For API 30+
using

enableEdgeToEdge()

Solves most of the issues. Regarding system StatusBar and SystemNavigationBar

Try “Go Easy”, its a vscode extension which works similar to nodemon but for Golang.

And its pretty good, your server logs after reloading stays there in the terminal and have shortcuts to run it too. Simple and Convenient.

The documentation for useState says (emphasis mine):

If the new value you provide is identical to the current state, as determined by an Object.is comparison, React will skip re-rendering the component and its children. This is an optimization. Although in some cases React may still need to call your component before skipping the children, it shouldn’t affect your code.

It seems most likely that in the case of your component, the React is calling it an extra time. Why? That's an implementation detail.

In the new code, do you have empty inside the fetchCurrent or only in return after the semaphore wait? Try to put your ssidResult to some ref box.

Also try to wrap your NEHotspotNetwork.fetchCurrent inside the DispatchQueue.global().async {

Things seem to work as intended by DT v(0.34.0). The original code: datatable(m) %>% formatCurrency("A", "£", digits=1) yields the following (with pound currency, rather than Euro, which is what "\U20AC" will give you)

instead of using an emulator is try using an android device for running and testing your apps.

for this use the file explorer and land to where your flutter project exists.

open the project and copy the path of your project. next open the command prompt and use cd command and your folder location to land there

e.g cd flutter_projects/firstProject

after you reach your project folder from the command prompt

plug in your android device using the charging cable to your laptop/pc, turn on usb debugging and related settings from the developer options. some phones dont have developer options enabled by default so turn that on first and then you run flutter devices command

and see if your phone is visible there it will show the os of it alongside a number assigned to it.

once you see your phone there, you then run the flutter run command

it'll take a good 5-7 minutes the first time and dont let your phone go to standby keep it open.

itl'll install and apk file of your project into your project which you can then open to see your app realtime and interact with it too.

process is long for the first time but do it enough and you'll get the hang of it.

for hot reload and refresh options the command prompt will give you options like Q to quit, r for hot reload etc.

i use this method because usage makes lot of junk files and make good systems lag too.

If you’re looking to improve the text formatting or visual styling on the Google Forms submission page, you might find tools like https://onlinefontsgenerator.com/  useful. It lets you create custom text styles that can be copied into form descriptions or confirmation messages to make them stand out a bit more visually. It’s not a direct Google Forms feature, but it helps add a unique look when you need creative text formatting.

Tried different way , but none of them is working

jffs2reset

This will erase all settings and remove any installed packages. Are you sure? [N/y]

y

MTD partition 'rootfs_data' not found

firstboot

This will erase all settings and remove any installed packages. Are you sure? [N/y]

y

MTD partition 'rootfs_data' not found

Alright, if anyone stumbles upon this problem, I'll leave here what solved it for me !

Thanks to @miken32's suggestion I dug a bit more towards the requests' authentication in etherpad and its cookie needs. The sessionID cookie is necessary to allow a user to open a pad, it's a security feature (enabled in settings.json by setting requireSession to true).

The cookie in my case is set in PHP :

setcookie("sessionID", $session_id, 0, '/p/' . $pad_id,"",true);

It turns out that Etherpad reads the path of the cookie from the domain itself : myapp.eu/p/$pad_id. My Etherpad instance is in a sublevel : myapp.eu/etherpad. So I just need to replace the "path" argument in set cookie with the relevant value (WITH a / at the beginning) :

setcookie("sessionID", $session_id, 0, '/etherpad/p/' . $pad_id,"",true);

Note that samesite and secure must be set to "None" and true respectively, be it through the cookie or settings.json.

Thanks miken32 for your suggestion !

Since Doctrine 3, ClassUtils::getClass() has been removed.

Here's my current replacement.

use Doctrine\Persistence\Proxy;

// ...

    public static function getRealClass(object|string $entityOrClass): string
    {
        $class = is_object($entityOrClass)
            ? $entityOrClass::class
            : $entityOrClass;

        if (is_subclass_of($class, Proxy::class, true)) {
            return get_parent_class($class);
        }

        return $class;
    }

Make sure to have lark and deeplake installed.

%pip install --upgrade --quiet  lark

%pip install --upgrade --quiet  libdeeplake

After five days or so, I've decided the best solution to my own particular problem is to abandon Jupyter Book and run my app with Quarto (https://quarto.org/). I don't know if my problems stemmed from trying to run Jupyter Book in a virtual environment, the transition to Jupyter Book 2.0, or a combination of the two, but in the end the time it was taking me to figure things out was making the project less and less worthwhile. Quarto seems a more mature setup and it's better suited to my needs right now.

Best of luck to the Jupyter Book team in the further development of their product, and many thanks to everyone who took the time to answer my questions.

