How did you solve this error ??

This was a really silly question until last month when Distraction Control was added to Safari, and now I have not one but two clients with monetized content asking me for a method to block visitors from accessing their content if they are using Safari. Sigh.

For PHP later than 7.4? or so, try Foreign Function Interface, for all DLLs https://www.php.net/manual/en/book.ffi.php

As an excel/libreoffice formula:

# cell A1
1295576502
=FLOOR(LEN(A1)*LOG(LEN(A1)-1,2))
# result is:
31

check: 2^30=1073741824; 2^31 = 2147483648; ok.

Try changing the target framework ordering so that .net 4x is first:

<TargetFrameworks>net472;netstandard2.1</TargetFrameworks>

Also for projects that reference multi-targeted project, try referencing like this. (From: https://stackoverflow.com/a/53388855)

<ProjectReference Include="..\yourMultiTargetedProject.csproj" >
  <SetTargetFramework>TargetFramework=netstandard2.1</SetTargetFramework>

I tried a similar approach but am getting simulation errors. Can you let me know what do your get_lib_names.tcl and Makefile look like?

Poll API was added in late 2024. Check it out here: https://developers.google.com/youtube/v3/live/docs/liveChatMessages

Solved it using: sudo apt-get install libcgi-pm-perl Needed to install the library because apparently new perl versions doesn't have it from the start as they did before. But it is very strange, that there were nothing on the error logs.

I am facing an issue with cusstom model. I trained my invoice with custom model but in line itels my cells get merged two cells in a clolumn. It shown fine when I traned it using prebuilt_invoice but after training when I anayzy the same invoice it merged two cells

for validation using a cat file you can use sigcheck from sysinternals:

sigcheck.exe -f X1Search.cat -e

https://learn.microsoft.com/en-us/windows-hardware/drivers/install/verifying-the-signature-of-a-test-signed-catalog-file

There is a lot more elegant way of finding the scrollbar width. Let me show you how to do it:

1. Maximize your browser window
2. In your browser console enter the following: window.visualViewport.width //So this will equal your screen size in pixels minus the vertical scrollbar width, which in my case is 1903px on Google Chrome with the default styles of the scrollbar. You may have a styled scrollbar with a different width. Check it out!

enter image description here

The only one way to do this - is to use --user-data-dir switch and specify unique path to the user profile each time when you need separated browser process. This is for all chromium compatible browsers.

This is due to Android studio Lady bug 2024.2 using JDK 21 by default instead of 17, there are two solutions:

1. Downgrade to Android studio 2024.1 and wait until Flutter team throw an update to Flutter so that it can upgrade all the android dependencies to work with JDK 21

2. Do it manually by following this link How to fix Android Compilation issues

It depends on what library you're using and the specific endpoint for methods. They may be outdated.

I need more information:

Which library are you using for this? What is the request delay, and how many requests are you making per day? Are you using a proxy?

Colleagues,this is how it works as necessary:

const delay = (timeout = 0) => new Promise(r => setTimeout(r, timeout));
var StillWorking = false;
function First(){
  console.log('First started');
  StillWorking = true;
  setTimeout(Second, 1000);
  console.log('First fired Second and continue working');
  return;
}
function Second(){
  console.log('Second');
  StillWorking = false;
}
async function Main(){
    console.log('Main');
    First();
    while (StillWorking) await delay(10);
    console.log('All done');
}
Main();

First function should not wait until Second is finished. The only task is to make Main to wait for callong a function Second, so I canot use Promise when calling Second.

But all the code seems very dumb, with a global variable and infinite loop. Yet I do not see more accurate solution.

This one should do it:

Create file .htaccess inside root web directory.

<IfModule mod_rewrite.c>
    RewriteEngine on
    RewriteRule ^(.*)?$  http://newcomain.com/$1 [R=301,L]
</IfModule>

It will redirect all requests with parameters to the new server

First of all, don't use client-side storage if storing sensitive data.

Second, I prefer using caching and Zustand, as they will do almost the same thing in comparison to this.

Third, in my case, I am using session storage to store the reviews list for 1 minute, yeah, kind of a mix, just for better UX.

