<duration unit>: identifier for the unit of duration:
d: days
w: weeks
m: months
y: years
So if you want to get issues created less than 7 days ago, you would add this qualifier to your search query:
created:>@today-7d
Unfortunately this feature is not mentioned in the relevant documentation, so this related documentation for GitHub Projects is the closest thing to an authoritative reference:
<duration unit>: identifier for the unit of duration:
d: days
w: weeks
m: months
y: years
So if you want to get issues created less than 7 days ago, you would add this qualifier to your search query:
created:>@today-7d
Unfortunately this feature is not mentioned in the relevant documentation, so this related documentation for GitHub Projects is the closest thing to an authoritative reference:
You can do this by adding the following qualifier to your search query:
created:@today
Unfortunately this feature is not mentioned in the relevant documentation, so this related documentation for GitHub Projects is the closest thing to an authoritative reference:
After version 106, you have to specify a full path to a chrome profile through the --user-data-dir argument and it cannot be the default profile at all.
from fpdf import FPDF
# Create a PDF class
class PDF(FPDF):
def header(self):
self.set_font("Arial", "B", 12)
self.cell(0, 10, "", ln=True)
def footer(self):
self.set_y(-15)
self.set_font("Arial", "I", 8)
self.cell(0, 10, f"Page {self.page_no()}", 0, 0, "C")
# Create the cover letter PDF
pdf = PDF()
pdf.add_page()
pdf.set_auto_page_break(auto=True, margin=15)
pdf.set_font("Arial", size=12)
text = """Mohamed Ebrahim El-Said
Zagazig, Egypt
+20 112 686 3737
[email protected]
May 24, 2025
Recruitment Team
Deloitte
Dear Hiring Team at Deloitte,
I am writing to express my interest in joining Deloitte. I recently graduated from Zagazig University with a degree in Accounting, where I achieved an Excellent academic standing and a GPA of 3.6.
During my time at university, I developed strong analytical and problem-solving skills through both coursework and practical projects. My academic performance reflects my commitment to excellence, and I am eager to bring that same level of dedication to your esteemed firm. Deloitte’s global reputation for innovation, professional growth, and integrity makes it an ideal environment for me to begin my professional journey.
I am particularly drawn to Deloitte’s collaborative culture and its focus on continuous development. I am confident that my educational background, combined with my enthusiasm for learning and growth, makes me a strong candidate for your team.
Thank you for considering my application. I would welcome the opportunity to discuss how my skills and ambitions align with Deloitte’s goals. Please find my CV attached for your review.
Sincerely,
Mohamed Ebrahim El-Said
"""
# Add the text to the PDF
for line in text.split("\n"):
pdf.multi_cell(0, 10, line)
# Output the PDF to a file
pdf_path = "/mnt/data/Mohamed_Ebrahim_El-Said_Cover_Letter_Deloitte.pdf"
pdf.output(pdf_path)
pdf_path
This just drove me nuts also and I wish someone else had responded here... It appears that via the Clone Repo > Github it only shows personal repos. Instead, copy your Organization repo URL from Github and paste it into the Repository URL tab on Android Studio. It will then prompt you to login or use your access token and voila!
using BenchmarkDotNet.Configs;
using BenchmarkDotNet.Jobs;
public class CustomConfig : ManualConfig
{
public CustomConfig()
{
// Starts with an empty config adding only the jobs you want
AddJob(Job.Default.WithCustomBuildConfiguration("Benchmarks"));
}
}
I didn't realize that when I use EAS to build my app, EAS uses a specific Mac OS and Xcode version as specified in my app's eas.json file. Because I had the production iOS image set to "latest", EAS was using Xcode 16.3 to compile my app.
I was thinking that when I used EAS to build my app, it sent the version of Xcode that I had on my local computer to EAS to build with that.
This documentation was helpful to learn about the image options available when using EAS to build an app:
I put this in a javascript file that is called in every page on my site.
I tested it with Chrome & Firefox.
var ADS_URL = 'https://pagead2.googlesyndication.com/pagead/js/adsbygoogle.js';
function checkAdsBlocked(callback) {
var xhr = new XMLHttpRequest();
xhr.onreadystatechange = function () {
if (xhr.readyState == XMLHttpRequest.DONE) {
callback(xhr.status === 0 || xhr.responseURL !== ADS_URL);
}
};
xhr.open('HEAD', ADS_URL, true);
xhr.send(null);
}
checkAdsBlocked(function(adsBlocked) {
if (adsBlocked) {alert ('\n\nPlease Disable Your Ad Blocker\nThere are no annoying ads or popups on this site.\nSome features here are not available with your ad blocker on.\nThanks!\n\n');}
});
The solution provided above now has issues due to updates:
Thus, since the apk in Termux-app-bluetooth is based on the version 0.64 of the Termux project you must install a more recent version of it.
