Parametrizo timeout

pydatajson/constants.py

@@ -0,0 +1 @@
+REQUESTS_TIMEOUT = 30

pydatajson/core.py

@@ -25,7 +25,7 @@
 from pydatajson.response_formatters import format_response
 from pydatajson.validation import Validator, \
     DEFAULT_CATALOG_SCHEMA_FILENAME, ABSOLUTE_SCHEMA_DIR
-from . import documentation
+from . import documentation, constants
 from . import helpers
 from . import indicators
 from . import readers
@@ -53,7 +53,8 @@ class DataJson(dict):
 
     def __init__(self, catalog=None, schema_filename=None, schema_dir=None,
                  default_values=None, catalog_format=None,
-                 validator_class=Validator, verify_ssl=False):
+                 validator_class=Validator, verify_ssl=False,
+                 requests_timeout=constants.REQUESTS_TIMEOUT):
         """Lee un catálogo y crea un objeto con funciones para manipularlo.
 
         Salvo que se indique lo contrario, se utiliza como default el schema
@@ -79,15 +80,16 @@ def __init__(self, catalog=None, schema_filename=None, schema_dir=None,
                     }
         """
         self.verify_ssl = verify_ssl
-
+        self.requests_timeout = requests_timeout
         # se construye el objeto DataJson con la interfaz de un dicconario
         if catalog:
 
             # lee representaciones de un catálogo hacia un diccionario
             catalog = readers.read_catalog(catalog,
                                            default_values=default_values,
                                            catalog_format=catalog_format,
-                                           verify=self.verify_ssl)
+                                           verify=self.verify_ssl,
+                                           timeout=self.requests_timeout)
 
             # copia todos los atributos del diccionario hacia el objeto
             for key, value in iteritems(catalog):
@@ -1097,7 +1099,9 @@ def make_catalogs_backup(self, catalogs=None,
         pass
 
     def _read_catalog(self, catalog):
-        return readers.read_catalog(catalog, verify=self.verify_ssl)
+        return readers.read_catalog(catalog,
+                                    verify=self.verify_ssl,
+                                    timeout=self.requests_timeout)
 
 
 def main():

pydatajson/readers.py

@@ -27,7 +27,7 @@
 
 import pydatajson
 from . import custom_exceptions as ce
-from . import helpers
+from . import helpers, constants
 from .ckan_reader import read_ckan_catalog
 
 import urllib3
@@ -53,7 +53,7 @@ def read_catalog_obj(catalog):
 
 
 def read_catalog(catalog, default_values=None, catalog_format=None,
-                 verify=False):
+                 verify=False, timeout=constants.REQUESTS_TIMEOUT):
     """Toma una representación cualquiera de un catálogo, y devuelve su
     representación interna (un diccionario de Python con su metadata.)
 
@@ -87,13 +87,17 @@ def read_catalog(catalog, default_values=None, catalog_format=None,
             catalog_format = catalog_format or suffix
         if catalog_format == "xlsx":
             try:
-                catalog_dict = read_xlsx_catalog(catalog, verify=verify)
+                catalog_dict = read_xlsx_catalog(catalog,
+                                                 verify=verify,
+                                                 timeout=timeout)
             except openpyxl_exceptions + \
                     (ValueError, AssertionError, IOError, BadZipfile) as e:
                 raise ce.NonParseableCatalog(catalog, str(e))
         elif catalog_format == "json":
             try:
-                catalog_dict = read_json(catalog, verify=verify)
+                catalog_dict = read_json(catalog,
+                                         verify=verify,
+                                         timeout=timeout)
             except(ValueError, TypeError, IOError) as e:
                 raise ce.NonParseableCatalog(catalog, str(e))
         elif catalog_format == "ckan":
@@ -183,7 +187,8 @@ def _set_default_value(dict_obj, keys, value):
             variable[keys[-1]] = value
 
 
-def read_json(json_path_or_url, verify=False):
+def read_json(json_path_or_url, verify=False,
+              timeout=constants.REQUESTS_TIMEOUT):
     """Toma el path a un JSON y devuelve el diccionario que representa.
 
     Se asume que el parámetro es una URL si comienza con 'http' o 'https', o
@@ -201,7 +206,7 @@ def read_json(json_path_or_url, verify=False):
 
     parsed_url = urlparse(json_path_or_url)
     if parsed_url.scheme in ["http", "https"]:
-        res = requests.get(json_path_or_url, verify=verify)
+        res = requests.get(json_path_or_url, verify=verify, timeout=timeout)
         json_dict = json.loads(res.content, encoding='utf-8')
 
     else:
@@ -219,7 +224,8 @@ def read_json(json_path_or_url, verify=False):
     return json_dict
 
 
-def read_xlsx_catalog(xlsx_path_or_url, logger=None, verify=False):
+def read_xlsx_catalog(xlsx_path_or_url, logger=None, verify=False,
+                      timeout=constants.REQUESTS_TIMEOUT):
     """Toma el path a un catálogo en formato XLSX y devuelve el diccionario
     que representa.
 
@@ -239,7 +245,7 @@ def read_xlsx_catalog(xlsx_path_or_url, logger=None, verify=False):
 
     parsed_url = urlparse(xlsx_path_or_url)
     if parsed_url.scheme in ["http", "https"]:
-        res = requests.get(xlsx_path_or_url, verify=verify)
+        res = requests.get(xlsx_path_or_url, verify=verify, timeout=timeout)
         tmpfilename = ".tmpfile.xlsx"
         with io.open(tmpfilename, 'wb') as tmpfile:
             tmpfile.write(res.content)