import os
import requests
from bs4 import BeautifulSoup
import re
import zipfile
import glob
import re
import matplotlib.pyplot as plt
import matplotlib.style as style
# style.available
style.use('fivethirtyeight')
# ------------------------------------------------------------
# lendo a lista dos IDs e nome dos pesquisadores
df_idlist = readIdList()
# ------------------------------------------------------------
# roda as funcoes para pegar dados de cada pesquisador
for nid in range(len(df_idlist)):
zipfilename = str(df_idlist.iloc[nid, 0]) + '.zip'
getprojpesqext(zipfilename)
getprodtec(zipfilename)
getorient(zipfilename)
getperiod(zipfilename)
getlivro(zipfilename)
getcapit(zipfilename)
getnomecompleto(zipfilename)
getdiscip(zipfilename)
def getgrapho():
# lendo a lista dos IDs e nome dos pesquisadores
df_idlist = readIdList()
# df_idlist['ID_LATTES'] = df_idlist['ID_LATTES'].apply(ss)
config_file = open('./config.txt', 'r')
yyi = config_file.readlines()[5].split(':')[1]
yyi = yyi.rstrip('\n')
yyi = yyi.strip(' ')
yyi = float(yyi)
config_file.close()
config_file = open('./config.txt', 'r')
yyf = config_file.readlines()[6].split(':')[1]
yyf = yyf.rstrip('\n')
yyf = yyf.strip(' ')
yyf = float(yyf)
config_file.close()
# ------------------------------------------------------------
# importadando os data frames gerados pelo gettidy
# ------------------------------------------------------------
dfppe_uniq = pd.read_csv('./csv_producao/projetos_uniq.csv', header=0)
dfpaper = pd.read_csv('./csv_producao/periodicos_all.csv', header=0)
dfpaper_uniq = pd.read_csv('./csv_producao/periodicos_uniq.csv', header=0)
# paper uniq
dfpaper['ID'] = dfpaper['ID'].apply(ss)
dfpaper_uniq['ID'] = dfpaper_uniq['ID'].apply(ss)
# filtrando o ano
# projetos
dfppe_uniq['YEAR_INI'] = dfppe_uniq['YEAR_INI'].replace('VAZIO', -99)
num99 = dfppe_uniq[dfppe_uniq['YEAR_INI'] == -99]
if len(num99) >= 1:
print('------------------------------------------------------------')
print('ATENCAO: ' + str(len(num99)) + 'projetos sem ano inicial')
print('------------------------------------------------------------')
dfppe_uniq['YEAR_INI'] = dfppe_uniq['YEAR_INI'].apply(ff)
dfppe_uniq = dfppe_uniq[(dfppe_uniq['YEAR_INI'] >= yyi)]
# ------------------------------------------------------------
# periodicos
dfpaper['YEAR'] = dfpaper['YEAR'].replace('VAZIO', -99)
dfpaper_uniq['YEAR'] = dfpaper_uniq['YEAR'].replace('VAZIO', -99)
num99 = dfpaper[dfpaper['YEAR'] == -99]
if len(num99) >= 1:
print('------------------------------------------------------------')
print('ATENCAO: ' + str(len(num99)) + 'artigos sem ano de publicacao')
print('------------------------------------------------------------')
dfpaper['YEAR'] = dfpaper['YEAR'].apply(ff)
dfpaper_uniq['YEAR'] = dfpaper_uniq['YEAR'].apply(ff)
dfpaper = dfpaper[(dfpaper['YEAR'] >= yyi) & (dfpaper['YEAR'] <= yyf)]
dfpaper_uniq = dfpaper_uniq[(dfpaper_uniq['YEAR'] >= yyi)
& (dfpaper_uniq['YEAR'] <= yyf)]
