import xml
import json
import pytest

from grobid_tei_xml import teixml2json, parse_document_xml, GrobidDocument, GrobidCitation
from grobid_tei_xml.types import *


def test_teixml2json_small_xml():

    with open('tests/files/small.xml', 'r') as f:
        tei_xml = f.read()
    with open('tests/files/small.json', 'r') as f:
        json_form = json.loads(f.read())

    assert teixml2json(tei_xml) == json_form

    assert parse_document_xml(tei_xml).to_dict() == json_form

def test_teixml2json_small_xml():

    with open('tests/files/small.xml', 'r') as f:
        tei_xml = f.read()

    doc = parse_document_xml(tei_xml)
    expected = GrobidDocument(
        grobid_version='0.5.1-SNAPSHOT',
        grobid_timestamp='2018-04-02T00:31+0000',
        language_code='en',
        header=GrobidHeader(
            title="Dummy Example File",
            authors=[
                GrobidAuthor(
                    name="Brewster Kahle",
                    given_name="Brewster",
                    surname="Kahle",
                    affiliation=GrobidAffiliation(
                        department="Faculty ofAgricultrial Engineering",
                        laboratory="Plant Physiology Laboratory",
                        institution="Technion-Israel Institute of Technology",
                        address=GrobidAddress(
                            post_code="32000",
                            settlement="Haifa",
                            country="Israel",
                        ),
                    )
                ),
                GrobidAuthor(
                    name="J Doe",
                    given_name="J",
                    surname="Doe",
                ),
            ],
            journal=GrobidJournal(
                name="Dummy Example File. Journal of Fake News. pp. 1-2. ISSN 1234-5678",
            ),
            date="2000",
        ),
        abstract="Everything you ever wanted to know about nothing",
        body="Introduction \nEverything starts somewhere, as somebody [1]  once said. \n\n In Depth \n Meat \nYou know, for kids. \n Potatos \nQED.",
        citations=[
            GrobidCitation(
                index=0,
                id="b0",
                authors=[
                    GrobidAuthor(
                        name="A Seaperson",
                        given_name="A",
                        surname="Seaperson"
                    )
                ],
                date="2001",
                journal="Letters in the Alphabet",
                title="Everything is Wonderful",
                volume="20",
                pages="1-11",
            ),
            GrobidCitation(
                index=1,
                id="b1",
                authors=[],
                date="2011-03-28",
                journal="The Dictionary",
                title="All about Facts",
                volume="14",
            ),
        ],
    )

    assert doc == expected


def test_invalid_xml():

    with pytest.raises(xml.etree.ElementTree.ParseError):
        teixml2json("this is not XML")
    with pytest.raises(ValueError):
        teixml2json("<xml></xml>")


def test_grobid_teixml2json() -> None:

    with open("tests/files/example_grobid.tei.xml", "r") as f:
        blob = f.read()

    obj = teixml2json(blob, True)

    assert (
        obj["title"] ==
        "Changes of patients' satisfaction with the health care services in Lithuanian Health Promoting Hospitals network"
    )

    ref = [c for c in obj["citations"] if c["id"] == "b12"][0]
    assert ref["authors"][0] == {
        "given_name": "K",
        "name": "K Tasa",
        "surname": "Tasa"
    }
    assert ref["journal"] == "Quality Management in Health Care"
    assert ref["title"] == "Using patient feedback for quality improvement"
    assert ref["date"] == "1996"
    assert ref["pages"] == "206-225"
    assert ref["volume"] == "8"
    assert (
        ref["unstructured"] ==
        "Tasa K, Baker R, Murray M. Using patient feedback for qua- lity improvement. Quality Management in Health Care 1996;8:206-19."
    )