def hash_control(self):
"""This method will create a table that keeps control about all the
system uploaded data
"""
import hashlib
from datetime import datetime
import shutil
# Generate hash with file content
h = hashlib.md5()
f = open(self.response, 'r')
h.update(f.read())
# Copy file to repository
session = model.Session
#metadata = model.metadata
# Create table if it doesn't exists
setup_model()
# First check if hash is already in database
results = session.query(DataRepository.hash).filter_by(hash=h.hexdigest()).all()
#self.log(results)
if len(results) > 0:
#log.error('This file %s has the same hash of a file already in\
# database. Aborting' % self.response)
self.log( 'This file %s has the same hash of a file already in\
database. Aborting' % self.response)
os.remove(self.response)
return True
# Today's date
file_date = datetime.today()
# Filename hash to store
filename, extension = os.path.splitext(os.path.basename(self.response))
h2 = hashlib.md5()
h2.update(file_date.__str__() + filename)
filename = h2.hexdigest() + extension
# Now add full repository path to filename
filename2 = os.path.join(self.repository,filename)
# Now insert data and copy file to repository
#log.warning('Inserting file %s in repository' % self.response)
self.log('Inserting file %s in repository' % self.response)
# Copy file to repository
shutil.copy2(self.response,filename2)
# insert info in database
repository = DataRepository(hash=h.hexdigest(), creation_date=file_date.today(), original_file = filename2, package_file=self.response)
session.add(repository)
session.commit()
#log.warning('File inserted')
self.log('File inserted')
# Remove other file
os.remove(self.response)
self.response = filename2
return False
def log_error(self,file_dict):
"""
Keep a repository of import errors in repository
"""
import hashlib
from datetime import datetime
import shutil
# Today's date
file_date = datetime.today()
# Generate hash with file content
h = hashlib.md5()
f = open(file_dict['tmpfile'], 'r')
h.update(f.read() + file_date.__str__())
f.close()
# Copy file to repository
session = model.Session
# Create table if it doesn't exists
setup_model()
# First check if hash is already in database
results = session.query(ErrorRepository.hash).filter_by(hash=h.hexdigest()).all()
if len(results) > 0:
self.log( 'This file %s has the same hash of a file already in\
database. Aborting' % file_dict['filename'])
os.remove(file_dict['tmpfile'])
return
# Filename hash to store
filename3, extension = os.path.splitext(os.path.basename(file_dict['filename']))
filename3 = file_date.__str__() + '-' + filename3 + extension
# Now add full repository path to filename
filename2 = os.path.join(self.repository,os.path.join('import_errors',filename3.replace(' ', '-')))
# Now insert data and copy file to repository
#log.error('Error parsing file %s. Inserting in repository' % file_dict['filename'])
self.log('Error in file %s. Inserting in repository with message\n %s' % (file_dict['filename'],file_dict.get('errmsg')))
# Create base dir if it doesn't exist
if not os.path.exists(os.path.join(self.repository,'import_errors')):
os.mkdir(os.path.join(self.repository,'import_errors'), 0770)
# Copy file to repository
shutil.copy2(file_dict['tmpfile'],filename2)
# insert info in database
repository = ErrorRepository(
hash=h.hexdigest(),
creation_date=file_date.today(),
original_file=filename2,
errmsg=file_dict.get('errmsg'),
error_type=file_dict.get('error_type'),
package_file=file_dict.get('package_file')
)
session.add(repository)
session.commit()
#log.warning('File inserted')
self.log('File inserted')
# Remove other file
os.remove(file_dict['tmpfile'])
class lightbaseAdminController(AdminController):
"""
Add controller to log page
"""
def log(self):
import sqlalchemy
from ckan import plugins, model
from ckanext.datadaemon.model import setup as setup_model
