Godd evening @Grace, @Douglas, @csett86 , at the end of the day a last problem occurs in my todays construction. I saved the file while we wre descussiong the problem with the measurment list in seamlyme. Now I wanted to continue the construction and I opend the file, which I send you here
@Grace, @Douglas, @csett86 *,
finally I found out what happend, It is the measuremnt List. After we solved the problem with the position of the added measure AD today I did the calculation with the formula assistent and placed the measure pac to the top as I normally have my values in alphabetical order, makes it easier to find them. And that was not a problem until the latest release.
The measure was named an error after I saved and closed the List. And that destroyed my construction. I placed the AD values at the bttm of the list and there it worked and the construction is complete as before, thats interesting as releases ar normally made to make programms running better making it easier for users to work with and solve problems that occured before.
A rule of thumb… Both Seamly2D and SeamlyME operate in a strict order, so if everything reads in a strict linear fashion. By placing your formula (in SeamlyME) at the top of the list means that the program is reading it first, so it will give an error if it references any of the codes below it.
I normally place this type of code into an Increment where I can edit it, if need be, and the order of Seamly2D is that it loads the measurement files first, then the increments and lastly the pattern drafting. So by placing this code in Increments, it will have the measurements already in memory. Just remember that it must not use any drafted objects in the formula, because they will only be read later and the formula will revert back to 0 or an error (the same applies to SeamlyME).
Since I haven’t made any updates since weeks (or month) ago, the only changes that I believe could affect the build versions are in the CI… that is any changes in the qt ver & compiler and / or ver used.
That being said… you have more or less treat each weekly build as a pre-release or Beta ver - where there may be new bug or regressions. We’re working towards having a stable “Release” schedule.
And for the CI @Douglas I can assure that the Qt version 5.15.2 has not changed in the last year at the least. Changes in CI were that the mac version is signed so that macOS does not warn when starting for the first time, but that also does not habe any effect at all on the pattern parsing etc. So I think the behaviour was always like this (parsing formulas from top to bottom, as @Grace highlighted), but maybe my chance there were no dependencies in your past patterns that conflicted here
Since I’m now home - yea! - I can look at the issue… there is no issue. The warning message posted is expected behavior, and has been there for awhile now. In this case I’m sure a measurment was changed that resulted in an intersect not being found - thus the message that the origin is being used as place holder. If the origin is not used (could be any arbitrary point) the pattern will always crash and you could never fix the formula / measurement.
It’s because something changed in a formula - like changing a measutrmenet value - that caused an intersection tooll to NOT find an intersection point. This is expected behavior. So as to not cause the pattern to totally crash, the non existent intersect point is temporarily replaced with the origin point (0,0) until the formula or measurement is fixed. The message is there to warn you that something is wrong with the formula. Alsoi keep in mind that inputting sizes beyond a pattern’s limit may also cause this to happen.
