-
Notifications
You must be signed in to change notification settings - Fork 12
/
ChangeLog
10589 lines (7693 loc) · 361 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
2014-09-30 Gavin SALAM <[email protected]> + Gregory
* Release of FastJet 3.1.0
2014-09-30 Gregory Soyez <[email protected]>
* tools/fastjet/tools/Recluster.hh:
put the Recluster tool in the "generic tools" doxygen module
2014-09-29 Gavin SALAM <[email protected]> + Gregory
* Doxyfile
* configure.ac
* doc/fastjet-doc.tex
* CHECKLIST
* include/fastjet/config_win.h
updated version number to 3.1.0 in preparation for release
siscone now points to 3.0.0
* include/fastjet/Selector.hh:
fixed comment for InvalidArea (+minor typo fix)
* Doxyfile:
made sure certain files like README etc are linked to from the
file list.
* AUTHORS:
updated list of people who have helped out
2014-09-29 Gregory Soyez <[email protected]>
* include/fastjet/ClusterSequence.hh:
placed _Line and _Parabola in the "internal" doxygen doc
2014-09-25 Gregory Soyez <[email protected]> + Matteo
* NEWS:
further work on the 3.1.0 release notes
2014-09-24 Gregory Soyez <[email protected]>
* NEWS:
preliminary version of the 3.1.0 release notes
2014-09-18 Gavin Salam <[email protected]>
* configure.ac:
fixed typo in --enable-demangling help message
* regression-tests/nightly-check.pl:
added minimal check that --enable-demangling doesn't break anything
(commit done only 2014-09-29)
2014-09-18 Matteo Cacciari <[email protected]>
* regression-tests/nightly-check.pl
Added a test with gcc 4.9
2014-09-18 Gregory Soyez <[email protected]>
* doc/fastjet-doc.tex:
. in the WTA section, added a footnote saying that WTA_E_scheme
is sick and therefore not implemented
. added a few words here and there to highlight the fact that
"SISCone" means SISCone with SM (better avoid any confusion)
* plugins/SISCone/fastjet/SISConeBasePlugin.hh:
added a note that we recommend to implement
UserScaleBase::is_larger() whenev possible.
2014-09-18 Matteo Cacciari <[email protected]>
* src/Error.cc
modified the guard to LimitedWarning Error::_execinfo_undefined;
so that it's the same that is now used elsewhere and in Error,hh,
i.e. #if (!defined(FASTJET_HAVE_EXECINFO_H)) || defined(__FJCORE__)
Note that #ifndef FASTJET_HAVE_EXECINFO_H would actually be sufficient.
2014-09-18 Gregory Soyez <[email protected]>
* include/fastjet/Error.hh
* src/Error.cc:
restored some FJCORE guards to keep it more lightweight
* include/fastjet/Error.hh
* src/Error.cc:
fixed a few details:
. included the config.h header in Error.hh
. defined _demangle in the header
. removed unnecessary __FJCORE__ guards (since
FASTJET_HAVE_EXECINFO_H) is not defined in FJCORE
This allows to used -DFASTJET_HAVE_EXECINFO_H on
the command line.
* include/fastjet/RectangularGrid.hh:
added '_in' suffix to a few ctor arguments to prevent a series of
Wshadow warnings
2014-09-18 Matteo Cacciari <[email protected]>
* include/fastjet/Error.hh
* src/Error.cc
* scripts/mkfjcore.sh
- demangle() included in the Error class
- Added a warning to Error:set_print_backtrace in case
backtrace is asked for by the user but EXECINFO is not
available
- tweaked mkfjcore.sh to make the backtrace warning work
properly in fjcore
2014-09-11 Gavin Salam <[email protected]>
* src/ClusterSequence.cc:
now throws Error in case of use of N2MHTLazy9AntiKtSeparateGhosts
from fjcore (since it's no longer supported)
2014-09-11 Gregory Soyez <[email protected]>
* src/Error.cc:
added a missing FJCORE guard
* src/Error.cc:
* configure.ac:
* config.h.in:
* m4/demangle.m4: *** ADDED ***
added demangling of the backtrace symbols to the backtrace. This
is only activated if (in order of the tests)
. execinfo.h is found
. --enable-demangling is specified at configure time
. the GNU C++ ABI is supported
2014-09-10 Gregory Soyez <[email protected]>
* include/fastjet/JetDefinition.hh:
* src/JetDefinition.cc:
commented out the WTA_E_scheme
Energy-ordering can lead to dangerous situations with particles at
rest. We instead implement the WTA_modp_scheme
2014-09-08 Matteo Cacciari <[email protected]>
* src/Error.cc
* scripts/mkfjcore.sh
Added FJCORE guards around the backtrace block and the corresponding
include in Error.cc, so that they are never included in fjcore
(issues under Windows with MSVC, as reported by
Chaowaroj Wanotayaroj <[email protected]>). Also fully cleaned away
the EXECINFO guards from config_auto.h (not really necessary though)
* src/ClusterSequence.cc
Added FJCORE guards around Lazy9AntiKtSeparateGhosts strategy so
that it's never included in fjcore. Eliminated the files accordingly
from mkfjcore.sh
2014-09-08 Gregory Soyez <[email protected]> + Matteo + Gavin
* tools/fastjet/tools/Subtractor.hh:
* tools/Subtractor.cc:
added a Subtractor(rho, rho_m) constructor + worked on some error messages
* example/07-subtraction.cc:
added the mass to the printed output
* plugins/D0RunICone/fastjet/D0RunIConePlugin.hh:
* plugins/D0RunICone/fastjet/D0RunIpre96ConePlugin.hh:
* tools/fastjet/tools/Subtractor.hh:
explicitly included base.hh to help Doxygen [likely a bug in Doxygen!]