However, ~~after September~~ when I get some free time, I will try to work on a new version of Termux-app-bluetooth that supports update or installation of packages.
I realized the issue was happening due to the incorrect jdk version being installed. I completely uninstalled java from my computer <https://www.java.com/en/download/help/mac_uninstall_java.html> along with the jdk <https://docs.oracle.com/javase/8/docs/technotes/guides/install/mac_jdk.html#uninstall>. Then installed jdk 21 (jdk 24 is not compatible with kotlin) and the build succeeded
This means that if you imagine memory as the flat Earth and your array lays close to the edge, then beyond it there is a place for at least one additional element. ;•)
To change the language of std::filesystem::filesystem_error messages, adjust the system locale to your native language. But I think this method depends on your operating system.
Though I would update with my current best solution based on the responses. Thanks to @Jérôme Richard for the detailed analysis, I've re-implemented in a way that I think incorporates the most impactful suggestions:
cimport numpy as cnp
ctypedef cnp.npy_uint32 UINT32_t
from libc.math cimport exp
from libc.stdlib cimport malloc, free
cdef inline UINT32_t DEFAULT_SEED = 1
cdef enum:
# Max value for our rand_r replacement (near the bottom).
# We don't use RAND_MAX because it's different across platforms and
# particularly tiny on Windows/MSVC.
# It corresponds to the maximum representable value for
# 32-bit signed integers (i.e. 2^31 - 1).
RAND_R_MAX = 2147483647
##################################################################
cdef class FiniteSet:
"""
Custom unordered set with O(1) add/remove methods
which takes advantage of the fixed maximum set size
"""
cdef:
int* index # Index of element in value array
int* value # Array of contained element
int size # Current number of elements
def __cinit__(self, cnp.ndarray[int] indices, int maximum):
self.size = len(indices)
# Allocate arrays
self.index = <int*>malloc(maximum * sizeof(int))
self.value = <int*>malloc(maximum * sizeof(int))
if not self.index or not self.value:
raise MemoryError("Could not allocate memory")
# Initialize index array to -1 (invalid)
cdef int i
for i in range(maximum):
self.index[i] = -1
# Set up initial values
for i in range(self.size):
self.value[i] = indices[i]
self.index[indices[i]] = i
def __dealloc__(self):
"""Cleanup C memory"""
if self.index:
free(self.index)
if self.value:
free(self.value)
cdef inline bint contains(self, int i) nogil:
return self.index[i] >= 0
cdef inline void add(self, int i) nogil:
"""
Add element to set
"""
# if not self.contains(i):
if self.index[i] < 0:
self.value[self.size] = i
self.index[i] = self.size
## Increase
self.size += 1
cdef inline void remove(self, int i) nogil:
"""
Remove element from set
"""
# if self.contains(i):
if self.index[i] >= 0:
self.value[self.index[i]] = self.value[self.size - 1]
self.index[self.value[self.size - 1]] = self.index[i]
self.index[i] = -1
self.size -= 1
cdef class XORRNG:
"""
Custom XORRNG sampler that I copied from the scikit-learn source code
"""
cdef UINT32_t state
def __cinit__(self, UINT32_t seed=DEFAULT_SEED):
if (seed == 0):
seed = DEFAULT_SEED
self.state = seed
cdef inline double sample(self) nogil:
"""Generate a pseudo-random np.uint32 from a np.uint32 seed"""
self.state ^= <UINT32_t>(self.state << 13)
self.state ^= <UINT32_t>(self.state >> 17)
self.state ^= <UINT32_t>(self.state << 5)
