#pullrequest — Public Fediverse posts

Our next #JCON2026 session is live: 'Ten Angry #Comments Later: #Stories About #Code Reviews, Teams, and Yesterday’s Code' with Johannes Rabauer

You open a #pullrequest. Ten comments later, you’re wondering whether the code …

Grab your coffee and hit play: https://youtu.be/N5hGP4qQ30I

Our next #JCON2026 session is live: 'Ten Angry #Comments Later: #Stories About #Code Reviews, Teams, and Yesterday’s Code' with Johannes Rabauer

You open a #pullrequest. Ten comments later, you’re wondering whether the code …

Grab your coffee and hit play: https://youtu.be/N5hGP4qQ30I

@cazabon #Codeberg algorithm: If you type the same comment into the #pullrequest #PR descriptor field a few times you get rate limited: cannot do the PR. This was for preventing bot use. But I was using the same PR comment *manually* for identical operations. I didn't know this rule & was hampered til they told me. So when they say no #LLM code I am freaked out: are they going to use similar bad automation & unfairly break my workflow? #Cooperate not #cooperhate #codebergalternatives

@cazabon #Codeberg algorithm: If you type the same comment into the #pullrequest #PR descriptor field a few times you get rate limited: cannot do the PR. This was for preventing bot use. But I was using the same PR comment *manually* for identical operations. I didn't know this rule & was hampered til they told me. So when they say no #LLM code I am freaked out: are they going to use similar bad automation & unfairly break my workflow? #Cooperate not #cooperhate #codebergalternatives

🛠️🌳 Forgejo 16.0 a fost lansat: Platforma de dezvoltare aduce notificări granulare și recenzii mult mai bune pentru PR-uriForgejo, platforma lightweight și auto-găzduită (self-hosted) dedicată stocării depozitelor Git și colaborării pe proiecte software (un fork comunitar excelent al Gitea), a primit o actualizare majoră: Forgejo 16.0. Această versiune se concentrează pe reducerea „zgomotului” din fluxul zilnic de lucru, pe îmbunătățirea procesului de recenzie a codului (code review) și pe o securitate sporită. Iată care sunt cele mai importante noutăți din Forgejo 16.0:🔹 Notificări granulare pe depozite (Granular Watch) Una dintre cele mai mari probleme ale dezvoltatorilor pe platformele de colaborare este volumul uriaș de e-mailuri și alerte primite.În Forgejo 16.0, opțiunea de „urmărire” (watch) a unui depozit a fost complet reproiectată. Utilizatorii pot alege acum exact ce tip de activități vor să urmărească: pot bifa doar raportările de probleme (Issues), doar cererile de fuziune (Pull Requests), doar lansările noi (Releases) sau orice combinație dorită între acestea, evitând astfel alertele inutile.Totodată, arhitectura din spate (notification backend) a fost rescrisă pentru o performanță superioară.🔹 Recenzii de cod superioare și comentarii pe mai multe rânduriProcesul de evaluare a Pull Request-urilor (PR) a primit îmbunătățiri extrem de utile, rezolvând probleme vechi legate de poziționarea comentariilor:Comentarii pe mai multe linii: Recenzorii pot selecta acum o întreagă secțiune de cod din diferențe (diff) ținând apăsată tasta Shift pentru a atașa un singur comentariu pe mai multe linii modificate. Listă de commit-uri redesenată: Pagina de PR are acum o listă de commit-uri cu un layout mult mai curat și adaptabil la ecrane de diverse dimensiuni. 🔹 Monitorizarea în timp real a migrării depozitelorDacă decizi să îți muți proiectele de pe GitHub sau alte platforme către instanța ta Forgejo, procesul este acum mult mai transparent:Ecranul de migrare afișează numărul de probleme (issues) și PR-uri transferate pe măsură ce acestea sunt importate în loturi.Pentru a nu aglomera interfața, această bară de progres detaliată apare automat pentru migrările care conțin cel puțin 45 de elemente. 🔹 Optimizări pentru Forgejo ActionsSistemul CI/CD integrat evaluează acum anumite condiții de flux de lucru (workflow) înainte de a trimite joburile către un executant (Forgejo Runner), economisind resurse prețioase de calcul prin evitarea execuțiilor redundante.Fluxurile de lucru finalizate, împreună cu fișierele rezultate (artifacts) și logurile lor, pot fi acum șterse direct din interfața web sau prin intermediul API-ului HTTP. 🔹 Îmbunătățiri de securitate și integritate GitForgejo verifică acum automat obiectele Git primite pentru a detecta inconsecvențele, scăzând drastic riscul de corupere a depozitelor.S-au eliminat fișierele de exemplu (example hook files) din depozitele interne și s-au mutat scripturile automate globale (hooks) într-o locație centralizată și mult mai sigură.Atenție pentru administratori: Autentificarea prin proxy invers (Reverse Proxy) necesită acum configurarea explicită a rețelelor sau adreselor IP de încredere din care provine traficul, sporind siguranța accesului. Forgejo 16.0 se dovedește a fi o actualizare extrem de solidă, rafinând instrumentele zilnice de dezvoltare și oferind administratorilor o platformă Git auto-găzduită tot mai robustă, sigură și plăcută în utilizare. #OpenSource #Forgejo #Forgejo16 #GitForge #SelfHosted #Git #PullRequest #CI_CD #TechNews #Linuxiac

