[impala] 03/05: Revert "IMPALA-10012: ds_hll_sketch() results ascii codec decoding error fix"

[bo...@apache.org]

This is an automated email from the ASF dual-hosted git repository.

boroknagyz pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/impala.git

commit b7965d8240f3a5cd6972351b7fe1116e4a0835c1
Author: Csaba Ringhofer <cs...@cloudera.com>
AuthorDate: Fri Sep 4 12:27:58 2020 +0000

    Revert "IMPALA-10012: ds_hll_sketch() results ascii codec decoding error fix"
    
    This reverts commit 75146c9138051758176d3d4482f45a9d1056127b.
    
    Change-Id: I57f790389a8c847877999d2b9b8185939b416c07
    Reviewed-on: http://gerrit.cloudera.org:8080/16417
    Reviewed-by: Csaba Ringhofer <cs...@cloudera.com>
    Tested-by: Csaba Ringhofer <cs...@cloudera.com>
---
 shell/impala_client.py | 10 +---------
 shell/impala_shell.py  | 11 +----------
 shell/shell_output.py  | 21 +++------------------
 3 files changed, 5 insertions(+), 37 deletions(-)

diff --git a/shell/impala_client.py b/shell/impala_client.py
index 19a570f..cb7cb39 100755
--- a/shell/impala_client.py
+++ b/shell/impala_client.py
@@ -1095,15 +1095,7 @@ class ImpalaBeeswaxClient(ImpalaClient):
                                         self.fetch_size))
       if rpc_status != RpcStatus.OK:
         raise RPCException()
-
-      def split_row_and_decode_if_needed(row):
-        try:
-          return row.split('\t')
-        except UnicodeDecodeError:
-          return row.decode('utf-8', 'replace').split('\t')
-
-      yield [split_row_and_decode_if_needed(row) for row in result.data]
-
+      yield [row.split('\t') for row in result.data]
       if not result.has_more:
         return
 
diff --git a/shell/impala_shell.py b/shell/impala_shell.py
index 6cb9aa7..8c18408 100755
--- a/shell/impala_shell.py
+++ b/shell/impala_shell.py
@@ -1221,15 +1221,6 @@ class ImpalaShell(cmd.Cmd, object):
     except RPCException as e:
       # could not complete the rpc successfully
       print(e, file=sys.stderr)
-    except UnicodeDecodeError as e:
-      # An error occoured possibly during the fetching.
-      # Depending of which protocol is at use it can come from different places.
-      # Possibly occours because we try to display binary data which contains
-      # undecodable elements.
-      if self.last_query_handle is not None:
-        self.imp_client.close_query(self.last_query_handle)
-      print('UnicodeDecodeError : %s \nPlease check for columns containing binary data '
-          'to find the possible source of the error.' % (e,), file=sys.stderr)
     except QueryStateException as e:
       # an exception occurred while executing the query
       if self.last_query_handle is not None:
@@ -1357,7 +1348,7 @@ class ImpalaShell(cmd.Cmd, object):
     if self.readline and self.readline.get_current_history_length() > 0:
       for index in xrange(1, self.readline.get_current_history_length() + 1):
         cmd = self.readline.get_history_item(index)
-        print('[%d]: %s' % (index, cmd.decode('utf-8', 'replace')), file=sys.stderr)
+        print('[%d]: %s' % (index, cmd), file=sys.stderr)
     else:
       print(READLINE_UNAVAILABLE_ERROR, file=sys.stderr)
 
diff --git a/shell/shell_output.py b/shell/shell_output.py
index 31d91a0..c6c2e99 100644
--- a/shell/shell_output.py
+++ b/shell/shell_output.py
@@ -35,25 +35,11 @@ class PrettyOutputFormatter(object):
 
   def format(self, rows):
     """Returns string containing representation of the table data."""
-
-    def decode_if_needed(row):
-      # Checking if the values in row is decodable. If it is we just give back the row
-      # as it is, if not we generate a new row and give that back instead where the
-      # undecodable parts are swapped out.
-      try:
-        ''.join(str(row))
-        return row
-      except UnicodeDecodeError:
-        new_row = []
-        for entry in row:
-          new_row.append(entry.decode('UTF-8', 'replace'))
-        return new_row
-
     # Clear rows that already exist in the table.
     self.prettytable.clear_rows()
     try:
       for row in rows:
-        self.prettytable.add_row(decode_if_needed(row))
+        self.prettytable.add_row(row)
       return self.prettytable.get_string()
     except Exception as e:
       # beeswax returns each row as a tab separated string. If a string column
@@ -63,8 +49,7 @@ class PrettyOutputFormatter(object):
                    "embedded tabs. Reverting to tab delimited text output")
       print(error_msg, file=sys.stderr)
       print('{0}: {1}'.format(type(e), str(e)), file=sys.stderr)
-
-      return '\n'.join(['\t'.join(decode_if_needed(row)) for row in rows])
+      return '\n'.join(['\t'.join(row) for row in rows])
 
 
 class DelimitedOutputFormatter(object):
@@ -92,7 +77,7 @@ class DelimitedOutputFormatter(object):
                         lineterminator='\n', quoting=csv.QUOTE_MINIMAL)
     for row in rows:
       if sys.version_info.major == 2:
-        row = [val.encode('utf-8', 'replace') for val in row]
+        row = [val.encode('utf-8') for val in row]
       writer.writerow(row)
     rows = temp_buffer.getvalue().rstrip()
     temp_buffer.close()