GitLab

*fixed remaining problems (hopefully) with 2 arrays drawn in the same gop (need to test with more than 2 arrays?)
*bug for Jonathan--bar graph draws from top down, as opposed from bottom up when opening array gop

*improved fittograph for multiple arrays (still needs some work)

*further improvements on the garray selection and redraw--fixed garray contents not remaining selected after dialog properties have been applied
*still need to explore problems with multiple arrays in a single GOP

*fixed segfault when resizing array via dialog (regression from May commit). Still need to work on fittograph logic for situations where there are multiple arrays in the same GOP

*added recursive checking for when objects are selected to properly reselect arrays nested inside GOP subpatches/abstractions. This is important because the implementation should work for any level of depth. Later consider making this a more universal check should such prove necessary (I think once we move away from redrawing the array every time a point is moved, all this will become moot)

*added comments with future TODOs

- nested plots display properly
- plot is vis'd and unvis'd properly when deleting plot or struct
- commented out ugly plot_displace hack-- multiple plots should work just fine
- separated out garray dependencies into class "old_plot".  This way the changes I made don't affect garrays.  _Very_ ugly hack, but it was pretty simple to implement and will be easy to remove later.
- no crashes when non-existent arrays are entered into a [struct] that has scalars hanging around
- template_cancreate ported from 0.45
- added a function to get the offending struct (to use with "Find last error").  Not using it yet, though.
+ For creating scalars using an object box:
- scalar must have initial "float x float y" fields, in that order.  This means you can't accidentally type the name of the hidden garray structs and create a scalar.  (Nor any other structs that only have a "y" field, etc.)
- broken object box returned if the struct has a bogus array template.  (Although currently I'm just refusing to create such a struct.)

fixed segfault regression in 4.datatructures/05.array.pd example where clicking on the scalar array in a subpatch would cause a segfault

*fixed bug where with JOC enabled in subpatch this only works for the bbox of the array, not the entire window
*made intelligent offset to various forms of array representations so that scrubbing with mouse accurately applies to closest points
*disabled drawing of a GOP rectangle inside GOP array window
*disabled ability to insert objects inside a subpatch with an array

added Jonathan's revision of the array properties and options with fixed font sizes and other minor fixes. LINGERING PROBLEM: array does not redraw (only after apply does it re-appear). Likely shouldvis problem.

improved keyboard interaction with arrays (shift+click adds a point, ctrl+click deletes it--still need to figure out the role of shift when wonset is set) and instantaneous cursor update. minor refinements to drawsymbol.

*better getrect for scalars inside plot/array (entire points are now selectable instead of arbitrary rectangle around the object)

*fixed drawsymbol scalars' double names
*improved build process for tkpath
*improved to-front/back logic

first stab at porting core GUI to tkpath with SVG goodness. new scrollbars that will hopefully put an end to the old painful scrollbar nonsense. pdtk_tip clean-up and improvements. tons of other small bug fixes. a number of issues/wishlists still exist before the next big release:
*improve g_array lower/raise to use new functionality (BUG! doesn't match any items)
*scalar is drawn behind at first
*Scope~ window name "h16ffd10" already exists in parent (likely resize hook)
	*Change appearance of the resize hook
*better getrect for scalars inside plot?
*reenable bgerror and t_tkcmd catch
*Clean-up and merge cyclone documentation

fixed segfault when undoing changes to the array (need to design a new undo type for this, so for the time being undo is disabled for this action), fixed incorrect type of plot upon creation.

fixed 2 segfaults: one when trying to delete array that is being read/written to by a dsp process (e.g. tabwrite), and another where creating undo checkpoint when deleting array would generate bogus data that would crash pd-l2ork. Currently undoing array deletion is not supported by undo.

experimental implementation of disabling drawing outside the GOP array boundaries (currently disabled)

Fixed bug http://sourceforge.net/tracker/index.php?func=detail&aid=2457992&group_id=55736&atid=478070; hopefully also fixed incorrectly reported hit/selection boxes for nested structs (like the one reported in the previous bug report)

added automatic resize of arrays when adding/removing points via ctrl+click, fixed a bug where array is not properly redrawn after a new point is created

fixed bug where garray did not redraw its names when an array name has been changed via "rename" command

altered behavior of gop objects so that internal elements are not visibly selected. Also reworked the way array and scalar selection is dealt with fixing segfault and making a more fool-proof way of selecting them.

fixed memory leak for invisible 2 canvases with templates. they are now being freed when closing pd as they should be.