As an Indiekit supporter and user, I’m really appreciative of Paul for maintaining this open source project. After a short hiatus, he’s back to review pull requests and issues!

View full note: https://burgeonlab.com/notes/2026/0710-0120

#indiekit #opensource #foss #pullrequest

As an Indiekit supporter and user, I’m really appreciative of Paul for maintaining this open source project. After a short hiatus, he’s back to review pull requests and issues!

View full note: https://burgeonlab.com/notes/2026/0710-0120

#indiekit #opensource #foss #pullrequest

SOLUTION to above issue: #codeberg blocks pull requests and merges based on use of the same title or an unchanged title as this is used as a flag of spam... So this is why my PRs andmerges returned the error shown above... I was using no, or the same, title, and this triggered the block. This was kindly explained on an issue I raised. Excellent help from Codeberg. I feel like I have gotten out of the maze and can work again. #pullrequest #git #gitmerge

SOLUTION to above issue: #codeberg blocks pull requests and merges based on use of the same title or an unchanged title as this is used as a flag of spam... So this is why my PRs andmerges returned the error shown above... I was using no, or the same, title, and this triggered the block. This was kindly explained on an issue I raised. Excellent help from Codeberg. I feel like I have gotten out of the maze and can work again. #pullrequest #git #gitmerge

Shadow AI ve vývoji začne být problém ve chvíli, kdy do pull requestu přijde změna navržená přes soukromý AI účet a nikdo už nedoloží, jaký kód, logy nebo interní dotaz model viděl. Samotný zákaz nestačí; dohled musí být v nástroji, kontrole kódu a CI.

Shadow AI ve vývoji začne být problém ve chvíli, kdy do pull requestu přijde změna navržená přes soukromý AI účet a nikdo už nedoloží, jaký kód, logy nebo interní dotaz model viděl. Samotný zákaz nestačí; dohled musí být v nástroji, kontrole kódu a CI.

@raspimidihub

https://github.com/wamdam/raspimidihub/pull/4

#MIDI #PullRequest #LinuxAudio #RaspiMIDIHub

@raspimidihub

https://github.com/wamdam/raspimidihub/pull/4

#MIDI #PullRequest #LinuxAudio #RaspiMIDIHub

RE: https://androiddev.social/@luboganev/116578211442808240

So the thing about #PR review. It's an alternative / adjunct to #PairProgramming.

So if a colleague sends you something like that it's absolutely fine to book a meeting with them and have them walk through the #PullRequest with you.

It's not uncommon for the meeting to be cut short because during the explanation they realise they made a mistake, and it needs more work.

RE: https://androiddev.social/@luboganev/116578211442808240

So the thing about #PR review. It's an alternative / adjunct to #PairProgramming.

So if a colleague sends you something like that it's absolutely fine to book a meeting with them and have them walk through the #PullRequest with you.

It's not uncommon for the meeting to be cut short because during the explanation they realise they made a mistake, and it needs more work.

Getting a bit sick of fielding LLM PRs. I understand why many projects have a no-LLM policy. I'm generally fine getting good PRs, LLM-aided or not, but the poor ones are just complete wastes of time. They all read the same, good or not, so I have to spend quite a bit of time sussing through it to determine which ones are useless or not. They're all "bug fixes" or "security improvements".

No, it does not make sense for an CLI argument parsing library to scan through argv to make sure it doesn't contain null pointers. That's ridiculous. No, it does not make sense for a CLI argument parsing library to defend against an input argument size overflowing size_t. When the fuck would that ever happen?

The PRs that actually fix a missed fallthrough causing undefined behavior look exactly the same as the ones that defend against somebody passing a 4GiB string as a CLI argument. They all have descriptions full of nothing but fluff, they all add much more code than they remove, they all bloat the code with more abstractions and indirections, and they all take way longer to review than the equivalent human-written code.

