Progetto_2023-OVI/flask_be/engine/contexts.py

import json
import pandas as pd

from .basic_queries import basicQueries
from .utilities.format import formatAllContexts, formatContext


# Executes query sequences to recover single and multiple contexts
# Returns Pandas dataframes
class contexts(basicQueries):
    
    def __init__(self, dataConfig):
        super().__init__(dataConfig)

    #%% funzione contesti multipli cumulativa
    # Potrebbe essere unita alle cooccorrenze?
    def contestimultipli (self, tipo_ricerca, ricerca, index = None):
        ricercadf = pd.DataFrame(ricerca)
        textlist = self.findtexts(tipo_ricerca, ricercadf, index)
        contexts = self.findcontexts (textlist)
        bibliocontexts = self.findbib (contexts)
        highlights = formatAllContexts(bibliocontexts)
        return highlights.to_dict(orient='records')

    #%% funzione contesti singoli cumulativa
    def contestosingolo (self, contestimultipli, indice, parole, periodi, brani):
        ### droppa le colonne "highlight" che gli rompono le scatole###
        contestimultipli = {k: v for k, v in contestimultipli.items() if not k.startswith('highlight')}
        ###############################################################
        contestimultiplidf = pd.DataFrame(contestimultipli, index=[0])
        contestosingolo = self.singlecontexts(contestimultiplidf, indice, parole, periodi, brani)
        braniassociati = self.findlinks(contestosingolo)
        contestosingoloclean = self.findbib (braniassociati)
        contestosingoloclean = formatAllContexts(contestosingoloclean)
        return contestosingoloclean.to_dict(orient='records')
        
    #%% funzione reperimento e raffinamento contesti singoli
    def singlecontexts(self, textlist, index, parole, periodi, brani):
        context = textlist.iloc[index]
        contexts = []
        formats = []
        listOcc = self.listOcc
        sigla = textlist.loc[index, "sigla"]
        periodlocal = textlist.loc[index, "numperiod"]
        ntxlocal = textlist.loc[index, "ntx"]
        mappalocal = textlist.loc[index, "mappa"]
        linkslocal = textlist.loc[index, "links"]
        numbranolocal = textlist.loc[index, "numbrano"]
        if parole != 0:
            pointerlist = pd.DataFrame()
            for table in listOcc:
                queryData = {'queryType': 'singlecontext', 'querySubtype': 'parole', 'parole': parole, 'periodi': periodi, 'brani': brani, 'table': table, 'ntxlocal': ntxlocal, 'mappalocal': mappalocal, 'periodlocal': periodlocal, 'numbranolocal': numbranolocal}
                queryresponse = self.queryHandler.query(queryData, pandas=True)
                pointerlist = pd.concat([pointerlist, queryresponse])
            fileQueryData = {'sigla': sigla, 'minChar': pointerlist["pitxt"].min(), 'maxChar': pointerlist["pitxt"].max()}
            cont, form = self.queryHandler.textQuery(fileQueryData, True)
            contexts.append(cont)
            formats.append(json.dumps(form))
            context ['piniz'] = pointerlist["pitxt"].min()
            context ['pfin'] = pointerlist["pitxt"].max()
        elif periodi != 0:
            queryData = {'queryType': 'singlecontext', 'querySubtype': 'parole', 'parole': parole, 'periodi': periodi, 'brani': brani, 'table': table, 'ntxlocal': ntxlocal, 'mappalocal': mappalocal, 'periodlocal': periodlocal, 'numbranolocal': numbranolocal}
            queryresponse = self.queryHandler.query(queryData, pandas=True)
            fileQueryData = {'sigla': sigla, 'minChar': pointerlist["pitxt"].min(), 'maxChar': pointerlist["pitxt"].max()}
            cont, form = self.queryHandler.textQuery(fileQueryData, True)
            contexts.append(cont)
            formats.append(json.dumps(form))
            context ['piniz'] = queryresponse["piniz"].min()
            context ['pfin'] = queryresponse["pfin"].max()
        elif brani != 0:
            if linkslocal == 0 or linkslocal == 1:
                return "Nessun brano associato a questo contesto"
            else:
                queryData = {'queryType': 'singlecontext', 'querySubtype': 'brani', 'parole': parole, 'periodi': periodi, 'brani': brani, 'table': table, 'ntxlocal': ntxlocal, 'mappalocal': mappalocal, 'periodlocal': periodlocal, 'numbranolocal': numbranolocal}
                queryresponse = self.queryHandler.query(queryData, pandas=True)
                fileQueryData = {'sigla': sigla, 'minChar': pointerlist["pitxt"].min(), 'maxChar': pointerlist["pitxt"].max()}
                cont, form = self.queryHandler.textQuery(fileQueryData, True)
                contexts.append(cont)
                formats.append(json.dumps(form))
                context ['piniz'] = queryresponse["piniz"].min()
                context ['pfin'] = queryresponse["pfin"].max() 
        context['contesto'] = contexts[0]
        context['formattazione contesto'] = formats[0]
        return pd.DataFrame(context).T.reset_index(drop=True)
    
