I have a question. In my case, I need to retrieve the IP addresses of my workloads using nslookup (I have a DNS entry that resolves to all the workload IPs), and then add those IPs to a security group in Terraform to allow access to my RDS database. Do you have any suggestions on how to automate this?

You can try reaching out to the support team of LambdaTest®

Well, to tackle this i created my custom select2 with the checkboxes functionality. I attach my GitHub repository link below. Whenever you want to add this functionality, just add this class, customselect Suppose if you want to add multiple custom checkboxes, select2 don't worry, just add this class customselect, in the select tag

https://github.com/hammadshah01/Custom-Select2-with-Checkbox

I am also new to this fields but maybe I can be of some helps. I don't know if you resolve your problem since it has been a few days.

I am not sure to understand what you mean by

I want to normalise the bounding boxes so that objects of the same size appear similar in dimensions, regardless of their distance from the camera.

you want the object to be at a fix size? or the bounding box is at a fix size regardless of the size the object appear to be (the position of the object to the camera)?

If you just want to modify the size of the bounding box. You could do a depth estimation https://paperswithcode.com/task/depth-estimation and do a cross multiplication.

If you want to modify the size of the object, you could try to extract the object of the picture, resize it and reintegrate it. maybe there is easier way.

hoping I could help.

ChatGPT suggested that from_pretrained might be the cause, so I checked the packages responsible for loading the models.

It's caused by from_pretrained called inside the constructor of M3Embedder from FlagEmbedding package. This method seems to perform a fetch every time, and surprisingly it does use the cache (otherwise it would not just cost a few seconds, but spend much longer downloading the model again), yet it never skips the fetch step.

By setting a local path to the model and explicitly specifying local_files_only=True in the args of from_pretrained I managed to bypass this fetch process eventually.

The yet to be release 1.16.0 will offer parallelisation for many of the minimize methods, e.g. https://scipy.github.io/devdocs/reference/optimize.minimize-lbfgsb.html, where the workers keyword is used to parallelise numerical differentiation.

this is the right picture for FluidToAgent block, sorry the previous one has no completed code

Sorry everyone, checking and checking again I've realized that the mod rewrite of apache wasn't enabled. Once done, everything started!

Currently the query is made up of two linked queries. If I add an intermediate table

tblUser

as part of the Join, the query runs now correctly, and every time.

SELECT qselStat_Batting.cbtUserID_DB, Count(qselStat_Batting.plyMatchID_PC) AS ContribValue, 1 AS StatTypeID
FROM (qselStat_Batting INNER JOIN qselPlayerTeam ON (qselStat_Batting.cmdMatchTypeID_PC = qselPlayerTeam.cmdMatchTypeID_PC) AND (qselStat_Batting.sidTeamID_DB = qselPlayerTeam.plyTeamID_DB)) INNER JOIN tblUser ON (qselPlayerTeam.plyUserID_DB = tblUser.usrUserID_DB) AND (qselStat_Batting.cbtUserID_DB = tblUser.usrUserID_DB)
WHERE (((qselPlayerTeam.plyMatchID_PC)=6178576) AND ((qselStat_Batting.mtcMatchDate)<([qselPlayerTeam].[mtcMatchDate])))
GROUP BY qselStat_Batting.cbtUserID_DB;

Does this mean that the original source queries should not be joined directly?

If Android 11 (API level 30) or above add queries:

<queries>
    <package android:name="com.example.your_app"/>
</queries>

If Android 11 (API level 30) or above add queries:

<queries>
    <package android:name="com.example.your_app"/>
</queries>

I simply opened Task Manager and ended the already running PyCharm process, that's it. This worked for PyCharm Community Edition 2024.3.

I’ve encountered this exact same issue as well on Transsion devices (including Infinix, Tecno, and Itel) running Android 15 (SDK 34).

The crash appears to come from this line:
com.transsion.widget.v1.AIHelper$Impl$Callback$1.onViewAttachedToWindow

This is not part of my app or any libraries I'm using. It looks like it's internal OEM code from Transsion, possibly related to some "AI Helper" feature built into the firmware. The crash occurs during normal view attachment — likely when we show a dialog or add a view to the window.

Key points:

• It's an Android 15-only issue (SDK 34).

• Happens only on Transsion devices (Infinix, Tecno, etc.).

• The method setQueryFromAppProcessEnabled() is a new Android 15 API, and it seems Transsion is using it without null checking, which leads to a crash.

• Since this is part of their OEM customization, we can't fix it from our app directly.

Request:

If you’ve found a fix or workaround, or if you have more info on what exactly AIHelper does on these devices, could you please share it here?

We are a full-service digital marketing company dedicated to helping businesses thrive in the online world. Our team of experts specializes in SEO, social media marketing, PPC advertising, content creation, email campaigns, and website development. We create customized strategies tailored to each client’s goals, ensuring maximum reach, engagement, and return on investment. With a strong focus on creativity, data analysis, and the latest digital trends, we deliver campaigns that not only attract attention but also convert visitors into loyal customers. We believe in transparent communication, continuous improvement, and measurable results. Whether you're looking to build brand awareness, generate leads, or increase sales, we’re here to guide your digital journey every step of the way. Our mission is to empower businesses with the tools and knowledge they need to succeed in a fast-changing digital landscape. Partner with us to elevate your online presence, outshine your competition, and achieve lasting growth.

Ok, I have found a solution. The following approach works for me:

Remove recursively bin and obj folders from the projects:

rm -rf **/*/bin **/*/obj

Make sure that C# and C# Dev Kit VS Code extensions are in release versions. For me it's 2.76.27 and 1.19.63 respectively.

For more information look at the similar GitHub thread.

You can try replacing 'width' with 'min-width'

 .card {
    /* width: 174px; */
    min-width: 174px;
 }

Trying to solve this too! It shows me all oauth2 flows, while only one is supported - users get regularly confused about which one to use!

That sounds like a great idea for a fun and interactive birthday countdown app! 🎉 A few common issues could be around handling date formats, accounting for leap years, or calculating the difference between now and the next birthday correctly (especially if the birthday already passed this year).

When you get it working, it could even be expanded to show привітання з днем народження хлопчика школяра (birthday greetings for a school-age boy) or generate a custom birthday message along with the countdown — that would make it more engaging for younger users!

If you’d like, feel free to share your current code — happy to take a look and help debug it with you!

