Skip to content

Fixed bug which attributed all yields and lactation summary data to i… #1

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
stuvet wants to merge 1 commit into
base: master
Choose a base branch
from
Open

Fixed bug which attributed all yields and lactation summary data to i… #1

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Binary file added ageadj.o
View file
Open in desktop
Binary file not shown.
Binary file added ageadjs.o
View file
Open in desktop
Binary file not shown.
Binary file added aiplage.o
View file
Open in desktop
Binary file not shown.
Binary file modified bestpred
View file
Open in desktop
Binary file not shown.
Binary file added bestpred.o
View file
Open in desktop
Binary file not shown.
Binary file added bestpred_fmt4.o
View file
Open in desktop
Binary file not shown.
Binary file added bestpred_log.o
View file
Open in desktop
Binary file not shown.
187 changes: 134 additions & 53 deletions bestpred_main.f90
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -333,7 +333,73 @@ program bestpred_main
if ( nomore == 1 ) go to 200
! Fixed portion
11 if ( source .ne. 14 .and. source .ne. 24 ) then
if ( nread > 0 ) then
!!! This section was commented out below and moved upwards in the flow to solve
!!! a bug where the first cow in a source 10 file was being processed
!!! without any lactation records, and her records were being misattributed
!!! to the subsequent cow, etc. Caused by setting f248 at the first pass, but
!!! only reading seg(i) at the second, at which time f248 is overwritten
!!! before they were both written to format4 and sent to bestpred_fmt4().
!!! Putting this section here allows the first format4 to receive
!!! the correct nseg, ntd & seg(i).

!!! However, moving this section upwards breaks the ending logic since beforehand
!!! nomore was set *before* the last record was processed (though at the second pass
!!! for the final f248, but now it is set when all records have been processed. The
!!! logic of 50 has been changed to suit.

!!! WARNING: not tested with any source other than 10 & 15, but it seems
!!! that all which match ( source .ne. 14 .and. source .ne. 24 ) should
!!! have been affected by the same issue. I haven't checked the logic for
!!! source 14 & 24.
18 if ( source == 10 .or. source == 15 ) then
read(10,20,end=50) n248,nseg,(seg(i),i=1,nseg)
20 format(a248,i2,50a23)
! read(n248,12) cow,nherd,nfresh
! Removed as read below
!read(n248,181) cow,nherd,nfresh
!181 format(t3,a17,t107,a8,t128,a8)
! print *, '[bestpred_main]: n248 : ', n248
! This needs more thorough investigation when I have time -- I think
! that f248 and n248 are redundant and a single variable can be used
! consistently in bestpred_main.f90. This fixes Miels's problem with
! incomplete Format information being passed to bestpred_fmt4.f90. The
! values written to the format4 variable were being taken from f248 in
! all cases, never from n248, so the data from the input file were never
! being passed downstream.
!f248 = n248
else if(source == 12) then
read(12,25,end=50) cow,nherd,nfresh,nlacno, &
nseg,(seg(i),i=1,nseg)
25 format(a17,a8,a8,i1,i2,50a23)
write(n248,12) cow,nherd,nfresh,nlacno
else if(source == 13) then
read(13,30,end=50) cow,nherd,nfresh,nbirth,nlacno, &
lnwt,ndoprev,herd14,avgage, &
nseg,(shrtseg(i),i=1,nseg)
30 format(a17,3a8,i1,i3,i3,i5,i4,i4,i2,i2,50a14)
! Adjust herd average
ageadj = 1.0 + .0056*(72 - avgage)
if(avgage > 72) ageadj = 1.0
do 35 j=1,3
35 herd14(j) = herd14(j)*ageadj
! End adjustment
do 37 j=1,nseg
read(shrtseg(j),36) dimm,milk,fatpc,propc,scs,freq
36 format(i3,i4,3i2,i1)
37 write(seg(j),38) dimm,'11',freq,freq,freq,'01100', &
milk,fatpc,propc,scs
38 format(i3,a,3i1,a,i4,3i2)
write(n248,40) cow,nbirth,nherd,nfresh,nlacno,herd14,dev305
40 format(t3,a17,t71,a8,t107,a8,t128,a8,t159,i2, &
t188,i5,2i4,i6,2i5)
else
! Handle non-matched source
stop("Source Code Not Recognised")
end if
!!! Removed this line to allow all records which have (now) just been read
!!! to continue to receive ntd, seg etc.
!!! WARNING: Not tested for any source other than 10 & 15
!if ( nread > 0 ) then
herd305(:,:) = 0.d0
f248 = n248
! read(f248,12) cowid,herd,fresh
Expand Down Expand Up @@ -363,59 +429,67 @@ program bestpred_main
length = max(length,dims)
! if ( ntd > 20 ) print *, 'segment ', ntd, ' ', segment(ntd)
15 continue
end if
!end if
! print *, '[bestpred_main]: length read from format 4T:', length
write(f248(136:138),'(i3)') length
end if
! Read next record
18 if ( source == 10 .or. source == 15 ) then
read(10,20,end=50) n248,nseg,(seg(i),i=1,nseg)
20 format(a248,i2,50a23)
! read(n248,12) cow,nherd,nfresh
read(n248,181) cow,nherd,nfresh
181 format(t3,a17,t107,a8,t128,a8)
! print *, '[bestpred_main]: n248 : ', n248
! This needs more thorough investigation when I have time -- I think
! that f248 and n248 are redundant and a single variable can be used
! consistently in bestpred_main.f90. This fixes Miels's problem with
! incomplete Format information being passed to bestpred_fmt4.f90. The
! values written to the format4 variable were being taken from f248 in
! all cases, never from n248, so the data from the input file were never
! being passed downstream.
f248 = n248
end if

