制表符格式的嵌套字符串到嵌套列表〜Python

Question

Hello guys, after managing to get some data by scraping with Beautiful Soup... I want to format that data so as I could easily export it to CSV and JSON.

My Question here is how can one translate this :

Heading :
    Subheading :

AnotherHeading : 
    AnotherSubheading :
        Somedata

Heading :
    Subheading :

AnotherHeading : 
    AnotherSubheading :
        Somedata

Into this :

[
['Heading',['Subheading']],
['AnotherHeading',['AnotherSubheading',['Somedata']]],
['Heading',['Subheading']],
['AnotherHeading',['AnotherSubheading',['Somedata']]]
]

Indented for clarity

Any rescue attempt would be appreciated by a warm thank you !

So far with help we got:

def parse(data):
  stack = [[]]
  levels = [0]
  current = stack[0]
  for line in data.splitlines():
    indent = len(line)-len(line.lstrip())
    if indent > levels[-1]:
      levels.append(indent)
      stack.append([])
      current.append(stack[-1])
      current = stack[-1]
    elif indent < levels[-1]:
      stack.pop()
      current = stack[-1]
      levels.pop()
    current.append(line.strip().rstrip(':'))
  return stack

The problem with that code is that it returns...

[
'Heading ', 
['Subheading '], 
'AnotherHeading ', 
['AnotherSubheading ', ['Somedata'], 'Heading ', 'Subheading '], 'AnotherHeading ', 
['AnotherSubheading ', ['Somedata']]
]

Here is a repl: https://repl.it/yvM/1

Answer 1

Thank you both kirbyfan64sos and SuperBiasedMan

def parse(data):

  currentTab = 0
  currentList = []
  result = [currentList]

  i = 0
  tabCount = 0

  for line in data.splitlines():

    tabCount = len(line)-len(line.lstrip())

    line = line.strip().rstrip(' :')

    if tabCount == currentTab:
        currentList.append(line)

    elif tabCount > currentTab:
        newList = [line]
        currentList.append(newList)
        currentList = newList

    elif tabCount == 0:
        currentList = [line]
        result.append(currentList)

    elif tabCount == 1:
        currentList = [line]
        result[-1].append(currentList)

    currentTab = tabCount

    tabCount = tabCount + 1
    i = i + 1

  print(result)

Answer 2

Well first you want to clear out unnecessary whitespace, so you make a list of all the lines that contain something more than whitespace and set up all the defaults that you start from for the main loop.

teststring = [line for line in teststring.split('\n') if line.strip()]
currentTab = 0
currentList = []
result = [currentList]

This method replies on the mutability of lists, so setting currentList as an empty list and then setting result to [currentList] is an important step, since we can now append to currentList .

for line in teststring:
    i, tabCount = 0, 0

    while line[i] == ' ':
        tabCount += 1
        i += 1
    tabCount /= 8

This is the best way I could think of to check for tab characters at the start of each line. Also, yes you'll notice I actually checked for spaces, not tabs. Tabs just 100% didn't work, I think it was because I was using repl.it since I don't have Python 3 installed. It works perfectly fine on Python 2.7 but I wont put code I haven't verified works. I can edit this if you confirm that using \\t and removing tabCount /= 8 produces the desired results.

Now, check how indented the line is. If it's the same as our currentTab value, then just append to the currentList .

    if tabCount == currentTab:
        currentList.append(line.strip())

If it's higher, that means we've gone to a deeper list level. We need a new list nested in currentList .

    elif tabCount > currentTab:
        newList = [line.strip()]
        currentList.append(newList)
        currentList = newList

Going backwards is trickier, since the data only contains 3 nesting levels I opted to hardcode what to do with the values 0 and 1 (2 should always result in one of the above blocks). If there are no tabs, we can append a new list to result .

    elif tabCount == 0:
        currentList = [line.strip()]
        result.append(currentList)

It's mostly the same for a one tab deep heading, except that you should append to result[-1] , as that's the last main heading to nest into.

    elif tabCount == 1:
        currentList = [line.strip()]
        result[-1].append(currentList)

Lastly, make sure currentTab is updated to what our current tabCount is so the next iteration behaves properly.

    currentTab = tabCount

Answer 3

Something like:

def parse(data):
    stack = [[]]
    levels = [0]
    current = stack[0]
    for line in data.splitlines():
        indent = len(line)-len(line.lstrip())
        if indent > levels[-1]:
            levels.append(indent)
            stack.append([])
            current.append(stack[-1])
            current = stack[-1]
        elif indent < levels[-1]:
            stack.pop()
            current = stack[-1]
            levels.pop()
        current.append(line.strip().rstrip(':'))
    return stack[0]

Your format looks a lot like YAML, though; you may want to look into PyYAML.