[Rp] Gifa V.4: A complete package for NMR data set processing

[delsuc]

Original article: [Rp] Gifa V.4: A complete package for NMR data set processing

PDF URL: https://github.com/delsuc/Gifa/blob/master/article/article.pdf
Metadata URL: https://github.com/delsuc/Gifa/blob/master/article/metadata.yaml
Code URL: https://github.com/delsuc/Gifa/blob/master/code

Scientific domain: NMR, data processing
Programming language: Fortran77, C, Gifa macro
Suggested editor: Pierre de Buyl, Konrad Hinsen, Thomas Arildsen, Georgios Detorakis

[rougier]

@pdebuyl @khinsen Can you serve as editor for this submission for the Ten Years Reproducibility Challenge ?

[rougier]

@delsuc Thanks for your submission, we'll assign an editor soon.

[pdebuyl]

I can edit. @khinsen can you review?

[khinsen]

@pdebuyl Yes, I will do a review.

[khinsen]

@delsuc Could you add instructions for compiling, installing, and running the tests? The code archive looks like this should be straightforward, but it isn't obvious how to start. Also, the paper suggests that running the code requires a 32-bit system. That should also be stated in the README so that people don't waste time trying anything else.

[delsuc]

You're right, I should find some time today to do this.

[delsuc]

Thanks @khinsen , that was needed, and I should have done it in the first hand.
I added a part in the README which shows how to build the binary, and reproduce the figures in the manuscript.

[khinsen]

Thanks @delsuc, that helps a lot!

Following the instructions, I almost managed to compile Gifa. I had to install the package libxext-dev in order to succeed. Then I could run the tests.

I haven't been able so far to reproduce the figures. Here's what I do:

1. cd Gifa/article
2. gifa (this opens a few windows)
3. fig2a.g (this open a dialog box entitled "Display control")
4. Push the OK button.

I get the promised error message about dispcont_doit, but no figure. All windows remain blank.

Next:

1. cd Gifa/article
2. gifa (this opens a few windows)
3. figure3.g (this open a dialog box entitled "figure_3", asking me to close the 'gifa' bar and then the dialog)
4. I close the 'gifa' bar (by closing the window)
5. I close the dialog via the OK button.
6. Error message in the terminal: 'No such file or directory. Aborting execution of command file : figure3.g At line : 11 Error in processing command : READ'

It looks like the macro loads a data file under Gifa/data, so I tried again from my home directory (which contains Gifa), and then I do get figure 3.

BTW, it would be useful to tell adventurous explorers that "exit" is the command to get out of Gifa!

[delsuc]

Thank you @khinsen for pointing out these errors !

• install libxext-dev - of course - I just forgot to copy it in the README.

• fig2a.g is really badly named ! it actually contains the listing in Figure 2a of the original paper, and reproduces Figure 2c there, which is Figure 1 from the Rescience work (sorry for the strange numbering!) - I considered that there is no need to reproduce Figure 2a and 2b from the original as these are just plain listings.
  So the appearance of the dialog box is the whole purpose - I realise this is a bit strange, so I added additional resources so that the correct behaviour is reproduced as well.

• the README has been extended with basic Gifa usage - and how to reproduce the figures.