(older archive entries before 2007 are not shown here, but included in the onsite-search)
Mailing List - Entries of 2011
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[QF-Test] Features suggestion
Here are some questions/suggestions for QFTest.
I don't know if it is the right place or if I should have sent it to the support directly.
Anyway, if you can provide any feedback/workaround on this.
- Memory monitoring -
I have very large Suite files ans sometime my run-logs are quite big too.
(Mostly because of screenshots)
So I start QFTest with a reasonable amount of memory but sometime I end up with an "out of memory" error.
And if I have not saved my suites yet, it can become tricky to close the right files in order to get some memory back and be able to save my work.
It would be helpful to have somewhere in the GUI a "memory status" which indicate what amount of memory is free.
- Suites IO improvements -
I have a lot of image checks per Test-case, and as a consequence, my Suite-files are large files (30 to 50 MB)
The "problem" is that when I save my Suite it saves the whole file and it takes more time as the file becomes bigger.
(I already try to have only ~10 Test-Cases per file but each Test can have a lot of image-checks and my images are quite big (> 1000x1000))
It seems that this kind of issues is handled for the runlogs with qzp files, right ?
Would it be possible to have some kind of "qzp" storage for Suites ?
(Or anything that would minimize the amount of bytes to write on the disk at each ctrl+s )
Further more, files are so big that it is hard to use it outside of QFTest (diff / svn / text_search&replace)
- Data loops & image-check improvements -
I find the Data-loop system really helpful and very appreciable when scaling a test on several inputs.
But I would like to use images as "expected" result.
Input1= ### ; Result1 = $Image1;
Input2= ### ; Result2 = $Image2;
I can do that using a file_path variable as the result of each iteration.
And then write the "expected image" to this file if it does not exist, or read the "expected image file" and compare it to the actual-image.
It works great, but this is a bit painful:
- Test reading/recording
I can't see the "expected images" in QFTest (except in the runlog of course).
I have to open the image-file manually.
It can be time&brain consuming when dealing with complex tests.
- Test maintenance
It is "complex" to update the "expected image" with a newer result from a runlog.
The Ctrl+U is not available. (Can I implement a sort of resolver/hook to make this works ?)
Is there an easier way of doing this ?