I created a separate bundle for it: https://github.com/savinmikhail/symfony-profiler-response-bundle, see also https://github.com/symfony/symfony/issues/21168

👉 Free Image Uploader – NovaBuilder

It lets you drag & drop or paste an image instantly, and you’ll get a direct shareable URL you can copy right away.
No signup required, and it works well for embedding screenshots or examples in discussions.

Just add transform-style:preserve-3d; to the parent, and transform:translateZ(-10px); to the child.

To import boolean value, you have tu use "1" for tue and "0" for false in your CSV file.

But, ensure the column is correctly mapped to the corresponding attribute in Magento.

I found the cause and solution ✅

This issue wasn’t caused by Django or Python but by a Windows 11 update (October 2024 patches KB5066835 and KB5065789) that broke localhost connections using HTTP/2.

To fix it:

1. Open the Registry Editor (regedit).

2. Navigate to:

   HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\HTTP\Parameters
   
   

3. Add these DWORD (32-bit) values:

   EnableHttp2Tls = 0
   EnableHttp2Cleartext = 0
   
   

4. Restart your computer.

After restarting, Django’s development server and Live Server both worked again at http://127.0.0.1:8000.

This should help others who get “The connection was reset” or “127.0.0.1 refused to connect” after recent Windows updates

If we want on convert the date to human readable format in template

Then use the inbuilt angular date pipe

import { DatePipe } from '@angular/common';

@Component({
  selector: 'app-example',
  imports: [DatePipe], // required
  templateUrl: 'template.component.html'
})

Use date pipe in the template file as

<p>{{ dateVariable | date: 'medium' }}</p>
// output will look as Sep 13, 2025, 6:15 PM
// or 
<p>{{dateVariable | date : "dd/MM/yy HH:mm:ss"}}</p>

Follow the given below link to know other predefined angular date time format

The problem was caused by a bug from Raspberry Pi OS. I know this because when I installed the python application in a new RPI5 with OS version 2025-10-01 the RAM consumption did not increase at all.

 I am using xaml with PS, and I want to use my background image source as base64. I can't use the relative path anymore, hence my base64 approach, how do I change this to use the base64 as a source. 
    
<Window.Background>
        
    </Window.Background>
    <Grid x:Name="mygrid">
        <Grid.Background>
            <ImageBrush ImageSource="/Wallpapers1280x1024.jpg"/>
        </Grid.Background>

Solved: Reinstall Sourcetree

I had this problem after trying the refresh option that did not work. I uninstalled sources tree and reinstalled it. It solved the issue. As the state is anyway in Git. It did not have any issues.

it's simply :

git branch --show-current

The chrome.storage can now be accessed from the Chrome browser's Developer Tools as shown below.

Press F12 to open Developer Tools -> Application -> Storage -> Extension storage

For people who have the same problem while editing an css file in vs. I have used the answer in this link and that worked for me.

‫</div> <div class="eye"><div class="pupil"></div></div> </div> <div class="mouth"></div> </div> <button id="button">🎪 Test Tippe mich!</button> <script> const button = document.getElementById('button'); const

Fill drug columns with "skipped" to preserve the semantic meaning (drug not given) and use NaN only for true missing or unknown values in other columns. This separation helps downstream analysis or models distinguish between absence-by-design and missing data.

• Hi, please tell me, I have an INVALID AUTH KEY. What can I do? I replenished my wallet, now I can't withdraw money

From my experience of HAL code so far whoever wrote some of it should be shot.

Take for instance HAL_UART_Transmit_IT(UART_HandleTypeDef *huart, uint8_t *pData, uint16_t Size).

I would expect this to transmit any multi byte buffer perfectly from start to finish once kicked off - but no!

YOU have to make sure that pData is pointing to a STATIC buffer, (and you can't change the contents of that static buffer until the transmission has completed), i.e. YOU have to wait until you get the Tx Complete callback invoked before you can either reuse the buffer, or before the buffer goes out of scope on your stack\heap.

In other words if you want proper interrupt based buffer transmission YOU have to implement ring buffering yourself on top of the HAL code provided by ST. The ST code will only reliably work sending single bytes at a time which rather renders pointless using interrupt driven comms in the first place.

Same applies to how they implemented DMA Idle buffer handling... a real mess when it could have been done properly using industry standard ring buffering patterns.

Had the same issue and was on 9.4.0. Downgrading to mysql-connector-python==8.4.0 worked for me.

Please check the breaking and important changes of each major release. Probably https://docs.typo3.org/c/typo3/cms-core/main/en-us/Changelog/12.0/Breaking-97305-IntroduceCSRF-likeLoginToken.html is the reason for this if you have your own felogin template.

SOLVED - Thank you to hints from Jared Smith and Pointy.

async function getKeyIndex(storeName) {
    return new Promise((resolve, reject) => {
        const dbStore = [];
        let c = 0;
        const objectStore = db.transaction(storeName).objectStore(storeName);
        const cursorRequest = objectStore.openCursor();

        cursorRequest.onsuccess = (event) => {
            const cursor = event.target.result;
            if (cursor) {
                console.log("Name: " + cursor.key + " - id: " + cursor.value.id);
                dbStore.push(cursor.value.id);
                c++;
                cursor.continue();
            } else {
                resolve(dbStore);
            }
        };
    });
}

Fired by:

        keyIndex = await getKeyIndex('conversations');
        console.log("Database keys " + keyIndex);

no, you can't opening two separate file in the same window.
Each workspace file represent a complete configuration and folder set, and opening it always launches its own VS code window

However, you can opening a new folder in a new window, and switch between windows to work with multiple workspaces.

code folder1(workspace1) folder2(workspace2)

ex: code backend frontend

What you're seeing is a lint warning, try suppressing the lint as follows:

