That text is garbled. OP could provide the formatted text eithe.http://t.me/DigitalVa0lt

I’ve found out a useful tutorial for this issue here.

https://blog.codeutility.io/frontend-development/handling-label-width-dynamically-in-amcharts-4-column-charts-b995d90080

They use hook to amchart events and reset max width of the label.

Hope this help,

try pip install --upgrade ultralytics

For anyone looking for a solution like the one @juliomalves gave, and one that doesn't involve the use of middleware on every request, you can:

1. generateStaticParams as dynamic routes (statically generated) as explained here: use generateStaticParams() in layout.tsx NextJS 14 app directory to get dynamic rotes prerendered (PPR)
2. rewrite the paths dynamically if you want to get the default locale at "/" (instead of "en/" for example if your default locale is English):

    const nextConfig = {
      ...
      async rewrites() {
        return [
          // Handle all paths that don't start with a locale 
          // and point them to the default locale static param: (en)
          {
            source: '/:path((?!(?:de|fr|es|it|el|pl|nl|en)(?:/|$)).*)',
            destination: '/en/:path*',
          },
        ];
      },
    };

I'm using JMeter 5.6.3 My JSON extractor is below. I'm not able to extract values. Did I miss something ? [enter image description here][1]

Connect Solar Panel to Relay: Connect the positive output of the solar panel to the common (COM) terminal of the relay. Connect the normally open (NO) terminal to the positive input of the charge regulator. The negative wire of the solar panel goes directly to the negative input of the charge regulator.

Relay Control: Use a control circuit (e.g., a microcontroller or a voltage sensor) to activate the relay. This allows the relay to switch on when charging is needed.

Connect to Charge Regulator: Connect the charge regulator’s output to the battery, ensuring correct polarity.

Safety Check: Double-check all connections for correct polarity and secure fitting to avoid short circuits.

For quality components, you can get them from reliable solar panel suppliers in UAE like Yellow Door Energy, SirajPower, and Sunergy Solar.

Pi0/w Pi0-2/w works out of the box. for pi pico w try https://github.com/sidd-kishan/PicoPiFi

You can try to recover your database from the "SUSPECT" state:

1. Set the Database to Emergency Mode:

   ALTER DATABASE [MYDB] SET EMERGENCY;

2. Set the Database to Single User Mode:

   ALTER DATABASE [MYDB] SET SINGLE_USER WITH ROLLBACK IMMEDIATE;

3. Run DBCC CHECKDB with REPAIR_ALLOW_DATA_LOSS:

   DBCC CHECKDB ([MYDB], REPAIR_ALLOW_DATA_LOSS);

4. Set the Database to Multi-User Mode:

   ALTER DATABASE [MYDB] SET MULTI_USER;

I had the same issue. I installed another developer app and it overrode the settings of the factory developer app. When I uninstalled the installed developer app, the factory developer app became the default again. I was able to see the right settings, and my device appeared on Android Studio again.

upsertAsync is indeed asynchronous.

https://github.com/couchbase/couchbase-net-client

There is an easy approach for this. Assume we have topic A and topic B where A needs to use KafkaAvroDeserializer while B needs to use StringDeserializer. Utilize properties of @KafkaListner to set the required values.

@KafkaListener(
topics = "topicA",
properties = {
    "key.deserializer=io.confluent.kafka.serializers.KafkaAvroDeserializer",
    "value.deserializer=io.confluent.kafka.serializers.KafkaAvroDeserializer"
})
public void consumeTopicA() {}

@KafkaListener(
topics = "topicB",
properties = {
    "key.deserializer=org.apache.kafka.common.serialization.StringDeserializer",
    "value.deserializer=org.apache.kafka.common.serialization.StringDeserializer"
})
public void consumeTopicB() {}

Just fill in as the deeplink.

https://itunes.apple.com/app/idYOUR_APP_ID

I've already tested it and published an in-app event, feedback is welcome.

can refer to this:

https://docs.openvino.ai/2025/openvino-workflow/running-inference/inference-devices-and-modes/auto-device-selection.html

The Automatic Device Selection with OpenVINO selects the most suitable device for inference by considering the model precision, power efficiency and processing capability of the available compute devices.

check out the Issue links API - https://docs.gitlab.com/api/issue_links/ to get specifics about linked items.

I have developed a sms-rocket package for sending SMS through multiple providers, fully integrated with CodeIgniter 4. It might be useful.

