forked from danieljprice/giza
-
Notifications
You must be signed in to change notification settings - Fork 0
/
ChangeLog
4766 lines (3330 loc) · 199 KB
/
ChangeLog
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
2023-11-30 Daniel Price <[email protected]>
* LATEST.md, NEWS, configure, configure.ac, docs/news/index.html:
v1.4.1
2023-11-30 Daniel Price <[email protected]>
* src/giza-driver-mp4.c: (mp4) use -loglevel quiet in default ffmpeg flags
2023-11-28 Daniel Price <[email protected]>
* LATEST.md, NEWS: update release notes for v1.4.0
2023-11-28 Daniel Price <[email protected]>
* .github/workflows/api.yml, .github/workflows/build.yml,
.github/workflows/release.yml: update github workflows
2023-11-28 Daniel Price <[email protected]>
* src/giza-driver-mp4.c: (mp4) avoid possibility to overflow string length using environment
variable
2023-11-28 Daniel Price <[email protected]>
* ChangeLog, NEWS: updated ChangeLog + NEWS
2023-11-28 Daniel Price <[email protected]>
* .github/workflows/release.yml: (release) fix bump brew command
2023-11-28 Daniel Price <[email protected]>
* : Merge pull request #47 from danieljprice/mp4 mp4 device, v1.4.0
2023-11-28 Daniel Price <[email protected]>
* LATEST.md, configure, configure.ac, docs/documentation/api.html,
docs/index.html, docs/news/index.html: v1.4.0
2023-11-28 github-actions <41898282+github-actions[bot]@users.noreply.github.com>
* docs/documentation/api.html: [web-bot] updated auto-generated
documentation
2023-11-28 Daniel Price <[email protected]>
* src/giza-driver-mp4.c: (mp4) always replace existing mp4 file; allow user to set ffmpeg
flags using GIZA_FFMPEG_FLAGS environment variable
2023-11-28 Daniel Price <[email protected]>
* LICENSE => COPYING, src/Makefile.am, src/Makefile.in,
src/giza-driver-mp4-private.h, src/giza-driver-mp4.c,
src/giza-drivers-private.h, src/giza-drivers.c: (mp4) added mp4 device; tested and seems to work
2023-11-28 Daniel Price <[email protected]>
* COPYING, LICENSE: changed license to LGPL3
2023-10-10 Daniel Price <[email protected]>
* : Merge pull request #46 from
htrb/add_registered_and_copyright_sign add registered and copyright sign as symbols.
2023-09-29 Hiroyuki Ito <[email protected]>
* src/giza-hershey-to-utf.h: add registered and copyright sign as
symbols.
2023-09-29 Hiroyuki Ito <[email protected]>
* src/giza-box.c: Add support for '1' and '2' options to giza_box.
2023-03-10 Daniel Price <[email protected]>
* : Merge pull request #40 from bkmgit/empty-file Delete empty api-ref.html file
2022-09-09 github-actions <41898282+github-actions[bot]@users.noreply.github.com>
* docs/documentation/cpgplot-status.html,
docs/documentation/pgplot-status.html,
docs/documentation/pgplot.html: [web-bot] updated auto-generated
documentation
2022-09-09 Daniel Price <[email protected]>
* src/giza-cpgplot.c, src/giza-pgplot.f90: (docs) updated pgplot implementation status for PGSITF/PGQITF
2022-09-09 Daniel Price <[email protected]>
* Makefile.in, aclocal.m4, config.h.in, configure, src/Makefile.in,
test/C/Makefile.in, test/F90/Makefile.in, test/Makefile.in: (build) updated automake files
2022-09-09 Daniel Price <[email protected]>
* : commit 1902b4d55ffc1439fabc7e564f9d3e16cf2c8817 Merge: 4ff70a8
ace6b44 Author: Daniel Price <[email protected]> Date: Tue
Sep 6 08:32:19 2022 +1000
2022-09-06 Daniel Price <[email protected]>
* : Merge pull request #38 from kerel-fs/pr/fix_readme
2022-08-01 Harro Verkouter <[email protected]>
* src/Makefile.am, src/giza-cpgplot.c, src/giza-fortran.F90,
src/giza-itf.c, src/giza-itf.h, src/giza-pgplot.f90,
src/giza-private.h, src/giza-render-private.h, src/giza-render.c,
src/giza.h: Add support for PG[SQ]ITF, fix PGCONT PGPLOT has three image transfer functions defined for PGIMAG:
linear, log, sqrt. These are now implemented by this commit. The current render engine uses pixel-colour-indices to colour
pixels: based on the pixel value, limits, the transfer function
computes the "position" of the pixel on the 0->1 scale and
transforms that to the nearest colour index from the current
CIMIN/CIMAX colour index range. (To follow PGPLOT docs.) The old code would ignore the colour index range and for each pixel
interpolate the colour table that was set, basically always "true
colour" representation. In a subsequent commit I could provide
infrastructure to support both: code calling giza_render() directly
would return to this behaviour, the call via the PGPLOT interface
would force it through using colour index range. The code now has several flavours of transforming a pixel value to a
normalized colour (on the interval 0..1) or converting that directly
into a colour index between cimin, cimax. In the process fixed some minor bugs. Tested using pgdemo4, adding a PGSITF([123]) manually in the demo
program.
