<wpwrak>
whee, reverse world ! now, for some reason, TW doesn't stay at all if NOT using DISPLAY_TEXTWND, while DISPLAY_TEXTWND works as usual, i.e., not perfectly, but not 100% failure ;-) tried closing some windows, but it's not that.
<whitequark>
wtf
<wpwrak>
meh. now it works. actually ... it appeared at an unusual place ... lemme see if there's a pattern ...
<wpwrak>
(unusual) very low on the screen
<wpwrak>
grmbl. of course, now i can't reproduce the disappearance with DISPLAY_TEXTWND. some bugs cling on to life with remarkable persistence.
<whitequark>
is one of your monitors larger than the other?
<wpwrak>
no, same size
cr1901_modern has quit [Ping timeout: 276 seconds]
<wpwrak>
hmm ... interesting pattern. i'm making somewhat major changes to a sketch. that sketch is used in an extrusion which is later on used in a point-on-face (?) constraint. during the changes i break the outline by first deleting a constraint and then moving a point. i.e., without deleting any lines or arcs.
<wpwrak>
then i delete a distance. now it seems that slvs tries to recalculate the rest, and promptly complains about the point-on-face constraint, probably because that face no longer gets created. that error stays on until i close the outline again.
<wpwrak>
interestingly, deleting the line i used to "patch" the outline doesn't bring the error back. odd.
<whitequark>
sounds about what I expect
<wpwrak>
but all this suggests that a "defer updating future groups during major surgery" function would be useful. referential integrity would still be preserved, but unavoidable transient issues would not be reported. besides, everything would probably be a lot faster.
<whitequark>
that would be nicely doable with chiral constraints
<wpwrak>
(i now get delays in the order of ten seconds for things that trigger a recalculation)
<whitequark>
commercial parametric CAD packages will often give you delays on order of minutes too :p
<wpwrak>
well, i can just check the following groups for chirality changes later
<whitequark>
I'm really not happy with shipping a feature that breaks sketches silently
<wpwrak>
display a big warning when someone uses it for the first time :)
<whitequark>
or I could spend time working on chirality instead
<wpwrak>
heh :)
<whitequark>
you can just patch this thing in your local tree if you want it *right now*
<whitequark>
anyhow because even if I agreed to implement this, you wouldn't get it today...
<wpwrak>
oh, i can be patient ;)
nathan7_ is now known as nathan7
<wpwrak>
by the way, my suspicion that the TW disappearance is caused by it simply ending up off-screen, and me now looking for confirmation of that theory, caused the issue to completely disappear. i wonder of that bug is reading #solvespace :)
<GitHub>
[solvespace] whitequark commented on issue #193: The "unexpectedly" part is not actionable without a way to reproduce. In general SolveSpace switches to the free in 3d mode by itself in one case: when there isn't a workplane available. https://github.com/solvespace/solvespace/issues/193#issuecomment-278373214
<wpwrak>
tricky. if i knew what causes the switch, i would probably avoid it :) is there a place i could put an abort to immediately make me aware that an automatic switch would be happening ? then i'd know what exactly i did last
mifune has joined #solvespace
mifune has joined #solvespace
mifune has quit [Changing host]
<GitHub>
[solvespace] wpwrak commented on issue #193: Okay, I found one case where this happens: I have a sketch without issues (that I know of). In this sketch there is an arc with both endpoints fixed. I set the center with a point-to-point coincidence, which over-constrains the sketch and I get an error. At this point I'm still in the workplane.... https://github.com/solvespace/solvespace/issues/193#issuecomment-278387267
<GitHub>
[solvespace] wpwrak commented on issue #193: The plot thickens: the "undo" also changed the group (from g01f-sk-minus-base waaay out to g02d-balance). No surprise it doesn't find my old workplane there. The group change may in turn have been caused by reporting a transient SOLVE FAILURE of the #182 type. https://github.com/solvespace/solvespace/issues/193#issuecomment-278389835
<wpwrak>
those intertwined bugs are fun ;-) (at least once you know what's going on)
<wpwrak>
hmm, seems that "Show Degrees of Freedom" never shows anything in a group that is an assembly (without other things in it). bug or feature ?
<whitequark>
wpwrak: sounds vaguely like a bug
<wpwrak>
filing an issue ...
mifune has quit [Ping timeout: 245 seconds]
<wpwrak>
#103 will be such a big leap forward. just gave up on converting a circle into a D shape. no matter what i do, i can't cheat death (deleting the circle taking numerous other groups with it). fortunately, it's not a major issue to have that circle/cylinder there.