votca / ctp Goto Github PK

Undocumented options
constant, generate_from, generate_unsafe

it is useful, make it a separate option
<load help=""> 
    <file help=""></file>       
    <unsafe help=""></unsafe>
</load>

Adjust to use default values and extend the help xml file

currently a python script; rewrite in a similar way to Gaussian

Move all classes to namescpaces ctp, moo, kmc

the current version is correct only for 90deg angles 

1. Implement spheres method in eoutersphere calculator (so far only 
time-consuming dielectric method implemented).
2. Implement constant method in eoutersphere calculator.

Program hangs for large atomic numbers - QmTopoloqy or Crgunit classes?

complains ERROR: Estatify first.

Sasha: if first three atoms in segments.xml and map.xml do not match mapping is 
not working properly even if symmetry=3.

Add enum(helpLong, helpShort) 
Move PrintDescription() to QMApplication (out of the ctp_run)

1. Consistent Thole polarizabilities needed (AMOEBA <-> Thole)
2. Introduce constant scaling factors for charge states 

Enumeration sometimes starts with 0 sometimes with 1. This needs to be unified. 
The same applies for crgunit ids. Internal and external representation must 
also be identical.

ctp fits better to charge transport

UPDATE ... WHERE ... = ? AND ... = ? takes too long.

Check if all object created via an objectfactory Create method are destroyed at 
the end of the run (qmpackage, event)

Calculates energy and density profiles, the input should be restructured

<profiles>
    <axis>
        <direction>
        <min>
        <max>
        <bin>
    <particles>
        <type>
        <range>
    <output>
        <density>
        <energy>

 Or even output all in one file? See gromacs with xmgrace -nxy bla.agr

Check that
1) the options match the number of fragments
2) all positions are set
3) all positions are set only once

This check has to be a combination of moo and ctp fixes.

Should be a wrapper to calculators (current one will be for one hopper)

useful to check moo against the projection methods 

Base class for beads in CSG and neighbourlist based on this base class

Currently only a single frame can be stored in the database. The problem is 
connected to id labelling not unique for molecules and rigid fragments 
(qmbeads).

e.g.
--options in csg vs --opt in ctp

Needed to restart KMC runs

The problem is that sqlite is pulled in as dependency by tools, but the include 
pathes are not set properly.

Is there any other way than explicitly checking for sqlite in moo, ctp as well. 
What is the best way to fix this in cmake?

GDMA currently needs g03 (fixed by someone?)
Basis set provided by an intermediate object (e.g. orbitals)?
Not parallel. Worth parallelization?

Input files, state file, and commands have changed. Update the manual.

adapt to the base class QMPackage (currently does not compile) 

for atoms with residue names of 3 symbols and 2 digits as an atom number the 
HETATM line is shifted in traj_cg.pdb
ATOM     12 PTR1 PBT     1      53.536   1.292   1.230  0.00  0.00
HETATM   12 PTR12 REU     1      53.393   1.958  1.9622  0.00  0.00

similar to the EventFactory object for Carriers 

need to figure out how to filter xml tags using Property class

1) copy & paste license headers from csg into each file. Don't forget to adjust 
the names
2) copy LICENSE file
3) Add README file mentioning the citation. See csg as template.
4) NOTICE file for every file that has different licencing information

In mixtures we need a neighbor list based on cutoffs which depend on types of 
molecules

Add attributes to every option
help=""
default=""
unit=""

In ecoulomb.cc we have a loop over the Molecules

MoleculeContainer::iterator imol;
    for (imol = atop->Molecules().begin(); imol != atop->Molecules().end(); imol++)

because the atomistic beads can only be accessed by

Bead *bi = mol->getBead(i);

The Center of mass etc is taken from the corresponding CrgUnit
mol->getUserData<CrgUnit > ()->GetCom()

This works only if the Molecule is the CrgUnit.

->Change the loop to work only with CrgUnits (aka conjugated segments)

Similar for eoutersphere.cc

xml file format has been extended to better reference things in the latex 
manual. ctp_run --list now reads in the brief description from the xml file. 

will give improved error message

This can happen if e.g. a double value is nan, writing to the database fails, 
but no error is thrown. This leads to a broken database without the user 
realiying it.

Check how to notice the error in sqlite commands.

should not be the case. check QMtopology and JCalc

Needs a lot of clean up, example:
$ ctp_dipro 123 456
/home/christoph/votca/bin/ctp_dipro: line 189: do_external: command not found

- rename it to *.in for getting version in there
- make use of csg_call functions as ctp depends on csg anyway
- rm do_external calls or source function_common

Comments?

the link to the code was revised; a bug in the rotation of rigid fragments was 
fixed. New test is needed.

Currently it is not verz clear what the function does. This has already caused 
one bug in site energy histogram

Change Crg1 to something more meaningfull like getPBCGhostCopy (there has to be 
a better name)

What steps will reproduce the problem?
1.
2.
3.

What is the expected output? What do you see instead?


Please use labels and text to provide additional information.

Useful if a user wants to print some run-time/input information of a program or 
a calculator.

Move csg_property to tools (might be used by both ctp and csg parts)

Make sure nothing can go wrong with <map2md>0 (or 1)</map2md> switch: Rename to 
<use_coords>qm (or md)</use_coords>.

derive Gaussian, Turbomole from QMWrapper and register them in a factory
adjust edft, idf to use QMWrapper

shall be changed to 
list<string>::iterator findAttribute(string key)
type getAttribute<type>(pos) 
At the moment two lookups are needed and it is not safe

Input file should support multiple frames

Bad atom type   reported by ctp map if moo cannot process it

ctp_run --list
ctp_run --description calculator_name

... should only appear for Jortner rates.