Re: [XMLPipeDB-developer] GM Builder post taxon merge
Brought to you by:
kdahlquist,
zugzugglug
Menu
▾
▴
Re: [XMLPipeDB-developer] GM Builder post taxon merge
|
From: Richard B. <rbr...@gm...> - 2011-04-15 01:53:21
|
OK, found the method and reviewed it.
It makes sense and I believe it should work as is unless I'm reading the
code wrong.
I don't see how if the taxon id is included in the Mtuberculosis species
profile (which it is) why this shouldn't work?
Help please, what am i missing?
richard
On Thu, Apr 14, 2011 at 11:04 AM, Richard Brous <rbr...@gm...> wrote:
> Gotcha, will check it out later today as well as post gdb's up to the wiki.
>
> Rb
>
> Sent from my iPhone
>
> On Apr 13, 2011, at 11:32 AM, "John David N. Dionisio" <do...@lm...>
> wrote:
>
> > Greetings,
> >
> > Sorry for the delay; I ended up digging through the code a bit so I would
> be sure of the response.
> >
> > getSpeciesProfilesFound does indeed feed the combo box that you want to
> touch, but it turns out that the array returned by this method is not built
> here.
> >
> > The code that builds the array returned by getSpeciesProfilesFound is
> located in UniProtDatabaseProfile; the method is called checkRequirements.
> This is the method that performs the query which retrieves the available
> taxon IDs from the relational database.
> >
> > The intent of the method is this: if a custom species profile is found
> that matches the taxon ID, then that species profile is added to the list.
> If a custom species profile is *not* found, a generic species profile
> initialized to the taxon ID is added instead.
> >
> > That is the method's intent. As you have seen, it may need some cleanup
> and/or fixing. But there it is for you:
> UniProtDatabaseProfile.checkRequirements.
> >
> > John David N. Dionisio, PhD
> > Associate Professor, Computer Science
> > Loyola Marymount University
> >
> >
> >
> > On Apr 12, 2011, at 7:03 PM, Richard Brous wrote:
> >
> >> OK, in ExportPanel1.java i found the following:
> >>
> >> 1. COMBOBOX DEFINED ON LINE 59
> >> private
> >>
> >> JComboBox speciesComboBox;
> >> 2. THEN COMBOBOX IS INSTANTIATED AND ACTION LISTENER ASSIGNED STARTING
> ON LINE 141
> >>
> >> // Species | JComboBox | speciesFound | JLabel | Description
> >>
> >>
> >> // | JTextField | customizable name
> >>
> >> speciesComboBox = new JComboBox();
> >>
> >> speciesComboBox.addActionListener(new ActionListener() {
> >>
> >> public void actionPerformed(ActionEvent arg0) {
> >> speciesProfileSelected(
> >>
> >> speciesComboBox.getSelectedItem());
> >> }
> >>
> >> });
> >>
> >> 3. NEXT CODE BLOCK ADJUSTS THE DATABASE PROFILE WHICH IS ALWAYS UNIPROT,
> THEN BASED ON DATABASE PROFILE
> >> IT POPULATES THE SPECIES COMBOBOX STARTING LINE 210
> >>
> >>
> >> * Adjusts the dynamic content when a database profile is selected.
> >>
> >> *
> >>
> >> *
> >>
> >> @param selectedItem
> >> */
> >>
> >>
> >> protected void databaseProfileSelected(Object selectedItem) {
> >>
> >> if (selectedItem instanceof DatabaseProfile) {
> >> DatabaseProfile selectedProfile = (DatabaseProfile)selectedItem;
> >>
> >> ExportToGenMAPP.setDatabaseProfile(selectedProfile);
> >>
> >>
> >> profileDescriptionTextArea.setText(selectedProfile.getDescription());
> >>
> >> modSystemTextField.setText(selectedProfile.getMODSystem());
> >>
> >>
> >> //This is populating the list of items in the speciesProfile combobox
> >>
> >> // based on what database was selected.
> >>
> >> speciesComboBox.removeAllItems();
> >>
> >> for (SpeciesProfile speciesProfile :
> selectedProfile.getSpeciesProfilesFound()) {
> >>
> >> speciesComboBox.addItem(speciesProfile);
> >> }
> >>
> >> }
> >>
> >> }
> >>
> >> SO I THINK THE METHOD I NEED TO SOURCE IS:
> selectedProfile.getSpeciesProfilesFound())
> >>
> >> SOUND GOOD SO FAR??
