You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When PDB file is chosen as an option, check that the file is indeed a PDB file and give a clear error message indicating so.
Initially it was believed that the error message given by xssp would be enough, but it seems not to be the case. There are still users providing fasta files.
The text was updated successfully, but these errors were encountered:
The PDB parser ignores missing sections and ends up failing on invalid data because no chains could be parsed. Whilst this is robust, the error message shown isn't very useful:
if (mChains.empty())
throw mas_exception("empty protein, or no valid complete residues");
The PDB file specification has mandatory record types, such as HEADER. Ideally the parser should report missing mandatory sections, but that's a lot of effort.
@touwwouter
What do you think about adding a check where, if the HEADER record type isn't present (which according to the specification should be the first line and only appears once as a single line), that we raise a mas_exception with the text "Not a PDB file"?
(We talked about this, but given the recent activity I will comment here). More informative error messages are helpful. I think, however, that files should be allowed if they have ATOM/HETATM records. Even without HEADER.
This issue was copied from cmbi/xssp-api#64.
When PDB file is chosen as an option, check that the file is indeed a PDB file and give a clear error message indicating so.
Initially it was believed that the error message given by xssp would be enough, but it seems not to be the case. There are still users providing fasta files.
The text was updated successfully, but these errors were encountered: