Hello,
I don't think anything would have been corrupted. The binary data is
described in the data model (so its size, etc.) but the access is
completely independent of the data model. In particular, the referencing
information can be set completely independently of whether the data
(partly or completely) exists or not. The main effect would be that there
would sometimes be exceptions when drawing the windows so some things
(contours, peaks) wouldn't appear when they should have.
Wayne
On Fri, 22 Jun 2007, Patrick van der Wel wrote:
> Well, the problem was that it seemed to partially load the spectrum as far as
> Analysis was concerned: it showed up in the spectrum list, and I could go in
> and even specify the aliasing limits in the referencing popup. (I had
> selected the 'Skip Verification dialogs' option during the loading, so those
> were skipped, but it never got far enough the specify the type of experiment)
>
> I was glad to notice this problem this time, since I managed to 'back out' of
> the program without saving, but wonder if I might not have corrupted the
> project had I saved it or continued using it.
>
> Patrick
>
> On Friday 22 June 2007 11:04, Wayne Boucher wrote:
> > Hello,
> >
> > Those error messages are from just trying to calculate the noise estimate,
> > never mind trying to draw the thing! For this noise estimate error we
> > probably should trap it a bit better and put the message in a popup.
> > (Because it's a one-off.) But for similar messages you'd get with drawing
> > we probably should continue just to print to screen (it would get far too
> > annoying otherwise).
> >
> > Wayne
> >
> > On Thu, 21 Jun 2007, Patrick van der Wel wrote:
> > > I encountered the following error when I tried to open a spectrum file
> > > just now. Admittedly, this was a file which had gotten corrupted (i.e.
> > > truncated), but perhaps it would be preferable to pop up an error message
> > > rather than quietly fail (except for the feedback in the console window)?
> > >
> > > (this was a 2D 13C-13C spectrum in NMRPipe format, if that helps)
> > >
> > > Patrick
> > >
> > > >>> WARNING: file "/home/user/research/jun07i12_068.ft2" seems to be
> > > >>> mostly 0,
> > >
> > > isDataBigEndian() returning True
> > > Exception in Tkinter callback
> > > Traceback (most recent call last):
> > > File "/usr/lib/python2.5/lib-tk/Tkinter.py", line 1403, in __call__
> > > return self.func(*args)
> > >
> > > File
> > > "/home/user/research/software/ccpnmr/ccpnmr1.0/python/ccpnmr/analysis/Ope
> > >nSpectrumP opup.py", line 206, in openSpectra
> > > spectrum.isBigEndian = True
> > >
> > > File
> > > "/home/user/research/software/ccpnmr/temporaryReleaseDir/ccpnmr/ccpnmr1.0
> > >/python/cc p/api/nmr/Nmr.py", line 18004, in setIsBigEndian
> > >
> > > File
> > > "/home/user/research/software/ccpnmr/ccpnmr1.0/python/ccpnmr/analysis/Ana
> > >lysis.py", line 1407, in updatedSpectrumFileDetails
> > > self.initBlockFile(spectrum)
> > >
> > > File
> > > "/home/user/research/software/ccpnmr/ccpnmr1.0/python/ccpnmr/analysis/Ana
> > >lysis.py", line 1161, in initBlockFile
> > > Util.defaultContourLevels(spectrum)
> > >
> > > File
> > > "/home/user/research/software/ccpnmr/ccpnmr1.0/python/ccpnmr/analysis/Uti
> > >l.py", lin e 1258, in
> > > defaultContourLevels v = 3 * getNoiseEstimate(spectrum) /
> > > spectrum.root.currentAnalysisProject.globalContourScale
> > >
> > > File
> > > "/home/user/research/software/ccpnmr/temporaryReleaseDir/ccpnmr/ccpnmr1.0
> > >/python/cc pnmr/analysis/ExperimentBasic.py", line 444, in
> > > getNoiseEstimate d = block_file.getValue(pt)
> > > BlockFile.error: could not get point: 934 448
>
|