So now you can decide or ask for more info.

Thank you.

If anyone is still looking for a quick solution, here's a plugin to use: https://sparkpressstudios.com/custom-order-notes-checkout-plugin/

Most probable reason might be a group policy setting (Do not enumerate connected users on domain-joined computers) is applied to your computer. You may verify with rsop.msc console OR gpresult -z from cmd line on the computer that this policy is applied. Since this is a computer policy, it applies to all users (local or domain) logging onto that computer.

There is one thing you have missed. Mounted file should already exist inside docker host before mounting. Otherwise docker creates a directory instead.

So you need to create nginx.conf empty file before mounting

I have zscaler installed and it was blocking port 9000 Changing it to 9001 also did not resolve the issue.

What worked in suggestion above to use port 9005.

I had to add this, which I did not need before.

.vscode/settings.json

{
  "editor.codeActionsOnSave": {
    "source.fixAll.eslint": "always"
  }
}

Using ifelse and base R:

dat <- data.frame(Count = round(runif(100, 1, 100)))
dat$even_or_odd <- ifelse(dat$Count %% 2 == 0, "even", "odd")

few months ago I was a victim of scam . Invested on fx.tnr not knowing it’s a scam zone. They kept asking for more payment and taxes to withdraw my money and later I never heard from them , I contacted their support team but no reply for months I got scared of losing over 50 thousand euros , that’s all I have for over 15 years of working as a secretary. Thank God for a friend who introduced me to Hacknetcyberprofessor10 (@)gmail , at first I thought it wasn’t possible cause most articles online says it cannot be recovered . He assured me of his service and I decided to try I already lost my savings so why not take the risk I thought , I have it a trial and I couldn’t believe it when I got back my funds in less than 24 hours . I still feel like it was a miracle . I’m not one to share reviews online but I know a lot of people are in this mess and they can be redeemed using the service of Hacknetcyberprofessor10(@)gmailcom he is the best to do a recovery in digital assets .

For me, this is what finally fixed it. I set the setting remote.SSH.remoteServerListenOnSocket to true:

{
  //...
  "remote.SSH.remoteServerListenOnSocket": true
}

then killed the VS Code server on the remote and retried my connection.

Inno Setup 6.2.2 version had similar issue on Windows 11. Instead, install its 6.3.3 version. I hope it helps you.

One way with base R:

df = data.frame(Count = c(1, 2, 3, 4, 5))
# Use the modulo operator to return the remainder of dividing by 2
df$evenodd <- df$Count %% 2
# Relabel 0 = Even, 1 = Odd
df[df$evenodd == 0,]$evenodd = "Even"
df[df$evenodd == 1,]$evenodd = "Odd"

I don't know why all the errors I had were so unhelpful and all the documentation I found didn't say this, but the issue was this:

My custom repository block in pom.xml needed this change:

<url>file://SHAREDSERVER/custom/repo</url>

should have been

<url>file:\\SHAREDSERVER\custom\repo</url>

Note the direction of \ vs. /! This is the direction of the slashes as used in Windows file explorer! (This is for a repository that is on a shared network location.)

Vercel only gives you 4 sets of environment variables you can pick from:

• development
• preview
• production
• a specified feature branch (not what you need)

When you're writing your Bitbucket Pipeline file, you've gotta pick 1 of those 4. It's completely up to you on which set of the 4 you want your 'staging' branch to pull from when it's being built.

Derive from CustomNodeManager2 (I named it DataAccessNodeManager) and add the following method to it:

private NodeId AddDataType<T>() where T : struct, Enum
{
    var enumFields = Enum
        .GetNames<T>()
        .Zip(Enum.GetValues<T>())
        .Select(item =>
        {
            return new EnumField
            {
                Name = item.First,
                Value = Convert.ToInt32(item.Second)
            };
        })
        .ToArray();

    var enumDefinition = new EnumDefinition { Fields = enumFields };
    var nodeId = new NodeId(++_nodeId, NamespaceIndex);

    var enumType = new DataTypeState()
    {
        NodeId = nodeId,
        BrowseName = typeof(T).Name,
        DisplayName = typeof(T).Name,
        SuperTypeId = DataTypeIds.Enumeration,
        DataTypeDefinition = new ExtensionObject(enumDefinition)
    };
    
