'string.split()' causes issue in 'Pymol-script-repo/modules/ADT/MolKit /pdbParser.py'

[WeibinRen]

See line 1392 as example,
atom._coords = [ [ float(rec[5]), float(rec[6]), float(rec[7]) ] ]
the rec here comes from line 1308:
rec = split(rec)
Sometimes protein PDB file content looks like(protein 2p7a from PDBBind database):
ATOM 406 O ASP A 259 66.586 -29.223-100.911 1.00 14.83 O
If directly use string.split() here, the number -29.223-100.911 will not be split correctly.
It will raise error like:
ValueError: could not convert string to float: '-29.223-100.911'

[WeibinRen]

I solved this by re-write split function:

def split(s):
    s_list = s.split()
    idx = 0
    while idx < len(s_list):
        item = s_list[idx]
        if '-' in item[1:]:
            index_minus = item.find('-', 1)
            s_list[idx] = item[:index_minus]
            s_list.insert(idx+1, item[index_minus:])
            idx = 0
        else:
            idx += 1
    return s_list

There should be some better way, I put the code here just hope this can help somebody in case they need this.

[WeibinRen]

Also line 1309 seem not 100% correct to use ==10

[pslacerda]

Hi, can you rewrite using a fixed string length like said at PDB ATOM record and submit as a pull request?

rec[31:39], rec[39:47]...

[pslacerda]

I don't know how to test this plugin. If you rewrite like I said, i'll accept as is.