In my case, the problem happened because i changed some inline policy of my iam user so it could not delete some things the stack created. I identified those undeleted resources through the AWS UI and then added necessary permissions to my user and tried again and it worked.
The Realm Settings has a variable called SSE Session Idle. This is what implements the desired behavior and expires the token if the user is inactive for a period exceeding the SSE Session Idle.
The Cookie.ExpireTimeSpan just indicates how long till the Cookie is deleted, but as long as the user is within the SSE Session Idle time and no MaxAge has been set on the OpenIdConnect the user can request a new Cookie by making an HTTP request. This also resets the SSE Session Idle timer.
I was able to Load the live server into my mobile phone. I was able to do so by disabling Microsoft's live preview which was pointing at the host or computer itself, I uninstalled it And installed the live server by Ritwick Dey, Which was hosting the live server on the ipv4 address. Through it, I was able to use that into my mobile phone. Also, I had to enable that specific port from the Windows firewall defender, as it is now serving on different port. Previously, it was serving on port 3000 And now it is serving on port 5500.
For Seamless connectivity I had made my wifi network private so that it would be able to transmit data between the computer and the mobile phone.
This work took a lot of time but finally I would be able to see the outcome into my mobile, so now I would like to give this to the community, so that if someone is facing a problem like this he or she would take lesser time finding an answer.
Thankyou, Hope you all have a wonderful day. 😄😊
Public Function BinaryReverse(ByRef pNumber As Variant) As Variant
If IsNumeric(pNumber) = False Then Exit Function
With WorksheetFunction
BinaryReverse = .Bin2Dec(StrReverse(.Dec2Bin(pNumber)))
End With
End Function
Thanks to all contributors... with your help, anything possible !
A server can have multiple clients that connect to it. Each client will have a clientId and clientSecret as you could then use the server as a single sign on provider
OAuth is a specific protocol for granting authorization to an application.
SSO (Single-Sign On) is the broader concept of signing into a single identity provider (like Google or Microsoft) and then using that authentication to sign into 3rd party applications. SSO may use different protocols to authorize the user in a 3rd-party app, including OAuth, SAML, OIDC.
For some organizations, SSO extends deeper than verifying the user to a third-party app. It may also control which applications each user can sign into and even what access level they should have within each 3rd-party app.
I have the same issue when writing custom functions for XPath evaluation in .NET C#, using XSLT inheritance. When text() is provided as argument, it turns into the type XPathSelectionIterator that is mentioned in the question's title.
The solution for me was to match it with the XPathNodeIterator type in a pattern and access the string as Current?.Value.
var stringValue = arg[0] switch
{
string s => s,
XPathNodeIterator i => i.Current?.Value ?? ""
_ => ""
};
I’d like to ask:
Is it safe to keep this configuration long-term?
Could it cause any issues with future updates of .NET MAUI, Windows App SDK, or when publishing to platforms like the Microsoft Store?
I also have the same problem. As the screenshot of my two external monitors shows, in the Quick Launch and on the vertical (right hand side) monitor, Firefox Nightly appears with its orange icon, while the tray of the left screen shows the blue version.
The Change Icon dialog shows that the icon has no other version, just the orange one. Where does the blue one come from?
After reading the idea from @SidF, I tried to change the scaling of the monitors, which originally were 125% and 100% for the left and right monitor, respectively. I tried every possible combination of 125% and 100%, and some other values too, but the icons always remained, even if I restarted the app between the changes.
var chart = new Chart();
chart.FormatNumber += ChartFormatNumber;
In the event handler, filter the labels for the Y-Axis and format them as needed.
private void ChartFormatNumber(object sender, FormatNumberEventArgs e)
{
/* Process the elements of type 'AxisLabels'. Additionally, filter by Y-Axis. */
if (e.ElementType == ChartElementType.AxisLabels && sender is Axis axis && axis.AxisName == AxisName.Y)
{
/* Format the seconds as 'hh:mm:ss' */
e.LocalizedValue = TimeSpan.FromSeconds(e.Value).ToString();
}
}
This is not something that’s provided out of the box. You can, however, create HTML or JSX converters that can handle it for you. You won’t be able to click a button and swap views to edit, though. At least, not by default.