AnimatedContent(
    targetState = uiState.singleId,
    transitionSpec = {
        if (targetState > initialState)
            slideInHorizontally { it } + fadeIn() togetherWith slideOutHorizontally { -it }
        else
            slideInHorizontally { -it } + fadeIn() togetherWith slideOutHorizontally { it }
    },
    label = "Fade Content",
) { @Suppress("UnusedTransitionTargetStateParameter") _ ->
    Content(dat = dat, uiState = uiState, onUiEvent = onUiEvent)
}

It seems that there is no API to get or set git config using LibGit2Sharp. However, I was able to do it by simply starting git process with correct arguments:

var startInfo = new ProcessStartInfo
{
    FileName = "git",
    Arguments = "config --global user.email [email protected]",
    RedirectStandardOutput = true,
    RedirectStandardError = true,
    UseShellExecute = false,
    CreateNoWindow = true,
    WorkingDirectory = "."
};

var process = new Process { StartInfo = startInfo };
process.Start();
process.WaitForExit();

if (process.ExitCode != 0)
{
    throw new InvalidOperationException($"Process exited with non-zero code ({process.ExitCode}).");
}

In regards to running it on a Windows Arm64 device I was at least able to build a project with LibGit2Sharp as dependency using win-arm64 runtime flag:

dotnet build -r win-arm64

And during my quick research I've also found this article, that seems to prove it is possible to run .NET 8 apps on Arm64 machine. If I were you I would check if .NET SDK is correctly installed on the machine.

x = 5 + 10 is an expression but with when followred by semicolon x = 5 + 10; its called an "expression statement", This makes "“every expression can also be used as a statement" true, but then they will not be strictly called as "statement" but rather called as "expression statement" as statements are  are constructs that perform actions or control flow and do not necessarily yield a value, like if, for, or function declarations, so my takeaway and simple solution to tell if given piece of line is statment or expression is**,
[if you can put it in the console.log() without error/which gives some meaningful result  then its an expression and when you place that content with a semicolon then its technically called as "expression statement" anything else is just pure "statement" ]**

add to code

import dayjs from 'dayjs';
import 'dayjs/locale/ru';

dayjs.locale('ru', {
  ...dayjs.Ls.ru,
  weekStart: 1,
});

I solved it by clearing all the build cache/files in the same directory where the source code was in, and then ran the flet build command through the Developer Command Prompt for VS 2022.

Use native CSS unit:

CSS.pt(1).to('px').value

https://developer.mozilla.org/en-US/docs/Web/API/CSS

Refer to this Troubleshoot documentation

The reason can be either Bad access token or Embed type doesn't match token type. Looking into your code, the tokenType is 1 which is correct. but the comment says Aad Token and you're using Aad token. Change the tokenType: 0 for Aad Token. Refer to this documentation

This code checks which YouTube thumbnail size is available and returns the best one. If the highest quality thumbnail (maxresdefault.jpg) doesn’t exist, it automatically falls back to the next available resolution. It’s a smart and reliable way to always get the best possible YouTube thumbnail for any video.

did you find the omnik register list.

regards

marc

Replacing the wp-admin and wp-content folders from a fresh install solved the error (although I still have no idea what caused it).

This is still a problem in VS 2022.

Sometimes nothing helps, error window keeps pop up even when we have zero error.

In that case deleting the ".vs" folder in the solution root folder + restarting may helps.

Consider using afterNextRender:

afterNextRender(
  () => {
    this.foo();
  },
  { injector: this.injector }
);

"Register callbacks to be invoked the next time the application finishes rendering..."
https://angular.dev/api/core/afterNextRender

To fix the lighting issue from your question using the setup shown in this image, you should manually generate lighting in Unity 6.2 because the "Auto Generate" option is no longer availablel 111 2J.

### What to Do in This Scene

Open the Lighting window by going to Window > Rendering > Lighting.

In the Lighting window, locate and click

the "Generate Lighting" button (since "Auto Generate" does not exist now).

This action will bake the lighting data for your scene, making all objects and panels display their intended brightness and color (panels will appear white instead of gray) I

https://www.youtube.com/watch?v=FxeeCxQGW9U&feature=youtu.be.

https://www.youtube.com/watch?v=FxeeCxQGW9U.

Can someone please tell me what name this YouTube channel is?

Removing or renaming updater.exe resolves.

https://stackoverflow.com/search?q=couldn%27t+load+%22XPCOM%22

I had a similar problem lately.

And I would suggest taking a look at the difference between clearAllMocks() and resetAllMocks()

clearAllMocks is useful when you want to ensure that the call history of mocks is clean between tests, but you want to retain any mocked return values or implementations.

resetAllMocks is useful when you want a complete reset, effectively returning all mocks to their initial, unconfigured state.

In my case, I was using mockResolvedValueOnce, and it affected the next test. clearAllMocks was of no help at all, but resetAllMocks solved the problem**.**

I have the Same Problem.

I solve it by using an observable.

toObservable(mySignalFromStore).subscribe(...)

I dont know If this is THE solution.

If you want to print exactly \n in python , you should do this: \\n

Example :

b= "\\n"
print(b)

Below are the two statements from two different papers one is cited in other and they applied same process to combine multiple band images and created a 7 channel dataset I want to do the same. (" Sentinel-1 and Sentinel-2 images were finally geocoded on a common

coordinates grid and then apply amplitude normalization. In the end,

concatenate together the processed data, resulting in a five-channel

multi-sensor dataset, including B4 (red), B3 (green), B2 (blue), B8

(NIR), and VV. " )("The

SAR; slope data; and the red, green, blue, NIR, and NDWI

images from Sentinel-2 were stacked to create a seven-channel

dataset.) They stacked images then normalize the stacked-7-channel-image because because SAR band has values of -13db to 20db greater value range and others have 0 to 1 small value ranges. So for faster convergence and better performence they have normalized the stacked-7-channel-image.

You'll need to install "Code Runner" extension in VS code. Then try.