2022-08-01 Harro Verkouter <[email protected]>
* src/giza-colour-table.c: Fix giza_set_colour_table to set
colourindices The function would warn about inconsistent entries in the
controlPoints and return with an error, not setting the colour-index
table. PGPLOT pgdemo4 (demonstrating PGIMAG) has a table with controlpoints
outside the range 0->1 (-.5, 1.7) triggering this warning and error
exit. The colour table is installed nonetheless! Net effect: pixel
colourings in "true colour" - i.e. those that do not go through
colour indices - keep on working, but those relying on colour
indices don't. This fix prints the warning, checks that a sensible number of
control points remain, and then goes on computing the colours for
the set colour index range (PGSCIR) - as per documentation.
2022-08-01 Harro Verkouter <[email protected]>
* src/giza-colour-index.c: Simplified set_colour_index_range
2022-08-01 Harro Verkouter <[email protected]>
* src/giza-contour.c: Follow PGPLOT auto-linestyle in PGCONT
2022-07-27 Fabian P. Schmidt <[email protected]>
* README.md: (docs) fix typo in pgplot replacement usage section
2022-07-25 Harro Verkouter <[email protected]>
* src/giza-points.c: Address char height/line width scaling issues According to PGPLOT documentation, symbols are drawn using current
attributes character height and line width (see
https://sites.astro.caltech.edu/~tjp/pgplot/subroutines.html#PGPT). Using the pgdemo2 program, 2nd page (SUBROUTINE PGEX22 in
pgdemo2.f), this behaviour can be tested. I modified both the
linewidth (LW=8) and the character height (CH=2.5) to see how PGPLOT
handles this. Indeed PGPLOT makes the symbols larger and draws thicker lines. The
Giza master branch does not follow this behaviour. This commit seeks
to address that. Some base symbols had to be scaled up a bit wrt to their current
setting, after which character height scaling could be used to do
the rest. giza-points.c does not set a fixed line width anymore, in stead it
follows the current device's setting.
2022-03-14 Daniel Price <[email protected]>
* LATEST.md, configure, configure.ac: v1.3.2
2022-03-14 Daniel Price <[email protected]>
* src/giza-fortran.F90: (giza-fortran) cleanup unnecessary implicit none statements; keep
only the one at the top of the module
2022-03-14 Daniel Price <[email protected]>
* src/giza-drivers.c, src/giza-fortran.F90: (#35) added additional
trims for safety where required
2022-03-14 Daniel Price <[email protected]>
* src/giza-drivers.c: (giza-drivers) trim spaces at end of device string in C causing test
failures in Fortran interface, fixes #35
2022-01-28 github-actions <41898282+github-actions[bot]@users.noreply.github.com>
* docs/documentation/api.html: [web-bot] updated auto-generated
documentation
2022-01-28 Daniel Price <[email protected]>
* ChangeLog, LATEST.md, configure, configure.ac,
docs/news/index.html: v1.3.1
2022-01-28 Daniel Price <[email protected]>
* test/F90/test-pgaxis.f90: (pgaxis) better pgaxis test
2022-01-28 Daniel Price <[email protected]>
* : commit e27012112cb66166dfe3a50d86a22aeba1fbeab2 Author: Daniel
Price <[email protected]> Date: Fri Jan 28 15:51:26 2022
+1100
2022-01-28 github-actions <41898282+github-actions[bot]@users.noreply.github.com>
* docs/documentation/api.html: [web-bot] updated auto-generated
documentation
2022-01-28 Daniel Price <[email protected]>
* : commit 23d9c67e0caa0fafde228f2d9b20f1696d77326b Author: Daniel
Price <[email protected]> Date: Fri Jan 28 15:44:05 2022
+1100
2022-01-28 github-actions <41898282+github-actions[bot]@users.noreply.github.com>
* docs/documentation/api.html: [web-bot] updated auto-generated
documentation
2022-01-28 Daniel Price <[email protected]>
* : commit 6a52d44e30da0beb293e31ff5571b3fa60678bba Author: Daniel
Price <[email protected]> Date: Fri Jan 28 15:35:09 2022
+1100
2022-01-28 github-actions <41898282+github-actions[bot]@users.noreply.github.com>
* docs/documentation/pgplot.html: [web-bot] updated auto-generated
documentation
2022-01-28 Daniel Price <[email protected]>
* .github/workflows/api.yml, docs/documentation/{pgplot.shtml =>
pgplot-header.html}, scripts/generate-docs.sh: (docs) bug fix with auto-generated docs: use simple cat instead of
sed -r
2022-01-28 Daniel Price <[email protected]>
* : commit c3429bf2707b6a2ec55cf5d41a554bb7b9d2c1ad Author: Daniel
Price <[email protected]> Date: Fri Jan 28 15:15:08 2022
+1100
2022-01-28 github-actions <41898282+github-actions[bot]@users.noreply.github.com>
* docs/documentation/pgplot.html: [web-bot] updated auto-generated
