple modules by merging all the dictionaries into one:
import * as widgets from 'widgets';
import * as widgets2 from 'widgets2';
const registry = Object.assign({}, widgets, widgets2);
const widget = registry[this.props.template];
I would totally do this to get dynamic dispatch of react components. In fact I think I am in a bunch of projects.
I had the same problem, and found out the solution by myself. I don't know if is the "best pratice" but it works and I'm using it currently in my solution.
You can simply make use of the "evil" eval function to dynamically create an instance of a react component. Something like:
function createComponent(componentName, props, children){
var component = React.createElement(eval(componentName), props, children);
return component;
}
Then, just call it where you want:
var homeComponent = createComponent('Home', [props], [...children]);
If it fits your needs, maybe you can consider something like this.
Hope it helps.
I wanted to know how to create React classes dynamically from a JSON spec loaded from a database and so I did some experimenting and figured it out. My basic idea was that I wanted to define a React app through a GUI instead of typing in code in a text editor.
This is compatible with React 16.3.2. Note React.createClass has been moved into its own module.
Here's condensed version of the essential parts:
import React from 'react'
import ReactDOMServer from 'react-dom/server'
import createReactClass from 'create-react-class'
const spec = {
// getDefaultProps
// getInitialState
// propTypes: { ... }
render () {
return React.createElement('div', null, 'Some text to render')
}
}
const component = createReactClass(spec)
const factory = React.createFactory(component)
const instance = factory({ /* props */ })
const str = ReactDOMServer.renderToStaticMarkup(instance)
console.log(str)
You can see a more complete example here:
https://github.com/brennancheung/02-dynamic-react/blob/master/src/commands/tests/createClass.test.js
Here is the way it will work from a string content without embedding your components as statically linked code into your package, as others have suggested.
import React from 'react';
import { Button } from 'semantic-ui-react';
import createReactClass from 'create-react-class';
export default class Demo extends React.Component {
render() {
const s = "return { render() { return rce('div', null, rce(components['Button'], {content: this.props.propA}), rce(components['Button'], {content: 'hardcoded content'})); } }"
const createComponentSpec = new Function("rce", "components", s);
const componentSpec = createComponentSpec(React.createElement, { "Button": Button });
const component = React.createElement(createReactClass(componentSpec), { propA: "content from property" }, null);
return (
<div>
{component}
</div>
)
}
}
The React class specification is in string s. Note the following:
rce stands for React.createElement and given as a first param when callingcreateComponentSpec.
components is a dictionary of extra component types and given as a second param when callingcreateComponentSpec. This is done so that you can provide components with clashing names.
For example string Button can be resolved to standard HTML button, or button from Semantic UI.
You can easily generate content for s by using https://babeljs.io as described in https://reactjs.org/docs/react-without-jsx.html. Essentially, the string can't contain JSX stuff, and has to be plain JavaScript. That's what BabelJS is doing by translating JSX into JavaScript.
All you need to do is replace React.createElement with rce, and resolve external components via components dictionary (if you don't use external components, that you can skip the dictionary stuff).
Here is equivalent what in the code above. The same <div> with two Semantic UI Buttons in it.
JSX render() code:
function render() {
return (
<div>
<Button content={this.props.propA}/>
<Button content='hardcoded content'/>
</div>
);
}
BabelJS translates it into:
function render() {
return React.createElement("div", null, React.createElement(Button, {
content: this.props.propA
}), React.createElement(Button, {
content: "hardcoded content"
}));
}
And you do replacement as outlined above:
render() { return rce('div', null, rce(components['Button'], {content: this.props.propA}), rce(components['Button'], {content: 'hardcoded content'})); }
Calling createComponentSpec function will create a spec for React class.
Which then converted into actual React class with createReactClass.
And then brought to life with React.createElement.
All you need to do is return it from main component render func.
When you use JSX you can either render HTML tags (strings) or React components (classes).
When you do var Tag = Home, it works because the JSX compiler transforms it to:
var Template = React.createElement(Tag, {});
with the variable Tag in the same scope and being a React class.
var Tag = Home = React.createClass({
render () {
return (
<div>Home layout</div>
)
}
});
When you do
var Tag = this.props.template; // example: Tag = "aClassName"
you are doing
var Template = React.createElement("aClassName", null);
But "aClassName" is not a valid HTML tag.
Look here