You're encountering this behavior because `setlocal` does **not** affect the current working directory or the directory stack managed by `pushd` and `popd`.

### What's happening in your script

In `start.bat`:

```bat

@echo off

pushd 111

call 1.bat

popd

echo CD is wrong:%cd%

pause

In 1.bat:

@echo off

setlocal

pushd 222

echo 1.bat says we are in %cd%

When you run start.bat, it prints:

1.bat says we are in C:\ex1\111\222

CD is wrong:C:\ex1\111

This happens because you're doing a pushd inside 1.bat, but you never do a matching popd, so the directory stack isn't fully reverted.

Also, setlocal only controls the environment variables (like %PATH%, %TEMP%, etc.) — it has no impact on the current directory or the directory stack.

I have problem like this: "Bad owner or permissions on _PROGRAMDATA_\\ssh/ssh_config".

My problem is my account is add new to PC, them old user can using ssh but new account can not.

I fix by these steps:

• go to "C:\ProgramData\ssh"

• do like image

  enter image description here

@teylin Please read the Note also on Microsoft site XLOOKUP function. The fourth parameter is implicitely defined (default value), and the formula works in the data structure the OP posted. Otherwise I really like XLOOKUP because more flexible.

1. Go to https://developers.google.com/apis-explorer/#search/admin%20api/appengine/v1/appengine.apps.domainMappings.create

2. Fill the request: point appId to your app, set overrideStrategy to OVERRIDE and fill request body like this:

   {
     "id": "yourdomain.com"
   }
   

3. Execute the request.

4. SSL certificates will be generated in several minutes.

Ok, I believe I have the solution. It appears that floating point operations for this particular processor is extremely expensive (at least compared to available memory) and any use will increase the binary significantly. For reasons that I really don't understand, if I comment out the *pBuf++ = 0 line, the optimizer is somehow able to optimize out some of the floating point operations in this function and hence the memory is saved.

The bottom line is that, for this processor, I need to avoid floating point at all costs.

Thanks to all who took a look at this.

I found W3Schools and geeksforgeeks pretty helpful when I was starting on C++. Once you are familiar with those, I would recommend doing some simple coding exercises on leetcode or any other platform. There are some good talks from cppcon freely available on youtube, but I would go there only when my fundamentals are clear.

My approach is to

// modify your function makeCall(target)
function makeCall(target) {
    outgoingSession = userAgent.call(target, callOptions);
    setupMedia(outgoingSession);
}

and then implement

session.connection.addEventListener('track', ({ track, streams: [stream] }) => {
    // logic identical to your peerConnection.addEventListener("track", (event) => {});
});

within setupMedia.

The specific reason, I suspect, is that addEventListener only captures newly added tracks. In your method, during outgoing calls, the audio may not be considered newly added?

Anyone having this problem in 2025, the default username is "phpmyadmin". I installed it on a Debian based system with apache2 and it prompted me for password during installation so I knew my password. I accessed the necessary files and found it.

You may follow this guide. Hope it helps.

https://anurajapaksha.blogspot.com/2025/05/how-to-re-execute-failed-step-functions.html

heres the text string version...

select pg_get_function_identity_arguments(oid)
from pg_proc
where proname = 'name'

The tab_corr function takes in a correlation matrix as computed by the cor() function and formats it. The apa.cor.table() function takes in data, calculates the correlations and formats the output.

Having trouble integrating Facebook Login with a Vercel-deployed React app.

Pandas works for ASCII as well.

import pandas as pd
data=pd.read_fwf("/file/path.asc")

replace "res.Data.map" with "res.map".

That's what I'm seeing from your json anyway.

---------------------------

he procedure entry point OCIClientVersion could not be located in the dynamic link library C:\xampp\php\ext\php_oci8_11g.dll.

i am using PHP 7.3.2 I made all setting but above error shows

i want to connect to oracle 11g .. thanks

" - string (double quote)

var str string = "Hello world!" // valid

' - rune (equals to char in C/C++)

var str string = 'Hello world!' // invalid, error: more than one character in rune literal
var c rune = 'H' // valid

` - string (single quote)

var str string = `Hello world!`  // valid

This is one of the 💩est common bugs out there

I had the exact same issue. This isn’t a bug - Chrome automatically injects inline CSS when displaying XML files to make them look prettier with syntax highlighting and collapsible elements. Your CSP blocks these styles.

