Is there an implementation of speech recognition (I only need very basic functionality that identifies when it has been silent for X seconds)
in Kivy that works on both iOS and Android devices?
geom_density_2d_filled expects a discrete color scale since it bins the density into contours, so we need to tell scale_fill_viridis to provide a discrete, rather than continuous color scale.
It turns out that when updating to expo version 52.0.7, I did not finish updating all the libraries related to routing to the new version to be compatible.
The command that updated it automatically (in theory) did not work, so I have updated it manually, and I had to put the versions correctly, especially the expo router and react-navigation/drawer.
I know it's something like next trying to run backend code in frontend like others say. But it's not working for me when I try put net: false in next.config.js.
Then I compare a working project with the not working project line by line. I found out the working project has the above line disabled. Just like magic.
Postgres doesn't support automatically altering an integer field to a serial field, see here. While you can do the conversion manually, it's probably easier to continue using an integer field, or to drop the table and recreate it.
Why does my PageSpeed Insights report show "Failed to load resource: the server responded with a status of 404 (Not Found)" for image URLs using blogger_img_proxy, even though the images were previously loading fine?
This is a tricky one! It sounds like you're using blogger_img_proxy to serve your images, which can sometimes cause issues with PageSpeed Insights. Here's a breakdown of why this might be happening and how to fix it:
Possible Causes:
Caching Issues: PageSpeed Insights might be hitting a cached version of your page where the images weren't loading correctly. Try clearing your website's cache and running the test again.
Temporary Server Errors: Sometimes, the blogger_img_proxy server might experience temporary outages or glitches, resulting in 404 errors. This is usually resolved quickly, so try running the test later.
Incorrect Image URLs: Double-check that the image URLs you're using with blogger_img_proxy are absolutely correct. Any typos or incorrect formatting can lead to 404 errors.
Firewall or Security Issues: A firewall or security plugin on your site might be blocking requests to blogger_img_proxy. Temporarily disable these to see if that resolves the issue.
How to Fix and Prevent 404 Errors:
Verify Image URLs: Meticulously check every image URL in your PageSpeed Insights report. Make sure they are accurate and accessible.
Test Image URLs Directly: Open the image URLs in your browser to confirm they load correctly.
Clear Cache: Clear your website's cache (and your browser's cache) and re-run the PageSpeed Insights test.
Check blogger_img_proxy Status: If you suspect a problem with the proxy itself, try using a different image proxy service temporarily to see if the issue persists.
Review Firewall/Security Settings: If you have a firewall or security plugin, temporarily disable it and re-run the test. If the images load, adjust your firewall settings to allow requests to blogger_img_proxy.
Consider Alternative Image Optimization: For long-term reliability, consider using a more robust image optimization solution like a CDN (Content Delivery Network) with image optimization features. This can improve loading times and reduce the reliance on external proxies.
Remember:
User Experience: Always prioritize the user experience. Make sure your images are optimized for fast loading and display correctly on all devices.
PageSpeed Insights Score: While a good PageSpeed Insights score is important, it's not the only factor in website performance. Focus on providing a smooth and enjoyable browsing experience for your users.
If you've tried these steps and are still having issues, provide more details about your website setup and how you're using blogger_img_proxy. I'll do my best to help you troubleshoot further!
Seriously, does StackOverflow have to be so toxic? I asked chatGPT and it helped me only partially, pointing out post_max_size and www-data permissions, some configuration in .htaccess. There was no other resource and the error was vague. Ken's comment solved the problem pretty easily, was it that hard? What benefit have -1 without feedback and pesky comments brought about? I'm a member for 7 years and I have almost 5k points, not a noob who cannot speak English.
I believe it should be deined in connections.toml rather than config.toml, see my below walkthrough
this is what i wrote to myself with regards to snowflake native apps, for future maybe it will be helpful for you
you're supposed to use snowflakecli ( not snowsql ) the command for snowflake cli is "snow" rather than "snowsql" that is used above, not confusing at all right?
for "snow" to work you have to install it and use it in windows in command prompt started as admin
once it's working you have configure it, the way to do this is
issue $ snow connection list
go to the connection file it lists
add your configs in the .toml file the same way other look like ( all see my config below )
issue snow connection set-default (myconnection)
test it by running snow sql --query "select 'this is not painful at all snowflake team' as is_it_painful "
now you should be able to issue snow app run as the walkthrough tells you to do
connection.toml file
["myconnection"]
name = "your_cute_name_not_the_same_as_myconnection_above_which_is_the_real_name"
account = "dagass-graawa"
user = "MAGA47"
authenticator = "snowflake"
password = "will_not_tell_you"
Go to the Checkout Pages in WordPress, remove everything,
and add only this shortcode:
plaintext
[woocommerce_checkout]
This will ensure the WooCommerce checkout page works correctly.
In my case, my app is on react native, and the problem was due to firebase/app, I had to delete the dependency on react native, run npm i, modify Podfile to comment out the references to Firebase related pods that apparently only work if you use modular headers run pod cache clean --all and pod install to remove the dependencies and the error disappeared.
The thing it seemd to be, that using those modular headers somehow "brakes" the UIKit, resulting on several errors.
For now I was using firebase to distribute, but I'm very disapointed because I wanted to implement push notifications using firebase, and I haven't found anyone that could implement it succesfully on iOS, so I'm investigating other options, as codepush.
(For those who didn't understand the other answers)
Yes, I faced the same issue today about call and jmp issue and it was all about a simple thing that i did wrong and it was by declaring .code before .data section due to which it was raising this issue (A2107 cannot have implicit far jump or call to near label) so make sure your segements are correctly placed.
For taking X language and translating it to Y language you could use a compiler. The book Compilers: Principles, Techniques, and Tools by Alfred Aho and Jeffrey Ullman is where you'd want to start on this.
Using the lexical analysis part of the compiler, you could create a system where by as the tokens are made, you analyse them against a database. If the full length of the tokens match, call for that function. Otherwise continue with the other compiler steps.
In Next.js 13, Layouts (layout.js) are used for persistent UI components like headers, footers, or sidebars, remaining consistent across navigations within a route subtree. They are static and don’t re-render. Templates (template.js) are reusable structures that generate a fresh DOM for every navigation, ideal for dynamic UIs, animations, or resetting states. Use layouts for stable, shared structures and templates for UIs requiring re-creation. Both are essential tools for flexible and optimized app design.
use pip install ydata-profiling rather than pip install pandas-profiling
replace pandas-profiling by ydata-profiling in your pip requirements files (requirements.txt, setup.py, setup.cfg, Pipfile, etc ...)
if the pandas-profiling package is used by one of your dependencies it would be great if you take some time to track which package uses pandas_profiling instead of ydata_profiling for the imports
Fallback to application-scoped resources occurs when an application avoids using the container-managed JNDI datasource and instead relies on resources bundled and loaded within the application itself. This can happen due to explicit configuration or as a backup mechanism when JNDI resources fail to initialise.
If the ojdbc.jar is missing from $CATALINA_HOME/lib, Tomcat cannot initialise the JNDI datasource, however, the application can still function if it directly loads the driver from WEB-INF/lib. Many legacy web applications were not designed to use container-managed JNDI datasources by default. They often have hardcoded connection logic as a fallback.
Even with spark.sql.caseSensitive set to False, the schema must match the data structure exactly( As the PySpark documentation shows, this configuration applies to only Spark SQL).
The simple way is to transform the schema dynamically without modifying the data. Here is the working code.
from pyspark.sql.types import StructType, StructField, StringType
schema = StructType([
StructField("name", StringType(), True),
StructField("address", StructType([
StructField("addressline1", StringType(), True),
StructField("addressline2", StringType(), True)
]), True)
])
data = [
{
"name": "some name 1",
"address": {
"addressLine1": "some address line 1",
"addressLine2": "some address line 2"
}
}
]
def transform_keys(obj):
if isinstance(obj, dict):
return {k.lower(): transform_keys(v) for k, v in obj.items()}
elif isinstance(obj, list):
return [transform_keys(i) for i in obj]
return obj
transformed_data = [transform_keys(record) for record in data]
df = spark.createDataFrame(transformed_data, schema)
df.show(truncate=False)
Please let me know if you need more information.
