So basically, when you run Vulkan, it kinda “takes over” the window. Think of it like Vulkan puts its own TV screen inside your game window and says “okay, I’m in charge of showing stuff here now.”

When you switch to DirectX, you’re telling Vulkan “alright, you can leave now.” Vulkan packs up its things and leaves… but the problem is, it forgets to actually take its TV screen out of the window. So Windows is still showing that last frame Vulkan left behind, like a paused YouTube video.

Meanwhile, DirectX is there, yelling “hey, I’m drawing stuff!” — but Windows ignores it, because it still thinks Vulkan owns the window. That’s why you just see the frozen Vulkan image.

The fix is basically making sure Vulkan really leaves before DirectX moves in. That means:

• Wait until Vulkan is 100% done drawing before shutting it down.

• Make sure you actually destroy all the stuff Vulkan made for the window (its swapchain, framebuffers, images, etc).

• Sometimes you even need to “nudge” Windows to refresh the window (like forcing a redraw), so it stops showing the frozen Vulkan picture.

So in short: Vulkan isn’t secretly still running — it just forgot to give the window back to Windows. DirectX is drawing, but Windows isn’t letting it through until Vulkan fully hands over the keys.

Firebase Craslytcs does not run very easily on maui dotnet9 depending on the project context, many developers can use it with maui dotnet9, however for my context it does not work either, try with Sentry smooth implementation with compatibility, it ran very easily https://docs.sentry.io/platforms/dotnet/guides/maui/

As it turns out, the problem was not NextCloud. Using this tutorial I implemented a working login flow using only the `requests` package. The code for that is below. It is not yet handling performing any kind of API request using the obtained access token beyond the initial authentication, nor is it handling using the refresh token to get a new access token when the old one expired. That is functionality an oauth library is usually handling and this manual implementation is not doing that for now. However it proves the problem isn't with NextCloud.

I stepped through both the initial authlib implementation and the new with a debugger and the request sent to the NextCloud API for getting the access token looks the same in both cases at first glance. There must be something subtly wrong about the request in the authlib case that causes the API to run into an error. I will investigate this further and take this bug up with authlib. This question here is answered and if there is a bug fix in authlib I will edit the answer to mention which version fixes it.

from __future__ import annotations

from pathlib import Path
import io
import uuid
from urllib.parse import urlencode
import requests
from flask import Flask, render_template, jsonify, request, session, url_for, redirect
from flask_session import Session

app = Flask("webapp")

# app.config is set here, specifically settings:
# NEXTCLOUD_CLIENT_ID
# NEXTCLOUD_SECRET
# NEXTCLOUD_API_BASE_URL
# NEXTCLOUD_AUTHORIZE_URL
# NEXTCLOUD_ACCESS_TOKEN_URL

# set session to be managed server-side
Session(app)


@app.route("/", methods=["GET"])
def index():
    if "user_id" not in session:
        session["user_id"] = "__anonymous__"
        session["nextcloud_authorized"] = False
    return render_template("index.html", session=session), 200

@app.route("/nextcloud_login", methods=["GET"])
def nextcloud_login():
    if "nextcloud_authorized" in session and session["nextcloud_authorized"]:
        redirect(url_for("index"))

    session['nextcloud_login_state'] = str(uuid.uuid4())

    qs = urlencode({
        'client_id': app.config['NEXTCLOUD_CLIENT_ID'],
        'redirect_uri': url_for('callback_nextcloud', _external=True),
        'response_type': 'code',
        'scope': "",
        'state': session['nextcloud_login_state'],
    })

    return redirect(app.config['NEXTCLOUD_AUTHORIZE_URL'] + '?' + qs)

@app.route('/callback/nextcloud', methods=["GET"])
def callback_nextcloud():
    if "nextcloud_authorized" in session and session["nextcloud_authorized"]:
        redirect(url_for("index"))

    # if the callback request from NextCloud has an error, we might catch this here, however
    # it is not clear how errors are presented in the request for the callback
    # if "error" in request.args:
    #     return jsonify({"error": "NextCloud callback has errors"}), 400

    if request.args["state"] != session["nextcloud_login_state"]:
        return jsonify({"error": "CSRF warning! Request states do not match."}), 403

    if "code" not in request.args or request.args["code"] == "":
        return jsonify({"error": "Did not receive valid code in NextCloud callback"}), 400

    response = requests.post(
        app.config['NEXTCLOUD_ACCESS_TOKEN_URL'],
        data={
            'client_id': app.config['NEXTCLOUD_CLIENT_ID'],
            'client_secret': app.config['NEXTCLOUD_SECRET'],
            'code': request.args['code'],
            'grant_type': 'authorization_code',
            'redirect_uri': url_for('callback_nextcloud', _external=True),
        },
        headers={'Accept': 'application/json'},
        timeout=10
    )

    if response.status_code != 200:
        return jsonify({"error": "Invalid response while fetching access token"}), 400

    response_data = response.json()
    access_token = response_data.get('access_token')
    if not access_token:
        return jsonify({"error": "Could not find access token in response"}), 400

    refresh_token = response_data.get('refresh_token')
    if not refresh_token:
        return jsonify({"error": "Could not find refresh token in response"}), 400

    session["nextcloud_access_token"] = access_token
    session["nextcloud_refresh_token"] = refresh_token
    session["nextcloud_authorized"] = True
    session["user_id"] = response_data.get("user_id")

    return redirect(url_for("index"))

Starting with Android 12 (API 31), splash screens are handled by the SplashScreen API. Flutter Native Splash generates the correct drawable for android:windowSplashScreenAnimatedIcon, but Android caches the splash drawable only after the first run. So, if the generated resource is too large, not in the right format, or not properly referenced in your theme, Android falls back to background color on first launch.

I am not sure if you have resolved this but what you may facing is DynamoDB read consistency issue, I had the similar issue.

https://docs.aws.amazon.com/amazondynamodb/latest/developerguide/HowItWorks.ReadConsistency.html

I am also struggling to set "de" as the keyboard layout on ubuntu core. I am using ubuntu-frame along with chromium kiosk for my ui. In your example, I would recommend building your own snap, which serves as a wrapper script that runs the firefox browser. With the flags daemon set to simple and restart-always set inside your snapcaft.yaml file, it should at least come up again after the user closed it.