The Fix

Add these specific SHA256 hashes to your CSP policy:

Content-Security-Policy: style-src 'self' 
  'sha256-47DEQpj8HBSa+/TImW+5JCeuQeRkm5NMpJWZG3hSuFU=' 
  'sha256-p08VBe6m5i8+qtXWjnH/AN3klt1l4uoOLsjNn8BjdQo=';
  img-src 'self' data: https://www.w3.org/2000/svg;

How to Get These Hashes

Method 1: Check Chrome DevTools Chrome actually tells you the hash in the error message! Look at your console error - it says ('sha256-47DEQpj8HBSa+/TImW+5JCeuQeRkm5NMpJWZG3hSuFU='). That’s one of the hashes you need.

Method 2: Use Online CSP Hash Generator

1. Open your sitemap.xml in Chrome

2. Right-click → View Page Source

3. Copy any <style> content you see

4. Use a CSP hash generator tool to convert it to SHA256

5. Add the hash to your policy

Why These Specific Hashes?

• 'sha256-47DEQpj8HBSa+/TImW+5JCeuQeRkm5NMpJWZG3hSuFU=' - Hash for Chrome’s base XML styling

• 'sha256-p08VBe6m5i8+qtXWjnH/AN3klt1l4uoOLsjNn8BjdQo=' - Hash for Chrome’s tree structure styles

These are the actual CSS content hashes that Chrome’s XML viewer uses.

Don’t Use unsafe-inline

// Bad - opens security holes
style-src 'self' 'unsafe-inline';

// Good - only allows Chrome's XML viewer styles  
style-src 'self' 'sha256-47DEQpj8HBSa+/TImW+5JCeuQeRkm5NMpJWZG3hSuFU=';

Using 'unsafe-inline' defeats the whole purpose of having CSP. The hash approach only allows the exact styles Chrome needs.

Why Firefox Works Fine

Firefox doesn’t inject inline styles for XML display, so it doesn’t trigger CSP violations.

Result

After adding those hashes:

• ✅ Chrome displays your sitemap.xml with proper formatting

• ✅ No more CSP errors in console

• ✅ Your security policy stays strict

• ✅ Search engines can still crawl normally

Tested this on Chrome 136+ and it works perfectly. Your sitemap will look nice and formatted while keeping CSP protection active.

If you're ok with having a different user name in your RCS checkins, you can set the LOGNAME environment variable. You can put this in your .bashrc:

export LOGNAME="nameatexample.com"

The problem has been solved. Here are my personal insights. Please correct me if I'm wrong. This problem can be abstracted as: in all real-time planning problems, how to synchronously update the shadow variables when changing non-planned variables. In the changeProblemProperty() method of version 9.44.0, it automatically updates the shadow variables, but since I changed the non-planned variables, this modification did not trigger the modification of the listener. I was very troubled by this problem. It's impossible to set this non-planned variable as a planned variable, right? Later, after studying the shadow variables in the official Optaplanning manual, I noticed the description of Piggyback shadow variable, and understood that shadow variables can also influence each other. So, in the changeProblemProperty(), I manually updated the shadow variables, and as I expected, this update triggered the listener, and the subsequent shadow variables also changed, achieving the expected result.The problem has been solved. Here are my personal insights. Please correct me if I'm wrong. This problem can be abstracted as: in all real-time planning problems, how to synchronously update the shadow variables when changing non-planned variables. In the changeProblemProperty() method of version 9.44.0, it automatically updates the shadow variables, but since I changed the non-planned variables, this modification did not trigger the modification of the listener. I was very troubled by this problem. It's impossible to set this non-planned variable as a planned variable, right? Later, after studying the shadow variables in the official Optaplanning manual, I noticed the description of Piggyback shadow variable, and understood that shadow variables can also influence each other. So, in the changeProblemProperty(), I manually updated the shadow variables, and as I expected, this update triggered the listener, and the subsequent shadow variables also changed, achieving the expected result.

Here is the updated code.

private static void modifyTheOrderQuantityToProblem(SolverManager<ProductionSchedule, UUID> solverManager,
                                                        UUID problemId, Long sl1, Integer sl2) {
        solverManager.addProblemChange(problemId, (workingSolution, problemChangeDirector) -> {
            List<ProductionOrder> allOrder = workingSolution.getOrders();
            allOrder.stream().filter(order -> order.getOrderId().equals(sl1)).findFirst().ifPresent(order -> {
                problemChangeDirector.changeProblemProperty(order,orderOne->{
                    orderOne.setQuantity(sl2);
                    List<LocalDateTime> startTimeAndEndTime = orderOne.getStartTimeAndEndTime(workingSolution.getTimePeriodList());
                    orderOne.setEndTime(startTimeAndEndTime.get(1));
                });
            });
        });
    }

I've encountered this frustrating issue before! Often, it stems from inconsistent indentation where the space and tab sizes are set to the same value.

To fix it, try updating your tab size to 3 or 4 spaces. This usually resolves the problem.