# Use the modulo to make sure that we don't return a values greater than the
# maximum representable value for signed 32bit integers (i.e. 2^31 - 1).
# Note that the parenthesis are needed to avoid overflow: here
# RAND_R_MAX is cast to UINT32_t before 1 is added.
return <double>(self.state % ((<UINT32_t>RAND_R_MAX) + 1))/RAND_R_MAX
def xorrng_py(int seed=1):
cdef XORRNG prng = XORRNG(seed)
return prng.sample()
###############################################################
cdef XORRNG prng = XORRNG()
def spgreedy(cnp.ndarray[double, ndim=2] J,
cnp.ndarray[double, ndim=1] h,
FiniteSet s,
double temp,
):
cdef int d
d = J.shape[0]
cdef int j, k
cdef double dot, curr, prob
for j in range(d):
s.remove(j)
dot = h[j]
for k in range(s.size):
dot += J[j, s.value[k]]
curr = dot / temp
if curr < -100:
prob = 0.0
elif curr > 100:
prob = 1.0
else:
prob = 1.0 / (1.0 + exp(-curr))
if (prng.sample() < prob):
s.add(j)
return s
which is now better than the dense algorithm (in the problems I've tested on). It also takes a similar amount of time as numpy's matrix multiplication (just doing J@s+h) which is both reassuring, since that code is very optimized I assume, but a bit disappointing because I don't get much benefit from the sparsity after all. I also don't know much about Cython so it's very possible I did something silly in this implementation as well!
This is due to how sql and spark execute these functions differently.
In SQL, the rand(42) function is deterministic for each row within that session/query. Hence it produces the same result every time you run that query.
In a Python Notebook, the execution is distributed across Spark executors. Spark's rand(seed) is only deterministic per executor, not globally. Also see rand function's source code, they also mention there that this is non-deterministic in the general sense, and hence the behavior that you are seeing.
Let understand with examples, you are trying to send and receive messages between parts of your app — like one part says, "Hey, a new order came in!" and another part listens for that and does something (like sending an email).
What is a Message Bus?
A message bus is like a middleman that helps parts of your app talk to each other without knowing about each other directly. It's like a delivery service: you drop off a message, and it makes sure the right person gets it.
What is MassTransit?
MassTransit is a tool (or framework) that helps you use a message bus more easily in your .NET apps. It gives you a friendly, consistent way to send and receive messages.
But MassTransit doesn’t deliver messages itself — it uses real delivery services behind the scenes.
What’s the Messaging Library?
This is the actual delivery service — like:
-RabbitMQ
-Azure Service Bus
-Amazon SQS
MassTransit connects to one of these behind the scenes and handles the setup, sending, and receiving for you.
So yes — you're right:
MassTransit is a message bus framework that works on top of real messaging systems (like RabbitMQ or Azure Service Bus). It lets you switch between those systems easily without changing much of your code.
public sealed class CustomMySqlSqlGenerationHelper : MySqlSqlGenerationHelper
{
public CustomMySqlSqlGenerationHelper(
RelationalSqlGenerationHelperDependencies dependencies,
IMySqlOptions options)
: base(dependencies, options)
{
}
public override string GetSchemaName(string name, string schema)
=> schema; // <-- this is the first part that is needed to map schemas to databases
}
Hi, since you already have the driver file/inf, then have gander with this
#find device
$Device = Get-PnpDevice -FriendlyName "Name of your device"
#update driver
Update-PnpDevice -InstanceId $Device.InstanceId -Driver $DriverPath
Event library that you must use:
package0 and replication.
Inside the replication library, you can find events related to the agents, including the distribution agent.
Key event for your case:
replication.agent_message.
This event captures messages that the replication agents (snapshot, log reader, distribution) generate. It is useful because many times the distribution agent writes to its output when the execution mode or the number of threads changes.
You can filter messages containing things like:
“Switching to single thread”
“Switched from multi-threaded to single-threaded”.
SQL Server doesn't have a direct event that says “switched from multi-threaded to single-threaded”, but the agent messages (which are captured with replication.agent_message) do reflect that behavior internally, and that's what you can track.
Simply refreshing the content of the QTextBrowser / QTextEdit with 'the next' picture slide, using a QTimer delay, does achieve the desired goal - displaying an animation.
You need to ask tap payment support to register your bundle ID. After they register it, you will get new secret keys for both production and sandbox. kindly use those keys, and also, make sure to put the correct Sdk configurations.
This way, the payment sheet will be opened
All of above mentioned solutions are unreliable and should not be accepted.
All of them will fail whenever process requests console input from user.
In this case console message will stuck in standard input stream and and won't get into OutputDataReceived event. This is because input prompt is unflushed string in stream buffer and is not committed as output data until user responded. Asynchronous read of process output using events doesn't allow to read unflushed data in stream buffer. The only way to make reliable process output read is to redirect output to file and monitor file changes instead of reading directly from standard output stream. But this workaround may be not suitable for case when there may be sensitive data in output stream.
The Parrot equations do not include raising many of the functions to a power. Huffington left off all the power expressions in the three equations, as they do in most Yeganeh examples.
Can you explain why the use of URL fixes the issue? I guess it is because URL loads the file as an absolute path while the previous method would load the .env file from the incorrect directory when nested inside module files?
