Export All

VRML2 sensitivity

Using export_all.py which uses GiGaRunActionGDML (possibly to be renamed GiGaRunActionExport ) find vertex count and precision differences in the WRL between exporting singly and when exported after DAE+GDML in the same process ?

Maybe:

1. missing vis initialistion
2. ordering effect ? is the DAE+GDML dump changing the geometry OR some parameters relevant to GetPolyhedron ?

[blyth@belle7 gdml]$ ll g4_00.* tmp/g4_00.*
-rw-rw-r-- 1 blyth blyth  5126579 Nov 15 12:24 tmp/g4_00.dae
-rw-rw-r-- 1 blyth blyth  4111332 Nov 15 12:24 tmp/g4_00.gdml
-rw-rw-r-- 1 blyth blyth 85400082 Nov 15 13:06 tmp/g4_00.wrl
-rw-rw-r-- 1 blyth blyth  4111332 Nov 15 14:38 g4_00.gdml
-rw-rw-r-- 1 blyth blyth  5126579 Nov 15 14:38 g4_00.dae
-rw-rw-r-- 1 blyth blyth 86458076 Nov 15 14:38 g4_00.wrl
-rw-rw-r-- 1 blyth blyth   217259 Nov 15 14:53 tmp/g4_00.wrl.10k
-rw-rw-r-- 1 blyth blyth   217194 Nov 15 14:53 g4_00.wrl.10k

[blyth@belle7 gdml]$ du -hs g4_00.* tmp/g4_00.*
5.0M    g4_00.dae
4.0M    g4_00.gdml
83M     g4_00.wrl
220K    g4_00.wrl.10k
5.0M    tmp/g4_00.dae
4.0M    tmp/g4_00.gdml
82M     tmp/g4_00.wrl
220K    tmp/g4_00.wrl.10k

[blyth@belle7 gdml]$ head -10000 tmp/g4_00.wrl > tmp/g4_00.wrl.10k
[blyth@belle7 gdml]$ head -10000 g4_00.wrl > g4_00.wrl.10k
[blyth@belle7 gdml]$ diff tmp/g4_00.wrl.10k  g4_00.wrl.10k

Question

1. does the changed WRL better match the DAE, in vertex counts/offsets ?

Order dependance : interference between DAE and WRL exports

1. no effect on GDML, same no matter what the order
2. wrl alone matches wrl first
3. dae alone matches gdml+dae first
4. looks like interference both ways, between dae and wrl
   • dae exported after wrl is smaller than dae alone
   • wrl exported after dae is larger than wrl alone

[blyth@belle7 gdml]$ l wrl_gdml_dae/    # single process : wrl followed by gdml+dae
total 92260
-rw-rw-r-- 1 blyth blyth 85400082 Nov 15 17:24 g4_00.wrl
-rw-rw-r-- 1 blyth blyth  4852486 Nov 15 17:24 g4_00.dae
-rw-rw-r-- 1 blyth blyth  4111332 Nov 15 17:24 g4_00.gdml

[blyth@belle7 gdml]$ l gdml_dae_wrl/    # single process : with gdml+dae followed by wrl
total 93780
-rw-rw-r-- 1 blyth blyth  4111332 Nov 15 14:38 g4_00.gdml
-rw-rw-r-- 1 blyth blyth  5126579 Nov 15 14:38 g4_00.dae
-rw-rw-r-- 1 blyth blyth 86458076 Nov 15 14:38 g4_00.wrl

[blyth@belle7 gdml]$ l tmp/             # wrl output separate from gdml+dae
total 92748
-rw-rw-r-- 1 blyth blyth 85400082 Nov 15 13:06 g4_00.wrl

-rw-rw-r-- 1 blyth blyth  4111332 Nov 15 12:24 g4_00.gdml
-rw-rw-r-- 1 blyth blyth  5126579 Nov 15 12:24 g4_00.dae

Comparing the interfered

[blyth@belle7 gdml]$ daedb.py --daepath wrl_gdml_dae/g4_00.dae
[blyth@belle7 gdml]$ daedb.py --daepath gdml_dae_wrl/g4_00.dae
[blyth@belle7 gdml]$ vrml2file.py -cx gdml_dae_wrl/g4_00.wrl     ## this is taking >20 min ???

