Added variable all_cells to be able to initialize new cell_data.
moved old_cells, max_cells, all_cells to phreeqc class.
initialized old_cells, max_cells, all_cells in init of phreeqc class.
Added to copy constructor.
changed logic for switching to numerical derivatives by adding && numerical_fixed_volume == false
if (iterations > 99 && numerical_fixed_volume == false)
Added
gammas_pz();
jacobian_sums();
For restart in pitzer.cpp
Added
gammas_sit();
jacobian_sums();
for restart in sit.cpp.
git-svn-id: svn://136.177.114.72/svn_GW/phreeqc3/trunk@8835 1feff8c3-07ed-0310-ac33-dd36852eb9cd
commented lines are directives for visual studio _CRTDBG_MAP_ALLOC, _CrtSetDbgFlag
VarClear in GetSelectedOutputHeading.
RebalanceLoadPerCell fixed for serial version, clear clock times.
delete out stream and punch stream before new, RunCells
also set streams to NULL after deleted.
also serial version.
VarClear in RunCellsThreadNoPrint
timing only if rebalance cell, RunCellsThreadNoPrint
VarClear in RunCellsThread.
changed methods to return pointer instead of reference, Get_out_stream
~IPhreeqcPhast, delete streams if necessary.
phast_manager did not delete module for flow-only.
phast_worker did not delete module for flow-only.
return pointers:
Get_out_stream
Get_punch_stream
IPhreeqc,
delete default_data_base and sformatf_buffer in clean_up, not in operator = or ~phreeqc
git-svn-id: svn://136.177.114.72/svn_GW/phreeqc3/trunk@8784 1feff8c3-07ed-0310-ac33-dd36852eb9cd
Also, the logic for using -isotopes data was changed so that the last defined applies to all remaining solutions.
git-svn-id: svn://136.177.114.72/svn_GW/phreeqc3/trunk@8741 1feff8c3-07ed-0310-ac33-dd36852eb9cd
Branching at 7862 for multiple SELECTED_OUTPUT, USER_PUNCH
........
Have multiple SELECTED_OUTPUT working.
Need to have multiple USER_PUNCH.
........
Think multiple instances of both SELECTED_OUTPUT and USER_PUNCH are working.
........
Fixed up inverse.cpp a bit.
........
Fixed missing -pressure in solution dump_raw and read_raw.
........
Merging revisedfandg back into trunk. Makes Excel spreadsheet for carbon-14 dating.
OK by Niel. Has Revised_Fontes_ and_Garnier_ in_NetpathXL.docx to add for documentation.
........
Added Revised_Fontes_ and_Garnier_ in_NetpathXL.docx to installation
Fixed bug that wouldn't allow uninstall
........
updated ftp targets
........
NetpathXL 1.4
........
checkin from make
........
latest test with change in Pitzer pressure, -pressure in SOLUTION_RAW, runs modify_test, has error
........
bugs linux caught, something different about when to print headings
........
Working on multipunch
........
working on high_precision, using SELECTED_OUTPUT 1
........
dump precision
........
Added basic functions:
EQ_FRAC
KIN_TIME
SPECIES_FORMULA
STR_F
STR_E
Added //{{NEW BASIC HERE}} for updating BASIC routines
........
Making backward compatible.
Copying all but lists.
Revised tidy_punch logic.
........
Still working to make backward compatible.
Think there will be one diff, headings will not be rewritten (ddl_flux).
........
updated dump tests (new -pressure line)
........
multi_punch branch
........
Added a Fortran callback.
Will try to revise so that the same callback works for Fortran and C.
........
Revised name to BasicFortran
Added methods for C call SetBasicCallback.
Need to check C side and documentation.
........
git-svn-id: svn://136.177.114.72/svn_GW/phreeqc3/trunk@7987 1feff8c3-07ed-0310-ac33-dd36852eb9cd
Will try to revise so that the same callback works for Fortran and C.
git-svn-id: svn://136.177.114.72/svn_GW/phreeqc3/trunk@7883 1feff8c3-07ed-0310-ac33-dd36852eb9cd
Fixed error with Pitzer paramenters that were not recalculated with
a change in pressure. Parameters depend on density of water I think.
