read.pdb

[Lars Skjærven]

proposing to change default value of multi in read.pdb() to TRUE. I guess the only reason for having this FALSE is speed (?), but you dont save too much anyway:

> get.pdb("1d1d")

> system.time(read.pdb("1d1d.pdb", multi=FALSE))
   user  system elapsed 
  0.539   0.003   0.544 

> system.time(read.pdb("1d1d.pdb", multi=TRUE))
   user  system elapsed 
  0.680   0.000   0.683 

Also, I exchange calls to print with message (or cat). ok?

[Barry Grant]

How wold this effect regular pdbaln() - pdbfit() - etc workflows? Also I am sure what your print() to cat() change is about?

[Lars Skjærven]

neh.. pdbsplit() would be affected, so that we would need to include a multi argument there as well. thats a hassle, so I take it back. the other functions would be ok though.

there are calls to print in read.pdb so that the messaging gets a bit ugly:

> read.pdb("2RSC", multi=TRUE)
  Note: Accessing on-line PDB file
[1] "PDB has multiple END/ENDMDL records"
[1] "multi=TRUE: 'read.dcd/read.ncdf' will be quicker!"
  HEADER    HYDROLASE                               19-DEC-11   2RSC               

[Barry Grant]

I agree, pdbsplit() changes could get painful.

Also changing the print() call would be good.