- User Since
- Mar 24 2017, 4:05 AM (64 w, 3 d)
Fri, Jun 15
I just saw the big wiki changes happened today. Two questions:
Sun, May 27
Fixed bad indexing of select_history in modified select_prev, updated benchmark results with code changes
Mon, May 21
May 3 2018
Applied with commit:
May 2 2018
Apr 23 2018
Apr 15 2018
Apr 14 2018
@Philipp Oeser (lichtwerk) I was not entirely sure if you were suggesting a rollback or not, but Ako was.
@Philipp Oeser (lichtwerk), I checked and AkoZ was correct, it's a layer management problem. Part of the problem is Blender's weird "view-dependent layers" option. For some reason Blender sets any layers (in "context.space_data.layers") attached to the current view to "False" if you disable the "Scene Layer Lock" option and switch to Local View. That is what prevents any measurements from showing up. I don't think rolling back rBA8ca82b7d or adding a quick fix like:
"if True in bpy.context.space_data.layers do old_behaviour"
are good options as both of those brings back the view layer inconsistency bug which is (IMO) worse than this Global / Local issue.
Apr 13 2018
Which version of MeasureIt? I can't reproduce the bug with MeasureIt 1.7.0 (included with 2.79b). For me the measurements show in both "Local" and "Global" modes (what "Numpad /" toggles).
Apr 6 2018
Mar 12 2018
Jan 25 2018
@Philipp Oeser (lichtwerk) I thought I remembered reading that Python does 1 or 2 additional operations behind the scenes somewhere, but I can't remember the details. I didn't see any significant change in performance with this hastily written test code though, so I say leave the last commit as is. :)
I forgot MeasureIt runs continuously when the add-on is enabled and measurements are set to "Show". That's why I think I originally left this as two separate statements than using boolean logic to join them as I was not sure how Python handled boolean logic behind the scenes.
I tested and can confirm this fixes the bug from T53766. This change looks fine to me.
but I think this might be cleaner \ easier to follow:
Jan 12 2018
Dec 10 2017
Try this link:
Dec 8 2017
Support for XP was completely dropped after 2.76. I think XP 64 support was dropped much earlier, maybe with 2.69 or 2.70?
Nov 10 2017
Nov 7 2017
Interesting proposal, but this type of discussion would be better suited for the Blender mailing lists (maybe bf-committers or bf-funboard) or a more general Blender / Blender feature discussion board (like rightclickselect or blender artists).
Oct 23 2017
So to clarify, these 2 results were intended?
- Having the supplementary angle (assuming degree measure: 180 - angle_measure) instead of returning the "interior" angle.
- Having calc_edge_angle not raise an error when called with a vertex that has only 1 edge and instead return pi (~3.14).
Oct 19 2017
This is caused by an older bug with custom visibility layers that differ from scene visibility settings. See this post for more info.
Sep 30 2017
Sep 26 2017
Sep 25 2017
@Antonio Vazquez (antoniov) Sure. I have a rough idea for a fix.
I found the issue, it looks like this may be a bug with MeasureIt after all. The problem is MeasureIt only checks Scene layer visibility settings for determining whether or not to display measurements (measureit_main.py, lines 2014 through 2040). This becomes an issue if the "Scene lock" button to the right of the layer manager is disabled which allows screen layouts to have seperate layer visibility settings seperate from the scene layer settings. This is the case with the "houseplan_24Aug17.blend" file which has all the MeasureIt data stored on layer 6 which is not enabled in the scene's layer settings, but is enabled in the active "2-Model " screen layout settings. You can verify this by enabling the "Scene lock" button with "houseplan_24Aug17.blend" open in Blender.
Sep 19 2017
I missed an update on this again :( I need to check my email settings...
Sep 15 2017
I did a more exhaustive test and opened "houseplan_24Aug17.blend" in Blender 2.78 and copied all the objects from that file into a seperate instance of 2.78 keeping the same layer organization and I'm not seeing the problems with MeasureIt anymore. The object and geometry count did go way up in the process:
@Antonio Vazquez (antoniov) According to the submitter this was not made with Archimesh, but the default add mesh tool. For some more strangeness, I noticed if I move the one of the HOUSE_WALLS objects from layer 0 back to layer 5 the measurements break again.
@Nicholas Spies (nspies11) I was wrong, "HOUSE_WALLS" was the only mesh object that was duplicated. It looks like the duplicate (HOUSE_WALLS.001) is what most of the measurements were attached to. Also, to submit answers you can just enter them in the reply box at the bottom of this page. But if that is not working for some reason, editing the description works to.
Sep 12 2017
I found what I believe is the source the problem. There appears to be an issue with MeasureIt when geometry is on different layers. When I moved all the geometry to the first layer all the measurements in the file work as intended. I'm not sure if this is related, but there also appears to be duplicates of most of the geometry scattered over different layers.
Bah, didn't know this was reopened. Well this is definitely a weird issue. MeasureIt seems to work just fine if you add new geometry or annotations to the blend file, but fails with most of the existing geometry. I don't know of any limitation preventing MeasureIt from working with joined or grouped geometry. Did you use Archimesh or another addon to create the geometry or the stock "Add mesh" option?
Aug 12 2017
Aug 11 2017
Aug 2 2017
Aug 1 2017
I pushed a partial fix that should fix the Bug 1, but (if you can) could you please give this commit a quick look over (mainly the changes in measureit_main.py):
lijenstina also reported the following two errors to the Blender Console when MeasureIt was being used, but I did not include the second in the task description as I have not yet figured out exactly what causing it to show up:
Jul 26 2017
Just attempted to push the change to master and it appears I do not have commit access to blender-addons :(
Jul 25 2017
@Antonio Vazquez (antoniov) I have gone ahead and uploaded a differential of all my changes here as you requested:
Here is the redesign I mentioned in my last reply.
Jul 24 2017
Somewhat related, I also have a minor redesign / change request for the "Mesh Debug" panel designed for ease of use and with more features, but the code for it is not currently in a usable state. I hape to have that ready today or tomorrow.
Jun 7 2017
Jun 5 2017
Jun 2 2017
Updated diff patch with Blendify's recommendations.
I mostly agree with all your suggestions, I left a comment for the others.
May 31 2017
May 30 2017
May 29 2017
May 25 2017
May 3 2017
Thanks for the info mont. After asking on IRC and searching google went nowhere I assumed this was a bug. This might make for a good addition to the Blender-Python API Doc's "Gotcha’s" page.