data:image/webp;base64,UklGRloFAABXRUJQVlA4WAoAAAAQAAAAWQAAWgAAQUxQSKgBAAABkLtt2/E3d8zatt09GGu3q90x/0C2uh0zYaqt3TZW2+b3wxO+eJ67U5eImAAoHbyy/tp7EUm/u1a/ajBY9q7LSMBMXW9z095LxPfTDHW4KtGvdrAxNS0q01P1jc6K2uwYXfmfRfXnfEW7RP0uLQVpMZguUDFbjM5RcELMnoz2Sgy/ivRbTP+Okhbj6QiJmE+CvRGCHwKdEYrngiwVkksDdBCaHfyEqNdpJsc9ugnVrm4ZLhmnJUJ2iYvQddjAZ2N7QridakbVbSWMkjZyhHJOq+2ctrcS0uTWs1oPvGP1HhDa1CbwmlDLq/YSr0sZXhn5j/IJr6eNvBoX8FqQxysPvEBtN6vdQH9W/QGwAoCHnB62Gs9pQitwQpsNjBraAiO0+4DPg/by+OS1h1tsbsOVDZw3cdnsBipZeHZh0sUHW3hshv9nFp8RkgWClnAoC4ORDEYidIW9CoSvtFaFmBNsTUDcLpa6IPofK3+gcb+N/dA5zMIwqD2p7SQ0F3zW9LkAyvv/1vK7Pwx2eqjhYSdYrYlVA9sbvoX6tgEMcxYdTdySg/Ny8G8HVlA4IIwDAADwEwCdASpaAFsAPpVAm0gtKicnLjgNMLASiWwAvCmT/unTdcl5x+WHtXWt+7/hTKf+knJJ6wvE46XnmA/XL9gPey9DH+N9QDpCPQA8tH2Pf3b/bD2czAss0nq5lKwsB7yYLghaAEa+5qZlTQaqx8tG+PeXDYocuAx5BnTwGnAK9AaeLIwlUGy/TyjtD/pQu7H7QeDUS4QQwcmB5CIDyRTPPHe5nCQf5QAA/v6G9j/eQK57ZQs6Aw1pvQ6qPnfAfLQAuI/Ga48h6DPXldWhZpC/JEYDMxD9oe+UUQM6pOhZuxKGC7EtbFN6+fAX8Up4EdbbVNMEAIAmAfrveMW3Ij34v8omOCEdWeNPf/lSHKW3cS4K9MX5wdhYM7xUmSbLPb1vVER1IovDOgCVZa6wsvj5co0KcSHp9pPLrx7dx+RJRfDrGGzMaJFIKRn0mGygNX9imMhdjqrUgxCmKFoOwVHeoHGsDjMgsmzKGTTx6r33rM7t+/60Z0mCM1MsBQgXPkURS4UbuJH0M5M1kncw9FfHQBZH3sN5K+7VKmciLeM9huyqSI4akaLqSYH8qVAuU82l8P/LTwYkogmheF9iLNoYLcQ4RGEWVU78SUAZ+AeHwWFQ4B4eyOPnk9dDq5l6fYeGn99RciiSnvGCGnzJVhoKs088OLSaXlqBnYj+xqYF/7ZTAlLZ4MFcuIBNu25anIcXaKeM//i/1LrIHk5eW8DUZW1IDxGbWJBgZuXXtqQQVHl+HQ1f/nAg1rIs1kgq027+6Edtdrf8hH8nnnU5ZiR1TDcMfSadLVtp9bClV/fnjCillbmBiIRwnQsJoH5VLbn38cHk4nb0M31OgmZjZxg+MVGzbFBZ96iG1io2SyrP/jvAfOXAryOWD+3XfeHKrXgIL+tHAykLtr8mVr1VAM+9SCzstRvHfGCDOKcO+TnYHRBeIuPEM9DFpTEgcCEfOVoSZgMcr1N31sXxO2j91QgbToGl7AiEm

I think you'd be better off using this repository.

https://github.com/abbasi0abolfazl/forexfactory-scraper

If you have EDR/antivirus or other security software, your laptop may be quarantined! Also, the rules for your laptop can been modified !

Have you tried to ping a server or computer connected on the same VPN ? To check, if this protocol works also ?

I'm using Elsa Workflows 3 for a project, and there are four types of triggers (source):

• HTTP Endpoint: triggers the workflow when a given HTTP request is sent to the workflow server.

• Timer: triggers the workflow each given interval based on a TimeSpan expression.

• Cron: triggers the workflow each given interval based on a CRON expression.

• Event: triggers when a given event is received by the workflow server.

I'm trying to figure out how to implement a Cron based trigger in code, but the only documentation I can find (here and here) is for an HTTP Based Trigger.

How do I implement a Cron based trigger in Elsa Workflows 3?

Apparently, I have to implement isAllowEmpty method as below to get the desired behavior:

public function isAllowEmpty(\Phalcon\Filter\Validation $validation, $field): bool
{
    $value = $validation->getValue($field);
    return $this->allowEmpty($field, $value);
}

Now this is my output:

Executing allowEmpty! // from preChecking method in Phalcon\Filter\Validation class.
Executing allowEmpty! // one from my validate method.
Field `foo` is invalid!

I found out that there is a check for isAllowEmpty method existance in Phalcon source code of Phalcon\Filter\Validation class. But isAllowEmpty method could only be found in Phalcon\Filter\Validation\Validator\File\AbstractFile and not in AbsctractValidator.

Later I will try to open the issue at Phalcon's GitHub repo.

Similar to using :empty you can also check if the element has no children. It comes in handy if the element has extra whitespace, but no content which causes :empty evaluate as false:

.container:not(:has(.grid-element--3 > *)) {
  margin-top: -20px;
}

Do I correctly understand that your code looks as follows:

PropertiesConfig.ts

export class PropertiesConfig {

}

index.ts

function f(p: PropertiesConfig){

}

If so, auto imports should operate correctly. Could you please create a new ticket on YouTrack and share your small example and the IDE logs(Help | Collect Logs and Diagnostic Data) there? We will research the issue and follow up with you.

I discovered what the issue was! It was a problem in our dockerfile. We use Liquibase and had frozen our version at 4.28.0, which was fine until recently, because their apt repository stopped supplying it. Because our dockerfile was one huge run statement with a few supporting smaller ones, failing one bit failed the rest, and so the devcontainer did not have docker installed. Upgrading liquibase fixed this issue.

Inspired by @walidtlili's accepted answer (which uses the DockerFile syntax) as an inspiration, here is something I use for GitHub Actions, but my guess is that something very similar can be adapted for other CI/CD workflows.

GitHub has several choices for Linux 'runners' (the virtual machines on Microsoft Azure that run these scripts) and one of them has been the latest LTS version of Ubuntu (at the time of writing, Ubuntu noble 24.04.3 LTS). GitHub already adds lots of packages (the exact list is provided after a successful run), but, of course, you still only get ImageMagick 6.9.

Previously, I avoided compiling everything from scratch. Instead, I used the Debian ('universal') repositories to directly install it. However, recently, I noticed that Debian had released so many upgrades to those .deb packages since I first started to use them, that the versions I had were not only obsolete, but not even available from Debian's repositories (and mirrors!). As such, this method no longer worked.

I made an attempt to bring them up to date, but the problem is that they now have several conflicts with the (also updated) packages installed by Ubuntu: the versions don't match (which is understandable, since, to support ImageMagick 7 — released in 2016!! — you might need to have post-2016 versions of dependancy packages...). You can juggle around a bit with the GitHub runners' configuration (up t a point; a few things are off-limits) and certainly remove mismatched packages, replacing the original Ubuntu packages with shiny new ones from Debian, but... I quickly succumbed to the 'dependency hell'.

