使用Python解析不同的数据类型

Question

I want to parse the following line:

#3 = IFCPERSONANDORGANIZATION(#4, #5, $);

and want to extract the numbers: 3, 4 , 5 to integer values and 'IFCPERSONANDORGANIZATION' as string value, in order to save these attributes in a graph with networkx.

This my code:

data = []
with open('test.ifc') as f:
    for line in f:
        if line.startswith('#'):
            words = line.rstrip().split('#')
            print(words)
            node = int(words[0])
            data.append(node)

Error: ValueError: invalid literal for int() with base 10: ''

How can I use regex, if the line structure is different every time? Like this:

#3 = IFCPERSONANDORGANIZATION(#4, #5, $);
#2 = IFCOWNERHISTORY(#3, #6, $, .NOTDEFINED., $, $, $, 1348486883);
#4 = IFCPERSON($, 'Bonsma', 'Peter', $, $, $, $, $);
#5 = IFCORGANIZATION($, 'RDF', 'RDF Ltd.', $, $);
#6 = IFCAPPLICATION(#5, '0.10', 'Test Application', 'TA 1001');****

Answer 1

You could use regex:

import re
line = '#3 = IFCPERSONANDORGANIZATION(#4, #5, $);'
node, name, a, b = re.search(r'(\d+) = (\w+)\(#(\d+), #(\d+), \$\)', line).groups()
node, a, b = map(int, [node, a, b])
print(node, name, a, b)

prints

3 IFCPERSONANDORGANIZATION 4 5

Answer 2

May be a late comment but I come up with your question and given answers when I am making a similar search. @user3926906 an IFC file structure is generally changing for each different file. When you are using re.search() did you experience any challenge to split the # of entities? I am asking because some of the entities does not # for referring other entities. Thanks