* src/JetDefinition.cc:
For WTA_modp scheme, added a safeguard that prevents divisions by
zero if both objects are at rest (the energy of the sum will
therefore be the mass of pa)
For WTA_E scheme, if the hardest object is at rest, keep the sum
at rest and set the energy to the sum of energies: this implies
that we break the rule of the mass being that of the higher-energy
particle.
2014-09-05 Gavin Salam <[email protected]>
* include/fastjet/JetDefinition.hh:
amplified doxygen documentation on WTA schemes. Also mentioned
that all pt^2, WTA, etc. schemes are collinear safe only for
algorithms with an appropriate recombination order.
* plugins/SISCone/SISConeSphericalPlugin.cc (description):
* plugins/SISCone/SISConePlugin.cc (description):
small tweak so that if there is a user scale for progressive
removal and it has a description, that description gets printed
out as part of the SISCone description
2014-09-05 Gregory Soyez <[email protected]>
* example/02-jetdef.cc:
mentioned the WTA schemes in the example
2014-09-04 Gregory Soyez <[email protected]>
* src/JetDefinition.cc:
implemented the recombination for WTA_{E,modp}_scheme
* includec/fastjet/JetDefinition.hh:
* src/JetDefinition.cc:
started the addition of WTA recombination schemes
. WTA_pt_scheme fully done
. WTA_{E,modp}_scheme have a description and a (inexistant)
preprocess step but still lack the actual recombination
2014-09-04 Gregory Soyez <[email protected]> + Gavin
* plugins/SISCone/SISConeBasePlugin.cc: *** ADDED ***
* plugins/SISCone/Makefile.am:
* plugins/SISCone/makefile.static:
SISConeBasePlugin::UserScaleBase::is_larger now uses the cached
value and has been moved in a proper .cc file (allows to declare
StructureType at the end of the class)
* plugins/SISCone/fastjet/SISConeSphericalPlugin.hh:
* plugins/SISCone/SISConeSphericalPlugin.cc:
implemented the user-defined scale support in the spherical
version too
* plugins/SISCone/SISConePlugin.cc:
* plugins/SISCone/fastjet/SISConePlugin.hh:
* plugins/SISCone/fastjet/SISConeBasePlugin.hh:
. worked on the description of the plugin
. moved the UserScaleBase class to SISConeBasePlugin
derived plugins now use a template specialisation
SISConePlugin::UserScaleBaseStructureType<type_of_siscone_jet>
* plugins/SISCone/SISConePlugin.cc:
* plugins/SISCone/fastjet/SISConePlugin.hh:
* plugins/SISCone/fastjet/SISConeBasePlugin.hh:
carried on with the support of user-defined scales
[current version under testing].
2014-09-04 Gregory Soyez <[email protected]>
* plugins/SISCone/SISConePlugin.cc:
* plugins/SISCone/SISConeSphericalPlugin.cc:
* plugins/SISCone/fastjet/SISConePlugin.hh:
* plugins/SISCone/fastjet/SISConeSphericalPlugin.h:
added support for SISCone with progressive removal (use
plugin.set_progressive_removal() to activate)
started the work towards support for user-defined scale to decide
which is the hardest stable cone in progressive-removal
mode. [breaks compilation for now... committed to carry on
development with Gavin]
2014-09-03 Gavin Salam <[email protected]> + Gregory
* plugins/SISCone/SISConePlugin.cc (run_clustering):
* plugins/SISCone/SISConeSphericalPlugin.cc (run_clustering):
added comment to explain that jets are transferred such that
CS::inclusive_jets() returns them in the order they were created
by SISCone.