!***
!if(cow /= 'HOUSA00093WPJ3927') go to 18
if(source == 12) then
read(12,25,end=50) cow,nherd,nfresh,nlacno, &
nseg,(seg(i),i=1,nseg)
25 format(a17,a8,a8,i1,i2,50a23)
write(n248,12) cow,nherd,nfresh,nlacno
end if

if(source == 13) then
read(13,30,end=50) cow,nherd,nfresh,nbirth,nlacno, &
lnwt,ndoprev,herd14,avgage, &
nseg,(shrtseg(i),i=1,nseg)
30 format(a17,3a8,i1,i3,i3,i5,i4,i4,i2,i2,50a14)
! Adjust herd average
ageadj = 1.0 + .0056*(72 - avgage)
if(avgage > 72) ageadj = 1.0
do 35 j=1,3
35 herd14(j) = herd14(j)*ageadj
! End adjustment
do 37 j=1,nseg
read(shrtseg(j),36) dimm,milk,fatpc,propc,scs,freq
36 format(i3,i4,3i2,i1)
37 write(seg(j),38) dimm,'11',freq,freq,freq,'01100', &
milk,fatpc,propc,scs
38 format(i3,a,3i1,a,i4,3i2)
write(n248,40) cow,nbirth,nherd,nfresh,nlacno,herd14,dev305
40 format(t3,a17,t71,a8,t107,a8,t128,a8,t159,i2, &
t188,i5,2i4,i6,2i5)
end if

!!! This section was commented out and moved upwards in the flow to solve
!!! a bug where the first cow in a source 10 file was being processed
!!! without any lactation records, and her records were being misattributed
!!! to the subsequent cow, etc. Caused by setting f248 at the first pass, but
!!! only reading seg(i) at the second, at which time f248 is overwritten
!!! before they were both written to format4 and sent to bestpred_fmt4()