When using the tiffsep1 driver, how can I set the independent color separation plate angles C=15, M=75, Y=0, K=45? I have tried many methods but none of them worked. The converted color separation plate images all have dots at the same Angle

Found the Fix to it. Go to Disk Utility when screen loads up and format the disk with supported Apple File Format System!

Refer here: https://support.apple.com/guide/disk-utility/file-system-formats-dsku19ed921c/22.7/mac/26

i having the same problem if you find the solution can you help me out with it

Try this one, it solve some bugs on launch4j

https://github.com/kaffamobile/gjg-maven-plugin

You can find the solution for device targeting API level >=36 in here - https://github.com/flutter/flutter/issues/168635#issuecomment-3417866156

based on your requirements you need children from particular parent so as far as my knowledge you should use findMany method because the findUnique with include is comparatively slower than the findMany and in your case as you said children can be in big numbers i prefer to use findMany

# Given values

current_age = 39

current_size = 9.47

# Calculate the rate of change (m)

rate_of_change = current_size / current_age

print(rate_of_change)

adsadsdasasd
DAS
DA
AD
S
DSA
SD

@Topaco in the first comment got the right answer. When you apply operations to data, its important to remember your order of operations.

In decrypt_data(), unpadding may only take place after decryption!

    @staticmethod
    def decrypt_data(password, salt, encrypted_data):
        print(f"{password=}")
        print(f"{salt=}")
        print("Encrypted Data: ")
        print(f"{encrypted_data}")
        key = PasswordManager.generate_aes_key(password, salt)
        print(f"{key=}")
        cipher = Cipher(algorithms.AES(key), modes.CBC(salt))
        decryptor = cipher.decryptor()
        decrypted_data = decryptor.update(encrypted_data) + decryptor.finalize()
        original_data = PasswordManager.unpad_data(decrypted_data)
        print(original_data)
        return original_data.decode(PasswordManager.encoding)

If anyone ever encounters the error using the OCI CLI

ERROR: Could not find config file at [DEFAULT]

user=ocid1.user.xxxx

...

Do you want to create a new config file? [Y/n]: Abort:

I hit this issue in a gitlab pipeline job, whereby I was using the contents of a pipeline variable named OCI_CLI_CONFIG_FILE to seed ~/.oci/config

Unfortunately that OCI_CLI_CONFIG_FILE variable happened to also be a reserved environment variable name in the context of the OCI CLI : https://docs.oracle.com/en-us/iaas/Content/API/SDKDocs/clienvironmentvariables.htm

... and was thus causing the OCI CLI to fail when making requests, as it was looking for a config file at that messed up location.

So be very careful with environment variables!

just use winboat, its a windows in a linux

The "NqSQL Database" term sometimes confuses, because at its root, it means "Not relational" databases: data are not organized as rows. SQL-like queries u can find for different storage, S3, for example. Another example is the Cassandra database with its CQL, which looks like SQL, but the database is not relational. Don't confuse between query language, which is an interface, and the internal engineering of the DB

Questions 1 and 4 are duplicate:

Answer of Q1: Inserting RecyclerView items at zero position - always stay scrolled to top

Answer of Q4: How to fix RecyclerView shuffling issue:

overriding these two methods in Adapter will solve the shuffle issue.

@Override
public long getItemId(int position) {
return position; // or item id
}

@Override
public int getItemViewType(int position) {
return position; // or item id
}

I recently wrote a C-to-Brainfuck compiler. Here's my write-up on how it works.

I was able to resolve the issue. I was using the latest AzureRM provider (v4.49.0), which was released just yesterday.

I restarted the setup (again), this time pinning the provider to a previous stable version (v4.48.0). After that, everything worked perfectly. The push succeeded without any errors and .gitignore was considered.

If others encounter similar problems with the latest v4.49.0 Terraform AzureRM provider, it may be caused by a build or packaging issue. Reverting to an earlier version seems to resolve it.

<video controls> <source src="Gemini_Generated_Video_2ymkw42ymkw42ymk.mp4" type="video/mp4"> Your browser does not support the video tag. </video>

here is a updated version of unity of work from svn repository.
Update up to .NET Framework 4.8
https://github.com/waltereidi/NHibernateUnitOfWork_NetFramework4.8_Rework

I encountered this issue at work, turns out my internet connection was fine, but our internal networks were down. Including the VPN. It's worth nonting that we use Azure DevOps, but I don't have admin access to any of it, so I wasn't able to try the accepted answer. So, my solution was to file a network support ticket with IT so they're aware our network is down, and go home early once I ran out of things to do that didn't require the secure access to our network.

I have just the thing for you.
https://github.com/rajames440/StarForth is an ANSI C99 with minimal dependencies I wrote to port into a microkernel environment. You'll also see some inline asm for amd64 and arm64 with plans down the road for RISC-V inline asm as well. Feel free to snoop around it. The documentation is pretty good. My approach was to build it as a totally static monolithic build unlike gForth & others so it ports almost anywhere fairly easily but will still be usable as an executable on Linux without a lot of difficulty and it was designed with Ras Pi 4b in mind as well. You could clone the repo onto a Raspberian install and build it without much difficulty.

Are you looking for something like this ?

