from datetime import date as Date
import json
import os
import io
import re
import shutil

import pydantic
import pytest
import time_machine
import ruamel.yaml

from rendercv import data_models as dm

from .conftest import update_testdata


@pytest.mark.parametrize(
    "date, expected_date_object, expected_error",
    [
        ("2020-01-01", Date(2020, 1, 1), None),
        ("2020-01", Date(2020, 1, 1), None),
        ("2020", Date(2020, 1, 1), None),
        (2020, Date(2020, 1, 1), None),
        ("present", Date(2024, 1, 1), None),
        ("invalid", None, ValueError),
        ("20222", None, ValueError),
        ("202222-20200", None, ValueError),
        ("202222-12-20", None, ValueError),
        ("2022-20-20", None, ValueError),
    ],
)
@time_machine.travel("2024-01-01")
def test_get_date_object(date, expected_date_object, expected_error):
    if expected_error:
        with pytest.raises(expected_error):
            dm.get_date_object(date)
    else:
        assert dm.get_date_object(date) == expected_date_object


@pytest.mark.parametrize(
    "date, expected_date_string",
    [
        (Date(2020, 1, 1), "Jan. 2020"),
        (Date(2020, 2, 1), "Feb. 2020"),
        (Date(2020, 3, 1), "Mar. 2020"),
        (Date(2020, 4, 1), "Apr. 2020"),
        (Date(2020, 5, 1), "May 2020"),
        (Date(2020, 6, 1), "June 2020"),
        (Date(2020, 7, 1), "July 2020"),
        (Date(2020, 8, 1), "Aug. 2020"),
        (Date(2020, 9, 1), "Sept. 2020"),
        (Date(2020, 10, 1), "Oct. 2020"),
        (Date(2020, 11, 1), "Nov. 2020"),
        (Date(2020, 12, 1), "Dec. 2020"),
    ],
)
def test_format_date(date, expected_date_string):
    assert dm.format_date(date) == expected_date_string


@pytest.mark.parametrize(
    "key, value",
    [
        ("cv.phone", "+905555555555"),
        ("cv.email", "test@example.com"),
        ("cv.sections.education.0.degree", "PhD"),
        ("cv.sections.education.0.highlights.0", "Did this."),
        ("cv.sections.this_is_a_new_section", '["This is a text entry."]'),
        ("design.page_size", "a4paper"),
        ("design", '{"theme": "engineeringresumes"}'),
    ],
)
def test_set_or_update_a_value(rendercv_data_model, key, value):
    dm.set_or_update_a_value(rendercv_data_model, key, value)

    # replace with regex pattern:
    key = re.sub(r"sections\.([^\.]*)", 'sections_input["\\1"]', key)
    key = re.sub(r"\.(\d+)", "[\\1]", key)

    if value.startswith("{") and value.endswith("}"):
        value = eval(value)
    elif value.startswith("[") and value.endswith("]"):
        value = eval(value)

    assert eval(f"rendercv_data_model.{key}") == value


@pytest.mark.parametrize(
    "key, value",
    [
        ("cv.phones", "+905555555555"),
        ("cv.emssdsail", ""),
        ("cv.sections.education.99.degree", "PhD"),
        ("dessssign.page_size", "a4paper"),
    ],
)
def test_set_or_update_a_value_invalid_keys(rendercv_data_model, key, value):
    with pytest.raises((ValueError, KeyError, IndexError, AttributeError)):
        dm.set_or_update_a_value(rendercv_data_model, key, value)


@pytest.mark.parametrize(
    "key, value",
    [
        ("cv.phone", "+9999995555555555"),
        ("cv.email", "notanemail***"),
        ("cv.sections.education.0.highlights", "this is not a list"),
        ("design.page_size", "invalid_page_size"),
    ],
)
def test_set_or_update_a_value_invalid_values(rendercv_data_model, key, value):
    with pytest.raises(pydantic.ValidationError):
        dm.set_or_update_a_value(rendercv_data_model, key, value)


def test_read_input_file(input_file_path):
    # Update the auxiliary files if update_testdata is True
    if update_testdata:
        # create testdata directory if it doesn't exist
        if not input_file_path.parent.exists():
            input_file_path.parent.mkdir()

        input_dictionary = {
            "cv": {
                "name": "John Doe",
            },
            "design": {
                "theme": "classic",
            },
        }

