use env MSYS_NO_PATHCONV=1

MSYS_NO_PATHCONV=1 adb -s xxx push local_file /mnt/sdcard

Cant we just rename the java.exe and other related executables to the version they are in like java7 if it is v7 and java19 if it is java v19.0.1 or something like that. I did that physically and i dont know if it has any side effect or something

According to https://cloud.google.com/storage/docs/deleting-objects#delete-objects-in-bulk one should use "lifecycle configuration rules":

To bulk delete objects in your bucket using this feature, set a lifecycle configuration rule on your bucket where the condition has Age set to 0 days and the action is set to delete . Once you set the rule, Cloud Storage performs the bulk delete asynchronously.

To just delete a directory, "prefix" can be used.

in windows i change this file : C:\Users<Your-User>.gradle\gradle.properties

tnx

These are my routes

Route::group(['middleware' => ['VerifyCsrfToken']], function () { Route::get('login/{provider}', [SocialController::class, 'redirectToProvider']); Route::post('login/{provider}/callback', [SocialController::class, 'handleProviderCallback']); });

awesome, what about the g4f for the generate-image endpoint? I had it working on the server but for some reason after adding web-search abilities, the /generate-image endpoint isn't working for the Flux model? I ran commands to the API and it came back finding the multiple versions like Flux and Flux-artistic endpoints for image generation but its throwing an error not found on my end. Everything else works, its strange.

you can add this in your yaml file

assets: - lib/l10n/

I recently completed a project called WordSearch.diy, an AI-powered word search generator. It uses artificial intelligence to dynamically create word search puzzles based on custom inputs. Feel free to check it out at WordSearch.diy.

Would love to hear any feedback or answer questions about its development process!

let connectedScenes = UIApplication.shared.connectedScenes
let windowScene = connectedScenes.first(where: { $0.activationState == .foregroundActive }) as? UIWindowScene
let window = windowScene?.keyWindow

This worked well for me.

The simplest way that I achive this is:

SimpleDateFormat( "yyyy-MM-dd'T'HH:mm:ss.SSSXXX", Locale.getDefault() ).format(System.currentTimeMillis())

Permanent Residency (PR) visas offer long-term residency, citizenship pathways, free education, healthcare, and more. Explore skilled, family, employer, or investment-based options. Learn more: https://www.y-axis.com/visa/pr/

[105] apct.a(208): VerifyApps: APK Analysis scan failed for com.example.fitnessapp com.google.android.finsky.verifier.apkanalysis.client.ApkAnalysisException: DOWNLOAD_FILE_NOT_FOUND_EXCEPTION while analyzing APK at aolk.b(PG:1287) at aolk.a(PG:13) at org.chromium.base.JNIUtils.i(PG:14) at bjrd.R(PG:10) at aonn.b(PG:838) at aonn.a(PG:307) at bjmf.G(PG:40) at bjmf.t(PG:12) at apct.a(PG:112) at aonn.b(PG:788) at bjkp.nQ(PG:6) at bjrm.run(PG:109) at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1133) at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:607) at mwa.run(PG:1014) at java.lang.Thread.run(Thread.java:761) 2025-01-17 21:26:31.806 3410-3410 System com.example.fitnessapp W ClassLoader referenced unknown path: /data/app/com.examp i am facing this error plz help me

Vayfiyi şifresini değiştiremiyorum

Hi I just want to echo this issue It is exactly happening on our MSKC as well. our MSK is 3.6.0 and MSKC 3.7.x and this periodic AdminClient Node disconnect -1 and MSK Connect graceful shutdown is seen every 6mins as you mentioned. Currently we rollback the MSKC back to 2.7.1 But need an explanation by AWS. Thanks for sharing this issue. so I can tell I am not along.

Import global css file into root layout.jsx file app/layout.jsx

I've decided to write a program that runs into an infinite execution of the method where I was testing (to automate the testing process).

Here's the code snippet I was working with to achieve what Ken White and Frank van Puffelen recommend.

import 'dart:math';

import 'package:flutter/material.dart';
import 'dart:developer' as developer;

class FourDigitCodeGenerator extends StatefulWidget {
  const FourDigitCodeGenerator({super.key});

  @override
  State<FourDigitCodeGenerator> createState() => _FourDigitCodeGeneratorState();
}

class _FourDigitCodeGeneratorState extends State<FourDigitCodeGenerator> {
  String? _code;

  void _generateCode() { // version 1
    Set<int> setOfInts = {};
    var scopedCode = Random().nextInt(9999);
    setOfInts.add(scopedCode);

    for (var num in setOfInts) {
      if (num < 999) return;
      if (num.toString().length == 4) {
        if (mounted) {
          WidgetsBinding.instance.addPostFrameCallback((_) {
            setState(() {
              _code = num.toString();
              developer.log('Code: $num');
            });
          });
        }
      } else {
        if (mounted) {
          WidgetsBinding.instance.addPostFrameCallback((_) {
            setState(() {
              _code = num.toString();
              developer.log('Code: not a 4 digit code');
            });
          });
        }
      }
    }
  }

  void _generateCode2() { // version 2
    Set<int> setOfInts = {};
    var scopedCode = 1000 + Random().nextInt(9000);
    setOfInts.add(scopedCode);

    for (var num in setOfInts) {
      // if (num < 999) return;
      if (num.toString().length == 4) {
        if (mounted) {
          WidgetsBinding.instance.addPostFrameCallback((_) {
            setState(() {
              _code = num.toString();
              developer.log('Code: $num');
            });
          });
        }
      } else {
        if (mounted) {
          WidgetsBinding.instance.addPostFrameCallback((_) {
            setState(() {
              _code = num.toString();
              developer.log('Code: not a 4 digit code');
            });
          });
        }
      }
    }
  }

  @override
  Widget build(BuildContext context) {
    return Builder(builder: (context) {
      _generateCode2();
      return Scaffold(
        appBar: AppBar(
          title: Text('Device ID'),
        ),
        body: Center(
          child: Column(
            mainAxisAlignment: MainAxisAlignment.center,
            children: [
              Text('Code: ${_code ?? 'Nothing to show'}'),
            ],
          ),
        ),
      );
    });
  }
}

Here's the output it provides using version 2 method:

gif

Note: I can't embed a GIF file yet, so it is converted into a link instead (by default).