You can easily visualize the indentation in your Makefile by using the command:

Bash

cat -A Makefile

This command will reveal how your indentation is represented. Specifically, if your commands are correctly indented with tabs, you'll see a ^I character at the beginning of those lines, like this:

^I@echo "Current WORKSPACE: $(WORKSPACE)"$

This ^I signifies a true tab character, which is crucial for Makefiles.

First, let’s reiterate that unsafe multithreaded access to C++ standard library containers is undefined behavior. Simultaneous read operations are safe for multithreading, but simultaneous writes or a read and write occurring simultaneously are unsafe for multithreading.

Why does unsafe multithreaded use of an std::unordered_map crash more often than unsafe multithreaded use of a std::map?

Additionally, do not use operator[] on std::(unordered)map. It is the most convenient syntax, but is not often what you actually want.
The std::map subscript operator is a convenience, but a potentially dangerous one
The operations for reading and writing single elements for C++ standard library maps
A simplified overview of ways to add or update elements in a std::map

Okey here any one get number to location in bd
dm me in Session Messenger
my id : 058e8c7e8c1d0fa15773b77dfb4530c7e1ae0c192addf0b5597ba08bafecc79962

Managed to fix with retentions using loki helm 6.30. Below the reference.

    runtimeConfig:    
        overrides:
            tenant_1:
                retention_stream:
                  - KEY: VALUE
                  - KEY: VALUE
            tenant_2:
                retention_stream:
                  - KEY: VALUE

https://community.grafana.com/t/loki-retention-overrides-in-helm-not-working/136728/8?u=heshand

Great capabilities, what about the headaches when you need to Implement Merge replication on temporal tables in an environment with 3 or more nodes? Almost impossible to accomplish. All the workarounds suggested out there simply don’t work. Any advice?

I had to add the user email from the Google Cloud service account to the credential, like this:

credential = GoogleCredential.FromStream(stream)
    .CreateScoped(scopes)
    .CreateWithUser("[email protected]");

were you able to implement Transbank OnePay in Chile? I managed to implement WebPay. OnePay is quite similar, and I should be implementing it soon. If you still need help, I can help. Best regards.

Did you solve this?
We are working with a partner organization that use AWS SES. Their emails to us get rejected by our server with that exact message.
The server hosting our email definitely supports STARTTLS; I confirmed this by forcing my Outlook client to specify STARTTLS when sending.
All the reading I've done thus far suggests the sender is the one that needs to change something.

If your storyboard is stuck in XML but View->Show Code Review isn't active, then Vivek Patel's advice will fix it.

My program is showing output but it for some reason it just skips the section "Please enter your User Name or your Official Name ". How can I fix this?

The reason your program skips is because it uses operator >> to extract the userid.

// From the OP:
char userid[100];
cin >> userid;

That operator only extracts one "word" at a time. By "word" I mean a non-empty sequence of charaters that does not contain a space or other whitespace character.

Extraction stops when a whitespace character is encountered, leaving the whitespace character in the stream. Note that the newline character ('\n') is a whitespace character. Extraction also stops if end-of-file file is encountered, but, in general, that is not an issue when using std::cin to get user input from the keyboard.

In your trial run, you entered two words for userid. Thus, "John Smith" was placed into the input stream. When you pressed Enter, that keystoke was also placed into the stream (as a newline). The stream contents at this point were "John Smith\n".

The statement cin >> userid; caused first word, "John", to be extracted, and stored in userid. The second word, "Smith", along with the space that precedes it, was left in the input stream.

The statement

cin.ignore();

extracts and discards a single character, in this case the space before "Smith."

The next input is read using std::getline.

string username;
getline(cin, username);

std::getline reads an entire line of characters, stopping either when a newline is encountered or when the stream hits end-of-file. If getline stops because it found a newline, it extracts, and discards, the newline.

Blank lines are okay, as are lines containing whitespace. getline happily extracts and stores whatever it finds.

In your program, getline extracted "Smith", and stored it in the string username. getline also extracted the the newline character that followed "Smith". It was discarded.

Note that all of this happens without any need for the console window to pause for user input. That's because keyboard input from cin is buffered. The only time cin causes the console window to pause, is when its buffer is empty. In this case, the buffer was not empty. It still contained "Smith\n". So getline read from the buffer, without pausing for keyboard input from the user.

From the outside, it may look like your program skipped username, but it did not. "Smith" was extracted from the buffer, and stored, in username.

After that, the buffer was empty. Thus, the program paused, as expected, to get keyboard input for passphrase.