After you have loaded the Chrome Extension in the Browser , you will have to reload the site with the domain mail.google.com . After reloading you can open the console and check it will show the log.
Incase you do not see the logs , you can also check if the content script has been injected in the tab . Open the devtools and click on sources . Under it select content script . If your file is injected then it will be visible there.
Are you sure your ic_notification exists in the drawable directory?
Add this to your main activity :
<intent-filter>
<action android:name="android.intent.action.VIEW" />
<category android:name="android.intent.category.DEFAULT" />
</intent-filter>
This should always redirect the users to your app instead of settings
Also sometimes its tricky because permissions are messed up when you are simulating vs actually running it locally on a device
Add this to your <service> tag :
android:name="com.ryanheise.audioservice.AudioService"
android:foregroundServiceType="mediaPlayback"
android:exported="true">
Make sure "AndroidManifest" includes :
<uses-permission android:name="android.permission.FOREGROUND_SERVICE_MEDIA_PLAYBACK" />
Should fix the issues you have, but also make sure that you don't have local and flutter notifications conflicting, check your flutter_local because remember its not a simulation when its local :)
No there is not, and the introduction of ISO/IEC 13211-2 forecloses on the possibility:
This International Standard defines syntax and semantics of modules in ISO Prolog. There is no other International Standard for Prolog modules.
You can see the list of ISO standards under development published on their website. There is nothing to suggest a revision or supersession of 13211-2 is in the works.
I found this article while looking for a way to script changes for an email domain name change on local AD that was synced to 365. I wanted to be able to have a precise lookup and be able to use variables to fill the right alias into the account in preparation for the migration to the new domain. While this got me closer than anything else I found online (which I found so hard to believe), I was challenged with getting to operate properly on a server that had way too many differences between SamAccountName and GivenName. I also wanted to have something that was less specific and provided options for future use across multiple clients we support that are Azure ADSynced. Enjoy.
$proxydomain = "@domain.com"
$OUpath = 'OU=Users,OU=FOLDER,DC=domain,DC=local'
Get-ADUser -Filter {(Enabled -eq $true -and UserPrincipalName -notlike '*.local')} -SearchBase $OUpath -SearchScope Subtree -Properties * | foreach-object {
# Create user alias from user account information.
# Various options provided, adjust comment header to select the format you wish
# NOTE THAT "smtp" ON THESE ARE LOWER CASE, MAKING THEM ALIASES, NOT PRIMARY EMAIL ADDRESSES
$alias = ("smtp:{0}{1}$proxydomain" -f $user.GivenName.Trim()[0], ($user.Surname)) #[email protected]
#$alias = ("smtp:{0}{1}{2}$proxydomain" -f ($user.GivenName), ".", ($user.Surname)) #[email protected] WATCH FOR MIDDLE INITIALS! You may need to replace GivenName with SamAccountName or use other trim methods.
#$alias = ("smtp:{0}$proxydomain" -f ($user.GivenName)) #[email protected]
# -match is a regex operator, escape appropriately
if ($_.ProxyAddresses -match [regex]::Escape($alias)) {
Write-Host "Result: $alias value already exists for user $($_.displayname); No action taken."
}
else {
# Now we only have a single variable that needs to be expanded in the string
Set-ADUser -Identity $_.SamAccountName -Add @{proxyAddresses = "$alias"}
Write-Host "Result: Added $alias value to $($_.displayname)"
}
}
The purpose of DTO is that your controller returns an object in your API that is separate from your core entities.
If you have a core entity called Transaction that contains an object of type Currency, it is a bad practice to return Transaction and leak to the outside that your transaction has a type Currency in it, instead you create a DTO and in the DTO format the currency to a String, this way your entities preserved and hidden from what your controller returns. Now your API and core entities can change independently without breaking each other.
The correct order is that services will return model entities, and in your controller, they are mapped to a DTO.
Check out Google Workspace and the 'ways to build' section. You have options to use API and UI toolkits or using low-code/ no-code app options to start building calendar functions into another app.
The problem is that SentenceTransformer relies on the indices present in the Series. Therefore, to eliminate this issue, you need to reset the indices before calling encode(sentences):
I author of CLI/TUI is a utility on Go for managing files in the terminal, supports flexible filters by modification time, size and extensions, and also allows you to quickly navigate through directories, select files and save sets of rules for reuse, and safely move files to the trash.
Did you ever get an answer or solution for this? I am having the same issue. I have a SQL task that the error doesn't propagate to the package it is in, but that package is called by another parent package that the error is still propagating to.
