bert_tokenizer_params in Subword tokenizers guide

I follow the guide to Generate the vocabulary step and modify bert_tokenizer_params to
bert_tokenizer_params=dict(max_bytes_per_word=42, max_chars_per_token=9)
base on text.BertTokenizer docs cause the bert_vocab_args code has a comment mention that this is the arguments for text.BertTokenizer.

bert_vocab_args = dict(
    ...
    # Arguments for `text.BertTokenizer`
    bert_tokenizer_params=bert_tokenizer_params,
    ...
)

Then I use bert_vocab_args to generate the vocab file. But It shows error, turn out the bert_tokenizer_params in bert_vocab_args is the input of bert_tokenizer.BasicTokenizer() which doesn't have max_bytes_per_word nor max_chars_per_token in the __init__() of BasicTokenizer code.
I update the bert_tokenizer_params inside the bert_vocab_args to an empty dict

bert_vocab_args = dict(
    ...
    # Arguments for `text.BertTokenizer`
    bert_tokenizer_params={},
    ...
)

and move the bert_tokenizer_params=dict(max_bytes_per_word=42, max_chars_per_token=9) to Build the tokenizer step. I'm still waiting bert_vocab_from_dataset() code run and see how my vocab file look like and how my tokenizer behave.
I think you should clearly state which are the text.BertTokenizer params and which are the BasicTokenizer params or update the bert_vocab_from_dataset.py code to match what the guide means.

Start with the Subword tokenizers guide sections on generating the vocabulary and building the tokenizer, then compare their parameter examples with bert_vocab_from_dataset.py and bert_tokenizer.py. Confirm which arguments belong to BasicTokenizer versus text.BertTokenizer, and document the distinction or align the guide and code so the example runs without the reported error.