This validator checks the amount of text in a RichTextField, by stripping away basic formatting markup. It also incorporates an adjustment for markup that causes line breaks.

models.py

from wagtail.core.models import Page
from wagtail.core.fields import RichTextField

from proj.richtext_validator import RichTextLengthValidator

class AuthorPage(Page):

    biography = RichTextField(
        max_length=3600,
        validators=[RichTextLengthValidator(3000, line_length=50)],
        features=['bold', 'italic', 'link'],
        blank=False,
        null=True,
        help_text="Full rich-text Biography"
    )

richtext_validator.py

from django.core.exceptions import ValidationError
from django.utils.deconstruct import deconstructible

from bs4 import BeautifulSoup

@deconstructible
class RichTextLengthValidator:
    """ImageField dimensions validator."""
    
    def __init__(self, max_length, line_length, tag_weights=None):
        """
        Constructor
        Args:
            max_length (int): maximum number of renders chars
            line_length (int): nominal number of characters in a line
        """
        self.max_length = max_length
        self.line_length = line_length
        if tag_weights:
            self.tag_weights = tag_weights
        else:
            self.tag_weights = {
                'li': line_length // 2,
                'p': line_length // 2,
                'br': line_length // 2,
            }

    def __call__(self, value):
        bs_value = BeautifulSoup(value, 'html5lib')

        length = 0
        for string in bs_value.strings:
            length += len(string)

        tag_length = 0
        for tag in bs_value.find_all():
            if tag.name in self.tag_weights:
                tag_length += self.tag_weights[tag.name]

        total_weighted_length = length + tag_length
        if total_weighted_length > self.max_length:
            raise ValidationError(
                "Ensure this richtext value is shorter than {} character equivalents (actual characters: {}, from formatting: {})".format(
                    self.max_length,
                    length,
                    tag_length
                ),
                code='invalid',
                params={'value': value}
            )