The main problem as far as I see it is that you have a misunderstanding of what the A*-algorithm actually does. As far as I see it, you have some kind of rectangular field with discrete points on it and you want to have the shortest path between two points. Why do you use A* at all? You can just draw a horizontal and a vertical line and that's it.

The main point of route planning algorithms is to find the shortest path in an environment where NOT all paths from source to target are actually valid. For example, you could insert obstacles onto your raster (=points that are inside your rectangle, but are invalid).

Now to your code: You are computing the g-values incorrectly. g is actually the length of the path you have already taken. You cannot just use your heuristic-function here. You really have to build the correct sum. Also, you are using closedList.contains(n) and n2.equals(n), but your Node-class doesn't override hashCode() and equals(). These are the two major flaws I've found immediately. (There may be more...)

Side note: Your code is very verbose. You could for example collapse all cases in your nextNodes()-method into one loop that has four iterations. I'm just suggesting that, because you may get more helpful answers in the future if your code is cleaner. Also please improve your formatting for better readability.

I optimized XOR maximization algorithm reduces complexity from O(log r) to O(log l) by iterating only over the smaller number L. It uses bitwise comparison to maximize the XOR value, flipping bits of l or r as needed while respecting the range [L,R] . The final XOR result is calculated using a fast power function, ensuring minimal operations and efficiency. This method ensures the maximum XOR value is computed while preserving the range constraints.

code with more detalied explain: https://github.com/MohamedHussienMansour/Maximum-XOR-problem-with-O-log-L-

I run in the same error message, and it was due the absence of the javac executable. Apparently, in my Linux distro (OpenSUSE), they have separated the JRE from the JDK in two different packages. I had java-21-openjdk installed, but not so java-21-openjdk-devel . Installing this last package solved the issue.

You're very close to getting Tailwind working. The issue you're seeing (Unknown at rule @tailwind) is just a warning from your editor, not an actual error. To fix everything, make sure you create a tailwind.config.js file by running

npx tailwindcss init

and update the content field to include your index.html (e.g. content: ["./index.html"]). Also, ensure you're linking the compiled output.css in your HTML file like this: . Lastly, install the Tailwind CSS IntelliSense extension in VS Code to get rid of those warnings and enjoy autocomplete support. Everything else you’ve done looks fine!

what how and can you tell me ow to do it

change

PERSIAN_EPOCH = 1948320.5;

to

PERSIAN_EPOCH = 1948321.5;

until the end of 1404

an then return it to 1948320.5

Move your _buildItem(context) from the bottom navigation to the body of the scaffold. Your tappable widgets are under the top stack widget, which is why your onTap of InkWell isn't working properly here. When you move the _buildItem to the body, it will be fixed.

This is because the message library was designed for Python2.

If you create a python=2.7 env, you can successfully install the package.

After trying many of the things mentioned above, what worked for me was updating the WSL extension in VScode.

If you want to switch the startup object in a more practical way consider using the Change Startup Object Visual Studio extension.

It gives you a command to change the project startup object directly in the context menu:

thank you, this save me, same problem solved...

I managed to fix the errors by deleting the following files in %localappdata%\vcpkg:

vcpkg.user.props
vcpkg.user.targets

It seems MSVS was allowing vcpkg to import custom properties with the following

  <Import Condition="Exists('C:\Scripts\vcpkg\scripts\buildsystems\msbuild\vcpkg.targets') and '$(VCPkgLocalAppDataDisabled)' == ''" Project="C:\Scripts\vcpkg\scripts\buildsystems\msbuild\vcpkg.targets" />

Use Cloudflare WARP, its work for me when I install pytorch.

Only after set options:

builder.setCharset(StandardCharsets.UTF_8);
builder.setMode(HttpMultipartMode.BROWSER_COMPATIBLE);

"name" sent correctly, and does not turn into "????? ?????".

*org.apache.httpcomponents httpclient version 4.5.14

Full code of using MultipartEntityBuilder with UTF-8:

var filePath = Path.of("<CyryllicFileName>");
var contentType = ContentType.TEXT_PLAIN.withCharset("UTF-8");
var builder = MultipartEntityBuilder.create();
builder.setCharset(StandardCharsets.UTF_8);
builder.setMode(HttpMultipartMode.BROWSER_COMPATIBLE);
var entity = builder
        .addTextBody("name", name, contentType)
        .addBinaryBody("file",
                       filePath.toFile(),
                       ContentType.APPLICATION_OCTET_STREAM,
                       filePath.toFile().getName())
        .addTextBody("collectionId", collectionId)
        .addTextBody("parentDocumentId", parentDocumentId)
        .addTextBody("publish", "true")
        .build();

You forget to add border style also will be good to use border shorthand to never forget it again:

--> border: width style color;*

example:

border:1px solid red;

I ran into this same issue a while back, and yeah, it’s a bit of a pain that macOS doesn’t have the `allowsPictureInPictureMediaPlayback` property like iOS does for `WKWebView`. The [Apple docs](https://developer.apple.com/documentation/webkit/wkwebviewconfiguration) make it pretty clear it’s iOS-only, so we’ve got to get creative.