Reasons:
RegEx Blacklisted phrase (2.5): Please let me know
I realize that this question is from a year ago, but in my search for a small tts model, I came across this question, and wanted to share a model that in my experience trains pretty well on just a few hours of audio.
In my case I had to train a language with no prior pre-trained model, so that was a challenge, but still with just a 2 hours dataset of audio and transcript, I managed to get quite intelligible speech out of training it for less than 15 hours on my GPU (RTX 4080 with 16GB of vRam). It was not a model that's ready for the market, but still is quite decent. And that was training on a brand new language for the model.
Not sure if this still has your interest now, a year later, but take a look at Piper TTS (Open Source). That is at least one model that can work pretty well on a small dataset. I would be cautious of expecting a lot, if you only have 20 sentences of audio, but if you train on a language that has a pre-trained checkpoint, I'm pretty sure it would work with somewhat decent quality, if you train on top of that. https://github.com/rhasspy/piper
It is small enough to run on a Raspberry PI, and performs really well. Also take a look at this tutorial, on training the model: https://www.youtube.com/watch?v=b_we_jma220
What worked for was to set permissions (the accepted answer), then copy file to a Desktop, for example. There I was able to edit it using Xcode. Then I copied the file back using cp command and that's it.
This error typically indicates an issue with the SSH handshake and also this could happen if the SSH server on the VM is blocked or misconfigured and also I would recommend you to reset the daemon with: service sshd restart and also try to restart the VS code and check if it is working.Also could you please confirm whether you have checked this link?
The "DeviceTokenNotForTopic" error typically indicates a mismatch between your app’s bundle ID, VoIP certificate, or APNs environment (sandbox vs. production). Start by verifying that the VoIP certificate in the Apple Developer Portal matches your app’s bundle ID exactly and that you’re using a VoIP Services Certificate, not a standard push certificate. Ensure the environment aligns with your certificate type: use the sandbox endpoint (api.sandbox.push.apple.com) for development and the production endpoint (api.push.apple.com) for live use. Double-check that the device token you’re using comes from PKPushRegistry for VoIP pushes and matches your app’s bundle ID without any formatting errors. Test your certificate by exporting it as .p12, converting it to .pem with openssl, and using tools like PushTry or curl to send test notifications. If you resolved trust issues with your Keychain certificate, confirm it’s now fully trusted and re-export the .pem file if needed. If problems persist, consider deleting and recreating the VoIP certificate in the Developer Portal, then updating your server with the new certificate. Also, review APNs logs in the macOS Console app or your server logs for more details. Finally, if all else fails, reach out to Apple Developer Support with detailed information about the issue and the steps you’ve taken to resolve it.
Just like everyone said when we have import and export in same file we get this TS error.
Sometimes I want to aggregate all file exports in index.ts file and export them again in one place (index.ts)
In this case exporting as default will fix the problem.
import {CONFIG} from './config.ts';
import {COLORS} from './colors.ts';
// and other imports
export default {
CONFIG,
COLORS
}
To load-balance conferences across multiple FreeSWITCH servers, it's a bit trickier than regular call routing because all participants need to join the same conference instance. If you just route calls to the same conference number from different servers, each server will create its own isolated conference. To allow participants from different servers to communicate, you’ll need to set up conference cascading—essentially linking the conference instances together between FreeSWITCH servers, typically using SIP trunks or bridging.
Kamailio can help route conference calls to the right FreeSWITCH server, but you’ll need to ensure all participants join the same conference instance. For large conferences, you might want to route all calls to a primary server or use cascading to connect multiple servers together.