You might want to put some logging in your event listener to debug this 

Log::info('Event fired', [
    'request_id' => request()->id() ?? uniqid(),
    'path' => request()->path(),
    'server' => gethostname(),
    'memory' => memory_get_usage(true)
]);


additionally might want to check that there are no multiple event registration on your EventServiceProvider

Also can you check that you dont  have multiple instance of the app service plan deployed ? 

    showReferenceStrip: true,
          referenceStripPosition: "BOTTOM_LEFT",
          referenceStripScroll: "vertical",
          referenceStripWidth: 30,

i am using this for reference and i need disable hover exam when i hover reference margin left 0 and when moved over then margib left become -66px

can you pls tell me how to disable hover ?

#include <bits/stdc++.h>
using namespace std;
//Santa


int main() {
    ios::sync_with_stdio(false);
    cin.tie(nullptr);

    int t; cin >> t;  // number of test cases
    while (t--) {
        int n; cin >> n;
        long long a; cin >> a;  // gift allocation can be large
        vector<int> score(n);
        for (int i = 0; i < n; i++) cin >> score[i];

        vector<long long> gifts(n, 1);

        // Left to right pass
        for (int i = 1; i < n; i++) {
            if (score[i] > score[i-1]) {
                gifts[i] = gifts[i-1] + 1;
            }
        }

        // Right to left pass
        for (int i = n - 2; i >= 0; i--) {
            if (score[i] > score[i+1]) {
                gifts[i] = max(gifts[i], gifts[i+1] + 1);
            }
        }

        long long total = accumulate(gifts.begin(), gifts.end(), 0LL);

        if (total <= a) {
            cout << "YES\n";
        } else {
            cout << "NO\n";
        }
    }

    return 0;
}

Here is my approach:

terraform state list # to extract all terraform state list
terraform state show packet_device.worker # don't use quote(')

And one more thing packet_device.worker should be in output of terraform state list

Could you please check Which Database You're using?

tenancy()->initialize($tenant);
dd(DB::connection()->getDatabaseName());

Also, make sure your tenancy package is set up correctly with the right database information for each tenant.

In my case the issue was that I was using the legacy Text game object. Use TextMeshPro instead, it's better and works out of the box.

For now I have reintroduced an older version where the entire solution is .NET6 based and that seems to solve the problem for now.

In that case first we need to check data load in database if that condition true then you can use your print function. I hope this will be useful.

Thanks

#include <bits/stdc++.h>
using namespace std;
//Long ago in the digital realm of Byteonia

int main() {
    ios::sync_with_stdio(false);
    cin.tie(nullptr);

    int N; cin >> N;
    unordered_map<int, string> hashTable;

    for (int i = 0; i < N; i++) {
        string cmd; cin >> cmd;

        if (cmd == "INSERT") {
            int key; string value;
            cin >> key >> value;
            hashTable[key] = value;  // Insert or update
        }
        else if (cmd == "DELETE") {
            int key; cin >> key;
            hashTable.erase(key);    // Delete if exists, else no effect
        }
        else if (cmd == "SEARCH") {
            int key; cin >> key;
            auto it = hashTable.find(key);
            if (it == hashTable.end())
                cout << "NOT_FOUND\n";
            else
                cout << it->second << "\n";
        }
    }

    return 0;
}

In your table (tbl_item ), the rowdata itemid appears more than twice. You can use more precise filtering conditions to avoid these duplicates

i find there is no explicit info.plist today in my project folder

but i solve it

i goto MyAppName - PROJECT - TARGETS - xxx Watch App(my project is a watch project) - All - Info.plist Values - Add the info for the "Privacy - Health .. "

If you are looking for better ways to customise Keycloak. I would suggetst checking out Keycloakify they really have something special

[email protected] foram dev

'dev' is not recognized as an internal or external command, operable program or batch file.

Problem solved.

someone or some apps changed my environment variable: HTTP_PROXY and FTP_PROXY

i think this is the quick approach you can take and show Loading... text of the basis of api response there are multiple way to handle this case its up to you

{user ? (
      <h2 className="mb-4">Hi, {user.firstName}!</h2>
    ) : (
      <h2 className="mb-4">Loading...</h2>
    )}

I cant comment on anybody's answers, as i am new here.
We recently changed Battery on our device and since then the Modem has stopped working. Checking the status in nmcli dev sh, it says that it is unavailable. I tried to start it, also restarting network manager but its not helping and shows its unavailable. how do i solve this?

You can start Elasticsearch with the old data directory and disable security settings. # elasticsearch.yml

path.data: /data/elasticsearch/data

xpack.security.enabled: false

In my case, it is working after updating package name from javax.persistance.Entity to jakarta.persistence.Entity

Then run ./gradlew clean build

Line 12: Char 29: error: cannot use type 'void' as a range
   12 |         for(auto [dy, dx] : {{-1,0}, {1,0}, {0,1}, {0,-1} } ) {
      |                             ^

I got similar problem, what to do here?

(I am using c++23)

how to add git repos to the dependency, ex: https://github.com/checkout/checkout-ios-components

Manage i18next is always headache, last time I saw AutoLocalise which doesnt require these kinds of setup and no need to manage translation files, maybe you can take a look https://www.npmjs.com/package/react-native-autolocalise

You don't really need to generate an id, you can usually do:

<input
 ...
 #inputEl
/>
<label [for]="inputEl">{{ labelTxt }}</label>

{"previousPageCursor":null,"nextPageCursor":"eyJzdGFydEluZGV4IjoxMCwiZGlzY3JpbWluYXRvciI6ImtleXdvcmQ6JHtpZGt9c2Vzc2lvbklkOiIsImNvdW50IjoxMH0KMTUzMWU2M2FiMzFhOGI4MzEzODE4YTM0MzIyMmYyYjJhMDhjZGY3ZGQ4NGVlZjM1MTlhYzViMmQyOTNlYWM0NA==","data":[{"previousUsernames":[],"hasVerifiedBadge":false,"id":7897820696,"name":"Idkidcbutiluvbaddies","displayName":"Idk"},{"previousUsernames":[],"hasVerifiedBadge":false,"id":7596464558,"name":"Idk_121113","displayName":"Idk"},{"previousUsernames":[],"hasVerifiedBadge":false,"id":8575065021,"name":"idk11111111103","displayName":"Idk"},{"previousUsernames":[],"hasVerifiedBadge":false,"id":7561426038,"name":"IDKACC291","displayName":"Idk"},{"previousUsernames":[],"hasVerifiedBadge":false,"id":8587081239,"name":"idk_forfun27","displayName":"idk"},{"previousUsernames":[],"hasVerifiedBadge":false,"id":6161117505,"name":"IDK1516176","displayName":"IDK"},{"previousUsernames":[],"hasVerifiedBadge":false,"id":8313385241,"name":"idk119734","displayName":"idk"},{"previousUsernames":[],"hasVerifiedBadge":false,"id":674737081,"name":"idkhow2dothiz","displayName":"Idk"},{"previousUsernames":[],"hasVerifiedBadge":false,"id":3628195102,"name":"Idk_yuhhhhh0","displayName":"Idk"},{"previousUsernames":[],"hasVerifiedBadge":false,"id":8512322155,"name":"Idk_V0737","displayName":"Idk"}]}

