From: Rajarshi G. <rg...@in...> - 2008-10-26 13:46:29
|
Is this approach the best way? It seems that as a result of this, one should in general use the new method (perceive..Unset..), since it may not always be apparent that some properties have been unset. From this point of view should the old perceive method be deprecated? Begin forwarded message: > From: "SourceForge.net" <no...@so...> > Date: October 26, 2008 5:00:48 AM EDT > To: no...@so... > Subject: [ cdk-Bugs-1969156 ] atomtype macthing and configuring > overwrites exact mass > > Bugs item #1969156, was opened at 2008-05-21 23:53 > Message generated for change (Settings changed) made by egonw > You can respond by visiting: > https://sourceforge.net/tracker/? > func=detail&atid=120024&aid=1969156&group_id=20024 > > Please note that this message will contain a full copy of the > comment thread, > including the initial issue submission, for this request, > not just the latest update. > Category: cdk-atomtype > Group: cdk-1.2.x/ >> Status: Closed >> Resolution: Fixed ------------------------------------------------------------------- Rajarshi Guha <rg...@in...> GPG Fingerprint: D070 5427 CC5B 7938 929C DD13 66A1 922C 51E7 9E84 ------------------------------------------------------------------- All science is either physics or stamp collecting. -- Ernest Rutherford |
From: Egon W. <ego...@gm...> - 2008-10-26 13:50:45
|
On Sun, Oct 26, 2008 at 2:46 PM, Rajarshi Guha <rg...@in...> wrote: > Is this approach the best way? Dunno, but as said, I don't want to change the behavior of the old code, regarding already configured fields... the idea of the 'old' perceive is to configure according to the given IAtomContainer... which does imply overwriting old values. > It seems that as a result of this, one > should in general use the new method (perceive..Unset..), since it > may not always be apparent that some properties have been unset. Sorry... you lost me... can't you just check that? > From this point of view should the old perceive method be deprecated? Post 1.2, sure. Egon -- ---- http://chem-bla-ics.blogspot.com/ |
From: Rajarshi G. <rg...@in...> - 2008-10-26 14:02:51
|
On Oct 26, 2008, at 9:50 AM, Egon Willighagen wrote: > On Sun, Oct 26, 2008 at 2:46 PM, Rajarshi Guha <rg...@in...> > wrote: >> >> It seems that as a result of this, one >> should in general use the new method (perceive..Unset..), since it >> may not always be apparent that some properties have been unset. > > Sorry... you lost me... can't you just check that? Right, but you also need to know when to check :) For example, the source of this report was the fact that SDF reading sets the exact mass. I didn't realize this till it was overwritten. Obviously one could write super defensive code and check every property beforehand, but that's not always practical (and also inefficient). Probably the simplest solution is to update the Javadocs >> From this point of view should the old perceive method be deprecated? > > Post 1.2, sure. Is this merged into trunk? ------------------------------------------------------------------- Rajarshi Guha <rg...@in...> GPG Fingerprint: D070 5427 CC5B 7938 929C DD13 66A1 922C 51E7 9E84 ------------------------------------------------------------------- Artificial intelligence has the same relation to intelligence as artificial flowers have to flowers. -- David Parnas |
From: Egon W. <ego...@gm...> - 2008-10-26 14:17:54
|
On Sun, Oct 26, 2008 at 3:02 PM, Rajarshi Guha <rg...@in...> wrote: > On Oct 26, 2008, at 9:50 AM, Egon Willighagen wrote: >> On Sun, Oct 26, 2008 at 2:46 PM, Rajarshi Guha <rg...@in...> > For example, the source of this report was the fact that SDF reading > sets the exact mass. I didn't realize this till it was overwritten. Right... that actually sounds like a but... did you file a report for that? > Obviously one could write super defensive code and check every > property beforehand, but that's not always practical (and also > inefficient). No, it depends on the application... no idea why the SDFile reader is actually perceiving atom types... not sure if it should at all... > Probably the simplest solution is to update the Javadocs > >>> From this point of view should the old perceive method be deprecated? >> >> Post 1.2, sure. > > Is this merged into trunk? No. Briefly tried updating trunk for recent 1.2 changes.. but ran into some conflicts for the builder3d stuff... Need to give that another attempt later... Egon -- ---- http://chem-bla-ics.blogspot.com/ |
From: Rajarshi G. <rg...@in...> - 2008-10-26 14:24:12
|
On Oct 26, 2008, at 10:17 AM, Egon Willighagen wrote: > On Sun, Oct 26, 2008 at 3:02 PM, Rajarshi Guha <rg...@in...> > wrote: >> On Oct 26, 2008, at 9:50 AM, Egon Willighagen wrote: >>> On Sun, Oct 26, 2008 at 2:46 PM, Rajarshi Guha <rg...@in...> >> For example, the source of this report was the fact that SDF reading >> sets the exact mass. I didn't realize this till it was overwritten. > > Right... that actually sounds like a but... did you file a report > for that? No > >> Obviously one could write super defensive code and check every >> property beforehand, but that's not always practical (and also >> inefficient). > > No, it depends on the application... no idea why the SDFile reader is > actually perceiving atom types... not sure if it should at all... Well one could argue that exact mass is not really an atom type property so it is OK for it to be set (though it should then be set by all readers) ------------------------------------------------------------------- Rajarshi Guha <rg...@in...> GPG Fingerprint: D070 5427 CC5B 7938 929C DD13 66A1 922C 51E7 9E84 ------------------------------------------------------------------- All life evolves by the differential survival of replicating entities. -- Dawkins |
From: Egon W. <ego...@gm...> - 2008-10-26 14:30:29
|
On Sun, Oct 26, 2008 at 3:23 PM, Rajarshi Guha <rg...@in...> wrote: >> no idea why the SDFile reader is >> actually perceiving atom types... not sure if it should at all... > > Well one could argue that exact mass is not really an atom type > property so it is OK for it to be set (though it should then be set > by all readers) What I mean, is that SDFileReader should just reproduce exactly what is in the file, and nothing more. So, exact masses in the file should be read from the file only, and not set from (some) common knowledge... Egon -- ---- http://chem-bla-ics.blogspot.com/ |
From: Rajarshi G. <rg...@in...> - 2008-10-26 14:45:13
|
On Oct 26, 2008, at 10:30 AM, Egon Willighagen wrote: > On Sun, Oct 26, 2008 at 3:23 PM, Rajarshi Guha <rg...@in...> > wrote: >>> no idea why the SDFile reader is >>> actually perceiving atom types... not sure if it should at all... >> >> Well one could argue that exact mass is not really an atom type >> property so it is OK for it to be set (though it should then be set >> by all readers) > > What I mean, is that SDFileReader should just reproduce exactly what > is in the file, and nothing more. > So, exact masses in the file should be read from the file only, and > not set from (some) common knowledge... Oh, right. Yes, I agree. I'll take a look at it today ------------------------------------------------------------------- Rajarshi Guha <rg...@in...> GPG Fingerprint: D070 5427 CC5B 7938 929C DD13 66A1 922C 51E7 9E84 ------------------------------------------------------------------- Every nonzero finite dimensional inner product space has an orthonormal basis. It makes sense, when you don't think about it. |