From what I’ve dug up, there’s no direct built-in way to enable Picture-in-Picture (PiP) in `WKWebView` on macOS, but you can lean on the HTML5 Picture-in-Picture API. If your web content supports it (think `videoElement.requestPictureInPicture()` in JavaScript), it should work in `WKWebView`—kinda like how it does in Safari. I’ve tested this with some basic video pages, and it’s solid when the stars align. Check out [MDN’s docs](https://developer.mozilla.org/en-US/docs/Web/API/Picture-in-Picture_API) for the nitty-gritty on that.

Now, here’s the catch: if you’re building for the Mac App Store, sandboxing can throw a wrench in things. Some folks have tried using the `com.apple.PIPAgent` entitlement to force PiP—there’s chatter about it on [Apple’s forums](https://developer.apple.com/forums/thread/756472)—but Apple’s rejected apps for it unless it’s super justified. So, if you’re not going App Store, you might experiment with that, but it’s dicey.

Another workaround I’ve seen is pulling the video URL out of the `WKWebView` and playing it with `AVPlayer`, which does support PiP on macOS. It’s a hassle, though, and not always practical if the video’s buried in tricky web code. [Kodeco’s got a guide](https://www.kodeco.com/24247382-picture-in-picture-across-all-platforms) that touches on this if you’re curious.

Honestly, macOS just doesn’t seem as PiP-friendly for web stuff as iOS. If you control the webpage, the HTML5 API is your best shot. Otherwise, you might be stuck tweaking things case-by-case or waiting for Apple to throw us a bone in a future update. Anyone else found a slicker way around this?

### Key Citations

1. [Apple WKWebViewConfiguration Docs](https://developer.apple.com/documentation/webkit/wkwebviewconfiguration)

2. [MDN Picture-in-Picture API](https://developer.mozilla.org/en-US/docs/Web/API/Picture-in-Picture_API)

3. [Apple Developer Forums](https://developer.apple.com/forums/thread/756472)

4. [Kodeco PiP Guide](https://www.kodeco.com/24247382-picture-in-picture-across-all-platforms)

Ive found this Stripe API request builder tool that helps with understand the attributes related to Checkout Sessions, which of course includes `expires_at` with the description of:

The Epoch time in seconds at which the Checkout Session will expire. It can be anywhere from 30 minutes to 24 hours after Checkout Session creation. By default, this value is 24 hours from creation.

Reasoning and explanation:if the question requires explanation or justification, be sure to provide reasons along with your answar.

In PyCharm, the automatic import suggestions are usually based on shared libraries. Unfortunately, like you said, PyCharm doesn’t automatically recognize custom aliases like tf for tensorflow. I did find a workaround method that can allow you to configure PyCharm to recognize your input. You must use "Live Templates" for custom text.

Steps to re-create for tensorflow:

1. Go to File > Settings in PyCharm.

2. Navigate to Editor > Live Templates.

3. Click the + button to add a new live template.

4. Set the abbreviation to trigger the import. This will be used as the shortcut you want to type every time. For this example, set it to 'tf'

5. In the Template text area, write the statement you want. This will be what your shortcut will be replaced with. For this example, set it to 'import tensorflow as tf'

6. At the bottom, click 'Define' to set the language you want this to be usable in. Select Python.

7. Apply the changes and press 'OK'

This should allow you to type 'tf' and press tab, setting in the text 'import tensorflow as tf'.

After many trials and going through online documentation on Google, Flutter sides, I concluded that MobileNetV3 is not supported in Flutter.

Finally decided to pursue the process using ResNet50

Found the problem! When trying to reproduce the problem in a minimal code snippet, I noticed that when using only the first part of the module everything is OK. But when I add the rest of the code it fails again.

It turns out that I have, at some point, by mistake pasted an extra import numpy as np deep down in the module. So, my mistake. But hard to find.

Thank you for your comments - those actually led to the finding of the issue.

Try out https://dartpm.com/

Someone is trying to create a dart package manager taking inspiration from node package manager.
It is in beta currently, but best part is it is free right now, for private packages as well. Also it doesn't look pricy.

To publish the package (docs)
1. Create account

2. Create org, click organization in user icon dropdown

3. Install dartpm cli tool
dart pub global activate --source hosted --hosted-url "https://dartpm.com" dartpm

4. Login dartpm cli
dartpm login

5. Update pubspec.yaml of a package you want to publish
publish_to: https://dartpm.com/registry/org-name

6. dart pub publish

Also sharing the package is very easy as well.

