|
1 |
| -- fix up autotrace |
2 |
| - |
3 |
| - falsecolour output, drag sliders, occasional segv ... unreffing a region in |
4 |
| - maplut_stop on an image that's already been freed? ouch |
5 |
| - |
6 |
| - try: |
7 |
| - |
8 |
| - - load image |
9 |
| - - extract green |
10 |
| - - size down to 128x128 |
11 |
| - - save as x.v |
12 |
| - - restart nip2 |
13 |
| - - load x.v |
14 |
| - - click trace |
15 |
| - - ^Q |
16 |
| - |
17 |
| - see: |
18 |
| - |
19 |
| -(nip2:29365): GLib-GObject-WARNING **: instance of invalid non-instantiatable type '(null)' |
20 |
| -(nip2:29365): GLib-GObject-CRITICAL **: g_signal_handler_is_connected: assertion 'G_TYPE_CHECK_INSTANCE (instance)' failed |
21 |
| -(nip2:29365): GLib-GObject-CRITICAL **: g_object_unref: assertion 'G_IS_OBJECT (object)' failed |
22 |
| -(nip2:29365): GLib-GObject-CRITICAL **: g_object_unref: assertion 'G_IS_OBJECT (object)' failed |
23 |
| -(nip2:29365): GLib-GObject-CRITICAL **: g_object_unref: assertion 'G_IS_OBJECT (object)' failed |
24 |
| - |
25 |
| - try: |
26 |
| - |
27 |
| - $ valgrind vips system "autotrace %s -output-format svg -output-file %s |
28 |
| --despeckle-level 9 -line-threshold 1" --in x.v --out x2.v --in-format %s.ppm |
29 |
| ---out-format %s.svg[scale=2] |
30 |
| - |
31 |
| - runs cleanly, so the error is probably in vips_call |
32 |
| - |
33 |
| - try: |
34 |
| - |
35 |
| - - load x.v |
36 |
| - - vips_call "system" ["autotrace %s -output-format svg -output-file %s"] [$in => [A1.value], $in_format => "%s.ppm", $out => true, $out_format => "%s.svg"] |
37 |
| - |
38 |
| - see: |
39 |
| - |
40 |
| -(nip2:30392): GLib-GObject-CRITICAL **: g_object_unref: assertion 'G_IS_OBJECT |
41 |
| -(object)' failed |
42 |
| - |
43 | 1 | - expose kernel option in vips_resize(), then use that in nip2 resize ... will
|
44 | 2 | need to be 8.4
|
45 | 3 |
|
46 | 4 | need to be able to set enums to vips8 calls, maybe set as strings?
|
47 | 5 |
|
48 | 6 | - sharpen should use new interface?
|
49 | 7 |
|
50 |
| -- redo scale along the new vipsthumbnail lines |
51 |
| - |
52 |
| - need to be able to get window_size from an interp |
53 |
| - |
54 |
| - just call vips_resize()? |
55 |
| - |
56 | 8 | - can we call affine from nip2 vips_call? do we need a double array?
|
57 | 9 |
|
58 |
| -- fix up menu tear-off, again |
59 |
| - |
60 |
| - bug in gtk and won't be fixed, tear-off menus are deprecated |
61 |
| - |
62 |
| - https://bugzilla.gnome.org/show_bug.cgi?id=627511 |
63 |
| - |
64 |
| - oh argh, might as well remove all tear-off code and direct people to the tk |
65 |
| - browser instead |
66 |
| - |
67 | 10 | - hough_circle etc. don't get cached ... they use the vips8 API and the vips
|
68 | 11 | cache only works for vips8
|
69 | 12 |
|
|
0 commit comments