explore_liborigins_object.py

#!/usr/bin/python3
#-*- coding: utf-8 -*-
""" 
Invocation: 
    explore_liborigins_object.py file.opj
    
Results in recursive printout of all the nested structure of the object provided by liborigin. 
Code can be modified and reused for other complex objects. 
"""

from pathlib import Path
import liborigin
import sys
style = 'abbrev' #$False
if style == 'color':
    normal = r"\034[1;0m"
    bold   = r"\033[1;1m"
    grey   = r"\033[2;2m"
else:
    normal, bold  , grey   = '', '', ''
maxlistlen = 200
allstrings = []

branchdict = {'│':'├', '║':'╟', '┃':'┣'}


def save_sheetdata(data, head, out_dir_name):
    outdir = Path(out_dir_name)
    outdir.mkdir(exist_ok = True)
    #print(lineindent,f"SAVING INTO {filename}/{sheetname.decode('utf8')}.dat: {len(sheetdata)} columns") #TODO
    with open(outdir / (sheetname + '.dat'), 'w') as outfile:
        outfile.write('#' + '\t'.join(sheethead) + '\n')
        for vals in zip(*sheetdata):
            if isinstance(vals[0], float) and vals[0] != 0 and abs(vals[0])<1e-100: continue  # origin encodes NaNs as *very* small numbers
            outfile.write('\t'.join(str(val) for val in vals)+'\n')

def explore(obj, parentstr, indent):
    global sheetdata
    global sheethead
    global sheetname 
    enc = 'cp1250' ## not specified by liborigin, set manually!

    lineindent = indent[:-1]+ branchdict.get(indent[-1:],'') + parentstr

    if style == 'color': indent = indent + grey + parentstr + normal ## prints ancestry objects with grey color for better clarity (terminal only)
    elif style == 'abbrev':        indent = indent + ' '*(len(parentstr)+1) ## replaces ancestry objects with whitespace for clarity
    else: indent = indent +  parentstr  ## prints ancestry objects with grey color for better clarity (terminal only)
    if len(indent)>200: 
        print(lineindent,'                          --------------- nesting too deep - halted here ----------------')
        return

    if isinstance(obj, list):
        print(lineindent,bold+'[', normal)
        for n, item in enumerate(obj[:maxlistlen]):
            #print(lineindent,f'n {n} item {item}')
            explore(item, 'item #%d' % n,  indent[:-1]+'║')
        if len(obj) > maxlistlen:
            print(lineindent+bold+'--- listing truncated here, %d values would follow ---' % (len(obj)-maxlistlen))
        print(lineindent, bold+']', normal)
    elif isinstance(obj, dict):
        print(lineindent,bold+'{', normal)
        for key, value in obj.items():
            #print(lineindent,f'K {key} V {value}')
            explore(value, '["'+key+'"]', indent[:-1]+'┃')
        print(lineindent,bold+'}', normal)
    elif isinstance(obj, str):
        print(lineindent,bold+'"'+obj+'"', normal)
        allstrings.append(obj)
    elif isinstance(obj, bytes):
        print(lineindent,bold+' b"'+obj.decode(enc).strip()+'"')
    elif type(obj) in (float, int, bool):
        print(lineindent,bold,obj, normal)
    else:
        print(lineindent,bold, obj, grey, type(obj), normal)
        
        #print(lineindent,' '*(len(indent)-len(parentstr)), grey, usefulattrs, normal)
        if isinstance(obj, liborigin.SpreadSheet):
            if sheetdata: save_sheetdata(sheetdata, sheethead, out_dir_name)
            sheetdata = []
            sheethead = []
            try:
                sheetname = obj.label.decode(enc).split('@${[')[0]
            except AttributeError:
                sheetname = obj.name.decode(enc)

            #if getattr(obj, 'label', None):
                #print(f"---- sheet named {sheetname} has label {getattr(obj, 'label', b'').decode('cp1252').split('@${[')[0]}")
        if isinstance(obj, liborigin.SpreadColumn):
            if getattr(obj, 'type', None) != 6:
                sheetdata.append(obj.data)
                sheethead.append(obj.name.decode(enc))
        usefulattrs = [attrname for attrname in dir(obj) if not attrname.startswith('__')]
        for attrname in usefulattrs:
            explore(getattr(obj, attrname), '.'+attrname, indent[:-1]+'│')


for opj_name in sys.argv[1:]:
    print('PROCESSING', opj_name)
    try:
        opj = liborigin.parseOriginFile(opj_name)
    except IndexError:
        print(lineindent,'Please specify the origin file'); quit()

    sheetdata, sheethead, sheetname = [], [], None

    out_dir_name=opj_name.replace(' ','_').replace('.opj','_exported')
    explore(opj, '', '')
    if sheetdata: save_sheetdata(sheetdata, sheethead, out_dir_name)
    

An example of how the printout looks like. This helps to dig through all the nested "tables", "graphs", "layers" etc. Still quite an involved task, though, as the format differs across OPJ versions.