documentation
2022-01-28 Daniel Price <[email protected]>
* .github/workflows/api.yml: (api) fix github workflow
2022-01-28 Daniel Price <[email protected]>
* .github/workflows/api.yml: (api) fix github workflow
2022-01-28 Daniel Price <[email protected]>
* .github/workflows/api.yml: (api) fix github workflow
2022-01-28 Daniel Price <[email protected]>
* .github/workflows/api.yml: (api) fix github workflow
2022-01-28 Daniel Price <[email protected]>
* .github/workflows/api.yml: (api) fix github workflow
2022-01-28 Daniel Price <[email protected]>
* .github/workflows/api.yml, {doc => scripts}/api.pl, {doc =>
scripts}/cpgplot_status.pl, {doc => scripts}/generate-docs.sh, {doc
=> scripts}/get-fortran-params.pl, {doc =>
scripts}/get-source-files.pl, {doc => scripts}/pgplot_status.pl: (docs) moved doc->scripts to avoid confusion
2022-01-28 Daniel Price <[email protected]>
* LATEST.md, doc/generate-docs.sh, docs/documentation/pgplot.html,
docs/documentation/pgplot.shtml: (docs) broken links fixed
2022-01-28 Daniel Price <[email protected]>
* doc/generate-docs.sh, docs/documentation/cpgplot-status.html,
docs/documentation/index.html, docs/documentation/pgplot.html,
docs/index.html, src/giza-cpgplot.c: (docs) updated cpgplot interfaces for giza_axis and giza_tick; fix
issues in documentation
2022-01-28 Daniel Price <[email protected]>
* docs/documentation/gettingstarted.html, docs/news/index.html: (docs) fix broken links
2022-01-27 Daniel Price <[email protected]>
* docs/documentation/api.html,
docs/documentation/pgplot-status.html: [web-bot] updated
auto-generated documentation
2022-01-27 Daniel Price <[email protected]>
* docs/news/index.html: updated version
2022-01-27 Daniel Price <[email protected]>
* LATEST.md, src/Makefile.am, src/Makefile.in, src/giza-axis.c,
src/giza-fortran.F90, src/giza-pgplot.f90, src/giza-tick-private.h,
src/giza-tick.c, src/giza.h: (#12) implemented giza_tick replacement
functionality for PGTICK
2022-01-21 Daniel Price <[email protected]>
* LATEST.md, configure, configure.ac: v1.3.0
2022-01-21 Daniel Price <[email protected]>
* : Merge pull request #34 from danieljprice/motion-callback Motion callback
2022-01-21 Daniel Price <[email protected]>
* src/giza-drivers.c, src/giza-fortran.F90, src/giza.h: (motion
callback) added giza_end_motion_callback
2022-01-21 Daniel Price <[email protected]>
* src/giza-driver-xw.c: (motion callback) bug fix: make sure
transform is to world coords
2022-01-19 Daniel Price <[email protected]>
* src/giza-driver-xw-private.h, src/giza-driver-xw.c: (motion
callback) bug fix: restore clipping at end of Xwindows loop
2022-01-19 Daniel Price <[email protected]>
* src/giza-axis.c, src/giza-box.c, src/giza-character-size.c: (giza_axis) bug fixes; everything now works according to the API
2022-01-18 Daniel Price <[email protected]>
* src/Makefile.am, src/Makefile.in, src/giza-band.c,
src/giza-box.c, src/giza-fortran.F90, src/giza-pgplot.f90,
src/giza-render.c, src/giza-stroke-private.h, src/giza.h: (giza_axis) implemented giza_axis to replace PGAXIS
2022-01-17 Daniel Price <[email protected]>
* src/giza-driver-xw.c: (motion callback) expand clipping during
Xwindows loop
2022-01-17 Daniel Price <[email protected]>
* src/giza-driver-xw.c, src/giza-drivers.c, src/giza-fortran.F90,
src/giza-private.h, src/giza.h: (motion callback) send band mode
2022-01-14 Daniel Price <[email protected]>
* src/giza-fortran.F90: (fortran) do not trim strings when passing text to giza
2022-01-14 Daniel Price <[email protected]>
* src/giza-fortran.F90: (motion callback) fortran interface to
giza_set_motion_callback added
2022-01-14 Daniel Price <[email protected]>
* src/giza-driver-xw.c, src/giza-drivers.c, src/giza-private.h,
src/giza-window.c, src/giza.h: (motion callback) implemented
giza_set_motion_callback specifying function to be called during
cursor movement (works for xw device)
2021-06-18 Daniel Price <[email protected]>
* .github/workflows/release.yml: updated brew update to point to
giza-x11 in danieljprice/all
2021-06-18 Daniel Price <[email protected]>
* LATEST.md, configure.ac: v1.2.1
2021-06-18 Daniel Price <[email protected]>
* .github/workflows/release.yml: (github) updated API token
2021-06-18 Daniel Price <[email protected]>
* src/giza-colour-palette.c: (giza_colour_palette) compiler warning/bug fixed for GIZA_PALETTE=8
(matlab-inspired)
2021-06-18 Daniel Price <[email protected]>
* src/giza-drivers.c: (device) bug fix specifying device with directory path e.g.