2014-09-03 Gavin Salam <[email protected]>
* src/PseudoJet.cc (join):
* include/fastjet/CompositeJetStructure.hh (join):
in join with a fixed number of input PseudoJets, reserved space
ahead of time in the temporary pieces vector.
2014-08-15 Gavin SALAM <[email protected]> + Matteo
* Doxyfile
* configure.ac
* doc/fastjet-doc.tex
* CHECKLIST
* NEWS
switched version number back to 3.1.0-devel
2014-08-15 Gavin SALAM <[email protected]> + Matteo
* Doxyfile
* configure.ac
* doc/fastjet-doc.tex
* CHECKLIST
* NEWS
update of version number for
Release of FastJet fastjet-3.1.0-beta.1
2014-08-15 Gregory Soyez <[email protected]>
* tools/Filter.cc:
added a few safeguards for situations where there is no subjets
2014-08-15 Gavin Salam <[email protected]> + Matteo
* tools/fastjet/tools/Recluster.hh:
switched default for acquire_reombiner_in in constructor with
new_jet_def so that it is now false (i.e. the jet_def is used
exactly as it is supplied).
2014-08-15 Gavin Salam <[email protected]>
* regression-tests/test-contrib.pl: *** ADDED ***
added script to facilitate testing of fjcontrib
2014-08-14 Gavin Salam <[email protected]>
* tools/Recluster.cc:
* tools/fastjet/tools/Recluster.hh:
added functions set_acquire_recombiner(...) and set_keep(...) to
control behaviour after construction.
Fixed issue in case where there are no subjets and keep() is false
(now returns empty jet).
In comments and variable nameing replaced subjet_def ->
new_jet_def and subjets with inclusive jets.
Various other small changes to comments.
2014-08-13 Gavin Salam <[email protected]> + Matteo
* plugins/CDFCones/CDFcode/JetCluAlgorithm.cc:
* plugins/CDFCones/CDFcode/MidPointAlgorithm.cc:
small changes, labelled in code, to avoid compiler warnings about
unused variables and comparisons between unsigned int and int
* src/AreaDefinition.cc:
replaced cerr + exit with a throw.
* src/LimitedWarning.cc:
introduced max_warn() and n_warn_so_far() functions and changed
warning text so that it says "WARNING from FastJet" rather than
just "WARNING".
* include/fastjet/ClusterSequence.hh:
* src/ClusterSequence.cc (exclusive_jets_up_to):
rephrased warning for exclusive jets to reflect the fact that it
may be used with C/A and genkt. Also switched it over to the
LimitedWarning system.
* include/fastjet/internal/DynamicNearestNeighbours.hh:
DnnError now derives from fastjet::Error
2014-08-13 Gavin Salam <[email protected]>
* include/fastjet/RectangularGrid.hh:
* src/RectangularGrid.cc:
made sure that all member variables are set even for the default
constructor (issue revealed by Coverity when testing SoftKiller).
Added is_initialised() member to base class and RectangularGrid,
to allow classes using the grid to check the grid is sensible.
* tools/fastjet/tools/GridMedianBackgroundEstimator.hh:
throws exception if contructed with an uninitialised grid.
* plugins/GridJet/GridJetPlugin.cc:
throws on uninitialised grid.
* testsuite/run_tests.cc:
* testsuite/TestGrids.hh: *** ADDED ***
added a few basic tests of the grid initialisation, GridJetPlugin,
etc.
2014-08-13 Matteo Cacciari <[email protected]>
Files manually merged from the 3.0.X branch:
* scripts/mkfjcore.sh
* scripts/Makefile-fjcore.txt *** ADDED ***
* scripts/preamble-fjcore.txt
switched mkfjcore to -O2 (flag also added in Makefile-fjcore.txt)
Added new Lazy* TilingExtent and files to mkfjcore.sh
2014-08-13 Gavin SALAM <[email protected]> + Matteo & Gregory
* NEWS:
reshuffled entries to try to get an order reflecting importance of
different changes.
2014-08-12 Gregory Soyez <[email protected]> + Gavin
* example/07-subtraction.cc:
switched on use_rho_m and _safe_mass
* NEWS:
worked on the list of changes since FastJet-3.0.6
* include/fastjet/internal/LazyTiling9SeparateGhosts.hh:
* src/LazyTiling9SeparateGhosts.cc:
declared ghost_pt2_threshold as a static (public) member of the
class. [with a note that this is use-ar-your-own-risk material]
* tools/Subtractor.cc (_amount_to_subtract):
small change to warning message about non-zero rho_m.
