Tag: WebDev

So i got this code inside Mapview for mapmarkers. This error only happens in android devices. <MapView key={darkMode ? 'dark' : 'light'} ref={mapRef} style={{ flex: 1 }} initialRegion={mapRegion} onRegionChange={() => { isUserInteractingRef.current = true; }} onRegionChangeComplete={(region) => { isUserInteractingRef.current = false; handleRegionChangeComplete(region); }} customMapStyle={darkMode ? darkMapStyle : undefined} moveOnMarkerPress={false} rotateEnabled={false} pitchEnabled={false} scrollEnabled={true} // Always enabled zoomEnabled={true} // Always enabled loadingEnabled={false} loadingIndicatorColor={tint} loadingBackgroundColor={background} showsBuildings={true} showsIndoors={true} onMapReady={() => { console.log('Map is ready'); // Give map time to initialize properly setTimeout(() => { if (mapRegion) { mapRef.current?.animateToRegion(mapRegion, 0); }

input: { "request": { "jobData": { "jobId": "2736" } }, "results": [ { "jobReqId": "1234", "applicationId": "1987" }, { "jobReqId": "2736", "applicationId": "1181" }, { "jobReqId": "7879", "applicationId": "1280" } ] } output: { "applicationId": "1181", "applied": "true" } Summary: The values of jobId, jobReqId, and applicationId are dynamic. Compare the request.jobData.jobId value with each jobReqId in the res

I have two overlapping rectangles: The red one (higher z) has a DragHandler with dragThreshold: 0. The blue one (lower z) has a PointHandler. When I click the overlapped area, I expect only the red rectangle's handler to respond, but the blue PointHandler activates instead. (If I drag, then both are active simultaneously). import QtQuick Window { width: 300; height: 300; visible: true Rectangle { width: 100; height: 100; color: "blue" PointHandler { onActiveChanged: if (active) print("blue PointHandler activated") } } Rectangle { x: 50; y: 50; z: 1; width: 100; height: 100; color: "red" DragHandler { target: null; dragThreshold: 0 } } } Setting grabPermissions: PointerHandler.CanTakeOverFr

I'm using HTML tag to create an accordion with multiple tabs. I'm also using Javascript with Web Animation API (WAAPI) to animate the height to open/close each tab. The issue is, I can't find the proper way to close the open active tab when opening a new one. Right now when you open a new tab, the previous one still keeps open. class Accordion { constructor(el) { this.el = el; this.summary = el.querySelector('summary'); this.content = el.querySelector('.content'); this.animation = null; this.isClosing = false; this.isExpanding = false; this.summary.addEventListener('click', (e) => this.onClick(e)); } onClick(e) { e.preventDefault(); this.el.style.overflow = 'hidden'; if (this.

I have multiple web applications hosted on different subdomains of the same parent domain: ONE.abc.ai TWO.abc.ai THREE.abc.ai Each application has its own login form (email + password). Problem When I log in to ONE.abc.ai and save the credentials in the browser, those credentials appear as autocomplete / password suggestions when I focus the login fields on TWO.abc.ai or THREE.abc.ai. I want to prevent passwords saved for one subdomain from appearing as suggestions on the other subdomains. What I’ve tried Setting autocomplete="off" on the form and inputs Using different name attributes for email/password fields Using: autocomplete="new-password" Verified that each app has its own backend and login logic None of these reliably prevent the browser from showing saved credentials from other subdomains. Constraints / Notes

I know Lucy is based on Swing and the problem should be reframed as "How to make e2e test for a Swing app". However, below are the differences between my programme and a general Swing app. Since Lucy Framework is primarily an OSGi framework, I tried researching into the ways we perform tests for OSGi frameworks in general. However, it seems to me that modularity implies testing separately. What I want is a way to test e2e fully within an OSGi framework. May I know if there's a way/tool to achieve that? Constraints: I'm stuck with Swing for sure. No negotiations on this. It's an OSGi framework which means jar files are loaded on demand. The main app can launch another Swing/JavaFX window. This affects frameworks like AssertJ-Swing, where it is assumed to test only 1 main class. I tried looking into blackbox testing using WinAppDriver and ins

I am working on a product where our APIs evolve pretty quickly. New fields get added, response shapes adjust, and validation rules shift as the domain model matures. Automation coverage is decent at the contract and integration level, but the friction shows up in how we structure and manage the test cases themselves. When tests are too granular, even small schema changes create a lot of maintenance churn. When they are too high level, we miss edge cases and validation gaps. I am trying to find a structure that keeps tests valuable without becoming brittle over time. A few specific areas I would love input on: Do you organize API tests by endpoint, business workflow, or risk area? How do you handle optional or frequently changing fields in assertions? Are you relying on schema validation, custom matchers, or snapshot style comparisons? When does a response change fail a test versus get treated as

I'm using PolarDB for PostgreSQL and experiencing a query performance issue where the planner severely underestimates the result size of a JOIN operation. CREATE TABLE orders ( order_id BIGINT PRIMARY KEY, customer_id INT, order_date DATE, status VARCHAR(20) ); CREATE TABLE order_items ( item_id BIGINT PRIMARY KEY, order_id BIGINT, product_id INT, quantity INT, price DECIMAL(10,2) ); CREATE INDEX idx_orders_customer_date ON orders(customer_id, order_date); CREATE INDEX idx_order_items_order_id ON order_items(order_id); CREATE INDEX idx_order_items_product_id ON order_items(product_id); -- orders: ~2M rows -- order_items: ~15M rows Query: SELECT o.order_id, o.customer_id, oi.product_id, oi.quantity FROM orders o JOIN order_items oi ON o.order_id = oi.order_id WHERE o.order_date >= '2024-01-01' AND o.order_date < '2024-07-01' AND o.status = 'completed' AND oi.product_id I

I have two buttons. Open btn opens div element and close btn closes the div element. But as I found out if I touch the close btn the browser triggers the touchend event but emediatelly it triggers also the open event. I dont understand it. I have e.preventDefault() and e.stopPropagation(). The code looks like. Open btn // SET UP EVENT LISTENERS FOR OPEN BTN AND CLOSE BTN // TOUCHSTART IS USED TO AVOID COLLISION BETWEEN OPEN BTN AND CLOSE BTN WHICH LISTEN ON TOUCHEND ['click', 'touchstart'].forEach(function (event) { if ( self.config_activation_btn ) { self.html_open_btn.addEventListener(event, function (e) { __openCustomApp(false, self.config_items[0].source); }, {'signal': self.open_event_abort_controller.signal}); } else if ( self.config_activation_touch ) { document.body.addEventListener(event, function (e) {

The one definition rule in the C++20 standard states that there can be no two definitions of a class type in a program unless the two definitions are in different translation units and have the same sequence of tokens (6.3/13.8) (plus some other requirements). The requirement does not mention scope or linkage. At face value, this would exclude code like this if a.cpp, b.cpp and c.cpp were part of the same program: a.cpp class T{}; b.cpp namespace N { class T{int i;}; } c.cpp void f() { class T{float f;}; } I'm quite sure that the example is fine; how could I justify that with the standard? I know that "a declaration of a name in a nested declarative region hides a declaration of the same name in an enclosing declarative region" (6.4.10); but visibil