Create a mapper

Mapper type - Attribute Importer
Attribute Name - ATTRIBUTE_FORMAT_URI
User Attribute Name - X509SubjectName

1. []T is a slice type. A variable with type []interface{} has a specific memory layout different from interface
2. cannot quickly assign something of type []MyType to something of type []interface{};

It turns out, the issue is the read size. Even though the file is one byte, I need to ask for 4K and only get back one byte. Here is the same example working by changing cb.aio_nbytes on line 111: wandbox.org/permlink/HgMjLPeQoPlgxarn

The final code that works in Cygwin is:

for file in $(find -name '[0-9]*.*');
do
filename=$(basename "$file")
name=${filename%.*}
dir=$(dirname "$file")
extension=${filename##*.}
new_name=`printf %04d.%s ${name} ${extension}`
new_name="$dir/$new_name"
echo $file
echo $new_name
mv -n $file $new_name
done

And yes, the code is an ugly hack. It can be polished a lot. But I don't care at this point as it works. Also, I left the echo statements in just to see what is the code doing during execution. They are not necessary.

Kill the peer node start process and run peer note pause -c channel. If you are running under a kubernetes POD, edit the deployment:

kubectl edit deployment -n namespace peer1-org-allshitrandomname and add the the following directive as container entrypoint / start peer:

command: ["sleep", "infinite"].

After that, get a pod shell:

k exec -i -t -n fabric peer1-f594fcfc5-46hlc -- /bin/sh

And run the: peer node pause -c channel_name

Command.

WITH tb AS ( 
SELECT LEVEL AS num -- n tree
FROM DUAL
CONNECT BY LEVEL <= 4)
SELECT 
SUBSTR('AaaaaBbbbbCccccDdddd', (5 * (num - 1)) + 1, 5) AS abcd
FROM tb;

Yo lo solucioné especificando en el modelo de usuario el campo específico en la base de datos, yo lo nombré password_hash en lugar de password.

namespace App\Models;

use Illuminate\Foundation\Auth\User as Authenticatable;

class Usuario extends Authenticatable
{
    protected $table = 'usuarios';
    protected $fillable = ['username', 'password_hash'];

    // Especifica el nombre de la columna de la contraseña
    public function getAuthPassword()
    {
        return $this->password_hash;
    }
    public function productos()
    {
        return $this->belongsToMany(Producto::class, 'usuario_producto', 'usuario_id', 'producto_id')
            ->withPivot('cantidad_asignada');
    }
}

Thanks @Mike M. Ctrl+K, M worked to get the menu where I could then select the language.

I just improved the message in https://github.com/GradleUp/shadow/pull/1287.

To flesh out Ryan's answer.

Install the required libraries via composer, as fpdi needs fpdf:

% composer require setasign/fpdi

% composer require setasign/fpdf

Using the library:

$pdf = new \setasign\Fpdi\Fpdi();

Select substring([xxx], 1, 5), substring([xxx], 6, 5), substring[xxx], 11, 5)

did you solve the problem? i have exactly the same issue.

die llllloooolllllllllllll.....

You can use UDF BS_CLOUD of Add-in A-Tools to get data from Google Sheets or Excel Online. Information of BS_CLOUD : https://bluesofts.net/Kien-thuc-Add-in-A-Tools/Cloud/Huong-dan-ham-BS_CLOUD-lay-du-lieu-tu-Google-Sheets-va-Excel-Online-ve-Excel

Since I can't use @IM_NAME LIKE ZZPATTN in the SELECT statement, I use another alternative: TYPES, RANGE OF and IN. I split the IM_NAME and append them to an internal table, and use IN to get the results. The reason I used TYPES, RANGE OF not TYPE, TABLE OF is to solve error "The line structure of the table "LT_PATTERNS" is incorrect."

Not straightforward as LIKE, but could provide similar result. Here are the sample code:

TYPES:
  ty_pattern TYPE STRING,  " Define type for pattern
  ty_pattern_range TYPE RANGE OF ty_pattern.  " Range table for pattern
DATA:
  lt_patterns TYPE ty_pattern_range,    " Range table for patterns
  lv_pattern TYPE STRING,       " Current pattern
  lv_length TYPE i.

" Initialize the pattern with '*' (wildcard)
lv_pattern = '*'.
CLEAR lt_patterns.
APPEND VALUE #( sign = 'I' option = 'EQ' low = lv_pattern ) TO lt_patterns.  " Add pattern to range

lv_length = strlen( IM_NAME ).  
  WHILE lv_length >= 1.
    lv_pattern = substring( val = IM_NAME len = lv_length ) && '*'.  
    lv_length = lv_length - 1.  " Reduce the length
    APPEND VALUE #( sign = 'I' option = 'EQ' low = lv_pattern ) TO lt_patterns.  
  ENDWHILE.