! Read next record
! 18 if ( source == 10 .or. source == 15 ) then
! read(10,20,end=50) n248,nseg,(seg(i),i=1,nseg)
! 20 format(a248,i2,50a23)
!! read(n248,12) cow,nherd,nfresh
! read(n248,181) cow,nherd,nfresh
! 181 format(t3,a17,t107,a8,t128,a8)
!! print *, '[bestpred_main]: n248 : ', n248
! ! This needs more thorough investigation when I have time -- I think
! ! that f248 and n248 are redundant and a single variable can be used
! ! consistently in bestpred_main.f90. This fixes Miels's problem with
! ! incomplete Format information being passed to bestpred_fmt4.f90. The
! ! values written to the format4 variable were being taken from f248 in
! ! all cases, never from n248, so the data from the input file were never
! ! being passed downstream.
! f248 = n248
! end if
!
!!***
!!if(cow /= 'HOUSA00093WPJ3927') go to 18
! if(source == 12) then
! read(12,25,end=50) cow,nherd,nfresh,nlacno, &
! nseg,(seg(i),i=1,nseg)
! 25 format(a17,a8,a8,i1,i2,50a23)
! write(n248,12) cow,nherd,nfresh,nlacno
! end if
!
! if(source == 13) then
! read(13,30,end=50) cow,nherd,nfresh,nbirth,nlacno, &
! lnwt,ndoprev,herd14,avgage, &
! nseg,(shrtseg(i),i=1,nseg)
! 30 format(a17,3a8,i1,i3,i3,i5,i4,i4,i2,i2,50a14)
!! Adjust herd average
! ageadj = 1.0 + .0056*(72 - avgage)
! if(avgage > 72) ageadj = 1.0
! do 35 j=1,3
! 35 herd14(j) = herd14(j)*ageadj
!! End adjustment
! do 37 j=1,nseg
! read(shrtseg(j),36) dimm,milk,fatpc,propc,scs,freq
! 36 format(i3,i4,3i2,i1)
! 37 write(seg(j),38) dimm,'11',freq,freq,freq,'01100', &
! milk,fatpc,propc,scs
! 38 format(i3,a,3i1,a,i4,3i2)
! write(n248,40) cow,nbirth,nherd,nfresh,nlacno,herd14,dev305
! 40 format(t3,a17,t71,a8,t107,a8,t128,a8,t159,i2, &
! t188,i5,2i4,i6,2i5)
! end if
!
! If the source is 24 then we need to loop over the contents of the file and
! handle each row as a separate source 14 file. IF we've read all of the
! records from a source 14 file then it's time to read the next filename.
Expand Down Expand Up @@ -598,7 +672,13 @@ program bestpred_main
if ( nread > obs ) go to 50
go to 52
50 nomore = 1
if ( nread == 0 ) go to 200
!!! Done to prevent the last cow being duplicated, as with the changes in flow
!!! in 11 nomore=1 is now only set *after* the last record has already been
!!! processed. Failing to change this logic leads to the last record being
!!! duplicated in the output files.
!!! WARNING: I have not tested this for any source other than 10 & 15
go to 200
! if ( nread == 0 ) go to 200
!!! This was used to skip the loop at 52, but we actually need
!!! to enter that code in order to catch the last record in a
!!! file with no TD data.
Expand Down Expand Up @@ -744,9 +824,10 @@ program bestpred_main
TDflag = 0
if ( TDstatus(i) ) TDflag = 1
! Write the record to the OUTfile.
!!! Required for compilation with g95
write(45,"(a7,1x,i3,1x,i1,1x,f5.1,1x,f3.1,1x,f3.1,1x,f3.1,1x, &
3(i5,1x,i4,1x,i4,1x,i4,1x), &
3(i3,1x),4(f5.2,1x),8(i2,1x),f5.1,1x,f3.1,1x,f3.1,1x,f3.1)") &
&3(i5,1x,i4,1x,i4,1x,i4,1x), &
&3(i3,1x),4(f5.2,1x),8(i2,1x),f5.1,1x,f3.1,1x,f3.1,1x,f3.1)") &
cowid7, i, TDflag &
, TEMPestyld(1), TEMPestyld(2), TEMPestyld(3), TEMPestyld(4) &
, int(YLDvec(1,9)), int(YLDvec(1,10)), int(YLDvec(1,11)), int(YLDvec(1,12)) &
Expand Down
Binary file added bestpred_parm.o
View file
Open in desktop
Binary file not shown.