from ckanext.datadaemon.model import ErrorRepository
# Adding pagination
q = c.q = request.params.get('q',
u'') # unicode format (decoded from utf8)
try:
page = int(request.params.get('page', 1))
except ValueError, e:
abort(400, ('"page" parameter must be an integer'))
limit = 20
# most search operations should reset the page counter:
params_nopage = [(k, v) for k, v in request.params.items()
if k != 'page']
def search_url(params):
url = '/ckan-admin/log'
params = [(k, v.encode('utf-8') if isinstance(v, basestring) else str(v)) \
for k, v in params]
return url + u'?' + urlencode(params)
def drill_down_url(**by):
params = list(params_nopage)
params.extend(by.items())
return search_url(set(params))
c.drill_down_url = drill_down_url
def remove_field(key, value):
params = list(params_nopage)
params.remove((key, value))
return search_url(params)
c.remove_field = remove_field
def pager_url(q=None, page=None):
params = list(params_nopage)
params.append(('page', page))
return search_url(params)
params = request.params
setup_model()
session = ckan.model.Session
# Variáveis de entrada:
# data: Postagem. Valores: 24, 168, 720
# data_inicio
# data_final
# tipo: Tipo de Error. Valores: None, ParsingError, FileRetrievalError, FileCollectionError
# Construir uma consulta com base nas variáveis de entrada
# 1 - Criar consulta que traz todos os valores
query = "SELECT e.* FROM dt_errors e WHERE 1 = 1"
# 2 - Filtrar pela variável data. Ou eu filtro pelo período ou por data de início
# e data de fim
value_inicio = ""
value_final = ""
if params.get('data'):
query = query + "AND creation_date >= (now() - interval '%s hours')" % params.get(
'data')
else:
if params.get('data_inicio'):
query = query + "AND creation_date >= '%s'" % params.get(
'data_inicio')
value_inicio = params.get('data_inicio')
if params.get('data_final'):
query = query + "AND creation_date <= '%s'" % params.get(
'data_final')
value_final = params.get('data_final')
# 3 - Filtrar por tipo de erro
if params.get('tipo') == 'None':
query = query + "AND error_type is NULL "
elif params.get('tipo'):
query = query + "AND error_type = '%s'" % params.get('tipo')
# 4 - Adicionar paginação à consulta
query_pagination = query + "ORDER BY creation_date DESC LIMIT %s OFFSET %s" % (
limit * page, (limit * page) - limit)
error_list = session.query(ErrorRepository).from_statement(
query_pagination).all()
error_list2 = session.query(ErrorRepository).from_statement(
query).all()
error_list3 = session.query(
ErrorRepository.original_file).from_statement(
query_pagination).all()
from os.path import basename
retorno = list()
for lista in error_list:
lista.original_file = basename(lista.original_file)
retorno.append(lista)
tipos_de_erros = session.query(
ErrorRepository.error_type).distinct().all()
data = params.get('data')
tipo = params.get('tipo')
x = {
'valor': retorno,
'valor2': tipos_de_erros,
'valor3': value_inicio,
'valor4': value_final,
'valor5': data,
'valor6': tipo
}
c.page = h.Page(collection=error_list2,
page=page,
url=pager_url,
item_count=len(error_list2),
items_per_page=limit)
#--------------Download do arquivo-------------
from paste.fileapp import FileApp
import mimetypes
if params.get('arquivo'):
filepath = session.query(ErrorRepository.original_file).filter(
ErrorRepository.hash == params.get('arquivo')).all()
content_type = mimetypes.guess_type(str(filepath[0][0]))
if content_type:
headers = [
('Content-Disposition',
'attachment; filename=\"' + str(filepath[0][0]) + '\"'),
('Content-Type', '\'' + str(content_type[0]) + '\'')
]
else:
headers = [
('Content-Disposition',
'attachment; filename=\"' + str(filepath[0][0]) + '\"'),
('Content-Type', 'aplication/octet-stream')
]
fapp = FileApp(str(filepath[0][0]), headers)
return fapp(request.environ, self.start_response)
else:
return render('admin/log.html', extra_vars=x)