As simple as this?
Application.bringToFront;

Works for me (Windows 10)

I've fixed with the following:

Added app.UseStatusCodePagesWithRedirects("/error-page/{0}"); to the Program.cs.

Added the page CustomErrorPage.razor with the following content:

@page "/error-page/{StatusCode:int}"

<div>content</div>

@code {
    [Parameter]
    public int StatusCode { get; set; }

    public bool Is404 => StatusCode == 404;
    public string Heading => Is404 ? "Page not found 404" : $"Error {StatusCode}";
}

ElastiCache supports Bloom filters with Valkey 8.1, which is compatible with Redis OSS 7.2. You can see https://docs.aws.amazon.com/AmazonElastiCache/latest/dg/BloomFilters.html for more information.

Olá, se estiver usando algum programa de backup em nuvem desative ele na hora de compilar.

mailto:[email protected],[email protected],[email protected]&cc=...

All other examples did not work for me. This one seems to work.

As of August 2025, Visual Studio 2017 community edition can be downloaded from this link https://aka.ms/vs/15/release/vs_community.exe without login in to a subscription.

Also, the professional version can be downloaded here https://aka.ms/vs/15/release/vs_professional.exe

I got this error with django (AlterUniqueTogether) and mariadb when adding unique_together={('field1', 'field2')} constraint, where field2 was varchar(1000). Size of that field (1000x4) was too big for max index key length of 3072 bytes. field1 was fk so somehow i was getting that error and spent lot of time debugging it.

Other types and usages are:

Built-in Type Usage

• Opaque arbitrary user-defined data

• kubernetes.io/service-account-token ServiceAccount token

• kubernetes.io/dockercfg serialized ~/.dockercfg file

• kubernetes.io/dockerconfigjson serialized ~/.docker/config.json file

• kubernetes.io/basic-auth credentials for basic authentication

• kubernetes.io/ssh-auth credentials for SSH authentication

• kubernetes.io/tls data for a TLS client or server

• bootstrap.kubernetes.io/token bootstrap token data

Source: https://kubernetes.io/docs/concepts/configuration/secret/#secret-types

I know it's an old question, however have the impression it's still an issue ...
If I understood correctly C23 provides "bigint" types or similar, however not
all users are on C23, not all like it and it's still without I/O support?
I puzzled together a little library, vanilla-C, header only, which provides
I/O from / to binary, octal, decimal and hex strings and some other functions
for gcc builtin 128-bit integer types: Libint128, I/O for 128-bit intger types

I found a cool video about it: https://shre.su/YJ85

It cant also happen due to broken text indexes. Please try invalidating the caches (File > Invalidate caches, Invalidate and Restart).

https://intellij-support.jetbrains.com/hc/en-us/community/posts/16256641441938/comments/16259399023122

If Outline does not work
Go to the base file from where you want the path in the terminal or PowerShell, then run this command

tree /A /F > structure.txt

This will generate a txt file named structure.txt in that base location

I had this in my schema.prisma generator client instead of removing it.

  output   = "../app/generated/prisma"

I change the below like this. 

import { PrismaClient } from '@prisma/client'
import { PrismaClient } from "../../generated/prisma";

<security:http name="apis" 
    pattern="/**"

is also matched?
try commenting out this part...

maybe ..change your request path or exclude these two paths when verifying the token?

<security:http name="employeeSecurityChain" 
    pattern="/auth/employees/token"

<security:http name="visitorSecurityChain" 
    pattern="/auth/visitors/token"

<security:http name="apis" 
    pattern="/api/**"

interface PageProps {
  params: Promise<{ id: string }>;
}

const Page = async ({ params }: PageProps) => {
  const { id } = await params;

  return <div>Page for {id}</div>;
};

export default Page;

does not work when i run build

Not sure about the third party's audit verdict. But imp, you should verify their suggestion. Perhaps its not required at all. The configuration file should contain the runtime and other configuration related values.

However, if you insist deleting these from your config file, you can do so. But keep in mind that -

• On .NET 4.0+ apps, it defaults to v4.0 runtime (then rolls forward to whatever 4.x is installed).

• On .NET 2.0/3.5 apps, it defaults to v2.0 CLR.

This usually works fine unless your app was explicitly depending on a particular runtime behavior.

You need to listen to autoUpdater events, example:

import { autoUpdater } from 'electron';

...

autoUpdater.on('update-downloaded', (event, releaseNotes, releaseName) => {
  const dialogOpts = {
    type: 'info',
    buttons: ['Restart', 'Later'],
    title: 'Application Update',
    message: process.platform === 'win32' ? releaseNotes : releaseName,
    detail:
      'A new version has been downloaded. Restart the application to apply the updates.'
  }

  dialog.showMessageBox(dialogOpts).then((returnValue) => {
    if (returnValue.response === 0) autoUpdater.quitAndInstall()
  })
})

More in this tutorial.

You probably should do it outside of the _checkForUpdates function, so you don't attach multiple listeners to one event.

Use https://pub.dev/packages/json_factory_generator

import 'package:flutter/material.dart';
import 'generated/json_factory.dart'; // Contains generated JsonFactory

void main() {
  // No initialization needed! 🎉
  runApp(const MyApp());
}

// Parse single objects
final user = JsonFactory.fromJson<User>({"id": 1, "name": "Alice"});

// Parse lists with proper typing
final posts = JsonFactory.fromJson<List<Post>>([
  {"id": 10, "title": "Hello", "content": "Content"},
  {"id": 11, "title": "World", "content": "More content"},
]);

// Type-safe list parsing
final userList = JsonFactory.fromJson<List<User>>([
  {"id": 1, "name": "Alice"},
  {"id": 2, "name": "Bob"}
]);

You can solve it by putting all of them into different partitions of a single module. e.g., M:A 、 M:B 、M:C。

• • Connected Redmi Note 14 for testing

  • Emulator performance lags behind an older Redmi Note 12

  • Tried changing emulator RAM/graphics, updated SDK tools and USB drivers

  • No major improvement

• Expectation: Better performance on the newer device

• Meta information:

  • Tags: androidperformance, android-emulator

Does this help https://docs.pytorch.org/data/0.7/generated/torchdata.datapipes.iter.ParquetDataFrameLoader.html

Pytorch used to have a torchtext library but it has been deprecated for over a year. You can check it here: https://docs.pytorch.org/text/stable/index.html

Otherwise, your best bet is to subclass one of the base dataset classes https://github.com/pytorch/pytorch/blob/main/torch/utils/data/dataset.py

Here is an example attempt at doing just that https://discuss.pytorch.org/t/efficient-tabular-data-loading-from-parquet-files-in-gcs/160322

My issue was simply that I was using a program to compile all of my docker-compose files into 1. This program only kept the "essential" parts and didn't keep the command: --config /etc/otel/config.yaml part of my otel-collector so the config wasn't being loaded properly into the collector.

I’m facing the same issue, and setting "extends": null didn’t solve it for me either. I created the app using Create React App (CRA). When I run npm run dist, everything builds correctly, but when I execute myapp.exe, I get the error: enter image description here

Can someone help me figure out what’s going wrong?

My package.json is:

{

 (...)

 "main": "main.js",

 (...)

 "scripts": {

    (...)

    "start:electron": "electron .",
    "dist": "electron-builder"
  }

  (...)

  "build": {
    "extends":null,
    "appId": "com.name.app",
    "files": [
      "build/**/*",
      "main.js",
      "backend/**/*",
      "node_modules/**/*"
    ],
     "directories": {
      "buildResources": "public",
      "output": "dist"
    },
     },
    "win": {
      "icon": "public/iconos/logoAntea.png",
      "target": "nsis"
    },
    "nsis": {
      "oneClick": false,
      "allowToChangeInstallationDirectory": true,
      "perMachine": true,
      "createDesktopShortcut": true,
      "createStartMenuShortcut": true,
      "shortcutName": "Datos Moviles",
      "uninstallDisplayName": "Datos Moviles",
      "include": "nsis-config.nsh"
    }
  }
} 

I know a lot of time has passed since this problem was discussed, however, I got the same error with WPF today. It turned out that when I set DialogResult twice, I got this error on the second setting. DialogResult did not behave like a storage location that one sets values to multiple time. The error message that results is very misleading. A similar situation was discussed in this chain of answers, however, in my case I was setting DialogResult to "true" both times, to the same value.

Adding to Asclepius's answer here a way to view the commit history up to the common ancestor (including it).

I find this helpful to see what has been going on since the fork.

$ git checkout feature-branch
$ git log HEAD...$(git merge-base --fork-point master)~1

To use the latest stable version, run:

fvm use stable --pin

Source: FVM | Commands Reference | use | Options

I found the answer by fiddling around. If anyone is interested:

I had to hover over the link in my Initiator column to retrieve the full stack trace, then right click on zone.js and Add script to ignore list.

Since TailwindCSS generates the expected CSS code perfectly, the issue is that the expected CSS code itself does not work properly. The expected CSS code is correct:

Input

<div class="cursor-pointer">...</div>

Generated CSS (check yourself: Playground)

.cursor-pointer {
  cursor: pointer;
}

Since the syntax is correct and other overrides can be ruled out, the only remaining explanation is: a browser bug.

Some external sources mentioning a similar browser bug in Safari:

• https://discussions.apple.com/thread/250283073?sortBy=rank
• https://www.reddit.com/r/osx/comments/b9fc3x/safari_cursor_not_changing_to_a_hand
• https://forums.macrumors.com/threads/mouse-pointer-not-changing-to-hand-or-text-cursor-macos-14-2.2415236/
• https://www.reddit.com/r/webdev/comments/1fj36kf/comment/mnrz3ad (latest? 4 months ago)

Answering my own question. Adding the org.freedesktop.DBus.Properties interface to my xml did not work, as the QDbusAbstractorAdaptor or anyone else is already implementing theses methods. But the signal will not be emitted. At least I did not succeed in finding a "official" way.

But I found a workaround which work for me: https://randomguy3.wordpress.com/2010/09/07/the-magic-of-qtdbus-and-the-propertychanged-signal/

My Adaptor parent class is using the setProperty and property functions of QObject.
Overloaded the setProperty function, calling the QOBject ones and as an addition emitted the PropertiesChanged signal manually like this:

    QDBusMessage signal = QDBusMessage::createSignal(
    "/my/object/path",
    "org.freedesktop.DBus.Properties",
    "PropertiesChanged");
    signal << "my.inter.face";

    QVariantMap changedProps;
    changedProps.insert(thePropertyName, thePropertyValue);
    signal << changedProps;

    QStringList invalidatedProps;
    signal << invalidatedProps;

    QDBusConnection::systemBus().send(signal);

Not a very nice way, but at least the signal is emitted.
Anyway I would be interessted in an more official way of doing it....

Cheers
Thilo

Django has PermissionRequiredMixin which can be derived for each View. PermissionRequired mixin has class property "permission_required". So you can individually define required permission for each view. Also you can tie users to permission groups and assign multiple permissions for each group.

https://docs.djangoproject.com/en/5.2/topics/auth/default/#the-permissionrequiredmixin-mixin

https://docs.djangoproject.com/en/5.2/topics/auth/default/#groups

I found this issue too but this is entirely different level issue and totally my careless mistake

I changed the IP of the machine. Then tried to connect using ssms.

Turns out, i forgot to change the IP too in TCP/IP protocols in SQL Server Network Config, but the locked out login error was really misleading for my case.

Just in case anyone did the same and didnt check. I almost created new admin acc just for that.

sudo apt install nvidia-cuda-dev

The "Test Connection" in the Glue Console only verifies network connectivity, not whether the SSL certificate is trusted during job runtime.

The actual job runtime uses a separate JVM where the certificate must be available and trusted. If AWS Glue can’t validate the server certificate chain during the job run, it throws the PKIX path building failed error.

This typically happens when:

The SAP OData SSL certificate is self-signed or issued by a private CA.

