GitLab

BUGFIX: formatted not updated in set_rawdata

See merge request !11

When the rawdata were set in MyFig::set_rawdata I forgot to add self.formatted = False before self.formatRawData and therefore if self.reformat was set to false the data were not reformatted... also add some debug info.

corrected the BUG Figure.__init__ was complaining from user arguments

See merge request !10

I now separates between MyFig and MyFig_base. MyFig_base can be used by overwrit…

See merge request !9

I now separates between MyFig and MyFig_base. MyFig_base can be used by overwritting the __init__ function and the plot function, very much like in v1.0.0, you just don't need update, remark the axes should be created in __init__ not in plot. The MyFig class however is smarter, and the user just need to overwrite defineKeywords and addAxes, yes, you don't need __init__! as their names suggests in defineKeywords you should update self.keywords, in addAxes you should add the objects MyAxes. Now the figure object reads a tuple of data object as argument (not optional anymore) each index for the axes. This behaviour can be overwritten in the function MyFig::formatRawdata, and MyFig::set_rawdata. Remark now the MyFig_base keywords became attributes: self.keywords['fignum'] -> self.fignum...

Some testing for myguilib

See merge request !8

mypltlib is now restructured with tests in test. I also changed MyFig and MyAxes such that they can not be used alone. This is done such that now as seen in test figTest1 a figure consists only of an __init__ function. Nothing else is required. I also changed the refresh function of MyWins such that the figure does not destroy it axis but redraw them. So now the axes are create only ones. This has the side effect that when the window is resized the axes ratio is not respected anymore.

correction of the import bug for Tk

import TK dependent on mpl version

See merge request !6

Dev -> undertest: v2.0.0a

See merge request !5

while merging I recognised that I could avoid the issue in replacing the canvas of the figure while MyFig::print2file by storing it in a local value now called winCanvas. Therefore the rebounding in MyWin::refresh is not necessary anymore. I add a warning message incase such situation happend. This should not be!

Verify print2file

See merge request !4

Tested WXAgg and GTKAgg, both worked! Modified myFig.print2file to take into account the case when the user put the path into the filename. Make the test for these feature (test306).

add print2file in MyFig class, tested it (test302, 303, 304, 305), and verfy that it works in combinasion with MyWin see test403. modified print2file in myTool (untested) and only TkAgg has been tested but WXAgg and GTKAgg are also implemented.

Object oriented to testing phase

See merge request !2

Rewrote myWin as three different classes depending on the backend changed the imports in __init__ to make it cleaner.

This version is object oriented, the interactive mode is now ready and seems to work. It requires more testing though. In anycase tests have to be added. myTool may also be removed, and another class MyCanvasToPrint may be created instead of the print2file function. MyWin.py should be commented and a header is needed, as well as a correction of the other headers. remarks: the default value of fignum may be removed and set to -1 hard coded (indeed another number may lead to truble in interactive mode)

this version is clean. no more bugs and should allow full potential. The remaining problem is that it is not object oriented. should not be a big deal.