Home | History | Annotate | Download | only in gslib 
      1 # -*- coding: utf-8 -*-
      2 # Copyright 2014 Google Inc. All Rights Reserved.
      3 #
      4 # Licensed under the Apache License, Version 2.0 (the "License");
      5 # you may not use this file except in compliance with the License.
      6 # You may obtain a copy of the License at
      7 #
      8 #     http://www.apache.org/licenses/LICENSE-2.0
      9 #
     10 # Unless required by applicable law or agreed to in writing, software
     11 # distributed under the License is distributed on an "AS IS" BASIS,
     12 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
     13 # See the License for the specific language governing permissions and
     14 # limitations under the License.
     15 """Helper for cat and cp streaming download."""
     16 
     17 from __future__ import absolute_import
     18 
     19 import sys
     20 
     21 from gslib.exception import CommandException
     22 from gslib.wildcard_iterator import StorageUrlFromString
     23 
     24 
     25 class CatHelper(object):
     26 
     27   def __init__(self, command_obj):
     28     """Initializes the helper object.
     29 
     30     Args:
     31       command_obj: gsutil command instance of calling command.
     32     """
     33     self.command_obj = command_obj
     34 
     35   def CatUrlStrings(self, url_strings, show_header=False, start_byte=0,
     36                     end_byte=None):
     37     """Prints each of the url strings to stdout.
     38 
     39     Args:
     40       url_strings: String iterable.
     41       show_header: If true, print a header per file.
     42       start_byte: Starting byte of the file to print, used for constructing
     43                   range requests.
     44       end_byte: Ending byte of the file to print; used for constructing range
     45                 requests. If this is negative, the start_byte is ignored and
     46                 and end range is sent over HTTP (such as range: bytes -9)
     47     Returns:
     48       0 on success.
     49 
     50     Raises:
     51       CommandException if no URLs can be found.
     52     """
     53     printed_one = False
     54     # We manipulate the stdout so that all other data other than the Object
     55     # contents go to stderr.
     56     cat_outfd = sys.stdout
     57     sys.stdout = sys.stderr
     58     try:
     59       for url_str in url_strings:
     60         did_some_work = False
     61         # TODO: Get only the needed fields here.
     62         for blr in self.command_obj.WildcardIterator(url_str).IterObjects():
     63           did_some_work = True
     64           if show_header:
     65             if printed_one:
     66               print
     67             print '==> %s <==' % blr
     68             printed_one = True
     69           cat_object = blr.root_object
     70           storage_url = StorageUrlFromString(blr.url_string)
     71           if storage_url.IsCloudUrl():
     72             self.command_obj.gsutil_api.GetObjectMedia(
     73                 cat_object.bucket, cat_object.name, cat_outfd,
     74                 start_byte=start_byte, end_byte=end_byte,
     75                 object_size=cat_object.size, generation=storage_url.generation,
     76                 provider=storage_url.scheme)
     77           else:
     78             cat_outfd.write(open(storage_url.object_name, 'rb').read())
     79         if not did_some_work:
     80           raise CommandException('No URLs matched %s' % url_str)
     81       sys.stdout = cat_outfd
     82     finally:
     83       sys.stdout = cat_outfd
     84 
     85     return 0
     86