:root {
  --angle: 45deg;
  --opacity: 0.5;
}

*,
*::before,
*::after {
  box-sizing: border-box;
}

.rainbow {
  width: 40px;
  height: 30px;
  border-radius: 10px;
  padding: 2rem;
  margin: auto;

  display: grid;
  place-content: center;
  text-align: center;

  font-size: 1.5em;

  --border-size: 0.3rem;
  border: var(--border-size) solid transparent;

  /* Paint an image in the border */
  border-image: conic-gradient(
      from var(--angle),
      #d53e33 0deg 90deg,
      #fbb300 90deg 180deg,
      #377af5 180deg 270deg,
      #399953 270deg 360deg
    )
    1 stretch;
  background: rgb(255 255 255 / var(--opacity));
}

@property --opacity {
  syntax: "<number>";
  initial-value: 0.5;
  inherits: false;
}

@property --angle {
  syntax: "<angle>";
  initial-value: 0deg;
  inherits: false;
}

@keyframes opacityChange {
  to {
    --opacity: 1;
  }
}

@keyframes rotate {
  to {
    --angle: 360deg;
  }
}

.rainbow {
  animation: rotate 4s linear infinite, opacityChange 3s infinite alternate;
}

html,
body {
  padding: 0;
  margin: 0;
  height: 100%;
}

body {
  display: grid;
  place-content: center;
  gap: 0.5em;
}

<div class="rainbow">

</div>

or you can sign up for multiple accounts

I really wish I could help you. I'm a FORTH programmer from way back. I haven't found any good syntax highlighting utilities. I'm simply using a block editor in my terminal. I would like to ask you what you would like to see in a syntax highlighter. I wouldn't even know what syntax highlighting would do for FORTH, but you have me interested myself!

This is a CORS error: the server is refusing requests from other domains. Fix it by proxying the images through your own backend.

Not sure this helps but I'm on 2025.2.3 and I have the Code Style set to defaults according to the below. This works quite well and certainly doesn't wrap single lines of SQL like in your example. Have you tried that already?

• Set the General settings (Editor, Code Style, SQL, General) to the defaults (Cogwheel, Reset Defaults)

• For all dialects below General, "Inherit general SQL style" is checked

You could also try

.onTapGesture(count:1) {location in
let tapLocation: CGPoint = location
}

This should make sure it doesn't trigger when you double-tap

I believe code below can handle the task. I tested with a couple test cases and seemed to work. Please let me know in case it fails with any specific array.

Algorithm:

1. It locates the first positive and negative indices and stores them in pos and neg respectively.

2. Starts iterating from this point on by

   1. Swaps the number at pos index with the number at neg index.

   2. Stores the index ahead among pos and neg in a hash set swapped so that any index already swapped does not get swapped again.

   3. Locates the next negative an positive indices for swapping.

Note that I implemented the algorithm in C# and let AI convert to to Java since I'm rusty with Java. Please let me know if anything seems odd with the Java code.

import java.util.HashSet;

public class Main {
    public static void main(String[] args) {
        int[] nums = { -1, 2, 3, -4, 5, 6, -7, -8, -9, 10 };

        int pos = 0;
        while (pos < nums.length && nums[pos] < 0)
            ++pos;

        int neg = 0;
        while (neg < nums.length && nums[neg] > 0)
            ++neg;

        HashSet<Integer> swapped = new HashSet<>();
        while (pos < nums.length && neg < nums.length) {
            // swap
            int temp = nums[pos];
            nums[pos] = nums[neg];
            nums[neg] = temp;

            if (pos < neg)
                swapped.add(neg);
            else
                swapped.add(pos);

            do {
                ++pos;
            } while (pos < nums.length && (swapped.contains(pos) || nums[pos] < 0));

            do {
                ++neg;
            } while (neg < nums.length && (swapped.contains(neg) || nums[neg] > 0));
        }

        for (int i = 0; i < nums.length; i++)
            System.out.print(nums[i] + " ");
    }
}

nevermind, this worked:

while \
  read \
    attribute_one \
    attribute_two \
; do

I'm getting exactly the same issue as you describe. If I have over ~70 slides, swiping through the slides on Safari iOS (iPhone 13 mini 26.0.1) will cause it to crash with the following message, `A problem repeatedly occurred on "https://192.168.0.210/bins/008b1c6-4463-4b94-862b-7d74d22dab3e/".` (For those reading this, the URL is irrelevant.)

I'm assuming this must be an issue of scale. And indeed it is. Looking at the swiper documentation at https://swiperjs.com/swiper-api#virtual-slides , it looks like Virtual slides was specifically designed to tackle this issue of scale. Heck, even their demo over at https://swiperjs.com/demos#virtual-slides shows 600 slides to prove the point. After I enabled Virtual slides, the issue instantly went away!

poisson surface reconstruction the method will need a point_set input. How to create point set from point is here https://doc.cgal.org/latest/Point_set_3/Point_set_3_2point_set_8cpp-example.html . You can use Advancing front without creating point set. https://doc.cgal.org/latest/Advancing_front_surface_reconstruction/Advancing_front_surface_reconstruction_2reconstruction_fct_8cpp-example.html

There may be a bug when trying to transform FancyArrowPatch to 3D for an curved path (connectionstyle arc3). Instead I'm using a workaround much like this one: https://stackoverflow.com/a/79787179/31611660, which is to just draw a very short FancyArrowPatch to render the arrow by itself. Not my favorite, but its the only thing I've seen, so far, that works.

I used this with matplotlib 3.9.2 and numpy 1.26.4

import mpl_toolkits.mplot3d.art3d as art3d
import numpy as np
from matplotlib import pyplot as plt
from matplotlib.patches import Arc, FancyArrowPatch
from mpl_toolkits.mplot3d import proj3d