dir/file.png
2021-01-22 Daniel Price <[email protected]>
* .github/workflows/api.yml: (doc-bot) second attempt to stop failing github action if no files
committed
2021-01-22 Daniel Price <[email protected]>
* .mailmap: updated .mailmap
2021-01-22 Daniel Price <[email protected]>
* .github/workflows/api.yml: (doc-bot) do not fail build if nothing to commit
2021-01-22 Daniel Price <[email protected]>
* .github/workflows/release.yml, LATEST.md: added release notes and
github workflow for automated release-on-tag
2021-01-22 Daniel Price <[email protected]>
* docs/documentation/installation.html, docs/download/index.html: (docs) updated links
2021-01-22 Daniel Price <[email protected]>
* : commit 91a67f39b63eb537066ec0e584034b36e0f4d98b Author: Daniel
Price <[email protected]> Date: Fri Jan 22 15:23:06 2021
+1100
2021-01-22 Daniel Price <[email protected]>
* docs/documentation/api.html: [doc-bot] updated auto-generated
documentation
2021-01-22 Daniel Price <[email protected]>
* : commit 9ad57ce8a0eff64e9c31462cea960d45c0cfa242 Author: Daniel
Price <[email protected]> Date: Fri Jan 22 04:00:17 2021
+0000
2021-01-22 Daniel Price <[email protected]>
* .github/workflows/api.yml: Create api.yml
2021-01-22 Daniel Price <[email protected]>
* ChangeLog: updated ChangeLog
2021-01-22 Daniel Price <[email protected]>
* doc/api.pl: fix comment about api not being stable (not true, it
is very stable)
2021-01-22 Daniel Price <[email protected]>
* docs/samples/index.html: updated sample images on webpage
2021-01-22 Daniel Price <[email protected]>
* INSTALL, Makefile.am, Makefile.in, aclocal.m4, configure,
src/Makefile.in, test/C/Makefile.in, test/F90/Makefile.in,
test/Makefile.in: updated automake and autoconf, restored necessary
INSTALL file and added rule for README to Makefile.am
2021-01-22 Daniel Price <[email protected]>
* : commit 03c20ab804e59a6d9d247083efacb8bbf4577f0f Author: Daniel
Price <[email protected]> Date: Fri Jan 22 14:22:37 2021
+1100
2021-01-22 Daniel Price <[email protected]>
* Makefile.am: remove INSTALL from Makefile.am
2021-01-22 Daniel Price <[email protected]>
* .github/workflows/build.yml: Update build.yml
2021-01-22 Daniel Price <[email protected]>
* .github/workflows/build.yml: Create build.yml
2021-01-22 Daniel Price <[email protected]>
* README, INSTALL => README.md: (docs) updated README.md for github; include contents of old INSTALL
file
2020-04-03 Daniel Price <[email protected]>
* : Merge pull request #30 from chafarderix/hack Avoid some compile warnings in tests
2020-04-03 chafarderix <[email protected]>
* : Merge pull request #1 from danieljprice/master following upstream
2020-04-02 chafar <[email protected]>
* src/giza-box.c, src/giza-drivers.c, src/giza-format-number.c,
src/giza-private.h, src/giza.h: new giza_set_number_format() sets
format to use by giza_box()
2020-02-17 Daniel Price <[email protected]>
* : Merge pull request #28 from aardk/master For PNG images sometimes the internal logic would classify a single
page
2019-03-20 Daniel Price <[email protected]>
* : Merge pull request #25 from markahutch/master Avoid compilation errors with Splash
2019-03-20 Daniel Price <[email protected]>
* : Merge pull request #26 from haavee/issue-21 Issue 21
2019-01-17 Harro Verkouter <[email protected]>
* src/giza-box-time.c: Fix unused expression result warning A no-op statement was still left in the code; it is now removed
2019-01-17 Harro Verkouter <[email protected]>
* src/giza-box-time.c: Remove C99 for-loop initial declarations There were still two C99-only idioms left which are now rewritten to
use a single, not-quite-as-transient-as-hoped, variable char* to
upcase the x/y option input strings.
2019-01-11 Mark Hutchison <[email protected]>
* src/giza-colour-palette.c: added Matlab style colour palette
2019-01-11 Mark Hutchison <[email protected]>
* src/giza-box-time.c, src/giza-drivers.c: define variables outside
of for loop to avoid the 'initial declarations are only allowed in
C99 mode' error
2018-11-28 Daniel Price <[email protected]>
* src/giza-driver-xw.c, src/giza-format-number.c: more compiler
warnings fixed
2018-11-28 Daniel Price <[email protected]>
* src/giza-box-time.c: (giza-box-time) compiler warnings fixed
2018-11-28 Daniel Price <[email protected]>
* src/giza-box-time.c: (giza-box-time) rename routines to follow internal routine
conventions consistent with rest of giza; whitespaces removed at end
of lines
2018-11-28 Daniel Price <[email protected]>
* : commit fb8c0e00805f73dc7cde7f09d2484df722c887cb Author: Daniel
Price <[email protected]> Date: Wed Nov 28 11:15:29 2018
+1100
2018-11-28 Daniel Price <[email protected]>
* : commit 34cbae10601f412fbdf46de33f47baf433aec317 Author: Daniel
Price <[email protected]> Date: Wed Nov 28 10:11:47 2018
+1100
2018-11-28 Daniel Price <[email protected]>
* : Merge pull request #24 from haavee/issue-21 This should fix issue #21 "missing braces"
2018-11-27 Harro Verkouter <[email protected]>
* src/giza-driver-xw.c, src/giza-drivers.c: Synchronize XW device id
with giza_device_t The global variable 'id' is now also used to address the instance of
the GIZA_XWindow struct where giza_device_t Dev[id]'s properties are
stored. This can be done because: - giza-driver-xw.c implements a shadow array, "XW", of euqal length
as the Dev[...] array (GIZA_MAX_DEVICES - `giza_open_...()` now searches for an unused entry in Dev[...]
when opening a device - `giza_close_device()` closes the correct device when it's, well,
closed. If the entry happened to be an XWindow, the corresponding
XW[id] struct will now also be properly closed.
