Fr0sT.Brutal

You can create a visual presentation - show support complexity (and thus productivity, possibility of errors, etc) with old code and shorter new one, list bugs that could've been avoided with new arch, benchmarks that show speedup if speed is of value... IOW, try to sell your boss an upgrade 🙂 Coders are weak sellers usually but it's the only way

My point was that updating OpenSSL libs is far not so painful than updating to a newer compiler or env (unless OpenSSL drops support of old env's).

Depends. As for me, your own is your own and any 3rd party service is just another link in the chain that you can't control. For the quick start they're nice though

That's weird. Try to trace FastMM sources where the messagebox is shown

Riding Firebirds 🙂

Could be useful // Prepare string value for FastReport Variable assignment // If s is empty, return Null (FR dislikes empty ('') vars) // If s is one-line string (without CRLF) it must be quoted // If s is multiline, no change is required function FRStrVar(const s: string): Variant; begin if s = '' then Result := Null else if Pos(sLineBreak, s) = 0 then Result := QuotedStr(s) else Result := s; end;

Well, probably older OpenSSL support should be dropped then. Unless they abandon some env's (like Win7) I doubt anyone wants to stick to older version

Not only this function... just think about all these dd-mm-yyyy patterns that will break

DPKs almost do not change across versions, it's the *PROJ that cause real pain... I believe most DPKs could be easily cloned 1:1

@TC Just remember the streamed data could be broken into different TCP packets at literally any place. So even converting a chunk to a string is not reliable unless you use 1-byte encodings (no, UTF-8 is not one of them). Otherwise you must use some kind of concatenating buffer/stream which makes TBytes application useless

Any example when it could be useful?

Yeah, I looked at it yesterday and was shocked. Probably that's because the author made completely separate engine? Updating existing OpenSSL engine to new version shouldn't touch so much code.

https://markwschaefer.medium.com/20-entertaining-uses-of-chatgpt-you-never-knew-were-possible-3bc2644d4507 Some more applications https://marketplace.visualstudio.com/items?itemName=DanielSanMedium.dscodegpt&ssr=false VSCode extension that uses Subj to generate comments or query code snippets

BTW, is that you an original author? I've been using old-style version of ResHacker since, maybe, mid-00s!

Plz don't take as offense or criticism, but one user have done the big ready-to-merge PR almost 3 years ago that is still pending 🙂

What were users trying to find in that system folder? I personally can't tell when I was visiting Windows\System last time. The mentioned folder contains translated resources for system apps, as I understood. That's why the question.

Just curious. What did they do in system folder?

Only Hercules could clean up Augeas stables but even he cheated.

Mac doesn't allow setting a custom proxy

What's so wrong with OpenSSL? ICS could compile this lib statically with additional paid 3rd party lib. Having read the code I have (10.3), I'm afraid you can't easily plug into RTL version as it doesn't operate sockets calling more high-level OS functions instead. You could run HTTP CONNECT request to a custom proxy but then you'll have problems doing TLS to a destination host. So I suspect all you have to do is first curse MacOS designers as they deserve and then lower your requirements to include 3rd party.

I have some Run Lola RUN!! 🙂

If you just want to jump over controls by cursor arrows, it's easier to post crafted messages to a form via PostMessage or Form.Perform

ICS does all that and even includes login form using a browser

Hmm, probably finding bugs in code would be very useful task for this tool. Another boring task that I think of is generating test cases for unit tests - could someone check this?

ICS works on MacOS but not on Linux and uses OpenSSL. Why not implement proxy in RTL client by yourself?