# ------------------------------------------------------------
# ordenando por ano (crescente)
dfppe_uniq_pesq = dfppe_uniq[dfppe_uniq['NATUREZA'] == 'PESQUISA']
dfppe_uniq_pesq = dfppe_uniq_pesq.sort_values(['YEAR_INI'])
dfppe_uniq_ext = dfppe_uniq[dfppe_uniq['NATUREZA'] == 'EXTENSAO']
dfppe_uniq_ext = dfppe_uniq_ext.sort_values(['YEAR_INI'])
dfpaper = dfpaper.sort_values(['YEAR'])
dfpaper_uniq = dfpaper_uniq.sort_values(['YEAR'])
# ------------------------------------------------------------
# carregando df com dados pessoais
lscsv_fullname = glob.glob('./csv_producao/*fullname.csv')
# df com nome completo, sobrenome e id
dffullname = pd.DataFrame()
for i in range(len(lscsv_fullname)):
a = pd.read_csv(lscsv_fullname[i], header=0, dtype='str')
dffullname = dffullname.append(a, ignore_index=False)
# passando ID para string, para poder comparar com dfpaper
dffullname['ID'] = dffullname['ID'].apply(ss)
dffullname = dffullname.reset_index(drop=True)
# verificando a interacao de periodicos entre integrantes
lsid = []
lsid_tocompare = []
lsinter_qtd = []
for m in range(len(df_idlist)):
idd = str(df_idlist.iloc[m, 0])
lname = dffullname[dffullname['ID'] == idd]
lname = lname.iloc[0, 1]
lname = lname.upper()
# lname = lname.split(';')
# print(lname)
dfids_tocompare = dffullname[dffullname['ID'] != str(idd)]
for n in range(len(dfids_tocompare)):
idd_tocompare = dfids_tocompare.iloc[n, 0]
dd = dfpaper[dfpaper['ID'] == idd_tocompare]
lsid.append(str(idd))
lsid_tocompare.append(idd_tocompare)
# DANGER ATTENTION FIX lname deve ser o nome completo
# removendo caract desnecessarios
interac = 0
for o in range(len(dd)):
authors = dd.iloc[o, 7].upper()
authors = authors.replace('[', '')
authors = authors.replace(']', '')
authors = authors.replace("'", '')
authors = authors.split(',')
# print(authors)
for op in range(len(authors)):
# print(authors[op])
if len(authors[op]) > 0:
if authors[op][0] == ' ':
authors[op] = authors[op][1:]
# interac = 0
inpaper = list(set([lname]) & set(authors))
if len(inpaper) >= 1:
interac = interac + 1
# print(interac)
# print(lname)
# print(authors)
lsinter_qtd.append(interac)
dfinterac = pd.DataFrame({
'IDD': lsid,
'IDD_COMP': lsid_tocompare,
'WEIGHT': lsinter_qtd
})
# data frame para profissionais sem interacao em periodicos
lsnointer_period = []
for m in range(len(df_idlist)):
aano = dfinterac[dfinterac['IDD'] == df_idlist.iloc[m, 0]]
aasum = aano['WEIGHT'].sum()
aano_a = dfinterac[dfinterac['IDD_COMP'] == df_idlist.iloc[m, 0]]
aasum_a = aano_a['WEIGHT'].sum()
if aasum == 0 and aasum_a == 0:
nointer = dffullname[dffullname['ID'] == df_idlist.iloc[
m, 0]].reset_index(drop=True)
nointer = nointer.iloc[0, 1]
lsnointer_period.append(nointer)
dfnointerac = pd.DataFrame({'NOME': lsnointer_period})
dfnointerac.to_csv('./csv_producao/periodicos_nointer.csv',
index=False,
sep=',')
# DANGER ATTENTION
# dfinterac.to_csv('test.csv', index=False)
# eliminando linhas sem interacao
indexremove = []
for i in range(len(lsid)):
if lsinter_qtd[i] == 0:
indexremove.append(i)
for index in sorted(indexremove, reverse=True):
del lsid[index]
del lsid_tocompare[index]