With the solution provided by Ken White (method version 1) and Frank van Puffelen (method version 2) produces the same outcomes.

So a big thanks to them!

I hope my answer would help the other future readers as well!

I made a package to take spectral derivatives using either the Chebyshev basis or the Fourier basis: spectral-derivatives. The code is open sourced with sphinx docs here, and I've put together a deep explanation of the math, so you can really know what's going on and why everything works.

I also encountered a problem very similar to yours, but my program does not use TerminateThread, and my stack information is

ntdll!RtlpWakeByAddress+0x7b
ntdll!RtlpUnWaitCriticalSection+0x2d
ntdll!RtlLeaveCriticalSection+0x60
ntdll!LdrpReleaseLoaderLock+0x15
ntdll!LdrpDecrementModuleLoadCountEx+0x61
ntdll!LdrUnloadDll+0x85
KERNELBASE!FreeLibrary+0x16
combase!FreeLibraryWithLogging+0x1f
combase!CClassCache::CDllPathEntry::CFinishObject::Finish+0x33
combase!CClassCache::CFinishComposite::Finish+0x51
combase!CClassCache::FreeUnused+0x9f
combase!CCFreeUnused+0x20
combase!CoFreeUnusedLibrariesEx+0x37
combase!CoFreeUnusedLibraries+0x9
mfc80u!AfxOleTermOrFreeLib+0x44
mfc80u!AfxUnlockTempMaps+0x4b
mfc80u!CWinThread::OnIdle+0x116
mfc80u!CWinApp::OnIdle+0x56
mfc80u!CWinThread::Run+0x3f
mfc80u!AfxWinMain+0x69
zenou!__tmainCRTStartup+0x150
kernel32!BaseThreadInitThunk+0x24
ntdll!__RtlUserThreadStart+0x2f
ntdll!_RtlUserThreadStart+0x1b

Did you finally solve it? Feeling the reply

The numpy.zeros shape parameter has type int or tuple of ints so it shoud be

image = np.zeros((480, 640, 3), dtype=np.uint8)

Since you have CUDA installed you could do the same with

gpu_image = cv2.cuda_GpuMat(480, 640, cv2.CV_8UC3, (0, 0, 0))
image = gpu_image.download()

Did you try to encode your url? For example using cyberchef: https://cyberchef.org/#recipe=URL_Encode(true)

You can try to test with Encode all special chars flag enabled and disabled.

I was moving resources from a DigitalOcean cluster to a Microk8s cluster, I decided to check the logs of the velero node agent and I found this error; An error occurred: unexpected directory structure for host-pods volume, ensure that the host-pods volume corresponds to the pods subdirectory of the kubelet root directory then I remembered I added this line of code after my installation;

/snap/bin/microk8s kubectl --kubeconfig="$kube_config_path" -n velero patch daemonset.apps/node-agent --type='json' -p='[{"op": "replace", "path": "/spec/template/spec/volumes/0/hostPath/path", "value":"/var/snap/microk8s/common/var/lib/kubelet/pods"}]'

This was interfering with the original hostpath value var/lib/kubelet/pods needed for normal Kubernetes cluster. I added a condition as to when the patch is to be done and it started working properly.

You can just ask whether it's in there:

["Hello", ["Hi"]] in mylist

Attempt This Online!

I found what is the different between header and result and try to fix it, I don't know this is a correct way to verify the result or not but it seems working

parseSignedRequest(signedRequest: string): {
  user_id: string;
  algorithm: 'HMAC-SHA256';
  issued_at: number;
} {
  const [encodedSig, payload] = signedRequest.split('.', 2);

  // decode the data
  const data = JSON.parse(this.base64UrlDecode(payload));

  const secret = "app-secret";

  // confirm the signature
  const expectedSig = crypto
    .createHmac('sha256', secret)
    .update(payload)
    .digest('base64')
    .replace('=', '')
    .replace(/\+/g, '-')
    .replace(/\//g, '_');

  if (expectedSig !== encodedSig) {
    throw new BadRequestException('Bad Signed JSON signature!');
  }

  return data;
}

private base64UrlDecode(input: string): string {
  const base64 = input.replace(/-/g, '+').replace(/_/g, '/');

  return Buffer.from(base64, 'base64').toString('utf-8');
}

I will be happy to get any suggestions

conda install pygraphviz

works for me.

Use a text Editor

If you just need to view the contents of the .csv file, you can use a text editor app.

Open the text editor app and navigate to the .csv file on your SD card to open it.

so actualy problem was in proc_addr() function, id dosen`t work in this way

As others mentioned, it will be tough to understand the problem without heap dump. You can verify below and if any of them is true that may be one of the causes to the issue.

1. Is there any objects being initialised during application startup. If yes check the size of the objects being loaded and avoid loding high payload into memory.
2. Are you loading any objects in memory and adding objects to the same objects which kept in memory.
3. There is possibility that any third party library could have stored high payload in memory. If that's the case then you might not be using it correctly.

However, I would still recommend to capture the heap dump and check the area which is consuming high memory and handle.

If it is feasible to connect to visualvm connect to it and there you can analyse or capture heap dump and analyse it using Eclipse-MAT app.

May be this link helps in capturing heap dump: How to take heap dump?

Hope this helps to understand the problem.

The dead links at aa-asterisk.org can be found again at web.archive.org

Youre changing system variable but in lesson you should do it with environment variable I think. Sorry for my english its not my native language by the way. I have same problem but I can`t change path environment variables, but can in system.

The name "gmon" likely refers to the GNU Monitoring tool.