    AddPredefinedNode(SystemContext, enumType);

    return nodeId;
}

Then, in your derived class, use it as follows:

class DataAccessNodeManager : CustomNodeManager2
{
    public DataAccessNodeManager(IServerInternal server)
        : base(server, ...)
    {
        //
    }

    public override void CreateAddressSpace(IDictionary<NodeId, IList<IReference>> externalReferences)
    {
        lock (Lock)
        {
            var myNodeId = AddDataType<HlbCompressor>();

            // ... add other nodes and use myNodeId for the `DataType` property 
        } 
    }
}

The often used UaExpert client software then shows the enum name next to its value:

I too had the issue and after LOTS of troubleshooting, I found an import that I can swear I did not add on my own.

import { send } from 'vite';

That killed the code, and almost me for 4 days. Now we are back on track.

I had a similar issue with TASKING compiler. You require to end the select statement with an asterisk. As shown below:

select ".bss.user_test_bss*";

Hope it helps.

Similar issue here, but the file needed to be in /config/nginx/

So, basically, steps were:

1.) Look for logs/log location Located in /config/log/nginx/error.log

2.) See error where failed login attempt was looking for .htpasswd file in /config/nginx

3.) moved existing .htpasswd file to that location For me 'mv /var/www/html/.htpasswd /config/nginx/.htpasswd'

4.) Restart swag

5.) ???

6.) Profit!

Added/updated to provide more information more than just a couple days out and if already expired

https://github.com/debaxtermsft/debaxtermsft/blob/main/KeyvaultExpiry.ps1

Name Category ExpirationDate KeyVaultName

ted2 SecretExpired 12/2/2022 11:59:59 PM keyvault01 wildcard SecretExpired 12/2/2022 11:59:59 PM KeyVault1 spncert 60-180DaySecretExpiry 1/14/2025 4:45:57 PM KeyVault1 start 180-360DaySecretExpiry 9/12/2025 2:13:48 PM KeyVault1 newsecret GT1YearDaySecretExpiry 1/3/2026 8:03:11 PM KeyVault1 new21424 GT1YearDaySecretExpiry 2/15/2026 9:24:49 PM KeyVault1

"It sounds like you've already tried quite a few options! If the TV isn’t connected to the internet, a Chromecast might not work unless you can set up a local hotspot for both your phone and the Chromecast. Alternatively, some hotel TVs block certain inputs even in non-hospitality mode, so check the settings for HDMI inputs or try different HDMI ports.

I have experienced a lot of issues working with virtualized TreeView, mainly:

• scrollbar jumping
• scrollbar thumb randomly resizing
• scrolling over the end (when using wheel or ctrl+end)
• not immediately displaying all items at the bottom after applying filter

Googling just confirmed people have a lot of similar and unresolved problems. And googling put this case as a first hit, so, I'll say here what helped me the most (besides setting VirtualizingStackPanel.ScrollUnit="Item", of course):

Even though all my TreeViewItems have equal height (I've checked), I needed to explicitly set Height to something on my top element under the TreeViewItem. After doing so, the first two problems were reduced to minimum and other two problems disappeared completely.

Maybe it will help someone...

P.S. I have seen this advice in one comment that hasn't received much attention in one of the threads I've encountered while googling...

Ron maupin

Thanks for your response,

I know is expected behaviour as per RFC,But I want like it should use it's assign static ipv6 address not link local(fe80) address as a source for neighbour reachability purpose,In neighbour solicitation...

Is it possible in Linux machine ?

Try FAB-9000 . :)

https://en.wikipedia.org/wiki/FAB-9000

Greetings from EU!