char passphrase[300];
cin.getline(passphrase, 300);

I think all the answers are amazing and valid in case of me trying to register a set of options based on another set of options

but what I want to get an instance of IOptions<AppSettings> to do something like this:

var appSettings = sp.GetRequiredService<IOptions<AppSettings>>().Value;
if (appSettings.EnableDebugTools){
    // some code here 
}

this is during ConfigureServices()

Did you found the solution for this? I'm struggling with the same issue.

Found this solution but is not working for me: https://community.amazonquicksight.com/t/embedurl-not-working/15197/5

As Alexandru suggested,

The stack trace shows that another updateStep() method is used to enter WorkflowService, without @Transactional.

And @Transactional annotations have never been working for local, inner-service calls, thus we get no transaction.

Just need to add:

    [assembly: Parallelize]

In the AssemblyInfo.cs

@MarzSocks not allowed to comment. I used the solution of @Vince

import { Directive, HostBinding, Input } from '@angular/core';
import { MatMenuPanel, MatMenuTrigger } from '@angular/material/menu';


@Directive({
    selector: '[matMenuTriggerForContext]',
    host: {
        'class': 'mat-mdc-menu-trigger',
        '[attr.aria-haspopup]': 'menu ? "menu" : null',
        '[attr.aria-expanded]': 'menuOpen',
        '[attr.aria-controls]': 'menuOpen ? menu.panelId : null',
        '(contextmenu)': '_handleContextMenu($event)',
      },
      exportAs: 'matMenuTriggerContext'
})
export class MatMenuTriggerForContextDirective extends MatMenuTrigger {
    @Input('matMenuTriggerForContext')
    get _matMenuTriggerForContext(): MatMenuPanel | null {
        return this.menu;
      }
    set _matMenuTriggerForContext(v: MatMenuPanel | null) {
        this.menu = v;
    }

    _handleContextMenu($event: MouseEvent): boolean {
        $event?.stopPropagation();
        $event?.preventDefault();
        $event?.stopImmediatePropagation();
        this._handleClick($event);
        return false;
    }
}

and added

/** Handles click events on the trigger. */
override _handleClick(event: MouseEvent): void {
    if (event.button !== 2) {
        return;
    }
    if (this.triggersSubmenu()) {
        // Stop event propagation to avoid closing the parent menu.
        event.stopPropagation();
        this.openMenu();
    } else {
        this.toggleMenu();
    }
}

to filter for right click (event.button == 2) This works, maybe some more changes are needed for perfectly user experience

Linux buffers are implemented in memory to improve I/O performance by reducing direct disk access and the buffer cache stores block device data, while the page cache handles file data. Linux OS uses structures like "buffer_head" and "page" in the kernel to manage these buffers. Data in buffers is flushed to disk by background processes like "pdflush" or "flush-kthreads"

Understanding what data is shared is easier when you visualize your data using memory_graph, a simple example:

See how memory_graph integrates in Google Colab.

<sub>Visualization made using memory_graph, I'm the developer.</sub>

I have this step-by-step git repo to create the most maintainable Firebase Functions. [https://github.com/felipeosano/nestfire-example][1]

If you have a project created on console.firebase.google.com then just clone the repo and change the .firebaserc with your project id.

This repository uses NestJS and NestFire. NestFire is an npm library that allows you to deploy Nest modules to Firebase Functions. This npm library also allows you to create fast and maintainable triggers. I recommend reading the documentation in its repository.

i have been having the same problem. I want to use -pb_rad charmm and -pb_rad roux, . But the given patch does not work . could you please help me with this. Thank you. Also what does "I just had to remove the last eval from your answer " means ? what is the replacement that actually works.

set ar_args [list assign_radii $currmol $pb_rad]
if { $pb_rad eq "charmm" } {
    # foreach p $parfile { lappend ar_args $p }
    eval [linsert $parfile 0 lappend ar_args] ;# Avoid a loop
} elseif { $pb_rad eq "parm7" } {
    lappend ar_args $topfile
}
eval $ar_args

You can apply the radius to the image using imageStyle:

<Image
  source={{ uri }}
  style={[StyleSheet.absoluteFill, { resizeMode }]}
  imageStyle={{ borderRadius: radius }}
/>

Looks like the Agenda component is broken in Expo SDK 53. Here's a PR you can take a look at and patch: https://github.com/wix/react-native-calendars/pull/2664

does anyone know how to adjust the legend generated by the Kcross graph? Is it possible to include some commands to make this adjustment? I would like to organize the location of the legacy.

Please try below command and see if it is working. in fact, It worked for me.

$ zz -FF IBM_SFG_OLD.zip --out IBM_SFG_NEW.zip -fz

The answer is Yes!

You can get the contents from fetch, change the visible location with function history.pushState() and totally replace document.documentElement.innerHTML with fetched data. The only thing that will be lost - some http headers which could affect the behavior of the page.