For example, a page at the U. South Carolina site describing its use is titled:

GPROF Monitoring Program Performance

And says it will:

... demonstrate the use of the GPROF performance monitoring tool.

could you solve it?. getting the same error

In the latest version of scipy, shape is changed from a function to an attribute of csr_matrix.

https://docs.scipy.org/doc/scipy-1.15.0/reference/generated/scipy.sparse.csr_matrix.shape.html

Pls describe your error properly. Your error was not properly stated here. You only showed what you see when you describe the backup and not seeing any error in the description.

If you are using Symfony and EasyAdmin within the FileUploadType, you have to add the file upload javascript from easyadmin manually.

TextField::new('upload')
   ->setFormType(FileUploadType::class)
   ->addJsFiles(Asset::fromEasyAdminAssetPackage('field-file-upload.js'))

Maybe you can try https://www.vpadmin.org/, it provides the amdin interface to manage the SSG site

Product uptime: This metric measures the time that your software is working over a given period of time.

Bug response time: This is how quickly your team takes to identify a bug, find a patch solution, and push the fix into production. Issues can range from quick, five-minute fixes to full-fledged projects.

Daily active users: This is the number of users that use your software daily. This can help you understand how many of your customers actually use and value your software. If there is a large gap between the number of customers and the number of daily active users, then your customers may not be finding value in your product.

my friend I'm pretty sure signInWithRedirect is gonna stop working soon (or already stopped), but I'm not 100% sure. I think I saw that in google auth interface.

If that come to be true, what about you use popUp?

const provider = new GoogleAuthProvider();
 try {
   
    const result = await signInWithPopup(auth, provider);
    const user = result.user;

    if(!user.uid){
      console.log("Erro");
      setExibirErro("Erro login google.");
      return
    }

    const { displayName, email, uid, emailVerified, photoURL } = user; 
    const providerId = result.providerId;
    const { firstName, lastName } = splitName(displayName || '');

}

yabai -m query --spaces --space | jq '.index'

If we want two add two matrices we need only two square matrices so,we can give columns and rows of A or B because both are with same rows and columns

I was trying to do the same as you. I found this tutorial How to Delete All WooCommerce Orders

It basically says that the hook 'manage_posts_extra_tablenav' is 'woocommerce_order_list_table_extra_tablenav' for HPOS orders.

If you can figure out how to show the added custom button on mobile (it hides) let me know!

my issue was resolve by fixing a typing mistake in the Redis container variable where I had a ";" instead of the ":"

The variable is referred to in the Space Invader immich tutorial pt2.

Update is called once per frame so you shouldn't be setting the player to alive there.

To know for sure that a player is dead, you'll need to access your game manager through your player, and upon death, call a public game manager method that will set the player alive to false. This is a more efficient way than to have the game manager check in Update() whether the player is alive.

You can also just keep score on the player object itself, and just allow the game manager / other objects to get it and update it. Then when the game is over, and you destroy the player object, the score will reset to 0 when you create a new game.

So, installing Team Explorer is the correct answer as there are dependencies required by the add-in that are not normally available with a base install of Windows. Team Explorer in lieu of the full flagship Visual Studio, which most stakeholders do not need. Team Explorer requires only "Stakeholder" permissions normally to access at a Azure DevOps limited, basic access level... but a CAL really depends on how you use the queries and underlying data. Normally this is not an issue because such users already have the sufficient access granted via Azure DevOps anyways.

just add .html to the files example instead http://docloo.s3-website.eu-west-2.amazonaws.com/docs/getting-started/introduction which gives error because in s3 you don't have file called introduction you have introduction.html so http://docloo.s3-website.eu-west-2.amazonaws.com/docs/getting-started/introduction.html perfectly works but at the end of the end you have to implement way you add .html automatically you don't expect your users to add that

New-AzResourceGroupDeployment -Name ExampleDeployment -ResourceGroupName ExampleResourceGroup -TemplateFile C:\MyTemplates\storage.bicep -TemplateParameterFile C:\MyTemplates\storage.bicepparam -storageAccountType Standard_LRS

Link

For those having issues... This solves some situations. Don't downvote just because it doesn't solve this specific OP--there are dozens of side cases.

So... many users are seeing an integration issue because the Azure DevOps Office Integration in of itself is not enough. Most folks using it have Visual Studio. However, those that don't may pull the Visual Studio Team Explorer 2022 for instance. Depending upon how you use it, you may need a CAL, but normally for "Stakeholder" access (users who require limited, basic access) may not need a CAL.

This installation will provide the other "bits" need to enable the Office addin to work for Excel and Project. This worked when all of the other steps seen throughout stackoverflow didn't.

Based on this thread:

NewFilePath = PathNoExtension & "pdf"
If Dir(NewFilePath) <> "" Then
    Dim res As Long
    res = GetAttr(NewFilePath)
    If res = 33 Then
        MsgBox "File is read only"
        Exit Sub
    End If
End If

Not clear, share screenshot please.

Prefabs can have scripts same as regular objects. You can add a new script to the prefab by dragging and dropping it on the prefab in the unity editor, or selecting the prefab and using "add component" in the inspector window then adding a "new script".

to move the instantiated objects in a random direction, just add the translate into the Update() method of this new script. In a collider on trigger event (or similar) check if self is normal and the other is a virus, and if so, then instantiate a new virus, set its transform to that of the current object, save any other properties you want to save into the new virus, then destroy this.gameobject.

still happening. as of today, i even subscribed to Google Colab Pro. No luck

No way to do that on FEN.
FEN is intended to represent a position plus validation status [turn, castling options, en passe option]

This was happening to me frequently while testing a free/$300/trial, no matter which model I used, sometimes after only 1 or 2 small queries. I modified my (python) code to just retry after waiting 1 second, and now while I can see in my logs that it sometimes does 1-3 retries all the queries succeed.

if anyone has this problem, just add the line id "dev.flutter.flutter-gradle-plugin" to plugin section in android/app/build.gradle

After you install the prompt package go to package.json and add "type": "module" right before the keywords, like this:

"type": "module",

"keywords": [],

"author": "",

"license": "ISC",

"description": "",

"dependencies": { "prompt-sync": "^4.2.0"

If you are downloading the file via a web browser, it is possible that the browser is auto-decompressing the file because browsers know how to handle web pages that are gzip-compressed.