In my case, the application was built using .NET 6.0, but my Visual Studio was set to use .NET 8.0, which caused the error. Later, I switched to IIS Express, which revealed the actual issue. After identifying it, I installed .NET 6.0, and that resolved the problem.

this error was resolved by adding the following mandatory field setting in both actions:

clocation.LocType = defaultLocation.LocType;

clocation.Status = defaultLocation.Status;

Hope this helps, thank you!

how ot open the folder in different folders form the root folder
1) launch the jupyter lab from its root directory
2) File --> New --> Terminal --> type the following command from the terminal shell:enter image description here
Create new Terminal
          jupyter lab --notebook-dir  <your desired directory name>
Type this terminal command: jupyter lab --notebook-dir  <your desired directory name>

3) Enter press
4) wait until it loads
you have to wait untill the URL displayed and correctly configured

5) continue working
Congratulations you are going to work with the following folder structure:

You use the wrong tool - it is exportsqlce you want to use.

https://github.com/ErikEJ/SqlCeToolbox/wiki/Command-line-tools

Thank you for pointing out the root of the issue @Lukasz Szozda and @Frank Heikens

In PostgreSQL, unquoted identifiers are automatically folded to lowercase. So, this command:

CREATE ROLE myDatabase_IntegrationServicesAccount LOGIN WITH PASSWORD 'test@12345';

is actually interpreted as :

CREATE ROLE mydatabase_integrationservicesaccount LOGIN WITH PASSWORD 'test@12345';

However, if you explicitly create the role with double quotes, then PostgreSQL preserves the exact case, making the role name case-sensitive.

DROP USER "myDatabase_IntegrationServicesAccount";

This will correctly identify and drop the role with exact casing.

Don't install python from the Microsoft Store. Microsoft python version doesn't add the py in your path for you. VirtualBox also assumes it can access the py command from your path. Otherwise you'll have to figure out how to do this manually.

1. Spring Framework Version

Spring Boot 1.x relies on Spring Framework 4.x, while Spring Boot 2.0 uses Spring 5.x. This update includes significant improvements, particularly with regards to reactive programming.

2. Reactive Programming Support

Spring Boot 2.0 adds support for Spring WebFlux, allowing reactive, non-blocking web applications to be created using Project Reactor. Spring Boot 1.x does not.

3. Java Version Requirement

Spring Boot 1.x supports Java 7 and later. Conversely, Spring Boot 2.0 supports Java 8 and later and provides Java 9+ features support.

4. Embedded Server Updates

Whereas Spring Boot 1.x has embedded Tomcat 8 as a default, Spring Boot 2.0 has embedded Tomcat 8.5 as default and also updates Jetty and Undertow to newer versions.

5. Actuator Improvements

In Spring Boot 2.0, the Actuator module has been totally revamped. Endpoints such as /health, /metrics, etc., now fall under the base url /actuator/. They are even secured by default and need to be made explicit in configuration. In 1.x, they were enabled and exposed by default.

6. Configuration Improvements

Spring Boot 2.0 has improved support for relaxed binding and enhanced more powerful mechanisms to bind configuration properties with @ConfigurationProperties. It's more flexible and consistent than 1.x.

7. Gradle Plugin

The legacy Gradle plugin was used by Spring Boot 1.x. Spring Boot 2.0 brings an official and enhanced Gradle plugin that follows contemporary Gradle conventions.

8. Security Defaults

In 2.0, Spring Security has enhanced defaults. It provides more configuration options and safe defaults. For instance, endpoints are default-locked unless they are explicitly allowed.

9. Data Access Improvements

Spring Boot 2.0 provides support for Reactive Repositories via Spring Data WebFlux. This enables completely reactive data access. Spring Boot 1.x only allows blocking JPA and traditional Spring Data.

10. Logging Enhancements

Spring Boot 2.0 has improved control and formatting of logs. Although 1.x was also SLF4J + Logback-based, 2.0 has simpler customization and structured logging support.

We went with staging option(internal blob) enabled as part of the copy activity which resolved the DNS issue. Internally, the copy activity first created file in blob container and then uploads files to external azure blob storage.

Thanks all

Comment the following line (because it prevents displaying the list of databases):

# proxy_set_header X-Database main

el porcentaje es un valor que funciona dependiendo al valor de su padre y miro que a la className con id=outer tiene un heigth=100% eso funcionaria si le pones una altura al body pero si el body no tiene altura tiene un valor por default como heigth=auto eso significa que aumenta por su contenido y eso significa que no tiene un valor de altura obligatorio lo que tendrás que hacer es darle un valor obligatorio al padre del hijo que quieres darle una altura en porcentaje si el es heigth=1900px el hijo con un heigth=100% ocupára el 100% de 1900px. con esta logica si en ves de poner porcentajes a los hijos en tu codigo ponle valores obligatorios en ves de 100% => 100px o 1000px lo que quieras. esto significa que si pones los valores en pixeles harás que el body al estar con una altura por default se obligue a estar del grande de su hijo que el cual tiene una latura obligatoria. te recomienda poner border:2px solid black; para mirar como estan maquetados tus div,body,nav,headre,footer etc

On the upper right, change the engine. My solution here worked for MySQL.

SELECT
  MAX(IF(occupation = 'Doctor', name, NULL)) AS Doctor,
  MAX(IF(occupation = 'Professor', name, NULL)) AS Professor,
  MAX(IF(occupation = 'Singer', name, NULL)) AS Singer,
  MAX(IF(occupation = 'Actor', name, NULL)) AS Actor
FROM (
  SELECT 
    name, occupation,
    ROW_NUMBER() OVER (PARTITION BY occupation ORDER BY name) AS rn
  FROM occupations
) AS t
GROUP BY rn
ORDER BY rn;

This issue is resolved after upgrading my local Docker and Docker Compose from 19.03 and 1.21 to 28.2.2 and v2.36.2, respectively. The old versions come with Debian Bullseye as the default. I have to override the default versions.

public static void main(String[] args) {
    String name ;megs
    int age ;25
    double height;5'2

Xcode writes out IDE workspace data constantly on a timer, on the main thread (WHY ON THE MAIN THREAD APPLE!?!), which is what causes this stuttering. Seeing it even on a M4 Pro (as revealed by using Instruments to sample Xcode).

Can you delete the yarn.lock file and run yarn install again?