The answer is No!

In most cases once you received the response from server on your POST-request its state will never be the same. If you just walk to new location /someform server will not receive POST and the reply will differ from the reply you already consumed by fetch. Of course there are some servers that always reply the same ( something like 'OK'). But don't get your hopes up.

So, which answer do you prefer?

What archX are you using for your OS? 32 or 64 bit? I'm using Debian 64 on ARM64, and I think I'm running into a depreciation in regards to the "sizeof()" or "strlen()" functions to correctly allocate buffer space for the yolo object arguments. It looks like the original code was bound to a 32 bit environment.

The kicker is that this is my SECOND time going through this journey... accidentally deleted my live and backup images.. "DOH!"

This fellow was helpful... Segmentation fault (core dumped) when using YOLOv3 from pjreddie

The blastula package enables displaying an html string in the RStudio viewer. This avoids having to save an html object to a file.

my_html <- "<h1>Here's some HTML</h1>
<p>Here's a paragraph</p>
</br>
</br>"
blastula::compose_email(blastula::md(my_html))

Import matplotlib.pyplot as plt

Services = ['Консультации', 'Материальная помощь', 'Социальное сопровождение', 'Психологическая помощь', 'Юридическая помощь']

Service_counts = [1750, 600, 350, 500, 250]

Plt.pie(service_counts, labels=services, autopct='%1.1f%%')

Plt.title(«Структура оказанных социальных услуг Отделением 'Служба семьи' в 2023 году»)

Plt.show()

ThisWorkbook.Path will provide the path of the current workbook.

EC2Launch is required to utilize user_data. You'll want to troubleshoot that before expecting this to run.

You should be able to achieve this using worklets: https://github.com/software-mansion/react-native-reanimated/discussions/7264

Here's an article you can read: https://dev.to/ajmal_hasan/worklets-and-threading-in-reanimated-for-smooth-animations-in-react-native-98

Changing self.__size = __first.__size to self.__size = first._First__size should correctly handle the name mangling and avoid giving you an error.

For others finding this question in the future, there is a step-by-step tutorial here: https://lopez-ibanez.eu/2024-redheur/

The only possibilities you have are:

- An incomplete recovery to a point in time before the command was executed, or
- Restore a cold backup of a timestamp prior to the time of the command
- Manually add a column (and insert the values).

but before do that you must drop the column first

Just as a test, try the file:// prefix, but with an absolute file path rather than a 'shared' folder.

I had the same problem and fixed it by making the enum Codable. I'm wondering if anyone knows why an enum has to conform to codable in SwiftData when it doesn't when using structs. Also, this error message was completely unhelpful in locating the issue, is this the kind of thing Apple wants feedback about?

Sterling File Gateway will maintain the file transfer statical information along with Event codes. These event codes are specifically designed for IBM Control Center. Each code will particular describe the status of the file at each phase during the file transmission process like File Arrived, Replayed, redelivered, completed/success or Failed. If we integrate SFG with IBM control center, by default all these events will be captured by ICC.

Solution#1: You can configure CDC in ICC database so that real-time events will be captured and make sure these captured into another database within the same database instance. Now, you can write a program in any programming language (python preferable) to get these events from new database. In python program, you can subscribe to pub/sub topic to push them to bigquery and the further we can project them to looker studio.

Solution#2: Once ICC captures events from SFG, have your SQL query get the required events from ICC database by using python program. Please make sure you python program should send events/messages in json format to the pub/sub topic which is already subscribed to. And then, create bigquery dataset to pull the pub/sub messages from topic subscription. And then finally , we can project these events to Google looker studio for user interface.

It happened to us today. Same message, the Whatsapp number is flagged but not blocked. The template was tested today and was fine. Now I´ve just tested it and it´s fine again. Can it be a massive temporary glitch in the system?..

Use the following code:

val view = ComposeView(this).apply {
    setContent {
        // content
    }

    // Trick The ComposeView into thinking we are tracking lifecycle
    val lifecycleOwner = ComposeLifecycleOwner()
    lifecycleOwner.performRestore(null)
    lifecycleOwner.handleLifecycleEvent(Lifecycle.Event.ON_CREATE)
    setViewTreeLifecycleOwner(lifecycleOwner)
    setViewTreeSavedStateRegistryOwner(lifecycleOwner)
}

ComposeLifecycleOwner:

import android.os.Bundle
import androidx.lifecycle.Lifecycle
import androidx.lifecycle.LifecycleRegistry
import androidx.savedstate.SavedStateRegistry
import androidx.savedstate.SavedStateRegistryController
import androidx.savedstate.SavedStateRegistryOwner

class ComposeLifecycleOwner : SavedStateRegistryOwner {
    private var mLifecycleRegistry: LifecycleRegistry = LifecycleRegistry(this)
    private var mSavedStateRegistryController: SavedStateRegistryController = SavedStateRegistryController.create(this)

    /**
     * @return True if the Lifecycle has been initialized.