To fully test what is happening, you should download the file via the AWS CLI and then check the file contents.

You could also compare the size of the file shown in S3 vs the size on your local disk.

See: Is GZIP Automatically Decompressed by Browser?

I think you're using a pretty old version of hibernate validator.

Do you need to use org.hibernate.validator.constraints.Email (this is deprecated in later releases for jakarta.validation.constraints.Email)

Going against training I received, Whether "View Native Query" is enabled is not a reliable indicator for query mode.

In my case, I have verified by means of a trace on the database server that the method described in the "Another way" section of my question actually does leave the report (table?) in Direct Query mode. Oddly, "View Native Query" was not enabled, but there was no big, yellow warning about being forced into Import mode like on my other attempts.

Hopefully the warning message IS a good indicator. I may investigate if it also lies.

Unfortunately, the sum total of this question and follow-up means that for anything interesting I'll need to write SQL rather than using Power Query for transformations if I want to use Direct Query.

@rehaqds made the comment that answered my question:

sharey=True works for me

Occam's razor, I was so wrapped up in my attempts I confused myself.

I wanted to post my current working solution in case it helps anyone else, or so someone can tear it apart if I'm missing something. I was able to resolve with Spring MVC by defining an external resource handler and redirecting users to this:

@Configuration
public class CustomWebMvcConfigurer implements WebMvcConfigurer {
    @Value("${external.resource.location:file:///external/foo/}")
    private String externalLocation;

    @Override
    public void addResourceHandlers(ResourceHandlerRegistry registry) {
        registry.addResourceHandler("/foo/**")
                .addResourceLocations(externalResourcesLocation);
    }
}

After adding this configuration and placing my external file content in /external/foo/[index.html|js/|css], I'm now able to access my externalized content when navigating to localhost:<port>/foo/index.html without requiring it to be packaged or deployed within my .war file.

This looks like a Pylance bug. Your code is fine, and mypy accepts it without complaint.

In my case I needed a combination of some answers here.

First, having config.assets.debug = true in the config/environments/development.rb file.

Then, I needed plugin :tailwindcss if ENV.fetch("RAILS_ENV", "development") == "development" on the puma.rb file. This way it runs the application and runs Tailwind running on "watch" mode.

I have ruby 3.3.4, rails 7.1.5 and node 20.18.0 versions.

See https://stackoverflow.com/a/78393579/12955733 and https://stackoverflow.com/a/78250287/12955733.

Laravel Guards: A Mechanism for Multi-Level Access Control

Guards in Laravel provide a powerful and flexible mechanism to implement multi-level access control within your application. They allow you to define how users are authenticated, handle different user roles, and ensure secure access to specific parts of your application based on the user type or context.

This is particularly useful in scenarios where your application has multiple levels of users, such as administrators, editors, and regular users, or in cases of multi-tenancy, where each tenant has its own authentication logic.

How Guards Work in Laravel Guards define how authentication is managed for a given request. By default, Laravel provides two main guards:

Web Guard: Uses session and cookies for stateful authentication, ideal for web applications. API Guard: Utilizes tokens for stateless authentication, often used in API-based applications.

You will need to let shortcuts run scripts and allow assistive access to control your computer. but heres the code

tell application "System Events"
keystroke "Your text here"
end tell

you can also add a second line of keystroke.

To follow up about this, as it turns out my issue was that I had an old broken version of MacPorts installed on my machine. Removing MacPorts fixed the problem for me.

See this thread for more info: https://github.com/pyenv/pyenv/issues/3158

Selecting debug on the app.run allows for auto reloading. So the whole codebase runs again, and it didnt close the cam, this is supposed to be for code edits but I found it just did it all the time.

In theory moving the init to the main bit should stop it, it doesnt.

I have an init function that checks if the picam var is nothing, if so it sets it, otherwise does nothing, as its been set already. This works.

As does setting the debug to false.

What I would do personally (if you are on windows) is use Visual Studio instead of GCC with a make file since Visual Studio lets you compile C and C++ files together or port all C++ code to C. But the most reliable way you could potentially get this to work is to use a regular C++ class.

For me, it was Nginx. The problem was Nginx trying to connect to port 80 while the port was occupied by another process, so I freed up the port 80:

netsh http add iplisten ipaddress=::

Resources: Nginx- error: bind() to 0.0.0.0:80 failed. permission denied

Its probably better to convert the image into an sf symbol. Although i think image sizes inside a tabItem is limited.

If you have the svg of your image, you can use this to convert into a valid SF symbol: https://github.com/swhitty/SwiftDraw

Then import in your assets and call as named

I realized I was using the wrong endpoint to obtain a token for the Graph API. The endpoint I was using is intended for acquiring tokens for Delegated or Custom APIs that require a token to respond.

Below is the correct endpoint to obtain a token for the Graph API:
https://login.microsoftonline.com/{tenant-id}/oauth2/v2.0/token

You can use https://appreviews.dev/ The most I could extract was ~20k reviews for X app (ex Twitter) It has a limit though, to 500 reviews per country

Another answer, since you're using Sail, is to run sail artisan storage:link instead of php artisan storage:link. That worked for me, 404 stopped and the files showed fine.

This behaviour seems to occur because of two different integer stringification methods within the interpreter. The slowdown might be fixed in a future perl release: https://github.com/Perl/perl5/pull/22927

why is there a compatibility issue?

Loki4j v1.6.0 requires Logback v1.4.x. If I remember correctly, ILoggingEvent.getNanoseconds() was introduced back in Logback v1.3.0. So your project is probably using Logback v1.2.3. That's why you have compatibility issue.

You can either force Maven to use Logback v1.4.x, or downgrade Loki4j to version that supports Logback v1.2.x (see compatibility matrix).