        # dump the dictionary to a yaml file
        yaml_object = ruamel.yaml.YAML()
        yaml_object.dump(input_dictionary, input_file_path)

    data_model = dm.read_input_file(input_file_path)

    assert isinstance(data_model, dm.RenderCVDataModel)


def test_read_input_file_directly_with_contents(input_file_path):
    input_dictionary = {
        "cv": {
            "name": "John Doe",
        },
        "design": {
            "theme": "classic",
        },
    }

    # dump the dictionary to a yaml file
    yaml_object = ruamel.yaml.YAML()
    yaml_object.width = 60
    yaml_object.indent(mapping=2, sequence=4, offset=2)
    with io.StringIO() as string_stream:
        yaml_object.dump(input_dictionary, string_stream)
        yaml_string = string_stream.getvalue()

    data_model = dm.read_input_file(yaml_string)

    assert isinstance(data_model, dm.RenderCVDataModel)


def test_read_input_file_invalid_file(tmp_path):
    invalid_file_path = tmp_path / "invalid.extension"
    invalid_file_path.write_text("dummy content", encoding="utf-8")
    with pytest.raises(ValueError):
        dm.read_input_file(invalid_file_path)


def test_read_input_file_that_doesnt_exist(tmp_path):
    non_existent_file_path = tmp_path / "non_existent_file.yaml"
    with pytest.raises(FileNotFoundError):
        dm.read_input_file(non_existent_file_path)


@pytest.mark.parametrize(
    "theme",
    dm.available_themes,
)
def test_get_a_sample_data_model(theme):
    data_model = dm.get_a_sample_data_model("John Doe", theme)
    assert isinstance(data_model, dm.RenderCVDataModel)


def test_get_a_sample_data_model_invalid_theme():
    with pytest.raises(ValueError):
        dm.get_a_sample_data_model("John Doe", "invalid")


def test_generate_json_schema():
    schema = dm.generate_json_schema()
    assert isinstance(schema, dict)


def test_generate_json_schema_file(tmp_path):
    schema_file_path = tmp_path / "schema.json"
    dm.generate_json_schema_file(schema_file_path)

    assert schema_file_path.exists()

    schema_text = schema_file_path.read_text(encoding="utf-8")
    schema = json.loads(schema_text)

    assert isinstance(schema, dict)


# def test_if_the_schema_is_the_latest(root_directory_path):
#     original_schema_file_path = root_directory_path / "schema.json"
#     original_schema_text = original_schema_file_path.read_text()
#     original_schema = json.loads(original_schema_text)

#     new_schema = dm.generate_json_schema()

#     assert original_schema == new_schema


@pytest.mark.parametrize(
    "start_date, end_date, date, expected_date_string, expected_date_string_only_years,"
    " expected_time_span",
    [
        (
            "2020-01-01",
            "2021-01-01",
            None,
            "Jan. 2020 to Jan. 2021",
            "2020 to 2021",
            "1 year 1 month",
        ),
        (
            Date(2020, 1, 1),
            Date(2021, 1, 1),
            None,
            "Jan. 2020 to Jan. 2021",
            "2020 to 2021",
            "1 year 1 month",
        ),
        (
            "2020-01",
            "2021-01",
            None,
            "Jan. 2020 to Jan. 2021",
            "2020 to 2021",
            "1 year 1 month",
        ),
        (
            "2020-01",
            "2021-01-01",
            None,
            "Jan. 2020 to Jan. 2021",
            "2020 to 2021",
            "1 year 1 month",
        ),
        (
            "2020-01-01",
            "2021-01",
            None,
            "Jan. 2020 to Jan. 2021",
            "2020 to 2021",
            "1 year 1 month",
        ),
        (
            "2020-01-01",
            None,
            None,
            "Jan. 2020 to present",
            "2020 to present",
            "4 years 1 month",
        ),
        (
            "2020-02-01",
            "present",
            None,
            "Feb. 2020 to present",
            "2020 to present",
            "3 years 11 months",
        ),
        ("2020-01-01", "2021-01-01", "2023-02-01", "Feb. 2023", "2023", ""),
        ("2020", "2021", None, "2020 to 2021", "2020 to 2021", "1 year"),
        ("2020", None, None, "2020 to present", "2020 to present", "4 years"),
        ("2020-10-10", "2022", None, "Oct. 2020 to 2022", "2020 to 2022", "2 years"),
        (
            "2020-10-10",
            "2020-11-05",
            None,
            "Oct. 2020 to Nov. 2020",
            "2020 to 2020",
            "1 month",
        ),
        ("2022", "2023-10-10", None, "2022 to Oct. 2023", "2022 to 2023", "1 year"),
        (
            "2020-01-01",
            "present",
            "My Custom Date",
            "My Custom Date",
            "My Custom Date",
            "",
        ),
        (
            "2020-01-01",
            None,
            "My Custom Date",
            "My Custom Date",
            "My Custom Date",
            "",
        ),
        (
            None,
            None,
            "My Custom Date",
            "My Custom Date",
            "My Custom Date",
            "",
        ),
        (
            None,
            "2020-01-01",
            "My Custom Date",
            "My Custom Date",
            "My Custom Date",
            "",
        ),
        (None, None, "2020-01-01", "Jan. 2020", "2020", ""),
        (None, None, "2020-09", "Sept. 2020", "2020", ""),
        (None, None, Date(2020, 1, 1), "Jan. 2020", "2020", ""),
        (None, None, None, "", "", ""),
        (None, "2020-01-01", None, "Jan. 2020", "2020", ""),
        (None, "present", None, "Jan. 2024", "2024", ""),
        ("2002", "2020", "2024", "2024", "2024", ""),
    ],
)
@time_machine.travel("2024-01-01")
def test_dates(
    start_date,
    end_date,
    date,
    expected_date_string,
    expected_date_string_only_years,
    expected_time_span,
):
    entry_base = dm.EntryBase(start_date=start_date, end_date=end_date, date=date)