1. Create a granular token from organization settings.
2. Add package and access.
3. Give the token to anyone to use the packages. Even client don't need to use dartpm account to get access.

This does not seem to work in 2025 with version 1.85 of n8n. Maybe it is because I am running a docker version of n8n. sigh...

Heimdall Data has a Database Proxy solution that supports caching.

your "userRepository.save(user)" returns the actual User object and not the "Optional<User>". Wrap it with Optional to match your return type.

return Optional.ofNullable(userRepository.save(user));

Test your add-on with external accounts by using a test deployment. See the details at https://developers.google.com/workspace/add-ons/how-tos/testing-editor-addons#create_a_test_deployment.

One of the possible findings is that your domain might be blocking something your add-on requires. You should solve this before continuing.

If everything is working fine, submit your add-on for OAuth verification.

Related

• Apps Script - OAuth consent screen and GWM Store Listing. Which is published first?

This works for me. I was populating a readonly textbox but the text would always be selected even if the textbox was readonly. TextBox.Select(0, 0) deselected the text.

u may have a file named pandas.py in your current working directory

try running this in a new cell:

import os

os.listdir()

if u see anything named pandas.py or pandas.pyc, delete or rename:

if u renamed or deleted, also remove the .pyc cache

!find . -name "*.pyc" -delete

restart the kernel and try again

In my clangd, I also get the same error with that code, I solved it by using:

#include <iostream>
#include <functional>

using namespace std;

int main() {
    function<bool()> fn;
}

Here is link to your solution in reddit: https://www.reddit.com/r/excel/comments/13wzhod/pythons_fstring_in_excel/

=LAMBDA(f_string, values, LET(
  is_labeled, NOT(ISNUMBER(FIND("{}", f_string))),
  is_numbered, IFERROR(ISNUMBER(VALUE(
    TEXTAFTER(TEXTBEFORE(f_string, "}"), "{"))), FALSE),
  values, IF(AND(ROWS(values) = 1, OR(is_numbered, NOT(is_labeled))),
    TRANSPOSE(values),
    values),
  value_pairs, IFS(
    is_numbered, HSTACK(SEQUENCE(ROWS(values)),
      IF(ROWS(values) = 1, TRANSPOSE(values), values)),
    COLUMNS(values) = 1, WRAPROWS(values, 2),
    TRUE, values),
  pair_rows, SEQUENCE(ROWS(value_pairs)),
  unlabeled, REDUCE(f_string, values, 
    LAMBDA(string, value,
      IFERROR(SUBSTITUTE(string, "{}", value, 1), 
        string))),
  labeled, REDUCE(f_string, pair_rows, 
    LAMBDA(string, pair_row, LET(
      label, INDEX(value_pairs, pair_row, 1),
      value, INDEX(value_pairs, pair_row, 2),
      IFERROR(SUBSTITUTE(string, "{" & label & "}", value), 
        string) ))),
  IF(is_labeled, labeled, unlabeled) ))

Found the solution with one of the issues that was raised:https://github.com/vercel/next.js/issues/10285

To add on this thread, to anyone struggling with this in 2025 and with the lack of information from google and thanks to some info spared on internet I found that you can just set your headers as follow (Here I'm using express)

it is important that status is 200. Now this is streaming for me on a prompt ui for IA using vercel ui sdk.

Not need of SSE or GRCP

  res.status(200);
  res.setHeader('transfer-encoding', 'chunked');
  res.setHeader('Content-Type', 'plain/text');
  res.setHeader('cache-control', 'no-cache');
  res.setHeader('x-accel-buffering', 'no');

Clear and Precise response: your answar should be clear and specific,directly addresing the question

why are trees important for the environment?"

Dave is correct. If you have only a theory question, please just Google it. Use StackOverflow only for a practical question that you faced and could not get an answer of from google. AI tools also made it easier to search for such theory answers.

Also, please remove the spring-boot tag here.

For your Question,

Records in Java were introduced in the Java 14 version, which simply saves efforts on making your class Immutable. Records eliminate boilerplate code like constructors, getters, setters, hashCode(), equals(), and toString() methods that you typically write for simple data-carrier classes.

Great deep dive! You're absolutely right—static final values are locked in at class loading, so anything relying on runtime args won't cut it unless passed as system properties. Loved your note on immutability too—people often overlook the mutability of collection contents. Solid insight!

class Param {
  constructor(...args) {
    this.firstName = args[0]
    this.lastName = args[1]
  }
  static parse(_, ...values) {
    return new Param(...values)
  }
}

let [name1, last1] = ["Ankit", "Anand"]

console.log(Param.parse`firstName:${name1}, lastName:${last1}`)

The advice helped to solve the problem, but the xfun package had to be reinstalled manually. After that, everything worked.