Found the solution on: https://wiki.nixos.org/w/index.php?title=OBS_Studio

{ config, pkgs, ... }:

{
  environment.systemPackages = with pkgs; [
    obs-studio
  ];

  programs.obs-studio = {
    enable = true;
    plugins = with pkgs.obs-studio-plugins; [
      obs-backgroundremoval
      obs-shaderfilter
      obs-composite-blur
      obs-scale-to-sound
      obs-move-transition
      obs-gradient-source
    ];
  };
}

It's a bug in the new Add-On implementation that occurs with pages with big memory footprints, often due to embedded images (which is the default for a ipynb noetbook) https://github.com/readthedocs/readthedocs.org/issues/11693

Suggested workarounds from the ticket:

Temporarily, if you are hitting this error you can either disable Addons for your project or try to reduce the response size for the page. If you are using a notebook, you could try to output image as files instead of using inline data in the HTML for the images.

I tried many things, and the last one worked for me;

dart pub global activate flutterfire_cli

I concluded that this error is related to Flutterfire, and the actions related to it solved the problem.

https://firebase.flutter.dev/docs/cli/

I was able to fix this bug by downgrading to version 4.1.6 of create-single-spa. I recommend that you do the same, surely in versions later than 5.X this problem will be solved.

Thanks for all comments that I learned from and I edited the query to be:

 UPDATE DBPaths SET Path_Image = 'E:\pics2024' +  RIGHT(Path_Image, (CHARINDEX('\', REVERSE(Path_Image)) )) from DBPaths 

The result will be :

 Path_Image
 E:\pics2024\1.png
 E:\pics2024\22.jpg
 E:\pics2024\947141.JPG

Figured the issue out with the help of some outside source that directed me to look at the header.php file. Found some old code on there that redirected on certain conditions.

This issue can also come up when trying to create a venv on existing broken venv, for example after ensurepip is not available error. Make sure to remove the existing one:

rm -rf .venv

Here is my professor's exercise examples on MGU. Her rule of thumb is that MGU must have variable on the left side and constant (or value or function) on the right side. You can take a look at this to practice more.

So based on her answer, I would say the number 2 is the right one.

https://source.android.com/docs/security/features/apksigning/v2

android extends the zip format to include the signature. there is no file name for it so it can't be read using tools that only understand standard zip.

Same issue with Cloudflare WARP October release. After driving me nuts during deadline crunch time, this worked:

ADB_MDNS_OPENSCREEN=0 adb start-server

It turns out, having justify-content: flex-end; will make scrollbars disappear. For now I can do without, but it is unfortunate.

In an effort to enforce tighter memory safety, iOS 18 introduced new memory allocation features:

The system memory allocator (malloc(3)) has switched to a new implementation for most allocation sizes. Users might experience exposure of latent memory access bugs due to changes in heap layout, differences in performance for allocation-heavy workloads, and changes in fragmentation. (127493322)

This likely includes changes in how memory allocation for zero-sized buffers is handled to guard against information disclosure and other memory safety issues.

According to the malloc documentation:

Crashes in memory allocators are almost always related to heap corruption, such as overflowing an allocated chunk...

You are not checking the return status of the AudioConverterGetProperty function. Check that maxPacketSize is initialized and not zero as a result of this call. Also check that the size is actually sufficient.

iOS & iPadOS 18 Release Notes
malloc(3) — Linux manual page

Are you using a private key for ssh? If so try to give group and others reading rights.

e.g. chmod go+r ~/.ssh/myprivatekey

Worked for me.

I run this problem and solve it

here is the step:

1st: All my working dir has src folder so inside add a folder name: utils 2nd: Delete the components.json files and run the " npx shadcn@latest init"

it will reinstall the package again and add the utils files and It will fix

As wrote luwes, you can add proxy_pass, and I deducted useful assets:

 location ~ \.php$ {
        proxy_pass $scheme://$host$uri/$args;
 }

1. Instead of using static link, you can write dynamically by variables $scheme, $host, $uri;
2. Following WordPress official Nginx Handbook and jamfire's example, variable $args should be preffered over $request_uri.

I run this problem and solve it

here is the step:

1st: All my working dir has src folder so inside add a folder name: utils 2nd: Delete the components.json files and run the " npx shadcn@latest init"

it will reinstall the npx and add the utils the files and It will fix

Duda from Weaviate here!