Proxy the root / path instead of /myapp

• proxy / to. your app on localhost:3000, so all root-relative urls work naturally

• then serve your other apps on different subdomains or different ports

• downsides: the app owns the entire root path

Could someone perhaps also elaborate on the properties? There are zero comments in source, so it is difficult to know what the intended purpose is. Especially in comparison and/or contrast of ReturnedClass, which I take to mean the intended CLR type in question, DefaultValue.

DefaultValue, somewhat obvious albeit without comments, in my case looking at NodaTime.Duration.Zero, right. So if we're talking about NodaTime.Duration, then what is PrimitiveClass in that picture? Could be for instance string? Or System.TimeSpan? Or what else?

Thank you...

abstract System.Type PrimitiveClass { get; }
abstract object DefaultValue { get; }

I am also trying to get this working, either locally or on colab. I have not succeeded yet.

I do not have enough karma to comment, so I post an answer.

It looks like mmpose was developed with PyTorch 1.8. Colab has a different PyTorch version.

Also, mmpose was developed with Python 3.8.

Going to the PyTorch old versions page (https://pytorch.org/get-started/previous-versions/), I noticed there is a version 1.8.2 with LTS support. See below for relevant commands to install the correct versions of all python packages.

Here is the command that I used (Conda on Linux):
conda install python=3.8 pytorch torchvision torchaudio cudatoolkit=11.1 -c pytorch-lts -c nvidia

(Snippet from: https://pytorch.org/get-started/previous-versions/)

v1.8.2 with LTS support

Conda

OSX

macOS is currently not supported for LTS.

Linux and Windows

# CUDA 10.2
# NOTE: PyTorch LTS version 1.8.2 is only supported for Python <= 3.8.
conda install pytorch torchvision torchaudio cudatoolkit=10.2 -c pytorch-lts

# CUDA 11.1 (Linux)
# NOTE: 'nvidia' channel is required for cudatoolkit 11.1 <br> <b>NOTE:</b> Pytorch LTS version 1.8.2 is only supported for Python <= 3.8.
conda install pytorch torchvision torchaudio cudatoolkit=11.1 -c pytorch-lts -c nvidia

# CUDA 11.1 (Windows)
# 'conda-forge' channel is required for cudatoolkit 11.1 <br> <b>NOTE:</b> Pytorch LTS version 1.8.2 is only supported for Python <= 3.8.
conda install pytorch torchvision torchaudio cudatoolkit=11.1 -c pytorch-lts -c conda-forge

# CPU Only
# Pytorch LTS version 1.8.2 is only supported for Python <= 3.8.
conda install pytorch torchvision torchaudio cpuonly -c pytorch-lts

# ROCM5.x

Not supported in LTS.

Wheel

OSX

macOS is currently not supported in LTS.

Linux and Windows

# CUDA 10.2
pip3 install torch==1.8.2 torchvision==0.9.2 torchaudio==0.8.2 --extra-index-url https://download.pytorch.org/whl/lts/1.8/cu102

# CUDA 11.1
pip3 install torch==1.8.2 torchvision==0.9.2 torchaudio==0.8.2 --extra-index-url https://download.pytorch.org/whl/lts/1.8/cu111

# CPU Only
pip3 install torch==1.8.2 torchvision==0.9.2 torchaudio==0.8.2 --extra-index-url https://download.pytorch.org/whl/lts/1.8/cpu

# ROCM5.x

Not supported in LTS.

Use \K before (\d+), like this:

(?<=fill:#00d5ff)[\s\S]*?\K(\d+)(?=<\/tspan>)

\K resets the starting point of the reported match. Any previously consumed characters are no longer included in the final match.

Ah yeah, Git’s being a bit too smart for what you’re trying to do.

Even though you removed the remote, your new repo still shares the exact same commit hash as the template — so when you fetch from it again (even with --shallow-since), Git recognizes that commit and says, “Hey! I know more about this history,” and pulls it in. That’s why your shallow history gets unshallowed.

Git identifies commits by their SHA-1, and since both repos share that SHA, Git treats them as part of the same history graph, no matter what remotes you have set.

So how to fix it?
To stop Git from linking the two histories, you basically need to make the commits look unrelated. Let me show you couple ways.

At first, create a new root (break history).

Use --orphan to create a new branch that doesn’t share any history.

Like this =>

git checkout --orphan new-main

git commit -m "Start fresh"

git cherry-pick main # or cherry-pick a few commits you care about

git branch -M new-main

Now your repo doesn’t share any SHAs with the template — so Git can’t accidentally “help” you.

Next, don’t clone, just copy.

If you're using this as a template anyway, you could just copy the files instead of cloning.

Like this =>

rsync -av --exclude='.git' template/ new-repo/

cd new-repo

git init

git add .

git commit -m "Initial commit from template"

That gives you a clean, standalone repo with no shared history.

Hope this helps someone out there dealing with the same Git weirdness — don't worry, you're not crazy, Git's just very good at remembering things 😄

While I generally agree with @jsbueno 's answer that overloading will require `typing.overload` I'm not sure it is needed for your example.

Sure, regarding the part of your question about "Can I somehow 'overload' decorator?" yes, you can and you would need `typing.overload` (as already mentioned) and you can read more about it in the official PEP-484 (python-enhancement-proposal) specifically the part about overloading: https://peps.python.org/pep-0484/#function-method-overloading

However, decorators are going to behave more like a C++ generic constructor for functions in your example through some syntactic sugar. So, regarding the remaining part of your question: "Can I use some nice solutions in this case?" yes, just make the method static.

The only issue with your example code I see is the class method is not marked as 'static' but implemented as static, yet called on the instance when used.

e.g, more decorators is simple and works in this case

   # double decorate static class method to drop the unused self argument
   @nice
   @staticmethod
   def sumup(a, b):
       return a + b

This way your @nice decorator takes the "static" "member" method (e.g., instead of the C++ covariant-like argument) output by the @staticmethod decorator which takes the static class function as input, you could also move the scope of a, b, to the decorator like so:

def nice(f, *args, **kwargs):
   @wraps(f)
   def decorator(*args, **kwargs):
       result = f(*args, **kwargs)
       return 'result is: %s' % str(result)
   
   return decorator

from my light testing:

from functools import wraps
def nice(f, *args, **kwargs):
    @wraps(f)
    def decorator(*args, **kwargs):
        result = f(*args, **kwargs)
        return 'result is: %s' % str(result)
    
    return decorator


@nice
def sumup(a, b):
    return a+ b


# >>> sumup(9, 8)
# 'result is: 17'
# >>> 

# >>> class Test:
# ...    def __init__(self):
# ...        pass
# ...    
# ...    @nice
# ...    def sumup(self, a, b):
# ...        return a + b
# ...        
# >>> cls = Test()
# ... print(cls.sumup(4, 8))
# ... 
# result is: 12
# >>> 

sorry for the rushed answer, hopefully this helps.

