Python Dataclass: Why can't a method use another instance of the Dataclass as a parameter

Question

I have a dataclass called ReportParts. I wrote an add() function that mashes another ReportParts into the the first ReportParts. It works perfectly.

@dataclass
class ReportParts():
    headers: []
    columns: [[]]

    def add(self, newParts):
        self.headers.extend(newParts.headers)
        self.columns.extend(newParts.columns)

petNames = reportParts(['cat','dog'], [['Gato','Sabrina','Meow'],['Spot','Cerberus','Argos']])

bunnyNames = reportParts(['bunny'], [['Hopper','Flopsy','Snowball']])

petNames.add(bunnyNames)

#Pretty Print
for animal, names in zip(petNames.headers, petNames.columns):
        print (animal + ': ' + str(names))


>>>cat: ['Gato', 'Sabrina', 'Meow']
>>>dog: ['Spot', 'Cerberus', 'Argos']
>>>bunny: ['Hopper', 'Flopsy', 'Snowball']

Why is it that if I specify that the parameter for add() is another ReportParts, Python says it doesn't know what a ReportParts is?

@dataclass
class ReportParts():
    headers: []
    columns: [[]]

    def add(self, newParts: ReportParts): #<--- newParts is now specified as a ReportParts
        self.headers.extend(newParts.headers)
        self.columns.extend(newParts.columns)

>>>Exception has occurred: NameError
>>>name 'reportParts' is not defined
>>>  File "/Users/test/GitHub/WildSort/Sandbox.py", line 11, in reportParts
>>>    def add(self, newParts:reportParts):
                           ^^^^^^^^^^^

Also, more subjective I know, but is there a good way around this? Should I be checking for type(newParts) != ReportParts and raising a TypeError? Should I just leave it be?