On Wed, 2006-11-01 at 02:59 +1100, Edward d'Auvergne wrote:
pymol.cartoon() doesn't work for me. The commands go to pymol (they
appear in the tcl gui at least) but have no apparent effect. Just
doing:
hide everything
show cartoon
on the pymol cmd line has the expected effect.
This could be a PyMOL version issue. Which version are you using?
I'm using the open source v0.99rc6 version (that you have to compile
yourself). The PyMOL 'cmd.hide()' function also works on the non open
source binary build of the same version.
I've tracked this one down. My PDB file is not in the current working
directory, so path details need to be stripped before the cmd.hide(),
cmd.show() functions. So I have:
relax> pymol.view('test')
relax> PyMOL>reinitialise
Traceback (most recent call last):
File "/home/chris/bin/pymol/modules/pymol/parser.py", line 370, in
parse
exec(com2[nest]+"\n",pymol_names,pymol_names)
File "<string>", line 1, in ?
NameError: name 'reinitialise' is not defined
PyMOL>load ../abp_600MHz_tensorFrame.pdb
CmdLoad: "../abp_600MHz_tensorFrame.pdb" loaded as
"abp_600MHz_tensorFrame".
This is all good so far... but:
relax> pymol.cartoon('test')
relax> PyMOL>cmd.hide('everything','../abp_600MHz_tensorFrame')
PyMOL>cmd.show('cartoon','../abp_600MHz_tensorFrame')
PyMOL>util.cbss('../abp_600MHz_tensorFrame', 'red', 'yellow', 'green')
fails because there is no such object (in pymol)
as ../abp_600MHz_tensorFrame. It should be
cmd.hide('everything','abp_600MHz_tensorFrame')
etc.
Ah. I think we need a new function within the IO file 'io.py'. Maybe
something called 'self.file_root()'. It should be pretty simple to
implement. Maybe split by the path separator 'path.sep', take the
last element, split by '.', then return the first element of the new
array. Would you like to have a go at adding it Chris? I won't have
commit access for a while. A complementary function 'self.file_ext()'
could be added for future use as well.
I've added the io.file_root() function to both the 1.3 line and
tensor_pdb branch, and made the fix to pymol.cartoon().
I deemed file_ext() a bit trivial to demand its own function, the
desired functionality being available as:
file_ext = os.path.splitext(file)[1]
And finally, if I do pymol.tensor_pdb(), then close pymol and then do
pymol.view() in the same run, I get the tensor back again (I was just
expecting to get my protein).
This is a feature. The PyMOL functions have inherited the command
history feature of the Molmol code. Any commands generated by relax
are rerun if you reopen PyMOL from relax. There is the user function
'pymol.clear_history()' which I forgot to mention in the original
email.
Fine. As long as it is run aware. If I have different tensors (or
different structures?) in separate runs but the same relax instance
(full_analysis.py, eg.), I don't want the two getting confused when I
start looking at them. The fact that pymol.clear_history() doesn't take
a run argument suggests that there might be the potential for such
confussion?
Well, it currently accepts the run argument but if you have a close
look at the code, you'll see that the pymol and molmol IO pipes
('self.relax.data.pymol' and 'self.relax.data.molmol') aren't
associated with a run. It's not worth worrying about as the proposed
redesign of the relax data model will cause the IO pipes to be tied to
a run.
Oh, the IO pipes aren't to be confused with the part of the proposal
about data pipes where 'runs' will be renamed 'pipes'.
Edward
P.S. I'll create a new post for the rest of your message.
Re: PyMOL is now supported.