Join devRant
Do all the things like
++ or -- rants, post your own rants, comment on others' rants and build your customized dev avatar
Sign Up
Pipeless API
From the creators of devRant, Pipeless lets you power real-time personalized recommendations and activity feeds using a simple API
Learn More
Feed
All
Post Types
- Rants
- Jokes/Memes
- Questions
- Collabs
- devRant
- Random
- Undefined
Cancel
All
-
Lucky Apple x2500: el atractivo de los símbolos frutales
Los juegos de temática frutal siguen teniendo un lugar especial entre las tragamonedas online gracias a sus gráficos reconocibles y su presentación sencilla. Lucky Apple x2500 de 3 Oaks es una propuesta que utiliza esta estética con una estructura pensada para partidas dinámicas. En https://juga-bet.com/services/... puedes consultar las características del título y descubrir cómo funciona su sistema de giros.
La interfaz está diseñada para mostrar los rodillos, el importe seleccionado y el botón principal de juego de manera clara. Las combinaciones se calculan según las reglas de 3 Oaks, mientras que la tabla de pagos permite conocer el valor de los símbolos y las posibles funciones especiales.
Un aspecto práctico es la posibilidad de ajustar el importe de cada giro. Esto permite configurar la experiencia según las preferencias de cada jugador. También se ofrece una modalidad demo para familiarizarse con los controles y observar el desarrollo de las rondas.
Lucky Apple x2500 está disponible desde el navegador móvil y la aplicación Android de JugaBet. Su temática clásica, sus controles directos y su presentación visual lo convierten en una opción interesante dentro de la categoría de slots.1 -
Can we please stop treating “processing” as a complete explanation?
Request accepted? Sitting in a queue? Waiting for another service? Failed, but nobody updated the status?
Those are very different situations, yet somehow they all get the same spinner.
A timeout makes this even more fun. The client doesn’t know whether the operation failed or whether the response just got lost. Now “try again” might mean “recover gracefully” or “accidentally do it twice.”
For an exchange integration, BYDFi or otherwise, I’d want the docs to explain request IDs, retry behavior, and how to check an operation’s actual state before showing me another happy-path example.
Give the frontend enough information to say something useful. Even “we received your request; don’t resubmit it” beats an endlessly rotating circle.
The spinner is an animation. It should not be the error-handling strategy.
What’s the least helpful API status you’ve had to work around? -
so you want an app that can upload your tiktok videos for you?
well, you'll need to setup a production tiktok app
but to do that you'll need to record a video of the full upload flow working
but to do that you'll need to setup a sandbox tiktok app
but to do that you'll need a tiktok account set to private so you are allowed to send a test example
but to do that you'll need to hunt down an old unused email and password
but to do that you'll need to confirm your test email
you did it! now you can create your private tiktok account, run your full upload flow against the sandbox account and then FINALLY, FINALLY use that recorded video to submit the production app and even then run some bullshit risk of not being approved.
welcome to 2026.
so good that we have oauth standards that are secure and usable, would be very useful in cases like these
🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡🤡1 -
Hot take after sitting on both sides of the table: most candidates who 'bomb' coding interviews don't lack skill, they just never practiced thinking out loud. Rehearse explaining your approach before you type, state the brute force first, then optimize. Talking to a rubber duck works; I ended up building a mock-interview tool (linkjob.ai) because I kept doing this for friends.2
-
Un compañero del trabajo me comentó el otro día que estaba siguiendo partidos en una versión en español y que las cuotas se veían claras. Me dio curiosidad saber cómo estaba organizada la interfaz y qué mercados ofrecía sin complicaciones. Después de revisar las opciones disponibles y cómo se actualizaban las líneas en vivo decidí probar con una apuesta pequeña en un partido de la noche. Todo se liquidó sin problemas y el proceso resultó sencillo de seguir. Quedé satisfecho con lo directo que se sentía manejar las selecciones desde el principio.
-
Why does our subtitle checker refuse markup in a WebVTT file but let the same tags through in an SRT file?
I work on OnePassApps, the team behind our own browser apps. We ship a Subtitle Workbench that reads the subtitle files you already have and tells you what is wrong with them before you export. I measured its format gate today, 12 November 2026, signed out, one build, one browser.
The finding is short: the markup gate is a format rule, not a markup rule. The same tags that stop a WebVTT file are accepted by an SRT file without a complaint.
## The same tags, two formats
I pasted one SRT of 198 characters with three cues. Cue 1 carries a voice tag, cue 2 a bold tag and a colour tag, cue 3 is plain:
```
1
00:00:01,000 --> 00:00:03,000
<v Roger Bingham>Voice tagged line
2
00:00:04,000 --> 00:00:06,000
<b>Bold</b> and <font color="#ff0000">coloured</font>
3
00:00:07,000 --> 00:00:08,000
Plain line
```
The panel read 3 cues - 0:01–0:08, then 3 cues, then No issues. Export SRT was enabled, I pressed it, and the confirmation line said Downloaded SRT and named the pasted file. The file that landed on disk was 198 bytes, and the markup was still there byte for byte - the voice tag, the bold tag and the font colour tag.
Then the same tags in a WebVTT file (221 B) are refused, one message per problem:
```
Error: WebVTT NOTE blocks are unsupported. Export is blocked to preserve their meaning; download the original instead.
Error: Rich WebVTT markup (voice, class, ruby, or inline timestamps) is unsupported. Download the original to preserve it.
```
The header counted 3 cues - 221 B and 3 errors, and the panel added a line of its own: Resolve source errors before exporting. The original file is still available. The NOTE block gets its own sentence, separate from the rich-markup one, and the rich-markup sentence names four kinds by name - voice, class, ruby, or inline timestamps. The voice tag and the ruby tag each produced one of those errors.
The file list said 1 with errors to resolve before export. The export button is named after the format that is selected: with the SRT file loaded the panel showed Export all ZIP and Export SRT, and with the WebVTT file loaded it showed Export all ZIP and Export WebVTT.
## The cue that produced no error at all
The positioned cue was the third case and the interesting one, because it raised nothing. Its WebVTT cue carried line:85% align:center position:50%, and the app printed the cue and its positioning settings on a single line in the cue list: Positioned cue line:85% align:center position:50%. No error was attached to it, even though the file carried three errors in all - one for the NOTE block, one for the voice tag and one for the ruby tag.
The guide's sentence about positioning is conditional - refused when it would lose information - and this cue produced no refusal and no error. I did not read the exported bytes for it, so I am not claiming what it contains. What I can say is what the panel did: it raised nothing, and it showed the positioning joined to the cue text.
## Nothing left the tab
I captured every request the tab made at the browser's own network layer, before I typed anything. The only requests were GETs and OPTIONS to our own app, API and auth hosts - the app's runtime and feature-flag reads - and there was no POST and no request body anywhere.
## What I did not test
One build, one browser, one signed-out session, one day, 12 November 2026. I pasted subtitle text rather than using the file picker, so the 5 MB per file and 20 files per batch limits are the app's own wording and not something I measured against a file of that size. The rich-markup error names four kinds and I exercised voice and ruby, so class and inline timestamps are untested.
The full write-up, with the timings and the export steps, is on our DEV Community account; this post is the format gate on its own. Our profile lists the nine apps we ship and carries the link to the tool.1 -