This is the actual vector of your object. So considering you never passed a vector when you created that object, it was generated by your transformers model.

this endpoint http://localhost:8080/v1/objects?include=vector will yield all the objects in your cluster.

So this is a list of objects, and one of the properties is the class it belongs to.

Let me know if this helps :)

your program is running as expected (with Python 3.12), but I had to change identation for all your 'while True' statements. If all your files are in the same directory, you should be able to run your calculator.

Don't use flutter_blue, flutter_blue_serial, and any dependencies that display the above error. Some of them haven't been updated in 3 years. You should use appropriate alternatives. If you're using BLE, I recommend flutter_blue_plus. For Classic, use flutter_blue_serial.

Batch Transform does not take any python code for input. It takes model name (model is created using Model class - from sagemaker.model import Model and CreateModelStep). Model class accepts an inference script (not mandatory but is one of the parameters) which just focuses on handling incoming API requests. Input data is an S3 location (directory) only. All the records, in files within this directory, are sent to model for inference and corresponding ".out" files are created in output directory. The content type like 'text/csv' defines the kind of files read for inference. These differences make batch transform step/job unique.

For set ylim use ylim=(low_value, high_value) keyword in mpf.plot() or mpf.make_addplot()
For set yticks use next sample of code:

fig, axlist = mpf.plot(**foo_kwargs, returnfig=True)

axlist[target_panel_index].set_yticks(np.arange(low_value, high_value, step))

Note 0.
In my case I have three panels (0 - candelsticks, 1 - volumes, 2 - custom indicator), but for custom indicator I should be set target_panel_index = 4. Why? I don't understand.
Note 1.
And for help: all keywords mpf.plot() and, perhaps, mpf.make_addplot()

Have you tried hinting the format(s) you wish to scan?

const hints = new Map();
hints.set(ZXing.DecodeHintType.POSSIBLE_FORMATS, [
    ZXing.BarcodeFormat.EAN_13
]);
const codeReader = new ZXing.BrowserMultiFormatReader(hints);

CBR to PDF Convertor in Python (single file | batch) :)

https://github.com/AshlanChidester/Holocron-Transformer-CBR-to-PDF-converter/tree/main

Try text-align:start instead of text-align:left, it would work

the automatic test needs two lines of output to be graded correctly. Changing this part of your code:

operation = input("1 - add an entry, 2 - read entries, 0 quit\nFunction:")

to

print("1 - add an entry, 2 - read entries, 0 - quit")
input("Function: ")

will correct your submission.

But how we can see how many free processes has a server name from FM TH_SERVER_LIST? Namely, we choose a server name from the list and we want to see for this server name how many free processes we have.
Thanks
Elias

I just copied your code to my machine and it is running fine. Only in utils.py I found an error as the indentation is wrong (while should be indented). Could you please share a picture of your file system? Other than that I can not help further.

def get_input(prompt):
   while True:
       try:
           return float(input(prompt))
       except ValueError:
           print("Invalid input. Please enter a number.")

The issue comes from how the shell interprets quotes when passing commands to exec, Try triple quotes :

Execute('print("""hello world""")')

The ORA-12170: TNS:Connect timeout occurred error typically indicates that the database link is unable to establish a connection to the target database within the specified timeout period. This can happen due to various reasons. Here are some steps you can take to troubleshoot and resolve the issue:

Check network connectivity:

Ensure that the security groups and network ACLs allow traffic between the two RDS instances. Verify that the VPCs in both accounts are properly connected (e.g., via VPC peering or AWS Transit Gateway). Verify DB link configuration:

Double-check the DB link configuration, including the hostname, port, and service name. Ensure that the username and password used in the DB link are correct and have the necessary permissions. Check RDS instance status:

As @dbugger in the comments helped me realize, I updated rspec from 5.0.3 to 7.0.1 and that solved the issue.

std::thread's destructor will terminate the program if you don't join or detach it before it is destroyed. std::jthread joins on destruction instead of terminating the program. Perhaps your std::thread object is being destroyed before it is detached or joined?