del lsinter_qtd[index]
# ------------------------------------------------------------
# Grapho
plt.figure(figsize=(12, 9.5))
G = nx.Graph()
for i in range(len(lsid)):
G.add_edge(lsid[i], lsid_tocompare[i], weight=lsinter_qtd[0])
pos = nx.spring_layout(G, 1.75)
# colors for nodes
colours = [
'#5a7d9a', 'red', 'green', 'yellow', 'gray', 'orange', 'blue',
'magenta', '#00555a', '#f7d560', 'cyan', '#b6b129', '#a1dd72',
'#d49acb', '#d4a69a', '#977e93', '#a3cc72', '#c60acb', '#d4b22a',
'#255e53', '#77525a', '#c7d511', '#c4c22b', '#c9b329', '#c8dd22',
'#f75acb', '#b1a40a', '#216693', '#b1cd32', '#b33acb', '#c9a32b',
'#925e11', '#c5dd39', '#d04205', '#d8a82a', '#373e29'
]
lsgroup_uniq = df_idlist['GROUP'].unique()
dic_colours = {}
for i in range(len(lsgroup_uniq)):
dic_colours[lsgroup_uniq[i]] = colours[i]
a = list(G.nodes())
node_colours = []
for i in range(len(a)):
x = df_idlist[df_idlist['ID_LATTES'] == a[i]]
x = x.iloc[0, 2]
c = dic_colours[x]
node_colours.append(c)
# nodes
nx.draw_networkx_nodes(G,
pos,
node_size=400,
node_shape='o',
node_color=node_colours,
alpha=0.7)
# labels
nn = list(G.nodes)
diclabel = {}
for i in range(len(nn)):
x = df_idlist[df_idlist['ID_LATTES'] == nn[i]]
xid = x.iloc[0, 0]
xname = x.iloc[0, 1]
diclabel[str(xid)] = xname
# edges
nx.draw_networkx_edges(
G,
pos, # edgelist=lsinter_qtd,
width=1,
edge_color='orange')
# labels
nx.draw_networkx_labels(G,
pos,
labels=diclabel,
font_size=16,
font_family='sans-serif')
plt.axis('off')
plt.tight_layout()
plt.savefig('./relatorio/figures/grapho.png')
def capes_indprodart():
# lendo pesquisadores
df_idlist = readIdList()
num_dp = len(df_idlist)
# lendo periodicos_uniq
df = pd.read_csv('./csv_producao/periodicos_uniq.csv',
header=0, sep=',')
num_period_tot = len(df['QUALIS'])
df.query('QUALIS != "XX"', inplace=True)
df.query('QUALIS != "C "', inplace=True)
df['YEAR'] = df['YEAR'].apply(iint)
num_period_semqualis = num_period_tot - len(df['QUALIS'])
print('Numero de periodicos sem QUALIS = ', num_period_semqualis)
# definindo os quadrienios
year_fquadrien = 2013
ls_quadri = [year_fquadrien]
for i in range(5):
year_fquadrien = year_fquadrien + 4
ls_quadri.append(year_fquadrien)
# print(ls_quadri)
# calculo para cada trienio
ls_yini_quad = []
ls_yfin_quad = []
ls_indprodart = []
for i in range(len(ls_quadri)):
yini = ls_quadri[i]
yfin = ls_quadri[i] + 3
df_qtdby_qualis = df.groupby(['YEAR', 'QUALIS'])[
'TITLE'].count().reset_index()
df_qtdby_qualis.columns = ['YEAR', 'QUALIS', 'AMOUNT']
# print(df_qtdby_qualis)
df_qtdby_qualis.query('YEAR >= @yini and YEAR <= @yfin', inplace=True)
if len(df_qtdby_qualis) < 1:
pass
else:
# print('Quadrienio', yini, ' - ', yfin)
df_qtdby_qualis['PESO'] = df_qtdby_qualis['QUALIS'].apply(
fun_indprodart_classif)
df_qtdby_qualis['PROD_AMOUPESO'] = df_qtdby_qualis['AMOUNT'] * \
df_qtdby_qualis['PESO']
# verificando representatividade B4 e B5 deve ser <= 0.2 por ano
# print(df_qtdby_qualis)
df_grade_tot_year = df_qtdby_qualis.groupby(
['YEAR'])['PROD_AMOUPESO'].sum().reset_index()