%dw 2.0
output application/json
---
{
  data: payload groupBy $.EMP_ID pluck ((value, key) -> {
    empID: value[0].EMP_ID,
    jobData: value groupBy ($.EMP_ID ++ '|' ++ $.DEPT_ID ++ '|' ++ $.OPRT_UNIT_CODE ++ '|' ++ $.PROD_CODE) pluck ((subValue, subKey) -> {
      DEPT_ID: subValue[0].DEPT_ID, 
      OPRT_UNIT_CODE: subValue[0].OPRT_UNIT_CODE,
      PROD_CODE: subValue[0].PROD_CODE,
      orgData: subValue map ((item, index) -> {
        OrgCode: item.NODE_CODE,
        AllocPct: item.ALLOC_PCT
      })
    })
  })
}

Your code has a major issue: the while thread_request_token.is_alive(): pass line creates a blocking loop that freezes the GUI until the thread finishes. This is not ideal, as it prevents Tkinter from updating the interface properly. Instead, you should use a callback function once the thread completes.

Here’s a better version of your code using threading and Tkinter's after() method, which allows the UI to remain responsive:

import threading

def change_login_to_wait():
    frame_waiting.grid(row=0, column=0, padx=10, pady=10)
    frame_discord_login.grid_forget()

def change_wait_to_session():
    frame_discord_session.grid(row=0, column=0, padx=10, pady=10)
    frame_autoticket.grid(row=0, column=1, padx=10, pady=10)
    frame_waiting.grid_forget()

def request_token():
    """Function that runs in a separate thread to request token."""
    bls_at.request_discord_token(input_email.get(), input_password.get())

def on_login_complete():
    """Called once the login process completes."""
    lbl_discord_username.configure(text=bls_at.USER_NAME)
    change_wait_to_session()

def login_user():
    change_login_to_wait()

    # Start thread for request, then check its completion using after()
    thread_request_token = threading.Thread(target=request_token)
    thread_request_token.start()

    # Schedule a periodic check without freezing the GUI
    check_thread_completion(thread_request_token)

def check_thread_completion(thread):
    """Checks if the thread is alive, then schedules another check."""
    if thread.is_alive():
        frame_waiting.after(100, lambda: check_thread_completion(thread))
    else:
        on_login_complete()

btn_login = CTkButton(
    frame_discord login,
    text="Login",
    command=login user,
    width=LOGIN_ELEMENTS_WIDTH
)

In your react code (if using axios), are you setting parameter withCredentials: true?

import React, { useEffect, useState } from 'react';

export default function Dashboard() {
  const [user, setUser] = useState(null);
  const [loading, setLoading] = useState(true);

  useEffect(() => {
    fetch('/api/user')
      .then((res) => res.json())
      .then((data) => {
        setUser(data);
        setLoading(false);
      });
  }, []);

  return (
    <h2>
      Hi, {loading ? 'loading...' : user.firstName}!
    </h2>
  );
}

You can also use a ternary to conditionally render based on data availability:

{user ? (
  <h2>Hi, {user.firstName}!</h2>
) : (
  <h2>Hi, loading...</h2>
)}

I’ve faced this challenge before when building dashboards that rely on asynchronous API data. Finding a balance between clean code and good user experience is key. From my experience, managing explicit loading states and using the nullish coalescing operator (??) often leads to the most reliable and readable solution.

I hope my answer helps you.

I learned a lot from you. Thank you.

i found one of the solution, if you know more, please let me know -solution is very simple, as greetings is simple list, let make it empty using this line of code :

greeting.current.controls=[]

so that updated fragment  will be this :
    def hello_here(e):
        greeting.current.controls=[]
        greeting.current.controls.append(ft.Text(f'hello {first_name.current.value} {last_name.current.value}'))
        first_name.current.value =""
        last_name.current.value =""
        first_name.current.focus()

i have same problem

00:00:38 CRITICAL  [php] Uncaught Error: Unknown named parameter $value ["exception" => Error { …}]

In AbstractLoader.php line 104:

  Unknown named parameter $value

I think that a native PBI solution is not available. Here is one workaround:

Exchange Manager name and Manager rank. This way Manager rank will be in small multiples and you will be able to sort descending by this rank. It will also display the name of the manager.

to solve the issue, 1 download the newest PowerShell 2. go back android studio terminal, 3 inside the terminal settings go to shell path and locate where you have downloaded the newest PowerShell and select that path way and the issue is solve.

⚠️ This guide is for developers who want to keep their Windows clean — no side hustle with NPM, no unnecessary Node.js setup, and no extra tech they'll barely use or never touch again.

🔥 Part 1: Firebase CLI Setup (No NPM, No Node, No BS)

1. Download the Firebase CLI
   Go to: https://github.com/firebase/firebase-tools/releases
   ⚠️ Avoid firebase-tools-instant-win.exe. That version uses Firepit and will likely break stuff or throw confusing errors like FormatException or SyntaxError: Unexpected end of JSON. Go for firebase-tools-win.exe for an easy life.

2. Rename the file
   Rename your downloaded file to:

   firebase.exe
   

3. Move it to a folder you control
   Example path: C:\Users\<yourusername>\firebase\firebase.exe

4. Copy the path of the folder
   Example: C:\Users\<yourusername>\firebase\

5. Add it to your Windows PATH

   • Open Start Menu → search “environment variables”
   • Click Environment Variables
   • Under System Variables, select Path → click Edit
   • Click New → paste the path from step 4

6. Test if it works
   Open a new terminal window and run:

   firebase --version
   

   If it shows a version number, you're golden.
   If not… maybe scream into the void or talk to your favorite AI assistant.

7. (Optional reset if you've failed before)

   firebase logout
   

8. Now log in properly

   firebase login
   

9. Confirm it's working

   firebase projects:list
   

   You should see your Firebase projects listed here. If not, something's off.

Part 2:🔥 Flutter + Firebase Setup (No NPM Required)

At this point, you can either:

Follow the official instructions shown in your Firebase Console (go to your project → click Add app → select Flutter → follow Step 2),
OR
Just follow these simplified, no-nonsense steps below:

1. Activate FlutterFire CLI

   dart pub global activate flutterfire_cli
   

   (Don’t ask why. Just trust the process.)

2. From your Flutter project root, run:

   flutterfire configure --project=<YOUR_PROJECT_ID>
   

   It’ll:

   • Ask what platforms you’re targeting (Android, iOS, Web, etc.)
   • Auto-generate lib/firebase_options.dart
   • Set everything up inside your Firebase Console

3. Follow what the terminal says
   It’ll walk you through the rest — way better than Google Docs, tbh.

🗯 Final Words

📣 Dear Google,
Please write docs for actual humans, not only for engineers who graduated from Hogwarts.
We just want Firebase in our app without sacrificing our sanity. Thanks.