I'm even not against getting LLM contributions, but I'm getting sick of the fucking spam, and I'm losing patience with figuring out the difference between the two.

#LLM #LLMs #LLMSlop #AI #Spam #Code #Programming #PR #PullRequests #PullRequest

Getting a bit sick of fielding LLM PRs. I understand why many projects have a no-LLM policy. I'm generally fine getting good PRs, LLM-aided or not, but the poor ones are just complete wastes of time. They all read the same, good or not, so I have to spend quite a bit of time sussing through it to determine which ones are useless or not. They're all "bug fixes" or "security improvements".

No, it does not make sense for an CLI argument parsing library to scan through argv to make sure it doesn't contain null pointers. That's ridiculous. No, it does not make sense for a CLI argument parsing library to defend against an input argument size overflowing size_t. When the fuck would that ever happen?

The PRs that actually fix a missed fallthrough causing undefined behavior look exactly the same as the ones that defend against somebody passing a 4GiB string as a CLI argument. They all have descriptions full of nothing but fluff, they all add much more code than they remove, they all bloat the code with more abstractions and indirections, and they all take way longer to review than the equivalent human-written code.

I'm even not against getting LLM contributions, but I'm getting sick of the fucking spam, and I'm losing patience with figuring out the difference between the two.

#LLM #LLMs #LLMSlop #AI #Spam #Code #Programming #PR #PullRequests #PullRequest

AI code review zvládá první průchod — shrnout diff, najít rutinní vzory a upozornit na chybějící test. Jako náhrada lidského review ale neobstojí: data ukazují vyšší abandonment a často nízký poměr signálu k šumu. Nasazení proto stojí na pravidlech, úzkém mandátu a měření, ne na výběru dodavatele.

AI code review zvládá první průchod — shrnout diff, najít rutinní vzory a upozornit na chybějící test. Jako náhrada lidského review ale neobstojí: data ukazují vyšší abandonment a často nízký poměr signálu k šumu. Nasazení proto stojí na pravidlech, úzkém mandátu a měření, ne na výběru dodavatele.

Good :blobnomcookie:

#opensource #pullrequest #GitHub #Dependabot

Good :blobnomcookie:

#opensource #pullrequest #GitHub #Dependabot

Vibe-Coder entdecken die Langsamkeit für sich

Stellt euch vor, ihr bekommt einen Pull-Request mit einem Patch für eure Video-Library, der in Assembly-Code geschrieben und dann rund 1,5 bis 2,7 mal langsamer als der C-Code ist.

Ist den VideoLAN-Leuten mit ihrer AV1-Library dav1d so passiert.

War natürlich Vibe-Coding und der Nutznießer (Ersteller ist er ja nicht) hat den Code selbst nicht mal getestet (mangels passender Hardware), geschweige denn einen Benchmark gefahren (weil Vibe-Coder aka dumm).

Um wie viel wollen wir wetten, dass da draußen tagtäglich gut laufender Code durch anderen ersetzt wird, der mehr Ressourcen und Kosten frisst, zusätzlich zu den steigenden AI-Kosten? Aber das wird toll!

🔗

#shortpost #blogpost #blog #AV1 #AI #Vibe-Coding #PullRequest #VideoLAN #Claude #Benchmark #Ressourcen #Langsamkeit #Kosten

Vibe-Coder entdecken die Langsamkeit für sich

Stellt euch vor, ihr bekommt einen Pull-Request mit einem Patch für eure Video-Library, der in Assembly-Code geschrieben und dann rund 1,5 bis 2,7 mal langsamer als der C-Code ist.

Ist den VideoLAN-Leuten mit ihrer AV1-Library dav1d so passiert.

War natürlich Vibe-Coding und der Nutznießer (Ersteller ist er ja nicht) hat den Code selbst nicht mal getestet (mangels passender Hardware), geschweige denn einen Benchmark gefahren (weil Vibe-Coder aka dumm).

Um wie viel wollen wir wetten, dass da draußen tagtäglich gut laufender Code durch anderen ersetzt wird, der mehr Ressourcen und Kosten frisst, zusätzlich zu den steigenden AI-Kosten? Aber das wird toll!

🔗

#shortpost #blogpost #blog #AV1 #AI #Vibe-Coding #PullRequest #VideoLAN #Claude #Benchmark #Ressourcen #Langsamkeit #Kosten

Ever since setting up #Indiekit for adding #Micropub capabilities to my #Hugo blog, I have only used its web UI to post notes, photos, and likes (haven’t tried other clients yet).

The interface is not bad, but one thing that that bugs me is how links are handled. I am so used to directly pasting in URLs into selected text...

