Total number of milliseconds for "tika_1_6" by file extension

FILE_EXTENSION ELAPSED_TIME_MILLIS_B TOTAL_FILES Average number of milliseconds per file
pptx
30891
35
882.60 
docx
8617
18
478.72 
pdf
18180874
52020
349.50 
pps
49379
160
308.62 
zip
301
1
301.00 
ppt
2980024
11884
250.76 
vrml
123
1
123.00 
doc
1910495
16156
118.25 
xls
756048
7868
96.09 
java
4744
54
87.85 
xlsx
160
2
80.00 
text
21171
321
65.95 
odp
65
1
65.00 
ttf
50
1
50.00 
py
48
1
48.00 
gz
69080
1567
44.08 
log
35645
830
42.95 
rtf
10441
246
42.44 
data
84
2
42.00 
eps
5627
153
36.78 
swf
2782
78
35.67 
fits
2785
79
35.25 
ps
74559
2396
31.12 
tmp
311
10
31.10 
html
1631883
52983
30.80 
bmp
450
16
28.12 
jpg
396129
14894
26.60 
txt
724971
27671
26.20 
xml
46605
1862
25.03 
kmz
1112
47
23.66 
dwf
365
16
22.81 
wp
2125
95
22.37 
png
15930
728
21.88 
unk
19075
879
21.70 
fm
64
3
21.33 
csv
38144
1937
19.69 
f
4955
257
19.28 
troff
740
39
18.97 
gif
75548
4103
18.41 
gls
183
10
18.30 
tex
1680
98
17.14 
sql
302
19
15.89 
g3
14
1
14.00 
wk1
14
1
14.00 
kml
563
42
13.40 
dbase3
4571
372
12.29 
hlp
241
24
10.04 
sgml
125
15
8.33 
pub
15
3
5.00 




SELECT comparisons.FILE_EXTENSION, sum(ifnull(ELAPSED_TIME_MILLIS_B, 0)) as ELAPSED_TIME_MILLIS_B, extensions_total.NUM_FILES as TOTAL_FILES, printf("%.2f ", (1.0*sum(ifnull(ELAPSED_TIME_MILLIS_B, 0))/extensions_total.NUM_FILES)) as 'Average number of milliseconds per file' from comparisons left outer join extensions_total on comparisons.FILE_EXTENSION=extensions_total.FILE_EXTENSION where JSON_EX_B is null and SORT_STACK_TRACE_B is null group by comparisons.FILE_EXTENSION order by (1.0*sum(ifnull(ELAPSED_TIME_MILLIS_B, 0))/extensions_total.NUM_FILES) desc;