The solution was to inspire myself on @walidtlili's answer and do the equivalent for GitHub Actions. Note the slight differences in the packages being fetched; @walidtlili, for some reason, does three separate sudo apt-get update, possibly to guarantee that the packages are being installed in the correct order, but, in my case, it suffices listing all that are needed (considering that a few extra ones will be retrieved anyway — dependencies! — and many, such as compilers, are already pre-installed).

name: 'Compile ImageMagick 7 on Ubuntu 24.04.3'

on:
  push:
    branches: [ "master" ]
  pull_request:
    branches: [ "master" ]

jobs:

  build:
    runs-on: ubuntu-latest
    steps:
    - uses: actions/checkout@v4

    - name: Install packages needed for compilation
      run: |
        sudo apt-get update
        sudo apt-get install -y --quiet wget autoconf pkgconf build-essential curl pkgconf libbz2-dev libfontconfig-dev libfreetype-dev libgs-dev libgvc6 libjpeg-dev libpng-dev libtiff-dev libxml2-dev

    - name: Download ImageMagick source files to /tmp
      run: |
        cd /tmp
        wget https://github.com/ImageMagick/ImageMagick/archive/refs/tags/7.1.2-7.tar.gz
        tar xzf 7.1.2-7.tar.gz
        rm 7.1.2-7.tar.gz

    - name: Configure the ImageMagick build
      run: |
        cd /tmp/ImageMagick-7.1.2-7
        sh ./configure --prefix=/usr/local --with-bzlib --with-fontconfig --with-freetype --with-gslib --with-gvc --with-jpeg --with-png --with-tiff --with-xml --with-gs-font-dir

    - name: Compile ImageMagick
      run: |
        cd /tmp/ImageMagick-7.1.2-7
        make -j

    - name: Install ImageMagick
      run: |
        cd /tmp/ImageMagick-7.1.2-7
        sudo make install
        sudo ldconfig /usr/local/lib/

    - name: Check if ImageMagick was built successfully
      run: $(which identify) -list configure

    # Moving on to the rest of the build...

Also note that I have broken up the full sequence in several steps, and, for each, I have to force cd /tmp/ImageMagick-7.1.2-7 each time¹ (each step runs independently from the others, albeit in sequence). There is no real "need" for having separate steps, but it was more convenient for me to debug. The above lines will also be quite chatty, to show what is being done, but I'd guess you could run them with --quiet flags (or the equivalent) and just ignore all the messages.

As of late 2025, the GitHub 'runners' currently take around 5 minutes to download everything, install the packages, run the ImageMagick autoconf building configuration, and finally compile all the components. I've also done all the checks (with make check) just to be 100% sure everything worked. But, to make things easier, I just list the compilation result by running identify -list configure: if that works, and shows all that I expect it to show, I know it's working correctly! 😉

If you just need to do all the above on your command line, or put it inside a script, then all you need is the following:

sudo apt-get update
sudo apt-get install -y --quiet wget autoconf pkgconf build-essential curl pkgconf libbz2-dev libfontconfig-dev libfreetype-dev libgs-dev libgvc6 libjpeg-dev libpng-dev libtiff-dev libxml2-dev

cd /tmp
wget https://github.com/ImageMagick/ImageMagick/archive/refs/tags/7.1.2-7.tar.gz
tar xzf 7.1.2-7.tar.gz
rm 7.1.2-7.tar.gz

cd /tmp/ImageMagick-7.1.2-7
sh ./configure --prefix=/usr/local --with-bzlib --with-fontconfig --with-freetype --with-gslib --with-gvc --with-jpeg --with-png --with-tiff --with-xml --with-gs-font-dir
make -j
sudo make install
sudo ldconfig /usr/local/lib/
identify -list configure

(the last step is, again, entirely optional)

Note that you don't need to do the actual compilation under the superuser (I'd seriously recommend not to); it's just the package installation that require superuser privileges, as well as the installation step. The ldconfig was added mostly because @walidtlili placed it at the end 😂

The point is helping the dynamic libraries to be found by existing packages, especially if you recompile everything and might have other things that depend on ImageMagick's dynamically loaded libraries.

Under Ubuntu, you can also do a simple

sudo echo "/usr/local/lib/ImageMagick-7.1.2/modules-Q16HDRI/filters" > /etc/ld.so.conf.d/ImageMagick.conf

... which should guarantee that the Linux code interpreter (which runs all compiled ELF binaries) correctly finds the location of the libraries. Just check that the above is the right directory — it was on my case, but YMMV.

This last step cannot be done on GitHub Acions, however, because not even the superuser is allowed to write to /etc on the virtual machine launched by a runner, for obvious security reasons. But sudo ldconfig /usr/local/lib/ works!

A last tip, you can host your own 'runners' locally, even when using GitHub Actions, allowing you full control over the virtual machine being launched. But that is definitely not relevant to the OP's original question!

I hope this can be useful for others as well.

¹ It's not strictly necessary to do it under /tmp. You can build everything locally, i.e., on whatever the virtual machine considers to be your "home". It's not as if everything will be permanently cluttered with litter — the virtual machine, after all, will be completely purged by GitHub Actions once it finishes the run. Technically speaking, even removing the downloaded source code is absolutely unnecessary; again, I just tried to keep everything as close as possible to @walidtlili solution.

For anyone searching for this in the future, PrimeFaces now uses PrimeFaces.current().ajax().update(id) for Server Side dynamic updates.

More Information

solved by changing

signingOption: default

Exposed does not support CTEs at this time. You can follow Exposed issue 868 to be kept up to date on this feature.

As a temporary solution, this adds CTE support to Exposed : https://gist.github.com/SalomonBrys/61fed3a1206d50e9865c3e76f274296b

React Query Builder has had a React Native extension for a couple of years now:

https://www.npmjs.com/package/@react-querybuilder/native

(I maintain React Query Builder.)

