Skip to content

Updates to the MEGAN specifier string parser #339

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

Merged
jedwards4b merged 3 commits into from
Jan 20, 2023
Merged

Updates to the MEGAN specifier string parser #339

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
9197fd2
updates to MEGAN namelist parser
fvitt Jan 18, 2023
9514b39
log MEGAN settings
fvitt Jan 19, 2023
690cf28
code clean up
fvitt Jan 19, 2023
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
166 changes: 92 additions & 74 deletions cesm/nuopc_cap_share/shr_expr_parser_mod.F90
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,12 +1,12 @@
!=============================================================================
! expression parser utility --
! for parsing simple linear mathematical expressions of the form
! X = a*Y + b*Z + ...
! X = a*R + b*S + c*(X + Y + Z) ...
!
!=============================================================================
module shr_expr_parser_mod
use shr_kind_mod,only : r8 => shr_kind_r8
use shr_kind_mod,only : cx => shr_kind_cx
use shr_kind_mod,only : CXX => shr_kind_cxx

implicit none
private
Expand Down Expand Up @@ -35,82 +35,122 @@ function shr_exp_parse( exp_array, nitems ) result(exp_items_list)
integer, optional, intent(out) :: nitems ! number of expressions parsed
type(shr_exp_item_t), pointer :: exp_items_list ! linked list of items returned

integer :: i,j, jj, nmax, nterms, n_exp_items
character(len=cx) :: tmp_str
integer :: i,j, n_exp_items
type(shr_exp_item_t), pointer :: exp_item, list_item
integer :: ndxs(512)
integer :: nelem, j1,j2,k
character(len=CXX) :: tmp_str, tmp_name
character(len=8) :: xchr ! multipler
real(r8) :: xdbl
real(r8) :: coeff0
logical :: more_to_come
character(len=CXX), allocatable :: sums_grps(:)
character(len=CXX) :: sum_string

allocate(sums_grps(size(exp_array)))

nullify( exp_items_list )
nullify( exp_item )
nullify( list_item )

n_exp_items = 0
nmax = size( exp_array )
sums_grps(:) = ' '

do i = 1,nmax
if (len_trim(exp_array(i))>0) then
! combine lines that have a trailing "+" with the next line
i=1
j=1
loop1: do while( len_trim(exp_array(i)) > 0 )

j = scan( exp_array(i), '=' )
k = scan(exp_array(i), '+', back=.true. )
more_to_come = k == len_trim(exp_array(i)) ! line ends with "+"

if ( j>0 ) then
if ( more_to_come ) then
sums_grps(j) = trim(sums_grps(j)) // trim(adjustl(exp_array(i)))
else
sums_grps(j) = trim(sums_grps(j)) // trim(adjustl(exp_array(i)))
j = j+1
endif

i = i+1
if ( i > size(exp_array) ) exit loop1

n_exp_items = n_exp_items + 1
end do loop1

allocate( exp_item )
exp_item%n_terms = 0
exp_item%name = trim(adjustl(exp_array(i)(:j-1)))
n_exp_items = j-1

tmp_str = trim(adjustl(exp_array(i)(j+1:)))
! a group is a summation of terms

nterms = 1
jj = scan( tmp_str, '+' )
do while(jj>0)
nterms = nterms + 1
tmp_str = tmp_str(jj+1:)
jj = scan( tmp_str, '+' )
enddo
! parse the individual sum strings... and form the groupings
has_grps: if (n_exp_items>0) then

allocate( exp_item%vars(nterms) )
allocate( exp_item%coeffs(nterms) )
! from shr_megan_mod ... should be generalized and shared...
grploop: do i = 1,n_exp_items

tmp_str = trim(adjustl(exp_array(i)(j+1:)))
! parse out the term names
! from first parsing out the terms in the summation equation ("+" separates the terms)

sum_string = sums_grps(i)
j = scan( sum_string, '=' )
nelem = 1
ndxs(nelem) = j ! ndxs stores the index of each term of the equation

! find indices of all the terms in the equation
tmp_str = trim( sum_string(j+1:) )
j = scan( tmp_str, '+' )
do while(j>0)
nelem = nelem+1
ndxs(nelem) = ndxs(nelem-1) + j
tmp_str = tmp_str(j+1:)
j = scan( tmp_str, '+' )
enddo
ndxs(nelem+1) = len(sum_string)+1

if (j>0) then
call set_coefvar( tmp_str(:j-1), exp_item )
tmp_str = tmp_str(j-1:)
else
call set_coefvar( tmp_str, exp_item )
endif
allocate( exp_item )

else
exp_item%n_terms = nelem ! number of terms

tmp_str = trim(adjustl(exp_array(i))) ! assumed to begin with '+'
exp_item%name = trim(adjustl( sum_string(:ndxs(1)-1))) ! thing to the left of the "=" is used as the name of the group

endif
! now that we have the number of terms in the summation allocate memory for the terms
allocate( exp_item%vars(nelem) )
allocate( exp_item%coeffs(nelem) )