2014-08-12 Gavin SALAM <[email protected]> + Gregory
* plugins/GridJet/GridJetPlugin.cc:
* plugins/GridJet/fastjet/GridJetPlugin.hh:
switched this over to the use of RectangularGrid internally, plus
a new constructor that takes a RectangularGrid. Added code to
handle cases where some grid tiles are deselected. Also removed an
apparently redundant copy constructor.
Passes the test-all-algs tests, and also examined behaviour
with deselected tiles in
issue-tracker/2014-08-08-testing-rect-grid/test-gridjet.cc
* include/fastjet/RectangularGrid.hh:
added rapmin(), rapmax(), drap() and dphi() enquiry functions.
2014-08-12 Gregory Soyez <[email protected]>
* tools/Subtractor.cc:
* tools/fastjet/tools/Subtractor.hh:
* doc/fastjet-doc.tex:
used safe_mass instead of safe (or be_safe)
2014-08-12 Gavin Salam <[email protected]>
* src/LimitedWarning.cc:
* include/fastjet/LimitedWarning.hh:
introduced versions of warn that take a const char *, in addition
to the existing ones with a const string &, so that calls of the
form warn("some message") don't need to create an intermediate
string (which could be slow). The master code is now the char *
version, and the string version just pass the resulting c_str().
* tools/fastjet/tools/Subtractor.hh:
* tools/Subtractor.cc (_amount_to_subtract):
added a warning for cases where rho_m is available and
substantially non-zero, but not used.
2014-08-12 Gregory Soyez <[email protected]>
* tools/Subtractor.cc:
fixed typo in the treatment of CHS events
* configure.ac:
used -O2 rather than -O3 as a default (more stable timings. O3 was
only giving a small gain, sometimes being worse than -O2 and being
much more compiler-version-dependent than O2)
* tools/Filter.cc:
reinserted the missing bit of code needed to include internal
subtraction!
2014-08-11 Gavin Salam <[email protected]> + Gregory & Matteo
* include/fastjet/PseudoJet.hh:
* src/PseudoJet.cc:
operator* and operator*= now premptively call
_ensure_valid_rap_phi() so that we can always ensure that rap()
and phi() will be conserved after a *= operation (previously, if
they hadn't yet been cached, and you multiplied by 0, you would
get a non-sensical rap() and phi()).
2014-08-11 Gavin Salam <[email protected]>
* scripts/preamble-fjcore.txt:
small tweaks to fjcore preamble
2014-08-11 Gregory Soyez <[email protected]>
* tools/Subtractor.cc:
* tools/fastjet/tools/Subtractor.hh:
worked on the description; tidied a few loose ends.
* tools/Subtractor.cc:
* tools/fastjet/tools/Subtractor.hh:
added several options to the original Subtractor:
. support for rho_m (set_use_rho_m(bool)) [off by default]
. support for mass-safety tests (set_safe(bool)) [off by default]
. support for CHS events using
set_known_selectors(const Selector &sel_known_vertex,
const Selector &sel_leading_vertex)
2014-08-11 Gavin SALAM <[email protected]> + Gregory
* src/RectangularGrid.cc:
* include/fastjet/RectangularGrid.hh:
index() -> tile_index()
is_good() -> tile_is_good()
* tools/GridMedianBackgroundEstimator.cc:
* tools/fastjet/tools/GridMedianBackgroundEstimator.hh:
took into account above name changes
2014-08-08 Gavin Salam <[email protected]> (later)
* include/fastjet/RectangularGrid.hh:
* src/RectangularGrid.cc:
constructor can now take a Selector that causes only certain tiles
to be considered "good"
* tools/fastjet/tools/GridMedianBackgroundEstimator.hh
* tools/GridMedianBackgroundEstimator.cc:
Added constructor that just takes a RectangularGrid.
Implemented support for cases where some tiles are discarded
through the mechanism of Selectors for good tiles.
Tests in issue-tracker/2014-08-08-testing-rect-grid
2014-08-08 Gavin Salam <[email protected]>
* include/fastjet/RectangularGrid.hh:
added mean_tile_area() and in the base class, by default
tile_area(itile) returns mean_tile_area() (since
all_tiles_equal_area() also returns true).
* tools/GridMedianBackgroundEstimator.cc:
* tools/fastjet/tools/GridMedianBackgroundEstimator.hh:
switched GMBGE over to the RectangularGrid class (enabled with
the FASTJET_GMBGE_USEFJGRID symbol). See
2013-particle-level-sub/fj31-devel-timing-tests for timing
tests, which suggest a slight advantage for the new setup.
