WebKit VS flex

Since 2023 Chrome announced the View Transition API, and it looks like Safari is also going to support it soon.

One heap per type.

Here’s an allocator optimized for that use case.

https://github.com/WebKit/WebKit/blob/main/Source/bmalloc/li...

To use this in Bun, you’d have to start Bun with the environment variable “BUN_JSC_useDollarVM=1” and then $vm.createBuiltin(mySourceCodeString)

When using this intrinsic, if any of the arguments are incorrect or it cannot otherwise enable it, the entire process will probably crash. In debug builds of JSC it will have a nicer assertion failure but that is not enabled in release builds

Example code: https://github.com/WebKit/WebKit/blob/17351231b4dedb62d81721...

also happy to answer any questions about Bun

Ah, good, let me introduce you to the wonderful world of the Chrome Devtools Protocol! (fka Chrome Remote Debugging Protocol)

I love this API for almost everything browser related. I built my RBI product atop this (BrowserBox: https://dosyago.com), and I think it's a drastically underrated API.

Also, it works out of the box in Edge, Brave, Chromium, and many parts of CRDP are supported by Firefox and Safari^1

1: See for example: https://github.com/WebKit/webkit/tree/main/Source/JavaScript...

No, it's unrelated.

https://github.com/WebKit/WebKit/commit/064df1a9f395f8c6e32c...

It's being worked on now: https://github.com/WebKit/WebKit/pull/17320

It is different. The cross-site navigation flag is a couple of years old. It was enabled by default for iOS in November 2018 for example https://github.com/WebKit/WebKit/commit/e191fc8c412850cb9fd0...

> That's excessive scope creep. Adding site-specific workarounds for some sites feels uncomfortable.

Not to Google and its fellow corporations apparently.

https://news.ycombinator.com/item?id=29707078

https://github.com/WebKit/WebKit/blob/main/Source/WebCore/pa...

They have site-specific fixes for their own sites. Why can obviously apply the exact same strategy to dealing with every single website out there. If a website is broken or generally annoying to use, just fix it by providing a site specific version of the browser.

> Who decides what websites get "fixed", and how?

Whoever develops the browser or its extensions. Arguably the whole of uBlock Origin and its filter lists is just a database of site specific fixes. If people can maintain an extremely huge list of advertisers and blockers for every website out there, surely they can maintain something like this too.

> Also, remember how Mozilla is funded.

I remember Mozilla has about a billion dollars in the bank. Who cares about their Google funding? I doubt they're gonna drop them anyway. I bet they pay them just to ward off risk of antitrust lawsuits.

I am using Flex++, which is Flex for C++ and I am having trouble setting the input file. Flex++ uses the FlexLexer class provided in FlexLexer.h to create the lexer object(https://github.com/westes/flex/blob/master/src/FlexLexer.h). In my main function I have

Further applications of DFA and NFA: lex or flex, yacc or bison, and POE :-)

How about writing a programming language using Flex and Bison? There are lots of good tutorials and examples out there.

Some lexer generators (notably Flex) take input from a file handle by default. While you can always read a file into a string before passing it to the generated lexer, this is not seen as "the best" since you have to read in all the data into memory, which can be a lot.

Ages ago, I loved writing domain-specific toy languages, and almost always used flex to generate lexers and GNU bison to generate the parser. I've begun a new toy project and I don't think those two will cut it this time, so I'm looking for other tools that integrate well with C++.

I don't understand :c don't lexers like Flex work off of regex rules? Isn't this the correct first step to parse it?

I wrote a C compiler using flex [1] and bison [2]. The glue between them is a bit hacky.

At some point ANTLR [3] looked promising, but these days I'd probably write a lexer and recursive descent parser by hand, then generate LLVM IR.

[1] https://github.com/westes/flex

github.com/westes/flex/releases

Un commento in flex, uno storico software, tanto per mostrare che anche i migliori fanno le cose alla buona.