AutoHotkey Community

I am considering going from Autohotkey classic to Autohotkey_L. Could anyone indicate the maturity of the project. The last release seemed to have bugs in quite basic parts of functionality. Like:


I believe that Autohotkey_L is the way to go, and I will utilize it on any new project of experimental character, but being on deadlines and strict demands for stability means that I am worried about migrating just yet.

i've never had any issues with Autohotkey_L

as far as im aware, everything in Autohotkey classic is there, and theres extra functions ontop. so only the new functions may contain bugs.

You should forget about AHK Basic, especially in professional scenarios.

_________________

I really hope that you are right that Autohotkey_L is more stabile than Autohotkey Classic, but I do not believe that the following issues are present in Autohotkey Classic:

Yes, but notice it says

David

_________________
With mixed feelings I am stepping down from all moderation responsibilities: http://www.autohotkey.com/forum/viewtopic.php?t=82906

Yes. But the issues were fixed just three weeks ago, (see http://www.autohotkey.net/~Lexikos/Auto ... ngeLog.htm).

I really hope that these were the final bugs ever in Autohotkey_L, but I guess they are not. My question still stands: Could someone close to the development of Autohotkey_L please comment on the maturity of Autohotkey_L?

Because AHK_L is still in development, it will always contain bugs. Even AHK basic with a year of bug reports and fixing with no major development still has bugs.

Even software made by huge companies with thousands of pounds/dollars pumped into them have bugs (how many bugs in Windows?)

You may want to take a look at the AHK_L Roadmap for further details of it's development cycle.

_________________
With mixed feelings I am stepping down from all moderation responsibilities: http://www.autohotkey.com/forum/viewtopic.php?t=82906

Thanks for the link to the roadmap. It is certainly an ambitious project.

I know that all software has bugs, but some software has more bugs (and more critical ones) than others and I am after a classiciation like:
- Experimental
- Alpha
- Beta
- Stabile

Btw. I believe it is called Autohotkey Classic, to not get confused with the Basic language.

I think the more important question is not about bugs but about differences, and whether they will affect the correctness of your existing code.

For example, if you have MCode or DllCalls that expect 8-bit character strings, they might behave incorrectly when given UTF-16 strings. You will have to check your calls to ensure that the strings are in the right encoding.

Lexikos' new numbering scheme should help with identifying the difference between newer and more stable versions. If stability is at a premium, you may want to go with the last version before the next .features release.

Because he is constantly adding features, I wonder if it might be useful to denote some versions as beta. It might even be easy to do.

As a matter of fact, the ClipboardAll issue is present in AutoHotkey Basic and I believe it has been for a very long time.
AutoHotkey_L ANSI builds aren't as well tested, as I use the Unicode build and recommend others do the same.
Conversely, because AutoHotkey Basic is not still in development, its bugs will never be fixed. Perhaps it's a moot point.
FYI, Chris himself labelled it "AutoHotkey Basic". Seriously, "basic" is a common English word. Even if AutoHotkey Basic was aimed at programmers (who might be familiar with Basic), I don't understand how anyone could seriously get confused.
It is under active development, more or less experimental.

(Also, "classiciation" and "stabile" should be "classification" and "stable".)

Thanks a lot Lexikos! Because of your relatively clear answer, I will start converting my existing Autohotkey Basic code to Autohotkey_L shortly.

It was important to me to ask about the stability directly as I have almost wasted a full year trying to get MySQL (so-called stable releases) to work. Afterwards I discovered that the developer of some of the code (the Federated Storage Engine) described this code as proof-of-concept code, and it was included by default in their stable releases.