For SoftKiller there appears to be little difference.
2014-08-04 Gavin Salam <[email protected]>
* include/fastjet/RectangularGrid.hh:
* src/RectangularGrid.cc:
* & corresponding Makefile.am files
added a new RectangularGrid class, intended to provide a common
core for at least some of the uses we have of rectangular grids.
2014-08-02 Gavin Salam <[email protected]>
* regression-tests/test-all-algs.pl (setRefResults):
for areas checksums, added a comment about results coming out
differently on Gavin's laptop, and in one case checked that this
is due to a single last digit change.
* src/ClusterSequence.cc (_best_strategy):
added handling of cambridge_for_passive_algorithm (treat it like
kt). This resolves today's nightly-check failure on cambridge
passive areas.
* include/fastjet/JetDefinition.hh:
noted in comments that cambridge_for_passive_algorithm is not
intended for the end-user, but rather for internal games.
Also added a comment that genkt_for_passive_algorithm is not
currently implemented.
* regression-tests/test-all-algs.pl:
now handles comma-separated strategy lists (not just colon-separated).
2014-08-01 Gavin SALAM <[email protected]>
* testsuite/TestRecombiners.hh:
added some very basic tests of recombiners, that setting to a
default recombination scheme works sensibly and that copying a
pointer to a recombination scheme also works. OK also with
valgrind.
2014-08-01 Gregory Soyez <[email protected]>
* tools/GridMedianBackgroundEstimator.cc:
* tools/fastjet/tools/GridMedianBackgroundEstimator.hh:
replaced _disable_rho_m with _enable_rho_m internally
* tools/JetMedianBackgroundEstimator.cc:
* tools/fastjet/tools/JetMedianBackgroundEstimator.hh:
replaced _disable_rho_m with _enable_rho_m internally
+ has_rho_m now also returns false if there is a density class
2014-08-01 Gavin SALAM <[email protected]> & Gregory
* tools/fastjet/tools/JetMedianBackgroundEstimator.hh:
* tools/fastjet/tools/GridMedianBackgroundEstimator.hh:
disable_rho_m() replaced with set_compute_rho_m()
* tools/fastjet/tools/BackgroundEstimatorBase.hh:
changes to comments
* tools/fastjet/tools/Recluster.hh:
* tools/Recluster.cc:
work on comments and also slight change to interface relating to
C/A optimisation.
* src/ClusterSequence.cc:
_best_strategy will now assert(false) if the jet alg is not one of
kt, antikt, cam or genkt, and drops back to kt or antikt choices
depending on genkt's p value.
The routine that calls it now replaces the NlnN strategy with
N2MHTLazy25 if NlnN isn't available.
2014-08-01 Gregory Soyez <[email protected]>
* src/ClusterSequence.cc:
best_strategy() now follows the transitions sugegsted by the
latest timing run
* include/fastjet/Selector.hh:
* src/Selector.cc:
added Selector::scalar_pt sum that sums the pt of the objecs
passing the Selector.
2014-07-31 Gregory Soyez <[email protected]>
* tools/JetMedianBackgroundEstimator.cc:
fix: enabled rho_m in reset()
do not compute rho_m if a jet density class is provided.
fix: critical "pt" v. "dt" typo in _compute()
* tools/JetMedianBackgroundEstimator.cc:
* tools/fastjet/tools/JetMedianBackgroundEstimator.hh:
added support for rho_m, sigma_m (on by default but can be turned
off using disable_rho_m())
TODO: disable rho_m automatically when a density class is provided
[do not forget to set it back if an empty density class is passed]
* tools/fastjet/tools/BackgroundEstimatorBase.hh:
fixed doxygen comment
* tools/GridMedianBackgroundEstimator.cc:
* tools/fastjet/tools/GridMedianBackgroundEstimator.hh:
. no longer cache the scalar pt in each grid cell (<~5% effect on
timing, omitting the call to _percentile in the version with
caching)
. added support for rho_m, sigma_m (on by default but can be
turned off using disable_rho_m())
* tools/fastjet/tools/BackgroundEstimatorBase.hh:
added rho_m(), sigma_m(), rho_m(jet), sigma_m(jet) [throw by
default] and has_rho_m() [false by default].
2014-07-30 Gregory Soyez <[email protected]>
* tools/fastjet/tools/Recluster.hh:
worked on the description of the class
* tools/fastjet/tools/Filter.hh:
properly handled the case where C/A optimisation has been used and
the jet has a non-explicit-ghost jet area
* tools/Recluster.cc:
no longer keep track of the jet definition used to obtain the
subjets. It can be accessed through the subjets themselves.