df_qtdby_qualis_b4b5 = df_qtdby_qualis.query(
'QUALIS == "B4" or QUALIS == "B5"')
# print(df_grade_tot_year)
# print(df_qtdby_qualis_b4b5)
ls_years_b4b5_uniq = df_qtdby_qualis_b4b5['YEAR'].unique()
for ia in range(len(ls_years_b4b5_uniq)):
year_b4b5 = ls_years_b4b5_uniq[ia]
df_yearb4b5 = df_qtdby_qualis_b4b5.query('YEAR == @year_b4b5')
grade_tot_year_b4b5 = df_yearb4b5['PROD_AMOUPESO'].sum()
df_temp = df_grade_tot_year.query('YEAR == @year_b4b5')
grade_tot_year = df_temp['PROD_AMOUPESO'].sum()
# print('Ano ', str(year_b4b5), 'B4 e B5 representam: ',
# str(round(grade_tot_year_b4b5 / grade_tot_year, 2)))
if grade_tot_year_b4b5 / grade_tot_year > 0.2:
print('Para o ano ', str(year_b4b5),
'artigos B4 B5 glosados, maior que 0.2')
df_qtdby_qualis.query(
'YEAR != @year_b4b5 and QUALIS != "B4"', inplace=True)
df_qtdby_qualis.query(
'YEAR != @year_b4b5 and QUALIS != "B5"', inplace=True)
df_qtdby_qualis = df_qtdby_qualis.groupby(
['YEAR'])['PROD_AMOUPESO'].sum() / num_dp
df_qtdby_qualis = df_qtdby_qualis.reset_index()
df_qtdby_qualis.columns = ['YEAR', 'INDPRODART']
indprodart = df_qtdby_qualis['INDPRODART'].mean()
ls_indprodart.append(indprodart)
ls_yini_quad.append(yini)
ls_yfin_quad.append(yfin)
df_indprodart = pd.DataFrame({'QUADRIENIO_INI': ls_yini_quad,
'QUADRIENIO_FIM': ls_yfin_quad,
'INDPRODART': ls_indprodart})
pathfilename = str('./csv_producao/' + 'capesindex_indprodart' '.csv')
df_indprodart.to_csv(pathfilename, index=False)
print(pathfilename, ' gravado com', len(df_indprodart), ' quadrienios')
def capes_distindproddp():
# lendo pesquisadores
df_idlist = readIdList()
num_dp = len(df_idlist)
# lendo periodicos_uniq
df = pd.read_csv('./csv_producao/periodicos_uniq.csv',
header=0, sep=',')
num_period_tot = len(df['QUALIS'])
df.query('QUALIS != "XX"', inplace=True)
df.query('QUALIS != "C "', inplace=True)
df['YEAR'] = df['YEAR'].apply(iint)
num_period_semqualis = num_period_tot - len(df['QUALIS'])
print('Numero de periodicos sem QUALIS = ', num_period_semqualis)
# definindo os quadrienios
year_fquadrien = 2013
ls_quadri = [year_fquadrien]
for i in range(5):
year_fquadrien = year_fquadrien + 4
ls_quadri.append(year_fquadrien)
# print(ls_quadri)
# calculo para cada trienio
ls_yini_quad = []
ls_yfin_quad = []
df_indprodart_full = pd.DataFrame(columns=['QUADRIENIO',
'FULL_NAME',
'INDPRODART',
'CLASSIF'])
for i in range(len(ls_quadri)):
yini = ls_quadri[i]
yfin = ls_quadri[i] + 3
df_qtdby_qualis = df.groupby(['YEAR', 'FULL_NAME',
'QUALIS'])['TITLE'].count().reset_index()
df_qtdby_qualis.columns = ['YEAR', 'FULL_NAME', 'QUALIS', 'AMOUNT']
# print(df_qtdby_qualis)
df_qtdby_qualis.query('YEAR >= @yini and YEAR <= @yfin', inplace=True)
if len(df_qtdby_qualis) < 1:
pass
else:
# print('Quadrienio', yini, ' - ', yfin)
df_qtdby_qualis['PESO'] = df_qtdby_qualis['QUALIS'].apply(
fun_indprodart_classif)
df_qtdby_qualis['PROD_AMOUPESO'] = df_qtdby_qualis['AMOUNT'] * \
df_qtdby_qualis['PESO']
# verificando representatividade B4 e B5 deve ser <= 0.2 por ano
# print(df_qtdby_qualis)
df_grade_tot_year = df_qtdby_qualis.groupby(