! at this point tmp_str begins with '+'
j = scan( tmp_str, '+' )
coeff0 = 1._r8 ! default multiplier

if (j>0) then
! now parse out the multiplier from the terms
elmloop: do k = 1,nelem

! remove the leading + ...
tmp_str = tmp_str(j+1:)
j = scan( tmp_str, '+' )
exp_item%coeffs(k) = coeff0

do while(j>0)
! get the term name which follows the '*' operator if the is one
tmp_name = adjustl(sum_string(ndxs(k)+1:ndxs(k+1)-1))

call set_coefvar( tmp_str(:j-1), exp_item )
j = scan( tmp_name, '*' )
if (j>0) then

tmp_str = tmp_str(j+1:)
j = scan( tmp_str, '+' )
xchr = tmp_name(1:j-1) ! get the multipler (left of the '*')
read( xchr, * ) xdbl ! convert the string to a real
exp_item%coeffs(k) = xdbl ! store the multiplier

enddo
j1 = scan( tmp_name, '(' )
if (j1>0) then
coeff0 = xdbl
tmp_name = trim(adjustl(tmp_name(j1+1:))) ! get the term name (right of the '*')
else
coeff0 = 1._r8
tmp_name = trim(adjustl(tmp_name(j+1:))) ! get the term name (right of the '*')
endif

call set_coefvar( tmp_str, exp_item )
endif

endif
j2 = scan( tmp_name, ')' )
if (j2>0) then
coeff0 = 1._r8
tmp_name = tmp_name(1:j2-1)
endif

exp_item%vars(k) = trim(tmp_name)

enddo elmloop

if (associated(exp_item)) then
if (associated(exp_items_list)) then
Expand All @@ -124,13 +164,16 @@ function shr_exp_parse( exp_array, nitems ) result(exp_items_list)
endif
endif

endif
enddo

enddo grploop
endif has_grps

if ( present(nitems) ) then
nitems = n_exp_items
endif

deallocate(sums_grps)

end function shr_exp_parse

! -----------------------------------------------------------------
Expand All @@ -157,29 +200,4 @@ subroutine shr_exp_list_destroy( list )

end subroutine shr_exp_list_destroy

!==========================
! Private Methods

! -----------------------------------------------------------------
! -----------------------------------------------------------------
subroutine set_coefvar( term, item )
character(len=*), intent(in) :: term
type(shr_exp_item_t) , intent(inout) :: item

integer :: k, n

item%n_terms = item%n_terms + 1
n = item%n_terms

k = scan( term, '*' )
if (k>0) then
item%vars(n) = trim(adjustl(term(k+1:)))
read( term(:k-1), *) item%coeffs(n)
else
item%vars(n) = trim(adjustl(term))
item%coeffs(n) = 1.0_r8
endif

end subroutine set_coefvar

end module shr_expr_parser_mod
37 changes: 21 additions & 16 deletions cesm/nuopc_cap_share/shr_megan_mod.F90
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5,22 +5,22 @@ module shr_megan_mod
! MEGAN = Model of Emissions of Gases and Aerosols from Nature
!
! This reads the megan_emis_nl namelist in drv_flds_in and makes the relavent
! information available to CAM, CLM, and driver.
! - The driver sets up CLM to CAM communication for the VOC flux fields.
! - CLM needs to know what specific VOC fluxes need to be passed to the coupler
! information available to CAM, CLM, and driver.
! - The driver sets up CLM to CAM communication for the VOC flux fields.
! - CLM needs to know what specific VOC fluxes need to be passed to the coupler
! and how to assemble the fluxes.
! - CAM needs to know what specific VOC fluxes to expect from CLM.
!================================================================================

use ESMF , only : ESMF_VMGetCurrent, ESMF_VM, ESMF_VMGet
use ESMF , only : ESMF_LogFoundError, ESMF_LOGERR_PASSTHRU, ESMF_SUCCESS
use shr_kind_mod , only : r8 => shr_kind_r8, cl=>shr_kind_cl, cx=>shr_kind_cx, cs=>shr_kind_cs
use ESMF , only : ESMF_LogFoundError, ESMF_LOGERR_PASSTHRU
use shr_kind_mod , only : r8 => shr_kind_r8, cl=>shr_kind_cl, cx=>shr_kind_cx
use shr_sys_mod , only : shr_sys_abort
use shr_log_mod , only : shr_log_getLogUnit
use shr_mpi_mod , only : shr_mpi_bcast
use shr_nl_mod , only : shr_nl_find_group_name
use shr_expr_parser_mod , only : shr_exp_parse, shr_exp_item_t, shr_exp_list_destroy

implicit none
private

Expand Down Expand Up @@ -68,6 +68,9 @@ module shr_megan_mod
! switch to use mapped emission factors
logical :: shr_megan_mapped_emisfctrs = .false.