* src/JetDefinition.cc:
reinserted the missing genkt_algorithm in n_parameters_for_algorithm
* tools/Filter.cc :
adapted to use the Keep enum in Recluster.
TODO: fix the "discard_area" part for non-explicit-ghost areas
with C/A reclustering
Note: we could get rid of the sub jet def returned by
Recluster::get_new_jets_and_def since it can be obtained from the
subjets.
* tools/fastjet/tools/Recluster.hh:
* tools/Recluster.cc:
. added an enum (Keep) that takes values keep_only_hardest or
keep_all, replacing the 'single' member
. moved the ctor structure to
Recluster(JetDef, acquire_recombiner = true,
Keep keep_in = Recluster::keep_only_hardest)
Recluster(JetAlg, R, keep_in = Recluster::keep_only_hardest)
Recluster(JetAlg, keep_in = Recluster::keep_only_hardest)
simplified quite a bit the structure accordingly
. used the new description functionalities in JetDefinition
. added Recluster::disable_cambridge_optimisation to disable the
attempt to optimise for C/A
. introduced get_new_jets_and_def(...) and generate_output_jet().
These are the two steps of result() and this should ease the
Filter.
* include/fastjet/JetDefinition.hh:
* src/JetDefinition.cc:
added JetDefinition::n_parameters_for_algorithm (static),
JetDefinition::algorithm_description (static) and
description_no_recombiner().
Slightly simplified other bits of the code accordingly.
2014-07-29 Gregory Soyez <[email protected]>
* include/fastjet/ClusterSequenceArea.hh:
replaced an error message printed to stdout (followed by exit(-1))
by an error thrown when the area type is unknown.
Note that this uses ostringstream, so we could try to move that to
the cc (or introduce AreaDefinition::area_type_str())
2014-07-29 Gavin Salam <[email protected]>
* include/fastjet/JetDefinition.hh:
* src/JetDefinition.cc:
moved new set_recombiner code into the .cc file, included a test
of the JetDef invariants and ensured that recombination_scheme()
returns correct result in case of external recombiners
* testsuite/TestRecombiners.hh:
* testsuite/run_tests.cc:
added first elements of testing for Recombiners.
2014-07-29 Gregory Soyez <[email protected]>
* tools/Recluster.cc:
* tools/Pruner.cc:
* tools/fastjet/tools/Recluster.hh:
* tools/fastjet/tools/Pruner.hh:
arranged the code to use JD::set_recombiner(const JetDefinition&)
instead of JD::set_recombiner(Recombiner*)
* include/fastjet/JetDefinition.hh:
introduced set_recombiner(const JetDefinition&) that correctly
copies the recombiner from another jet definition. When
applicable, this version is strongly recommended as the
alternative (set_recombiner(Recombiner*)) can lead to memory
corruption if the recombiner is owned by the jet definition.
removed set_shared_recombiner and shared_recombiner (superseded
by the above option)
2014-07-29 Gavin SALAM <[email protected]> + Gregory
* src/LazyTiling25.cc
* src/LazyTiling9.cc
* src/LazyTiling9Alt.cc
* include/fastjet/internal/LazyTiling25.hh
* include/fastjet/internal/LazyTiling9.hh
* include/fastjet/internal/LazyTiling9Alt.hh
replace is_near_zero_phi() test, based on double-precision
arithmetic for each tiles with a use_periodic_delta_phi bool for
each tile. This appears to solve the
2014-07-26-Lazy9-largeR-clustering issue. (Tests ongoing with 10^7
events).
* src/ClusterSequence_Delaunay.cc (_delaunay_cluster):
reworked comments around the assert for cases where the clustering
strategy has not been recognized (all cases for which this would
be called should be recognized).
2014-07-29 Gregory Soyez <[email protected]>
* plugins/SISCone/SISConePlugin.cc:
* plugins/SISCone/SISConeSphericalPlugin.cc:
* plugins/SISCone/fastjet/SISConeBasePlugin.hh:
* plugins/SISCone/fastjet/SISConePlugin.hh:
* plugins/SISCone/fastjet/SISConeSphericalPlugin.hh:
* include/fastjet/ClusterSequenceArea.hh:
* src/ClusterSequence_Delaunay.cc:
* example/fastjet_timing_plugins.cc:
* fortran_wrapper/fastjetfortran.cc:
rolled back to the auto_ptr (The idea would be to switch to a
unique_ptr when we include c++11 support). For now, we just
deprecate its use in the interface
* example/fastjet_timing_plugins.cc:
replaced rapmax->grid_rapmax in a local declaration to eliminate a warning.