Reasons:
RegEx Blacklisted phrase (3): Did you ever get an answer
There's no module called "RNPermissione", it should be "RNPermissions" (Ending with an S instead of an E). Check the Android/iOS configuration to see if maybe you have something there with this typo
After some testing, I found out that the most upvoted answer isn't good for reusability and is ugly, so I ended up going with one of the other suggested solutions and created this method in my utils:
Phone numbers can get rather long when you have to insert dialing pauses (usually represented by ","), outside vs inside access, access numbers, extensions, confirmation codes (which might include "#" and "*"), etc. The best test of a phone number is actually to dial it or text to it and verify that it is legitimate and request a return tone, text, or click through to a verification site. Yes, 7-15 seems right, if you aren't going to verify it otherwise.
I have decided that instead of asking this question, I should have adjusted my code to remove the need for it entirely. Some comments kindly alluded to this which helped me arrive at this conclusion (thank you). My function is better for it.
In the Group name field, enter a long name, like This is a long group name, so long that it causes a validation error to appear that specifies the field's maximum length.
A validation error should appear that specifies the field's maximum length.
set @v_xml_nest_level_pos = CHARINDEX ('<tsql_stack><frame nest_level', @p_additional_info, 0);
-- nest_level found
if @v_xml_nest_level_pos > 0
begin
-- Replace single quotes with double quotes to make it valid XML
set @xmlString = REPLACE(@p_additional_info, '''', '"');
-- Now cast to XML
set @xml = TRY_CAST(@xmlString AS XML);
-- Extract nest_level as string
set @nest_level_str = @xml.value('(/tsql_stack/frame/@nest_level)[1]', 'NVARCHAR(100)');
-- Try casting to INT safely
SET @nest_level = TRY_CAST(@nest_level_str AS INT);
if @nest_level is null
begin
set @P_SEO_CURR_USER = 1;
end;
if @nest_level is not null
begin
if @nest_level < 1
begin
set @P_SEO_CURR_USER = 1;
end;
end;
end
else -- nest_level not found
begin
set @P_SEO_CURR_USER = 1;
end;
where:
p_additional_info - is field additional_information of fn_get_audit_file, within a loop of which the above code is run.
@P_SEO_CURR_USER is set as default 0, meaning non-top-level SQL. Set to 1 when top-level.
Thank you to all commenters who helped me on this.
I am using flink to consume events from kafka and having a sliding window assigner for 1 hour window slides every 5 mins and then it has to write result in cassandra. I want to understand how this works internally ??
For example we have 10000 events and 5 taskmanagers and let's say each taskmanager gets 2000 event so there will be 5 entries in cassandra or Flink internally will aggregate all the outputs from 5 taskmanager and then create a single entry in cassandra.
The most straightforward approach is to hook into the deletingevent of the User Eloquent model and delete the Sanctum tokens there.
// app/Models/User.php
use Illuminate\Foundation\Auth\User as Authenticatable;
use Laravel\Sanctum\HasApiTokens;
class User extends Authenticatable
{
use HasApiTokens;
protected static function booted(): void
{
self::deleting(function ($user) {
$user->tokens()->delete();
});
}
}
Now your proposed line will work as expected, no need to call $user->tokens()->delete() explicitly anymore:
Note: if you extend the User model with child classes and still want this behavior, you'll want to use static::deleting instead of self::deleting (Understanding Static vs Self in PHP).
If you are using SimpleAuthManager, the user name is defined by simple_auth_manager_users in airflow.cfg. Default is admin:admin meaning a user named "admin" that has the "admin" role. The password is auto-generated and can be found in the webserver's log. See airflow's document here
In case anyone came here as I did while debugging an ASP.NET Web Forms application, what worked for me was properly configuring the project properties Web tab, to start the application on the same URL as the project URL. You can either check the Start URL option and duplicate the project URL in that field, or as I did check Specific Page and enter a forward slash in the field.
It sounds like you didn’t open the correct terminal. Double check it is the correct MSYS2 terminal as the installer places many in there (it is specified in the getting started doc).
If I understand your question correctly, this is what typically happens:
1. T1 modifies x but does not commit - it sets a write intent on x
2. T2 reads x - T2 will block here for T1 to complete
3. T1 commits
4. T2 resumes and reads x for the first time. if it reads x again, it will see the same value.
As far as I know it has sth to do with sensors for example starting a session of running will enable the GPS some others will enable the gyroscope but it is not clear which ones enable what. There is a service I used for this found on rapidAPI that turns the workout to hkWorkoutActivityType I used it to get all the predefined workout types once.