The certificate isn’t properly loaded at runtime for the job to trust it.

✅ What You’ve Done (Good Steps):

You're already trying to add the certificate using:

"JdbcEnforceSsl": "true",

"CustomJdbcCert": "s3://{bucket}/cert/{cert}"

✅ That’s correct — this tells AWS Glue to load a custom certificate.

📌 What to Check / Do Next:

1. Certificate Format

Make sure the certificate is in PEM format (.crt or .pem), not DER or PFX.

2. Certificate Path in S3

Ensure the file exists at the correct path and is publicly readable by the Glue job (via IAM role).

Example:

s3://your-bucket-name/cert/sap_server.crt

3. Permissions

The Glue job role must have permission to read the certificate from S3. Add this to the role policy:

{

"Effect": "Allow",

"Action": "s3:GetObject",

"Resource": "arn:aws:s3:::your-bucket-name/cert/*"

}

4. Recheck Key Option Names

Make sure you didn’t misspell any keys like CustomJdbcCert or JdbcEnforceSsl. They are case-sensitive.

5. Glue Version Compatibility

If using Glue 3.0 or earlier, try upgrading to Glue 4.0, which has better support for custom JDBC certificate handling.

6. Restart Job after Changes

After uploading or changing the certificate, restart the job — don’t rely on retries alone.

I had this problem when I had the expected type in a file named alma.d.ts in a folder that also contained a regular alma.ts file. When I renamed the alma.ts file the error went away.

Go to Run -> Edit Configurations -> Additional options -> Check "Emulate terminal in the output console"

KeyStore Explorer (https://keystore-explorer.org/) could be used to extract the private key into a PEM file.

1. Open the certificate PFX file that contains the public and private key.

2. Right-click on the entry in KeyStore Explorer and select the Export | Export Private Key

3. Select OpenSSL from the list

4. Unselect the Encrypt option and choose the location to save the PEM file.

Did you find a solution yet? I had the same problem and still try to figure out. Which version of spark do you have?

A simple workaround is to transform the geometry column to wkt or wkb and drop the geometry column.

In case of reading you have to tranform it back. Its not nice but functional.

df = df.withColumn("geometry_wkt", expr("ST_AsText(geometry)"))

You could use <iframe>` to load the websites and animate them with css transition or @keyframes

See: https://www.w3schools.com/tags/tag_iframe.asp and https://www.w3schools.com/cssref/css3_pr_transition.php or https://www.w3schools.com/cssref/atrule_keyframes.php

The Places Details API only returns up to 5 reviews for a place. That limit is hard and there is no pagination for the reviews array. The next_page_token you are checking applies to paginated search results, not to reviews in a Place Details response. To fetch all reviews for your own verified business, you must use the Google Business Profile API’s accounts.locations.reviews.list, which supports pagination.

https://developers.google.com/maps/documentation/places/web-service/reference/rest/v1/places#Review:~:text=List%20of%20reviews%20about%20this%20place%2C%20sorted%20by%20relevance.%20A%20maximum%20of%205%20reviews%20can%20be%20returned.

I guess you need to install their Code Coverage plugin too:
https://bitbucket.org/atlassian/bitbucket-code-coverage

https://nextjs.org/docs/app/api-reference/functions/redirect

I'm new to NextJS myself, but maybe something like this could work? Maybe perform the request whenever the request is triggered and await the response and use the function accordingly?

For some file formats like `flac` pydub requires ffmpeg to be installed. And it throws this error when ffmpeg is not found.

Access via window.ZOHO

In the script, where you have used ZOHO will not work directly, as the SDKs will not be supported, to use it! make the zoho library global and use window.ZOHO

In your script, just replace ZOHO with window.ZOHO

In your vs code go to settings, search for javascript.validation and uncheck the checkbox
close and reopen your vs code, if required.

From AWS WEB console -

And the link to create the repository after the latest changes.

Create a Repository in AWS CodeCommit

close android studio

open via command

open -a "Android Studio"

linux: Pulling from library/hello-world

198f93fd5094: Retrying in 1 second

error pulling image configuration: download failed after attempts=6: dialing docker-images-prod.6aa30f8b08e16409b46e0173d6de2f56.r2.cloudflarestorage.com:443 container via direct connection because disabled has no HTTPS proxy: connecting to docker-images-prod.6aa30f8b08e16409b46e0173d6de2f56.r2.cloudflarestorage.com:443: dial tcp: lookup docker-images-prod.6aa30f8b08e16409b46e0173d6de2f56.r2.cloudflarestorage.com: no such host

I've solved similar problem by editing nginx.conf:

sudo nano /etc/nginx/nginx.conf

then change 'user www-data' to 'user sudo_user' where sudo_user it's your configured sudo user.

Do this simply

input {
    field-sizing: content;
    text-align: center;
    min-width: 25%;
}

from typing import get_origin, get_args
    
origin = get_origin(klass)
args = get_args(klass)

if origin is list and args:
    return _func1(data, args[0])
elif origin is dict and len(args) == 2:
    return _func2(data, args[1])

                messagebox.showerror(
                    "Ruta requerida",
                    "Debes indicar una ruta completa. Usa 'Examinar...' o escribe una ruta absoluta (por ejemplo, C:\\carpeta\\archivo.txt)."
                )
                return
            # Evitar que se indique una carpeta como archivo
            if os.path.isdir(archivo_path):
                messagebox.showerror(
                    "Error",
                    "La ruta indicada es una carpeta. Especifica un archivo (por ejemplo, datos.txt)."
                )
                return
            # Verificar/crear carpeta
            try:
                dir_path = os.path.dirname(os.path.abspath(archivo_path))
            except (OSError, ValueError):
                messagebox.showerror("Error", "La ruta del archivo destino no es válida")
                return
            if dir_path and not os.path.exists(dir_path):
                crear = messagebox.askyesno(
                    "Crear carpeta",
                    f"La carpeta no existe:\n{dir_path}\n\n¿Deseas crearla?"
                )
                if crear:
                    try:
                        os.makedirs(dir_path, exist_ok=True)
                    except OSError as e:
                        messagebox.showerror("Error", f"No se pudo crear la carpeta:\n{e}")
                        return
                else:
                    return