> >>
> >> Richard
> >>
> >>
> >>
> >>
> >>
> >> On Tue, Apr 12, 2011 at 6:18 PM, Richard Brous <rbr...@gm...>
> wrote:
> >>
> >> OK, I'll look around and attempt to sniff out the code regarding
> populating the drop down menu on the first export screen.
> >>
> >> But that said, I was surprised it didn't work since the functionality
> 'seemed' to be in place for the two working species (P_a and S_a). And I
> noticed that the working species don't always show in the drop down box, so
> figured they likely aren't hard coded in.
> >>
> >> Richard
> >>
> >>
> >> On Tue, Apr 12, 2011 at 6:04 PM, John David N. Dionisio <do...@lm...>
> wrote:
> >>
> >> Greetings,
> >>
> >> First off, recall that the taxon ID that is given to the species profile
> is not involved in building the dropdown menu; it is used to match the
> chosen taxon ID from that menu with a species profile to use. That dropdown
> menu is built from a database query that retrieves all available taxon IDs
> and species names in the PostgreSQL database.
> >>
> >> With that in mind, I would tackle this as follows:
> >>
> >> - First, track down the code that builds the data displayed in that
> JComboBox. It should be trackable from the wizard panel code. If you're
> having trouble finding it, let me know and I can take a stab at following
> the breadcrumbs, or you can ask Don.
> >>
> >> This code should ultimately lead you to the database query that derives
> the overall list of available taxon IDs and species names in the PostgreSQL
> database.
> >>
> >> - Now, armed with this database query, you can verify the validity of
> this query directly in PostgreSQL. Essentially you need to make sure that
> this query does indeed retrieve the list of all taxon IDs and species names
> that are present in the PostgreSQL database.
> >>
> >> If the query is correct, then the issue is in the way the JComboBox is
> put together. This may need a little more knowledge of Swing that you have,
> in which case send the issue my way and I can take a look.
> >>
> >> If the query is incorrect, then it should be corrected. If the Swing
> code is correct, then things should work. If the Swing code is also
> incorrect, then you are at the previous case, and the Swing code now needs
> to be addressed (safe in the knowledge that the data part has been fixed).
> >>
> >> Let me know how that plan sounds to you.
> >>
> >> John David N. Dionisio, PhD
> >> Associate Professor, Computer Science
> >> Loyola Marymount University
> >>
> >>
> >>
> >>
> >> On Apr 12, 2011, at 5:56 PM, Richard Brous wrote:
> >>
> >>> Great, glad we are on the same page!
> >>>
> >>> I went ahead and added the taxon id to the Mtuberculosis species
> profile, recompiled and initiated an export but the expected taxon id didn't
> show in the species drop down box.
> >>>
> >>> The coded modified is (as shown above):
> >>>
> >>>
> >>> public MycobacteriumTuberculosisUniProtSpeciesProfile() {
> >>>
> >>> super("Mycobacterium tuberculosis", 83332,
> >>>
> >>> "This profile customizes the GenMAPP Builder export for Mycobacterium
> Tuberculosis data loaded from a UniProt XML file.");
> >>> }
> >>>
> >>> Attached is a ss of the dialog box.
> >>>
> >>> I also checked to make sure the species was listed in the
> UniProtDatabaseProfile and Mtuberculosis and the rest of the newer species
> are present.
> >>>
> >>> Any idea where i should look next aside from assuring that the change
> is actually in my working build?
> >>>
> >>> Thanks!
> >>>
> >>> Richard
> >>>
> >>>
> >>>
> >>>
> >>> On Tue, Apr 12, 2011 at 3:40 PM, John David N. Dionisio <do...@lm...>
> wrote:
> >>>
> >>> Hi Rich,
> >>>
> >>> Thanks for the update. I agree that we can move past the merge stage
> now.
> >>>
> >>> Regarding the species constructor, yes, having the required change
> consist solely of the addition of the taxon ID really was the goal.
> "Separation of concerns" and "don't repeat yourself" and all that :) Note
> how, in the end, the designation of the taxon ID is truly the only piece of
> additional information that is required of species profiles in order to get
> the work done. So, this ideal is reflected in the code.
> >>>
> >>> Of course, there is a possibility that certain types of customizations
> may require additional work; however that depends on the type of
> customization involved and the way the species profile implements it. We
> can approach those on a per-species-profile basis.