I know this is a super old thread, but I wanted to put the solution in that I have been using. I borrowed it from https://github.com/RamblingCookieMonster/PSSQLite/blob/master/PSSQLite/Invoke-SqliteQuery.ps1
this project scrubbed it out using a bit of imbedded c# code that is very efficient and clean:

if ($As -eq 'PSObject') {
    #This code scrubs DBNulls.  Props to Dave Wyatt
    $cSharp = @'
using System;
using System.Data;
using System.Management.Automation;

public class DBNullScrubber
{
    public static PSObject DataRowToPSObject(DataRow row)
    {
        PSObject psObject = new PSObject();
        if (row != null && (row.RowState & DataRowState.Detached) != DataRowState.Detached)
        {
            foreach (DataColumn column in row.Table.Columns)
            {
                Object value = null;
                if (!row.IsNull(column))
                {
                    value = row[column];
                }
                psObject.Properties.Add(new PSNoteProperty(column.ColumnName, value));
            }
        }

        return psObject;
    }
}
'@

    try {
        if ($PSEdition -eq 'Core') {
            # Core doesn't auto-load these assemblies unlike desktop?
            # Not csharp coder, unsure why
            # by fffnite
            $Ref = @( 
                'System.Data.Common'
                'System.Management.Automation'
                'System.ComponentModel.TypeConverter'
            )
        } else {
            $Ref = @(
                'System.Data'
                'System.Xml'
            )
        }
        Add-Type -TypeDefinition $cSharp -ReferencedAssemblies $Ref -ErrorAction stop
    } catch {
        if (-not $_.ToString() -like "*The type name 'DBNullScrubber' already exists*") {
            Write-Warning "Could not load DBNullScrubber.  Defaulting to DataRow output: $_"
            $As = 'Datarow'
        }
    }
}


# to use it to convert the DBNull into $null I use it like this:

try {
    $CSBuilder = New-Object System.Data.Odbc.OdbcConnectionStringBuilder
    $Cred = Get-Credential -Message "Please provide credentials for $Server"
    $OdbcDriverName = 'SQL Server'
    $CSBuilder['driver'] = $OdbcDriverName
    $CSBuilder['DSURL'] = $Server
    #$CSBuilder['Database'] = 'master'
    $CSBuilder['Database'] = $Database
    $CSBuilder['uid'] = $cred.UserName
    $CSBuilder['pwd'] = $cred.GetNetworkCredential().Password
    $CSBuilder['EncryptedPassword'] = 2
    $CSBuilder['ConnectionIdleTimeout'] = 600

    $conn = New-Object System.Data.Odbc.OdbcConnection
    $conn.ConnectionString = $CSBuilder.ConnectionString
    $conn.ConnectionTimeout = 30
    $conn.Open()            
    #$conn.ChangeDatabase($Database)

    $cmd = New-Object System.Data.Odbc.OdbcCommand($QryText, $conn)
    $cmd.CommandTimeout = 30

    $ds = New-Object System.Data.DataSet
    $da = New-Object System.Data.Odbc.OdbcDataAdapter($cmd)
    [void]$da.Fill($ds)

    switch ($As) {
        'DataSet' { return $ds }
        'Table' { return $($ds.Tables[0]) }
        'Row' { return $($ds.Tables[0].Rows[0]) }
        'PSObject' {
            foreach ($row in $ds.Tables[0].Rows) {
                #--- DBNull scrubber conversion ---#
                [DBNullScrubber]::DataRowToPSObject($row)
            }
        }
        'SingleValue' {
            return $ds.Tables[0] | Select-Object -ExpandProperty $ds.Tables[0].Columns[0].ColumnName
        }
        default { return $ds }
    }
} catch {
    Write-Warning "Query failed: $($_.Exception.Message)"
} 

No, you cannot achieve automatic Azure Web App regional failover using only a Private DNS Zone. Private DNS zones are for internal name resolution and do not provide failover or public routing.

For public apps, use Azure DNS Public Zone with manual or scripted updates, or use Traffic Manager / Front Door for automatic failover. https://learn.microsoft.com/en-us/azure/dns/dns-overview

In VS 2022 the default behaviour is that Enter submits the chat. There's no built-in shortcut documented to insert a new line instead while staying in the chat input.

"In the Copilot Chat window, type a coding related question in the Ask Copilot text box. Press Enter or select Send to ask your question." - Microsoft Learn

Try to close your modal inside a finalize() :

onSubmit() {
isError = false;
...
            this.updateSub = this.bewirtungService.updateCatering(cateringCreate)
.pipe(
finalize(()=> {
    if(!isError) {
    this.dialogRef.close(b);
    }
})
)
.subscribe(
                (b: Bewirtung) => {
                    ...
                    // don't close it here
                },
                () => {
                    isError = true;
                    this.notificationService.error('Die Bewirtung konnte nicht geändert werden.');
                }
            );
        }
        else {...}
    }

Very good example for the composition and aggregation.

This answer worked for me: https://stackoverflow.com/a/63022606/415551

sudo apt install xclip

I wonder if this is officially supported as I can not find any clear references to it.

by the way the <Extensions> tag needs to be inside the <Application> tag

Here's another more recent solution in case anyone still wondering - ref: https://moderniser.repo.cont-aid.com/en/How-to-use-the-latest-latest-AWS-icons-in-Mermaid.html)

Example:

flowchart TB

subgraph ACCOUNT[AWS Account]
  subgraph GRP1[" "]
    ELB@{ img: "https://api.iconify.design/logos/aws-elb.svg", label: "ELB", pos: "b", w: 60, h: 60, constraint: "on" }
  end
  subgraph GRP2[" "]
    EC2@{ img: "https://api.iconify.design/logos/aws-ec2.svg", label: "EC2", pos: "b", w: 60, h: 60, constraint: "on" }
  end
  subgraph GRP3[" "]
    RDS@{ img: "https://api.iconify.design/logos/aws-rds.svg", label: "RDS", pos: "b", w: 60, h: 60, constraint: "on" }
  end
  ELB --- EC2 --- RDS
end

classDef vpc fill:none,color:#0a0,stroke:#0a0
class ACCOUNT vpc

classDef group fill:none,stroke:none
class GRP1,GRP2,GRP3 group

When rendered, it looks like this:

AWS Example Output

For Azure you have to install from Debug console !!!!

Nice explanation! Managing external data files reminds me of how https://summerrtimesagamodapk.com/ handles player progress and story saves — separate, yet seamlessly loaded.

feel free to ask this question in our GitHub discussions channel - the Langfuse maintainers are happy to help you there.

https://github.com/orgs/langfuse/discussions

@Stas Simonov's response contains a key finding: that objcopy -O binary doesn't correctly handle multiple sections inside the object file.

However, this is not just on Windows but also on Linux. It happens even for my example but it's a bit hidden.