If no pods are running, ensure the deployment is properly configured and try scaling it:

bash Copy code kubectl scale deployment --replicas=1 If the service is misconfigured (incorrect selectors, ports, etc.), redeploy the service with the correct configuration.

If pods are failing, check the Docker image or application logs for potential issues causing crashes.

Upgrade torch to 2.4.1 can solve it.

pip install --upgrade torch torchvision

If you are facing this kind of issue even after you update your jdk version and change the path as well. What you can do is change the run configuration setting. You can access this by running the program, then on the run window beside the rerun option there would be 4 more options, you can click on the last option which is a 3 dots hamburger symbol it presents you with an option Modify RUN Configurations click on it, then you can change the JDK version path.

The option you are looking for is:

.setHiddenDimensionStrategy(Charts.ChartHiddenDimensionStrategy.SHOW_BOTH)

reference: https://developers.google.com/apps-script/reference/spreadsheet/embedded-chart#getHiddenDimensionStrategy()

Не могу открыть ссылку на изображение в чём может быть причина что-то нужно поменять либо что-то нужно удалить в этой ссылке помогите https://tinki.vip/p/982ndfsh4n2l12fdsjsdf

The correct solution to the problem is the following:

class A {
    var x : int 
    var y : int

  }

  class B 
  {
    const l : seq<A>

    method m()
        modifies (set a | a in l)`x
        ensures forall a :: a in l ==> a.x == 10
        ensures forall a :: a in l ==>  unchanged(a`y)
    {
       var it := 0;
       while it < |l|
            invariant 0 <= it <= |l|
            invariant forall a :: a in l[..it] ==> a.x == 10
        {
            l[it].x := 10;
            it := it + 1;
        }
    }
  }

The actual source of the problem ended up being my data type conversion. The time was not being included in the conversion for large datasets. I defined the format with esheet$DateTime<-as.POSIXlt(esheet$DateTime,format="%Y-%m-%d %H:%M") and my subsetting worked within the loop.

myView.isLaidOut() // USE THIS NOW

ViewCompat.isLaidOut(myView) // DEPRECATED

When handling background actions like receiving a notification, the isolate that runs in the background may not have direct access to your providers, which are initialized in the main isolate. To solve this issue and play audio in the background using Riverpod, you can follow these steps:

1. Enable background usage of flutter_local_notifications and just_audio_background

2. Configure the background Isolate to run the necessary code

3. Alternative: Using a Background Service: If the logic you need to run is more complex or if you need to handle more shared state, you can implement a background service that maintains the application state or uses just_audio_background directly. This approach is a bit more complex but gives you more control over background audio playback.

Reflect.set(FixturesManager, 'instance', undefined)

If your setup is not too picky, here is one example:

• BlueJ suggesting OpenJDK from Adoptium
• Brew for Adoption OpenJDK is archived https://github.com/AdoptOpenJDK/homebrew-openjdk
• suggested replacement is:

brew install --cask temurin

If ca:autocomplete="off" doesn't work, try ca:autocomplete="new-password" instead.

<zk xmlns:ca="client/attribute">
    <grid>
        <!-- removed for brevity -->
        <textbox ca:autocomplete="new-password" />
    </grid>
</zk>

This may help;

const myElement = document.querySelector('[data-node-key="holders"]')

myElement.className = 'ant-tabs-tab ant-tabs-tab-active';

It clunky but you could define a custom tailwind utility class that applies both the color and padding in your tailwind config, and then just use the [&>button]:custom-class-name. Here is documentation: https://tailwindcss.com/docs/adding-custom-styles#adding-custom-utilities

This work for me.

1. generate the dacpac file from server
2. Replace it with the local DB project bin\debug and bin\release
3. schema compare drift issue is gone

I had one of the tests with the wrong ints in convert after check50, the deal in my case was that I was forgetting to test the convert function with any int value and only tested for error msgs.

So we need to keep always in mind was it was asked of us, and what we are giving.

I need to do the same thing but make it a V2. How do I do that? Seems like it's something within the content.

Also is there a way of doing this with just using the command line and not an external file?