Also please note that recent versions of Loki4j support JSON layout natively, so you don't have to specify message pattern like this.

Adding to the discussion, here are those tests in the linked page reproduced with some minor changes to see if anything has changed since that original post was made almost 8 years ago and python and many of its libraries have upgraded quite a bit since then. According to python.org the newest version of python available at the time of his post was 3.6 .

Here is the source code, copied from the linked page and updated to be runnable as posted here, plus a few minor changes for convenience.

import pandas
import matplotlib.pyplot as plt
import seaborn
import numpy

import sys
import time

NUMBER_OF_ITERATIONS = 10
FIGURE_NUMBER = 0

def bench_sub(mode1_inputs: list, mode1_statement: str, mode2_inputs: list, mode2_statement: str) -> tuple[bool, list[float], list[float]]:

    mode1_results = []
    mode1_times = []
    mode2_results = []
    mode2_times = []

    for inputs, statementi, results, times in (
            (mode1_inputs, mode1_statement, mode1_results, mode1_times),
            (mode2_inputs, mode2_statement, mode2_results, mode2_times)
    ):
        for inputi in inputs:
            ast = compile(statementi, '<string>', 'exec')
            ast_locals = {'data': inputi}
            start_time = time.perf_counter_ns()
            for _ in range(NUMBER_OF_ITERATIONS):
                exec(ast, locals=ast_locals)
            end_time = time.perf_counter_ns()

            results.append(ast_locals['res'])
            times.append((end_time - start_time) / 10 ** 9 / NUMBER_OF_ITERATIONS)

    passing = True
    for results1, results2 in zip(mode1_results, mode2_results):
        if not passing:
            break
        try:
            if type(results1) in [pandas.Series, numpy.ndarray] and type(results2) in [pandas.Series, numpy.ndarray]:
                if type(results1[0]) is str:
                    isclose = set(results1) == set(results2)
                else:
                    isclose = numpy.isclose(results1, results2).all()
            else:
                isclose = numpy.isclose(results1, results2)
            if not isclose:
                passing = False
                break
        except (ValueError, TypeError):
            print(type(results1))
            print(results1)
            print(type(results2))
            print(results2)
            raise
    return passing, mode1_times, mode2_times

def bench_sub_plot(mode1_inputs: list, mode1_statement: str, mode2_inputs: list, mode2_statement: str, title: str, label1: str, label2: str, save_fig: bool = True) -> tuple[bool, list[float], list[float]]:
    passing, mode1_times, mode2_times = bench_sub(mode1_inputs, mode1_statement, mode2_inputs, mode2_statement)

    fig, ax = plt.subplots(2, dpi=100, figsize=(8, 6))
    mode1_x = [len(x) for x in mode1_inputs]
    mode2_x = [len(x) for x in mode2_inputs]

    ax[0].plot(mode1_x, mode1_times, marker='o', markerfacecolor='none', label=label1)
    ax[0].plot(mode2_x, mode2_times, marker='^', markerfacecolor='none', label=label2)
    ax[0].set_xscale('log')
    ax[0].set_yscale('log')
    ax[0].legend()
    ax[0].set_title(title + f' : {"PASS" if passing else "FAIL"}')
    ax[0].set_xlabel('Number of records')
    ax[0].set_ylabel('Time [s]')

    if mode1_x == mode2_x:
        mode_comp = [x / y for x, y in zip(mode1_times, mode2_times)]
        ax[1].plot(mode1_x, mode_comp, marker='o', markerfacecolor='none', label=f'{label1} / {label2}')
        ax[1].plot([min(mode1_x), max(mode1_x)], [1.0, 1.0], linestyle='dashed', color='#AAAAAA', label='parity')
        ax[1].set_xscale('log')
        ax[1].legend()
        ax[1].set_title(title + f' (ratio)\nValues <1 indicate {label1} is faster than {label2}')
        ax[1].set_xlabel('Number of records')
        ax[1].set_ylabel(f'{label1} / {label2}')
    plt.tight_layout()
    # plt.show()

    if save_fig:
        global FIGURE_NUMBER
        # https://stackoverflow.com/a/295152
        clean_title = ''.join([x for x in title if (x.isalnum() or x in '_-. ')])
        fig.savefig(f'outputs/{FIGURE_NUMBER:06}_{clean_title}.png')
        FIGURE_NUMBER += 1

    return passing, mode1_times, mode2_times

def _print_result_comparison(success: bool, times1: list[float], times2: list[float], input_lengths: list[int], title: str, label1: str, label2: str):
    print(title)
    print(f'  Test result: {"PASS" if success else "FAIL"}')
    field_width = 15
    print(f'{"# of records":>{field_width}} {label1 + " [ms]":>{field_width}} {label2 + " [ms]":>{field_width}} {"ratio":>{field_width}}')
    for input_length, time1, time2 in zip(input_lengths, times1, times2):
        print(f'{input_length:>{field_width}} {time1 * 1000:>{field_width}.03f} {time2 * 1000:>{field_width}.03f} {time1 / time2:>{field_width}.03f}')
    print()

def bench_sub_plot_print(mode1_inputs: list, mode1_statement: str, mode2_inputs: list, mode2_statement: str, title: str, label1: str, label2: str, all_lengths: list[int], save_fig: bool = True) -> tuple[bool, list[float], list[float]]:
    success, times1, times2 = bench_sub_plot(
        mode1_inputs,
        mode1_statement,
        mode2_inputs,
        mode2_statement,
        title,
        label1,
        label2,
        True
    )
    _print_result_comparison(success, times1, times2, all_lengths, title, label1, label2)
    return success, times1, times2


def _main():

    start_time = time.perf_counter_ns()