2018-11-27 Harro Verkouter <[email protected]>
* src/giza-driver-null.c, src/giza-driver-pdf.c,
src/giza-driver-png.c, src/giza-driver-ps.c, src/giza-driver-svg.c,
src/giza-driver-xw.c: Fix device specific close fn's giza_open() now detects and handles errors occuring during setup of
the new device. The device specific _giza_close_<device>() functions
needed to be updated to deal with partially/not fully initialized
devices. For the XWindow driver, its shadow administration of GIZA_XWindow
devices (the "XW[...]" array) needed to be managed slightly
different, not unlinke how devices are now
initialized/de-initialized in giza-drivers.c The global 'xid' variable was removed and replaced by a lookup in
the (already existing) giza_xw_id[...] table. Thinking about it, it seems to me the whole shadow 'xid'
administration can be simplified if Dev[N], if Dev[N] refers to an
XWindow, stores its peculiars in XW[N]. `giza_open()` now already
takes care of finding an empty slot and there should be no reason
for the XWindow driver to use that assigned slot as well. A separate commit implementing that will follow, which makes that
change opt-in.
2018-11-26 Harro Verkouter <[email protected]>
* src/giza-drivers.c, src/giza-private.h,
src/giza-set-font-private.h, src/giza-set-font.c, src/giza.h: Font
memory management changed In order to (1) prevent excessive font unload/load sequences and (2)
proper support for per-device font handling and (3) simplifying font
loading and (4) get a handle on memory management, the giza_set_font()
implementation was changed. Fonts are now cached internally based on the primary key
(font-family, slant, weight) and reference counted across open
devices. Up to three mechanisms of loading a font are tried: FreeType,
FontConfig and cairo_toy_font - depending on cairo version and
wether cairo was compiled with support for that. FreeType/FontConfig
library initialization code made more robust and handling error(s)
and releasing allocated memory in such a case was improved. Extra code to ensure cache unloading at program end was added.
2018-11-23 Harro Verkouter <[email protected]>
* src/giza-box.c, src/giza-cpgplot.c, src/giza-format-number.c,
src/giza-fortran.F90, src/giza.h: `giza_format_number()` string
buffer length The `giza_format_number()` function writes unconditionally to a
destination string buffer without being informed how large that
buffer is. This commit adds an extra parameter to be passed to
`giza_format_number()` - the size of the string buffer. It changes
the implementation to take care of not writing outside the
destination string's boundaries. It should be noted that this may have an effect on
cpgnumb()/PGNUMB() because there is no officially documented
necessity that the *string_length parameter in cpgnumb() has to
contain the size of the buffer upon entry; the documentation only
mentions the contents of that parameter are set on output to the
number of characters written. I have checked the official PGPLOT libcpgplot binding and
perl5-PGPLOT which both do this: both write the size of the FORTRAN
string in the parameter before calling cpgnumb() or use a temporary
variable and pass that size in. I have updated giza-fortran.F90 to
do the same. The Python-PGPLOT binding (ppgplot) does not expose cpgnumb() so at this point that's not an
issue.
2018-11-23 Harro Verkouter <[email protected]>
* src/giza-cpgplot.c, src/giza-drivers.c, src/giza.h: Fix multi dev
handling (SIGSEGV, libc corruption) The giza library, specifically in `giza_open_device_size()` and
`giza_close_device()`, did not handle multiple devices being opened
and closed in random order - which is what my own application
happens to do. The internal bookkeeping got out of sync with the user's device
identification, leading to API calls wreaking havoc because they
would apply to an unitialized giza_device_t entry or overwrite
properties of a giza_device_t they weren't supposed to. (E.g.
cpgopen() could easily overwrite an already open device entry,
cpgclos() could fail because the internal bookkeeping had forgotten
that the user's open device actually existed). Long story short: SIGSEGV and program termination because libc
malloc had detected corruption of its internal structures ensued. In order to fix this a slightly larger than normal edit had to take
place. Entries in the global giza_device_t array are now properly
initialized and de-initialized to a known state, even when opening a
device fails halfway through. PGEND() / cpgend() did not, contrary to documentation, close all
current open devices; now it does.
2018-11-20 Daniel Price <[email protected]>
* docs/documentation/api.html: [web-bot] updated auto-generated
documentation
2018-11-20 Daniel Price <[email protected]>
* configure, configure.ac: updated version number
2018-11-17 Daniel Price <[email protected]>
* : commit 2733835a13c73b272a5ff999f86906b7c8992bcc Author: Harro
Verkouter <[email protected]> Date: Fri Nov 16 17:55:41 2018
+0100
2018-11-15 Harro Verkouter <[email protected]>
* src/giza-drivers.c: Remove _giza_change_size() because it's not used anywhere anymore.
2018-11-15 Daniel Price <[email protected]>
* : Merge pull request #20 from haavee/issue-19 This fixes #19 build failure macports PPC
2018-11-14 Harro Verkouter <[email protected]>
* src/giza-cpgplot.c: Revert "Args swapped between API and giza
implementation" This reverts commit c323737e7d49dd998d068b2964ffe76c86f648c5.
Should be on a completely different branch!
2018-11-14 Harro Verkouter <[email protected]>
* src/giza-driver-xw-private.h, src/giza-driver-xw.c,
src/giza-drivers.c, src/giza-paper.c, src/giza-private.h: Attempt to
be more PGPLOT compatible * PGPAP() does not resize the window immediately anymore, only sets
.resize flag. Only when moving to a new page the change in size
takes effect * PGPAGE() skips asking for confirmation the first time * XWindow specific "_giza_change_size_xw()" removed - now handled
better in giza_set_paper_size(). The resize flag is detected in _giza_change_page_xw() and handled correctly there.