So, if I do

gcc -c

then the sections are in the order .data, .comment and .note.gnu.property.

If I do

gcc -r

then the sections are in the order .note.gnu.property, .data and .comment.

When .data is first, it's written but then overriden by .note.gnu.property.

When .note.gnu.property is first, it's partially overriden by data because .data is smaller. That's why I see 0102030405 when I use -r.

A possible solution to this issue is to use the -j flag so we can select the .data section i.e.

objcopy -j .data -O binary ...

This way, just the .data section is copied to the binary file.

I'm honestly not sure if this is a bug, a limitation or simply a counterintuitive intended behaviour of objcopy.

For me the issue was resolved by going into build phase under "{name} Extension (macOS)" there is a "Copy Bundle Resource", it contained all other files except "content.js" and "background.js" once I added them, the error went away.

enter image description here

In a sync context, the user context is now available in tools as of Spring AI 1.1.0-M1

https://github.com/spring-projects/spring-ai/releases/tag/v1.1.0-M1

specifically this commit:

• Configured McpSyncServer to use immediate execution when running in a servlet context 568ca96

If anyone still has this issue, I found an easy answer that at least solved my problem. The temp directory was full, with 65,536 files. I cleaned it out and the migration worked fine after that

You also need the share folder that contains themes and icons

When you run npm ls:

[email protected]
+-- [email protected]
`-- [email protected] -> ../common-components
    `-- [email protected]

• [email protected] is installed in the root of the example app — correct, satisfies the peer dependency.

• The extra [email protected] under common-components is not actually installed again in a separate copy; it’s just how npm shows the peer dependency link (even though it's using the root version).

In other words: npm ls reports it under both packages, but in reality there’s only one copy used.

If you want to verify that only one copy of my-api is being used at runtime

1. Use require.resolve(CommonJS) or import.meta.url(ESM)

Since your project is ESM ("type": "module"), you can do:

// In App.tsx or any example file
import * as MyApi from 'my-api';

console.log('my-api path:', import.meta.resolve ? await import.meta.resolve('my-api') : MyApi);

• This will show you the absolute path where my-api is being imported from.

• If both common-components and common-components-example resolve to the same path, there’s only one copy in use.

2. Compare references at runtime

A more React/JS way:

import * as MyApi from 'my-api';
import { something } from '../common-components/src/SomeComponent';

console.log('Same my-api instance?', MyApi === something.__myApiInstance);

• If your library common-components exposes a reference to my-api internally (or you temporarily attach it to window), you can compare the objects.

• If they are strictly equal (===), then both the library and your app are using the same copy.

3. Quick hack with node_modules paths

Run this in your example app:

node -p "require.resolve('my-api')"
node -p "require.resolve('../common-components/node_modules/my-api')"

• If the second one fails (can’t find), it confirms that common-components is not installing a separate copy in its own node_modules.

Actually we can see wrapper.jar version using:

java -classpath /path/to/jar/gradle-wrapper.jar org.gradle.wrapper.GradleWrapperMain --version

At least works with 7.1 that was unknown in my case.

I found it renaming gradle directory, so gradlew --version show error with path of main wrapper class (not found). So I think this will work with other versions.

With JavaScript :

/^(.)\1+$/i.test(value)

Works for upper and lower case mix, for a string of at least two characters :

/^(.)\1+$/i.test('a')
false
/^(.)\1+$/i.test('aa')
true
/^(.)\1+$/i.test('AA')
true
/^(.)\1+$/i.test('Aa')
true
/^(.)\1+$/i.test('Abc')
false

Changing the text of a ttk.Labelframe causes it to redraw completely, which makes the window flash. To avoid that, keep the Labelframe title static and show the “Message X of Y” info in a separate Label inside the frame instead. This removes the flicker.

For me simplest code was:

val isTestRun = Thread.currentThread().stackTrace.any { it.className.contains("androidx.test.runner") }

As a workaround, switching to a QListWidget (which has it's own model) works just fine :

# getting actual order
def qlistwidget_iter_items(lst: QListWidget, role=Qt.DisplayRole):
    for i in range(lst.count()):
        list_item = lst.item(i)
        item = list_item.data(role)
        yield item

adding item adapted :

pix = self.get_image(item.filename)
list_item = QListWidgetItem(QIcon(pix), "")
list_item.setData(Qt.ItemDataRole.DecorationRole, pix)  # image
list_item.setData(Qt.ItemDataRole.UserRole, item)  #
list_item.setSizeHint(thumbnail_size) #  self.ui.lstViewAddedItems.gridSize())
flags = list_item.flags() | Qt.ItemFlag.ItemIsDragEnabled
flags &= ~Qt.ItemFlag.ItemIsDropEnabled
list_item.setFlags(flags)

how do we remake your program we receive sytax error on line Dim swb As Workbook: Set swb = Set swb = Workbooks.Open( _ Filename:=SRC_FILE_PATH, UpdateLinks:=True, ReadOnly:=True)

No this is not supported by IBM, but a way used by a lot of people in Legacy (11.7) Datastage, though most people prefer to do this in an DSX instead of XML Export. Also this is normally done to search & replace code already there, while adding a whole new stage is way more complex. For adding a new stage, datastage is very copy&paste friendly in the UI.

You need to use an older version of react-native-maps that support eh old architecture.

The version 1.20.1 will make the markers appear again.

You can try this one
https://www.jsdelivr.com/package/npm/@use-pico/graphql-codegen-zod

it worked straight!
I needed to change only the plugins: on codegen.yml

How do I use this adb shell pm revoke com.android.systemui android.permission.SYSTEM_ALERT_WINDOW because it doesn't work

This is a verified bug that hasn't been fixed

https://bugs.mysql.com/bug.php?id=108582

quoting here:

[23 Sep 2022 12:43] MySQL Verification Team

Hi Mr. Power Gamer, It turns out that you are correct. You can no longer use ANSI option for mysqldump, due to this bug. This is due to the reason that ANSI option enforces ONLY_FULL_GROUP_BY. We do not know whether that particular query will be changed in mysqldump, or that ANSI mode will be disabled. This, however, has nothing to do with the fact that you are correct regarding the ANSI option. This report is now a verified bug.

