Thank's for the report (-in1new is not really recommended in general anymore). I guess the fix is easy and the lines below should do it in make_inso_lapw.

#     set numberL = `head -$lineint < $filein1 | tail -1 |cut -c 11-13`
      set numberL1 = `head -$lineint < $filein1 | tail -1`
      set numberL = $numberL1[2]

On 04/25/2016 05:21 PM, Ulrich Wedig wrote:
Dear colleagues,

I'm using WIEN2k 14.2.
After a SCF run with the -in1new switch, initso_lapw (make_inso_lapw)
doesn't interpret the case.in1 file correctly, due to the following reason:

write_in1_lapw, invoked with the -in1new switch, writes line 3 of
case.in1 free format.
E.g. in my case:

  .2370444277   4   0      global e-param with N other choices, napw

make_inso_lapw, called by initso_lapw, reads case.in1 in a formatted way
(line 86 in my version), if RLO should be added:

         set numberL = `head -$lineint < $filein1 | tail -1 |cut -c 11-13`
         @ lineint = $numberL + $lineint

With the example above, make_inso_lapw sets numberL = 77, i.e assumes 77
lines of exceptions in case.in1.

As a consequence, the script tries to read 77 lines, prints all lines
with l=1, prints '@: Expression Syntax.' and returns to initso_lapw,
continuing this script.
case.inso is not finalized beyond line 4.

As a workaround I truncated the Etrial values in case.in1 to get the
required format.

Best regards,
Ulrich  Wedig


Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-1-58801-165300             FAX: +43-1-58801-165982
Email:    WIEN2k:
Wien mailing list

Reply via email to