Also changed SIT, although I am not sure the parameters have a pressure
dependence.
git-svn-id: svn://136.177.114.72/svn_GW/phreeqc3/trunk@7837 1feff8c3-07ed-0310-ac33-dd36852eb9cd
kin_delta/kin_time is the average rate over the time interval.
git-svn-id: svn://136.177.114.72/svn_GW/phreeqc3/trunk@7676 1feff8c3-07ed-0310-ac33-dd36852eb9cd
MAX_ITERS 40
hnew = hg
predict_fail
CVMEM cvode_error == TRUE || predict_fail after CVnls
if (CVMEM cvode_error == TRUE) after CVNewtonIteration
CVDoErrorTest exit(4);
model
Phreeqc.cpp
pitzer.cpp
sit.cpp
new status ifdef'd with NPP
250 ms delay
new global variable std::string string_status
!!prep: switch_bases 10 to 100 orders of mag. Need to check
utilities.cpp: new status
git-svn-id: svn://136.177.114.72/svn_GW/phreeqc3/trunk@7585 1feff8c3-07ed-0310-ac33-dd36852eb9cd
Running ion association model with equilibrium phases
Running Pitzer.
git-svn-id: svn://136.177.114.72/svn_GW/phreeqc3/trunk@7449 1feff8c3-07ed-0310-ac33-dd36852eb9cd
All data members are now initialized in init.
Almost all allocation is in initialize.
There is a little bit of duplication in initialize, but not much.
git-svn-id: svn://136.177.114.72/svn_GW/phreeqc3/trunk@7005 1feff8c3-07ed-0310-ac33-dd36852eb9cd
Guy had problems with static variables with OpenMP.
Also removed other static variables fpunchf_user s_warning and buffer, spread length, utilities spinner. Put in Phreeqc.h.
Removed static fnbuf from Pbasic.cpp, put in Pbasic.h.
I think only static variables are in Keywords.h and are necessary for initialization.
git-svn-id: svn://136.177.114.72/svn_GW/phreeqc3/trunk@6770 1feff8c3-07ed-0310-ac33-dd36852eb9cd
This is really just 6324 without deleting Phreeqc.h and keeping the history. In order
to keep the history you need to do a reverse merge.
see http://stackoverflow.com/questions/2863028/recover-a-folder-or-file-in-tortoisesvn-whilst-also-retaining-all-history
answer attached:
Do a "reverse merge":
1.Ensure that your working copy is updated to HEAD and completely clean (not strictly necessary, but always a good idea before trying to merge)
2.Right click on the folder and select TortoiseSVN > Merge...
3.Select "Merge a range of revisions" and hit "Next"
4.The URL to merge from is the repo path for your current directory (ie. the one containing the deleted file)
5.The revision range to merge is the revision in which you deleted the file (show log might help here)
6.Be sure to select "Reverse merge"
7.Hit "Next"
8.All the default merge options should be sufficient, so hit "Merge"
The file should now be added and still retain all of its previous history.
Explanation:
This "rolls back" the revision containing the deletion, but adds the merge information to the directory properties. SVN can track the original file through this metadata. On the command line you would to do a backwards merge of the revision in which you deleted it:
svn merge -c -<revision-number> path/containing/file
Note the hyphen before the rev number (ie. a "negative" revision nunber).
git-svn-id: svn://136.177.114.72/svn_GW/phreeqc3/trunk@6328 1feff8c3-07ed-0310-ac33-dd36852eb9cd
All reactant structs have been removed.
Tony's pressure uses mu in pressure term of log_k.
Test cases run, discriminant check at 1e-8.
Still want to optimize out some k_temp calls and checks for same T, P, mu.
git-svn-id: svn://136.177.114.72/svn_GW/phreeqc3/trunk@6269 1feff8c3-07ed-0310-ac33-dd36852eb9cd
Need to compile on Linux and run all test cases.
git-svn-id: svn://136.177.114.72/svn_GW/phreeqcpp/branches/ErrorHandling@6070 1feff8c3-07ed-0310-ac33-dd36852eb9cd