This package is for use with API KEY. That means you are accessing Firestore unauthenticated and security rules are applied. So you can access only "public" collections.

enter image description hereverde_transparente.pngverde_transparente.pngverde_transparente.pngverde_transparente.pngverde_transparente.pngverde_transparente.pngverde_transparente.pngverde_transparente.pngverde_transparente.pngverde_transparente.pngverde_transparente.pngverde_transparente.pngverde_transparente.pngverde_transparente.pngverde_transparente.pngverde_transparente.pngverde_transparente.png

focus-visible:ring-0 this fixed it for me in the Trigger Component

It works when you specify plot.background in the theme inside patchworkGrob():

grid.arrange(
  patchworkGrob(pA + 
                  plot_annotation(tag_levels = list(c('A')) )
  ),
  patchworkGrob( plot_spacer() + pD +
                   plot_layout(widths=c(-0.35, 1)) + 
                   plot_annotation(tag_levels = list(c('D'))) & 
                   theme(plot.background = element_rect(fill = "transparent", color="#808080"))
  ),
  nrow=1
)

My package versions:
ggplot2_3.5.2
gridExtra_2.3
patchwork_1.3.0
ggpubr_0.6.0

I believe you can easily achieve your goal using the Power Dropdown Excel add-in. You can find it inside Excel, using the Add-ins button or get it from Appsource. You can try it completely free of charge for up to 3 months.

The user is seeing an "Access blocked" error when trying to sign in to Garena Free Fire with their Google account. The error message, "Error 403: disallowed_useragent," indicates that the app's sign-in request does not comply with Google's "Use secure browsers" policy.

This issue often occurs when an application uses an embedded browser (webview) to handle the Google sign-in process, which is no longer permitted by Google's security policies. Google requires that authorization requests be made from a full-featured, secure web browser.

Here are the suggested actions to address this issue:

• Contact the Developer: The error message itself suggests contacting the developer, Garena, to inform them that their app needs to be updated to comply with Google's policies.

• Use a Web Browser: If Garena Free Fire has a website, the user can try signing in from there using a secure web browser.

• Update and Clear Data: In some cases, clearing the cache and data for the Google Play Store and Google Play Services, or uninstalling and reinstalling the game, can help resolve login issues.

• Check Browser Settings: Ensure that a secure browser like Chrome is set as the default browser on the device.

The issue appears to be on the app's side, as it is using an outdated method for Google sign-in. Developers are required to use secure methods, such as native OAuth libraries or Google Sign-in for their platforms, rather than embedded browsing environments.

Ok there may be a small confusion about how env_file: and --env-file work so let's try to explain this in few steps

services:
  db:
    image: mysql:8.0
    restart: always
    env_file:
      - .env.local
    ports:
      - 5432:5432

This will in fact mean, that you have following compose file

services:
  db:
    image: mysql:8.0
    restart: always
    environment:
        - DATABASE_USERNAME=dummy
        - DATABASE_PASSWORD=secret
        - DATABASE_NAME=demo
        - MYSQL_DATABASE=demo
        - MYSQL_USER=dummy
        - MYSQL_PASSWORD=secret
        - MYSQL_ROOT_PASSWORD=supersecret
    ports:
      - 5432:5432

Which you may reiterate to

services:
  db:
    image: mysql:8.0
    restart: always
    environment:
        - DATABASE_USERNAME=${USER_NAME}
        - DATABASE_PASSWORD=${SECRET}
        - DATABASE_NAME=${DB_NAME}
        - MYSQL_DATABASE=${DB_NAME}
        - MYSQL_USER=${USER_NAME}
        - MYSQL_PASSWORD=${SECRET}
        - MYSQL_ROOT_PASSWORD=${ROOT_PASS}
    ports:
      - 5432:5432

Prepare my_envs file like:

USER_NAME = dummy
SECRET = secret
DB_NAME = demo
ROOT_PASS = supersecret

And the use the podman-compose --env-file my_envs -f compose.yaml config which will "substitute" your variables inside the compose file.

podman-compose version: 1.0.6
['podman', '--version', '']
using podman version: 4.3.1
services:
  db:
    environment:
      DATABASE_NAME: demo
      DATABASE_PASSWORD: secret
      DATABASE_USERNAME: dummy
      MYSQL_DATABASE: demo
      MYSQL_PASSWORD: secret
      MYSQL_ROOT_PASSWORD: supersecret
      MYSQL_USER: dummy
    image: mysql:8.0
    ports:
    - 5432:5432
    restart: always

Maybe this is what you are looking for?

We can insert tables and charts as images using both the Office.js Common API and the PowerPoint API. However, the PowerPoint API offers more advantages since the inserted image can have all the properties of a shape.

What we need is simply a Base64 string. Using the example from https://learn.microsoft.com/en-us/office/dev/add-ins/powerpoint/bind-shapes-in-presentation, we can fill any shape with the image we want to insert. Additionally, we can assign a name and ID to the shape, which is helpful for tracking the image later.

I resolved the issue on my end by switching from release to pre-release version of extensions C# and C# Dev-kit.

While not really answering the question on why the shader misbehaves, I did find a solution for my problem.

It comes down to me misunderstanding how stencil tests work. It turns out you can control stencil buffer writes by discarding the fragment in the fragment shader. This lets me have one subpass that writes to the stencil buffer in appropriate places and another one after it that uses to discard unneeded fragments.

I implement a cursor-like inline diff editor with accept/reject undo/redo
https://github.com/Dimitri-WEI-Lingfeng/monaco-inline-diff-editor-with-accept-reject-undo

pyenv install 3

will install the latest stable (non-alpha/beta/dev) version of Python 3 (so Python 3.x.y).

Chrome is not published as single monolithic APK file. Instead it is published a split APK, thus if you have one APK this isn't enough. If you install it the installed chrome version is incomplete and thus crashes.

Where did you got the Chrome APK from? If you download it from a third party site like ppkpure.com you get an xapk apkm or apks file which requires a special installer or unpack the file and install it using adb install-multiple <apk files>.

Before installing a new chrome app version clearing the chrome app data/cache is recommended.

that may because of vcpkg and msys2(MinGW) confliction. for my case, i uninstalled msys2 and use msvc + cmake and it is worked.

I had this exact issue and after trying to error_log(...) everything related to flush_rewrite_rules(...) and ruling out all the plugins and Cron jobs within Wordpress, I started looking outside of the Wordpress admin interface.