View full note: https://burgeonlab.com/notes/2026/0430-104551

Syndicated via https://getindiekit.com

This is post 37 of #100DaysToOffload

#foss #indieweb #opensource #pullrequest #markdown

Ever since setting up #Indiekit for adding #Micropub capabilities to my #Hugo blog, I have only used its web UI to post notes, photos, and likes (haven’t tried other clients yet).

The interface is not bad, but one thing that that bugs me is how links are handled. I am so used to directly pasting in URLs into selected text...

View full note: https://burgeonlab.com/notes/2026/0430-104551

Syndicated via https://getindiekit.com

This is post 37 of #100DaysToOffload

#foss #indieweb #opensource #pullrequest #markdown

💬✨ In a groundbreaking 4-minute treatise, Jake Worth reveals the ancient secrets of pull request commenting: If it's all nitpicks and nonblocking, approve it anyway! 🎉 Who knew that caring about code could be as simple as leaving comments and hitting the approve button simultaneously? 🙃🔍
https://www.jakeworth.com/posts/on-commenting-and-approving-pull-requests/ #pullrequest #commenting #codingtips #softwaredevelopment #codequality #HackerNews #ngated

💬✨ In a groundbreaking 4-minute treatise, Jake Worth reveals the ancient secrets of pull request commenting: If it's all nitpicks and nonblocking, approve it anyway! 🎉 Who knew that caring about code could be as simple as leaving comments and hitting the approve button simultaneously? 🙃🔍
https://www.jakeworth.com/posts/on-commenting-and-approving-pull-requests/ #pullrequest #commenting #codingtips #softwaredevelopment #codequality #HackerNews #ngated

TurboQuant model weight compression now graces #Llamacpp, but only if you speak fluent Metal! 🏋️‍♂️ Meanwhile, everyone else waits for TheTom to bless us with a #CUDA port, assuming he ever emerges from the GitHub labyrinth of Pull Request 45. How many engineers does it take to compress a llama? 🤔
https://github.com/TheTom/llama-cpp-turboquant/pull/45 #TurboQuant #Metal #PullRequest #HackerNews #ngated

TurboQuant model weight compression now graces #Llamacpp, but only if you speak fluent Metal! 🏋️‍♂️ Meanwhile, everyone else waits for TheTom to bless us with a #CUDA port, assuming he ever emerges from the GitHub labyrinth of Pull Request 45. How many engineers does it take to compress a llama? 🤔
https://github.com/TheTom/llama-cpp-turboquant/pull/45 #TurboQuant #Metal #PullRequest #HackerNews #ngated

🎉 Oh joy, another #C++ #backend for OCaml! Because what the world really needs is more ways to make the already impenetrable C++ even more unbearable! 🙄 Congrats to the brave souls ready to venture into the abyss of Pull Request 14701, where #GitHub buzzwords go to die. 🚀
https://github.com/ocaml/ocaml/pull/14701 #OCaml #PullRequest #DeveloperHumor #TechCommunity #HackerNews #ngated

🎉 Oh joy, another #C++ #backend for OCaml! Because what the world really needs is more ways to make the already impenetrable C++ even more unbearable! 🙄 Congrats to the brave souls ready to venture into the abyss of Pull Request 14701, where #GitHub buzzwords go to die. 🚀
https://github.com/ocaml/ocaml/pull/14701 #OCaml #PullRequest #DeveloperHumor #TechCommunity #HackerNews #ngated

idea regarding #code #review:
intentionally adding #bugs into #code to check if the #reviewer discovers them.
if the reviewer does not find them, then their #review is questionable.
of course the added bugs should be removed before merging ;-)

I am wondering if a concept like this already exists ...
#codequality #softwareengineering #quality #software #git #pullrequest #mergerequest

idea regarding #code #review:
intentionally adding #bugs into #code to check if the #reviewer discovers them.
if the reviewer does not find them, then their #review is questionable.
of course the added bugs should be removed before merging ;-)

I am wondering if a concept like this already exists ...
#codequality #softwareengineering #quality #software #git #pullrequest #mergerequest

#GitHub backs down, kills #GirhubCopilot #PR ‘tips’ after backlash
Following backlash from developers, GitHub has removed #Copilot's ability to stick #ads - what it calls "tips" - into any #pullrequest that invokes its name.
GitHub VP of developer relations Martin Woodward explained in a post on X later on Monday that Copilot inserting ads into PRs isn't actually new behavior - it's been doing so in the ones it creates for a while.
https://www.theregister.com/2026/03/30/github_copilot_ads_pull_requests/