['YEAR'])['PROD_AMOUPESO'].sum().reset_index()
df_qtdby_qualis_b4b5 = df_qtdby_qualis.query(
'QUALIS == "B4" or QUALIS == "B5"')
# print(df_grade_tot_year)
# print(df_qtdby_qualis_b4b5)
ls_years_b4b5_uniq = df_qtdby_qualis_b4b5['YEAR'].unique()
for ia in range(len(ls_years_b4b5_uniq)):
year_b4b5 = ls_years_b4b5_uniq[ia]
df_yearb4b5 = df_qtdby_qualis_b4b5.query('YEAR == @year_b4b5')
grade_tot_year_b4b5 = df_yearb4b5['PROD_AMOUPESO'].sum()
df_temp = df_grade_tot_year.query('YEAR == @year_b4b5')
grade_tot_year = df_temp['PROD_AMOUPESO'].sum()
# print('Ano ', str(year_b4b5), 'B4 e B5 representam: ',
# str(round(grade_tot_year_b4b5 / grade_tot_year, 2)))
if grade_tot_year_b4b5 / grade_tot_year > 0.2:
print('Para o ano ', str(year_b4b5),
'artigos B4 B5 glosados, maior que 0.2')
df_qtdby_qualis.query(
'YEAR != @year_b4b5 and QUALIS != "B4"', inplace=True)
df_qtdby_qualis.query(
'YEAR != @year_b4b5 and QUALIS != "B5"', inplace=True)
df_qtdby_qualis = df_qtdby_qualis.groupby(
['YEAR', 'FULL_NAME'])['PROD_AMOUPESO'].sum() / num_dp
df_qtdby_qualis = df_qtdby_qualis.reset_index()
df_qtdby_qualis.columns = ['YEAR', 'FULL_NAME', 'INDPRODART']
df_qtdby_qualis = df_qtdby_qualis.groupby(
['FULL_NAME'])['INDPRODART'].mean().reset_index()
df_qtdby_qualis['CLASSIF'] = df_qtdby_qualis['INDPRODART'].apply(
fun_indori_classif)
quadr = str(str(yini) + '-' + str(yfin))
df_qtdby_qualis['QUADRIENIO'] = np.repeat(
quadr, len(df_qtdby_qualis))
df_qtdby_qualis = df_qtdby_qualis[[
'QUADRIENIO', 'FULL_NAME', 'INDPRODART', 'CLASSIF']]
df_indprodart_full = pd.concat(
[df_indprodart_full, df_qtdby_qualis], axis=0)
pathfilename = str('./csv_producao/' + 'capesindex_distindproddp_doce.csv')
df_indprodart_full.to_csv(pathfilename, index=False)
print(pathfilename, ' gravado com', len(
df_indprodart_full), ' pesquisadores para todos os quadrienios')
# qd = df_indprodart_full['QUADRIENIO'].unique()
df_distindproddp = df_indprodart_full.groupby(['QUADRIENIO', 'CLASSIF'])[
'FULL_NAME'].count().reset_index()
df_distindproddp.columns = ['QUADRIENIO', 'CLASSIF', 'COUNT']
df_distindproddp['DISTINDPRODDP'] = 100 * \
df_distindproddp['COUNT'] / num_dp
df_distindproddp.query('CLASSIF != "FRACO" and CLASSIF != \
"DEFICIENTE"', inplace=True)
df_distindproddp = df_distindproddp.groupby(
['QUADRIENIO'])['DISTINDPRODDP'].sum().reset_index()
pathfilename = str('./csv_producao/' + 'capesindex_distindproddp' '.csv')
df_distindproddp.to_csv(pathfilename, index=False)
print(pathfilename, ' gravado com', len(
df_distindproddp), ' quadrienios')
def capes_indautdis():
# nome ppg
# config_file = open('./config.txt', 'r')
config_file = open('./config.txt', 'r', encoding='utf-8')
name_ppg = config_file.readlines()[8].split(':')[1]
name_ppg = name_ppg.rstrip('\n')
name_ppg = name_ppg.strip(' ')
name_ppg = fun_uppercase(name_ppg)
config_file.close()
# lendo pesquisadores
df_idlist = readIdList()
num_dp = len(df_idlist)
# lendo orientacoes_all
df = pd.read_csv('./csv_producao/orientacoes_all.csv',
header=0, sep=',')