In short, while Kamailio can load-balance incoming conference calls, you’ll need to implement cascading or link servers together to manage a single, unified conference. For scaling, you may also want to consider using media relays or distributed processing to handle the media between servers.
Do the following checks, maybe this will sort it out:
Case Sensitivity: Make sure your folder is actually named components (all lowercase). If it’s named components but you’re importing it as @/Components/NavBar (with a capital C), that will break. Try updating your import to:
import { NavBar } from "@/components/NavBar"; // lowercase "components"
Check Your tsconfig.json: If you're using the @ alias, check your tsconfig.json file to make sure it includes this under compilerOptions:
"paths": {
"@/*": ["./*"]
}
This configuration ensures TypeScript can resolve paths starting with @/. If it’s not there, add it, save, and restart your TypeScript server. (In VS Code, press Ctrl+Shift+P and search for TypeScript: Restart TS Server.)
File Location: Double-check that NavBar.tsx exists exactly where you’re pointing to. It should be inside a folder named components at the root (or wherever your alias resolves). Also, make sure there are no typos in the file name, and it’s using the correct extension (.tsx or .js).
Clear the Cache: If everything looks fine and it’s still not working, try restarting your Next.js server and clearing the cache. Run:
npm run dev -- --reset-cache
Hopefully, one of these steps fixes the issue! Let me know how it goes! 😊
I've tried to replicate your code using SDL preview-3.1.6 and both the extension number and the required extensions get reported correctly. The steps I followed are:
Add an include directory that points to SDL/include.
Add a linker path to SDL/VisualC/x64/Debug
Copy SDL/VisualC/x64/Debug/SDL3.dll into the same folder as the exe.
If after following these steps you get the same error you need to check that you have a good Vulkan installation and that you are linking correctly with SDL.
First thing I would try is to check that the demos in the Vulkan SDK work. You can find them at %VULKAN_SDK%/Demos.
After that I would test a basic triangle sample you can find online. For example:
Then set the Triangle demo as the default project, then build and run it
If the Vulkan Demos or the Triangle Example work it means you have a problem with SDL. Check that you are linking with the correct lib/dll or try with SDL2.
Since the issue with conference_auto_record is happening intermittently, it’s likely due to timing, resource conflicts, or certain call scenarios. To troubleshoot, try adding a small delay (1-2 seconds) in the sched_api command to ensure the conference is fully set up before recording starts. Compare channel variables for successful and failed calls using uuid_dump to spot inconsistencies, especially with the conference_auto_record variable. Turn on detailed logging to catch any errors or unusual behavior during the process. Also, check your system resources (CPU, memory, disk) to rule out performance issues, and test under higher call volumes to see if the problem only happens under load. Finally, make sure the variable is being passed reliably, using export if needed, and enable conference event logging to confirm when recordings are being triggered. These steps should help you pinpoint the cause.
How can I efficiently merge multiple dictionaries in Python, considering Python 3.9+ and older versions?
I'm working on a project where I need to merge several dictionaries into one. The number of dictionaries can vary, and some of them may contain overlapping keys. What is the most efficient way to merge these dictionaries while maintaining the values from later dictionaries when there are conflicts?
For example, I want to merge the following dictionaries:
python
Copy code
dict1 = {'a': 1, 'b': 2}
dict2 = {'b': 3, 'c': 4}enter code here
dict3 = {'d': 5}
And expect the merged result to be:
python
Copy code
{'a': 1, 'b': 3, 'c': 4, 'd': 5}
Bonus: Is there a one-liner for this in Python 3.9 and later, and how can I achieve the same in older versions of Python?
This question touches on the topic of merging dictionaries, including handling conflicts, which is a common task in Python. It also encourages discussion about compatibility between newer and older versions of Python, which is often helpful in a community setting.
In the Signing & Capabilities tab, make sure the following capabilities are enabled:
Push Notifications
Background Modes (Remote notifications option must be checked)
There is always a way :-) but maybe sometimes a long one. Please open an issue on github and provide as much information as possible and i will try to help.
