Opened 2 years ago

Closed 2 years ago

#340 closed defect - wrong answer (fixed)

SPECIES command issues

Reported by: Gary J. Ferland Owned by: nobody
Priority: blocker Milestone: C17_branch
Component: etc Version: trunk
Keywords: Cc:


The SPECIES command allows details of the treatment of database species to be changed. Tests show the following:

The command is not defended against user error. The command

species "ty3" lvls 3

does not result in any error message, even though the species name is invalid, and the option is misspelled. This is especially important because we now are case sensitive. It will be common for a user to make a mistake with the species name, since we were not case sensitive in the past.

The command parser does the following

species "Si+" levels 10
species "Si+" levels all
species "Si+" levels=all

The first and last work as expected while the middle does not work and does not result in an error.

species level does not work for Fe, probably because we use more levels for this element, as explained in Lykins+13, section 2.9.

Change History (2)

comment:1 Changed 2 years ago by rjrw

species "Si+" levels all

in fact prints the diagnostic

Incorrect type for 'LEVELS' option to species

I'll make this more chatty.

The problem in parsing 'levels all' (as against 'levels=all') is to distinguish whether <word> in

levels <word>

is an argument to the 'levels' option, or a distinct option (such as 'lte') with the argument to 'levels' missing in error. Requiring an '=' in the case of text arguments disambiguates this. This could grouping be achieved in different ways, e.g. using commas.

species level certainly doesn't work for Fe II. This has been waiting on the merge of the stout6 branch for a little while.

comment:2 Changed 2 years ago by Gary J. Ferland

Resolution: fixed
Status: newclosed
Note: See TracTickets for help on using tickets.