" Now use the range table in the SELECT statement
  SELECT ZZPRICE, ZZEFDAT INTO ( @PRICE, @EFDAT ) FROM ZZPRICE
        UP TO 1 ROWS
        WHERE ( ZZNAME = @IM_NAME OR ZZPATTN IN @lt_patterns ) 
ENDSELECT.

I simply created the directory "portainer-compose-unpacker" Enable relative path volumes set path. For some reason it this is what was denied but cloning worked without issue after manually doing that.

To help with this error message, include in your python file: import keras; import tensorflow.keras; and do not include import tensorflow as tf, whereas in the bash file include the version of your tensorflow for example include module load tensorflow//2.16.1-pip-py312-cuda122, I hope this helps to solve the problem.

I would say the REST API Client is the way to go. The Guidewire documentation for 10.2.4 (latest on-premise release) and Las Lenas (latest GWCP release) point to the exact same REST API Client documentation. Link below (GW partner/customer login required).

https://docs.guidewire.com/is/restapiclient/release/guide/REST-API-Client/topics/c_overview-rest-client.html

So using the REST API Client provided by Guidewire is the best option if you want your development effort on-premise to be able to move to GWCP. Setting up and using the REST API Client takes a bit more of time initially, but should be easier to maintain over the long run. Good luck!

This is how you can turn off this in Mac Vscode/ Cursor.

1. Go to setting

2. Search for “dart” in the search bar. Look for the setting labelled "Dart: Open Dev Tools" and select the option "Never."

Simon Mourier's comments is the answer is used. I have decompiled the generated tlb dll back into C# using ILSpy . Then pseudo-manually and with help of AI to speed it up I converted attributes on the interfaces to using GeneratedComInterface instead of ComImport. Fortunately, the COM interface I am using is relatively basic, so this wasn't that difficult with no custom marshalling required.

Simon's comment provides a link with documentation for the basic cases I needed: https://learn.microsoft.com/en-us/dotnet/standard/native-interop/comwrappers-source-generation

@drodir

Thank you so much, with your input, I was able to discover that I needed the import() method in my conanfile.py, to import files from lib_b to lib_a, as shown below. Your suggestion of knowing the underlying conan install led me to the answer.

   def imports(self):
      self.copy("*.h", dst="src", src="include") 

Using the ARN works instead of using the names and this will throw an error aws ssm get-parameter --name "ec2-image-builder/devops/rhel8-stig-golden-image"

An error occurred (ValidationException) when calling the GetParameter operation: Parameter name: can't be prefixed with "ssm" (case-insensitive). If formed as a path, it can consist of sub-paths divided by slash symbol; each sub-path can be formed as a mix of letters, numbers and the following 3 symbols .-_

Instead use: $ aws ssm get-parameter --name arn:aws:service:us-east-1:0000000000:parameter/myparam

NOTE: If the parameter has a forward slash already do not add it $ aws ssm get-parameter --name arn:aws:service:us-east-1:0000000000:parameter${myparam}

There's a much easier way to do this. If you add an @key attribute, Blazor will know that when the key value has changed, it needs to remove the old DOM element and insert a new for the new URL. So all you'd need here is this:

<video @key="currentVideoUrl" id="videoTagId" autoplay width="1080" height="720">

Once you have saved your .html file to a location on your computer, navigate to that file location and click the document file. It should open to your webpage with your default browser.

https://pypi.org/project/bmdf/

For those coming across: this is a more currently supported tool for click documenting

Put internalA and internalB in a seperate package and make them package-private. that way, the client cannot @Autowire them.

Same problem, here is a sample extension: https://github.com/justinmann/sidepanel-audio-issue

https://learn.microsoft.com/fr-fr/powershell/module/microsoft.powershell.core/about/about_execution_policies?view=powershell-7.5

Very usefull when PS block script execution.

I read that assert.deepEqual() is deprecated. And we should use assert.deepStrictEqual() instead.

I've seen an answer somewhere that I highly agree with, but I can't find it now so I'll put it here.

The practical reason that RB tree is more widely used than AVL tree in standard libraries is probably that the RB tree, and not the AVL tree, is elaborated in CLRS.

Believe it or not, publicity is a critical factor in winning, especially when alternatives are similar in merit.

I'm getting the exact same issue, and I wasn't before. What Mac OS are you using? I recently updated Xcode and OS and I'm thinking it might be related.

I been researching on this from very long time could you help me more on this personally ?

enter code hereenter image description here

enter link description here

enter link description here

I am not sure how to install Light ingest, on the document there is link that take you to a githib repo where you see a installer for windows but for some reason I am unable to install LightIngest

If you use react, check if you need to include"css.enabledLanguages": ["typescriptreact"] to your settings.json in vscode

Thanks to @john Bollinger! His suggestion got me on track.