This might be a controversial take, but at the heart of it, yes, I think it is a bit of a fancy yet powerful data structure (it inherently uses data to create an 'algorithm' unlike most conventional algorithms). The "training" is all about using data to set the weight and biases across the neural network and hence create this 'algorithm'. This essentially means that the weights and biases are in fact your data, massaged and extruded to a humanly unrecognisable level.
We strongly recommend reinstalling the SAP .NET Connector 3.0 as the first step. It comes in both 32-bit and 64-bit versions, so ensure that the connector version is compatible with your operating system (How do I tell if my computer is running a 32-bit or a 64-bit version of Windows?). Additionally, during the installation process, in the ‘Optional setup steps‘ section, make sure to check the ‘Install assemblies to GAC’ option; it is not installed by default**.**
The source if my issue was an existing flutter_launcher_icons.yaml file. The system was finding & using this file 1st and values in pubspec.yaml were ignored.
Sorry for creating this turmoil.
Thank you for the answers and time spent looking at this.
As for why you should not do it anyway, e.g. by trying to make both sides of the relationship the "owner", as suggested by the other answer that says "You actually CAN use @JoinTable on both sides" (can't comment under it due to lacking reputation): If you don't designate either one of the sides as the non-owning side using the mappedBy element of the @ManyToMany annotation, you will get two unidirectional relationships, instead of one bidirectional relationship.
"But I can trick Hibernate into using the same junction table for both of these unidirectional relationships, by using @JoinTable on both sides and specifying the correct columns, so what's the problem?" If Hibernate thinks there are two relationships, it will try to persist it twice.
Consider this scenario, you have the movie and category tables in your database in a many-to-many relationship, tracked by the movie_category table connecting them together. Now you'd like to fetch the movie entity Borat and the category entity comedy, add the category to Borat, and also add the movie to comedy to keep the in-memory model consistent with the database (for some unknown reason you have a bidirectional relationship in this weird example). Hibernate will eventually flush the changes and it will try to write the same thing to the movie_category table twice, once for each relationship, and you will get an error due to the duplicate row (something like <borat_id>, <comedy_category_id> already exists in this table).
I could also imagine it causing more sophisticated, harder-to-debug surprises.
Vercel does not have that feature at the moment. But the vercel domain can be assigned to another branch and this way you can switch your environment and make your productive enironment unreachable. Before you need that, you need create a branch in GIT that serves a "website in maintanance mode" site.
You can assign a different branch in Vercel this way:
Project > Settings > Domains > Your Domain > Connect to environment > Preview > Branch > Select a branch that serves a "maintanance site"
It sounds like the issue may lie in configuration differences between the sample project and your existing one—especially around view resolvers or resource handling in Spring Boot. Double-check that your project’s application.properties (or YAML) file has the correct JSP view settings and that the src/main/webapp directory is properly structured. Also, make sure you’ve included the necessary dependencies for JSP support in Gradle. While you troubleshoot that, why not take a break and browse the delicious options on the Waffle House menu?
I just had the same problem. In my code I have hx-swap="outerHTML transition:true". This works and is the right solution for my project. Everywhere were I want to have page transitions I put this there, but at one place I had it wrong: hx-swap="innerHTML transition:true". Then It goes wrong. So this could might help someone :-)
AOSP utilises EXT4 for its root filesystem (like /data/media/0/; basically, all except mounted ones, for which I believe that solely exFAT is a possible alternative). EXT4 has no filesystem-defined maximum path length. [1] However, the Linux kernel does, exposed via PATH_MAX. It is usually 4096 characters. [2] Unfortunately, programmatically verifying (and, by extension, relying upon) it isn't recommended. [3]
I am having literally this exact same problem... json web token cookie is definitely being sent and appears in the response body, but when i check devtools the cookie is not there but i can clearly see the user is logged in in my app. did you ever find an answer for this?
Reasons:
RegEx Blacklisted phrase (3): did you ever find an answer
It is frustrating that there are so many moving targets and incompatibilities between versions.
I am very much dissuaded whenever I think "I might be able to do that in Python" because I so often run into roadblocks of the forms: you should upgrade to the latest version of Python, or the library you want is not available, or the library you want is not available for the current version.
I only wanted a way to list the titles of the open windows to a text file.
Winlister can't do it. Windows provides no easy way to do it.
I --eventually -- got something to work in Python.