Depending on your needs, consider switching to a more feature-rich CNI:

CNI PluginHighlightsCalicoNetwork Policies, BGP, IPIP, eBPF, IPv6, great performanceCiliumeBPF-based, observability, security, DNS-aware policiesFlannelSimple, works well for smaller clustersWeave NetEasy to set up, encrypted networkingAmazon VPC CNINative AWS networking, great for EKS

More Infomation here : https://gacor.rtpug8live.xyz/

It depends on the gradient.
For example: If a horizontal lineGradient is assigned for filling and the line has x1 equal to x2 then the line is not shown, because there is no width for gradient.
The plotter should use fill="#colorHex" instead of fill="url(#gradient)" in SVG line element.

I'm facing similar issue like @terraform-ftw
I even added Artifact Registry Reader to ALL of my service accounts - Still, no positive result
I even tried the solution proposed by @xab , but there's no such a a thing like "Nodes" tab in single cluster's page image

The response some sort of cloud worker is getting is 403 Forbidden as logs suggest (kubectl describe [podName])

(I've altered some details in the logs but I hope everyone gets what are they about):

  Warning  Failed     7m8s (x4 over 8m33s)    kubelet                                Failed to pull image "us-docker.pkg.dev/{MyProjectId}/docker-images/{MyImageName}:latest": failed to pull and unpack image "us-docker.pkg.dev/{MyProjectId}/docker-images/{MyImageName}:latest": failed to resolve reference "us-docker.pkg.dev/{MyProjectId}/docker-images/{MyImageName}:latest": failed to authorize: failed to fetch oauth token: unexpected status from GET request to https://(******)%3Apull&service=us-docker.pkg.dev: 403 Forbidden

Muito obrigado funcionou assim ;)

[global]

protocol = NT1

min protocol = NT1

max protocol = NT1

client min protocol = NT1

client max protocol = NT1

i was able to fix it by setting

editor.semanticHighlighting.enabled

from "configuredbytheme" to "true" in the vscode settings

You could print out the path where the executable is looking for the assets. For example.

use std::os;

fn main() {
    let p = os::getcwd();
    println!("{}", p.display());
}

Then, either change the path in the program to be the one you want, or move the assets to the path where it's looking.

Have the same issue on all environments...
Can only completely reinstall them, unfortunately.

Use --capabilities and explicitly set all parameters

aws cloudformation deploy \
  --template-file template.yaml \
  --stack-name my-stack \
  --capabilities CAPABILITY_IAM \
  --parameter-overrides MyKey=ABC123

As @Nate Eldredge pointed out, putting assembly code into "as" and then "objdump" doesn't lead to the same code. "objdump" unfortunately messes with jump offsets. But my assumption that 0xebfe is an endless loop is correct.

What exactly are you asking? Because I'm not sure this is really a question, you literally just described how GitHub Pages works.

The "Rate Limit Exceeded" error typically indicates that your requests have surpassed the allowed quota for your OpenAI API plan. Even if it’s your first manual query, background processes like file ingestion or agent setup might have consumed the limit. Check your rate limits and usage in the OpenAI dashboard.

The Shadcn Command uses the pacocoursey/cmdk library, and according to their docs it should be:

https://github.com/pacocoursey/cmdk?tab=readme-ov-file#input-cmdk-input

const [search, setSearch] = React.useState('')

return <Command.Input value={search} onValueChange={setSearch} />

Remove 's' from the following config name spring.batch.job.name

If godaddy is is your domain provider, here's what others have said: in godaddy put send grid name and value in quotes. Remove any and all reference to your own domain name which godaddy automatically attaches. But will cause sendgrid to return an error.

\documentclass{article}

\usepackage{amsmath}

\begin{document}

\begin{equation*}

\Huge

9 + 8 = \left( \frac{\left( 16 + 5i \right)^{(4+7i)}}{\sqrt{\left( 11 + 3i \right)^{2}}} \right) + \frac{\log_{\left( 3 + 6i \right)}\left( e^{\pi i} \right)}{5} + \left( 7 + 4i \right)

\end{equation*}

\end{

document}

if a background BASH shell script has a "#!/bin/bash -i" shebang (the "-i" option tells that the script is an interactive script), then it will be stopped even if it is started via nohup

1)install react using viteintsall npm create vite@latest my-app --template react

2)downgrade react to version 18 npm install react@18 react-dom@18

3)intall chakra version 2.0 npm i @chakra-ui/react@2 @emotion/react @emotion/styled framer-motion

4)use provider to get chakra ui(check the synax in the chakra website)

import { ChakraProvider } from '@chakra-ui/react'

function App() {
  // 2. Wrap ChakraProvider at the root of your app
  return (
    <ChakraProvider>
      <TheRestOfYourApplication />
    </ChakraProvider>
  )

The previously suggested solution did not work for me anymore.
Instead, setting the environment variable

PMA_ABSOLUTE_URI=https://your.domain.tld/phpmyadmin/

fixed the issue behind a reverse proxy.
Tested with phpMyAdmin 5.2.2.