2018-11-12 Harro Verkouter <[email protected]>
* src/giza-driver-xw.c, src/giza-environment.c: Savegame issue #18
2018-11-12 Harro Verkouter <[email protected]>
* src/giza-cpgplot.c: Args swapped between API and giza
implementation giza_render has calling convention giza_render_*( ..., valMin, valMax, ...) whereas PGPLOT API cpggray(..., float fg, float bg, ...) but the parameters 'fg', 'bg' were passed to giza_render in that
order as 'valMin', 'valMax' i.e. fg, bg are reversed
2018-11-09 Harro Verkouter <[email protected]>
* src/giza-box-time.c: This fixes #19 build failure macports PPC I erroneously used scoped for loop variables in a non-100% portable
way.
2018-11-09 Daniel Price <[email protected]>
* : Merge pull request #16 from haavee/symbol-issues Fixes #5 Symbol issues
2018-11-08 Harro Verkouter <[email protected]>
* src/giza-points.c: Scale up symbols #-3..-8 They came out (noticeably) smaller in giza than in PGPLOT as far as
I could tell.
2018-11-08 Harro Verkouter <[email protected]>
* src/giza-points.c: Scaled up the arrows (#28-31) a bit The arrows looked markedly smaller in giza than on the PGPLOT
original. Scaled them up a bit.
2018-11-08 Harro Verkouter <[email protected]>
* src/giza-points.c: Fix #19, update #0, 6 Symbol #19 was not a hexagon with a cross but a larger open square
than #6. The squares are also drawn with a slightly thinner line.
2018-11-08 Harro Verkouter <[email protected]>
* src/giza-points.c: Fix symbols #20 - 27 (inclusive) The open circles are now exponentially scaled and drawn with a
slightly thinner line than before
2018-11-08 Harro Verkouter <[email protected]>
* src/giza-points.c: Fix symbols 12, 18 (5-pt star), #20 initiated Symbols 12 and 18 now are five-point stars, open and filled. Symbol
20 was entered as seven-point star but it should be a small open
circle. Already changed it from _giza_star() => _giza_circle() but
symbols 20-27 are a bit off anyway, so that's going to be fixed in a
different commit.
2018-11-08 Harro Verkouter <[email protected]>
* src/giza-points.c: New line widths only used if you actually
stroke
2018-11-08 Harro Verkouter <[email protected]>
* src/giza-points.c: Fix symbol 14 (open plus sign)
2018-11-08 Harro Verkouter <[email protected]>
* src/giza-points.c: Fix symbol #10 (concave rect)
2018-11-07 Harro Verkouter <[email protected]>
* src/giza-points.c: Fix symbols 7,13,15 The triangles 7, 13 were pointing the wrong way and symbol 15 is the
star of David formed by overplotting two open triangles offset a
little bit
2018-11-07 Harro Verkouter <[email protected]>
* : Merge remote-tracking branch 'upstream/master'
2018-10-31 Daniel Price <[email protected]>
* : Merge pull request #3 from haavee/master giza_parse_string() fix, add cpgqinf() support, fix critical bug
found doing this
2018-09-06 Harro Verkouter <[email protected]>
* src/giza-cpgplot.c, src/giza-drivers-private.h,
src/giza-drivers.c, src/giza-fortran.F90, src/giza.h: Support
cpgqinf(), fix critical bug Found out that the fortron giza pgplot binding implemented PQQINF()
(to query version &cet) but the C-language binding didn't. Whilst implementing cpgqinf() found that giza_query_device() and
_giza_int_to_device() contained a serious bug, with SIGSEGV looming
or loss of information. Both quoted functions copy information into a user allocated string
buffer. Extract of relevant code with 'noise' deleted: _giza_int_to_device (int numDevice, char *DeviceName) { case GIZA_DEVICE_NULL: strncpy(DeviceName, "/null", sizeof(DeviceName)-1); break; .... The user allocated buffers are passed by pointer ("char*
DeviceName") and the 'sizeof(...)' operator returns the size of the
_pointer_ not the size of the buffer it's pointing at! Depending on the platform "sizeof(char*) - 1" evaluates to 3 (32
bit) or 7 (64 bit), irrespective of how many characters the user
really allocated. This can lead to either: - overestimating the amount of characters available in case the
user allocated less than 3 (or 7) bytes for their buffer. A
strncpy() with a source longer than 3 (or 7) would then overwrite
memory that it shouldn't, which in turn would lead to unpredictable
behaviour at best and program termination via SIGSEGV at worst. - underestimating the amount characters available if the user
allocated more than 3 (or 7) characters for their target buffer.
In this case, the strncpy() with a source string > 3 (or 7)
characters would truncate the output - e.g. the file name of a
device would easily be > 7 characters long but at most 3 (or 7) of
them would be copied. This would lead to unnecessary loss of
information. In order to fix this the target buffer size must be explicitly
passed to the functions by the caller. Thus the prototype of
giza_query_device() needed to be altered and as a result the Fortran
wrapper needed adapting to the new signature and pass in the extra
variable accordingly.