     */
    val isInitialized: Boolean
        get() = true

    override val lifecycle = mLifecycleRegistry

    fun setCurrentState(state: Lifecycle.State) {
        mLifecycleRegistry.currentState = state
    }

    fun handleLifecycleEvent(event: Lifecycle.Event) {
        mLifecycleRegistry.handleLifecycleEvent(event)
    }

    override val savedStateRegistry: SavedStateRegistry
        get() = mSavedStateRegistryController.savedStateRegistry

    fun performRestore(savedState: Bundle?) {
        mSavedStateRegistryController.performRestore(savedState)
    }

    fun performSave(outBundle: Bundle) {
        mSavedStateRegistryController.performSave(outBundle)
    }
}

Source: https://gist.github.com/handstandsam/6ecff2f39da72c0b38c07aa80bbb5a2f

Thanks to @CommonsWare for giving me the idea!

you should check the apache log if there is no log for that problem could be related to your Client DNS Server. maybe your dns server respond faster than your local machine.

As Sam Nseir suggested you can use Dynamic M query parameters, you can check the following question that seems to be similar to yours:

How to change the power query parameter from the Power BI Interface?

A part from MS the reference given by Sam Nseir you can check this article that can be helpful:

https://blog.crossjoin.co.uk/2023/01/29/passing-any-arbitrary-value-from-a-power-bi-report-to-a-dynamic-m-parameter/

I have resolved it by downgrading version of react-native-screen.

"react-native-screens": "^2.18.1",

I found the solution to this issue on Github;
https://github.com/coderforlife/mingw-unicode-main/

That short, simple code fragment solved the problem completely!!

So, going from here, I have to update my environment variable from

e TESTCONTAINERS_HOST_OVERRIDE=172.17.0.3

to

e TESTCONTAINERS_HOST_OVERRIDE=host.docker.internal

and it worked

go to %temp% and delete hsperfdata_<os_user> folder

check your package.json, it should have the autolinking paths 'searchPaths' so that the packages in these paths are autolinked by expo

  "expo": {
    "autolinking": {
      "searchPaths": [
        "../../packages"
      ],
      "nativeModulesDir": "../../packages"
    }
  },

This topic was very confusing to me, so I looked it up and found this link - which confused me more.

My understanding is that:

1. git rm removes the file from both the working directory (aka worktree, disk) and the staging area (aka index).

2. git rm -cached removes the file from only the staging area

So if you created a whole new git repository, created a new file, added the new file to the staging area, then ran git rm -cached, basically all you've done is unstage the new file and the new file still exists in the working directory so it can be added back to the staging area. Had you ran git rm instead, you would have removed the file from the staging area AND the working directory so it couldn't be added back to the staging area without creating the file again.

I'm guessing the OP's original question was why PaginationResult.java was still listed in the staging area after running git rm --cached. At least that's where I'm still confused.

You need to be using spherical kmeans. Yes, minimizing the euclidean distance between two l2 normalized vectors is the same with minimizing cosine, but you also have to l2 normalize the centroids, which you dont have access to in sklearn Kmean.

Adding to Florian's answer above, I recently attempted some more reverse-engineering on this, which has led me to a lot of new information.

What SciPy calls the function_workspace is actually called the subsystem. The subsystem appears in the MAT-file as a normal variable of mxUINT8_CLASS, i.e., its written as a uint8 variable with just one caveat - this variable has no name. Whether the subsystem is present or not is indicated by a 64-bit integer in the MAT-file header. This integer is a byte marker pointing to the start of the subsystem data in the MAT-file.

Coming to its contents, the subsystem contains all necessary data required to construct object instances of mxOPAQUE_CLASS objects. This includes both user-defined classes (using classdef), as well as datatypes like string, datetime, table, etc.

The way it achieves this is by encoding instructions on object construction using methods of a class FileWrapper__. These instructions include information such as the class type, its properties and their values, and the construction order (in case of nested objects). Each constructed object is mapped to a unique object ID, which is used to link this constructed object to its corresponding variable when reading them in the normal portion of the MAT-file.

The exact format of the subsystem data is extremely convoluted to get into, I would refer you to my documentation which goes into it in great detail.

I did this for my computer vision project at university to train a model using the max size for YOLO. I would use pillow to break the image in to smaller images, which you can do easily with list slicing.

As for question 2, the resolution is going to be different because you sliced the image into smaller images which directly affects the resolution, but the PPI will remain the same resulting in an image with high clarity yet smaller in size.