With Raku/Sparrow this is just:

begin:
machine api.mydomain.com
:any:
end:

code: <<RAKU
!raku
for streams().value -> $s {
   say $s<>[0]<data>;
   say $s<>[1]<data>;
}
RAKU

There is no way we can remove cookies from chromium, other way to prevent attacker stealing it, you can set webview incognito=true and use proguard rule to obfuscate your code, this will make your code very hard to read or reverse engineer.

In redshift you unnest the super array into rows in order to search them

https://docs.aws.amazon.com/redshift/latest/dg/query-super.html

permitir colagem ( se estiver em português )

How can d162c240 or c240d162 be 6.07456254959106 if the most significant bit is active, meaning it has a negative number?

My formula (change ‘x’ to the required cell): =(-1)^(BITAND(HEX2DEC(x);HEX2DEC("80000000"))/2^31) * (1 + BITAND(HEX2DEC(x);HEX2DEC("007FFFFF"))/2^23) * 2^(BITAND(HEX2DEC(x);HEX2DEC("7F800000"))/2^23-127)

import speech_recognition as sr
import pyttsx3
import os

# आवाज़ बोलने वाला इंजन सेट करें
engine = pyttsx3.init()

def speak(text):
    engine.say(text)
    engine.runAndWait()

# वेलकम मैसेज
speak("नमस्ते! मैं आपका वॉइस असिस्टेंट हूं।")

# कमांड सुनना और चलाना
while True:
    r = sr.Recognizer()
    with sr.Microphone() as source:
        print("मैं सुन रहा हूँ...")
        audio = r.listen(source)

        try:
            command = r.recognize_google(audio, language="hi-IN")  # हिंदी में समझेगा
            print(f"आपने कहा: {command}")

            if 'नोटपैड खोलो' in command:
                speak("नोटपैड खोल रहा हूँ")
                os.system("notepad")

            elif 'बंद करो' in command:
                speak("ठीक है, अलविदा!")
                break

            else:
                speak("माफ कीजिए, मैं वह नहीं समझ पाया।")

        except:
            speak("माफ कीजिए, कुछ गड़बड़ हुई।")

Assuming you have the RGP in the numpy array and the shape of of your array with thousands of pictures with 32*32 Pixels is (50000,32,32,3), then you can create gray scale by average of RGB

X_gray = np.sum(X_color/3, axis=3, keepdims=True)

Thanks to Mayukh Bhattacharya that solved the problem by suggesting to wrap the idx inside a MAX(idx, 1) function when using it inside MAKEARRAY!

Final function looks like that:

=LET(
dataAll;        PromoTbl[[SO_inizio]:[SO_ fine]];
daySel;         D$6;
cliente;        DASHBOARD!$B$2;
filtroCliente;  (PromoTbl[[L6]:[L6]]=cliente);
filterWk;   ((BYROW(PromoTbl[[SO_ fine]:[SO_ fine]];LAMBDA(row;MIN(daySel+7;row)))-BYROW(PromoTbl[[SO_inizio]:[SO_inizio]];LAMBDA(row;MAX(daySel;row)))+1)>=1);
filterWkBef;  ((BYROW(PromoTbl[[SO_ fine]:[SO_ fine]];LAMBDA(row;MIN(daySel-1;row)))-BYROW(PromoTbl[[SO_inizio]:[SO_inizio]];LAMBDA(row;MAX(daySel-7;row)))+1)>=1);
dataWk;         SORT(FILTER(dataAll;filtroCliente*filterWk;"NA"));
dataWkBef;      SORT(FILTER(dataAll;filtroCliente*filterWkBef;"NA"));
listWk;         BYROW(dataWk;LAMBDA(row;IFERROR(TEXT(CHOOSECOLS(row;1);"gg-mmm-aa") & "|" & TEXT(CHOOSECOLS(row;2);"gg-mmm-aa");"")));
listWkBef;      BYROW(dataWkBef;LAMBDA(row;IFERROR(TEXT(CHOOSECOLS(row;1);"gg-mmm-aa") & "|" & TEXT(CHOOSECOLS(row;2);"gg-mmm-aa");"")));
firstRow;       TAKE(listWk;1);
idx;            MAX(IFERROR(MATCH(firstRow;listWkBef;0);1);1);
spaces;         MAKEARRAY(idx;1;LAMBDA(r;c;""));
newArr;         DROP(VSTACK(spaces;listWk);1);
newArr
)