            self._mostrar_progreso_gen()

            header = (
                "ID|Nombre|Email|Edad|Salario|FechaNacimiento|Activo|Codigo|Telefono|Puntuacion|Categoria|Comentarios\n"
            )
            with open(archivo_path, 'w', encoding='utf-8') as f:
                f.write(header)
                tamano_actual = len(header.encode('utf-8'))
                rid = 1
                while tamano_actual < tamano_objetivo_bytes:
                    linea = self._generar_registro_aleatorio(rid)
                    f.write(linea)
                    tamano_actual += len(linea.encode('utf-8'))
                    rid += 1
                    if rid % 1000 == 0:
                        # Actualización periódica del progreso para no saturar la UI
                        try:
                            if self.root.winfo_exists():
                                progreso = min(100, (tamano_actual / tamano_objetivo_bytes) * 100)
                                self.progress['value'] = progreso
                                self.estado_label.config(
                                    text=f"Registros... {rid:,} registros ({progreso:.1f}%)")
                                self.root.update()
                        except tk.TclError:
                            break

            tamano_real_bytes = os.path.getsize(archivo_path)
            tamano_real_mb = tamano_real_bytes / (1024 * 1024)

            try:
                if self.root.winfo_exists():
                    self.progress['value'] = 100
                    self.estado_label.config(text="¡Archivo generado exitosamente!", fg='#4CAF50')
                    self.root.update()
            except tk.TclError:
                pass

            abrir = messagebox.askyesno(
                "Archivo Generado",
                "Archivo creado exitosamente:\n\n"
                f"Ruta: {archivo_path}\n"
                f"Tamaño objetivo: {tamano_objetivo_mb:,.1f} MB\n"
                f"Tamaño real: {tamano_real_mb:.1f} MB\n"
                f"Registros generados: {rid-1:,}\n\n"
                "¿Deseas abrir la carpeta donde se guardó el archivo?"
            )
            if abrir:
                try:
                    destino = os.path.abspath(archivo_path)
                    # Abrir Explorer seleccionando el archivo generado
                    subprocess.run(['explorer', '/select,', destino], check=True)
                except (OSError, subprocess.CalledProcessError) as e:
                    print(f"No se pudo abrir Explorer: {e}")

            try:
                if self.root.winfo_exists():
                    self.root.after(3000, self._ocultar_progreso_gen)
            except tk.TclError:
                pass

        except (IOError, OSError, ValueError) as e:
            messagebox.showerror("❌ Error", f"Error al generar el archivo:\n{str(e)}")
            try:
                if self.root.winfo_exists():
                    self.estado_label.config(text="❌ Error en la generación", fg='red')
                    self.root.after(2000, self._ocultar_progreso_gen)
            except tk.TclError:
                pass

    # ------------------------------
    # Lógica: División de archivo
    # ------------------------------
    def _dividir_archivo(self):
        """Divide un archivo en múltiples partes respetando líneas completas.

        Reglas y comportamiento:
            - El tamaño máximo de cada parte se define en "Tamaño por parte (MB)".
            - No corta líneas: si una línea no cabe en la parte actual y ésta ya tiene
              contenido, se inicia una nueva parte y se escribe allí la línea completa.
            - Los nombres de salida se forman como: <base>_NN<ext> (NN con 2 dígitos).

        Manejo de errores:
            - Valida ruta de origen, tamaño de parte y tamaño > 0 del archivo.
            - Muestra mensajes de error/aviso según corresponda.
        """
        try:
            src = self.split_source_file.get()
            if not src or not os.path.isfile(src):
                messagebox.showerror("Error", "Selecciona un archivo origen válido")
                return

            part_size_mb = self.split_size_mb.get()
            if part_size_mb <= 0:
                messagebox.showerror("Error", "El tamaño por parte debe ser mayor a 0")
                return
            part_size_bytes = int(part_size_mb * 1024 * 1024)

            total_bytes = os.path.getsize(src)
            if total_bytes == 0:
                messagebox.showwarning("Aviso", "El archivo está vacío")
                return

            self._mostrar_progreso_split()

            base, ext = os.path.splitext(src)
            part_idx = 1
            bytes_procesados = 0
            bytes_en_parte = 0
            out = None

            def abrir_nueva_parte(idx: int):
                nonlocal out, bytes_en_parte
                if out:
                    out.close()
                nombre = f"{base}_{idx:02d}{ext}"
                out = open(nombre, 'wb')  # escritura binaria
                bytes_en_parte = 0

            abrir_nueva_parte(part_idx)

            line_count = 0
            with open(src, 'rb') as fin:  # lectura binaria
                for linea in fin:
                    lb = len(linea)

                    # Si excede y ya escribimos algo, nueva parte
                    if bytes_en_parte > 0 and bytes_en_parte + lb > part_size_bytes:
                        part_idx += 1
                        abrir_nueva_parte(part_idx)

                    # Escribimos la línea completa
                    out.write(linea)
                    bytes_en_parte += lb
                    bytes_procesados += lb
                    line_count += 1

                    # Actualizar progreso cada 1000 líneas
                    if line_count % 1000 == 0:
                        try:
                            if self.root.winfo_exists():
                                progreso = min(100, (bytes_procesados / total_bytes) * 100)
                                self.split_progress['value'] = progreso
                                self.split_estado_label.config(
                                    text=f"Procesando... {line_count:,} líneas ({progreso:.1f}%)")
                                self.root.update()
                        except tk.TclError:
                            break

            if out:
                out.close()

            try:
                if self.root.winfo_exists():
                    self.split_progress['value'] = 100
                    self.split_estado_label.config(text="¡Archivo dividido exitosamente!", fg='#4CAF50')
                    self.root.update()
            except tk.TclError:
                pass