ARCSTYLE = dict(linestyle="--", color="black", lw=1)
ANGSTYLE = dict(radius=0.6, linestyle="-", color="black", lw=1, mutation_scale=15)
ARROWOPTS = {
    "mutation_scale": 20,
    "arrowstyle": "-|>",
    "color": "k",
    "shrinkA": 0,
    "shrinkB": 0,
}
TEXTOPTS = {
    "horizontalalignment": "center",
    "verticalalignment": "center",
    "fontsize": 14,
}


class Arrow3D(FancyArrowPatch):
    def __init__(self, xs, ys, zs, *args, **kwargs):
        super().__init__((0, 0), (0, 0), *args, **kwargs)
        self._verts3d = xs, ys, zs

    def do_3d_projection(self, renderer=None):
        xs3d, ys3d, zs3d = self._verts3d
        xs, ys, zs = proj3d.proj_transform(xs3d, ys3d, zs3d, self.axes.M)
        self.set_positions((xs[0], ys[0]), (xs[1], ys[1]))
        return np.min(zs)


class Arc3D(Arc):
    """
    Arc patch transformed to 3D, draws circular arc
       xdir: starting position of arc (3x1)
       ydir: ending position of arc (3x1)
       xy: origin as tuple (x,y)
       radius: radius of arc
       zs: Z-direction offset applied after rotation to 3D
       z: Z-direction offset applied before rotation to 3D
    """

    def __init__(self, xdir, ydir, xy=(0, 0), radius=1, *args, zs=0, z=0, **kwargs):
        xdir = np.array(xdir) / np.linalg.norm(xdir)
        ydir = np.array(ydir) / np.linalg.norm(ydir)
        super().__init__(
            xy,
            radius * 2.0,
            radius * 2.0,
            theta1=0,
            theta2=np.degrees(np.arccos(xdir.dot(ydir))),
            *args,
            **kwargs,
        )
        transform_2d_to_3d(self, rotation_matrix(xdir, ydir), zs=zs, z=z)


def arrowarc3d(
    ax, xdir, ydir, xy=(0, 0), radius=1, *args, zs=0, z=0, arrowfactor=100.0, **kwargs
):
    """
    draws an Arc3D and an arrowhead of Arrow3D
    """

    def radlen(q):
        return q * radius / np.linalg.norm(q)

    kw_arrow = {}
    # which kwargs are for *only* Arrow3D?
    for kw in ("mutation_scale",):
        if kw in kwargs:
            kw_arrow[kw] = kwargs.pop(kw)
    # settings in common
    color = kwargs.get("color", "blue")
    kwargs["color"] = color
    # compute begin & end of arrow
    xdir = np.array(xdir)
    ydir = np.array(ydir)
    # distance between points for arrow is (1/arrowfactor) * |ydir-xdir|
    ym = ydir - (ydir - xdir) / arrowfactor
    ax.add_artist(Arc3D(xdir, ydir, xy, radius, *args, zs=zs, z=z, **kwargs))
    ax.add_artist(Arrow3D(*zip(radlen(ym), radlen(ydir)), color=color, **kw_arrow))


def rotation_matrix(x, y):
    """
    Rotation matrix: [ X  ((X cross Y) cross X)  (X cross Y) ]
       x: X-axis of new coordinate system
       y: orients new coordinate system, new Y-axis is cross(cross(x,y),x)
    rotation_matrix(x,y).dot((1,0,0)) || x
    rotation_matrix(x,y).dot((0,0,1)) || cross(x,y)
    """
    x = np.array(x) / np.linalg.norm(x)
    z = np.cross(x, y)
    z /= np.linalg.norm(z)
    y = np.cross(np.array(z), x)
    return np.array([x, y, z]).T


def transform_2d_to_3d(pathpatch, rotation, zs=0, z=0):
    """
    Transforms a 2D Patch to a 3D patch.
      rotation: a 3x3 rotation matrix from 2D to 3D.
      zs: Z-direction offset applied after rotation
      z: Z-direction offset applied before rotation
    """
    # https://stackoverflow.com/a/18228967/31611660
    path = pathpatch.get_path()  # Get the path and the associated transform
    trans = pathpatch.get_patch_transform()
    path = trans.transform_path(path)  # Apply the transform
    pathpatch.__class__ = art3d.PathPatch3D  # Change the class
    pathpatch._code3d = path.codes  # Copy the codes
    pathpatch._facecolor3d = pathpatch.get_facecolor  # Get the face color
    verts = path.vertices  # Get the vertices in 2D
    # apply the rotation matrix and any "z" or "zs" offsets
    pathpatch._segment3d = np.array(
        [np.dot(rotation, (x, y, z)) + (0, 0, zs) for x, y in verts]
    )


def Rx(phi):
    return np.array(
        [[1, 0, 0], [0, np.cos(phi), -np.sin(phi)], [0, np.sin(phi), np.cos(phi)]]
    )


def Ry(theta):
    return np.array(
        [
            [np.cos(theta), 0, np.sin(theta)],
            [0, 1, 0],
            [-np.sin(theta), 0, np.cos(theta)],
        ]
    )


def Rz(psi):
    return np.array(
        [[np.cos(psi), -np.sin(psi), 0], [np.sin(psi), np.cos(psi), 0], [0, 0, 1]]
    )


def vecs(key, arc):
    """input is in degrees"""
    return vecsr(key, np.array(arc) * np.pi / 180.0)


def vecsr(key, arc):
    """input is in radians"""
    arc = np.array(arc)
    if key == "z":
        return np.array([np.cos(arc), np.sin(arc), arc * 0])
    elif key == "-z":
        return np.array([-np.sin(arc), np.cos(arc), arc * 0])
    elif key == "y":
        return np.array([np.cos(arc), arc * 0, -np.sin(arc)])
    elif key == "-y":
        return np.array([np.sin(arc), arc * 0, np.cos(arc)])
    elif key == "x":
        return np.array([arc * 0, np.cos(arc), np.sin(arc)])
    elif key == "-x":
        return np.array([arc * 0, -np.sin(arc), np.cos(arc)])
    else:
        raise ValueError(f"unknown key {key}")


# define origin
o = np.array([0, 0, 0])

# define ox0y0z0 axes
x0 = np.array([1, 0, 0])
y0 = np.array([0, 1, 0])
z0 = np.array([0, 0, 1])

# define ox1y1z1 axes
psi = 20 * np.pi / 180
x1 = Rz(psi).dot(x0)
y1 = Rz(psi).dot(y0)
z1 = Rz(psi).dot(z0)

# define ox2y2z2 axes
theta = 10 * np.pi / 180
x2 = Rz(psi).dot(Ry(theta)).dot(x0)
y2 = Rz(psi).dot(Ry(theta)).dot(y0)
z2 = Rz(psi).dot(Ry(theta)).dot(z0)

