PDB file recognition about pdbtbx HOT 6 CLOSED

I haven't really looked at the parsing function yet, the last time I tried it scared me :D
If I get around to it I'll take a look but no promises...

from pdbtbx.

I totally agree that such an error is useful. I propose to make them be triggered by the amount of atoms parsed, so if there are no atoms found when the parsing is finished it should generate an error. I would assume that a breaking error is more appropriate, but if you think otherwise please let me know.

from pdbtbx.

Yes, this is pretty much exactly what I was thinking.

from pdbtbx.

Great, then I will build it in in the future, but if you want you could also take a shot.

from pdbtbx.

So I just took a look at the parsers. It seems to me a good place for this validation would be the validate function since it's called during the parsing of PDB and mmCIF files and has &PDB as argument so checking for presence of any atoms would be trivial. However, the functions seems to only be used for validation of models if more than one is present. Do you think writing a separate function would be more appropriate or would expanding this one be suitable?

from pdbtbx.

I submitted a tentative pull request with the most straightworward way I could think of. I couldn't add any sensible context to the error since the validate function doesn't have access to the file name but I think that's not much of an issue.

from pdbtbx.