I have created global variable: pthread_cond_t condition;

Initialized this one with defaults: r = pthread_cond_init(&condition, NULL);

And went to sleep by this: rt = pthread_cond_timedwait(&condition, &lock, &ts);

Where lock is a global lockfile (which I was already using anyways): pthread_mutex_t lock;

And ts is of struct timespec ts;

Because pthread_cond_timedwait does not wait for seconds to pass, instead until a specified time has come I just added the sleeping time: ts.tv_sec += waitSecs;

In the other thread it was very easy. Just inform the waiting thread: pthread_cond_signal(&condition); while the lock is blocked by the sending thread.

Works like a charm! Thanks again!

You can use the shortcut to "Toggle Secondary Side Bar"

Ctrl + Alt + B

or click this

(Windows/English)

I found out that xcode had a different version of the GoogleService-Info.plist than what I had in my VSCode. I'm not sure how this can happen, but it can.

Watch out.

In an elevated shell, make a link named whatever you want inside AppData/Local/Microsoft/WindowsApps to the Notepad++ executable:

mklink %LOCALAPPDATA%\Microsoft\WindowsApps\npp.exe "C:\Program Files\Notepad++\notepad++.exe"

That location is in the user's default path.

But when your provider verifies this pact, it expects size of the list per key to be be one always. If your provider returns list of size more than 1 for any key then it fails saying expected size is 1 but found 2 size for any key.

This looks to be a bug.

I had same problem. What I could found:

1. learn.microsoft.com web page contains a lot of examples how to get data of user Xbox account
2. This GitHub project implement Xbox API using Python.
3. My Github project do same, but I'm using C#

Try this - Years and Month: DateDiff("yyyy",[Start_Date],Date()) & " Years and " & DateDiff("m",[Start_Date],Date()) Mod 12 & " months"

A, you need to use a global variable e.g Public Shared Variable name as string... B, Create a new event handler for key-press event of the textbox, and u would need to use e.handled.

I recommend using c sharp instead and give up.

1. Remover los node_modules ==>> rm -r node_modules

2. Remover package-lock.json

3. Limpiar la cache de instalacion de modulos npm cache clean --force

4. Reinstalar todos los modulos npm install

In the Posh Code site, their PowerShell Practice and Style guide, Don Jones, M. Penny, C. Perez, J. Bennett and the PowerShell Community suggest function documentation best practice places the comment-based help inside and at the top of the function it describes. Inside, so it does not get separated from the function, and at the top so developers see them and remember to update them.

In order to ensure that the documentation stays with the function, documentation comments should be placed INSIDE the function, rather than above. To make it harder to forget to update them when changing a function, you should keep them at the top of the function, rather than at the bottom.

Documentation and Comments

And if you prefer something closer to your style, this is also supported:

let s;
if condition {
    s = "first".to_string();
} else {
    s = "second".to_string();
}

This was not an issue with my ingress rules or any K8 configuration. It was with how the path was defined in ingress alongside how the express app serves requests. Express app was trying to serve static files on the root "/" path instead of the path I defined in ingress /webui.

Had to adjust static file middleware in express to serve on /webui path and adjusted necessary routes in express and I am able to access things properly.

I have got this error.

[error] 2547571#2547571: *1971306 [client 172.31.***.***] ModSecurity: Access denied with code 400 (phase 2). Matched "Operator `Eq' with parameter `0' against variable `MULTIPART_STRICT_ERROR' (Value: `1' ) [file "/etc/nginx/conf/modsecurity.conf"] [line "59"]

sudo vim /etc/nginx/conf/modsecurity.conf

and comment all out for "MULTIPART_STRICT_ERROR" part. and

sudo systemctl restart nginx

@jason Thanks!! you saved my day.

f

f's in the chat for formatting assistant

I can confirm that Rich's solution is partially correct. As of today, Bootstrap 5 does not seem to recognize the class "overflow-x-scroll". I have tried it many times without success. The solution is to keep it as "overflow-auto", which will work perfectly. Thanks, Rich for the great start though!

A link to a solution that is not mine https://catchts.com/union-array

Create a custom function like so:

def get_element(object, index, fallback):
    try:
        return object[index]
    except IndexError:
        return fallback

Then, you call:

get_element(foo_val, 3, None)

And if 3 is in range, it returns the correct value; if not, it returns None.

A sustainable solution would be to deploy a PyPi Proxy mirror. Sonatype offers a great solution to this, but, proxy mirror will need some form of access to PyPi either by downloading the packages (I've used git for updating packages as needed) to some internal store which nexus references or by using a proxy directly to PyPi.

