Recent submissions

Comment · 1 day ago
Shipping a bug like "WHERE 4 = 5" returning rows says it all to me
Comment · 6 days ago
That's a nice idea. I have been using DataGrip in the past, it is heavy. Then started to take a look at DBPro, it's much lighter in terms of UI, but again costs a lot of RAM due to electron usage and I have noticed a few issues there in terms of performance and capabilities.

Overall, from what I've tried, it's a nice project. It's a bit unfair to compare, but it lacks a things I'd need to switch from DataGrip. I hope this doesn't sound too harsh:

The optimiser itself is AI. Which I don't like, rewrites could be proved from the catalog (an unused unique LEFT JOIN, DISTINCT over a key, NOT IN to NOT EXISTS, OR to UNION). A model can suggest a rewrite, but it can't guarantee the result set is the same.

Also vim mode is editor only, for me its important to have the whole app keyboard native

The formatter is basic compared with DataGrip's code-style settings

DataGrip can step through PL/pgSQL and Oracle PL/SQL code. As far as I can tell TablePro has nothing similar, and in my work that's a blocker.

Comment · 3 months ago
Hi @khurs, I think that it is related how Tor and Safari Lockdown handles SW. If SW usage is blocked this is incompatible with current architecture. I will think what we can do about it and come back to you. Meanwhile open to PR's in helping to resolve it!
Comment · 3 months ago
It would also be cool if such bang lists were supported by websites officially, allowing a fast navigation
Comment · 3 months ago
Exactly, there’s so much potential beyond search. Bangs could act as general browser-side shortcuts to deep links, so something like !ghsettings could take you directly to GitHub’s settings instead of making you do this manual navigation through the site you are visiting
Comment · 3 months ago
Yes, I agree, thanks for testing on Firefox. I use Zen on a daily basis, it is Firefox based and my result is also interestingly around (0.56ms median, 0.59ms mean, 0.78ms P95, 0.94ms p99, min 0.48ms, max 1.3ms) for prefix bangs. I think that the differences come in how each browser may have implemented SW. On benchmarking suite specifically the tail can also come from scheduling, IPC, extensions, timer precision, or other browser work
Comment · 8 months ago

Live from the official Hacker News API

A modern reader for Hacker News · SvelteKit + Skeleton