    # In [2]:
    iris = seaborn.load_dataset('iris')


    # In [3]:
    data_pandas: list[pandas.DataFrame] = []
    data_numpy: list[numpy.rec.recarray] = []
    all_lengths = [10_000, 100_000, 500_000, 1_000_000, 5_000_000, 10_000_000, 15_000_000]
    # all_lengths = [10_000, 100_000, 500_000] #, 1_000_000, 5_000_000, 10_000_000, 15_000_000]
    for total_len in all_lengths:
        data_pandas_i = pandas.concat([iris] * (total_len // len(iris)))
        data_pandas_i = pandas.concat([data_pandas_i, iris[:total_len - len(data_pandas_i)]])
        data_pandas.append(data_pandas_i)
        data_numpy.append(data_pandas_i.to_records())

    # In [4]:
    print('Input sizes [count]:')
    print(f'{"#":>4} {"pandas":>9} {"numpy":>9}')
    for i, (data_pandas_i, data_numpy_i) in enumerate(zip(data_pandas, data_numpy)):
        print(f'{i:>4} {len(data_pandas_i):>9} {len(data_numpy_i):>9}')
    print()

    # In [5]:
    mb_size_in_bytes = 1024 * 1024
    print('Data sizes [MB]:')
    print(f'{"#":>4} {"pandas":>9} {"numpy":>9}')
    for i, (data_pandas_i, data_numpy_i) in enumerate(zip(data_pandas, data_numpy)):
        print(f'{i:>4} {int(sys.getsizeof(data_pandas_i) / mb_size_in_bytes):>9} {int(sys.getsizeof(data_numpy_i) / mb_size_in_bytes):>9}')
    print()

    # In [6]:
    print(data_pandas[0].head())
    print()

    # In [7]:
    # ...

    # In [8]:
    success, times_pandas, times_numpy = bench_sub_plot_print(
        data_pandas,
        'res = data.loc[:, "sepal_length"].mean()',
        data_numpy,
        'res = numpy.mean(data.sepal_length)',
        'Mean on Unfiltered Column',
        'pandas',
        'numpy',
        all_lengths,
        True
    )

    # In [9]:
    success, times_pandas, times_numpy = bench_sub_plot_print(
        data_pandas,
        'res = numpy.log(data.loc[:, "sepal_length"])',
        data_numpy,
        'res = numpy.log(data.sepal_length)',
        'Vectorised log on Unfiltered Column',
        'pandas',
        'numpy',
        all_lengths,
        True
    )

    # In [10]:
    success, times_pandas, times_numpy = bench_sub_plot_print(
        data_pandas,
        'res = data.loc[:, "species"].unique()',
        data_numpy,
        'res = numpy.unique(data.species)',
        'Unique on Unfiltered String Column',
        'pandas',
        'numpy',
        all_lengths,
        True
    )

    # In [11]:
    success, times_pandas, times_numpy = bench_sub_plot_print(
        data_pandas,
        'res = data.loc[(data.sepal_width > 3) & (data.petal_length < 1.5), "sepal_length"].mean()',
        data_numpy,
        'res = numpy.mean(data[(data.sepal_width > 3) & (data.petal_length < 1.5)].sepal_length)',
        'Mean on Filtered Column',
        'pandas',
        'numpy',
        all_lengths,
        True
    )

    # In [12]:
    success, times_pandas, times_numpy = bench_sub_plot_print(
        data_pandas,
        'res = numpy.log(data.loc[(data.sepal_width > 3) & (data.petal_length < 1.5), "sepal_length"])',
        data_numpy,
        'res = numpy.log(data[(data.sepal_width > 3) & (data.petal_length < 1.5)].sepal_length)',
        'Vectorised log on Filtered Column',
        'pandas',
        'numpy',
        all_lengths,
        True
    )

    # In [13]:
    success, times_pandas, times_numpy = bench_sub_plot_print(
        data_pandas,
        'res = data[data.species == "setosa"].sepal_length.mean()',
        data_numpy,
        'res = numpy.mean(data[data.species == "setosa"].sepal_length)',
        'Mean on (String) Filtered Column',
        'pandas',
        'numpy',
        all_lengths,
        True
    )

    # In [14]:
    success, times_pandas, times_numpy = bench_sub_plot_print(
        data_pandas,
        'res = data.petal_length * data.sepal_length + data.petal_width * data.sepal_width',
        data_numpy,
        'res = data.petal_length * data.sepal_length + data.petal_width * data.sepal_width',
        'Vectorized Math on Unfiltered Column',
        'pandas',
        'numpy',
        all_lengths,
        True
    )

    # In [16]:
    success, times_pandas, times_numpy = bench_sub_plot_print(
        data_pandas,
        'res = data.loc[data.sepal_width * data.petal_length > data.sepal_length, "sepal_length"].mean()',
        data_numpy,
        'res = numpy.mean(data[data.sepal_width * data.petal_length > data.sepal_length].sepal_length)',
        'Vectorized Math in Filtering Column',
        'pandas',
        'numpy',
        all_lengths,
        True
    )

    end_time = time.perf_counter_ns()
    print(f'Total run time: {(end_time - start_time) / 10 ** 9:.3f} s')

if __name__ == '__main__':
    _main()

Here is the console output it generates:

Input sizes [count]:
   #    pandas     numpy
   0     10000     10000
   1    100000    100000
   2    500000    500000
   3   1000000   1000000
   4   5000000   5000000
   5  10000000  10000000
   6  15000000  15000000

Data sizes [MB]:
   #    pandas     numpy
   0         0         0
   1         9         4
   2        46        22
   3        92        45
   4       464       228
   5       928       457
   6      1392       686

   sepal_length  sepal_width  petal_length  petal_width species
0           5.1          3.5           1.4          0.2  setosa
1           4.9          3.0           1.4          0.2  setosa
2           4.7          3.2           1.3          0.2  setosa
3           4.6          3.1           1.5          0.2  setosa
4           5.0          3.6           1.4          0.2  setosa

Mean on Unfiltered Column
  Test result: PASS
   # of records     pandas [ms]      numpy [ms]           ratio
          10000           0.061           0.033           1.855
         100000           0.160           0.148           1.081
         500000           0.653           1.074           0.608
        1000000           1.512           2.440           0.620
        5000000          11.633          12.558           0.926
       10000000          23.954          25.360           0.945
       15000000          35.362          40.108           0.882

