Menu
▾
▴
animl-develop
|
From: Burkhard S. <b_...@us...> - 2006-04-12 00:16:33
|
Robert, > in conversion of JCAMP-DX files to AnIML I am concerned about > RESOLUTION as well since this is NOT required in JCAMP-DX > so will not be present in thousands of files already in existence. > > Our PE FTIR machines can be set at 1,2 or 4 cm-1 resolution so > even then I am not always sure what was the setting when the > info was recorded. > > What should go into the field in the AnIML file exported > from a JCAMP-DX file when it is given as REQUIRED? This is probably an error in the technique definition. I've filed a bug for this (#1468925). > Another problem I have.... > I am attaching a section of a file from Anh Dao I'm snipping out a few things for brevity. > > ============================================================== > <AnIML xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" > xsi:noNamespaceSchemaLocation="animl-core.xsd" > xmlns:ds="http://www.w3.org/2000/09/xmldsig#" version="1.0"> > <SampleSet> > <Sample sampleID="Polystyrene_IR" derived="false"> [...] > </Sample> > <Sample sampleID="Reference_IR" derived="false"> [...] > </Sample> > </SampleSet> > <MeasurementData> > <ExperimentStep name="Measurement_Polystyrene_IR"> > <Author type="device"> > <Name>Magna System 550</Name> > </Author> > <Timestamp>2002-11-06T13:03:04</Timestamp> > <SamplesUsed> > <SampleRef sampleID="Polystyrene_IR" > role="MeasurementSample" purpose="consumed"/> > </SamplesUsed> > =========================================================================================== > I can not understand how (without human intervention) it will be > possible to decide > which of the sample sampleID labels refers to a SAMPLE or BACKGROUND or > REFERENCE. This is determined by the Sample Role, as given in the role attribute of the SampleRef element. Note that this happens inside the ExperimentStep, not within the sample definition itself. At this point I am not sure if the list of sample roles for the IR technique is sufficient. [This is something where the technique NDRs will come into play.] > The labels used above are "polytyrene_ir" and "reference_ir" which can > be anything > the user puts there and not specific for a SAMPLE or REFERENCE. True. The sampleId does not contain any semantics. It is just an identifier. > How do I decide what the sample state is for the SAMPLE when I don't > know if it > is the SAMPLE or REFERENCE tag that I have just read?? You can have multiple SampleRef elements inside the ExperimentStep (as declared in the Technique Definition). For example, you could have a measured sample (role: Run Sample) and a sample used for background subtraction (role: Background Sample). The names of these roles will be standardized in the Technique NDRs. Best regards, Burkhard |
|
From: Burkhard S. <b_...@us...> - 2006-04-12 00:22:20
|
Robert, that's odd. I haven't had problems with SourceForge mail lately. We should keep an eye on this. I've replied to your message directly and sent a copy to the list so that other people can follow as well. > By the way > Have you looked at > jspecview.sf.net > > The application will read and write JCAMP-DX/AnIML files > and the applet will import both OK but file restrictions > are limiting writing so far since the applet is not signed. > > In addition I am having some problems with DOM routines > and think I am look at StAX instead Yes, I looked at it and find it a great project. I tried it with some JCAMP data and it worked very well. Haven't tried the AnIML support yet. I agree, DOM is a pain. If you'd like something with a similar interface, I suggest JDOM. It's quite a bit easier to use (http://www.jdom.org). I haven't used StAX yet, so I can't comment on it. Best regards, Burkhard |
|
From: Prof. R. J. L. <rob...@uw...> - 2006-04-12 01:07:24
|
Burkhard, this suggests processing the file twice or holding info in temp variables until I get down to the ExperimentStep to find the role to then try to find out which of the sampleIDs correspond to the actual sample? In my case lets say I am trying to read the file to fill in JCAMP-DX header items so that if I export as JCAMP-DX I know whether I have found the MP of the sample or the MP of the reference (for which there is NO JCAMP-DX label). That is, I've read in the sampleID's and the information given for each but don't know how to fill in the header labels until I reach the Role under ExperimentStep. It seems to me to be MUCH better to have the SampleID say exactly that this is the sample the reference or blank etc. There can't be too many alternatives to deal with. There can be another attribute that gives the info like "polystyrene_IR" but at least it is clear when processing the Sample section when I find a MP or Temp or refractive index whatever that the value is related to the SAMPLE and not the BLANK or REFERENCE. Doesn't this sound less confusing? Thanks Robert on 11/04/2006 07:16 PM Burkhard Schaefer said the following: > Robert, > >> in conversion of JCAMP-DX files to AnIML I am concerned about >> RESOLUTION as well since this is NOT required in JCAMP-DX >> so will not be present in thousands of files already in existence. >> >> Our PE FTIR machines can be set at 1,2 or 4 cm-1 resolution so >> even then I am not always sure what was the setting when the >> info was recorded. >> >> What should go into the field in the AnIML file exported >> from a JCAMP-DX file when it is given as REQUIRED? > > This is probably an error in the technique definition. I've filed a bug > for this (#1468925). > > > >> Another problem I have.... >> I am attaching a section of a file from Anh Dao > > I'm snipping out a few things for brevity. > >> >> ============================================================== >> <AnIML xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" >> xsi:noNamespaceSchemaLocation="animl-core.xsd" >> xmlns:ds="http://www.w3.org/2000/09/xmldsig#" version="1.0"> >> <SampleSet> >> <Sample sampleID="Polystyrene_IR" derived="false"> [...] >> </Sample> >> <Sample sampleID="Reference_IR" derived="false"> [...] >> </Sample> >> </SampleSet> >> <MeasurementData> >> <ExperimentStep name="Measurement_Polystyrene_IR"> >> <Author type="device"> >> <Name>Magna System 550</Name> >> </Author> >> <Timestamp>2002-11-06T13:03:04</Timestamp> >> <SamplesUsed> >> <SampleRef sampleID="Polystyrene_IR" >> role="MeasurementSample" purpose="consumed"/> >> </SamplesUsed> >> =========================================================================================== > > >> I can not understand how (without human intervention) it will be >> possible to decide >> which of the sample sampleID labels refers to a SAMPLE or BACKGROUND >> or REFERENCE. > > This is determined by the Sample Role, as given in the role attribute of > the SampleRef element. Note that this happens inside the ExperimentStep, > not within the sample definition itself. > > At this point I am not sure if the list of sample roles for the IR > technique is sufficient. [This is something where the technique NDRs > will come into play.] > > >> The labels used above are "polytyrene_ir" and "reference_ir" which can >> be anything >> the user puts there and not specific for a SAMPLE or REFERENCE. > > True. The sampleId does not contain any semantics. It is just an > identifier. > > >> How do I decide what the sample state is for the SAMPLE when I don't >> know if it >> is the SAMPLE or REFERENCE tag that I have just read?? > > You can have multiple SampleRef elements inside the ExperimentStep (as > declared in the Technique Definition). For example, you could have a > measured sample (role: Run Sample) and a sample used for background > subtraction (role: Background Sample). > > The names of these roles will be standardized in the Technique NDRs. > > Best regards, > Burkhard > -- Prof. Robert J. Lancashire mailto:rob...@uw... Department of Chemistry Tel (876) 512 3026 University of the West Indies, Kingston 7 Fax (876) 9771835 Mona Campus, JAMAICA. http://wwwchem.uwimona.edu.jm/chrl.html |
Thanks for helping keep SourceForge clean.
X