If anyone has yet this same problem, I found out in my datastax astraDb database that I just had to wait a little. I guess the command works but doesn't tell you when is it it finishes creating the index, returning back an error if you're querying it and it isn't up yet

I struggle a lot with that one, I hate my hesitation time on that topic when coding.

When I think about it, it feels like self variables should be avoided here are some reasons (from my perspective):

• Closer to a black box approach where inputs and outputs are clearly defined.
• Self variables are often left aside when defining method documentation, this can hide the impact of the inputs and the resulting outputs.
• (I like this one) Methods wich avoid the usage of self variables are easier to move outside the class. Removing a method from your class often simplify testing cases.

That being said, in some situation (even if you should normally avoid it) you might be a situation where many inputs are required by your method, at some point passing arguments starts to feel ridiculous.

In conclusion, try to apply a rule you follow or you will be stuck in a mixed approached like me which leads to dirty code :s

def distance_from_zero(x): if type(x) == int or type(x) == float: return abs(x) else: print("not possible")

def distance_from_zero(x): if type(x) == int or type(x) == float: return abs(x) else: print("not possible")

The problem was in a corporate antivirus. It somehow has blocked websocket control frames.

In some cases you may need to include the header inside the extern keyword. For example trying to use a C library like libpci inside a C++ program.

extern "C" {
#include <pci/pci.h>
}

Then you can link it against your C++ program like for example

g++ -std=c++11 -o testutils ../TestUtils.cpp -lpci -L/lib/x86_64-linux-gnu/

Came to this problem before. What i needed to realize is that by default, auto regression only predicts one next step. This is actually mentioned in documentations, but is easy to overlook. This is reason why only one row is added.

If you want to predict more than one next step, use n_forecasts parameter in neural prophet constructor call

since you need 720 days prediction for each consecutive prediction model needs also last predictions. So to create 720 predictions into future, you need to run "next step" prediction 720 times, always adding last guess into list of lags.

One more important thing that happened to me and cost me few days of headache, i used validation dataframe for predicting future same as shown in documentation, and got success result that i was not able to reproduce since. Problem was, that if you provide 720 lines dataframe, containing real 'y' values, NP does 720 individual prediction in silence using last n lags of real 'y' each time. This creates illusion that model can predict 720 days into future at once....which it can not, because in reality you need new predictions to create even newer predictions.

update your node version and re-create the project

This Unity Class Library Template seems to have disappeared in later Rider versions. I sure wonder why they removed it.

Hold control and scroll up with your wheel, just like a word doc or web page.

If batching promises, avoid using Promise.all, and prefer Promise.allSettled. Promise.all will reject if any promise fails returning the first rejection reason. Promise.allSettled allows all promises to run and collectes the results whether or not the promise succeeded or failed.

Adding a solution specific to my situation. Installed MySQL by way of MAMP (cause it's 'easy'). Needed workbench specifically for the ERD generation. Workbench couldn't connect to local mysql installed by mamp; other clients could connect.

MySQL installed by MAMP creates a root user using mysql_native_password plugin which is not supported anymore in MySQL 8.4.

Followed instructions here to change PWD storage for that same root user to sha2 cache:

https://php.watch/articles/fix-php-mysql-84-mysql_native_password-not-loaded

Relevant commands are:

1. Check that it is in fact native password problem:

   SELECT user, host, plugin from mysql.user WHERE plugin='mysql_native_password';

2. Change the user(s) to remove the need for native password plugin:

   ALTER USER 'USERNAME'@'HOST' IDENTIFIED WITH caching_sha2_password BY 'NEW_PASSWORD';

Use the pickMedia form the image_picker

_picker.pickMedia( maxWidth: 390, maxHeight: 844, imageQuality: 100, );

Issue is solved now!

the git post buffer seemed to be not sufficient!

git config --local http.postBuffer 524288000 solved the problem!!!

The blend file is not "that" big, about 38 MB, but obviously too much.

Thank you all for your help!

Another cause for this error might be too many open connections for the backend database to handle e.g. when there are too many instances of the service trying to connect at the same time.