Vectorised log on Unfiltered Column
  Test result: PASS
   # of records     pandas [ms]      numpy [ms]           ratio
          10000           0.124           0.056           2.190
         100000           0.507           0.493           1.029
         500000           3.399           3.441           0.988
        1000000           5.396           6.867           0.786
        5000000          27.187          38.121           0.713
       10000000          55.497          72.609           0.764
       15000000          88.406         112.199           0.788

Unique on Unfiltered String Column
  Test result: PASS
   # of records     pandas [ms]      numpy [ms]           ratio
          10000           0.332           1.742           0.191
         100000           2.885          21.833           0.132
         500000          14.769         125.961           0.117
        1000000          29.687         264.521           0.112
        5000000         147.359        1501.378           0.098
       10000000         295.118        3132.478           0.094
       15000000         444.365        4882.316           0.091

Mean on Filtered Column
  Test result: PASS
   # of records     pandas [ms]      numpy [ms]           ratio
          10000           0.355           0.130           2.719
         100000           0.522           0.672           0.777
         500000           1.797           4.824           0.372
        1000000           4.602          10.827           0.425
        5000000          22.116          57.945           0.382
       10000000          43.076         116.028           0.371
       15000000          68.893         177.658           0.388

Vectorised log on Filtered Column
  Test result: PASS
   # of records     pandas [ms]      numpy [ms]           ratio
          10000           0.361           0.128           2.821
         100000           0.576           0.758           0.760
         500000           2.066           5.199           0.397
        1000000           5.259          11.523           0.456
        5000000          22.785          59.581           0.382
       10000000          47.527         121.882           0.390
       15000000          75.080         187.954           0.399

Mean on (String) Filtered Column
  Test result: PASS
   # of records     pandas [ms]      numpy [ms]           ratio
          10000           0.636           0.192           3.304
         100000           4.068           1.743           2.334
         500000          20.954           9.306           2.252
        1000000          41.938          18.522           2.264
        5000000         217.254          97.929           2.218
       10000000         434.242         197.289           2.201
       15000000         657.205         297.919           2.206

Vectorized Math on Unfiltered Column
  Test result: PASS
   # of records     pandas [ms]      numpy [ms]           ratio
          10000           0.168           0.049           3.415
         100000           0.385           0.338           1.140
         500000           3.193           5.018           0.636
        1000000           6.028           9.539           0.632
        5000000          32.640          48.235           0.677
       10000000          69.748          99.893           0.698
       15000000         107.528         159.040           0.676

Vectorized Math in Filtering Column
  Test result: PASS
   # of records     pandas [ms]      numpy [ms]           ratio
          10000           0.350           0.234           1.500
         100000           0.926           2.494           0.371
         500000           6.093          15.007           0.406
        1000000          12.641          30.021           0.421
        5000000          71.714         163.060           0.440
       10000000         145.373         326.206           0.446
       15000000         227.817         490.991           0.464

Total run time: 183.198 s

And here are the plots it generated:

These results were generated with Windows 10, Python 3.13, on i9-10900K, and never got close to running out of memory so swap should not be a factor.

In my case my Docker Hub personal access token was read only. Changed to read/write and it worked.

Next.js 15 introduced changes to ESLint configuration, which can cause issues with VS Code integration. To simplify the setup and ensure ESLint and Prettier work correctly, I created a CLI that automates everything.

🔗 NPM: https://www.npmjs.com/package/eslint-prettier-next-15 💻 GitHub: https://github.com/danielalves96/eslint-prettier-next-15

You can install and run it, and it will configure everything properly. Plus, you can customize the setup as needed. Hope this helps!

I wondering what if the endpoint only accept form-data I cannot parse to JSON.

I've seen an issue in NextJS when I have enabled telemetry. For some reason formData has not been sent.

If someone knows more about it please share your comments.

The only way that I made it work is adding.

But I prefer to do not ignore it.

opentelemetry: { ignore: true, },

Been a while since I asked this but while I'm here, I'll just drop the solution I discovered here in case any one else ends up stuck.

After upgrading Gradle used with my old project, I no longer had the issue any more so I'm guessing it was just incompatible with the latest Android SDK.

Byte-level BPE (BBPE) utilizes UTF-8 to encode every characters into 1 to 4 bytes. To ensure base vocab size is 256 (which is 1 byte), BBPE only use 1 byte per token. So in case a character requires 2 or more bytes to represent, BBPE breaks down those bytes into individual tokens (which means 1 character is transformed into 2, 3 or 4 different tokens).

For example, the UTF-8 code of character "の" is E3 81 AE (3 bytes), so in BBPE, "の" is written as 3 different tokens: E3, 81, and AE.

(Note that these 3 tokens are individual to each other, and may not pair up again in BPE merging step)

BBPE tokenizer may cause the tokenized text to be up to 4x longer than that in BPE tokenizer (when every characters are 4 bytes in UTF-8), but it's a trade-off to keep the vocab size to as low as 256.

The above example is taken from Figure 1 of the original paper of Byte-level Text Representation.

In this topic of Dockers by the Github community, it is a must to have file location of your pid file, router.db, logs, crash folder etc. into your home directory, you may change the file location of your pid file by kubectl plugins to avoid denied permission.

I think this is the best option for react.js

Building an Infinite Scroll FlatList Component in ReactJS: A Comprehensive Guide

https://medium.com/@sazzadur/building-an-infinite-scroll-flatlist-component-in-reactjs-a-comprehensive-guide-3f25c0ec203b

This issue related to a missing environment variable. Kindly review the configuration map and ensure that the environment variable is properly set within it. example : username=XYZ. Once this is completed, you may verify.

Try getting rid of the .local in your query:

Resolve-DnsName test1

Needed this myself and couldn't find an answer. This is working for me.

