#1290 InChIGenerator endless loop

[CoderX]

InChIGenerator running endless loop when passed the attached sdf.

my (partial) code:

while(sdfreade.hasNext()) {
...some code getting sid, etc from sdf...

InChIGenerator inchigen = InChIGeneratorFactory.getInstance().getInChIGenerator(mol);

//stuck here

inchiKey = inchigen.getInchiKey(); // generate the inchiKey based on mol structure

...some more code...

}

[John May]

Unfortunately, this is a problem with the actual InChI algorithm. If we run it through the the C binary it also hangs for a while then times out. CDK uses the InChI C code via JNI so I don't think there is much we could do, sorry.

[intrepid ~/Downloads]: inchi subst_141382403.sdf 
InChI version 1, Software version 1.03 Build of June 15, 2010
Opened log file 'subst_141382403.sdf.log'
Opened input file 'subst_141382403.sdf'
Opened output file 'subst_141382403.sdf.txt'
Opened problem file 'subst_141382403.sdf.prb'
The command line used:
"inchi subst_141382403.sdf"
Generating standard InChI
Input format: MOLfile
Output format: Plain text
Full Aux. info
Timeout per structure: 60.000 sec
Up to 1024 atoms per structure

Error -30008 (Time limit exceeded) structure #1 component 1.
End of file detected after structure #1.   
Finished processing 1 structure: 1 error, processing time 0:01:00.00

It's probably (almost definitely) getting stuck in the symmetry group calculation.

[Egon Willighagen]

Dear CoderX,

please contact the InChI mailing list. Of course, double check that the file content is correct, but if it is, you have indeed run into a problem with the InChI library. Their mailing list can be found at:

https://sourceforge.net/mail/?group_id=173262

Egon

[Egon Willighagen]

Sorry, that was the wrong link. That was for the JNI-InChI mailing list...

The InChI mailing list is found here:

https://sourceforge.net/mail/?group_id=136669

[CoderX]

Thanks for the info... I didn't know how else to troubleshoot this.