use python re lib parsing "user define xpath" scripts

Question

I'm building a "xpath parsing tool" in Python for my team. In my case, the xpath script is not the normal xpath, the syntax that user input will be in a special struct, here is an example:

The input format will be like: (the element can be tuple-type or normal element)

sig = "(xpath_1_1, xpath_1_2), (xpath_2_1, xpath_2_2), xpath_3..."

which is edited in excel by users

And my goal is to parse the string into a list-type data with tuple or normal element:

[(xpath_1_1, xpath_1_2), (xpath_2_1, xpath_2_2), xpath_3...]

Then I can input this data into my selenium to snapshot img sequentially.

Here is one of my testing data:

sig = "(//div[@style='font-family:Arial;float: left;width:930px;font-size:12px;' and ./span[contains(text(),'005930')]], //table[@id='gv_flow_krKS0 1']),//table[@id='123456'],(//div[@style='font-family:Arial;float: left;width:930px;font-size:12px;' and ./span[contains(text(),'000660')]],  //table[@id='gv_flow_krKS0 2']),//table[@id='456789']"

I'm wondering is there any better way to implement this func without disrupting the order of list ?

First , I think eval() func is not a good idea since it may cause some security prob.

Now I'm trying to use re lib to solve it.

However I found it's quite difficult and have no idea how to start.

Anyone can help ? Thanks~

Answer 1

OK, I think this does what you want. You should try some different test strings.

sig = "(//div[@style='font-family:Arial;float: left;width:930px;font-size:12px;' and ./span[contains(text(),'005930')]], //table[@id='gv_flow_krKS0 1']),//table[@id='123456'],(//div[@style='font-family:Arial;float: left;width:930px;font-size:12px;' and ./span[contains(text(),'000660')]],  //table[@id='gv_flow_krKS0 2']),//table[@id='456789']"


gather = ''
element = []
elements = []
state = ''
for c in sig:
    if state:
        gather += c
        if c == state:
            state = ''
        continue

    if c == '(':
        in_tuple = True
        continue
    elif c == ')':
        in_tuple = False
        element.append( gather )
        gather = ''
        elements.append(tuple(element))
        element = []
        continue
    elif c == ',':
        if in_tuple:
            element.append( gather )
        else:
            elements.append( gather )
        gather = ''
        continue
    elif c == '[':
        state = ']'
    elif c == "'":
        state = "'"

    gather += c

# Handle leftover.
if element:
    elements.append( element )

for e in elements:
    print( e )

Output:

("//div[@style='font-family:Arial;float: left;width:930px;font-size:12px;' and ./span[contains(text(),'005930')]]", " //table[@id='gv_flow_krKS0 1']")
//table[@id='123456']
("//div[@style='font-family:Arial;float: left;width:930px;font-size:12px;' and ./span[contains(text(),'000660')]]", "  //table[@id='gv_flow_krKS0 2']")