Made KeyedServiceCache non-generic with generic methods to resolve keys from IServiceCollection, that way the in registration of KeyedServiceCache you can inject the IServiceCollection directly with having to register it
Added a ConcurrentDictionary<Type, IEnumerable<object>> for caching
Added filter for KeyedImplementationFactory and KeyedImplementationInstance as well as KeyedImplementationType
Made KeyServiceDictionary<TKey, TService> fully Lazy Loaded, will now resolve the Keys separate to the, resolving of services
Added support to resolve KeyServiceDictionary<TKey, IEnumerable<TService>>
Solved the issue of Captive Dependency by injecting IServiceProviderIsKeyedService into KeyServiceDictionary<TKey, TService> allowing filtering by scope using IsKeyedService method
public class KeyServiceDictionary<TKey, TService> : IReadOnlyDictionary<TKey, TService>
{
private readonly KeyedServiceCache _keyedServiceCache;
private readonly IServiceProvider _serviceProvider;
private readonly bool _isServiceEnumerable;
private readonly Type _serviceType;
private readonly IServiceProviderIsKeyedService _serviceProviderIsKeyedService;
//caching
private readonly Dictionary<TKey, TService> _resolvedServices = new Dictionary<TKey, TService>();
public KeyServiceDictionary(KeyedServiceCache keyedServiceCache, IServiceProvider serviceProvider, IServiceProviderIsKeyedService serviceProviderIsKeyedService)
{
_isServiceEnumerable = typeof(TService).IsGenericType &&
typeof(TService).GetGenericTypeDefinition() == typeof(IEnumerable<>);
_serviceType = _isServiceEnumerable? typeof(TService).GetGenericArguments()[0]: typeof(TService);
_keyedServiceCache = keyedServiceCache;
_serviceProvider = serviceProvider;
_serviceProviderIsKeyedService = serviceProviderIsKeyedService;
}
public IEnumerator<KeyValuePair<TKey, TService>> GetEnumerator()
{
return new Enumerator(Keys.GetEnumerator(), this);
}
IEnumerator IEnumerable.GetEnumerator()
{
return GetEnumerator();
}
public int Count => Keys.Count();
public bool ContainsKey(TKey key)
{
return Keys.Any(x => x.Equals(key));
}
public bool TryGetValue(TKey key, out TService value)
{
value = _serviceProvider.GetKeyedService<TService>(key);
return value != null;
}
public TService this[TKey key]
{
get
{
if (!_serviceProviderIsKeyedService.IsKeyedService(_serviceType, key))
throw new KeyNotFoundException($"Could not find Key:{key} in Dictionary");
if (_resolvedServices.TryGetValue(key, out var service))
return service;
try
{
service = _isServiceEnumerable
? (TService)_serviceProvider.GetKeyedServices(_serviceType, key)
: _serviceProvider.GetRequiredKeyedService<TService>(key);
}
catch (InvalidOperationException e)
{
throw new KeyNotFoundException($"Could not find Key:{key} in Dictionary", e);
}
_resolvedServices.Add(key, service);
return service;
}
}
public IEnumerable<TKey> Keys =>
(_isServiceEnumerable
? _keyedServiceCache.GetKeys<TKey>(_serviceType)
: _keyedServiceCache.GetKeys<TKey, TService>()).Where(k =>
_serviceProviderIsKeyedService.IsKeyedService(_serviceType, _serviceType));
public IEnumerable<TService> Values => Keys.Select(key => this[key]);
private class Enumerator : IEnumerator<KeyValuePair<TKey, TService>>
{
private readonly IEnumerator<TKey> _keys;
private readonly IReadOnlyDictionary<TKey, TService> _parentDictionary;
private KeyValuePair<TKey, TService> _currentKeyPair;
public Enumerator(IEnumerator<TKey> keys, IReadOnlyDictionary<TKey,TService> parentDictionary)
{
_keys = keys;
_parentDictionary = parentDictionary;
}
public bool MoveNext()
{
return _keys.MoveNext();
}
public void Reset()
{
_keys.Reset();
}
private KeyValuePair<TKey, TService> GetCurrent()
{
var service = _parentDictionary[_keys.Current];
_currentKeyPair = new KeyValuePair<TKey, TService>(_keys.Current, service);
return _currentKeyPair;
}
public object Current => GetCurrent();
KeyValuePair<TKey, TService> IEnumerator<KeyValuePair<TKey, TService>>.Current => GetCurrent();
public void Dispose()
{
}
}
}
public static class ServiceExtensions
{
public static void WithKeyServiceDictionarySupport(this IServiceCollection serviceCollection)
{
serviceCollection.AddSingleton(s => new KeyedServiceCache(serviceCollection));
serviceCollection.AddTransient(typeof(IReadOnlyDictionary<,>), typeof(KeyServiceDictionary<,>));
}
}
No, the Terraform provider does not track the scanning type, therefore manually changing it via the console or the CLI will not cause any state mismatch.