    assert entry_base.date_string == expected_date_string
    assert entry_base.date_string_only_years == expected_date_string_only_years
    assert entry_base.time_span_string == expected_time_span


@pytest.mark.parametrize(
    "date, expected_date_string",
    [
        ("2020-01-01", "Jan. 2020"),
        ("2020-01", "Jan. 2020"),
        ("2020", "2020"),
    ],
)
def test_publication_dates(publication_entry, date, expected_date_string):
    publication_entry["date"] = date
    publication_entry = dm.PublicationEntry(**publication_entry)
    assert publication_entry.date_string == expected_date_string


@pytest.mark.parametrize("date", ["2025-23-23"])
def test_invalid_publication_dates(publication_entry, date):
    with pytest.raises(pydantic.ValidationError):
        publication_entry["date"] = date
        dm.PublicationEntry(**publication_entry)


@pytest.mark.parametrize(
    "start_date, end_date, date",
    [
        ("aaa", "2021-01-01", None),
        ("2020-01-01", "aaa", None),
        ("2023-01-01", "2021-01-01", None),
        ("2022", "2021", None),
        ("2025", "2021", None),
        ("2020-01-01", "invalid_end_date", None),
        ("invalid_start_date", "2021-01-01", None),
        ("2020-99-99", "2021-01-01", None),
        ("2020-10-12", "2020-99-99", None),
        (None, None, "2020-20-20"),
    ],
)
def test_invalid_dates(start_date, end_date, date):
    with pytest.raises(pydantic.ValidationError):
        dm.EntryBase(start_date=start_date, end_date=end_date, date=date)


@pytest.mark.parametrize(
    "doi, expected_doi_url",
    [
        ("10.1109/TASC.2023.3340648", "https://doi.org/10.1109/TASC.2023.3340648"),
    ],
)
def test_doi_url(publication_entry, doi, expected_doi_url):
    publication_entry["doi"] = doi
    publication_entry = dm.PublicationEntry(**publication_entry)
    assert publication_entry.doi_url == expected_doi_url


@pytest.mark.parametrize(
    "doi",
    ["aaa10.1109/TASC.2023.3340648", "aaa", "A string"],
)
def test_invalid_doi(publication_entry, doi):
    with pytest.raises(pydantic.ValidationError):
        publication_entry["doi"] = doi
        dm.PublicationEntry(**publication_entry)


@pytest.mark.parametrize(
    "network, username",
    [
        ("Mastodon", "invalidmastodon"),
        ("Mastodon", "@inva@l@id"),
        ("Mastodon", "@invalid@ne<>twork.com"),
        ("StackOverflow", "invalidusername"),
        ("StackOverflow", "invalidusername//"),
        ("StackOverflow", "invalidusername/invalid"),
        ("YouTube", "@invalidusername"),
    ],
)
def test_invalid_social_networks(network, username):
    with pytest.raises(pydantic.ValidationError):
        dm.SocialNetwork(network=network, username=username)