            abrir = messagebox.askyesno(
                "División completada",
                "El archivo se dividió correctamente en partes con sufijos _01, _02, ...\n\n"
                f"Origen: {src}\n"
                f"Tamaño por parte: {part_size_mb:.1f} MB\n\n"
                "¿Deseas abrir la carpeta del archivo origen?"
            )
            if abrir:
                try:
                    # Si existe la primera parte, seleccionarla; si no, abrir carpeta del origen
                    base, ext = os.path.splitext(src)
                    primera_parte = f"{base}_{1:02d}{ext}"
                    if os.path.exists(primera_parte):
                        subprocess.run(['explorer', '/select,', os.path.abspath(primera_parte)], check=True)
                    else:
                        carpeta = os.path.dirname(src)
                        subprocess.run(['explorer', carpeta], check=True)
                except (OSError, subprocess.CalledProcessError) as e:
                    print(f"No se pudo abrir Explorer: {e}")

            try:
                if self.root.winfo_exists():
                    self.root.after(3000, self._ocultar_progreso_split)
            except tk.TclError:
                pass

        except (IOError, OSError, ValueError) as e:
            messagebox.showerror("❌ Error", f"Error al dividir el archivo:\n{str(e)}")
            try:
                if self.root.winfo_exists():
                    self.split_estado_label.config(text="❌ Error en la división", fg='red')
                    self.root.after(2000, self._ocultar_progreso_split)
            except tk.TclError:
                pass


def main():
    """Punto de entrada de la aplicación.

    Crea la ventana raíz, instancia la clase de la UI, centra la ventana y
    arranca el loop principal de Tkinter.
    """
    root = tk.Tk()
    GeneradorArchivo(root)

    # Centrar ventana
    root.update_idletasks()
    width = root.winfo_width()
    height = root.winfo_height()
    x = (root.winfo_screenwidth() // 2) - (width // 2)
    y = (root.winfo_screenheight() // 2) - (height // 2)
    root.geometry(f"{width}x{height}+{x}+{y}")

    root.mainloop()


if __name__ == "__main__":
    main()

https://forum.rclone.org/t/google-drive-service-account-changes-and-rclone/50136 please check this out - new service accounts made after 15 April 2025 will no longer be able to own drive items. Old service accounts will be unaffected.

I know this is an old post, but I am wondering what the state of play now (2025) is for using deck.gl with Vue.js (my specific use case is GeoJson visualisation)?

The suggested project at vue_deckgl still seems alive, but I also noticed another project vue-deckgl-suite.

Are there other alternatives?

Is vue-deckgl-suite the same thing as vue_deckgl, with a slightly different name?

And do the answers to my questions depend on Vue 2 vs Vue 3 compatability?

Using SSR + Pkce flow works. However make sure to have the used cookies whitelisted since i wasted 2 whole days not realizing “Auth season missing” since the cookies didn’t get placed in case you are using a cookie manager 😩

After spending ages trying to get this working where I set .allowsHitTesting(true) and tried to let the SpriteView children manage all interaction and feed it back to the RealityView when needed, I decided it just wasn't possible. RealityKit doesn't really want to play nicely with anything else.

So what I did was create a simple ApplicationModel:

public class ApplicationModel : ObservableObject {
    
    @Published var hudInControl : Bool
    
    init() {
        self.hudInControl = false
    }
    
    static let shared : ApplicationModel = ApplicationModel()
    
}

and then in the ContentView do this:

struct ContentView: View {
    
    @Environment(\.mainWindowSize) var mainWindowSize

    @StateObject var appModel : ApplicationModel = .shared

    var body: some View {
        ZStack {
            RealityView { content in
                // If iOS device that is not the simulator,
                // use the spatial tracking camera.
                #if os(iOS) && !targetEnvironment(simulator)
                content.camera = .spatialTracking
                #endif
                createGameScene(content)
            }.gesture(tapEntityGesture)
            // When this app runs on macOS or iOS simulator,
            // add camera controls that orbit the origin.
            #if os(macOS) || (os(iOS) && targetEnvironment(simulator))
            .realityViewCameraControls(.orbit)
            #endif

            let hudScene = HUDScene(size: mainWindowSize)
            
            SpriteView(scene: hudScene, options: [.allowsTransparency])
            
            // this following line either allows the HUD to receive events (true), or
            // the RealityView to receive Gestures.  How can we enable both at the same
            // time so that SpriteKit SKNodes within the HUD node tree can receive and
            // respond to touches as well as letting RealityKit handle gestures when
            // the HUD ignores the interaction?
            //
                .allowsHitTesting(appModel.hudInControl)
        }
    }
}

this then gives the app some control over whether RealityKit, or SpriteKit get the user interaction events. When the app starts, interaction is through the RealityKit environment by default.

When the user then triggers something that gives control to the 2D environment, appModel.hudInControl is set to true and it just works.

For those situations where I have a HUD based button that I want sensitive to taps when the HUD is not in control, I, in the tapEntityGesture handler, offer the tap to the HUD first, and if the HUD does not consume it, I then use it as needed within the RealityView.

The reason you don’t see the extra artifacts in a regular mvn dependency:tree is because the MUnit Maven plugin downloads additional test-only dependencies dynamically during the code coverage phase, not as part of your project’s declared pom.xml dependencies. The standard dependency:tree goal only resolves dependencies from the project’s dependency graph, so it won’t include those.

Options to capture them:

1. Run with verbose dependency plugin on the test scope

mvn dependency:tree -Dscope=test -Dverbose

This will at least show all test-scoped dependencies that Maven resolves from your POM.

2. List resolved artifacts for a specific phase
   Use:

mvn dependency:list -DincludeScope=test -DoutputFile=deps.txt

Then run the plugin phase that triggers coverage (munit:coverage-report) in the same build. This way you can compare which artifacts are pulled in.

3. Use dependency:go-offline

mvn dependency:go-offline -DincludeScope=test

This forces Maven to download everything needed (including test/coverage). Then inspect the local repository folder (~/.m2/repository) to see what was actually pulled in by the MUnit plugin.

4. Enable debug logging when running coverage

mvn -X test
mvn -X munit:coverage-report

With -X, Maven logs every artifact resolution. You’ll be able to see which additional dependencies the plugin downloads specifically for coverage.

✅ Key Point:
Those extra jars are not “normal” dependencies of your project—they are plugin-managed artifacts that the MUnit Maven plugin itself pulls in. So the only way to see them is either with -X debug logging during plugin execution, or by looking in the local Maven repo after running coverage.