Hola tambien tuve el mismo problema, use tu solucion, pero al guardar los cambios, me cambia el mes de la fecha actual

Try to 'Erase all contents and settings' option on your simulator from

Device -> Erase all contents and settings

You could use an AI enlarger. There are many. Here is an example from https://www.creatopy.com/tools/ai-image-resizer/

same problem, advise on resolving it-Unhandled Runtime Error

Error: useUploadHandlers must be used within UploadHandlersProvider

src\app(payload)\layout.tsx (26:3) @ Layout

24 | 25 | const Layout = ({ children }: Args) => (

26 | | ^ 27 | {children} 28 | 29 | ) Call Stack 3

Show 2 ignore-listed frame(s) Layout src\app(payload)\layout.tsx (26:3)

You will have to create a VBA module and add a function

Function something (a, b)
   something = a + b * 2
End Function

Then write it in the worksheet

Make sure the `Settings Location` has right path and file.

Have you found the solution? I have the same problem and I'm trying to find the solution.

Shameless plug, but https://github.com/relaycorp/dnssec-js has been around since 2022.

@Pizza_Potamus

SharePoint Online doesn't natively support a strict "allowlist-only" access model for individual sites when Conditional Access (CA) or even SharePoint Advanced Management (SAM) is involved.

@thewildman97 Use GET /drives/{driveId}/items/{uniqueId}

If you already have the file’s unique ID (a DriveItem ID or SharePoint file ID), then:

GET https://graph.microsoft.com/v1.0/sites/{siteId}/drive/items/{uniqueId}

Or, better yet, if you want to use the path (e.g., serverRelativePath or serverRelativeUrl), you can:

Use the path-based drive/root: endpoint

GET https://graph.microsoft.com/v1.0/sites/{siteId}/drive/root:{serverRelativePath}

Example:-

GET https://graph.microsoft.com/v1.0/sites/{siteId}/drive/root:/Shared Documents/ExampleFolder/movie.mp4

This is often more reliable when dealing with paths you already parsed from serverRelativeUrl or file.

It avoids ambiguity with GUID vs. item ID mismatches.

I know this is an old thread, but for anyone stumbling upon this, I've created a plugin for this and I'm using it on my own projects:

https://www.npmjs.com/package/prisma-prefixed-ids

This is because in <iframe> embed, Google Maps defaults to gestureHandling : 'cooperative' which requires the user to hold down the Ctrl key while scrolling to zoom the map since the API cannot determine whether the page is scrollable.

To enable scrolling without holding the Ctrl key, you need to use the Google Maps JavaScript API instead of an <iframe> and set an option gestureHandling: "greedy";

Make sure the Python interpreter selected in VS Code matches the one with which you installed tensorflow.

By default you will find the selected Python interpreter in the bottom right of VS Code if you have a Python file selected. More info: https://code.visualstudio.com/docs/python/environments.

I resolve it by myself.

_listener.service = .init(type: "h3")

is unnecessary.

NWListener.service is for the Bonjure that is only used in local network.

In 2020 I needed a portable binary of OpenSSL for Windows, so I compiled it with Cygwin and made the binaries available. If you don't need the GOST (Russian Algorithms) libraries, download OpenSSL-1.1.1h_win32(static)[No-GOST].zip. It's a static binary (doesn't depend on dynamic DLL libraries), portable to have on a USB stick. No installer, Open Source Excellence.
https://sourceforge.net/projects/openssl-for-windows/

First Point: The issue was that there was no comma after the allowedTags array. In JavaScript object literals, you need commas between properties.

Second point using [/.*/] is working fine on my side

It's possible to access for some builtin iterators via __reduce__ e.g.:

xs = iter([8, 23, 45])

for x in xs:
    print("item #{} = {}".format(xs.__reduce__()[-1], x))

But I'd prefer to use enumerate generally

For other builtin iterators (excluding i.e. Generators and others) that cannot have their index retrieved via reduce see my answer here if interested: Is it possible to get index of the next item from iterator?

pd.isnull() behaves differently depending on the type of input:

pd.isnull('nan') is False because 'nan' is just a string.

pd.isnull(float('nan')) is True because it’s a real NaN float.

In a DataFrame, if a column contains 'nan' as a string, pd.isnull() will treat it as a regular value. If you convert the column to Float64, pandas will try to coerce 'nan' to np.nan, which is treated as a missing value.

Example :

import pandas as pd

import numpy as np

df = pd.DataFrame([['1', pd.NA, 'nan']], columns=['a', 'b', 'c'])

print(pd.isnull(df['c'])) # False

df['c'] = df['c'].astype('Float64')

print(pd.isnull(df['c'])) # True

Okay, after testing for hours, i got it..
I have to free the reference hold by GJS in a destroy signal handler, all other tries result in memory leaks.
I dont know why, but simply setting your references in GJS to null after using Gtk.widget.destroy() doesnt work and if you try to remove the element first from the parent and then set its reference to null and let GJS destroy it, also results in memory leaks.