2014-07-28 Gregory Soyez <[email protected]>
* example/fastjet_timing_plugins.cc:
declared SubType globally to prevent a icpc (v10.1) error.
2014-07-28 Gregory Soyez <[email protected]>
* example/fastjet_timing_plugins.cc:
* fortran_wrapper/fastjetfortran.cc:
auto_ptr -> SharedPtr as below
* include/fastjet/ClusterSequenceArea.hh:
use a standard pointer to hold the specific underlying CSAB rather
than an auto_ptr
2014-07-27 Gregory Soyez <[email protected]>
* src/ClusterSequence_Delaunay.cc:
replaced the auto_ptr to the DNN object by a standard C pointer
(deleted at the end of ClusterSequence::_delaunay_cluster())
* plugins/SISCone/SISConePlugin.cc:
* plugins/SISCone/SISConeSphericalPlugin.cc:
* plugins/SISCone/fastjet/SISConePlugin.hh:
* plugins/SISCone/fastjet/SISConeSphericalPlugin.hh:
replaced the use of auto_ptr by SharedPtr (auto_ptr has been
deprecated by C++11. Ideally a unique_ptr should be used but
SharedPtr will do the job)
2014-07-26 Gavin Salam <[email protected]>
* example/fastjet_timing_plugins.cc:
Switched from fj:: to using namespace fastjet.
Added -compare-strategy for comparing two different strategies and
printing out events where they disagree. Currently does full
clustering history comparison (not suitable for C/A alg.)
2014-07-25 Gavin Salam <[email protected]>
* src/LazyTiling9.cc:
added FASTJET_LAZY9_MIN2TILESY option, which forces at least two
tiles in rapidity. Ran usual tests of 10k events on each of antikt,
cam & kt with R=1000. Found one event for kt in which the answers
do not agree with N2MinHeapTiled. This appears to be independent
of whether FASTJET_LAZY9_MIN2TILESY is set. Commands to see the
problem are
gunzip -c ../data/Pythia-PtMin50-LHC-10kev.dat.gz| time ./example/fastjet_timing_plugins -kt -R 1000 -strategy -7 -write -nev 7492 -skip 7491
gunzip -c ../data/Pythia-PtMin50-LHC-10kev.dat.gz| time ./example/fastjet_timing_plugins -kt -R 1000 -strategy -4 -write -nev 7492 -skip 7491
Detailed investigation still to be carried out.
Issue filed as 2014-07-26-Lazy9-largeR-clustering
2014-07-25 Gregory Soyez <[email protected]>
* tools/fastjet/tools/Recluster.hh:
* tools/Recluster.cc:
back to the case with C/A reclustering and non-ghosted areas:
instead of issuing a warning, discard the area info in the
composite jet.
* include/fastjet/CompositeJetStructure.hh:
commented out set_area_information: the comment said this was used
by "join" and cannot be used by an end-user while in practice,
join no longer uses this (has it at any time?)
instead, added a discard_area() method that allows to disable the
area information for composite jets made of pieces with area.
* tools/Pruner.cc:
* tools/fastjet/tools/Pruner.hh:
when creating the pruning jet definition, set the shared
recombiner in case it it owned by the parent jet definition.
* include/fastjet/JetDefinition.hh:
in set_shared_Recombiner, only copy the recombiner if we have
ownership of it [this allows us to copy the shared pointer even if we
have no ownership of the recombiner]
* tools/fastjet/tools/Filter.hh:
* tools/Filter.cc:
simplified things considerably to use Recluster... still work in
progress and checks to be made
* tools/fastjet/tools/Recluster.hh:
* tools/Recluster.cc:
issue a warning when we're using the simplified C/A reclustering
and that there is non-explicit-ghost area support. Areas can be
erroneous in this case.
2014-07-24 Gregory Soyez <[email protected]>
* tools/fastjet/tools/Recluster.hh *** ADDED ***
* tools/Recluster.cc: *** ADDED ***
added a "Recluster" tool [it is essentially the reclustering part
of Filter although this code has transited through the
RecursiveTools contrib]
2014-07-24 Gavin SALAM <[email protected]> + Matteo
* include/fastjet/ClusterSequence.hh:
removed an old commented-out constructor
2014-07-24 Gavin Salam <[email protected]>
* include/fastjet/Selector.hh
* include/fastjet/internal/Dnn2piCylinder.hh
* include/fastjet/internal/Dnn3piCylinder.hh
* include/fastjet/internal/Dnn4piCylinder.hh
* include/fastjet/internal/DnnPlane.hh
* include/fastjet/internal/DynamicNearestNeighbours.hh
* src/DnnPlane.cc
* src/Selector.cc
further replacements of const double & and const int & with plain
const double and const int.