If you want a consolidated dependency tree for test execution including MUnit coverage, run the build with:

mvn clean test munit:coverage-report -X

and parse the “Downloading from …” / “Resolved …” sections in the logs.

Would you like me to write a ready-to-run shell script that extracts just the resolved test dependencies (including MUnit coverage) from the Maven debug output?

how to get data from line x to line y where line x and y identify by name.

Example:

set 1 = MSTUMASTER

3303910000

3303920000

3304030000

3303840000

set 2 = LEDGER

3303950000

I want get data under set 1 as below

3303910000

3303920000

3304030000

3303840000

see my method here, i installed it successuflly in 2025 for visual studio 2022

https://stackoverflow.com/a/79742876/4801995

I locked myself out by the mistaken security setting and had to search for the config file without any hint from the web UI.

Mine (Windows 7) is surprisingly in a different location: C:\Users\<user name>\AppData\Local\Jenkins\.jenkins

I’m trying to figure out a 8 digit number code there are 1 2 3 4 5 6 7 8 9 0 that you can add to it these are the numbers I know 739463

In my case, enabling Fast Deployment fixed this error.

Project>Property>Android>Option>Fast Deployment

Reference: https://github.com/dotnet/maui/issues/29941

I have a solution here, you can use uiautomation to find the browser control and activate it, while starting a thread to invoke the system-level enter button. After uiautomation activates the browser window, it starts to perform carriage enter once a second, and the pop-up window of this browser can be skipped correctly.

React Navigation doesn't use the native tabs instead it uses JS Tabs to mimic the behaviour of the native tabs. If you want liquid glass tabs you need to use react-native-bottom-tabs library to replace React Navigation Tabs with Native Tabs. You then need to do a pod install to do the linking and you should be good to go

The problem is that your function pdf_combiner() never gets called. In your code try/except block is indented the function,so Python just defines the functions and exits without ever executing it.
You can fix it by moving the function call outside and passing the output filename.

Unfortunately, the ASG down scaling is controlled by the TargetTracking AlarmLow Cloudwatch alarm. It needs to see 15 consecutive checks, 1 minute apart before triggering a scale down. It would allow you to edit it since it is controlled by ECS CAS. I am trying to find an environment variable to change it but so far, nothing.
The mentioned ECS_ENGINE_TASK_CLEANUP_WAIT_DURATION and ECS_IMAGE_CLEANUP_INTERVAL don't seem to be related to ASG/EC2 scale down.

int deckSize = deck.Count;

// show the last 5 cards in order
for (int i = 0; i < 5; i++)
{
    var drawnPage = deck[deckSize - 1 - i]; // shift by i each time

    buttonSlots[i].GetComponent<PageInHandButtonScript>().setPage(drawnPage);
    buttonSlots[i].GetComponent<UnityEngine.UI.Image>().sprite = drawnPage.getSprite();

    Debug.Log($"Page added to hand: {drawnPage.element} rune");
}

// now remove those 5 cards from the deck
deck.RemoveRange(deckSize - 5, 5);

Debug.Log($"Filled up hand. New Deck size: {deck.Count}");

I installed .net8 for visual studio community 2022 in 2025

Follow these steps:

----------

(install and update "Assistant install on step 4)"

(upgrade to net8 for your current project https://www.c-sharpcorner.com/article/upgrade-net-core-web-app-from-net-5-0-3-1-to-8-0-by-ms-upgrade-assistant/)

they now added the value parameter (Chrome 117) that must match to be deleted

TextField(
    textAlignVertical: TextAlignVertical.center,
    decoration: InputDecoration(
      isDense: true,
      contentPadding: EdgeInsets.symmetric(vertical: 10, horizontal: 15),
    ),
)

I tried this, but did not work. Created the new sort column fine and sorted ASC and it worked in the table, but my matrix header is still sorted ASC. Ugh! Power BI version Aug 2025

enter image description hereThis fanart is Lord x as an emoji.

Art by: Edited Maker

(It’s on YouTube.)

https://i.sstatic.net/Egqi3kZP.png

They now have an example repo for React https://github.com/docusign/code-examples-react. I don't think it has all the examples listed in the node examples repo but it might be a good starting point to understand how to integrate Docusign on a React app

FAC

CitiTri

City3.net

FJR.CA

JRV

CAB

UMA

Nineteen7ty3

SYETETRES

Onyx

Uno

Batman

101073

191910

101010

Tripple10

This has been fixed in the latest version of python-build-standalone. Please try the 20250808 release or later and see if the problem persists.

Since this is still an issue and there are not many solutions out there, I am gonna post an answer here.

This is a known compatibility issue between google-cloud-logging and Python 3.11. The CloudLoggingHandler creates background daemon threads that don't shut down gracefully when GAE terminates instances in Python 3.11, due to stricter thread lifecycle management.

Solutions (in order of preference for me)

1. Switch to StructuredLogHandler (Recommended)

Replace your current logging configuration with StructuredLogHandler, which writes to stdout instead of using background threads:

  # In Django settings.py (or equivalent configuration)
  LOGGING = {
      'version': 1,
      'disable_existing_loggers': False,
      'handlers': {
          'structured': {
              'class': 'google.cloud.logging.handlers.StructuredLogHandler',
          }
      },
      'loggers': {
          '': {
              'handlers': ['structured'],
              'level': 'INFO',
          }
      },
  }

Remove the problematic setup:

# Remove these lines:  
logging_client = logging.Client()  
logging_client.setup_logging()

Benefits:

• No background threads = no shutdown errors
• More reliable on serverless platforms
• Same logging destination in Cloud Console
• Zero application code changes needed

2. Downgrade to Python 3.10

Change your app.yaml:
runtime: python310 # instead of python311

Benefits: Confirmed to resolve the issue immediately
Drawbacks: Delays Python 3.11 adoption

3. Upgrade google-cloud-logging

Update to the latest version in requirements.txt:
google-cloud-logging>=3.10.0

Benefits: May include Python 3.11 compatibility fixes
Drawbacks: Not guaranteed to resolve the issue

References

• https://github.com/googleapis/python-logging/issues/803
• https://cloud.google.com/logging/docs/structured-logging

The StructuredLogHandler approach is recommended as it's the most future-proof solution and completely avoids the threading architecture that causes these shutdown errors.