Este error está asociado con muchos factores, especialmente con el inicio del proyecto de tratamiento y la gestión de giro en su aplicación. Revisaremos algunas posibles razones y soluciones:

Problemas de limpieza de monos: [mono] contacte a Debgar-Agent Mensaje: UNA10. Esto puede estar vinculado al error al comienzo de la formación de red o la estación de procesamiento.

solución:

IP10 Si el emulador no se conecta a la planta de tratamiento, intente cancelar la purificación remota y ensamblarla sin opciones de purificación. Puede intentar restaurar el emulador o iniciar un nuevo proyecto para ver si el error se corrige. Mutex distinguido (control de muttex): error [no. Poppinornot] Muttex. CC: 432 para destruir Muttex con propietarios o competidores. Propietario: 10086 indica que Muttex tiene un problema, que puede ser el resultado del procesamiento incorrecto de recursos o recursos comunes.solución:

Verifique si su aplicación utiliza indicadores de interconexión y hay algún recurso que participe en múltiples indicadores interconectados sin una sincronización adecuada. Esto puede causar corrupción de memoria o errores en la memoria. Si está utilizando la tercera biblioteca de sección, asegúrese de actualizarla correctamente, ya que puede ser un error relacionado con parte de la versión de dependencia. Errores en la Mezquita de recolección: usted ha mencionado que ha deshabilitado "el uso de la mezquita de basura de contratación", pero asegúrese de implementar la capacitación correctamente y que no hay otras formaciones contradictorias.

solución:

Simplemente deshabilite la recolección de residuos en la formación de su proyecto. Intente usar un emulador con una carga baja o incluso un dispositivo real para ver si se soluciona el problema. Signo 6 (Sigbart): Sigabrt generalmente indica que la aplicación se ha cancelado debido a un caso de error agudo, como la mala memoria o la gestión de la interconexión.solución:

Verifique el registro de su aplicación antes de no encontrar una excepción o un posible error para controlar a las Naciones Unidas en el flujo de las instrucciones de su software. También puede usar herramientas como ADB Logkat para obtener información más detallada sobre la causa de la falla. Si aún no ha tenido éxito, le recomiendo probar un dispositivo real, porque a veces los problemas del emulador pueden estar relacionados con su formación o fuentes limitadas. Puede intentar limpiar y renovar el proyecto, y eliminar Android Studio (por archivo> Archivo> Lappel / no correcto re -restart).

This issue has different results depending on how many times/much duration in exposure each screen and each ad unit has. If each screen has many times/much duration in exposure, putting in a dedicated ad-unique unit for each screen will maintain revenue and aloso make analysis easier. However, if you distribute different ad units to screens with low exposure, it may help analysis, but overall revenue will be low. This issue has been a bit more lenient on Android since Apple introduced its ATT(AppTrackingTransparency) policy.

Please take a look at this article, it provides a more scalable approach is to store configurations in a JSON file and dynamically load them into Terraform, making the setup more modular and readable.

https://medium.com/@ameerahaiderrizvi/dynamic-aws-elastic-beanstalk-configuration-with-terraform-and-json-e38ac21627b7

You can do it this way:

We use Binding Dependency between ISomeInterface and ISomeInterface <Class1>. We also define a connection Realization between ISomeInterface<Class1> and Class1Repo which implies the use of Class1 where there was a parameter T (in our case, as a return type).

Tried using the new SSH for Infrastructure and VsCode couldn't finish the connection, was erroring with ssh child died and other errors. too bad

I had a different issue, we use Cloudflare SSH for infrastructure, like a tunnel, and seems this SSH tunneling was affecting VsCode, so I switched back to direct SSH connection and it worked.

too bad

This is the best article I found. https://www.confluent.io/blog/error-handling-patterns-in-kafka/ It explains DLQ pattern.

When I set <BlazorCacheBootResources>false</BlazorCacheBootResources> the site doesn't work on both - desktop and mobile platforms.

Select Case node.tagName should be elem.tagName

This suddenly happened to my win 11 entra id hybrid machine in Feb 2025. I don't know why--I found no applocker, SRP or WDAC policies, though it was acting like one was applied--I also could not run batch files.

To fix, I had to create a default policy in Local Security Policy under Software Restriction Policies. I just left the Security Levels at the default "unrestricted" and this fixed things after a reboot.

In my case I had to use AWS CLI to configure credential This guide will help

https://medium.com/@damipetiwo/seting-up-s3-bucket-in-aws-using-terraform-5444e9b85ce6

Given the exception stack trace, looks like the graphics subsystem is not properly initialized. When deserializing and instantiating the report, some of its properties try to initialize the graphics engine in order to determine some basic metrics like machine's DPI.

