llot95a_mod.f

49 lines · 1 definitions · on GitHub (7be4133)

Python version: llot95a_mod.py, 33 lines, on the right (switch to Python above the code).

Variables on this page (12)

cappa2, numd, numt, numx, numxm, numxz, numz, nz, r0gr, tgr, xxg, zzg

Purpose

LLOT95A_MOD holds all the OPAL95 opacity tables as read, with their grids in $\log T$, $\log R$, $X$ and $Z$, shared by 5 routines. It replaces COMMON/LLOT95A/ (MHP via Claude, 10/14/26). The dimensions NUMX = 10, NUMT = 70, NUMD = 19, NUMZ = 13 and NUMXZ = 126 (the number of tables) are PRIVATE.

What its variables hold, who sets them, who reads them.

CAPPA2(NUMXZ,NUMT,NUMD): $\log\kappa$ of every table, read by LL95TBL (missing entries set to 9.999); interpolated in $Z$ by OP95ZTAB, and in all four variables by OP954D and GETOPAL95.
TGR(NUMT), R0GR(NUMD): the $\log T$ and $\log R$ grids, read from the tables by LL95TBL; used by GETOPAL95 to locate a point.
XXG(NUMX), ZZG(NUMZ): the $X$ values (0 to 1) and $Z$ values (0 to 0.10) of the tables; LL95TBL checks each table's header against them, OP95XTAB, OP95ZTAB and GETOPAL95 interpolate in them. (For the last $X$, LL95TBL's check uses $1 - Z$ instead of XXG(NUMX).)
NZ(NUMZ): the offset of each $Z$'s tables in CAPPA2's first index (0, 10, … , 118); with nine slots for $Z$ = 0.06 and 0.08, the tables LL95TBL stores at NZ(IZ)+10 overwrite others (a suspected bug, on LL95TBL's page; line 47 ↗).
NUMXM(NUMZ): the number of $X$ tables for each $Z$ (10, but 9, 9 and 8 for the three largest, $Z$ = 0.06, 0.08, 0.1); read nowhere: unused?
XXG, ZZG, NZ and NUMXM carry initial values, moved from LL95TBL's DATA statements.

Design

A blanket USE LLOT95A_MOD. The module's header says its initializers came from PARMIN's DATA statements and are namelist variables; at rp_1 they were in LL95TBL, and they are in no namelist. At rp_1, main.f declared the block with its own NUMTT (also 70); that declaration is gone.

Definitions

moduleLLOT95A_MOD

lines 29–49 · GitHub

Used by (5 units)

GETOPAL95, LL95TBL, OP954D, OP95XTAB, OP95ZTAB

Change log

The changes since rp_1

Changed since rp_1. New file: COMMON/LLOT95A/ as a module, with its fixed grids moved from LL95TBL's DATA statements.

1C 2C 3C$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ 4C LLOT95A_MOD 5C$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ 6C 10/14/26 MHP via Claude 7C code changed to convert COMMON/LLOT95A/ into this module. 8C 9C Imported with a blanket USE, the convention for the conversion. 10C All 5 files declared this block identically, so the USE imports 11C exactly the names the declaration already made global in each 12C scoping unit and cannot change meaning. 13C 14C The types restate what the implicit rules supplied: L is 15C LOGICAL*4, I/J/K/M/N are INTEGER*4, everything else REAL*8. 16C SAVE is explicit rather than relying on module variables being 17C implicitly saved, which holds only from Fortran 2008 while 18C ifort Classic predates it. That matches COMMON's semantics. 19C 20C 4 of the 7 members carry an initializer, migrated 21C from parmin.f's DATA statements because DATA on a USE-associated 22C name is rejected. The literal text is copied verbatim: these 23C are namelist variables whose DATA supplied a default the 24C namelist read could override, and a module initializer is 25C applied once at load time, before PARMIN runs, which is the 26C same thing. Rewriting a literal such as 0.9828 as 0.9828D0 27C would change its stored value, so that is deliberately not 28C done here; see KNOWN_ISSUES.md issue 23. 29 MODULE LLOT95A_MOD 30 31 IMPLICIT NONE 32 SAVE 33 34 INTEGER*4, PARAMETER, PRIVATE :: NUMD = 19 35 INTEGER*4, PARAMETER, PRIVATE :: NUMT = 70 36 INTEGER*4, PARAMETER, PRIVATE :: NUMX = 10 37 INTEGER*4, PARAMETER, PRIVATE :: NUMXZ = 126 38 INTEGER*4, PARAMETER, PRIVATE :: NUMZ = 13 39 40 REAL*8 :: TGR(NUMT) 41 REAL*8 :: XXG(NUMX) = (/0.0D0, 0.1D0, 0.2D0, 0.35D0, 0.5D0, 0.7D0, 0.8D0, 0.9D0, 0.95D0, 1.0D0/) 42 REAL*8 :: R0GR(NUMD) 43 REAL*8 :: ZZG(NUMZ) = (/0.0D0, 0.0001D0, 0.0003D0, 0.001D0, 0.002D0, 0.004D0, 0.01D0, 0.02D0, 0.03D0, 0.04D0, 0.06D0, 44 * * 0.08D0, 0.10D0/) 45 REAL*8 :: CAPPA2(NUMXZ,NUMT,NUMD) 46 INTEGER*4 :: NUMXM(NUMZ) = (/10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 9, 9, 8/) 47 INTEGER*4 :: NZ(NUMZ) = (/0, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 109, 118/) 48 49 END MODULE LLOT95A_MOD

1"""Python version of llot95a_mod.f (YREC 7be4133): all the OPAL95 opacity 2tables and their grids, formerly COMMON/LLOT95A/. 3 4Translates: llot95a_mod.f#LLOT95A_MOD (7be4133) 5 6Routines that USE LLOT95A_MOD import this module and write 7``llot95a_mod.<name>``. Arrays keep Fortran's index order, index 0 unused. 8""" 9 10import numpy as np 11 12# Table dimensions (PRIVATE in the Fortran) (Fortran lines 34–38). 13_NUMD = 19 14_NUMT = 70 15_NUMX = 10 16_NUMXZ = 126 17_NUMZ = 13 18 19# The grids and the tables (Fortran lines 40–47). 20tgr = np.zeros(_NUMT + 1) # log T grid (ll95tbl) 21# X values of the tables (initializer). 22xxg = np.array([0.0, 0.0, 0.1, 0.2, 0.35, 0.5, 0.7, 0.8, 0.9, 0.95, 1.0]) 23r0gr = np.zeros(_NUMD + 1) # log R grid (ll95tbl) 24# Z values of the tables (initializer). 25zzg = np.array([0.0, 0.0, 0.0001, 0.0003, 0.001, 0.002, 0.004, 0.01, 0.02, 26 0.03, 0.04, 0.06, 0.08, 0.10]) 27cappa2 = np.zeros((_NUMXZ + 1, _NUMT + 1, _NUMD + 1)) # log kappa, every table (ll95tbl) 28# Number of X tables per Z (initializer; read nowhere). 29numxm = np.array([0, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 9, 9, 8]) 30# Offset of each Z's tables in cappa2's first index (initializer). 31# Suspected bug (see LL95TBL's page): Z = 0.06 and 0.08 get nine slots, but 32# ll95tbl stores their X = 1-Z tables at nz + 10. 33nz = np.array([0, 0, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 109, 118])