✅ You're now Firebase-powered.
❌ No NPM bloat.
🧼 System stays clean.
💯 Developers stay sane.

Okay, so your Airflow task is getting stuck on requests.post() when trying to grab a Keycloak token, even though it works fine locally and in a plain venv. This is a classic "it works on my machine" but not in Airflow!

Here's a rundown of what's likely going on and how to fix it, in a more Stack Overflow-friendly way:

It sounds like your Airflow task is hanging when it tries to make the requests.post() call to Keycloak. This usually points to network issues from the Airflow worker, how the request itself is configured, or sometimes even Keycloak being picky.

Here’s what I’d check, step-by-step:

1. The BIG One: Timeouts! Your requests.post(...) call doesn't have a timeout. If the network is a bit slow or Keycloak takes a moment longer to respond from the Airflow worker, your request will just sit there waiting... forever.

   Fix: Always add a timeout.

   Python

   try:
       response = requests.post(
           token_url,
           data=token_data,
           headers=headers,
           timeout=(10, 30),  # (connect_timeout_seconds, read_timeout_seconds)
           verify=False # We'll talk about this next!
       )
       response.raise_for_status() # Good for catching 4xx/5xx errors
       #... rest of your code
   except requests.exceptions.Timeout:
       logger.error("Request to Keycloak timed out!")
       raise
   except requests.exceptions.RequestException as e:
       logger.error(f"Something went wrong with the request: {e}")
       raise
   
   

2. Network Gremlins in the Airflow Worker: Your Airflow worker lives in a different environment than your local machine.

   • Firewalls/Security Groups: Can the worker actually reach your Keycloak server's address and port? Test this from inside the worker environment if you can (e.g., curl from a worker pod if you're on Kubernetes).

   • Proxies: Does your Airflow environment need an HTTP/HTTPS proxy to get to the internet or your Keycloak instance? If so, requests needs to be told about it (either via environment variables HTTP_PROXY/HTTPS_PROXY or the proxies argument in requests.post).

   • DNS: Can the worker resolve the Keycloak hostname? Again, test from the worker.

3. SSL Verification (verify=False): You've got verify=False. While this can bypass some SSL issues, it's a security risk (Man-in-the-Middle attacks). Sometimes, even with verify=False, underlying SSL libraries can behave differently in different environments.

   • Better Fix: Get the CA certificate for your Keycloak instance and tell requests to use it:

     Python

     response = requests.post(..., verify='/path/to/your/ca_bundle.pem')
     
     

     Or set the REQUESTS_CA_BUNDLE environment variable in your worker.

   • Debugging SSL: If you suspect SSL, you can try openssl s_client -connect your-keycloak-host:port from the worker to see handshake details.

4. Keycloak Server Being Strict:

   • Is Keycloak configured to only allow requests from certain IPs? Your worker's IP might not be on the list. Check Keycloak's admin console and server logs.

   • Any weird client policies or rate limiting on the Keycloak side?

5. Python Environment Mismatch: Are the versions of requests, urllib3, or even Python itself the same in your Airflow worker as in your working venv? Subtle differences can cause issues.

   • Fix: Use a requirements.txt to ensure your Airflow environment is built consistently.

6. Airflow Worker Resources: Less likely to cause a silent hang on requests.post itself, but if the worker is starved for CPU or memory, things can get weird or very slow. Check your worker logs for any OOM errors.

7. Super-Detailed Logging (for desperate times): If you're really stuck, you can enable http.client debug logging to see exactly what's happening at the HTTP level. It's very verbose, so use it sparingly.

   Python

   import http.client as http_client
   import logging
   http_client.HTTPConnection.debuglevel = 1
   logging.basicConfig() # You need to initialize logging, otherwise you'll not see debug output.
   logging.getLogger().setLevel(logging.DEBUG)
   requests_log = logging.getLogger("requests.packages.urllib3")
   requests_log.setLevel(logging.DEBUG)
   requests_log.propagate = True
   
   

   Put this at the top of your task file.

Quick Summary of Code Improvements:

Python

import requests
import logging
from airflow.models import Variable
from urllib.parse import urljoin
import urllib3

logger = logging.getLogger("airflow.task") # Use Airflow's task logger

def get_keycloak_token() -> str:
    # Consider if you really need to disable this warning. Best to fix SSL.
    # urllib3.disable_warnings(urllib3.exceptions.InsecureRequestWarning)

    keycloak_base_url = Variable.get("keycloak_base_url")
    #... (get other variables)...
    password = Variable.get("keycloak_password") # Ensure this isn't being double-JSON-encoded

    token_url = urljoin(keycloak_base_url, f"/realms/{Variable.get('keycloak_realm')}/protocol/openid-connect/token")

    token_data = {
        'grant_type': 'password',
        'username': Variable.get("keycloak_username"),
        'password': password,
        'client_id': Variable.get("keycloak_client_id"),
    }
    headers = {"Content-Type": "application/x-www-form-urlencoded"}

    logger.info(f"Attempting to get token from: {token_url}")
    try:
        response = requests.post(
            token_url,
            data=token_data,
            headers=headers,
            timeout=(10, 30),  # Connect, Read timeouts
            # BEST: verify='/path/to/your/ca_bundle.pem'
            # For now, if you must:
            verify=False
        )
        response.raise_for_status() # Will raise an exception for 4XX/5XX errors
        token_json = response.json()
        access_token = token_json.get('access_token')
        if not access_token:
            logger.error("Access token not found in Keycloak response.")
            raise ValueError("Access token not found")
        logger.info("Successfully obtained Keycloak token.")
        return access_token
    except requests.exceptions.Timeout:
        logger.error(f"Timeout connecting to Keycloak at {token_url}")
        raise
    except requests.exceptions.HTTPError as e:
        logger.error(f"HTTP error from Keycloak: {e.response.status_code} - {e.response.text}")
        raise
    except requests.exceptions.ConnectionError as e:
        logger.error(f"Connection error to Keycloak: {e}")
        raise
    except Exception as e:
        logger.error(f"An unexpected error occurred while getting Keycloak token: {e}", exc_info=True)
        raise

# Tip: Consider using Airflow Connections to store your Keycloak credentials
# instead of Variables for better security and management.