I found in the web hosting control panel (CPanel) there was a section called WordPress Management > WP Toolkit that was connected to the Wordpress installation. I found in these logs that this was performing daily maintenance at the exact time the permalinks were being broken.

By detaching the Wordpress installation from the WP Toolkit, the problem was resolved.

Yes, it is possible to embed Reddit posts programmatically in a web page. You can use Reddit's embed feature or third-party tools. Here’s how:

1. Find the Reddit Post: Go to the Reddit post you want to embed.

2. Get the Embed Code: Click on the "Share" button below the post, then select "Embed." This will provide you with an HTML code snippet.

3. Use the Embed Code: Copy the HTML snippet and insert it into your web page where you want the post to appear. Here’s a basic example of how the embed code looks:

enter image description here

4. Third-Party Tools: You can also use third-party tools like Embedly, Tagembed, IFTTT, or Taggbox to automate the embedding of Reddit posts. These tools allow you to integrate and customize embeds without extensive coding easily.

5. Adjust as Needed: You can modify the size and other attributes according to your design requirements.

Using either the built-in Reddit embed feature or third-party tools, you can programmatically include Reddit posts in your web pages with ease.

This is the official JetBrains documentation on sbt support in IntelliJ IDEA: https://www.jetbrains.com/help/idea/sbt-support.html

It provides a detailed guide on how to set up and manage Scala projects using sbt in IntelliJ IDEA — including how to import sbt builds, configure auto-import, delegate build actions to sbt, and run or debug your Scala code directly from the IDE.

tasks.dokkaHtml { dokkaSourceSets { named("main") { footerMessage.set("© 2025 MyProject — Subscribe on YouTube 🎬") } } } Youtube-plus-plus

//My welcome.ts file is:

import { Component, OnInit } from '@angular/core';

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

import { Router } from '@angular/router';

import {

SocialAuthService,

GoogleLoginProvider,

SocialUser,

} from '@abacritt/angularx-social-login';

@Component({

selector: 'app-welcome',

standalone: true,

// CRUCIAL: Solo CommonModule. Esto asegura que no haya doble inyección.

imports: [CommonModule],

templateUrl: './welcome.html',

styleUrls: ['./welcome.css']

})

export class WelcomeComponent implements OnInit {

user: SocialUser | null = null;

loggedIn: boolean = false;

message: string = 'Inicia sesión para continuar.';

constructor(

private authService: SocialAuthService, 

private router: Router 

) { }

ngOnInit(): void {

// Escucha los cambios de estado de autenticación

this.authService.authState.subscribe((user) =\> {

  this.user = user;

  this.loggedIn = (user != null);

  

  if (this.loggedIn && user.idToken) {

    // Uso de backticks (comillas inversas) para la plantilla literal

    this.message = '¡Hola, ${this.user?.firstName}! Validando credenciales...';

    

    // Aquí enviaremos el token al Backend

    this.sendTokenToBackend(user.idToken); 



  } else if (this.loggedIn) {

     // Si está logueado pero no hay idToken (raro en Google), forzamos navegación

     this.router.navigate(\['/dashboard'\]);

  } else {

    this.message = 'Inicia sesión para continuar.';

  }

});

}

/**

* Inicia el flujo de autenticación de Google.

*/

loginWithGoogle(): void {

this.authService.signIn(GoogleLoginProvider.PROVIDER_ID)

  .catch(err =\> {

    console.error('Error al intentar iniciar sesión:', err);

    this.message = 'Error de inicio de sesión. Inténtalo de nuevo.';

  });

}

/**

* Simulación del envío del ID Token de Google al Backend para su validación.

* @param idToken El ID Token de Google.

*/

sendTokenToBackend(idToken: string): void {

  console.log("Token de Google recibido. Llamando al Backend para validación...");

  

  // Por ahora, simulamos la respuesta exitosa y navegamos.

  setTimeout(() =\> {

      this.router.navigate(\['/dashboard'\]); 

  }, 1500); 

}

}

And my app.config.ts file is:

import { ApplicationConfig, provideBrowserGlobalErrorListeners } from '@angular/core';
import { provideRouter } from '@angular/router';
import { routes } from './app.routes';

// Importaciones necesarias para el login social
import { 
  SocialAuthServiceConfig, 
  GoogleLoginProvider,
  SocialAuthService, // Se necesita el servicio para poder inyectarlo
} from '@abacritt/angularx-social-login';

// ID de Cliente de Google para la aplicación frontend
const GOOGLE_CLIENT_ID = '1000000000000-example.apps.googleusrcont.com';

// 1. Objeto de configuración de autenticación.
const authConfig: SocialAuthServiceConfig = {
    autoLogin: false, 
    providers: [
      {
        id: GoogleLoginProvider.PROVIDER_ID,
        provider: new GoogleLoginProvider(GOOGLE_CLIENT_ID, {
            oneTapEnabled: false, 
            scopes: 'email profile',
            prompt: 'select_account' 
        }),
      },
    ],
    onError: (err: any) => {
      console.error('Error del Social Login:', err);
    },
};

export const appConfig: ApplicationConfig = {
  providers: [
    // 1. Inyección del Router
    provideRouter(routes),
    
    // 2. Otros Providers
    provideBrowserGlobalErrorListeners(),
    
    // 3. REGISTRO DEL SERVICIO Y SU CONFIGURACIÓN
    // Proveemos el servicio principal.
    SocialAuthService,
    // CRUCIAL: Este es el proveedor que no encuentra. Lo inyectamos usando el string token.
    {
      provide: 'SocialAuthServiceConfig', // <-- ESTA ES LA CLAVE
      useValue: authConfig,
    }
  ]
};

Just add
dependency_overrides:
analyzer_plugin: ^0.13.1

y in x = y must be a free variable as it is quantified in B : (y : A) → x ≡ y → Set b for this reasin it can not be specialize to x when doing an induction. So if you want to prove sth for equality, you must first prove sth for x = y to deduce it for x = x, which is not possible for UIP / axiom K.

I have the same problem. Once I enable BiDi, modal dialogs don't appear anymore.