You can opt in without worrying about your ECR getting destroyed.
If you've put the whole schema into a single .query() statement there should be quite a few responses generated. You can quickly debug the whole thing (each response will have its own index number), use .num_statements to get the total length handle them one at a time,
If still unsure I would just copy paste the whole schema into a &str and stick that in, see if that works and if so then work bit by bit to what you have with the bytes into a &str one step at a time and see if anything stands out.
And for the quickest iteration and without needing to connect to anything remote you can do the whole thing in memory to start:
Oh, here is the poster of the quesiton. I think I have found the reason. When we update ans, we need to compare cnt1 and cnt2. But before that, I have done cnt1 = (cnt1 % N + _nums1[p1] % N) % N and cnt2 = (cnt2 % N + _nums2[p2] % N) % N which changed how large they actually are.
Aleksandr Dubinsky's answer points to a syntax that can be nested multiple times, when you have more than two choices. In the following I need to return three possible values:
i = 2
s = "start" if i==0 else "end" if i == 10 else None
print(s)
Sadly, this is why PR environments, or having multiple staging/pre-prod environments became so popular. In my opinion, there is sadly no silver bullet for this issue. From my perspective, you will need to limit work that can go into QA, have another environment after QA with a specific branch, or have dynamic environments for each working branch. None of these are simple and require workflow changes, but will make life easier.
Thank you all for responding my questions ! I really appreciate the clarification.
So the sum it all up :
React treats the array like a tree inside the <App/> component,
and the paragraphs are its sub-nodes.
React.memo prevents re-renders unless its props change due to parent re-renders,
conditional rendering, or state changes from the child component.
If a child component updates its own state, it will re-render itself, but this does not cause the parent to re-render. As a result, a new virtual DOM tree is created, with the child component as the root of that tree.
Also a few defintions I understood Based on the explanations you all provided:
Re-rendering: This is the process of regenerating the virtual DOM tree after the initial render or following a state change.
Re-mounting: This refers to the actual update of the DOM itself when components are re-initialized.
Again, thank you so much for all the support— I really appreciate it.
I know my questions might seem basic, but they truly help me a lot. Thanks again! :)
I figured that the error was caused due to strict pydantic definitions, which stop us from defining any new instance variables in the inherited class definition. Although I could not find any workarounds that worked with llama_index, langchain had more flexible definition so switched packages in the end.
For ASP.NET Core with .NET 9, based on @LazZiya's answer and the asp.net docs on how to resolve a service at app startup, this is how I did it in the Program.cs file:
I had started to notice the same thing, though I don't know if it is new or not. You can even see it occur on the angular material site, so it isn't an implementation issue. Open that first dialog without animation button a few times. If you click outside to close, the overlay is leaked, not the dialog component it self, but if you click anywhere inside the dialog at all, that becomes a leak.
$git pull origin your-branch is a shorcut which runs two commands one after another:
$git fetch origin your-branch
$git merge origin your-branch
So:
fetch updates your-branch located on your local repository with the version of your-branch from the remote repository
merge merges the current local branch in your workspace with your-branch from your local repository
So, if you're checked out on a different branch in your workspace (your-other-branch) and you run $git pull origin your-branch, it'll update your-branch on your local repository, then it'll merge your-branch from the local repository into your-other-branch in your workspace.
tdlr; run $git switch/checkout your-branch before running $git pull origin your-branch.
However, the Convert Custom Object to JSON does not support datatable datatypes. It appears that the datatable must be first converted to an intermediate datatype first and there is where I'm hitting a wall. The CustomObject datatype appears to be a (pardon my terminology) a single row entity/instance. I've pursued several posted solutions and none of them work.
I tried the time_pulse_us function with a mixed result. I fed the microcontroller (Pi Pico) from a function generator. The duration I got was a mixed bag with times all over the place. I then used the same setup but with Arduino IDE and used the pulseInLong function. The result was exact to three decimals.
My conclusion is that MicroPython is not suited for time critical applikations. At least not with the time_pulse_us function. It is too unreliable.