2018-09-06 Harro Verkouter <[email protected]>
* src/giza-scanner.l, src/lex.yy.c: Fix backspace handling ... The meaning of the nGlyph index used in giza_parse_string changed
subtly during development and the handling of backspace was not
changed accordingly. The result was that "x^2\b_i" rendered fine but the unit superscript
over the decimal point in PGTBOX() now looked horrible because the
pen position was moved to an erroneous position. Sigh. In order to make "x-i-squared" render nicely it must now be spelled
as: x^2\b\b_i with two backspaces. The first backs up over the
superscript '2', the second backs up over the '^' superscript
operation; when entering the superscript the baseline of the font
and the current pen position is changed immediately. So if that is
not undone before staring to render further characters, the pen
position is not where you hoped it would be.
2018-09-06 Daniel Price <[email protected]>
* docs/news/index.html: updated docs
2018-09-06 Daniel Price <[email protected]>
* docs/documentation/api.html,
docs/documentation/cpgplot-status.html,
docs/documentation/pgplot-status.html: [web-bot] updated
auto-generated documentation
2018-09-06 Daniel Price <[email protected]>
* configure: updated configure script
2018-09-06 Daniel Price <[email protected]>
* AUTHORS, README, configure.ac, src/giza-cpgplot.c,
src/giza-pgplot.f90: updated docs and version number to following
merge
2018-09-06 Daniel Price <[email protected]>
* src/giza-points.c: (giza_points) revert marker size to behaviour on master
2018-09-06 Daniel Price <[email protected]>
* src/giza-driver-xw.c: (xw) spaces at end of lines removed
2018-09-06 Daniel Price <[email protected]>
* : Merge https://github.com/haavee/giza into haavee-master
2018-09-05 Harro Verkouter <[email protected]>
* : commit 553bf85db2f60aaf908113322b0cae4c0d0836de Author: Harro
Verkouter <[email protected]> Date: Wed Sep 5 15:35:47 2018
+0200
2018-09-05 Harro Verkouter <[email protected]>
* src/giza-set-font.c, src/giza.h: giza_set_font_*(...) take const
char* Sometimes compile time static strings are fed into
giza_set_font*(...) calls but their prototype is "char *", implying
mutable strings. None of the functions actually modify their argument so it probably
is better to make this explicit in their prototypes.
2018-09-04 Harro Verkouter <[email protected]>
* src/giza-driver-xw.c, src/giza-drivers.c: Better XWindow resizing
implementation It turned out that the XWindow resizing wasn't handled correctly in
all cases; notable pgdemo13 (with two x-windows) exposed a problem
in the code in the first window, the one with the choice menu. More often than not it would show ellipses where circles should be
drawn and the text was (way) ouside the window. What happened was two calls to giza_change_page_xw() quickly after
each other. The first had Dev[id].resize = 1 (from PGPAP(...) in
pgdemo13), which triggered an XResizeWindow(...) in the code, to
change the size of the XWindow to the requested device size. The
next call entered with Dev[id].resize = 0 (because that was already
handled) ... but that one detected a window geometry change! Upon
inspection it turned out that, as far as X11 was concerned, the
window still had the /old/ size whilst our internal bookkeeping was
already updated with the requested size. Turns out that the XResizeWindow() does not take effect immediately.
So the code now waits for ConfigureNotify events until the requested
window size has been achieved. Now it works reliably and windows can
be resized between plotting.
2018-08-30 Harro Verkouter <[email protected]>
* src/giza-io-private.h, src/giza-io.c: Enable formatted
warning/error messages (compat) By changing _giza_warning() and _giza_error() prototype (and
implementation) slightly it is now possible to make the 'message'
parameter be a formatting string (cf. printf("formatting string %s =
%d", s, i)) Now this is possible: _giza_warning("giza_parse_string", "Invalid Hershey number %d",
hershey); _giza_error("giza_parse_string", "Unrecognized token
#%d '%s'", token, yytext); Existing _giza_warning()/_giza_error() calls are completely
unaffected by this.
2018-08-30 Harro Verkouter <[email protected]>
* src/giza-driver-xw.c: no XWindow resize waiting for keypress Whilst the system is waiting for a keypress, do not honour window
resizing requests. PGPLOT also does not do this.
2018-08-30 Harro Verkouter <[email protected]>
* src/giza-box.c: Fixed earlier broken 'repair' Sometimes the PGPLOT doc is somewhat ambiguous but not for 'N' and
'M' options: both imply labels have to be drawn, only 'M' sais to do
so in the unconvential location. Ambiguity example: does 'L' (logarithmic labelling) *imply* 'N' or
'M' ("write labels")? Or is 'L' only honoured if either 'N' or 'M'
is specified (which is how the code currently works).
2018-08-30 Harro Verkouter <[email protected]>
* src/giza-box.c: Add support for 'M' option to PG(T)BOX When it comes to writing labels PG(T)BOX supports the following: -N :- Label the axis (conventional, below/left viewport) -M :- Put labels in the unconvential location (above/right
viewport) The 'M' option was not implemented by giza so I added it. Whilst at
it removed a few levels of somewhat unnecessary if()'s by adding a
high-level check and sometimes continue'ing from a loop if the rest
of the loop body is precluded from being executed anyway.
2018-08-30 Harro Verkouter <[email protected]>
* src/giza-box-time.c: Fixed minor bugs (logic, off-by-one, missing
/) There were two off-by-one lost-in-FORTRAN-to-C-translation: - using number of items as last accessible index ok in fortran, not so much in C (a classic) - end condition off by one because of pretty much the same in fortran "<" must be <= in C (if C-style counting with fortran meaning of index) I condensed what I thought three levels of fortran IF .. END IF into
a single C "if( x && y && y)" but missed that the 3rd IF was a
single-statement and did not introduce the 3rd 'and' ... IF x THEN IF y THEN IF z <stuff> <other stuff> END IF END IF at first erroneously translated by me into if( x && y && z ) { <stuff> <other stuff> } now reads if( x && y ) { if( z ) <stuff> <other stuff> } And finally in the specification of the super/subscript label
formats I had forgotten to escape the backslash with a backslash in
the strings.