# define ox3y3z3 axes
phi = 30 * np.pi / 180
x3 = Rz(psi).dot(Ry(theta)).dot(Rx(phi)).dot(x0)
y3 = Rz(psi).dot(Ry(theta)).dot(Rx(phi)).dot(y0)
z3 = Rz(psi).dot(Ry(theta)).dot(Rx(phi)).dot(z0)

ARROWOPTS = {
    "mutation_scale": 20,
    "arrowstyle": "-|>",
    "color": "k",
    "shrinkA": 0,
    "shrinkB": 0,
}

# produce figure
fig = plt.figure()
ax = fig.add_subplot(111, projection="3d")

# plot ox0y0z0 axes
a = Arrow3D([o[0], x0[0]], [o[1], x0[1]], [o[2], x0[2]], **ARROWOPTS)
ax.add_artist(a)
a = Arrow3D([o[0], y0[0]], [o[1], y0[1]], [o[2], y0[2]], **ARROWOPTS)
ax.add_artist(a)
a = Arrow3D([o[0], z0[0]], [o[1], z0[1]], [o[2], z0[2]], **ARROWOPTS)
ax.add_artist(a)

# plot ox1y1z1 axes
a = Arrow3D([o[0], x1[0]], [o[1], x1[1]], [o[2], x1[2]], **ARROWOPTS)
ax.add_artist(a)
a = Arrow3D([o[0], y1[0]], [o[1], y1[1]], [o[2], y1[2]], **ARROWOPTS)
ax.add_artist(a)
a = Arrow3D([o[0], z1[0]], [o[1], z1[1]], [o[2], z1[2]], **ARROWOPTS)
ax.add_artist(a)

# draw dotted arc in x0y0 plane
ax.add_artist(Arc3D(*vecs("z", [-5, 116]).T, **ARCSTYLE))

# mark z0 rotation angles (psi)
arrowarc3d(ax, *vecsr("z", [0, psi]).T, **ANGSTYLE)
arrowarc3d(ax, *vecsr("-z", [0, psi]).T, **ANGSTYLE)

# plot ox2y2z2 axes
a = Arrow3D([o[0], x2[0]], [o[1], x2[1]], [o[2], x2[2]], **ARROWOPTS)
ax.add_artist(a)
a = Arrow3D([o[0], y2[0]], [o[1], y2[1]], [o[2], y2[2]], **ARROWOPTS)
ax.add_artist(a)
a = Arrow3D([o[0], z2[0]], [o[1], z2[1]], [o[2], z2[2]], **ARROWOPTS)
ax.add_artist(a)

# draw dotted arc in x1z1 plane
ax.add_artist(Arc3D(*Rz(psi).dot(vecs("-y", [-5, 105])).T, **ARCSTYLE))

# mark y1 rotation angles (theta)
arrowarc3d(ax, *Rz(psi).dot(vecsr("y", [0, theta])).T, **ANGSTYLE)
arrowarc3d(ax, *Rz(psi).dot(vecsr("-y", [0, theta])).T, **ANGSTYLE)

# plot ox3y3z3 axes
a = Arrow3D([o[0], x3[0]], [o[1], x3[1]], [o[2], x3[2]], **ARROWOPTS)
ax.add_artist(a)
a = Arrow3D([o[0], y3[0]], [o[1], y3[1]], [o[2], y3[2]], **ARROWOPTS)
ax.add_artist(a)
a = Arrow3D([o[0], z3[0]], [o[1], z3[1]], [o[2], z3[2]], **ARROWOPTS)
ax.add_artist(a)

# draw dotted arc in y2z2 plane
ax.add_artist(Arc3D(*Rz(psi).dot(Ry(theta).dot(vecs("x", [-5, 125]))).T, **ARCSTYLE))

# mark x2 rotation angles (phi)
arrowarc3d(ax, *Rz(psi).dot(Ry(theta).dot(vecsr("x", [0, phi]))).T, **ANGSTYLE)
arrowarc3d(ax, *Rz(psi).dot(Ry(theta).dot(vecsr("-x", [0, phi]))).T, **ANGSTYLE)

# add label for origin
ax.text(0.0, 0.0, -0.05, r"$o$", **TEXTOPTS)

# add labels for x axes
ax.text(1.1 * x0[0], 1.1 * x0[1], 1.1 * x0[2], r"$x_0$", **TEXTOPTS)
ax.text(1.1 * x1[0], 1.1 * x1[1], 1.1 * x1[2], r"$x_1$", **TEXTOPTS)
ax.text(1.1 * x2[0], 1.1 * x2[1], 1.1 * x2[2], r"$x_2, x_3$", **TEXTOPTS)

# add labels for y axes
ax.text(1.1 * y0[0], 1.1 * y0[1], 1.1 * y0[2], r"$y_0$", **TEXTOPTS)
ax.text(1.1 * y1[0], 1.1 * y1[1], 1.1 * y1[2], r"$y_1, y_2$", **TEXTOPTS)
ax.text(1.1 * y3[0], 1.1 * y3[1], 1.1 * y3[2], r"$y_3$", **TEXTOPTS)

# add labels for z axes
ax.text(1.1 * z0[0], 1.1 * z0[1], 1.1 * z0[2], r"$z_0, z_1$", **TEXTOPTS)
ax.text(1.1 * z2[0], 1.1 * z2[1], 1.1 * z2[2], r"$z_2$", **TEXTOPTS)
ax.text(1.1 * z3[0], 1.1 * z3[1], 1.1 * z3[2], r"$z_3$", **TEXTOPTS)

# add psi angle labels
m = 0.55 * ((x0 + x1) / 2.0)
ax.text(m[0], m[1], m[2], r"$\psi$", **TEXTOPTS)
m = 0.55 * ((y0 + y1) / 2.0)
ax.text(m[0], m[1], m[2], r"$\psi$", **TEXTOPTS)

