Options
The future of QAP?
- The future of QAP?
-
The future of QAP?
joeNOR > 2023-05-04 11:23
This is not a support question, more of an open discussion or maybe a question.
This is not an attack on Jean, a developer whom I started talking with way back in the days of "FolderSomething..." (don't remember the exact name) and whom I respect a lot.
But, AHK, at least the v.1 branch sucks so bad it makes me pretty close to Jason (google Friday the 13th for those who don't get it)
No multithreading, everything is so single threaded that it makes my head hurt.
Me forking off jobs in QAP, but have to wait for the cycle to complete.
This is an AHK-problem, this is not a QAP-problem.
Access to Windows functions via truly arcane methods, no intelligent handling of "non-normal" events.
(the "network drive"- discussion is an excellent example)
...and the list goes on.
When upgrading to Win 11, the startup time for QAP is insanely slow compared to Win 10.
QAP is, and have been for a very long time, one of my essential apps.
Maybe AHK 2 will solve this, but QAP should be ported to a "real" language, not an interpreter.
So, I hope that at some point in time Jean moves to AHK 2, AND AHK 2 is better in those areas than v1.
....OR:
We start a talk on this forum with Jean on what it will cost to move to a "real" language.
Of course this should be based on a crowd founding model where we have to put up enough money for Jean to push on.
...just a thought.
-joeNOR -
RE: The future of QAP?
Horst > 2023-05-04 12:25
I don't think there will be real benefits for using any compiled language.
AHK2 makes a lot better than version 1
but it has big learning curve and a brings lot of work changing large existing scripts.
The existing converter is currently not a big help.
I don't understand what the speed factor means in your environment ?
I use QAP as its indented functions.
- Run programs
- Make hotkeys and hotstrings
- Show lists of files and dirs
- Make favorites for my file manager (TC)
For me none of this functions are slow because of AHK limits. -
RE: The future of QAP?
LaurentG > 2023-05-04 14:23
Hi @joeNOR,
let me begin with a remark : I was surprised with your statement about W11, and possible poor performances of QAP on this OS;
I'm not a W11 user, so I decided to do a test, and I installed QAP on two VM (under virtualbox), one is a W10 "naked" (ie. nothing else but W10 itself installed), and the other with W11, "naked" also.
And according to my tests, startup time of QAP is the exactly the same on W10 and W11 (and the same than on my "real" laptop, running W10). And response time of QAP is immediate on both VM, as it is on real laptop.
So, if you have performance issue on W11, maybe there's something wrong in your W11 install...
Now, on the initial question of your post, personnally I'm very happy with QAP as it is today. I agree with @Horst that I can't see any real advantage to use a compiled language.
If there are some features that cannot be implemented in QAP (and correspond to a need...) due to AHK1, question will be asked to move to AHK2 or other , but today, is there really something that Jean would like to implement and he is no able to ?
If the answer is No, let's stay as we are currently... -
RE: The future of QAP?
joeNOR > 2023-05-04 23:16
Thank You to both of you for good and useful replies.
To address a couple of your questions/remarks.
Speed/multi threading:
I use the QAP Group function to trigger some pretty large sequences of favorites.
When the Group runs, QAP is totally unresponsive, and I have to wait for the Group to complete.
Some of those take around 2min to complete. Not much time if you do it once, but try to do it 50 times.
Then the time waiting for QAP becomes an issue.
This is starting to annoy me to such an extent that I seriously considers moving those groups to Powershell scripts, and disconnecting QAP completely from the process (other than maybe trigger the script).
Windows 11:
You won't see this issue with 50 favorites in QAP, but try loading it up with 500 favorites.
This is not a problem with my install because I've seen this across several machines completely different from each other apart from a large QAP "menu".
An old laptop upgraded to Win 11, two different virtual machines, one upgraded from Win 10, another installed with Win 11 from scratch.
Starting QAP on Win 11 takes more time than on Win 10.
Not a huge problem doing it once, again, try doing it 50 or a 100 times.
Especially since I also use QAP to trigger some jobs at login (the startup setting).
I guess my initial post was just to vent some frustration!
Regards,
joeNOR -
RE: The future of QAP?
Horst > 2023-05-06 12:50
2joeNOR
You said
"Starting QAP on Win 11 takes more time than on Win 10.
Not a huge problem doing it once, again, try doing it 50 or a 100 times."
I don't fully understand this.
Starting QAP is typicall done once per login in Windows and not for every usage.
After a login QAP is ready as many other tools started at login time. -
RE: The future of QAP?
Jean Lalonde > 2023-05-08 19:16
Thank you joeNOR for your good words in your initial post. I decided to wait before replying but I think it is now time to jump in :-)
About AHK v1 and v2
AHK has its advantages. It is light, portable, free, open source and supported by a large community. Without this community, QAP would just not exist. But yes it also has its limits. It is slower than professional compiled languages (even if you convert the .ahk source file it to an .exe file).
But, in my view, the main goal of AHK is not to process huge amounts of data or to do intensive calculations. Its goal is more to automate Windows or other applications. And, in my experience, the limits in automating are more related to what Windows allows to do (using API, COM objects or, in last resort, by sending keystrokes) than limits of AHK itself.
Does AHK v2 improve this? AHK v2 is a better programming language but it is not intended to improve performance. Its goal is to make coding easier and more consistent. This being said, all the major changes between v1 and v2 make it very-very hard to port large v1 scripts like QAP to v2.
Would a compiled language do better in automating Windows? I don't think because it would be confronted to the same limits: security roadblocks, different application platforms and various quirks and bugs.
This is why I have no plan to port QAP to AHK v2 or to any other language. There would be very little benefits for a HUGE amount of work. And, from a personal point of view, I have more interesting projects that porting an existing application to a new language :-)
About your QAP usage
As I mentioned to you in an earlier message, I'm glad and impressed with what you can do with QAP to manage a large number of computers using QAP groups, snippets, shared menus, etc. But you must consider that QAP was initially designed for a single user loading the favorites once a day (or less) and, before the introduction of shared menus, on a single computer. A few hundred favorites looked like a an high number in my mind. Loading the QAP favorites from an ini file is the weak link of QAP in terms of performance compared to storing them in a database. But, again, this did not look like an issue and storing favorites in a text file has its advantages. Honestly, you really pushed the limits of QAP and, again, I applaud this. But I agree with you that it is a good idea to look for alternatives, at least for parts of your processes.
About Win 11
I won't comment on this issue, having not enough experience with Win 11. I only have it on a portable computer that I now use mainly to test QAP. In these tests, I never loaded very large ini files and the medium files always loaded with the expected performance.
About QAP future
Finally, speaking of the QAP future, I consider it a mature product... some could even say that it suffers from feature creep ;-) . This is why I don't plan any major additions to QAP. Rest assured that I will continue to maintain it and, from time to time, add new features from the wish list. But there will not be a new "generation" of QAP.
After almost 10 years (taking into account Folders Popup years), I'm still impressed with the interest shown in QAP and the support I receive from users whether through donations or good feedback. I thank you all for this and hope to continue to be useful to users with this application or with my other projects.