Since you run your report on AWS, by default Telerik Reporting will use Skia graphics engine, which requires installing the Telerik.Drawing.Skia NuGet package. Additionally, two other libraries must be installed:

• sudo apt-get install -y libfreetype6
• sudo apt-get install -y libfontconfig1

Check the help article for more info: https://docs.telerik.com/reporting/getting-started/installation/dot-net-core-support#deploying-on-linux

Macrobenchmark runs on a variant that is based off of the release variant. If your app is having those issues when running the benchmark, chances are it's also facing similar problems on your regular release build. I'd begin by checking if your app is behaving normally on this device with release configuration. This has been the problem I've encountered.

.ply files are not supported in Unity without plugins. You can convert to .obj, .fbx etc by downloading Blender, importing the .ply file, and exporting as whatever you want. The issue is that .ply files use vertex colouring, and .obj files need to use a texture file as a PNG or something.

If you convert to .obj using MeshLab, the colours will only show up in Meshlab (since Unity will not import vertex colours from .obj files since that is not part of the .obj spec, MeshLab just adds it the .obj anyway).

What you should do is export the .ply as an .fbx file using Blender, and then import that into Unity. Then to get the colours to show up, you need to go to your .fbx file, and go to the inspector. Then click "extract materials", and edit the material that pops out (should be called material_001.mat or something). Edit it by changing shader from "Standard" to custom/unlit vertex shader. And then you should get colours to show up in Unity.

I had the same error message. I was able to fix the issue with the following command:

serverless deploy --aws-profile PROFILE_NAME

For me, specifying the profile was the solution.

When it comes to high-performance rendering in Windows/WPF, Direct3D is the answer. But one needs to incorporate Compute Shaders to achieve the “most performant” rendering.

System.Windows.Interop.D3DImage offers a direct interface to Direct3D (one has to render to a texture target). Note, Winforms is faster and offers a more direct interface via the Device Context / direct Render Target, so using WindowsFormsHost to embed a Winform control is technically the fastest way to draw in WPF. I’d recommend your solution be interface independent but one could stick to D3DImage. This approach requires a mixture of c++ and c#.

Pass your data to the gpu side when creating the buffer/shader resource view (50 lines), set up your unordered access view (50 lines), and write your shader code hlsl (50 lines) to solve your problem in a parallelism methodology (your data buffer size to vertex buffer size should be a fixed proportion so break up the work in chunks of 500 points for example) and ultimately the shader produces your vertex buffer. You will also need to understand constant buffers, a simple pixel shader, and the inspiration hits when calling "Dispatch". There’s example code on creating your Device and Immediate context. All in all, no more than 750 lines of code. This is the fastest way to draw in WPF if you consider all possible solutions, which some readers should. Given many current and future computers will have Integrated GPUs, APUs/NPUs or real discrete GPUs, it's past time to start learning compute shaders and Direct3D and Vulkan. I’ve written a most performant way to draw in WPF and Winforms and it can be no-hassle simple-click experienced at Gigasoft.com for those interested in the most performant way to draw in WPF (100 million points fully re-passed and re-rendered) and optionally Winforms.

I had the same issue in PowerShell.

Full disclosure: I didn't have any luck finding the "proper" way to do this in PowerShell, so I had to hack something out...This is what I have so far. I wouldn't consider this to be the "proper" way, it's just a way that is actually working for me. I borrowed snippets from various examples to kludge this together.

# CookieManager isn't available until Initialization is completed.  So I get it in the Initialization Completed event.
        $coreWebView2Initialized = {
            $script:cookieManager = $web.CoreWebView2.CookieManager;
            $script:cookies = $script:cookieManager.GetCookiesAsync("");
            $script:coreweb2pid =  $web.CoreWebView2.BrowserProcessId;    #this I used later to find out if the webview2 process was closed so I could delete the cache.
        }
        $web.add_CoreWebView2InitializationCompleted($coreWebView2Initialized);


# Once the initial naviation is completed I hook to the GetCookiesAsync method.
        $web_NavigationCompleted = {
            $script:cookies = $script:cookieManager.GetCookiesAsync("");
        }
        $web.add_NavigationCompleted($web_NavigationCompleted)

# With my particular situation, I wanted to deal with MFA/JWT authentication with a 3rd party vendor talking to our MFA provider.  The vendor uses javascript to change pages which dosen't trigger a webview2 event.  I added a javascript observer that watched for the documentElement.innerText for the "You can close" text that the 3rd party provider would return indicating it's ok to close the browser.  Once this text came through I used the webview.postMessage('Close!') to send a message back to my script so it could close the form and cleanup everything.
#  The specific part of this that addressed the getting async cookies part is adding the GetCookiesAsync hookup once the initial page is loaded.  For me, the cookies I wanted were HTTP Only cookies so I had to do it this way to get at them.