# using  join

set a {abcd}
set b {efgh}
set l [list $a $b] # a list
set c [join  $l ]  # space sep (default)
set d [join  $l  {}]  # no sep

puts $c

abcd efgh

puts $d

abcdefgh

Try this:

from selenium import webdriver
from selenium.webdriver.chrome.options import Options

optionss = Options()
optionss.add_argument("--start-maximized")
optionss.add_argument("--disable-popup-blocking")
optionss.add_argument("--ignore-certificate-errors")
driver = webdriver.Chrome(options=optionss)

If you already doing this, please show us all code.

Does it work this way ?

buttons = driver.find_elements_by_xpath("//*[contains(text(), 'Download All')]")

for btn in buttons:
    btn.click()

Inspired from Click button by text using Python and Selenium

This appears to be a known bug. Linking PR, which has been sitting dormant for almost 2 years

https://github.com/apache/flink/pull/25410

unless i misunderstand, looks like you just need to restructure your result differently.

const result = {};
for (let i = 7; i < jsonData.length; i++) {
    const rowData = jsonData[i];
    const singleRow = {};
    singleRow.water = { total: rowData[2], startDate: rowData[6]; }; // add other data as needed
    singleRow.electricity = { total: rowData[14], startDate: rowData[19]; };
    singleRow.gas = { total: rowData[9], startDate: rowData[11]; };

    const date = rowData[0];
    result[date] = singleRow;  // set property as row date
}

it seems you have all the pieces you need already. is there a particular issue you are having?

I have exactly the same problem, incredible!!!
I had to rollback to version 17.13.7 (where I have the problem with Output-Window without line breaks)

Use the class

unreal.PixelStreaming2StreamerComponent

Check the api documentation

https://dev.epicgames.com/documentation/en-us/unreal-engine/python-api/

It's May 28, 2025 and my team is running into this same issue. It potentially impacts certs using 4096-bit keys (rather than 2048). I've also noticed that if I upload a .pem version of the same cert (instead of .pfx) it seems to work.

PEM format:

• Cert

• Encrypted Key

• Cert Chain

emailObj.AddAttachment "c:\windows\win.ini"

How can i use if file has ddmmyyhhmmss date and time stamp like win_05282025122200.ini ??

i have follow this video, it's very helpful https://www.youtube.com/watch?v=zSqGjr3p13M

I have done this, dir.create(“testdir”) even copy pasted it. It does not accept it. Have spent over an hour collectively trying to figure this out. Would it be wise to shut the app down and restart or am I doing something incorrectly?

Appreciate your time and help

Navigate to the github address that you listed above.

Click on the "Code" dropdown menu.

Click on the "Download Zip" button.

For removing at a index you could do:

    public static T[] removeArrAt<T>(T[] src, int idx) {
        return src.Take(idx).Concat(src.Skip(idx + 1)).ToArray();
    }

which returns the modified array

try this wait = WebDriverWait(driver, 10) button = wait.until(EC.presence_of_element_located((By.XPATH, "//button[contains(., 'Download All')]")))

driver.execute_script("arguments[0].scrollIntoView(true);", button) driver.execute_script("arguments[0].click();", button) can you share a screenshot of the element in the browser dev tools or the full HTML block from outerHTML

After speaking with the dev team the correct answer here is that tethering can mitigate the issue somewhat, but backpressure cannot be fully eliminated from aeron at this time

For anyone still having this problem in May 2025, upgrading to the latest gtrendsR dev version helped:

install.packages("remotes")
remotes::install_github("PMassicotte/gtrendsR")

I didn't have such a rule in place, but experienced a similar issue, and found adding the following rule to .eslintrc.json did make the annoying line-length error go away:

"rules": { "@typescript-eslint/max-len": "ignore" }

The solution was simple--I am not sure what originally caused the problem.

The solution is to use "git remote add" to restore the target and associate it with the correct URL.

<canvas width="100" height="100" style="image-rendering: pixelated;"> will do the trick.

Found some notes from past reports and it turns out to be more than just merging columns. The generated group column must be deleted without deleting the group then add a row at that level. Then the columns can be merged.

See Examples for information on how example functions are named.

The example ExampleMain refers to the identifier Main.

The go test command runs the go vet command. The go vet command reports an error because the package does not have an identifier named Main.

Fix by associating the example with an exported identifier or the package.

In the following code, the example is associated with the package and has the suffix "jasmine".

package main

func Example_jasmine() {
    main()
    //Output: Hello, world!
}

Run it on the Playground!!!

I checked your code, but can't find any problem.

Here is class inheritance in Django.