This works:

#!/usr/bin/cjs
"use strict";
import Gtk from 'gi://Gtk?version=3.0';
 
Gtk.init (null);

const window=  new Gtk.Window ();
let buttons=[], button, box=new Gtk.Box ({ orientation:1}), remove=true;

function finalize_destroy (button)
    {
    buttons[button.index]=null;
    }

function construct_buttons ()
   {
   for (let i=0;i<500;i++)
      {
      button=new Gtk.Button ({label:"hello"});
      box.add (button);
      button.index=i;
      button.connect ("destroy", finalize_destroy)
      buttons.push (button);
      }
   }
   
setInterval (function ()   
   {
   if (remove)
      {
      for (let i=0;i<500;i++) buttons[i].destroy ();
      buttons=[];
      }
   else
      {
      construct_buttons ();
      window.show_all ();
      }
   remove=!remove;
   }, 2000);

construct_buttons ();
      
window.add(box);
window.set_title ("Test");
window.connect ('destroy', () => { Gtk.main_quit (); });
window.set_size_request (740, 600);
window.show_all ();

Gtk.main ();

So if i construct my own widget, which consists of a lot of Gtk widgets, i collect all widgets in a "widgets" namespace container object.
After construction i connect a handler for every widget to the destroy signal and release the references in the destroy handler.
I dont know, if that is the right way, but it works.
A more complex example:

#!/usr/bin/cjs
"use strict";
import Gtk from 'gi://Gtk?version=3.0';
 
Gtk.init (null);

const window=  new Gtk.Window ();
let top_box=new Gtk.Box ({ orientation:1}), own_widgets= [], remove=true;

//-- constructor own widgets: which is a collection of structured gtk widgets
function OwnWidget ()
   {
   var widgets=new Object ();
   
   //-- collecting all widgets i construct in container namespace widgets
   widgets.eventbox_for_hbox= new Gtk.EventBox ();                                                           
    widgets.eventbox_for_vbox= new Gtk.EventBox ({halign:1,valign:1});                                          
    widgets.hbox=              new Gtk.Box ({ spacing: 2});
    widgets.vbox=              new Gtk.Box ({ orientation:1, spacing: 2});
    widgets.button_A=          new Gtk.Button ({ label:"Button_A", halign:3, valign:3 }),         
    widgets.button_B=          new Gtk.Button ({ label:"Button_B", halign:3, valign:3 }), 
    widgets.button_C=          new Gtk.Button ({ label:"Button_C", halign:3, valign:3 }), 
   
   //-- some structure: eventbox_for_vbox (top container) -> vbox -> eventbox_for_hbox -> hbox -> buttons                                       
   widgets.hbox.add(widgets.button_A);
   widgets.hbox.add(widgets.button_B);
   widgets.hbox.add(widgets.button_C);
   widgets.eventbox_for_hbox.add (widgets.hbox);
    widgets.vbox.add (widgets.eventbox_for_hbox);
    widgets.eventbox_for_vbox.add (widgets.vbox);
    
    /* connecting destroy handler for every widget -> OwnWidget.prototype.destroy.finalize_destroy is the template function
       and adding a property to every widget called OWNWIDGET_propertyID, which is the widgets property name in the "widgets" object.
       This is needed in finalie_destroy see below. */
    for (let propertyID in widgets)
       {
       widgets[propertyID].connect ("destroy", OwnWidget.prototype.destroy.finalize_destroy.bind (this));
       widgets[propertyID].OWNWIDGET_propertyID=propertyID;
       }
       
    this.widgets=widgets;
    this.top_container=widgets.eventbox_for_vbox;
   }

//-- destroys your own widget
OwnWidget.prototype.destroy=function ()
   {
   this.top_container.destroy ();
   this.top_container=null;
   }

//-- called on signal destroy, releaving the references of GJS in instance.widgets object
OwnWidget.prototype.destroy.finalize_destroy=function destroy (widget)
   {
   this.widgets[widget.OWNWIDGET_propertyID]=null;
   }
   
//-- construct 100 own widgets with the above structure
function construct ()
   {
   let own_widget;
 
   for (let i=0;i<250;i++)
      {
      own_widget=new OwnWidget ();
      top_box.add (own_widget.widgets.eventbox_for_vbox);
      own_widgets[i]=own_widget;
      }
   window.show_all();
   }