DB extend takinh ages and filling disk

Argh its filling the disk:

simon:gdml_dae_wrl blyth$ du -hs *
4.9M    g4_00.dae
2.4M    g4_00.dae.db
3.9M    g4_00.gdml
 82M    g4_00.wrl
253M    g4_00.wrl.db
simon:gdml_dae_wrl blyth$ rm g4_00.wrl.db

Add Sequence flexibility

Using sequence with WRL first:

G4DAE_EXPORT_SEQUENCE VVVDDDGGGVVVVDVDVD

Suggests that what matters is who exports first (perhaps creating Polyhedron that are then reused perhaps):

[blyth@belle7 gdml]$ cd  /data1/env/local/env/geant4/geometry/gdml/20131116-1645/
[blyth@belle7 20131116-1645]$ l *.wrl
-rw-rw-r-- 1 blyth blyth 85400082 Nov 16 16:49 g4_08.wrl
-rw-rw-r-- 1 blyth blyth 85400082 Nov 16 16:48 g4_07.wrl
-rw-rw-r-- 1 blyth blyth 85400082 Nov 16 16:48 g4_06.wrl
-rw-rw-r-- 1 blyth blyth 85400082 Nov 16 16:48 g4_05.wrl
-rw-rw-r-- 1 blyth blyth 85400082 Nov 16 16:48 g4_04.wrl
-rw-rw-r-- 1 blyth blyth 85400082 Nov 16 16:48 g4_03.wrl
-rw-rw-r-- 1 blyth blyth 85400082 Nov 16 16:48 g4_02.wrl
-rw-rw-r-- 1 blyth blyth 85400082 Nov 16 16:47 g4_01.wrl
-rw-rw-r-- 1 blyth blyth 85400082 Nov 16 16:47 g4_00.wrl
[blyth@belle7 20131116-1645]$
[blyth@belle7 20131116-1645]$ l *.dae
-rw-rw-r-- 1 blyth blyth 4852486 Nov 16 16:49 g4_05.dae
-rw-rw-r-- 1 blyth blyth 4852486 Nov 16 16:48 g4_04.dae
-rw-rw-r-- 1 blyth blyth 4852486 Nov 16 16:48 g4_03.dae
-rw-rw-r-- 1 blyth blyth 4852486 Nov 16 16:48 g4_02.dae
-rw-rw-r-- 1 blyth blyth 4852486 Nov 16 16:48 g4_01.dae
-rw-rw-r-- 1 blyth blyth 4852486 Nov 16 16:48 g4_00.dae
[blyth@belle7 20131116-1645]$
[blyth@belle7 20131116-1645]$ l *.gdml
-rw-rw-r-- 1 blyth blyth 4111332 Nov 16 16:48 g4_02.gdml
-rw-rw-r-- 1 blyth blyth 4111332 Nov 16 16:48 g4_01.gdml
-rw-rw-r-- 1 blyth blyth 4111332 Nov 16 16:48 g4_00.gdml

Changing sequence to DAE first DVVVDDDGGGVVVVDVDVD confirms this, the first export influences all:

[blyth@belle7 20131116-1659]$ l *.gdml
-rw-rw-r-- 1 blyth blyth 4111332 Nov 16 17:02 g4_02.gdml
-rw-rw-r-- 1 blyth blyth 4111332 Nov 16 17:02 g4_01.gdml
-rw-rw-r-- 1 blyth blyth 4111332 Nov 16 17:02 g4_00.gdml
[blyth@belle7 20131116-1659]$ l *.dae
-rw-rw-r-- 1 blyth blyth 5126579 Nov 16 17:03 g4_06.dae
-rw-rw-r-- 1 blyth blyth 5126579 Nov 16 17:02 g4_05.dae
-rw-rw-r-- 1 blyth blyth 5126579 Nov 16 17:02 g4_04.dae
-rw-rw-r-- 1 blyth blyth 5126579 Nov 16 17:02 g4_03.dae
-rw-rw-r-- 1 blyth blyth 5126579 Nov 16 17:02 g4_02.dae
-rw-rw-r-- 1 blyth blyth 5126579 Nov 16 17:02 g4_01.dae
-rw-rw-r-- 1 blyth blyth 5126579 Nov 16 17:01 g4_00.dae
[blyth@belle7 20131116-1659]$ l *.wrl
-rw-rw-r-- 1 blyth blyth 86458076 Nov 16 17:03 g4_08.wrl
-rw-rw-r-- 1 blyth blyth 86458076 Nov 16 17:02 g4_07.wrl
-rw-rw-r-- 1 blyth blyth 86458076 Nov 16 17:02 g4_06.wrl
-rw-rw-r-- 1 blyth blyth 86458076 Nov 16 17:02 g4_05.wrl
-rw-rw-r-- 1 blyth blyth 86458076 Nov 16 17:02 g4_04.wrl
-rw-rw-r-- 1 blyth blyth 86458076 Nov 16 17:02 g4_03.wrl
-rw-rw-r-- 1 blyth blyth 86458076 Nov 16 17:02 g4_02.wrl
-rw-rw-r-- 1 blyth blyth 86458076 Nov 16 17:01 g4_01.wrl
-rw-rw-r-- 1 blyth blyth 86458076 Nov 16 17:01 g4_00.wrl