ModelViewSet <- GenericViewSet <- GenericAPIView <- APIView

So if it works with APIView, it should work with ModelViewSet too.

Old question but a newbie, like me, might struggle with it.

So I created a directory, Production, then later renamed it production. My original files were stored in Github under Production, all new ones in another directory production. When I worked locally, (windows), it was all fine under production.

Sounds just like what you are asking.
Simply fix really, ignore all the git stuff.
Rename your local directory something else, let's call it test. Commit and push. All of your files and directories will be aggregated under the new directory test. Rename it back to your lower case, production, and commit and push. All issues resolved.

It’s a bit late, but I’d say you need to base64 encode the data you’re passing in on VersionData

(Edit your file content reference and pass it through base64(…) )

Because of using '\n' in scanf() function, you have problem on it. Just use this:

scanf("%d",&a);

CKEditor does not initialize on new CollectionField items in EasyAdmin Symfony — how to fix?

I'm using EasyAdmin 4 with Symfony, and I have a CollectionField that includes form fields with CKEditor (via Symfony's FOSCKEditorType or similar).

CKEditor loads fine for existing collection entries, but when I add a new item dynamically via the "Add new item" button, the CKEditor does not initialize on the new textarea.

🔍 What I tried:

• CKEditor is loaded and works on page load.

• New textareas appear when I add items, but CKEditor doesn't attach to them.

• Tried triggering CKEDITOR.replace() manually — no success.

• Using default Symfony + EasyAdmin JS setup, no custom overrides.

✅ Answer:

This is a common issue because CKEditor needs to be manually re-initialized on dynamically added fields inside CollectionField.

🔧 Fix: Manually re-initialize CKEditor when new fields are added

You can do this by listening to the ea.collection.item-added event that EasyAdmin dispatches.

Add this JavaScript to your EasyAdmin layout or as a custom Stimulus controller:

javascript

CopyEdit

document.addEventListener('ea.collection.item-added', function (event) { const newFormItem = event.detail.item; const textareas = newFormItem.querySelectorAll('textarea'); textareas.forEach(textarea => { if (textarea.classList.contains('ckeditor') && !textarea.id.startsWith('cke_')) { // Replace with CKEDITOR.replace or ClassicEditor.create depending on your setup CKEDITOR.replace(textarea.id); } }); });

🧠 Notes:

• Make sure the textarea has a unique ID, otherwise CKEditor won't attach.

• You might need to delay execution slightly using setTimeout() if CKEditor loads too early.

• If you’re using CKEditor 5, use ClassicEditor.create() instead of CKEDITOR.replace().

📦 Example (CKEditor 5):

javascript

CopyEdit

import ClassicEditor from '@ckeditor/ckeditor5-build-classic'; document.addEventListener('ea.collection.item-added', function (event) { const newFormItem = event.detail.item; const textareas = newFormItem.querySelectorAll('textarea'); textareas.forEach(textarea => { if (!textarea.classList.contains('ck-editor__editable')) { ClassicEditor.create(textarea).catch(error => { console.error(error); }); } }); });

Be aware that prompting a user with a native API in an infinite loop like this can cause app store and google play store rejection. You are requesting the user to setup / use a specific form of authentication, you must allow them to refuse that request - and both stores expect your app to remain usable despite the users decision. Some native APIs will just not show a message and fail automatically if you call them too many times. In your code that could cause a stack buffer overflow and your app could crash. But it depends on the API, others will just keep prompting, inwhich case your app will be rejected if it's reviewed thoroughly. Last I checked the max number of prompts allowed for permissions is 2. So your code is not usable either way, and you should not use it.

all scopes can be found here -https://developers.google.com/identity/protocols/oauth2/scopes

for firebase messaging - https://developers.google.com/identity/protocols/oauth2/scopes#fcm

I conditionally replace the Binary file in the Content column to create a column of tables:

= Table.ReplaceValue(Source,each [Content],

each if Text.Contains([Extension],".xl") then Table.TransformColumns(Source, {"Content", each Excel.Workbook(_)})

else if [Extension] = ".txt" then Table.FromColumns({Lines.FromBinary([Content])})

else if [Extension] = ".csv" then Csv.Document([Content],[Delimiter=","])

else [Content],

Replacer.ReplaceValue,{"Content"})

As of QTCreator 16:

Edit->Preferences

Then go to the "Text Editor" section on the left. (Below 'Environment', above 'FakeVim')

Then go to the "Display" tab. There is a block called 'Live Annotations'. Uncheck it to remove the inline warnings.

select document_mimetype, document_data  from resource_t where resource_id = :id

using a media resource did the trick where mimetype and the blob column are requested from the database table.