@pytest.mark.parametrize(
    "network, username, expected_url",
    [
        ("LinkedIn", "myusername", "https://linkedin.com/in/myusername"),
        ("GitHub", "myusername", "https://github.com/myusername"),
        ("Instagram", "myusername", "https://instagram.com/myusername"),
        ("ORCID", "myusername", "https://orcid.org/myusername"),
        ("Twitter", "myusername", "https://twitter.com/myusername"),
        ("Mastodon", "@myusername@test.org", "https://test.org/@myusername"),
        (
            "StackOverflow",
            "4567/myusername",
            "https://stackoverflow.com/users/4567/myusername",
        ),
        (
            "GitLab",
            "myusername",
            "https://gitlab.com/myusername",
        ),
        (
            "ResearchGate",
            "myusername",
            "https://researchgate.net/profile/myusername",
        ),
        (
            "YouTube",
            "myusername",
            "https://youtube.com/@myusername",
        ),
        (
            "Google Scholar",
            "myusername",
            "https://scholar.google.com/citations?user=myusername",
        ),
    ],
)
def test_social_network_url(network, username, expected_url):
    social_network = dm.SocialNetwork(network=network, username=username)
    assert str(social_network.url) == expected_url


@pytest.mark.parametrize(
    "entry, expected_entry_type, expected_section_type",
    [
        (
            "publication_entry",
            "PublicationEntry",
            "SectionWithPublicationEntries",
        ),
        (
            "experience_entry",
            "ExperienceEntry",
            "SectionWithExperienceEntries",
        ),
        (
            "education_entry",
            "EducationEntry",
            "SectionWithEducationEntries",
        ),
        (
            "normal_entry",
            "NormalEntry",
            "SectionWithNormalEntries",
        ),
        ("one_line_entry", "OneLineEntry", "SectionWithOneLineEntries"),
        ("text_entry", "TextEntry", "SectionWithTextEntries"),
        ("bullet_entry", "BulletEntry", "SectionWithBulletEntries"),
    ],
)
def test_get_entry_and_section_type(
    entry, expected_entry_type, expected_section_type, request: pytest.FixtureRequest
):
    entry = request.getfixturevalue(entry)
    entry_type, section_type = dm.get_entry_and_section_type(entry)
    assert entry_type == expected_entry_type
    assert section_type.__name__ == expected_section_type

    # initialize the entry with the entry type
    if entry_type != "TextEntry":
        entry = eval(f"dm.{entry_type}(**entry)")
        entry_type, section_type = dm.get_entry_and_section_type(entry)
        assert entry_type == expected_entry_type
        assert section_type.__name__ == expected_section_type


def test_sections(
    education_entry,
    experience_entry,
    publication_entry,
    normal_entry,
    one_line_entry,
    text_entry,
):
    input = {
        "name": "John Doe",
        "sections": {
            "arbitrary_title": [
                education_entry,
                education_entry,
            ],
            "arbitrary_title_2": [
                experience_entry,
                experience_entry,
            ],
            "arbitrary_title_3": [
                publication_entry,
                publication_entry,
            ],
            "arbitrary_title_4": [
                normal_entry,
                normal_entry,
            ],
            "arbitrary_title_5": [
                one_line_entry,
                one_line_entry,
            ],
            "arbitrary_title_6": [
                text_entry,
                text_entry,
            ],
        },
    }

    cv = dm.CurriculumVitae(**input)
    assert len(cv.sections) == 6
    for section in cv.sections:
        assert len(section.entries) == 2


def test_sections_with_invalid_entries():
    input = {"name": "John Doe", "sections": dict()}
    input["sections"]["section_title"] = [
        {
            "this": "is",
            "an": "invalid",
            "entry": 10,
        }
    ]
    with pytest.raises(pydantic.ValidationError):
        dm.CurriculumVitae(**input)


@pytest.mark.parametrize(
    "invalid_custom_theme_name",
    [
        "pathdoesntexist",
        "invalid_theme_name",
    ],
)
def test_invalid_custom_theme(invalid_custom_theme_name):
    with pytest.raises(pydantic.ValidationError):
        dm.RenderCVDataModel(
            **{
                "cv": {"name": "John Doe"},
                "design": {"theme": invalid_custom_theme_name},
            }
        )


def test_custom_theme_with_missing_files(tmp_path):
    custom_theme_path = tmp_path / "customtheme"
    custom_theme_path.mkdir()
    with pytest.raises(pydantic.ValidationError):
        os.chdir(tmp_path)
        dm.RenderCVDataModel(
            **{  # type: ignore
                "cv": {"name": "John Doe"},
                "design": {"theme": "customtheme"},
            }
        )