Update on this issue. The S3 trigger seemed to have a timing issue. I was able to resolve this by using Cloudwatch EventBridge to monitor S3 CreateObject event and use that as a trigger. Once I removed the S3 trigger, changed to this trigger (and adjusted code for the changes to JSON event), it works perfectly now.
I had a similar error (130 unknown data type) but it was because the original field was created as a char instead of varchar. When I dropped and re-created the field as varchar, the migration to SQL Server worked.
In your POST-triggered push version, you are not keeping the stream open in the GET method for the '/sse' route. As a result, the stream gets closed, and your frontend is no longer able to read from it.
The only required modification is adding a while (true) loop to keep the stream open:
Hi Did you ever get this Solved?
I have a similar issue now,. after from .net to .net core application, migrated LinqToXsd to using the XObjectsCore nuget, Same code base, but now i get the error, duplicate nodes, what is suggested fix this without recreating the cs file?
Reasons:
RegEx Blacklisted phrase (3): Did you ever get this Solved
Now you can go to a test run and click on the three dots on the right of a test case and select Report bug. The issue tracker of your GitLab project will then open with an issue containing all relevant information of the corresponding test case.
I do some research on online learning and I find this topic.
I have a question because in your example u have only one batch of data, am i right ? So if i understand right, you could use .fit because in your example it would do the same result no ?
So to use train_on_batch u have to use a for loop, no ?
time=2025-05-09T13:31:39.202Z level=ERROR source=main.go:185 msg="Missing MIB" mib=SNMPv2-TC from="At line 10 in mibs/WLC.mib"
time=2025-05-09T13:31:39.202Z level=ERROR source=main.go:185 msg="Missing MIB" mib=SNMPv2-TC from="At line 10 in mibs/WIFI.mib"
time=2025-05-09T13:31:39.202Z level=ERROR source=main.go:185 msg="Missing MIB" mib=SNMPv2-TC from="At line 10 in mibs/WAN.mib"
time=2025-05-09T13:31:39.202Z level=ERROR source=main.go:185 msg="Missing MIB" mib=SNMPv2-TC from="At line 10 in mibs/SFP.mib"
time=2025-05-09T13:31:39.202Z level=ERROR source=main.go:185 msg="Missing MIB" mib=SNMPv2-TC from="At line 10 in mibs/PEPVPN-SPEEDFUSION.mib"
time=2025-05-09T13:31:39.202Z level=ERROR source=main.go:185 msg="Missing MIB" mib=SNMPv2-TC from="At line 10 in mibs/LAN.mib"
time=2025-05-09T13:31:39.202Z level=ERROR source=main.go:185 msg="Missing MIB" mib=HCNUM-TC from="At line 12 in mibs/IPT-NETFLOW-MIB.my"
time=2025-05-09T13:31:39.202Z level=ERROR source=main.go:185 msg="Missing MIB" mib=SNMPv2-TC from="At line 14 in mibs/IPT-NETFLOW-MIB.my"
time=2025-05-09T13:31:39.202Z level=ERROR source=main.go:185 msg="Missing MIB" mib=SNMPv2-TC from="At line 9 in mibs/IPSEC-VPN.mib"
time=2025-05-09T13:31:39.202Z level=ERROR source=main.go:185 msg="Missing MIB" mib=SNMPv2-TC from="At line 10 in mibs/GRE.mib"
time=2025-05-09T13:31:39.202Z level=ERROR source=main.go:185 msg="Missing MIB" mib=SNMPv2-TC from="At line 10 in mibs/DEVICE.mib"
time=2025-05-09T13:31:39.202Z level=ERROR source=main.go:185 msg="Missing MIB" mib=SNMPv2-TC from="At line 10 in mibs/CELLULAR.mib"
time=2025-05-09T13:31:39.202Z level=ERROR source=main.go:137 msg="Failing on reported parse error(s)" help="Use 'generator parse_errors' command to see errors, --no-fail-on-parse-errors to ignore"
I downloaded the MIBs it said they were missing into the MIBs folder on the French site, but still no joy. Any guidance very much appreciated!.
It might be due to a mix of tabs and spaces in the indentation. Different codes or editors may use different types of indentation, so issues can occur when you copy and use code with mixed indentation styles. If you select all the code with Ctrl + A like in the image below, you'll be able to tell which type of indentation was used. Additionally, the 'Spaces: 4' label at the bottom of VS Code shows how indentation is currently configured in the editor. You can change it as needed.