#GitHub backs down, kills #GirhubCopilot #PR ‘tips’ after backlash
Following backlash from developers, GitHub has removed #Copilot's ability to stick #ads - what it calls "tips" - into any #pullrequest that invokes its name.
GitHub VP of developer relations Martin Woodward explained in a post on X later on Monday that Copilot inserting ads into PRs isn't actually new behavior - it's been doing so in the ones it creates for a while.
https://www.theregister.com/2026/03/30/github_copilot_ads_pull_requests/

🚀 Ah, another riveting tale of a techie's underwater adventure with fork() 🌊, where the "thrilling" climax is a pull request awaiting the hallowed review. 📜 The #suspense is palpable as we wait with bated breath for the groundbreaking revelation... after next week's staging tests, of course. 😂
https://tech.daniellbastos.com.br/posts/what-fork-actually-copies/ #techadventure #fork #pullrequest #stagingtests #HackerNews #ngated

🚀 Ah, another riveting tale of a techie's underwater adventure with fork() 🌊, where the "thrilling" climax is a pull request awaiting the hallowed review. 📜 The #suspense is palpable as we wait with bated breath for the groundbreaking revelation... after next week's staging tests, of course. 😂
https://tech.daniellbastos.com.br/posts/what-fork-actually-copies/ #techadventure #fork #pullrequest #stagingtests #HackerNews #ngated

Feature request: gh pr create --fill should use PR template if available...

#GitHub #FeatureRequest #PullRequest #QualityOfLifeBugFix

Feature request: gh pr create --fill should use PR template if available...

#GitHub #FeatureRequest #PullRequest #QualityOfLifeBugFix

ターミナルからPRを監視するTUIツール「prtop」を作った
https://qiita.com/kitamura-yzrh/items/243ecdd24e1c08ede528?utm_campaign=popular_items&utm_medium=feed&utm_source=popular_items

#qiita #Rust #pullrequest #tui #ClaudeCode

Excited for #JCON EUROPE 2026? See Johannes Rabauer at #JCON2026 in Cologne talking about 'Ten Angry #Comments Later: Stories About #Code Reviews, #Teams, and Yesterday’s Code'

You open a #pullrequest. Ten comments later, you’re wondering …

🎟️ 2026.europe.jcon.one/tickets
Free for #JUG members

Excited for #JCON EUROPE 2026? See Johannes Rabauer at #JCON2026 in Cologne talking about 'Ten Angry #Comments Later: Stories About #Code Reviews, #Teams, and Yesterday’s Code'

You open a #pullrequest. Ten comments later, you’re wondering …

🎟️ 2026.europe.jcon.one/tickets
Free for #JUG members

I've been writing this piece on and off for around five years now. I've rewritten it a bunch of times. I think it is okay to publish now.

https://www.ottorask.com/blog/pull-requests-quality-theater-with-questionable-results

#PullRequest #ContinuousIntegration #ContinuousDelivery #Software #SoftwareDevelopment #ProductDevelopment #ExtremeProgramming #PairProgramming #MobProgramming #Quality #QualityControl #QualityAssurance

I've been writing this piece on and off for around five years now. I've rewritten it a bunch of times. I think it is okay to publish now.

https://www.ottorask.com/blog/pull-requests-quality-theater-with-questionable-results

#PullRequest #ContinuousIntegration #ContinuousDelivery #Software #SoftwareDevelopment #ProductDevelopment #ExtremeProgramming #PairProgramming #MobProgramming #Quality #QualityControl #QualityAssurance

RE: https://social.heise.de/@heiseonlineenglish/116204871378593268

My preferred solution here is CI/CD without code reviews, with 100% code responsibility and AI disclosure per merge request. Anyone using AI must first disclose this and then explain how it has been used. If the code causes a failure, I, and I alone, am liable for it, even at 23:55.

#ai #llm #FckAI #FckLLMs #Amazon #MergeRequest #PullRequest #CICD #SoftwareDevelopment

Wow, found a great use case for GitHub Copilot. It can help Dependabot finish update bumps that require code changes!

#github #copilot #dependabot #update #upgrade #ai #migration #pullrequest #automation #developer #code #agents

Wow, found a great use case for GitHub Copilot. It can help Dependabot finish update bumps that require code changes!

#github #copilot #dependabot #update #upgrade #ai #migration #pullrequest #automation #developer #code #agents

What would happen if #PullRequest #feedback ACTUALLY hit you?
Johannes Rabauer put it to the test!

Catch his session “Ten Angry Comments Later: Stories About #Code Reviews, Teams, and Yesterday’s Code” at #JCON2026

🎟️https://2026.europe.jcon.one/tickets
#JCON #Tech