    #%% funzione reperimento note e brani associati
    def findlinks (self, context):
        linkslocal = context.loc[0, "links"]
        siglalocal = context.loc[0, "sigla"]
        ntxlocal = context.loc[0, "ntx"]
        pitxtlocal = context.loc[0, "pitxt"]
        pinizlocal = context.loc[0, "piniz"]
        pfinlocal = context.loc[0, "pfin"]
        if linkslocal == 0:
            return context
        if linkslocal == 1:
            queryData = {'queryType': 'links', 'querySubtype': 'nota', 'ntxlocal': ntxlocal, 'pinizlocal': pinizlocal, 'pitxtlocal': pitxtlocal, 'pfinlocal': pfinlocal}
            queryresponse = self.queryHandler.query(queryData, pandas=True)
            fileQueryData = {'sigla': siglalocal, 'minChar': queryresponse["piniz"].min(), 'maxChar': queryresponse["pfin"].max()}
            cont, form = self.queryHandler.textQuery(fileQueryData, True)
            context['nota'] = cont
            context['formattazione nota'] = json.dumps(form)
            context['nota formattata'] = formatContext(json.dumps(form))
            return context
        if linkslocal == 2:
            queryData = {'queryType': 'links', 'querySubtype': 'testo_associato', 'ntxlocal': ntxlocal, 'pinizlocal': pinizlocal, 'pitxtlocal': pitxtlocal, 'pfinlocal': pfinlocal}
            queryresponse = self.queryHandler.query(queryData, pandas=True)
            fileQueryData = {'sigla': siglalocal, 'minChar': queryresponse["piniz"].min(), 'maxChar': queryresponse["pfin"].max()}
            cont, form = self.queryHandler.textQuery(fileQueryData, True)
            context['testo associato'] = cont
            context['formattazione testo associato'] = json.dumps(form)
            context['testo associato formattato'] = formatContext(cont, json.dumps(form))
        if linkslocal == 3:
            queryData = {'queryType': 'links', 'querySubtype': 'nota', 'ntxlocal': ntxlocal, 'pinizlocal': pinizlocal, 'pitxtlocal': pitxtlocal, 'pfinlocal': pfinlocal}
            queryresponse = self.queryHandler.query(queryData, pandas=True)
            fileQueryData = {'sigla': siglalocal, 'minChar': queryresponse["piniz"].min(), 'maxChar': queryresponse["pfin"].max()}
            cont, form = self.queryHandler.textQuery(fileQueryData, True)
            context['nota'] = cont
            context['formattazione nota'] = json.dumps(form)
            context['nota formattata'] = formatContext(cont, json.dumps(form))
            queryData2 = {'queryType': 'links', 'querySubtype': 'testo_associato', 'ntxlocal': ntxlocal, 'pinizlocal': pinizlocal, 'pitxtlocal': pitxtlocal, 'pfinlocal': pfinlocal}
            queryresponse2 = self.queryHandler.query(queryData2, pandas=True)
            fileQueryData2 = {'sigla': siglalocal, 'minChar': queryresponse2["piniz"].min(), 'maxChar': queryresponse2["pfin"].max()}
            cont2, form2 = self.queryHandler.textQuery(fileQueryData2, True)
            context['testo associato'] = cont2
            context['formattazione testo associato'] = json.dumps(form2)
            context['testo associato formattato'] = formatContext(cont2, json.dumps(form2))
        return context