2014-07-23 Gavin Salam <[email protected]>
* src/LazyTiling25.cc:
fixed bug in positioning of tile centers that emerged when
_tiles_ieta_min*_tile_size and _tiles_eta_min were different
(which was previously never the case); having fixed this, the
scheme for always having at least 3 tiles in Y now works. Ran a
test on 10k events that anti-kt results are identical to other
schemes.
* src/LazyTiling9.cc:
fixed the same bug in positioning of tile centers; this should
have no effect here since, for now, _tiles_ieta_min*_tile_size and
_tiles_eta_min are always the same.
* scripts/update-headers.pl: *** ADDED ***
new script to update headers (including copyright date) of all .hh,
.cc, .icc and .f files.
The master copyright text is now stored here. It has been adapted
to mention the manual as well as the original fastjet paper, and
also to request that the FastJet version number be quoted in
scientific publications.
The header start/end strings are now FJSTARTHEADER and FJENDHEADER
(i.e. added "FJ") in order to limit chances of collisions with
other code.
* **/*.(cc|icc|hh|f):
updated headers
2014-07-22 Gavin SALAM <[email protected]> + Matteo + Gregory
* RoadMap: updated (plans for 3.1)
* include/fastjet/PseudoJetStructureBase.hh:
deprecated the use of "const double& dcut" in exclusive_subjets(...)
and exclusive_subjets_up_to(...) for future major releases. (to be
replaced by "const double dcut".
* include/fastjet/GhostedAreaSpec.hh:
added both (set)_ghost_maxeta and (set)_ghost_rapmax to ensure
symmetry
* BUGS:
removed the bug due to coincident points in strategies using CGAL
* include/fastjet/JetDefinition.hh:
* src/JetDefinition.cc:
. addded shared_recombiner() and set_shared_recombiner() methods.
. replaced _recombiner_shared with _shared_recombiner
. tidied up comments
* include/fastjet/ClusterSequence.hh:
fixed comment about delete_self_when_unused() (one should not be
in a context where the CS may just go out of scope).
+ deprecated the use of auto_ptr for plugin extras.
* include/fastjet/internal/LazyTiling9Alt.hh:
* src/LazyTiling9Alt.cc:
commented out unused variant of _distance_to_tile
* src/ClusterSequence.cc:
* include/fastjet/JetDefinition.hh:
renamed FJ30Best -> BestFJ30
2014-07-22 Gavin Salam <[email protected]>
* include/fastjet/ClusterSequence.hh
* include/fastjet/PseudoJet.hh
* include/fastjet/internal/LazyTiling25.hh
* include/fastjet/internal/LazyTiling9.hh
* include/fastjet/internal/LazyTiling9Alt.hh
* include/fastjet/internal/LazyTiling9SeparateGhosts.hh
* src/ClusterSequence.cc
* src/ClusterSequence_Delaunay.cc
* src/ClusterSequence_TiledN2.cc
* src/LazyTiling25.cc
* src/LazyTiling9.cc
* src/LazyTiling9Alt.cc
* src/LazyTiling9SeparateGhosts.cc
* src/PseudoJet.cc
replaced a number of "const double &" and "const int &" with
"const double" and "const int". There are still a number of others
to handle.
2014-07-22 Gavin Salam <[email protected]>
* testsuite/TestPseudoJet.hh:
added a set of tests of the ==0 and != 0 operators
* testsuite/cs_delete_self.cc:
now gives errors when things that should (not) throw do not (or
do).
* testsuite/PJtiming.cc:
replaced dynamic-sized C-array (which gave compilation errors)
with a dynamically allocated array.
* testsuite/run-all.sh: *** ADDED ***
this script runs everything in the directory and makes sure it all
exits OK.
* testsuite/README: *** ADDED ***
added a README to help us recall how to use what's in here
* include/fastjet/PseudoJet.hh:
added val==PseudoJet and val!=PseudoJet bool functions so that we
have the complete set.
* include/fastjet/internal/SearchTree.hh (circulator):
the following line
"template<class U> friend class SearchTree<U>::const_circulator;"
gave a warning, so made it template wrt class T
Things still work with Apple LLVM version 5.1 (clang-503.0.40) (based on LLVM 3.4svn)
* src/ClosestPair2D.cc:
removed unused
const unsigned int huge_unsigned = 4294967295U;
which was giving unused-variable warnings
Both warnings reported by Torbjorn Sjostrand.