2018-08-30 Harro Verkouter <[email protected]>
* src/giza-box.c: y-axis ticks on the right were drawn wrong side Obviously the previous edit introduced a sign error :-) If y-axis
ticks were drawn on the right hand side they'd be drawn the same
direction as the ones on the left-hand side. Fixed.
2018-08-30 Harro Verkouter <[email protected]>
* src/giza-box.c: Support for PGBOX 'I' and 'P' options and bug fix giza's pgbox(...) routine did not support the PGPLOT options 'I' and
'P': -I: Invert tick marks - write them outside the viewport in
stead of inside (default) -P: Project major tick marks outside the viewport (ignored if 'I'
also specified) Fixed bug where tickmarks on 'the other side' of the frame were
drawn in the same direction but they need to be inverted: If tick marks on the bottom are drawn inwards (i.e. up) then at the
top they must be drawn downwards to achieve the same effect. The
existing code drew the top tick marks outside the viewport because
it would draw them 'up' as well.
2018-08-30 Harro Verkouter <[email protected]>
* src/giza-box.c: Fix axis ticks not being drawn if axis 'reversed' In checking pgdemo1 certain axes did not display any tick marks at
all! This was traced to be when the window set through pgswin() was
such that Dev[id].Win.[xy]max < Dev[id].Win.[xy]min, i.e. the axis
is reversed: the values decrease towards the right (X) or top (Y)
2018-08-22 Harro Verkouter <[email protected]>
* : commit c00f33c3c943fba12f94916ed468f76afdd76bd1 Author: Harro
Verkouter <[email protected]> Date: Fri Aug 17 17:38:56 2018
+0200
2018-08-16 Harro Verkouter <[email protected]>
* src/giza-cursor-routines.c: fix pgdemo5 misbehaving of
pgolin/pglcur pgdemo5 with giza master branch draws all subsequent polygons in
black whereas PGPLOT draws subsequenct polygons in different colours
by incrementing colour index at each new polygon.
2018-08-16 Harro Verkouter <[email protected]>
* src/giza-viewport.c: Revert "fix PGPLOT incompatibility" This reverts commit e9644f6b098e54fb2cd246434d8d1c58fc6e1ba1.
Breaks pgdemo2 (pgvsiz() function). It was not specifically a bug or
fix and it doesn't hurt keeping the original code but it hurts not
keeping the original version ... ;-)
2018-08-16 Harro Verkouter <[email protected]>
* src/giza-driver-xw.c, src/giza-drivers.c: revert to drawing to
pixamp, XCopyArea to window Turns out there's code relying on drawing onto the pixmap and in
general, direct drawing to the window is probably best avoided
2018-08-16 Daniel Price <[email protected]>
* : commit da2f2eca704c4002386610f8f5b344f5f9465e17 Merge: 6c8ce0d
99aaa05 Author: Daniel Price <[email protected]> Date: Thu
Aug 16 10:35:33 2018 +1000
2018-08-15 eee software boss <[email protected]>
* src/giza-box-time.c: fix sigsegv in pgtbox() Using printf()'s Parameter field(*) I used 0-based counting for the
argument replacement fields but it has to be 1-based for argument 0
is the format string itself ... (*)
https://en.wikipedia.org/wiki/Printf_format_string#Parameter_field
2018-08-14 Harro Verkouter <[email protected]>
* src/giza-print-id.c: Fix compiler warning prototype for sprintf() missing
2018-08-14 Harro Verkouter <[email protected]>
* src/giza-cpgplot.c: fix off-by-one between external/internal
device id When using multiple open devices simultaneously, the returned device
id from cpgoen should be usable to later select that open device.
There was an extra +/- 1 mismatch between cpgopen and cpgslct.
2018-08-14 Harro Verkouter <[email protected]>
* src/giza-drivers-private.h, src/giza-drivers.c, src/giza-io.c: fix
compiler warning casting const away When splitting the device string sometimes 'const char*' was
assigned to 'char *' - which the compiler doesn't like. Fixed by changing prototype and destination data type of the split
string.
2018-08-14 Harro Verkouter <[email protected]>
* src/giza-io.c: fix warning about unused return value fgets() returns an error code and the caller is expected to check
that return valuer; compiler verifies this.
2018-08-14 Harro Verkouter <[email protected]>
* src/giza-box-time.c: Support for pgtbox() The existing pgtbox() redirected to pgbox(). For astronomy application(s) time labelling is very important and a
highly desirable property. The existing PGTBOX() from the PGPLOT source code was ported into an
initial working version, behaving identically, modulo porting bugs.
2018-08-14 Harro Verkouter <[email protected]>
* src/giza-scanner.l, src/lex.yy.c: Support for backspace (\b) in
printing strings For time labelling to look good it is 'necessary' to make the
superscript appear on top of e.g. the decimal point. This can be
done if 'backspace' is supported - backing up the text by one or
more characters. This necessitated rewriting the giza_parse_string() a bit because in
the current version backing up to a previous character's position
was impossible.