Okay so, I forgot that destructuring preventDefault from an Event object gives an unusable function (since it's not bound to the event object's this anymore)
Calling preventDefault on the mousedown event does work, woops '^^
We had same problem with multi language app that is on all platforms (android, iOS, desktop and web) so we came up with building gradle plugin for it.
It uses the defined strings.xml from different folders in the composeResources folder to generate languages along with global state for the current language and helper functions to list all languages in the app or find it with the language code.
You can check out our repository readme for more details of using plugin and see if it satisfying your needs.
The strategy described is not recommended, due to how package manager workspaces work. A workspace and, more broadly, the JS ecosystem, expect that the workspace's definition is at the root of the workspace. What you're doing may work in the short term, or on accident, but it's not recommended.
To share code across workspaces, it's recommended to publish a package to a registry and consume it from there, like you would for any other external package.
Note: The limitations/expectations involved in this decision are based on package managers, not Turborepo. Turborepo does adhere to expectations of the JavaScript ecosystem, so what's described in the question doesn't work because of how package manager workspaces work by convention.
I know this answer is late, but I hope my experience can help someone. In my case, the error occurred while using Plesk shared hosting. I had created an Excel template and pre-formatted thousands of rows in print preview, preparing for a large data export. However, the error disappeared once I removed the formatting. I add the formatting in C# during exporting of the data.
In your Databricks workspace unity Catalog enable, go to Compute > Cluster Policies, click Create Policy, and name it UC_Policy to set up a Unity Catalog-enabled cluster policy.
Attach the UC policy to your cluster.
The notebook in the child pipeline is using a Linked Service that's compatible with Unity Catalog.
The way the Linked Service is defined in each pipeline should be reviewed.
It should be verified the user identity Managed Identity or Personal Access Token—changes between pipeline levels to provide correct Authentication type and cluster details.
It should also be checked whether the notebook activity is directly associated with a policy or cluster pool, or if that association is lost when the notebook is invoked through the parent pipeline.
Once all references use the same linked service and policy:
Run Master pipeline — it should now work successfully.
I wanted to answer this in the hope it would help.
As Radoslav pointed out, what it did come down to was "something completely unrelated is using the same multicast IP/port". As it turns out something else on the network was using this port and somehow getting into jgroups processing causing the strange version to appear and mismatch in the packets. It was unfortunate because it did give off the impression that there was something wrong from a configuration standpoint i.e. The version imported to the jar was mismatching.
Switching to a completely unused port on the machine allowed this to work first time.
Hope it helps if someone else faces this in future.
I saw your question yesterday, I see no one has responded yet so I'll try to get you in the right direction. I'm honestly a bit lost in your code (not your fault) so I can't provide you with an exact solution, but I know where your problem is, the origin.
Every 3D object has an origin point in 3D space. That origin point is completely free from the object in space (it could be anywhere, depending on how it was made). This point however determines how certain changes get oriented, especially stuff like rotation. The object will rotate around this point. If you then look at your first and second screenshot again, you see where your origin point is, it is at the top right corner of your original wooden block. That's why if you turn it towards that top side (second screenshot), it will leave a space, and when you rotate it down, it will rotate 'into' itself.
Again, I'm just not experienced enough to determine how this is set in your code or how to fix it, but I hope you can do something with this.
As an extra, showcasing the expressive power of standard Scheme, here is an implementation that accepts any number of lists, including none at all.(In other words, it behaves just like Python's built-in zip()function.)
Let's go to the REPL and test that it works as advertised:
> (zip) ; no arguments at all
()
> (zip '(1 2 3)) ; a single argument
(1) (2) (3))
;; Let's conclude by trying four arguments.
> (zip '(1 2 3) '(I II III) '(one two three) '(uno dos tres))
((1 I one uno) (2II two dos) (3 III three tres))
Finally, we make sure that the two-argument tests in the original post continue to pass:
I ended up finding that what David said was on the right track, apparently the file env.js cant be in the same folder as the application, but if you set it in a subfolder for example env/env.js and configuring the ConfigMap to write the file actually works.
There is the BarcodeScanning.Native.MAUI package that works well for basic scanning. If you're looking for an enterprise-grade scanner, check out Scanbot SDK. They actually published a tutorial comparing both solutions.
