Dynamics failing to run in v3.0.2 when dynamics.length is set

[jevandezande]

Using conda CREST v3.0.2 on linux, the dynamics fails to run if length is set, while the same input runs dynamics with v3.0.1. Removing the length keyword allows the dynamics to run as expected.

To Reproduce
input.toml

input="geom.xyz"
runtype="md"

[dynamics]
length_ps=5

geom.xyz

3
H 0 0 0
O 0 0 1
H 0 1 1

Relevant output:

...

 $ crest input.toml

 reading input.toml
********************************************************************************
* INPUT FILE input.toml content (without comments):
********************************************************************************
* input = "geom.xyz"
* runtype = "md"
* [dynamics]
* length_ps = 500
********************************************************************************

> Setting up backup calculator ... done.
 ----------------
 Calculation info
 ----------------
> User-defined calculation level:
 : xTB calculation via tblite lib
 : GFN2-xTB level
 :   Molecular charge    : 0
 :   Fermi temperature   : 300.00000
 :   Accuracy            : 1.00000
 :   max SCC cycles      : 500


       _                             _
    __| |_   _ _ __   __ _ _ __ ___ (_) ___ ___
   / _` | | | | '_ \ / _` | '_ ` _ \| |/ __/ __|
  | (_| | |_| | | | | (_| | | | | | | | (__\__ \
   \__,_|\__, |_| |_|\__,_|_| |_| |_|_|\___|___/
         |___/

 =============================
  # threads =           1
 =============================

 Input structure:
  3

 H          0.7071067810       -0.6279826729        0.0000000000
 O          0.0000000000        0.0791241086        0.0000000000
 H         -0.7071067814       -0.6279826723        0.0000000000

 ----------------
 Calculation info
 ----------------
> User-defined calculation level:
 : xTB calculation via tblite lib
 : GFN2-xTB level
 :   Molecular charge    : 0
 :   Fermi temperature   : 300.00000
 :   Accuracy            : 1.00000
 :   max SCC cycles      : 500


> Molecular dynamics settings
 MD time /ps        :      0.00
 dt /fs             :      5.00
 temperature /K     :    300.00
 max steps          :         0
 block length (av.) :      1000
 dumpstep(trj) /fs  :    100.00    20
 # deg. of freedom  :         7
 thermostat         : berendsen
 SHAKE constraint   :        T
 # SHAKE bonds      :         2 (all bonds)
 hydrogen mass /u   :   2.00000

> Starting simulation

           time (ps)       <Epot>        Ekin     <T>       T            Etot

 average properties
 ----------------------
 <Epot> / Eh          :   0.0000000000000000
 <Ekin> / Eh          :   0.0000000000000000
 <Etot> / Eh          :   0.0000000000000000
 <T> / K              :   0.0000000000000000
 normal MD termination
 MD run completed successfully
 Trajectory written to crest_dynamics.trj

 -----------------
 Wall Time Summary
 -----------------
 CREST runtime (total)               0 d,  0 h,  0 min,  0.017 sec
 ------------------------------------------------------------------
 Molecular dynamics (MD)    ...        0 min,  0.001 sec (  5.227%)
 I/O and setup              ...        0 min,  0.016 sec ( 94.773%)
 ------------------------------------------------------------------
 * wall-time:     0 d,  0 h,  0 min,  0.017 sec
 *  cpu-time:     0 d,  0 h,  0 min,  0.255 sec
 * ratio c/w:    15.411 speedup
 ------------------------------------------------------------------
 * Total number of energy+grad calls: 1

 CREST terminated normally.

[pprcht]

Can you try setting the argument as a float please? I.e. length_ps=5.0

[jevandezande]

Switching to a float makes it properly read the length.