    ChromeOptions options = new ChromeOptions();
    //options.enableBiDi(); // uncomment this line to reproduce the error
    ChromeDriver driver = new ChromeDriver(options);
    driver.get("https://the-internet.herokuapp.com/javascript_alerts");
    driver.findElement(By.cssSelector("[onclick='jsAlert()']")).click();

    // throws "no such alert" if BiDi is enabled:
    Alert alert = driver.switchTo().alert();
    assertThat(alert.getText()).isEqualTo("I am a JS Alert");

if you are here in in 2025 the answer above still works but no need for adding the "serviceAccount:" test before the pricipals as shown in the image here .

Git has a built-in option to the commit command to add Trailer Messages with --trailer so you could use it to add Co-authored-by trailer messages to your last commit with git commit --amend --trailer="Co-authored-by: name <email>".

You can also see commits grouped by trailer message using the shortlog command with something like git shortlog --group=trailer:co-authored-by.

Try these steps to get rid of the problem:

1. in Settings (Ctrl+Alt+S) -> type "Python Debugger" in the search field -> enable "Gevent Compatible".

2. in Actions (Ctrl+Shift+A) -> type "Registry..." -> disable "python.debug.low.impact.monitoring.api".
   And retry debugging.

It helped me. Debug is working again now.

it work after add line: cmake.dir=android/sdk/cmake/4.1.2 in local.properties. before i used cmake 3.22.1

Although not related to IdeaVim plugin nor uncommenting lines, for `IntelliJ IDEA 2025.1.4.1 (Ultimate Edition)`, there is a bell sound on the terminal. It can be disabled through `Settings -> Tools -> Terminal` and un-toggle the option `Audible bell`

I think your slice might be off by one bit. 'high is an index, not a width, so 'high - reg_sum_low'high may give a wrong range.
Try using reg_sum_low'length - 1 to make sure the slice has the correct width — that usually fixes the sign-extension issue you described.

You can use [UpdateModelPackage](https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_UpdateModelPackage.html) API to update the metadata of existing model package without re-creating it, and it doesn't cause version increment.

try adding noopener

<a href="https://www.example.com/" target="_blank" rel="noopener noreferrer" >

and will open in new tab

Regards

After further investigation, I found that the problem was related to how the number of ranges in the LiDAR scan was being calculated. The formula used for calculating the number of ranges is:

"num_ranges = int((scan.angle_max - scan.angle_min) / scan.angle_increment)"

This formula calculates the number of ranges based on the angle range and angle increment, but because it uses integer division, it rounds down the result. The issue arises because the scan.angle_max and scan.angle_min values in the configuration file were set with limited precision (3.14), leading to a mismatch in the expected number of values.

So, the solution was to adjust the precision of the angle values in the configuration file. By increasing the number of decimal places for the angle_max and angle_min values (e.g., changing 3.14 to 3.1416), the calculation of num_ranges became more accurate and the number of ranges matched the expected value.

This small change allowed the merged scan to be processed correctly and the map was generated without issues.

The Canvas itself doesn't support server-less deployment, but you can export the notebook to train the model and use it for server-less deployment.
For more details, you can visit this AWS official document : https://docs.aws.amazon.com/sagemaker/latest/dg/canvas-notebook.html

We could use isPresented environment value to acheive this behaviour.

You can use isPresented environemnt value to acheive the same.

The underlying error is :

java.lang.NoClassDefFoundError: org/hibernate/hql/internal/QueryExecutionRequestException

I'm trying to implement the function with WebView within the IOS app, is there a way? As far as I know, webView.setWebChromeClient() does not support it.

Your assessment is correct. Lifecycle script runs everytime you create space. You can create your own container image to bake heavy initialization step.
You can visit AWS official document [How to bring your own image](https://docs.aws.amazon.com/sagemaker/latest/dg/studio-updated-byoi-how-to.html) how to achieve it.

I came up with this one-liner that works for me:

comm -13 \
  <(git ls-remote --tags origin | sed 's/.*refs\/tags\///g' | grep -v '\^{}$' | sort | uniq) \
  <(git tag | sort)

P.W.DM powdered by Tochukwu P.W.DM is knowledge to inputs it's spider it database dive(production Wed Development Members P.W.DM (crown it's spider friendly free first fine activeTochukwu John Ani MD founder

This error occurs because the version of JavaScript used to configure TypeScript doesn't recognize replaceAll(). This method was added in ES2021 . To resolve this, update the lib array in tsconfig.json by adding the following:

{
  ...,
  "compilerOptions": {
    "lib": [..., "ES2021"]
  },
  ...
}

I happen to come across the same problem and got it rectified by declaring type as date

var source=

{

name:’dob’ , type:’date’

};

$(“divid”).jqxgrid({

Columns:[ { text:’Date of Birth’, datafield:’dob’,cellsformat:‘dd-MM-YYYY’}]});

FWIW Just experienced this, and the answers here helped.

In my case, I had installed the STM32 Cube IDE extension, which doesnt play nicely with the microsoft Intellilsense, and asked if I wanted to disable it. I said yes, but didnt realise it disabled it as USER and not workspace etc - so it broke all the other projects.

Live and learn I guess :-)

Seems to be a bug in IntelliJ Maven Runner.

Disable maven.use.scripts helps but than the maven runner is used the legacy way.

Raised ticket:
https://youtrack.jetbrains.com/issue/IDEA-381163/Maven-Runner-did-not-parse-arguments-correct?draftId=25-7028245

When I first started learning programming, the phrase “Java is platform independent” kept popping up everywhere — but honestly, it sounded like a fancy buzzword. I used to wonder, how can one program magically run on every operating system?

It finally made sense when I understood what happens behind the scenes. Imagine you write your Java code — that’s your story, your idea. Instead of turning it into something that only one computer can read, Java turns it into bytecode, a universal language understood by the Java Virtual Machine (JVM).

Now, here’s where the magic happens: every operating system — Windows, macOS, Linux — has its own JVM. When you run your Java program, the JVM on that system reads your bytecode and translates it into the language that the specific computer understands. That’s how the same program works almost anywhere, without rewriting a single line.

Of course, there are exceptions — if your code uses system-specific features, that independence can break a little. But overall, it’s pretty close to the dream of “write once, run anywhere.”

You shouldn't use the test set since its no longer unseen anymore, it is supposed to only be used once at the end to see the true test of how well it generalizes.

Instead validation sets are used to tune k, which will help the model improve, however afterwards you only use the test set once at the end. Hope this helped!

from within neovim do

:echo $MYVIMRC

to display the full path to the currently used init.vim file