import difflib

import json

import time

from jsondoc.serialize import load_page

from jsondoc.utils import diff_strings, load_json_file, timer

PAGE_PATH = "../schema/page/ex1_success.json"

def remove_null_fields(string):

"""

This is just to make the test pass, because they get removed when

we serialize the model with exclude_none=True

"""

lines = string.splitlines()

to_remove = [

'"link": null',

'"href": null',

]

# Remove lines that contain the fields

lines = [line for line in lines if all(field not in line for field in to_remove)]

return "\n".join(lines)

def remove_commas_from_end_of_lines(string):

return "\n".join([line.rstrip(",") for line in string.splitlines()])

def test_load_page():

content = load_json_file(PAGE_PATH)

# This should not raise any errors

with timer("load_page", unit="ms"):

page = load_page(content)

assert page is not None

# Serialize it again

serialized = page.model_dump_json(

serialize_as_any=True, # This argument is needed to output nested models

exclude_none=True,

)

canonical_content = json.dumps(content, indent=2, sort_keys=True)

canonical_serialized = json.dumps(json.loads(serialized), indent=2, sort_keys=True)

canonical_content = remove_null_fields(canonical_content)

# Remove commas from the end of lines

canonical_content = remove_commas_from_end_of_lines(canonical_content)

canonical_serialized = remove_commas_from_end_of_lines(canonical_serialized)

# Indented versions

# canonical_content = json.dumps

diff = diff_strings(canonical_content, canonical_serialized)

assert len(diff) == 0, (

"The serialized content is different from the original content:\n" + diff

)

if __name__ == "__main__":

test_load_page()