def test_custom_theme(testdata_directory_path):
    os.chdir(
        testdata_directory_path
        / "test_copy_theme_files_to_output_directory_custom_theme"
    )
    data_model = dm.RenderCVDataModel(
        **{  # type: ignore
            "cv": {"name": "John Doe"},
            "design": {"theme": "dummytheme"},
        }
    )

    assert data_model.design.theme == "dummytheme"


def test_custom_theme_without_init_file(tmp_path, testdata_directory_path):
    reference_custom_theme_path = (
        testdata_directory_path
        / "test_copy_theme_files_to_output_directory_custom_theme"
        / "dummytheme"
    )

    # copy the directory to tmp_path:
    custom_theme_path = tmp_path / "dummytheme"
    shutil.copytree(reference_custom_theme_path, custom_theme_path, dirs_exist_ok=True)

    # remove the __init__.py file:
    init_file = custom_theme_path / "__init__.py"
    init_file.unlink()

    os.chdir(tmp_path)
    data_model = dm.RenderCVDataModel(
        **{  # type: ignore
            "cv": {"name": "John Doe"},
            "design": {"theme": "dummytheme"},
        }
    )

    assert data_model.design.theme == "dummytheme"


def test_custom_theme_with_broken_init_file(tmp_path, testdata_directory_path):
    reference_custom_theme_path = (
        testdata_directory_path
        / "test_copy_theme_files_to_output_directory_custom_theme"
        / "dummytheme"
    )

    # copy the directory to tmp_path:
    custom_theme_path = tmp_path / "dummytheme"
    shutil.copytree(reference_custom_theme_path, custom_theme_path, dirs_exist_ok=True)

    # overwrite the __init__.py file (syntax error)
    init_file = custom_theme_path / "__init__.py"
    init_file.write_text("invalid python code", encoding="utf-8")

    os.chdir(tmp_path)
    with pytest.raises(pydantic.ValidationError):
        dm.RenderCVDataModel(
            **{  # type: ignore
                "cv": {"name": "John Doe"},
                "design": {"theme": "dummytheme"},
            }
        )

    # overwrite the __init__.py file (import error)
    init_file = custom_theme_path / "__init__.py"
    init_file.write_text("from ... import test", encoding="utf-8")

    os.chdir(tmp_path)
    with pytest.raises(pydantic.ValidationError):
        dm.RenderCVDataModel(
            **{  # type: ignore
                "cv": {"name": "John Doe"},
                "design": {"theme": "dummytheme"},
            }
        )


def test_locale_catalog():
    data_model = dm.get_a_sample_data_model("John Doe")
    data_model.locale_catalog = dm.LocaleCatalog(
        month="a",
        months="b",
        year="c",
        years="d",
        present="e",
        to="f",
        abbreviations_for_months=[
            "1",
            "2",
            "3",
            "4",
            "5",
            "6",
            "7",
            "8",
            "9",
            "10",
            "11",
            "12",
        ],
    )

    assert dm.locale_catalog == data_model.locale_catalog.model_dump()


def test_if_local_catalog_resets():
    data_model = dm.get_a_sample_data_model("John Doe")

    data_model.locale_catalog = dm.LocaleCatalog(
        month="a",
    )

    assert dm.locale_catalog["month"] == "a"

    data_model = dm.get_a_sample_data_model("John Doe")

    assert dm.locale_catalog["month"] == "month"


def test_dictionary_to_yaml():
    input_dictionary = {
        "test_list": [
            "a",
            "b",
            "c",
        ],
        "test_dict": {
            "a": 1,
            "b": 2,
        },
    }
    yaml_string = dm.dictionary_to_yaml(input_dictionary)

    # load the yaml string
    yaml_object = ruamel.yaml.YAML()
    output_dictionary = yaml_object.load(yaml_string)

    assert input_dictionary == output_dictionary


def test_create_a_sample_yaml_input_file(tmp_path):
    input_file_path = tmp_path / "input.yaml"
    yaml_contents = dm.create_a_sample_yaml_input_file(input_file_path)

    assert input_file_path.exists()
    assert yaml_contents == input_file_path.read_text(encoding="utf-8")


def test_default_input_file_doesnt_have_local_catalog():
    yaml_contents = dm.create_a_sample_yaml_input_file()
    assert "locale_catalog" not in yaml_contents