Languages detected in pdfbox_1_8_8 by content type then language

DETECTED_CONTENT_TYPE_A LANG_ID1_A COUNT
application/pdf
en
217488
application/pdf
de
1445
application/pdf
es
961
application/pdf
fr
706
application/pdf
bn
560
application/pdf
it
173
application/pdf
vi
170
application/pdf
pt
165
application/pdf
th
74
application/pdf
sl
68
application/pdf
so
62
application/pdf
id
58
application/pdf
lt
58
application/pdf
tl
57
application/pdf
pl
55
application/pdf
sq
51
application/pdf
nl
47
application/pdf
tr
44
application/pdf
sv
43
application/pdf
hu
40
application/pdf
hr
33
application/pdf
ro
30
application/pdf
da
23
application/pdf
et
21
application/pdf
no
21
application/pdf
el
20
application/pdf
sw
17
application/pdf
ru
17
application/pdf
sk
14
application/pdf
hi
14
application/pdf
cs
11
application/pdf
ko
8
application/pdf
lv
8
application/pdf
af
5
application/pdf
ar
5
application/pdf
ja
4
application/pdf
fi
4
application/pdf
uk
2
application/pdf
ur
2
application/pdf
zh-cn
2
application/pdf
bg
2
application/pdf
zh-tw
2
application/pdf
mk
1
application/pdf
fa
1
text/html; charset=ISO-8859-1
en
1
text/plain; charset=ISO-8859-8
nl
1
text/plain; charset=windows-1252
en
6
text/plain; charset=windows-1255
en
3




select DETECTED_CONTENT_TYPE_A, LANG_ID1_A, count(1) as COUNT from comparisons where LANG_ID1_A is not null group by DETECTED_CONTENT_TYPE_A, LANG_ID1_A order by DETECTED_CONTENT_TYPE_A, COUNT desc