# add theta angle labels
m = 0.55 * ((x1 + x2) / 2.0)
ax.text(m[0], m[1], m[2], r"$\theta$", **TEXTOPTS)
m = 0.55 * ((z1 + z2) / 2.0)
ax.text(m[0], m[1], m[2], r"$\theta$", **TEXTOPTS)

# add phi angle labels
m = 0.55 * ((y2 + y3) / 2.0)
ax.text(m[0], m[1], m[2], r"$\phi$", **TEXTOPTS)
m = 0.55 * ((z2 + z3) / 2.0)
ax.text(m[0], m[1], m[2], r"$\phi$", **TEXTOPTS)

# show figure
ax.view_init(elev=-150, azim=60, roll=0)
ax.set(xlim=[-0.46, 1.06], ylim=[-0.59, 1.06], zlim=[-0.11, 1.03])
ax.set_axis_off()
plt.ion()
plt.show()

result: https://s3nd.pics/post/68f2b73c8b58d466e5f5998b

more info

• https://github.com/matplotlib/matplotlib/issues/30659
• https://stackoverflow.com/a/74122407/31611660
• https://stackoverflow.com/a/22867877/31611660
• https://stackoverflow.com/a/18228967/31611660
• FancyArrowPatch in 3D with curve (connectionstyle)

🧩 Why Safari needs double tap on links (and the simple fix)

In iOS Safari, tapping a link sometimes requires two taps — the first activates the :hover state, and the second actually follows the link.
This happens because Safari doesn’t initialize touch events until a touchstart listener exists on the page.

You can fix it globally with a single line of JavaScript:

document.addEventListener('touchstart', () => {}, { passive: true });

This tells Safari:

“This page handles touch events — treat taps as immediate clicks.”

After adding this, links and buttons respond instantly with a single tap, while the { passive: true } option ensures it won’t block scrolling or cause performance issues.

✅ Safe to use: yes, it’s a no-op listener.
📱 Affects: only iOS Safari (WebKit).
💡 Why it works: it forces Safari to initialize its touch event system early, avoiding the hover-delay behavior.

Hopefully this will soon be a problem of the past.
I just merged a mitigation in LLVM that should prevent this false positive error. See: https://github.com/llvm/llvm-project/pull/159618

The PR makes it so that catch parameters are not instrumented on Windows (instrumenting them correctly is complex), which should prevent this issue altogether. It means less coverage (inside a catch-block), but in practice it should mean we can run ASan on more unmodified Windows try-catch code.

Yes, it does. When you define an entity some fields can be null or empty. Suppose you design an model contact name field which can be null. The Django Template System (DTS) will render it as None. Therefore, you could provide a more adequate display value: a default_if_none:"--", for example.

No clue why CopyFromRecordset wouldn't dump the data for these 2 systems but with the help from @frankball, copying the data into a variant, then looping through each value in the variant and writing it to a cell got it to work. Since I'm writing the field names in the first row, need to skip down to second row when writing the array. Final code:

Public Sub SendRecordsettoExcel(Rs As DAO.Recordset)
Dim intMaxCol As Integer
Dim intMaxRow As Integer

Dim objXL As Object
Dim objWkb As Object
Dim objSht As Object
Dim vaTmp() As String
Dim x, y As Long
Dim records As Variant
Dim FineColumnLetter As String
Dim BirthYearColumnLetter As String

On Error GoTo ErrorHandler

    If Not Rs.EOF Or Not Rs.BOF Then

        intMaxCol = Rs.Fields.Count
        If Rs.recordCount > 0 Then
          Rs.MoveLast
          Rs.MoveFirst
          intMaxRow = Rs.recordCount + 1
          Set objXL = CreateObject("Excel.Application")
          With objXL
            '.Visible = True
            Set objWkb = .Workbooks.Add
            Set objSht = objWkb.Worksheets(1)

            ' get the column headers
            ReDim vaTmp(Rs.Fields.Count)
            For x = 0 To Rs.Fields.Count - 1
                vaTmp(x) = Rs.Fields(x).Name
                If (Rs.Fields(x).Name = "Fine") Then
                    FineColumnLetter = Chr(x + 65)
                End If
                If (Rs.Fields(x).Name = "BirthYear") Then
                    BirthYearColumnLetter = Chr(x + 65)
                End If
            Next
            objSht.cells(1, 1).Resize(1, Rs.Fields.Count) = vaTmp

            Rs.MoveFirst
            records = Rs.GetRows(intMaxRow)
            For x = 0 To Rs.recordCount - 1
                For y = 0 To Rs.Fields.Count - 1
                    objSht.cells(x + 2, y + 1).Value = records(y, x)
                Next
            Next
          End With
        End If
    End If

ErrorHandler:

    Set objWkb = Nothing
    Set objSht = Nothing
    Set objXL = Nothing
End Sub

Oh. Propably didn't check one more important think - whether my notebook is connected to the compute :) After switching from "Connect" to my Cluster it ran. I used unity catalog connection
enter image description here

Top Thanks for the code it finally works fine :)

I had same error in headless mode:

selenium.common.exceptions.ElementNotInteractableException: Message: element not interactable
         (Session info: chrome=141.0.7390.108);

✅ I had solve it by just adding argument: --start-maximized to chrome_options.

That is because you wrapped in a try/catch. So the before_insert didn't understand that it should abort. raise some Exception and lets see.

