Unreadable line in Restart/pdepnetworks.txt due to variable no. of defined Pressures in seed or restart

[nickvandewiele]

When reading in a Restart folder containing Pdep reactions (pdepnetworks.txt, pdepreactions.txt), an unreadable line is detected and RMG crashes.

Might be due to e7e56cc?

pdepnetworks.txt

---

UnitsOfEa: kcal/mol
NumberOfFameTemps: 8
NumberOfFamePress: 5
NumberOfChebyTemps: 6
NumberOfChebyPress: 4
NumberOfPLogs: Could be different for seed and library reactions, but default for FAME-generated rates is probably 5

---

Error trace:
Reading reactions from Restart folder
Reading seed mechanism from directory /home/nickv/JP10_pyrolysis/JP10-20jan_from
ERROR: java.lang.NumberFormatException: For input string: "Could"
at java.lang.NumberFormatException.forInputString(NumberFormatException.java
at java.lang.Integer.parseInt(Integer.java:481)
at java.lang.Integer.parseInt(Integer.java:514)
at jing.rxnSys.ReactionModelGenerator.readPDepNetworks(ReactionModelGenerato
at jing.rxnSys.ReactionModelGenerator.initializeReactionSystems(ReactionMode
at jing.rxnSys.ReactionModelGenerator.modelGeneration(ReactionModelGenerator
at RMG.main(RMG.java:96)

CRITICAL: For input string: "Could"

RMG execution terminated at 2012-01-20 09:44:54

[rwest]

I think it's due to ba70d26

When the "restart" code was originally written, all PLOG rates had the same number of pressures, because they had all been created by FAME and were simply written out and read in via the restart code. This number of pressures was stored at the top of the restart file, and used to parse the rest of the file.

HOWEVER, when we added pressure dependence capabilities to the seed mechanisms (the seedmech2 branch 1157b93), we inevitably allow each PLOG reaction to have a different number of pressures (they could all come from different sources; some have only two, some have up to six). This broke the restart functionality. Looks like it hasn't been fixed yet.

You'll have to relax the assumption that all PLOGS have the same number of pressures, when parsing the pdep reactions restart file (perhaps reuse the seed mechanism reading code, which mustn't have this problem?)

Dup: issue #169

[nickvandewiele]

Thanks for the quick update, Richard.