Re:My experience crafting an interpreter with Rust (2021)
by dhosek on Saturday March 25, 2023 @07:58PM
(#35308917)
I’m curious about the extent to which DOS is a concern for an interpreter like this. I can see it be an issue if you’re exposing a service to external actors, but how big of a concern is it really in this context?
[ Parent
| on HN ]
Re:My experience crafting an interpreter with Rust (2021)
by mastax on Saturday March 25, 2023 @08:56PM
(#35309395)
I'm going to assume that at least one popular interpreter uses a weak hash function for its hashtables/objects, and I haven't ever heard about it being an issue.
Re:My experience crafting an interpreter with Rust (2021)
by loeg on Sunday March 26, 2023 @11:34AM
(#35315564)
It was a big problem for web-facing services in interpreted languages "back in the day." All the dynamic languages basically fixed this problem ten years ago (or 20 years ago for Perl): https://lwn.net/Articles/474912/
If you weren't going to expose clox to the web or other untrusted input (and why would you), it's not a concern.
Re:My experience crafting an interpreter with Rust (2021)
by rstuart4133 on Sunday March 26, 2023 @09:25PM
(#35321801)
I went down that rabbit hole myself after reading the article. This sums up the situation: https://sudonull.com/post/127497-Non-cryptographic-hash-func...