   $web_NavigationCompleted = {
        $script:cookies = $script:cookieManager.GetCookiesAsync("");
        $web.CoreWebView2.ExecuteScriptAsync("
            //Setup an observer to watch for time to close the window
            function observerCallback(mutations) {
                if ( (document.documentElement.textContent || document.documentElement.innerText).indexOf('You can close') > -1 ) {
                    //send a Close! message back to webview2 so it can close the window and complete.
                    window.chrome.webview.postMessage('Close!');
                }    
            }    
            const observer = new MutationObserver(observerCallback);
            const targetNode = document.documentElement || document.body;
            const observerconf = { attributes: true, childList: true, subtree: true, characterData: true };
            observer.observe(targetNode, observerconf);
        ");
    }
        $web.add_NavigationCompleted($web_NavigationCompleted)

#   Once the form "Close!" message is generted, the cookie I want should be there.  This is ignoring any of the misc innerText events that happen and just waiting for the "Close!".
#   I grab the specific HTTP Only cookie and return the value.

   $web.add_WebMessageReceived({
        param($WebView2, $message)
 
        if ($message.TryGetWebMessageAsString() -eq 'Close!') {
            $result = ($cookies.Result | Where-Object {$_.name -eq "The_Name_of_the_HTTP_ONLY_Cookie_I_Wanted"}).value
            $web.Dispose()
            # Cleanup cache dir if desired - wait for the webview2 process to close after the dispose(), then you can delete the cache dir.
            if ($Purge_cache) {
                if ($debug) {write-host "form closing webview2 pid "$script:coreweb2pid -ForegroundColor blue} 
                    $timeout = 0
                    try
                    {
                        while ($null -ne [System.Diagnostics.Process]::GetProcessById($script:coreweb2pid)  -and $timeout -le 2000)
                        {
                            if ($debug) {write-host "Waiting for close pid "$script:coreweb2pid -ForegroundColor blue} 
                            Start-Sleep -seconds 1                      
                            $timeout += 10;
                        }
                    }
                    catch { }
                    if ($debug) {write-host "cleaning up old temp folder" -ForegroundColor blue} 
                    
                    $OriginalPref = $ProgressPreference 
                    $ProgressPreference = "SilentlyContinue"
                    $null = remove-item  "$([IO.Path]::Combine( [String[]]([IO.Path]::GetTempPath(), 'MyTempWebview2CacheDir')) )" -Recurse -Force 
                    $ProgressPreference = $originalpref            
            }
            $form.Close()
                
            return $result.tostring()
        }
    })

There's probably a cleaner way to do this. For now, it works. It drove me crazy because if you dig for anything about doing MFA authentication with PowerShell you end up with O365 examples...Like the only thing we'd use PowerShell for is O365? If/when I get my authentication module polished enough to post, I'll add it to GitHub and update this post. I spent a lot of time running around in circles trying to do this in PowerShell, hopefully this removes that barrier for folks.

Note: I've also tried setting incognito mode for the Webview2 browser (many ways). That doesn't work, not so far anyway. I don't like any of this authentication data being written to disk in cache or any other way, I want it to be momentary authorization for the use of the script and then gone...I am continuing to work on making this not cache things, but for now at least I have a path to delete the browser environment cache.

Cheers.

Thank you for the feedback! I did realize that the size is what was throwing things off...looks like GAM changed up their transcoding outputs and the formally working size no longer spits out.

In case, you create a Browser, can use

if "--headed" in str(sys.argv):
   Browser.browser_parameters["headless"] = False

From where you are trying to pull the code? If you are trying to pull the code for the first time then you need to clone the code for the first time so you can refer this link to checkout multiple repos from different code repository managers: https://learn.microsoft.com/en-us/azure/devops/pipelines/repos/multi-repo-checkout?view=azure-devops

I could also see you are using -script without any task it is good practice to use the task for such implementations for consistency.

is there any way to check if currently the device has airplane mode turned on through ADB?

You can try to execute those new commands (without requiring root permissions):

• If you want to know if the airplane mode is enabled or disabled:

  adb shell cmd connectivity airplane-mode

• If you want to enable the airplane mode:

  adb shell cmd connectivity airplane-mode enable

• If you want to disable the airplane mode:

  adb shell cmd connectivity airplane-mode disable

The issue was created on the repo link

This was suggested: browser_args.append(f"--headless=new")

I had the same issue thank you very much as it worked.