> >>>
> >>> John David N. Dionisio, PhD
> >>> Associate Professor, Computer Science
> >>> Loyola Marymount University
> >>>
> >>>
> >>>
> >>>
> >>> On Apr 11, 2011, at 11:57 PM, Richard Brous wrote:
> >>>
> >>>> Hi all,
> >>>>
> >>>> I recreated a multispecies postgres db for P_aeruginosa and S_aureus
> and reimported the species xml files:
> >>>> P_aeruginosa UniProtXML
> >>>> P_aeruginosa goa
> >>>>
> >>>> S_aureus UniProtXML
> >>>> S_aureus goa
> >>>>
> >>>> GO xml
> >>>> processed go data
> >>>>
> >>>> I then exported P_aeruginosa .gdb using both Don's name2taxon code
> and the new trunk (b64): both gdb's original row counts are identical
> >>>> Also exported S_aureus using both Don's name2taxon code and the new
> trunk (b64) : both gdb's original row counts are identical
> >>>>
> >>>> I think we can safely say that Don's name2taxon changes are working at
> the point where he left off and the successful merge into the trunk is now
> verfied.
> >>>>
> >>>>
> =============================================================================
> >>>>
> >>>> Now to move to species I am familiar with and ensure they still match
> the base line gdb's generated from b63.
> >>>>
> >>>> The first species I will work on is Mtuberculosis. From the readme
> file of the gdb previously posted, the taxon id is 83332.
> >>>>
> >>>> I will go ahead and make the change to the Mtuberculosis species
> profile which is to add the taxon id in the form:
> >>>>
> >>>>
> >>>> public MycobacteriumTuberculosisUniProtSpeciesProfile() {
> >>>>
> >>>> super("Mycobacterium tuberculosis", 83332,
> >>>>
> >>>> "This profile customizes the GenMAPP Builder export for Mycobacterium
> Tuberculosis data loaded from a UniProt XML file.");
> >>>> }
> >>>>
> >>>> When comparing to S_aereus species profile, it was the only change to
> make in the Mtuberculosis species profile.
> >>>>
> >>>> I rechecked my notes from this morning and don't see any other changes
> needed to enable our newer species to work with the post merge constructor.
> >>>>
> >>>> I'll attempt an export tomorrow and see what happens.
> >>>>
> >>>> Richard
> >>>>
> >>>> On Fri, Apr 1, 2011 at 12:51 PM, John David N. Dionisio <
> do...@lm...> wrote:
> >>>> Follow the way the name2taxon “branch” was done. Put a gmbuilder/
> directory under tags/ if it is not already there, and make sure that the
> copied/tagged/branched subdirectory under that corresponds to the top-level
> directory of a gmbuilder project (i.e., it contains the build.xml, other
> top-level directories, etc.).
> >>>>
> >>>> Subversion copies are “cheap:” it is OK to copy deep trees of files.
> You only start using additional storage space if the files themselves get
> modified. Otherwise, they are just references to their original versions.
> >>>>
> >>>> John David N. Dionisio, PhD
> >>>> Associate Professor, Computer Science
> >>>> Loyola Marymount University
> >>>>
> >>>>
> >>>>
> >>>> On Apr 1, 2011, at 12:47 PM, Richard Brous wrote:
> >>>>
> >>>>> ok, so should I start at the gmbuilder directory Or the src
> directory?
> >>>>>
> >>>>> On Fri, Apr 1, 2011 at 11:42 AM, John David N. Dionisio <
> do...@lm...> wrote:
> >>>>> Yes, in Subversion what they did is collapse both activities into a
> single operation called "Copy." Whether you use the copy as a branch for
> modifications, or use it as a tag for posterity, is left to the developer
> and not the system.
> >>>>>
> >>>>> This functional change is the subject of some debate in the
> community. Some folks still feel that "enforced" tags have their benefits.
> But in the end, this is how Subversion conducts things, so we'll just have
> to go with it.
> >>>>>
> >>>>> John David N. Dionisio, PhD
> >>>>> Associate Professor, Computer Science
> >>>>> Loyola Marymount University
> >>>>>
> >>>>>
> >>>>> On Apr 1, 2011, at 12:38 PM, Richard Brous wrote:
> >>>>>
> >>>>>> ok, makes sense because when I went through the process (without
> clicking finish) there was really no difference between the two... not even
> a specification to say this this is a "tag" vs "branch" unless I can only do
> that in the comments.
> >>>>>>
> >>>>>> On Fri, Apr 1, 2011 at 11:35 AM, John David N. Dionisio <
> do...@lm...> wrote:
> >>>>>> Yes, tags and branches are structurally identical in Subversion.
> The only difference between a “tag” and a “branch” is that the developers
> have agreed to *not* touch a tag after it has been copied out.
> >>>>>>
> >>>>>> In other words, “tags” and “branches” are only different in
> Subversion by consensus. The distinction is not system-enforced.
