Scripting

Apart from the command line interface all routines implemented in the fxs toolkit can also be accessed interactively.

In fact executing the command

$ xframe fxs reconstruct tutorial

is equivalent to the following python code

import xframe
xframe.select_project('fxs','reconstruct','tutorial')
xframe.run()

Calling xframe.select_project('fxs','reconstruct','tutorial') will load the settings file and setup the database module. This allows you to change settings and look at data that is used by fxs in a convenient way.

Modify Settings

As an example lets take another look at the tutorial settings file for the correlate worker.

structure_name: 'tutorial'

image_dimensions: [512,512]

n_processes: 32

max_n_patterns: 2000


compute: ['is_good','waxs_aver','ccf_q1q2']
pixel_size: 200.0                 #Detector pixel size in microns
sample_distance: 800.0            #Detector to sample distance in [mm]
wavelength: 1.23984               #X-ray wavelength in Å
detector_origin: [255.2,255.5]    #2D detector origin in pixel coordinates

interpolation_order: 2            #Cartesian to Polar spline interpolation order

phi_range:                        #Azimutal angle grid via (start,stop,npoints,???)
    command: '(0.0, 2*np.pi, 1024, "exact")'
fc_n_max: 70                      #maximal resolved harmonic order


IO:
  folders:
    home:
      command: 'xframe_opt.IO.folders.home'
    in_base:
      home: 'data/fxs/input_files/'
  files:
    input_file_list:
      name: patterns_list.txt
      folder: in_base

In python you have access to these settings via the xframe.settings.project attribute. All settings files in xFrame support . notation to access its attributes, as well as most of the features of a python dictionary, here is an example.

import xframe
xframe.setup_project('fxs',correlate','tutorial')

opt = xframe.settings.project

# Looking up properties in the settings file
n_processes_1 = opt.n_processes
n_processes_2 = opt['n_processes']
non_existing = opt.get('non_existing_settings',False)

#Changing settings on the fly
# via . notation
opt.n_processes = 10
# or like you would use a dictionary
opt['n_processes'] = 10

# Running the script with modified settings
xframe.run()

More over the xfrmae.settings.project contains the parsed version of the sittings file, that is all command: lines are replace with the output of eval() applied to their string and all default settings have been included.

The settings file to be saved alongside the routine results will correspond to the state of xframe.settings.project at the moment xframe.run() is called.

Accessing project files

You can also us the fxs database module stored in xframe.database.project after the call to select_project. This allows you easy access to all files defined in the IO: section of your settings file.

If, for example, you want to look at the patterns_list.txt file from the above settings you can use its name input_file_list to get access to it, e.g.

import xframe
xframe.setup_project('fxs','correlate','tutorial')
db = xframe.database.project

# loading the patterns_list.txt 
file_list = db.load('input_file_list')

# Overriding the patterns_list.txt
db.save('input_file_list',file_list)

In fact, all file access in fxs's routines is managed via the entries in the IO: portion of the settings files, they are not explicitly listed in the tutorial.yaml of the getting started page files to keep the tutorials simple.

More information on xFrame`s handling of settings and data access can be found under Settings