I found this looking for the same thing. Basically My window is 281 columns wide. I'd like for tmux to choose to split along the vertical axis (create side-by-side windows) and balance those until splitting them would create a window that is too narrow (80 columns in my case). Instead I get very wide but short windows which for looking at CLI results is less optimal I feel.
I wonder how hard this would be to code as a new layout?
On one of my test devices, this sometimes happens. One solution that sometimes works (2 times out of 2 so far) is to reboot the phone, and then no other change is needed (same Mac, same USB C cable, same android studio, etc.), and this time it remembers my previous authorizations.
The point of CQRS as @geek said, is the separation of write and query models. I think you should reconsider using CQRS as it focuses on separation rather than sharing logic. What you are looking for is specification pattern to reuse specs. But these two are not mixable. Read this link from Vladimir Khorikov for more information:
Were you able to resolve this issue? I'm encountering the same problem on Ubuntu. The directory definitely exists, and I can even create a new folder through the Dart code during initialization. However, the main issue persists. I’ve tested the same setup on Windows, and it works perfectly—I'm able to get the devices on the network there. Could this be related to a permission issue on Ubuntu? Any insights would be greatly appreciated!
Reasons:
Blacklisted phrase (1): appreciated
RegEx Blacklisted phrase (1.5): resolve this issue?
If you hope to extract "0129" from "AXB1234_CBS0129",
=MID(str,XMATCH(TRUE,ISERROR(--MID(str,SEQUENCE(LEN(str)),1)),,-1)+1,999)
enter image description here
Reasons:
Blacklisted phrase (1): enter image description here
If I were you I would consider using Celery a powerful asynchronous task queue for Python. It allows you to run functions in the background without blocking your main web request or causing the page to hang.
Here’s a basic approach (You can find more in the docs)
Install Celery and a message broker like Redis
Move your response-generating loop into a Celery task
Trigger the task from your Django view without waiting for it to complete.
Using the trail with MySql: I get no response at all. Python simply stops, and I have to start it up again from scratch.
Using ODBC, I get:
exec(open('pythonx.py').read())
1
2
Traceback (most recent call last):
File "", line 1, in
exec(open('pythonx.py').read())
~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "", line 25, in
pyodbc.InterfaceError: ('IM002', '[IM002] [Microsoft][ODBC-stuurprogrammabeheer] De naam van de gegevensbron is niet gevonden en er is geen standaardstuurprogramma opgegeven (0) (SQLDriverConnect)')
Or, after adjusting the driver mentioned
pyodbc.Error: ('HY000', 'The driver did not supply an error!')
You can go to https://lmarena.ai and ask your questions. Choose a suitable model (I liked deepseek-v2.5) and speak to it on the topic. It will provide you with detailed instructions.
Good luck.
email format to enable classic editor
to get permission to the organizational level on azure devops you have to an email in this format to azure team and this issue will resolve within 2 days
There are many architecture patterns and you can follow any one that suits your need. By default android follow mvc but most developers prefer mvvm since its also recommended by google itself.
Since you are new to android I would suggest you to follow the default architecture (mvc) you can place the business logic in your activity or fragment (especially if your building a small app).
First get familiar with android eco system by making different small apps then when you are confident move to advance topics. (Its my opinion, others can agree or disagree).
Incase you prefer to follow other architecture patterns like mvvm then you can put your buisness logic in repository or use clean architecture with mvvm and place your business logic in usecases.
I just managed to solve the problem for me (as far as I can tell with the testing I have done so far). I had the same app with two different ids on my phone, one for development, the other for (beta/performance) testing. They both used the same id resulting in some kind of conflict when accessing/writing data with mmkv. In any case, deleting the preduction build solved the problem. I will simply use two different mmvk storage ids to avoid the problem going forward.
When converting the code to IL, we observe that abstract methods are called using the callvirt instruction. This signifies that they are virtual by nature and rely on a runtime mechanism called the vtable (virtual table).