> >>>>>>
> >>>>>> John David N. Dionisio, PhD
> >>>>>> Associate Professor, Computer Science
> >>>>>> Loyola Marymount University
> >>>>>>
> >>>>>>
> >>>>>>
> >>>>>> On Apr 1, 2011, at 12:32 PM, Richard Brous wrote:
> >>>>>>
> >>>>>>> tagging is really creating a new branch for posterity?
> >>>>>>>
> >>>>>>> On Fri, Apr 1, 2011 at 10:56 AM, John David N. Dionisio <
> do...@lm...> wrote:
> >>>>>>> You'll want to tag from the repository, since your local copy has
> been modified. If you issue the tag command from the SVN Repository Viewing
> perspective, that should happen server-side.
> >>>>>>>
> >>>>>>> If you want to play it really safe, you can even do the tagging
> from a completely separate Eclipse workspace that does not have your local
> copies loaded up.
> >>>>>>>
> >>>>>>> John David N. Dionisio, PhD
> >>>>>>> Associate Professor, Computer Science
> >>>>>>> Loyola Marymount University
> >>>>>>>
> >>>>>>>
> >>>>>>>
> >>>>>>> On Apr 1, 2011, at 11:54 AM, Richard Brous wrote:
> >>>>>>>
> >>>>>>>> Am I tagging in my local project or in the repository?
> >>>>>>>>
> >>>>>>>> On Fri, Apr 1, 2011 at 10:51 AM, John David N. Dionisio <
> do...@lm...> wrote:
> >>>>>>>> Hi Rich,
> >>>>>>>>
> >>>>>>>> I think it's OK to commit to the trunk, with one tweak for easy
> management: before committing, copy out ("tag") what is in the trunk right
> now, and put it in /tags/gmbuilder/before-name2taxon-merge (or something
> like that). This way, you have an easy basis for comparison if you need to
> go back to how the trunk was before name2taxon came along.
> >>>>>>>>
> >>>>>>>> Also, it will save some time if your before- and after- .gdbs are
> available. Running our own exports, while of course doable as you said,
> would require things like getting old versions, building them, etc.
> >>>>>>>>
> >>>>>>>> Happy digital forensics...
> >>>>>>>>
> >>>>>>>> John David N. Dionisio, PhD
> >>>>>>>> Associate Professor, Computer Science
> >>>>>>>> Loyola Marymount University
> >>>>>>>>
> >>>>>>>>
> >>>>>>>> On Apr 1, 2011, at 11:42 AM, Richard Brous wrote:
> >>>>>>>>
> >>>>>>>>> Hi all,
> >>>>>>>>>
> >>>>>>>>> Started a new thread since this is Pfalciparum specific.
> >>>>>>>>>
> >>>>>>>>> I have double checked my manual merge to confirm that in fact my
> edits were correctly included for Mtuberculosis and Pfalciparum.
> >>>>>>>>>
> >>>>>>>>> So when running gdb exports for each I determined that in fact
> something broke in the merge.
> >>>>>>>>>
> >>>>>>>>> Before I start digging into this further I want to ask a couple
> questions:
> >>>>>>>>>
> >>>>>>>>> 1. Should I commit the changes up to Sourceforge yet or leave
> them locally on my laptop while troubleshooting? I know we can revert to
> earlier versions if needed but wanted to be sure we were all on same page.
> >>>>>>>>>
> >>>>>>>>> 2. I have before and after exports of Mtuberculosis and
> Pfalciparum on my laptop. Would you like me to post them for review or will
> you be running your own anyway?
> >>>>>>>>>
> >>>>>>>>> Richard
> >>>>>>>>> <ATT00001..txt><ATT00002..txt>
> >>>>>>>>
> >>>>>>>>
> >>>>>>>>
> ------------------------------------------------------------------------------
> >>>>>>>> Create and publish websites with WebMatrix
> >>>>>>>> Use the most popular FREE web apps or write code yourself;
> >>>>>>>> WebMatrix provides all the features you need to develop and
> >>>>>>>> publish your website. http://p.sf.net/sfu/ms-webmatrix-sf
> >>>>>>>> _______________________________________________
> >>>>>>>> xmlpipedb-developer mailing list
> >>>>>>>> xml...@li...
> >>>>>>>> https://lists.sourceforge.net/lists/listinfo/xmlpipedb-developer
> >>>>>>>>
> >>>>>>>> <ATT00001..txt><ATT00002..txt>
> >>>>>>>
> >>>>>>>
> >>>>>>>
> ------------------------------------------------------------------------------
> >>>>>>> Create and publish websites with WebMatrix
> >>>>>>> Use the most popular FREE web apps or write code yourself;
> >>>>>>> WebMatrix provides all the features you need to develop and
> >>>>>>> publish your website. http://p.sf.net/sfu/ms-webmatrix-sf
> >>>>>>> _______________________________________________
> >>>>>>> xmlpipedb-developer mailing list
> >>>>>>> xml...@li...