//-- removes the 100 own widgets from the top-box and destroys it
function remove_and_destroy ()
   {
   for (let i=0;i<250;i++) 
      {
      own_widgets[i].destroy ();
      /* to be sure, that all is freed, i also set the array reference on the own widget instance null,
         but this will be overwritten by the next construct () call, so i think its not necessary. */
      own_widgets[i]=null;
      }
   }
 
setInterval (function ()   
   {
   if (remove) remove_and_destroy ();
   else construct ();

   remove=!remove;
   }, 3000);

construct ();
window.add(top_box);
window.set_title ("Test");
window.connect ('destroy', () => { Gtk.main_quit (); });
window.set_size_request (740, 600);
window.show_all ();

Gtk.main ();

✓ Linting and checking validity of types unhandledRejection TypeError: n.createContextKey is not a function at 12410 (C:\Users\admin\Desktop\jamiakh.next\server\chunks\548.js:1:4635) at t (C:\Users\admin\Desktop\jamiakh.next\server\webpack-runtime.js:1:143) at 58548 (C:\Users\admin\Desktop\jamiakh.next\server\chunks\548.js:1:33454) at t (C:\Users\admin\Desktop\jamiakh.next\server\webpack-runtime.js:1:143) at s (C:\Users\admin\Desktop\jamiakh.next\server\pages_document.js:1:378) at (C:\Users\admin\Desktop\jamiakh.next\server\pages_document.js:1:405) at t.X (C:\Users\admin\Desktop\jamiakh.next\server\webpack-runtime.js:1:1285) at (C:\Users\admin\Desktop\jamiakh.next\server\pages_document.js:1:391) at Object. (C:\Users\admin\Desktop\jamiakh.next\server\pages_document.js:1:433) { type: 'TypeError' } The error occurs every time I build the project. Please provide a solution for this.

remove syntax in MySQL when working with int value in the array

UPDATE jsontesting
SET jsondata= (
    SELECT JSON_ARRAYAGG(value)
    FROM JSON_TABLE(
        jsondata,
        '$[*]' COLUMNS (
            value INT PATH '$'
        )
    ) AS jt
    WHERE value != 7

Try using the filter instead of drop

place resources/views/errors/503.balde.php with your desired template.
tested on laraval 11, works like a charm, need to create errors directory on views.

In addition to the answer above, some sites block internet connections but allow DNS queries to be sent, or only the internal recursive DNS server connects to the internet to resolve queries. In this case, you may use LicenseDNS since it works by querying licenses through DNS queries. LicenseDNS.net

You should use css media queries. You can set different breakpoints for different display sizes. Also there is a media query for printing called print. examples:

@media print {
    /* your css */
}
@media (max-width: 450px) {
    /* your css */
}

Link to documentation: https://developer.mozilla.org/en-US/docs/Web/CSS/CSS_media_queries/Using_media_queries#targeting_media_types

I reversed the ifstream and ofstream, it was a mistake on my part

ViJay's answer worked, after I clicked on a context popup in the source code. The red mark was removed on the import line in the source and the project works. Calls org.codehaus.plexus.util.DirectoryScanner. It was confusing that the red mark on the import statement did not disappear right away. I don't remember what the context popup said but it fixed it. Apparently downloaded the library and installed it.

Who knew. It's all just magic.

can anyone find a solution to this problem?

April 2025

Refer to https://github.com/firebase/flutterfire/issues/13228#issuecomment-2325843150 also update the FirebaseFirestore versions in podfile to 11.10.0

to solve it : download php_http.dll and place it to C:\xampp\php\ext\

but i also still looking where to download the extension too

Use FormSubmit

No additional JavaScript required and the system is imbedded in the html code itself. As for the painting ID you'll have to add some JS or figure out a way yourself.

What Alan Birtles said. The Xcode project you're trying to build is 14+ years old. Use ioQuake3.

How about this?

(?<![=|] )(?<![=|])\b\d{1,2} [a-zâéîôû]{3,9} \d{4}\b

Even though I have added the Request validation error as said by Chris, still it is throwing the default whole error content.

<?php

if (isset($_GET['domain'])) {

echo '\<pre\>';

$domain = $\_GET\['domain'\];

$lookup = system("nslookup {$domain}");

echo($lookup);

echo '\</pre\>';

}

?>

same facing the same issue

psql: error: connection to server at "127.0.0.1", port 6432 failed: FATAL: server login failed: wrong password type

Thanks to @The fourth bird that answered the question in the first post comments:

(?<!(?:\w+ *[=|] *|{{\w+ *\| *)[\w ]*)(?:(\d+(?:er)?|\d+{{er}}|{{\d+er}}) *)?(janvier|f[ée]vrier|mars|avril|mai|juin|juillet|a[ôo][ûu]t|septembre|octobre|novembre|d[ée]cembre) *(\d{3,4})

regex101.com/r/bqHMlI/1