Please backport grace (1:5.1.22-8build1) from Maverick to Lucid

New patch with support for double peak fitting. Builds and compiles OK. Test binaries:

https://launchpad.net/~feranick/+archive/ppa/+packages

Just to clarify: the patches are for grace 5.1.22-3 in lucid.

Hi Feranick,

Great work... please could you create a debdiff between the Debian package (5.1.22-3) and your new version (5.1.22-3ubuntu1) and attach it? Once you have done that, please subscribe ubuntu-universe-sponsors to the package so that someone can review the patch and add it to the repositories.

Also, it's probably a good idea to forward this upstream even though Evgeny's upstream development is currently stalled. If/when he or someone else resumes the development, it would be helpful to include all the downstream changes there.

Forwarded upstream: http://plasma-gate.weizmann.ac.il/Grace/phpbb/w3todo.php?action=view_report&project_id=1&todo_id=2220

Alex,

I will do as indicated. I am working on an improved patch. As soon as I am done with it I'll proceed with the subscription.

Thanks,
Nicola Ferralis (aka feranick)

Attached updated patch and debdiff. Subscribed ubuntu-universe-sponsors.

I significantly improved the current patch with the following:

1. For single Gaussian and Lorentzian fitting, now upon selection of the proper function, the user can use the mouse to point to the maximum of the peak (as indicated also in the footer of xmgrace window). This sets the initial values for the peak maximum and intensity. With these values, the fit becomes straightforward.
2. I added an short explanation in the user guide.

Attached are the new patch with the debdiff.

I significantly extended the library of non-linear fitting function, so that it includes:

Gaussian (single, double, chromatography), Lorentzian (single double), Pseudo Voigt 1, Pseudo Voigt 2, Asymmetric double sigmoidal, Log Normal Function, Gram-Charlier A-Series, Edgeworth-Cramer Series, Inverse Polynomial Function.

It also includes, Periodic Peak functions (Sine, Sine square, Sine damp), and Baseline Functions (Exponential Decay 1, Exponential Decay 2, Exponential Growth 1, Exponential Growth 2, Hyperbolic, Bradley, Logarithm 3 parameters).

The new patch and the debdiff are attached.

New binaries for testing available here:

https://launchpad.net/~feranick/+archive/ppa/+packages?field.name_filter=grace&field.status_filter=published

Sorry for the multiple attachment, please consider only the latest. I added support for triple Lorentzian and Gaussians.

I am done with the patch, no more modifications for now.

I created a new ppa for testing binaries:

https://launchpad.net/~grace-extended/+archive/ppa

Thanks Nicola, I have untagged the old patches to help avoid confusion. I'll give the ppa version a try today if I get a chance.

Any news about this?

- New debdiff against new upstream release (5.1.22-5)
- Fixed typo in patch.

Test builds available in grace-extended ppa:

https://launchpad.net/~grace-extended/+archive/ppa/+packages

Updated patch. Now user input (via mouse coordinates) for initial peak position/amplitude is available for single, double and triple Lorentzian and Gaussian. Previously it was available only for single peak fitting.

Test builds available in grace-extended ppa:

https://launchpad.net/~grace-extended/+archive/ppa/+packages

I re-uploaded the patch and debdiff, since the old ones were the wrong ones...

Sorry for the inconvenience.

I attach a screenshot highlighting the new "Library" entry in the nonlinear fitting panel, and the initialization parameters (in the console) used for a three peak Lorentzian fit.

Hi Nicola,

I finally got round to testing the new function library, and it looks great. I think the user guide should mention the mouse input for Lorentzian fitting too. It might even be a good idea to print instructions to the console window which pops up.

Thanks for the feedback, Alex, and please, keep it coming!

I implemented the suggested changes, specifically I added a brief note in the User guide about the user input, and instructions for user input are now printed both in the console and in the footer of the window.

Test builds available soon in the ppa.

I added another useful fitting function to the patch (asymmetric Doniach-Sunjic), which is used in fitting photoemission spectra. I do include a debdiff, as this patch will be soon available in Debian.