IL Abstract.
A vtable is a per-class table that holds references to the implementations of virtual and abstract methods. This table allows derived classes to override the base class's abstract or virtual methods. At compile time, the compiler ensures that the derived class refers to the base class’s vtable and updates the entries to point to the correct implementations from the derived class.
IL override
This mechanism inherently relies on object instances to resolve the method calls at runtime, which is why abstract methods are tied to instance-based polymorphism.
However, static methods work entirely differently in C#. They are not associated with instances of a class but rather with the class itself. Static methods are resolved at compile time based on the type, and they do not participate in the vtable mechanism. Because static methods are not tied to an instance, there is no runtime resolution, and they cannot be overridden or abstracted like instance methods.IL Static
Due to this fundamental difference, it is not possible to define static abstract methods in a class. The runtime has no mechanism to support such behavior, as it would require combining type-level resolution (for static methods) with instance-level polymorphism (for abstract methods), which is not supported in C#.
import mysql.connector
cnx = mysql.connector.connect(user='root', password='Aragorn_2024',
host='127.0.0.1', database='Aragorn')
cur = cnx.cursor()
cur.execute("INSERT INTO sakila.tempora SELECT sysdate()")
for row in cur.fetchall():
print (row[0])
cnxn.close())
Using ODBC, the code looks like this:
import pyodbc
cnx = pyodbc.connect('DRIVER={Devart ODBC Driver for SQL Server};Server=127.0.0.1;Database=Aragorn;Port=3306;UserID=root;Password=Aragorn_2024')
cur = cnx.cursor()
cur.execute("INSERT INTO sakila.tempora SELECT sysdate()")
for row in cur.fetchall():
print (row[0])
cnxn.close()
It seems the GPT-4 with browsing (GPT-4o) feature is either disabled or temporarily unavailable. You might be selecting the model correctly, but it's defaulting to regular GPT-4, which doesn't have internet access.
Try:
Logging out and back in.
Clearing your browser cache.
Checking OpenAI status for updates.
Contacting OpenAI support for clarification.
If browsing is down, responses will be immediate since the model isn't fetching real-time data.
this is what i wrote to myself with regards to snowflake native apps, for future maybe it will be helpful for you
you're supposed to use snowflakecli ( not snowsql ) the command for snowflake cli is "snow" rather than "snowsql" that is used above, not confusing at all right?
for "snow" to work you have to install it and use it in windows in command prompt started as admin
once it's working you have configure it, the way to do this is
issue $ snow connection list
go to the connection file it lists
add your configs in the .toml file the same way other look like ( all see my config below )
issue snow connection set-default (myconnection)
test it by running snow sql --query "select 'this is not painful at all snowflake team' as is_it_painful "
now you should be able to issue snow app run as the walkthrough tells you to do
connection.toml file
["myconnection"]
name = "your_cute_name_not_the_same_as_myconnection_above_which_is_the_real_name"
account = "dagass-graawa"
user = "MAGA47"
authenticator = "snowflake"
password = "will_not_tell_you"
As discussed in the comment section, without knowing ordering of data it would not be possible to retrieve the output, because select does not guarantee it will output in insertion order, unless you have a column to determine it.And every calculation done on grouping similar timestamp could lead to different output in different runs.
Usually Android applications follow MVVM design pattern or sometimes MVP pattern, in addition to those sometimes we add Use Case - Repository - Service a typical app using latest technology and built from the ground up in a decent pattern and ready for unit testing would be like this.
Model - View - ViewModel - UseCase - Repository - Service with each one having it's own unique functionality, and all with a Dependency Injection running as well for testing, something like Dagger or Hilt for larger apps, or maybe Koin for smaller ones
In addition to the existing solutions, I found that Assimp::Importer::ReadFileFromMemory only read file ptr but no handling outsource file, such as mtl file. It is commented in the library. Instead, I should use Assimp::Importer::ReadFile.