Why I Switched Back to jQuery in 2025 (And Why You’re Wrong for Not Doing It)

https://medium.com/@hhartleyjs/why-i-switched-back-to-jquery-in-2024-and-why-youre-wrong-for-not-doing-it-f247b9830fcb

I've wrote a Android in App in which I want to Integrate a Chat via XMPP. I've logged in Facebook with permission for XMPPLogin. But when I want to login in Facebook from XMPP I receive this XML Response:

RCV  (1095775528): <failure xmlns='urn:ietf:params:xml:ns:xmpp-sasl'><not-authorized/></failure></stream:stream>

causing this Error:

SASL authentication failed using mechanism X-FACEBOOK-PLATFORM: 
at org.jivesoftware.smack.SASLAuthentication.authenticate(SASLAuthentication.java:341)
at org.jivesoftware.smack.XMPPConnection.login(XMPPConnection.java:242)

Look at the migration guide info: https://docs.aws.amazon.com/sdk-for-java/latest/developer-guide/client-configuration.html#client-configuration-http

I hope you’re doing well. I’m currently encountering the same issue with Angular 17 and AngularFire when running unit tests using Jasmine + Karma. The error message says: “AngularFireModule is not provided in AppModule”, even though I’m properly configuring the Firebase providers inside the test setup.

I was wondering if you eventually found a reliable solution or workaround for this? Any insights or updated configuration examples for Angular 17 would be greatly appreciated. Thanks in advance for your help!

Follow-up on Firebase Realtime Database Leaderboard Solution :

Hi @Frank, thanks for the answer! 🙏

I was able to solve it by updating the query:

var task = reference.Child("users")  
    .OrderByChild("levelContainer/highestScore")  
    .StartAt(currentUserScore, userId)  
    .GetValueAsync();

What I have achieved :

• Retrieve the top 10 users ordered by highestScore.
• Retrieve the current user’s rank in the leaderboard (even if they are not in the top 10).

Follow-up Questions :

1. Would Firestore be a better choice for this type of use case where I need to query/search over a large list of users?
2. I’ve set the rule as you mentioned — but does setting .indexOn also sort the data in the Firebase console UI? (Because it didn’t appear sorted there.)

My tag is Gael_et_co sry about that ^^

I know this was from a while ago but this fixed the issue I was having of the wrong schema being used when bringing an existing report from one environment to another, which there doesn't seem to be a clear answer for anywhere but here. This was an ideal solution because it removed the schema from the query, which allows us to easily move the report around environments.

For your question, I'm sure you've already solved it but you can duplicate your error by copying your SQL query into a standard SQL Developer interface which will tell you where the problem is more specifically. I'm guessing you may be missing a synonym or something.

I hope that you have resolved your issue by now.

Please note that rbibutils doesn't depend on any R packages outside base R. In fact, it is pure C within base R. The fact that in the log provided by you, stuff is loaded from RcppEigen and StanHeaders (with which rbibutils has nothing to do) suggests that on your system some compiler related paths are corrupted to look into places where they should not.

You certainly don't need to change any R settings despite what some answers here suggest. Of course, if you can install on your system a pre-prepared binary of rbibutils that will do (there are such binaries for the major Linux distributions).

As the author of rbibutils, I am always ready to consider bug reports, though this doesn't seem to be the case.

The problem of huge tree created by looping is solved in my product Tabbydb from KwikQuery.

Please refer to KwikQuery llc

There is a sample note book button where you can compare the perf difference in realtime.

You are hitting undefined behaviour by dereferencing a pointer which is not valid. Anything can happen. And no, the other pointer is not reset to nullptr when the first one is, these are independent variables.

You can download prebuilt mesa3d dll's these days. One example. Cheers!

I have also seen this issue and don't understand it.
Despite "ms-python.debugpy" being declared in the .devcontainer.json customizations.vscode.extensions it is not installed. I have to manually install

You could also inline the compare directly if you prefer, see below:

enum State {
  A = "A",
  B = "B",
  C = "C"
}

const prio: [State, number][] = [
  [State.A, 23],
  [State.B, 2],
  [State.C, 5]
];

prio.sort((a, b) => b[1] - a[1]);
console.log(prio);

That error usually appears when Facebook’s Business Manager endpoint hits an internal validation issue — often related to permissions or the parent account’s ownership status. Try these steps:

1. Double-check that the parent Business Manager has admin access and valid payment setup.

2. Use the latest API version (v19 or higher). Older versions sometimes throw generic Code 1 errors.

3. Verify your access token scopes: business_management, pages_manage_metadata, and ads_management.

4. If you’re calling the endpoint in quick succession, add a short delay (Facebook occasionally rate-limits /owned_businesses).

5. Finally, clear any pending Business creation requests in the parent manager before retrying.

If everything checks out and it still returns Subcode 1690192, it’s likely a temporary backend fault — retry after a few hours or open a Business Support ticket with your request ID.

By the way, interesting how meta-platform integrations now mirror decentralized systems — data ownership and verification layers almost resemble transparent ledgers used in blockchain or projects like Lucky Block, a casino platform that also focuses on verifiable fairness. Different field, same logic: transparent data flow and trust.

It looks like url rewrite has been added:
https://aws.amazon.com/blogs/networking-and-content-delivery/introducing-url-and-host-header-rewrite-with-aws-application-load-balancers/

but I can't get it to work :(

To enable Windows built-in firewall rules independently of localization (i.e., regardless of system language or region), you can use PowerShell or Command Prompt with rule identifiers that are consistent across locales.

https://sps-support.honeywell.com/s/article/Sample-IPL-Format

This is basic syntax for IPL. hope it helps

I encountered the same problem in windows, i have try to set the proxy but not work

I think you should mount it before assigning it to your tablespace.

I had similar wondering and this is what I found on AWS:

• Strings are Unicode with UTF-8 binary encoding. The size of a string is (number of UTF-8-encoded bytes of attribute name) + (number of UTF-8-encoded bytes).

• Numbers are variable length, with up to 38 significant digits. Leading and trailing zeroes are trimmed. The size of a number is approximately (number of UTF-8-encoded bytes of attribute name) + (1 byte per two significant digits) + (1 byte).

So Names of the attribute can play significant role it seems.