Update to this, most of scikit-learn's weights functions have been updated to 'balanced', so you should be using something like:

svm = OneVsRestClassifier(LinearSVC(class_weight='balanced'))

X = [[1, 2], [3, 4], [5, 4]]
Y = [0,1,2]

svm.fit(X, Y)

For a typical cloud workload with similar server types, Least Connections is arguably the best "set-it-and-forget-it" algorithm. It is dynamic, efficient, and perfectly suited for the variable and scalable nature of cloud computing. It's a simple concept that delivers intelligent results.

For more details about other Algorithms this might be helpful 𝐋𝐨𝐚𝐝 𝐁𝐚𝐥𝐚𝐧𝐜𝐢𝐧𝐠 𝐀𝐥𝐠𝐨𝐫𝐢𝐭𝐡𝐦𝐬 𝐘𝐨𝐮 𝐌𝐮𝐬𝐭 𝐊𝐧𝐨𝐰

from moviepy.editor import VideoFileClip, concatenate_videoclips

# Carregar o vídeo enviado pelo usuário

input_path = "/mnt/data/VID-20250821-WA0001~2.mp4"

clip = VideoFileClip(input_path)

# Criar o reverso do vídeo

reverse_clip = clip.fx(vfx.time_mirror)

# Concatenar original + reverso para efeito boomerang

boomerang = concatenate_videoclips([clip, reverse_clip])

# Exportar resultado

output_path = "/mnt/data/boomerang.mp4"

boomerang.write_videofile(output_path, codec="libx264", audio_codec="aac")

output_path

I have a case where I am using SSIS to insert records and I a leaving out a timestamp column which has a default constraint of getdate(), strangely, when I run the insert, the column is still NULL.

Since iOS 26:

import UIKit
UIApplication.shared.sendAction(#selector(UIResponderStandardEditActions.performClose(_:)), to: nil, from: nil, for: nil)

Ensure that the UIApplication.shared responder hierarchy contains a valid first responder, otherwise app may not respond to system actions like closing

You can try the library called desktop_multi_window, I think it will solve your problem:

$ flutter pub add desktop_multi_window

You can see the documentation at desktop_multi_window

able to resolve this issue ?? if yes can you share me the details please.

Thanks,

Manoj.

Is this what do you want?

var x=[1,2,3,4];
var z=["1z","2z","3z","4z","5z","6z"];
var y=["1y","2y"];
for(i=0;i<Math.max(x.length, y.length, z.length);i++)
{
  console.log(x[i] ? x[i]:"");
  console.log(y[i] ? y[i]:"");
  console.log(z[i] ? z[i]:"");
}

Warning - I have never installed this extension and yet I found this executable running on my system. I verified that the hash of the executable on my local matches the known hash of e27f0eabdaa7f4d26a25818b3be57a2b33cbe3d74f4bfb70d9614ead56bbb3ea.

Again, I have never installed this extension (I only have a handful of Microsoft, GitHub, and AWS published extensions installed in VSCode) and so I find it very suspicious that it was running.

The Restler.exe file is a Windows executable. Since the docker container is running on a Linux kernel it cannot natively run Restler.exe. Instead, run: dotnet ./Restler.dll

C# Extensions by JosKreativ: He apparently continued with the project.

import numpy as np

import cv2

import os

# Path to the uploaded image

image_path = "/mnt/data/404614311_891495322548092_4664051338560382022_n.webp"

# Load the image

image = cv2.imread(image_path)

# Resize for faster processing

image_small = cv2.resize(image, (200, 200))

# Convert to LAB for better color clustering

image_lab = cv2.cvtColor(image_small, cv2.COLOR_BGR2LAB)

pixels = image_lab.reshape((-1, 3))

# KMeans to extract main colors

from sklearn.cluster import KMeans

kmeans = KMeans(n_clusters=6, random_state=42).fit(pixels)

colors = kmeans.cluster_centers_.astype(int)

# Convert colors back to RGB

colors_rgb = cv2.cvtColor(np.array([colors], dtype=np.uint8), cv2.COLOR_Lab2BGR)[0]

colors_rgb_list = colors_rgb.tolist()

colors_rgb_list

iPad userAgent string has no 'iPad' or 'iPhone' in the string any longer.

Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_6) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/15.6.7 Safari/605.1.15

Will the folliwing return true with iPads and false with all other Apple/Macs computers and iPhones?

.. = preg_match("/Macintosh;\sIntel\sMac\sOS\sX\s[\d]{2,4}_[\d]{1,8}_[\d]{1,4}/i", $_SERVER["HTTP_USER_AGENT");

If any one has a Mac that's not an iPad, please post user agent.

I am using PHP so can't use javascript

'ontouchstart' in window or navigator.msMaxTouchPoints or screen.width etc.

I haven't been able to figure out how to debug 32-bit Azure apps with Visual Studio 2022, but until a better solution is available a workaround to debug your app might be to create a console app or test project that includes your azure function app as a reference, and then call the relevant code from your console app.

Comma in the condition only test the last item, use && or loop.

table { will-change: transform; }

Did you try this?

First thing, clear WDT inside loops like reconnect() or in hangs forever.

Also, fix millis rollover by reboot using subtraction, not now > X, and eboot if MQTT is down for 1 min.

Sorry for contributing so late, But it might help others

You can do it, using a framework of robot framework, called RF Swarm, where you need to clone it from github, install rfswarm manager, agent just for running the test suite, you can also install reporter for html or doc report, also there will be logs in the logs directory with individual logs for a individual robot, the installing file will be there clone directory. Using RFSwarm, will help you to exert load of 25 to 40 robot if you have free 6 to 8 gb of ram, so you should have 16gb ram in you local machine (laptop/pc), as robot use selenium for UI testing, so the chrome browser eats a lot of ram when run in numbers. Suggestion to use chrome headless when performing load testing.

The answer is to wrap the content in the td cell with <div class="content">