Start with the timeout. That's the most common culprit for this kind of hang. Then move on to network and SSL. Good luck!

facing the same problem, i lately post my short solution:

(define (map-nested f L)

    (define (map-nested-aux x)
      (if (list? x)
           (map map-nested-aux x)
           (apply f (list x))))

    (map map-nested-aux L))

example:

(map-nested sin '(0.1 0.2 (0.3) 0.4))
'(0.09983341664682815 0.19866933079506122 (0.29552020666133955) 0.3894183423086505)

Answer / Workaround I Found
EF doesn’t enforce precision for computed columns unless you explicitly cast the result inside the SQL string:

.HasComputedColumnSql("CAST(([Price] * ([DurationInMinutes] / 60.0)) AS DECIMAL(18, 2))")

The .HasPrecision(...) method only affects non-computed columns unless your computed expression already returns the expected precision. EF doesn’t auto-wrap or warn you — so you have to handle that casting yourself.

when you install something from an external url that cannot be versioned or that the content can change, docker cannot cache the command and need to execute it at every build.

In your case, the package is not redownloaded because of the mount, the package is present so no need to be redownloaded but the docker run line is steel executed.

(i'm not english native so i hope it is comprehensible)

try add to your quasar.config

css: [
  'app.scss',
  '~quasar-ui-qcalendar/src/css/calendar-day.scss' // <---ADD
],
//-------
build: {
  webpackTranspile: true,
  webpackTranspileDependencies: [ /quasar-ui-qcalendar[\\/]src/ ]  // <---ADD
}

More information: https://qcalendar.netlify.app/getting-started/installation

Okay, I understand. It looks like you're grappling with a pretty common headache in ggplot2: how to make two dashed lines of different colors clearly visible when they overlap, especially if you want them to have the exact same dash pattern but just... offset a bit, so the dashes don't land right on top of each other. That's a totally reasonable thing to want!

Here's the gist of what's going on and what you can (and can't easily) do, based on how R and ggplot2 handle lines:

The Not-So-Great News: Direct "Dash Phase" Control Isn't Really a Thing in R/ggplot2

Unfortunately, R's graphics engine (which ggplot2 uses under the hood) doesn't have a simple switch or parameter to say, "Hey, for this second blue line, start drawing the dashes just a little bit later along the line than you did for the red one."

• How Linetypes Work: When you define a linetype (like 'dashed' or even a custom pattern like "44" which means "4 units on, 4 units off"), you're defining the repeating pattern itself, but not its starting point or "phase" along the line.

• Device-Dependent "Units": Those "units" in custom linetypes are also a bit tricky. They are generally proportional to the line width (linewidth or lwd), but how a basic unit (like for lwd=1) translates to actual pixels or points can vary depending on where you're looking at the plot (your RStudio plot pane, a PNG file, a PDF, etc.). This makes trying to create a "shifted" pattern by just slightly changing the numbers in the hex string (e.g., line1 = "44", line2 = "1344") very unreliable and not a general fix – it might look okay on your screen but break when you save it.

So, What Are Your Best Bets?

Since a direct "phase" control is off the table within R's standard graphics, here are the most practical ways people tackle this:

1. The "True but Complicated" Fix: SVG Export and stroke-dashoffset

   • If you absolutely need that perfect phase control, the way to get it is to step outside R's direct rendering. The SVG (Scalable Vector Graphics) format does support an attribute called stroke-dashoffset which does exactly what you want.

   • The process would be: create your ggplot2 plot -> export it to SVG (e.g., using the gridSVG package) -> then programmatically edit that SVG file to add the stroke-dashoffset to one of your lines -> then use the modified SVG.

   • This is powerful but definitely more involved than a simple ggplot2 tweak.

2. Making Lines Visually Distinct with ggfx (A Good ggplot2-Friendly Option)

   • The ggfx package lets you add some cool visual effects to ggplot2 layers. While it won't shift the dash phase, it can make one line stand out from the other.

   • A good candidate is with_outer_glow(). You could apply a subtle glow to one of your lines. It doesn't change the dashes themselves but makes the line appear slightly "thicker" or haloed, helping to distinguish it.

   • Conceptual Example:

     R

     # install.packages("ggfx") # if you haven't already
     library(ggplot2)
     library(ggfx)
     
     ggplot(data.frame(x=1:5,y=1:5)) +
       geom_point(aes(x=x,y=y)) +
       geom_abline(slope=1, intercept=1, linetype='dashed', color='red', linewidth=2) +
       with_outer_glow( # Apply glow to the blue line
         geom_abline(slope=1, intercept=1, linetype='dashed', color='blue', linewidth=2),
         colour = "lightblue", # Or even "blue" for a softer edge
         sigma = 1.5,        # Adjust for how much glow
         expand = 1          # Might need to tweak this
       )
     
     

   • This keeps you within the ggplot2 world and can be quite effective.

3. The Standard Workarounds (Which You Mentioned Wanting to Avoid, But Just for Completeness):

   • Transparency (alpha): geom_abline(..., alpha = 0.5) on the top line. Simple, but leads to mixed colors where dashes overlap.

   • Slightly Different Linetypes: Your example of a solid red line under a dashed blue line is one way. Or, using two different dash patterns (e.g., linetype = "dashed" vs. linetype = "longdash" or linetype = "dotdash"). This is often the most straightforward if you can accept a slight difference in the dash appearance.

   • Slight Positional Offset (If Your Data Allows): If it doesn't misrepresent your data, you could add a tiny numerical offset to the intercept of one of the lines so they aren't perfectly on top of each other.

Regarding That "Red Protruding" Antialiasing Thing:

Yeah, that little artifact where one color seems to "bleed" or "protrude" around the edge of another when they're close or overlapping is usually due to antialiasing. Antialiasing smooths out jagged edges, but how different graphics devices (your screen, PNGs, PDFs) do this can vary, and sometimes it results in these subtle visual quirks.

• Tip: Try saving your plot using a high-quality graphics device, like those provided by the ragg package (e.g., ragg::agg_png()) or Cairo-based devices (cairo_pdf(), cairo_png()). These often have better antialiasing and might reduce that effect.

In a Nutshell:

It stinks that there isn't a simple dash_phase argument in ggplot2. For making overlapping dashed lines distinct when you want the same dash pattern:

• ggfx (like with_outer_glow) is probably your best bet for a ggplot2-based solution that adds visual distinction without altering the line path or trying to simulate phase in an unreliable way.

• If you need true phase control, SVG export is the way, but it's a bigger lift.

Hope this helps clear things up, even if it's not the magic bullet solution we all wish R had for this one!

You could start with the flutter documentation. Starting any new language the documentation for that language will have all the important information you need to start your learning.

Learn Flutter documentation

Dart documentation

I see you tagged android, In the documentation it has flutter for android developers that helps you convert your understanding of android to flutter.