integer :: localPet = -huge(1)
integer :: logunit = -huge(1)

!--------------------------------------------------------
contains
!--------------------------------------------------------
Expand Down Expand Up @@ -100,7 +103,8 @@ subroutine shr_megan_readnl( NLFileName, megan_nflds)
! Example:
! &megan_emis_nl
! megan_specifier = 'ISOP = isoprene',
! 'C10H16 = myrcene + sabinene + limonene + carene_3 + ocimene_t_b + pinene_b + ...',
! 'C10H16 = myrcene + sabinene + limonene + carene_3 + ocimene_t_b + pinene_b + ',
! ' thujene_a + bornene + 0.5*(terpineol_4 + terpineol_a + terpinyl_ACT_a + myrtenal) + ...',
! 'CH3OH = methanol',
! 'C2H5OH = ethanol',
! 'CH2O = formaldehyde',
Expand All @@ -109,25 +113,22 @@ subroutine shr_megan_readnl( NLFileName, megan_nflds)
! megan_factors_file = '$datapath/megan_emis_factors.nc'
! /
!-------------------------------------------------------------------------

! input/output variables
character(len=*), intent(in) :: NLFileName
integer, intent(out) :: megan_nflds

! local variables
type(ESMF_VM) :: vm
integer :: localPet
integer :: mpicom
integer :: unitn ! namelist unit number
integer :: ierr ! error code
logical :: exists ! if file exists or not
integer, parameter :: maxspc = 100
character(len=2*CX) :: megan_specifier(maxspc) = ' '
integer, parameter :: maxspc = 200
character(len=CX) :: megan_specifier(maxspc) = ' '
logical :: megan_mapped_emisfctrs = .false.
character(len=CL) :: megan_factors_file = ' '
integer :: rc
integer :: logunit
integer :: i, tmp(1)
character(*), parameter :: F00 = "('(shr_megan_readnl) ',2a)"
character(len=*), parameter :: subname='(shr_megan_readnl)'
!--------------------------------------------------------------
Expand All @@ -140,12 +141,12 @@ subroutine shr_megan_readnl( NLFileName, megan_nflds)
end if

call ESMF_VMGetCurrent(vm, rc=rc)
if (ESMF_LogFoundError(rcToCheck=rc, msg=ESMF_LOGERR_PASSTHRU, line=__LINE__, file=__FILE__)) return
if (ESMF_LogFoundError(rcToCheck=rc, msg=ESMF_LOGERR_PASSTHRU, line=__LINE__, file=__FILE__)) return

call ESMF_VMGet(vm, localPet=localPet, mpiCommunicator=mpicom, rc=rc)
if (ESMF_LogFoundError(rcToCheck=rc, msg=ESMF_LOGERR_PASSTHRU, line=__LINE__, file=__FILE__)) return
if (ESMF_LogFoundError(rcToCheck=rc, msg=ESMF_LOGERR_PASSTHRU, line=__LINE__, file=__FILE__)) return
call shr_log_getLogUnit(logunit)
! Note the following still needs to be called on all processors since the mpi_bcast is a collective
! Note the following still needs to be called on all processors since the mpi_bcast is a collective
! call on all the pes of mpicom
if (localPet==0) then
inquire( file=trim(NLFileName), exist=exists)
Expand Down Expand Up @@ -204,6 +205,8 @@ subroutine shr_megan_init( specifier)
allocate(shr_megan_mechcomps(n_entries))
shr_megan_mechcomps(:)%n_megan_comps = 0

if (localPet==0) write(logunit,*) 'MEGAN entries:'

item => items_list
i = 1
do while(associated(item))
Expand All @@ -221,7 +224,9 @@ subroutine shr_megan_init( specifier)
shr_megan_mechcomps(i)%n_megan_comps = item%n_terms
allocate(shr_megan_mechcomps(i)%megan_comps(item%n_terms))

if (localPet==0) write(logunit,*) ' species : ', item%name
do j = 1,item%n_terms
if (localPet==0) write(logunit,'(f12.4,a,a)') item%coeffs(j),' * ', item%vars(j)
shr_megan_mechcomps(i)%megan_comps(j)%ptr => add_megan_comp( item%vars(j), item%coeffs(j) )
enddo
shr_megan_mechcomps_n = shr_megan_mechcomps_n+1
Expand Down