df = df.query('NATURE == "Dissertação de mestrado" \
or NATURE == "Tese de doutorado"')
df = df.query('TYPE != "CO_ORIENTADOR"').reset_index(drop=True)
df['COURSE'] = df['COURSE'].apply(fun_uppercase)
df['STUDENT'] = df['STUDENT'].apply(fun_uppercase)
df['FULL_NAME'] = df['FULL_NAME'].apply(fun_uppercase)
df = df.query('COURSE == @name_ppg')
# identificando os ppg dos pesquisadores
ls_ppgs = df['COURSE'].unique()
ls_ppgs.sort()
ls_ppgs = ", ".join(ls_ppgs)
# print('------------------------------------------------------------')
# print("PPGs listados nos curriculos dos pesquisadores: ", ls_ppgs)
# print('------------------------------------------------------------')
# print("PPG a ser avaliado: ", name_ppg)
# print('------------------------------------------------------------')
# definindo os quadrienios
year_fquadrien = 2013
ls_quadri = [year_fquadrien]
df_indautdisc_all = pd.DataFrame(
columns=['QUADRIENIO', 'STUDENT', 'DOCENTE', 'TYPE', 'AMOUNT'])
for i in range(5):
year_fquadrien = year_fquadrien + 4
ls_quadri.append(year_fquadrien)
# print(ls_quadri)
# calculo para cada trienio
for i in range(len(ls_quadri)):
ls_yini_quad = []
ls_yfin_quad = []
ls_disc = []
ls_doce = []
ls_disc_type_prod = []
ls_disc_amount_prod_period = []
yini = ls_quadri[i] # egressos ate 5 anos
yfin = ls_quadri[i] + 3
# print('Quadrienio', yini, ' - ', yfin)
df_disc_quadri = df.query('YEAR >= @yini+3-4 and YEAR <= @yfin+3')
df_disc_quadri
if len(df_disc_quadri) < 1:
print("sem orientacoes para o periodo")
pass
else:
ls_disc_period = list(df_disc_quadri['STUDENT'])
ls_doce_period = list(df_disc_quadri['FULL_NAME'])
df_period_all = pd.read_csv('./csv_producao/periodicos_all.csv',
header=0, sep=',')
df_period_all['AUTHOR'] = df_period_all['AUTHOR'].apply(
fun_uppercase)
for ia in range(len(ls_disc_period)):
period_count = 0
for ib in range(len(df_period_all)):
# print(ia.upper(), '---', df_period_all.iloc[ib, 7])
zdis = ls_disc_period[ia].split(' ')[-1]
zdoc = ls_doce_period[ia].split(' ')[-1]
zaut = df_period_all['AUTHOR'].iloc[ib]
if zdis in zaut and zdoc in zaut:
period_count += 1
disc_type_prod = 'periodico'
quadr = str(str(yini) + '-' + str(yfin))
ls_disc.append(ls_disc_period[ia])
ls_doce.append(ls_doce_period[ia])
ls_disc_type_prod.append(disc_type_prod)
ls_disc_amount_prod_period.append(period_count)
ls_quad = np.repeat(quadr, len(ls_disc))
df_indautdisc = pd.DataFrame(list(zip(ls_quad,
ls_disc,
ls_doce,
ls_disc_type_prod,
ls_disc_amount_prod_period)),
columns=['QUADRIENIO', 'STUDENT', 'DOCENTE', 'TYPE', 'AMOUNT'])
df_indautdisc_all = pd.concat([df_indautdisc_all,
df_indautdisc])
ls_indautdisc_quad = []
ls_indautdisc = []
ls_indis = []
quad_indautdisc = df_indautdisc_all['QUADRIENIO'].unique()
for i in range(len(quad_indautdisc)):
q = quad_indautdisc[i]
df_d = df_indautdisc_all.query('QUADRIENIO == @q')
disc_zero = len(df_d.query('AMOUNT == 0'))
E = (len(df_d) - disc_zero)
F = len(df_d)
G = df_d['AMOUNT'].sum()
indaut = E / F
indis = G / F
# print(indaut, '--', indis, '--', disc_zero, F)
ls_indautdisc_quad.append(q)