[POST] https://dev.azure.com/{organisation}/_apis/Contribution/HierarchyQuery?api-version=5.0-preview.1

body: { "contributionIds": ["ms.vss-test-web.testcase-results-data-provider"], "dataProviderContext": { "properties": { "testCaseId":{test_case_id}, "sourcePage": { "routeValues": { "project":"{project_name}" } } } } }

These are the commands that I've been using. Not sure if I'm doing this right though.

apt install -y acpica-tools
acpidump > acpidump.out
acpixtract -a acpidump.out
iasl -d dsdt.dat
OEM_ID=$(grep "OEM ID" dsdt.dsl | tail -n 1 | awk -F'"' '{print $2}' | cut -c1-4)
sed -i "s/'<WOOT>'/'$OEM_ID'/g" /opt/CAPEv2/installer/kvm-qemu.sh

This is a old question but it keeps coming up in searches. For a more up-to-date wireshark BLE 5.x sniffer, look at https://github.com/nccgroup/Sniffle It needs firmware reflashed onto various TI chipset developer boards but captures can be done directly from wireshark. Many of the Nordic and older sniffers were never updated beyond BLE 4.x TI also has a sniffer of their own for some boards, but it wasn't updated for BLE 5.x

I get the same error. Were you ever able to resolve the issue? I am running the latest version of Ubuntu.

queryPurchasesAsync() will only return non-consumed one-time purchases & active subscriptions. Per this doc queryPurchaseHistory() is deprecated in Billing v7, so it seems the only way to do it in-app is to track the history yourself.

The loss should explicitely require a gradient to be evaluated, so I would remove the loss.requires_grad = True line. Also, try to rewrite the first line

def get_pedalboard(self):
    highpass_freq = self.highpass_freq.clamp(20, 500) 

Are you sure the

board = Pedalboard([
            HighpassFilter(cutoff_frequency_hz=float(highpass_freq)),
        ])

works with PyTorch autograd?

I was thinking of something similar, it would be cool to have one of these! It looks like someone else is attempting one and has their frontend set up but maybe not the backend: https://mixtape.dj/

At any rate, before hitting the API, I would probably start by breaking the audio master into phrases. As you likely know almost all mixable tracks are in 4/4 time. Furthermore, most DJs will mix in phrasing blocks, so this would serve as a good segmentation point as per (1). Looks like these guys have built some functionality for doing this: https://github.com/mixxxdj/mixxx/wiki/ Might have to just mine the srcs looks like it's mostly compiled stuff.

The main issue I can see Shazam running into is deciphering songs during long blends. If you can figure out some way to identify which phrases are blends and remove them from the search area you'll be laughing. Fast cuts should be easy to do with FFT or even echoprint (https://github.com/spotify/echoprint-codegen) just look for a sudden change in the spectrum/print.

Once you've done that, you should have boundary points for the start of each new track. Then you can just feed tracks 1 x 1 into Shazam API.

Ultimately, I think the clincher is ID'ing those long blends. Maybe ML is an approach to that? Should be easy to train on by building a script to literally play 2 random songs at once, over and over in a bajillion diff permutations.

Best of luck! I will be curious as to how this works out :)

https://youtu.be/EPtY-mLpdfM?si=6W48WciTPws7bnXa

Same conding help please.

My mail - [email protected]

If you have a way to convert the files to jt-format, you should be able to use the 3d viewer from the Mendix marketplace: https://marketplace.mendix.com/link/component/118345

I looks like you skip the first datapoint since you write x = data[1:len(data),0] and so forth. As @trincot mentioned, you also have to care about the y[i-1] case for i=0. Maybe the following will help you:

tst = []
x = data[:,0]
y = data[:,1]  
intt = data[:,2]
for i in range(1,len(data)):
   if intt[i]!=0:         
      tst.append((x[i]**2.0+ y[i]-y[i-1])**2.0)
   else:
      break

This includes all data points in x, y, and intt, but the first data point will still be skipped since the loop starts with i=1.

Improving great answer by @dmackerman a bit (I cannot comment yet) by preventing to delete if there is only one row

HTML:

<form>
    <p class="form_field">
        <label>Name:</label> <input type="text">
        <label>Age:</label> <input type="text">
        <span class="remove">Remove</span>
    </p>
    <p>
        <span class="add">Add fields</span>
    </p>
</form>

and JS:

$(".add").click(function() {
    $("form > p:first-child").clone(true).insertBefore("form > p:last-child");
    return false;
});

$(".remove").click(function() {
     if ($(".form_field").length > 1) {
        $(this).parent().remove();
    }
});

1. Make sure you have the ENABLE_TEMPLATE_PROCESSING feature flag enabled.

FEATURE_FLAGS = {
    "ENABLE_TEMPLATE_PROCESSING": True,
}

2. Make sure pass the urlParam.

urlParams: {
    foo: 'bar',
}

1. Make sure you use the macro in your dataset

select fld from tbl where fld= '{{ url_param('foo') }}'

Here's some useful documentation:

• https://preset.io/blog/intro-jinja-templating-apache-superset/
• https://superset.apache.org/docs/configuration/sql-templating/

Gosh, missed this entirely. It's just a different kind of option for the action. Hopefully it will help someone else if they come looking for something similar.

Installing Android Studio after years, After installing the Android Studio Ladybug, the SDK was missing. File -> Setting -> No SDK found.

Used the search option to find SDK and then was able to select a SDK version and continue with SDK Installation.

I have released a version using comtypes. @mach

https://github.com/tifoji/pyrtdc/

win32com also works using an approach similar to the wrapper suggested by @Aincrad

When you forward external port 81 to internal port 80, then internal Traefik entrypoint needs to be 80.

ports:
  - "81:80"
command:
  - --entrypoints.http.address=:81

This issue occurred for me when I used imports/exports like this:

export { test } from './test';  
import { test } from '@/lib';

Having many similar patterns caused the error. Updating the import to:

import { test } from '@/lib/test';  

resolved the problem in my case.