You could also find sample projects you could download and go over the code in your own time and look at the documentation at parts you don’t understand that help me quickly understand flutter. You can find small sample projects anywhere for GitHub to YouTube to blogs.

Experiencing the same issue and wondering if you found a solution to this?

Can you tell me what your response is by using this console

console.log(typeof tx.wait); console.log(tx);

I think .wait() might not be valid here due to the nature of the function.

Microsoft - you only had to add additional permission...

I don't know my friend you will get the answer

You can edit the URL handling
Check out the following link:
https://medium.com/@iman.rameshni/django-health-check-89fb6ad39b0c

I wrote a walkthrough for that on medium here;

https://medium.com/@kevostein2k47/how-to-add-natural-voices-in-microsoft-text-to-speech-tts-67d0ee5a9973

To build on Pedro's discovery of support for unnamed sections in Python 3.13, an example:

import configparser

config = "config.ini"

cparser = configparser.ConfigParser(allow_unnamed_section=True)
cparser.read(config)

value = cparser.get(configparser.UNNAMED_SECTION, 'key1')

and while OP has no need to write/modify the original file, for those who do:

cparser.set(configparser.UNNAMED_SECTION, 'key1', 'new_val')

with open(config, 'w') as cf:
    cparser.write(cf)

In my case, I just deleted my Android 34 API emulator and created a new one with 33 API and it worked,

Movie download Answers generated by AI tools are not allowed due to Stack Overflow's artificial intelligence policy

I am seeing the same error. This is a new error because my code worked just a few days ago. I am also using Selenium with options:

        options = webdriver.ChromeOptions()
        options.add_argument('--headless=new')
        options.add_argument('ignore-certificate-errors')
        options.add_experimental_option('excludeSwitches', ['enable-logging'])
        browser = webdriver.Chrome(options=options)

Hopefully someone knows an work around.

that option doesn't disable yaml parsing, what am I doing wrong?

--from markdown-yaml_metadata_block

it is super easy
Check out the following link:
https://medium.com/@iman.rameshni/django-health-check-89fb6ad39b0c

possibly this is a bug in react-native-screens version 4.11.1, you should open an issue at "https://github.com/software-mansion/react-native-screens", for now you can use version 4.10 as a temporary solution. "yarn add [email protected]" or specify in package.json.

There is another part of the documentation called Adding someone else to such chat or channel.

I think that if you have a big team, it’s better to use a professional live translation tool that adapts to your team’s needs and comes with support for your meetings on MS Teams. I recently wrote a blog on MS Teams' capabilities in this area and how Interprefy can help businesses solve the lack of live translation options. It’s in-depth!
https://www.interprefy.com/resources/blog/live-translation-options-for-ms-teams-2025-interpretation-and-ai-captions

The following code worked just fine on Kali Linux for the purpose of preventing sleep. The above answer that involved setting the volume up and down did not work on Kali. I do not know about Windows or MacOS.

import pyautogui, time

while True:
    pyautogui.moveTo(100, 100)
    pyautogui.moveTo(500, 500)
    time.sleep(5)

I think I figured this out:

• "By default" the IDE creates "bare metal" project, which has this features disabled. You can develop you project as bare-metal but you are on your own and need to configure everything in the code, which might be fine for many use cases
• In order to get access to some advanced MCU-specific tools, such as configuring via GUI you have to install MCU-specific SDK (which is called RTD in the most recent incarnation of it)
• I only could install it from file (perhaps there is a way to download it through IDE, but didn;t work for me). In my case the file was `SW32K1_S32M24x_RTD_4.4_3.0.0_QLP02_D2505_DesignStudio_updatesite.zip `
• You need either to specify the SDK during project selection or add it later

Invalid input type <class 'langchain_core.messages.human.HumanMessage'>. Must be a PromptValue, str, or list of BaseMessages.

message = [HumanMessage("what is the capital of India")]

keep HumanMessage in Array.

Good idea. But after a 404, the browser instance automatically closes, and the URL is no longer accessible.

The way I did it (if you have a Windows machine) is to set it up in your ODBC Data Source Administrator and click Add

In the next screen, choose SQL Server and fill in your DSN information then hit next

Then toggle With Windows NT authentication and finish the rest

And you can test the connection from the wizard.

Then you can go into R with the following:

library(RODBC)
con <- odbcConnect(dsn="your_dsn")

Hope that works for you.

What about this?

This allows me to continue using `Enum`, but forces me to check whether the object I'm serializing has a `_to_db` method.

class MyEnum(Enum):
    VALUE = auto()

    def _to_db(self) -> Self:
        return self.value

A TYPO3 version later, but a similar problem with v13.4 - maybe the same cause. I have a dropdown in my navbar to login/logout on every page.
My setup:

lib.loginBox = USER
lib.loginBox {

plugin.felogin.showForgotPassword = 0
 felogin.showForgotPassword = 0
#page und showForgotPassword don't work
# 5 page with regular plugin
plugin.tx_felogin_login.settings.pages = 5

plugin.tx_felogin_login.settings.showLogoutFormAfterLogin = 1 
#outdated? https://docs.typo3.org/c/typo3/cms-core/main/en-us/Changelog/11.5.x/Important-98122-FixFeloginVariableNameInTypoScriptSetup.html
plugin.tx_felogin_login.settings.showForgotPasswordLink = 0
plugin.tx_felogin_login.settings.showForgotPassword = 0

  userFunc = TYPO3\CMS\Extbase\Core\Bootstrap->run
  extensionName = Felogin
  pluginName = Login
  
  settings < plugin.tx_felogin_login.settings 
  
  settings {
   #pages - ##Storage Page with the Website User Records## plugin
   pages = 7
   # does not work
   #    showForgotPassword = 1
   showForgotPassword = 0
   showLogoutFormAfterLogin = 1

  }
  #custom templates
  view {
    templateRootPaths {
       #10 = {$plugin.tx_felogin_login.view.templateRootPath}
       50 = EXT:ubsp/Resources/ext/fe_login/Resources/Private/Templates2/
        }
    partialRootPaths {   
       50 = EXT:ubsp/Resources/ext/fe_login/Resources/Private/Partials2/
        }
    } 
    
} 


  

in my navbar template I use <f:cObject typoscriptObjectPath="lib.loginBox" />
This work fine, but has some disadvantages
Problems I'm struggeling with:

• A user receives two emails when demands forgot password. I couldn't find a way to limt the actions in my lib.loginBox to login, overweiw and logout.

• The info-texts show up only in the dropdown form, not at the regular-plugin.

When I try to use a more simple template in the dropdown, I can't login (refer to https://t3forum.net/d/900-fe-login-in-navbar/23).

Do I have a mistake or is this a bug in felogin?
Thanks in advance for your help