This bug was fixed in the package grace - 1:5.1.22-6

---------------
grace (1:5.1.22-6) unstable; urgency=low

  * rules: Run sensible-editor inside x-terminal-emulator.
    (Closes: #582154) (LP: #582399)
  * New patches:
    - non-resizable-dialogs.patch: Set certain dialog boxes to fixed sizes,
      preventing display errors and possible concealment of action buttons.
      (Closes: #253087)
    - superscript-subscript-scale.patch: Adjust {sub,super}script scaling to
      bring them more in line with LaTeX's style. (Closes: #279359)
    - nonlinfit_lorentzian_gaussian_2.diff: Add support for non-linear
      Lorentzian and Gaussian fitting. Written and contributed by Nicola
      Ferralis. (Closes: #578435) (LP: #535459)
  * control: Add Build-Depends on texlive-latex-{base,recommended} for
    regeneration of DVI documentation.
 -- Ubuntu Archive Auto-Sync <email address hidden> Fri, 28 May 2010 11:22:55 +0100

Adding request to port this patch to grace 5.1.22-5 in lucid. The capabilities provided by this patch are significant since easy non-linear fitting is sorely missing. The patch has been significantly tested and it is included in more recent version of grace (it's included in version 5.1.22-6 and beyond). No side effects or potential regressions are found.

 The patch (attached) has been tested and found to work on 5.1.22-5. I also attach a debdiff for version 5.1.22-5ubuntu1.

grace (1:5.1.22-5ubuntu1) lucid; urgency=low

  * Add support for non-linear fitting.
    - Written and contributed by Nicola Ferralis. (LP: #535459)

 -- Nicola Ferralis <email address hidden> Tue, 23 Nov 2010 15:04:47 -0800

Requesting a "nomination for lucid".

Hi Nicola, and thank you for taking the time to report this bug and helping to make Ubuntu better.

Just two points:
 - The issue you are reporting is an upstream one and it would be nice if somebody having it could send the bug to the developers of the software. If you have done so, please tell us the number of the upstream bug (or the link), so we can add a bugwatch that will inform us about its status.
 - Would you mind to change the target release in your debdiff from 'lucid' to 'lucid-proposed'? You may find useful information information about Ubuntu's SRU here: https://wiki.ubuntu.com/StableReleaseUpdates#Procedure

Thanks in advance.

Unsubscribing sponsors team for now, please re-subscribe when ready.

@Alessio - This patch is really to add a new feature, rather than a bug fix. It was forwarded upstream some months ago at http://plasma-gate.weizmann.ac.il/Grace/phpbb/viewtopic.php?f=6&t=1898. Unfortunately, upstream development of Grace has stalled, and the patch is is not likely to be taken upstream in the foreseeable future.

Thanks, Alessio.

1. As Alex mentioned, the bug has been sitting upstream for a long while with no action (see also comment #7 above).

2. Done, see below. Re-subscribing sponsors-team.

grace (1:5.1.22-5ubuntu1) lucid-proposed; urgency=low

  * Add support for non-linear fitting.
    - Written and contributed by Nicola Ferralis. (LP: #535459)

 -- Nicola Ferralis <email address hidden> Wed, 24 Nov 2010 17:29:03 -0800

Hi Nicola,

I just noticed that the latest patch also adds the Weibull PDF, so it also closes bug #608789. Probably worth mentioning this in debian/changelog?

Hi Alex, done.

grace (1:5.1.22-5ubuntu1) lucid-proposed; urgency=low

  * Add support for non-linear fitting.
    - Written and contributed by Nicola Ferralis. (LP: #535459, #608789)

 -- Nicola Ferralis <email address hidden> Thu, 25 Nov 2010 14:09:02 -0800

I don't think that this patch is appropriate for an SRU. SRU's are supposed to fix "high-impact bugs", such as "severe regressions" and "loss of user data", and are specifically not supposed to be used as a mechanism to provide new features. (See https://wiki.ubuntu.com/StableReleaseUpdates#When)

Backports, on the other hand, are for including "significant new features", so I think that backporting grace from Natty to Lucid (and Maverick) is pretty clearly the way to go here.

I'm going to turn this into a backports bug instead of an SRU bug.

Oh, whoops - this got fixed in Maverick, so we only need to backport to Lucid.