ls_indautdisc.append(indaut)
ls_indis.append(indis)
df_indiscente = pd.DataFrame(list(zip(ls_indautdisc_quad,
ls_indautdisc,
ls_indis)),
columns=['QUADRIENIO', 'INDOUT',
'INDIS'])
pathfilename = str('./csv_producao/' + 'capesindex_indautdis' '.csv')
df_indiscente.to_csv(pathfilename, index=False)
print(pathfilename, ' gravado com', len(df_indiscente), ' quadrienios')
def capes_indori():
# nome ppg
# config_file = open('./config.txt', 'r')
config_file = open('./config.txt', 'r', encoding='utf-8')
name_ppg = config_file.readlines()[8].split(':')[1]
name_ppg = name_ppg.rstrip('\n')
name_ppg = name_ppg.strip(' ')
name_ppg = fun_uppercase(name_ppg)
config_file.close()
# lendo pesquisadores
df_idlist = readIdList()
num_dp = len(df_idlist)
# lendo orientacoes_all
df = pd.read_csv('./csv_producao/orientacoes_all.csv',
header=0, sep=',')
df = df.query('NATURE == "Dissertação de mestrado" \
or NATURE == "Tese de doutorado"')
df = df.query('TYPE != "CO_ORIENTADOR"').reset_index(drop=True)
df['COURSE'] = df['COURSE'].apply(fun_uppercase)
df = df.query('COURSE == @name_ppg')
# identificando os ppg dos pesquisadores
ls_ppgs = df['COURSE'].unique()
ls_ppgs.sort()
ls_ppgs = ", ".join(ls_ppgs)
# print('------------------------------------------------------------')
# print("PPGs listados nos curriculos dos pesquisadores: ", ls_ppgs)
# print('------------------------------------------------------------')
# print("PPG a ser avaliado: ", name_ppg)
# print('------------------------------------------------------------')
# definindo os quadrienios
year_fquadrien = 2013
ls_quadri = [year_fquadrien]
for i in range(5):
year_fquadrien = year_fquadrien + 4
ls_quadri.append(year_fquadrien)
# print(ls_quadri)
# calculo para cada trienio
ls_yini_quad = []
ls_yfin_quad = []
ls_indori = []
for i in range(len(ls_quadri)):
yini = ls_quadri[i]
yfin = ls_quadri[i] + 3
# print('Quadrienio', yini, ' - ', yfin)
df_qtdby_yradv = df.groupby(['YEAR', 'NATURE'])[
'STUDENT'].count().reset_index()
# print(df_qtdby_yradv)
df_qtdby_yradv.query('YEAR >= @yini and YEAR <= @yfin', inplace=True)
if len(df_qtdby_yradv) < 1:
pass
else:
df_qtdby_yradv['PESO_DEF'] = df_qtdby_yradv['NATURE'].apply(
fun_peso_defesa)
df_qtdby_yradv['PROD_STUPESO'] = (df_qtdby_yradv['STUDENT'] *
df_qtdby_yradv['PESO_DEF'])
df_qtdby_yradv = df_qtdby_yradv.groupby(
['YEAR'])['PROD_STUPESO'].sum() / num_dp
df_qtdby_yradv = df_qtdby_yradv.reset_index()
df_qtdby_yradv.columns = ['YEAR', 'INDORI']
indori_quad = df_qtdby_yradv['INDORI'].mean()
ls_yini_quad.append(yini)
ls_yfin_quad.append(yfin)
ls_indori.append(round(indori_quad, 3))
# print(df_qtdby_yradv)
# print(indori_quad)
df_indori = pd.DataFrame({'QUADRIENIO_INI': ls_yini_quad,
'QUADRIENIO_FIM': ls_yfin_quad,
'INDORI': ls_indori})
df_indori['INDORI_CLASSIFICACAO'] = df_indori['INDORI'].apply(
fun_indori_classif)
pathfilename = str('./csv_producao/' + 'capesindex_indori' '.csv')
df_indori.to_csv(pathfilename, index=False)
print(pathfilename, ' gravado com', len(df_indori), ' quadrienios')