Happens to me too. Clear cookies helped for the first screen, but when navigating to other screens it happened again.

Based on the comment by @Jeffrey D., I tried using the Safari browser, and indeed it helped, the problem did not reproduce.

I'm new to polars, so unfortunately I also don't know how to use the pull request 13747 updates, but issue 10833 had a code snippet and I tried to adapt your approach as well. I tried 3 different approaches shown below and got the following timings for a fake dataset of 25,000 sequences

• Took 40.06 seconds for the approach you show above
• Took 13.12 seconds for my adaptation of your approach
• Took 2.35 seconds for the map_elements_approach from the github issue

Here's the code:

#from memory_profiler import profile
import polars as pl
import numpy as np
import time

np.random.seed(0)
num_seqs = 25_000
min_seq_len = 100
max_seq_len = 1_000
seq_lens = np.random.randint(min_seq_len,max_seq_len,num_seqs)
sequences = [''.join(np.random.choice(['A','C','G','T'],seq_len)) for seq_len in seq_lens]
data = {'sequence': sequences, 'length': seq_lens}

df = pl.DataFrame(data)

ksize = 24


def op_approach(df):
    start = time.time()
    kmer_df = df.group_by("sequence").map_groups(
        lambda group_df: group_df.with_columns(kmers=pl.col("sequence").repeat_by("length"))
        .explode("kmers")
        .with_row_index()
    ).with_columns(
        pl.col("kmers").str.slice("index", ksize)
    ).filter(pl.col("kmers").str.len_chars() == ksize)
    print(f"Took {time.time()-start:.2f} seconds for op_approach")
    
    return kmer_df

def kmer_index_approach(df):
    start = time.time()
    kmer_df = df.with_columns(
        pl.int_ranges(0,pl.col("length").sub(ksize)+1).alias("kmer_starts")
    ).explode("kmer_starts").with_columns(
        pl.col("sequence").str.slice("kmer_starts", ksize).alias("kmers")
    )
    print(f"Took {time.time()-start:.2f} seconds for kmer_index_approach")
    
    return kmer_df
    

def map_elements_approach(df):
    #Stolen nearly directly from https://github.com/pola-rs/polars/issues/10833#issuecomment-1703894870
    start = time.time()
    def create_cngram(message, ngram=3):
        if ngram <= 0:
            return []
        return [message[i:i+ngram] for i in range(len(message) - ngram + 1)]

    kmer_df = df.with_columns(
        pl.col("sequence").map_elements(
            lambda message: create_cngram(message=message, ngram=ksize),
            return_dtype = pl.List(pl.String),
        ).alias("kmers")
    ).explode("kmers")
    print(f"Took {time.time()-start:.2f} seconds for map_elements_approach")
    
    return kmer_df

op_res = op_approach(df)
kmer_index_res = kmer_index_approach(df)
map_res = map_elements_approach(df)


assert op_res["kmers"].sort().equals(map_res["kmers"].sort())
assert op_res["kmers"].sort().equals(kmer_index_res["kmers"].sort())

The kmer_index_approach is inspired by your use of str.slice which I think is cool. But it avoids having to do any grouping and it first explodes a new column of indices which might require less memory than replicating the entire sequence before replacement with a kmer. Also avoids having to do the filtering step to remove partial kmers. This results in an extra column kmer_starts which needs to be removed.

The map_elements_approach is based on the approach mentioned in the github issue where mmantyla uses map/apply to just apply a python function to all elements.

I'm personally surprised that the map_elements approach is the fastest, and by a large margin, but again I don't know if there's a different better approach based on the pull request you shared.

Were you able to make any progress on this?

Try using another template, use "django_tables2/bootstrap4.html" instead of "django_tables2/bootstrap-responsive.html"

You shouldn't be using a variable before a value is assigned to it. Most compilers will inform you of that, which generally means a program logic mistake. Sticking a value on a declaration masks those errors, and leads to bug.

Personally, I don't like the idea.

As of Feb 2025, the following works for Google Finance

=GOOGLEFINANCE("Currency:BTC/USD")
=GOOGLEFINANCE("Currency:ETH/USD")

Well, this question was asked quite a while ago, and it happens that now I'm helping my kid learn programming by making minecraft plugins so I encountered the same issue :)

This Spigot guide for debugging local/remote server is very useful: https://www.spigotmc.org/wiki/intellij-debug-your-plugin/ I verified that the local server debugging works well.

Essentially, you define a run/debug configuration, which allows you to not only start/debug your minecraft server, but you can also "reload classes" (Default shortcut is Control+Shift+F9) which does "hot swapping" allowing you to modify code on-the-fly, reducing even further the overhead per code modification iteration in many cases.