Tag: WebDev

I have a docker setup with Laravel Application that works fine on local Mac however, I am facing issues with the deployment on Ubuntu server 24.0. All my paths are correct All my volumes are mapped correctly I checked the file location which exists I have correctly pointed the root for NGINX i.e. both the NGINX and PHP-FPM has the same folder structure /var/www/api/public User & Group ID matches the host OS i.e. 1000:1000 in all containers Permissions: api: 755 api/storage : 775 api/bootstrap/cache: 775 Where could be the issue? I have created a git repo for this setup here. The folder structure I have is: Project Folder |- dock

Some binary files, such as the .WAV files I'm working with, contain text strings. Some text editors, such as BBEdit are able to open the file, showing the text strings along with gibberish like "˚x‹(˝ÿËÔ§Û$˜ÿ¸,¿" for the binary data. When I try a naive approach such as "fs::read_to_string()" I get the understandable error "stream did not contain valid UTF-8". Is there a straight forward way to just pluck out the text strings I'm looking for?

I have an extensive background in several languages (java, C, python, perl, etc) and am starting to work more with Typescript (and React). I see lots of code which uses an assignment form of method declaration, such as: const handleChangeComplete = (color: { hex: any; }) => { setBackground(color.hex); onChangeComplete(color.hex); } Obviously this is a perfectly valid syntax, and so is: function handleChangeComplete(color: { hex: any; }) { setBackground(color.hex); onChangeComplete(color.hex); } Even though both forms work in the contexts I've encountered, the first form seems more prevalent. Are they not completely equivalent? Is there a particular reason to prefer one form over the other? So far I prefer the function syntax. It is more concise, and more familiar to devs experienced in other languages. As a reader, to me it seems to more directly expres

The problems I'm having: s3 tables maintenance fails, cannot access s3 tables in Athena s3 table maintenance fails: I make my s3 tables in a glue etl job, they show up in my s3 table bucket, but if i click on the table and go to the maintenance section, it says that maintenance failed and gives the following error: Encountered Iceberg validation exception when trying to read table. Ensure that your table is readable, adheres to the Iceberg specification, and contains only S3 paths that begin with your S3 Table alias. Athena error: My s3 tables appear in Athena, but i get the following when trying to query the table, PERMISSION_DENIED: null (Service: S3, Status Code: 403) Below is the configuration I've set for my glue etl notebook (using this [aws community answer](https://repost.aws/questions/QUwVF6859STAC

My laptop is ASUS Vivobook S14. System is UEFI, non-CSM. I have 2 USB that I did this test on : 16GB TOSHIBA, 8GB Lexar. The OS that I've been meaning to boot is the newest Tails OS (7.4.2) that's downloaded from the "Windows Installation" window from the official tails.org website itself. Both my devices has been flashed with : - Rufus, with the settings of GPT partition, and UEFI (non-CSM)* - BalenaEtcher And yes, it's fully cleaned before it was flashed. I tried both left side and right side port of the laptop. I've disabled secure boot and fast boot. There's no UEFI firmware settings user nor admin password. Both devices works on other machines, so clearly, it's the laptop's issue. It does the same thing when flashed with other OS including arch installer. I've tried the laptop's boot menu first. The USB devices did not show up at all. I tried the windows startup settings "Use other devices" menu. It also doesn't appear

I need some advice from C language gurus (not necessarily C development gurus). Earlier, I was fiddling with a new C23 feature typeof (I know it's been there for GCC/Clang for a while, but from the standard point of view, it's new). I thought anonymous struct types with the same exact members were compatible with each other (https://en.cppreference.com/w/c/language/compatible_type.html), so I thought maybe the following code would work; it didn't. so_question.c #include <stdio.h> struct { int x; double y; } __my_var1__; struct { int x; double y; } __my_var2__; void foo(typeof(__my_var2__) aaa) { printf("%d %lf\n", aaa.x, aaa.y); } int main() { typeof(__my_var1__) bbb = { 1, 2.0 }; foo(bbb); return 0; } The exact error message from Clang 18.1.3 wa

I've done some research, but couldn't find a reasonable answer to my problem. It's fairly simple. I have two Python dictionaries, d1 and d2. Let them be: d1 = {'a':1 , 'b':2} d2 = {'a':3 , 'b':4} Those dicts would be united as a new one, say new_dict: new_dict = {1:{'a':1 , 'b':2} , 2:{'a':3 , 'b':4}} Would that be possible? How? I know I can do this: new_dict = {1:d1 , 2:d2} But this doesn't seem to be smart/efficient, especially when the number of dictionaries to be appended to new_dict grow (d1, d2, d3, ..., dn). Thanks.

Problem A Microsoft moderator has confirmed that the semantic ranker not working well with vector search is a "documented limitation": "If vector similarity dominates the top results, the semantic reranker may not have sufficient textual context to apply meaningful reranking or boosting." "This is not a bug, but a documented limitation." Our testing with 8,068 real user queries confirms this — enabling the semantic ranker on hybrid search with text-embedding-3-large drops Hit@1 by 27 percentage points. Setup Azure AI Search, ~254K documents, 8 languages Embeddings: text-embedding-3-large (1536 dimensions) Evaluation: 8,068 real user queri

This is super weird and most likely a clang bug, but I want to be sure first, to avoid spamming devs with invalid issues. I've distilled my use case to a simple example, where clang produces the error, that I'm pretty sure should not be there. Weirdly enough, if I tinker with the number of alternatives the issue miraculously disappears. Other compilers don't emit any errors, as well as clang with libstdc++ (live on godbolt.com). // Set 1/2/3 to pick different use cases #define USE_CASE 1 #include <vector> #include <variant> struct Array; struct A1 { auto operator<=>(const A1 &) const = default; }; struct A2 { auto operator<=>(const A2 &) const = default; }; struct A3 { auto operator<=>(const A3 &) const = default; }; struct A4 { auto operator<=>(const A4 &) const = default; }; struct A5 { auto operator&

std::initializer_list is baked into the language. If std::tuple had the same treatment, std::initializer_list would become "speacial" case for std::tuple where all types are the same. I wanted to simplify some code that calls the same function with different set or params: auto val = QueryRegString(HKEY_LOCAL_MACHINE, L"Windows Kits\\Installed Roots", L"KitsRoot10"); /// work with val val = QueryRegString(HKEY_LOCAL_MACHINE, L"Windows Kits\\Installed Roots", L"KitsRoot81"); /// work with val val = QueryRegString(HKEY_LOCAL_MACHINE, L"Microsoft SDKs\\Windows\\v7.1", L"InstallationFolder"); /// work with val I wanted to rewrite it like that: const auto regs[] = { {HKEY_LOCAL_MACHINE, L"Windows Kits\\Installed Roots", L"