David Heffernan

What Mike is concerned about is that calling Replace is slower than filtering with Pos before calling Replace. In reality he is asking the wrong question. The right solution to the problem (see five year old SO question) is to iterate over the input string once rather than 5-10 times.

Well there's your problem! Run through the string once only and you'll get a huge boost in performance. I described how to do that in my comment to your Q.

Why not achieve faster in all cases then? Does the code that you are looking at, that is a bottleneck, still call StringReplace multiple times per input string, as per the SO question?

If the code is not a bottleneck then surely you'd reach the opposite conclusion.

You really need to learn about the extract method refactoring.

No. I'd consider writing one myself if I found that my program was spending a significant amount of time in an inefficient function. But we've had this discussion before and you don't believe in that approach.

Nothing needs to be fixed. StringReplace is giving correct output, isn't it? You absolutely don't want to be calling Pos inside an efficient implementation of StringReplace. All that you are learning is that StringReplace is not efficiently written. The correct way to deal with that is to find/write and use a more efficient function to replace text. That is of course if this is a bottleneck in your code. Have you checked? BTW, do you really have calls to Pos littered through your code whenever you call StringReplace?

Really? I mean true for bmp files but Windows API supports 32 bit bitmaps fine so far as I know.

Why don't you want to use a third party component?

That's what RPyC would gain you

Two different values can have the same checksum. Read my previous post again. What you are expecting is impossible. I mean it. Impossible.

What is a signature? All these terms being bandied about!!

Yeah, that's not a hash. And what you want (known max length) is clearly impossible. If you could store arbitrary data in a lossless way in a data of a fixed length, then well, you can see where I am going.

OK, maybe I misunderstood. I saw "hashing" and "indexed on", and inferred that you want to use a hashed collection like a dictionary. Hence the need to define data structure and what equality means. But perhaps you want to hash this data for some other reason. What will you do with this hash?

First of all, don't worry about hashing the data. Work out how to represent the data, what data structure to use. And then define what you mean by equality for this data. Once you have done that, and have a clear specification of that, defining a hash function should be obvious. But if not, we will be able to help at that point. But first you need to define the data structure and the equality operator.

Any code can terminate the process. The way you avoid this is to ensure that the code you run doesn't lead to fatal errors. In this case it sounds like the issue is that you need to make sure that your environment has the correct packages installed.

What is the advantage over Timsort?

I'm not sure what you want to do. Do you want to read the XML quickly? Or do you want to read the XML without using CPU resources? Because it seems like you want to do both but of course you can do one or the other but not both.

What algorithm?

No. This is of course easy to test using debugging techniques, trace logging etc. It's also documented in the language guide.

@Stefan Glienke already explained that quadword reads/writes are atomic for aligned data.

This is true for all atomic memory operations. Is your argument that we should not use atomic memory operations? That can't be what you mean. Like all code, there are pre-conditions for it to work. In this case we require the memory to be aligned. Onus for that falls on the user. And it's clear that Thomas knows this.

No. The polymorphism applies to method dispatch, and not to the types.

No. It's not counting lines. It's counting warnings. The compiler emits two distinct warnings. For whatever reason they are identical. There's doubtless some internal quirk that explains it.

They are all counted correctly. Look in the screen shots. Count the items marked error, warning and hint. Compare to the totals reported. In every single case these match. I think in this case you need to pay a bit more attention to the detail.