> >>>>>>> https://lists.sourceforge.net/lists/listinfo/xmlpipedb-developer
> >>>>>>>
> >>>>>>> <ATT00001..txt><ATT00002..txt>
> >>>>>>
> >>>>>>
> >>>>>>
> ------------------------------------------------------------------------------
> >>>>>> Create and publish websites with WebMatrix
> >>>>>> Use the most popular FREE web apps or write code yourself;
> >>>>>> WebMatrix provides all the features you need to develop and
> >>>>>> publish your website. http://p.sf.net/sfu/ms-webmatrix-sf
> >>>>>> _______________________________________________
> >>>>>> xmlpipedb-developer mailing list
> >>>>>> xml...@li...
> >>>>>> https://lists.sourceforge.net/lists/listinfo/xmlpipedb-developer
> >>>>>>
> >>>>>> <ATT00001..txt><ATT00002..txt>
> >>>>>
> >>>>>
> >>>>>
> ------------------------------------------------------------------------------
> >>>>> Create and publish websites with WebMatrix
> >>>>> Use the most popular FREE web apps or write code yourself;
> >>>>> WebMatrix provides all the features you need to develop and
> >>>>> publish your website. http://p.sf.net/sfu/ms-webmatrix-sf
> >>>>> _______________________________________________
> >>>>> xmlpipedb-developer mailing list
> >>>>> xml...@li...
> >>>>> https://lists.sourceforge.net/lists/listinfo/xmlpipedb-developer
> >>>>>
> >>>>> <ATT00001..txt><ATT00002..txt>
> >>>>
> >>>>
> >>>>
> ------------------------------------------------------------------------------
> >>>> Create and publish websites with WebMatrix
> >>>> Use the most popular FREE web apps or write code yourself;
> >>>> WebMatrix provides all the features you need to develop and
> >>>> publish your website. http://p.sf.net/sfu/ms-webmatrix-sf
> >>>> _______________________________________________
> >>>> xmlpipedb-developer mailing list
> >>>> xml...@li...
> >>>> https://lists.sourceforge.net/lists/listinfo/xmlpipedb-developer
> >>>>
> >>>
> >>>> <ATT00001..txt><ATT00002..txt>
> >>>
> >>>
> >>>
> ------------------------------------------------------------------------------
> >>> Forrester Wave Report - Recovery time is now measured in hours and
> minutes
> >>> not days. Key insights are discussed in the 2010 Forrester Wave Report
> as
> >>> part of an in-depth evaluation of disaster recovery service providers.
> >>> Forrester found the best-in-class provider in terms of services and
> vision.
> >>> Read this report now! http://p.sf.net/sfu/ibm-webcastpromo
> >>>
> >>> _______________________________________________
> >>> xmlpipedb-developer mailing list
> >>> xml...@li...
> >>> https://lists.sourceforge.net/lists/listinfo/xmlpipedb-developer
> >>>
> >>>
> >>
> >>> <Tb_export_screen.jpg><ATT00001..txt><ATT00002..txt>
> >>
> >>
> >>
> ------------------------------------------------------------------------------
> >> Forrester Wave Report - Recovery time is now measured in hours and
> minutes
> >> not days. Key insights are discussed in the 2010 Forrester Wave Report
> as
> >> part of an in-depth evaluation of disaster recovery service providers.
> >> Forrester found the best-in-class provider in terms of services and
> vision.
> >> Read this report now! http://p.sf.net/sfu/ibm-webcastpromo
> >> _______________________________________________
> >> xmlpipedb-developer mailing list
> >> xml...@li...
> >> https://lists.sourceforge.net/lists/listinfo/xmlpipedb-developer
> >>
> >>
> >>
> >> <ATT00001..txt><ATT00002..txt>
> >
> >
> >
> ------------------------------------------------------------------------------
> > Forrester Wave Report - Recovery time is now measured in hours and
> minutes
> > not days. Key insights are discussed in the 2010 Forrester Wave Report as
> > part of an in-depth evaluation of disaster recovery service providers.
> > Forrester found the best-in-class provider in terms of services and
> vision.
> > Read this report now! http://p.sf.net/sfu/ibm-webcastpromo
> > _______________________________________________
> > xmlpipedb-developer mailing list
> > xml...@li...
> > https://lists.sourceforge.net/lists/listinfo/xmlpipedb-developer
>
|
