comparison pipeline_telescope.py @ 0:a35056104c2c draft default tip

planemo upload for repository https://github.com/esg-epfl-apc/tools-astro/tree/main/tools commit da42ae0d18f550dec7f6d7e29d297e7cf1909df2

0:a35056104c2c

import pandas as pd
import re

from rdflib import Graph, Namespace
from rdflib.namespace import RDF, RDFS, SKOS

from aux_functions import compute_sensitivity, compute_sensitivity_int, list_tel

ODA = Namespace("https://odahub.io/ontology#")
g_label_site = len("http://www.w3.org/2000/01/rdf-schema#label")


def find_entity(g, class_, text_id_text, text_id_text_upper):
    dict_ = {"label": {"val": [], "URI": [], "Sensitivity": []}, "altLabel": {"val": [], "URI": [], "Sensitivity": []}, "hiddenLabel": {"val": [], "URI": [], "Sensitivity": []}}
    for i, [u_telescope, p, o] in enumerate(g.triples((None, RDF.type, class_))):

        exists_label_telescope = 0
        for s, p, label_telescope in g.triples((u_telescope, RDFS.label, None)):

            result = re.search("\\b(" + label_telescope.lower() + ")([1-2]{0,1})\\b", text_id_text)
            if result:
                exists_label_telescope = 1
                val_ = result.group(0)
                # add_ = val_[len(label_telescope):]
                dict_["label"]["val"].append(label_telescope)
                dict_["label"]["URI"].append(u_telescope)
                dict_["label"]["Sensitivity"].append(compute_sensitivity(list_tel(u_telescope, g)))

        exists_altlabel_telescope = 0
        if exists_label_telescope == 0:
            for s, p, altlabel_telescope in g.triples((u_telescope, SKOS.altLabel, None)):

                result = re.search("\\b(" + altlabel_telescope + ")\\b", text_id_text_upper)
                if result:
                    exists_altlabel_telescope = 1
                    val_ = result.group(0)
                    dict_["altLabel"]["val"].append(val_)
                    dict_["altLabel"]["URI"].append(u_telescope)
                    dict_["altLabel"]["Sensitivity"].append(compute_sensitivity(list_tel(u_telescope, g)))

            if exists_altlabel_telescope == 0:
                for s, p, hiddenlabel_telescope in g.triples((u_telescope, SKOS.hiddenLabel, None)):

                    result = re.search("\\b" + hiddenlabel_telescope.lower() + "\\b", text_id_text)
                    if result:
                        val_ = result.group(0)
                        dict_["hiddenLabel"]["val"].append(hiddenlabel_telescope)
                        dict_["hiddenLabel"]["URI"].append(u_telescope)
                        dict_["hiddenLabel"]["Sensitivity"].append(compute_sensitivity(list_tel(u_telescope, g)))

    return dict_


def rule_based_telescope_detector(text_id, text_id_text, telescope_ontology):
    g = Graph()
    g.parse(telescope_ontology, format="n3")

    text_id_text_lower = text_id_text.lower()

    dict_observatory = find_entity(g, ODA.observatory,     text_id_text_lower, text_id_text)
    dict_survey = find_entity(g, ODA.survey,               text_id_text_lower, text_id_text)
    dict_telescope = find_entity(g, ODA.telescope,         text_id_text_lower, text_id_text)
    dict_misctelescope = find_entity(g, ODA.misctelescope, text_id_text_lower, text_id_text)
    dict_telescopetype = find_entity(g, ODA.telescopetype, text_id_text_lower, text_id_text)

    dict_spacetelescope = find_entity(g, ODA.spacetelescope, text_id_text_lower, text_id_text)
    dict_instrument = find_entity(g, ODA.instrument,         text_id_text_lower, text_id_text)
    dict_institution = find_entity(g, ODA.institution,       text_id_text_lower, text_id_text)
    dict_radiotelescope = find_entity(g, ODA.radiotelescope, text_id_text_lower, text_id_text)

    tel_sur_obs = []
    type_key = []
    uri_list = []
    sens_list = []

    for key in ["label", "altLabel", "hiddenLabel"]:
        list_key = dict_institution[key]["val"] + dict_spacetelescope[key]["val"] + dict_telescope[key]["val"] + dict_survey[key]["val"] + dict_observatory[key]["val"] + dict_radiotelescope[key]["val"] + dict_instrument[key]["val"] + dict_telescopetype[key]["val"] + dict_misctelescope[key]["val"]
        tel_sur_obs += list_key

        list_uri_key = dict_institution[key]["URI"] + dict_spacetelescope[key]["URI"] + dict_telescope[key]["URI"] + dict_survey[key]["URI"] + dict_observatory[key]["URI"] + dict_radiotelescope[key]["URI"] + dict_instrument[key]["URI"] + dict_telescopetype[key]["URI"] + dict_misctelescope[key]["URI"]
        uri_list += list_uri_key

        sens_list += dict_institution[key]["Sensitivity"] + dict_spacetelescope[key]["Sensitivity"] + dict_telescope[key]["Sensitivity"] + dict_survey[key]["Sensitivity"] + dict_observatory[key]["Sensitivity"] + dict_radiotelescope[key]["Sensitivity"] + dict_instrument[key]["Sensitivity"] + dict_telescopetype[key]["Sensitivity"] + dict_misctelescope[key]["Sensitivity"]

        type_key += [key]*len(list_key)

    dict_data = {"TEXT_ID": [text_id] * len(tel_sur_obs), "Telescope": tel_sur_obs, "LabelType": type_key, "URI": uri_list, "Sensitivity": sens_list, "Total Sensitivity": [compute_sensitivity_int(sens_list)] * len(tel_sur_obs)}

    df_data = pd.DataFrame(dict_data)
    df_data.drop_duplicates(subset=['URI'], inplace=True)

    return df_data