You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@orc.apache.org by GitBox <gi...@apache.org> on 2022/08/14 17:52:45 UTC

[GitHub] [orc] cubter opened a new issue, #1220: How to set list's offsets correctly

cubter opened a new issue, #1220:
URL: https://github.com/apache/orc/issues/1220

   Greetings,
   I'm learning to work with ORC in C++, and I think I'm stuck and don't quite understand how to set array's offsets. Precisely, the following code, when executed, produces the following exception: "**Caught exception in test-file.orc: bad read in nextBuffer**"
   "):
   ```c++
   void write_orc()
   {
       using namespace orc;
   
       ORC_UNIQUE_PTR<OutputStream> outStream = writeLocalFile("test-file.orc");
       ORC_UNIQUE_PTR<Type> schema(
           Type::buildTypeFromString("struct<id:int,list1:array<string>>"));
       WriterOptions options;
       ORC_UNIQUE_PTR<Writer> writer = createWriter(*schema, outStream.get(), options);
   
       std::unique_ptr<Writer> writer = createWriter(*type, stream.get(), options);
   
       uint64_t batch_size = 1024, row_count = 2048;
   
       std::unique_ptr<ColumnVectorBatch> batch =
           writer->createRowBatch(row_count);
       StructVectorBatch &root_batch =
           dynamic_cast<StructVectorBatch &>(*batch.get());
       LongVectorBatch &id_batch =
           dynamic_cast<LongVectorBatch &>(*struct_batch.fields[0]);
       ListVectorBatch &list_batch =
           dynamic_cast<ListVectorBatch &>(*struct_batch.fields[1]);
       StringVectorBatch &str_batch =
           dynamic_cast<StringVectorBatch &>(*list_batch.elements.get());
       
       std::vector<std::string> vs{"str1", "str2"};
   
       char **data         = str_batch.data.data();
       int64_t *offsets    = list_batch.offsets.data();
       uint64_t offset     = 0, rows = 0;
       for (size_t i = 0; i < row_count; ++i) {
           offsets[rows] = static_cast<int64_t>(offset);
   
           id_batch.data[rows] = articles[i]->get_id();
   
           for (const auto &s : vs)
           {
               data[offset] = &s[0];
               str_batch.length[offset++] = s.size();
           }
   
           rows++;
           if (rows == batch_size) 
           {
               root_batch.numElements = rows;
               id_batch.numElements   = rows;
               list_batch.numElements = rows;
   
               writer->add(*batch);
               rows = 0;
               offset = 0;
           }
       }
   
       if (rows != 0) 
       {
           root_batch.numElements = rows;
           id_batch.numElements   = rows;
           list_batch.numElements = rows;
   
           writer->add(*batch);
           rows = 0;
           offset = 0;
       }
   
       writer->close();
   }
   ```
   
   My question is: what exactly am I doing wrong when settings list's offsets?


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: issues-unsubscribe@orc.apache.org.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


[GitHub] [orc] dongjoon-hyun commented on issue #1220: How to set list's offsets correctly

Posted by GitBox <gi...@apache.org>.
dongjoon-hyun commented on issue #1220:
URL: https://github.com/apache/orc/issues/1220#issuecomment-1220985376

   Given the above comment, I close this issue.


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: issues-unsubscribe@orc.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


[GitHub] [orc] coderex2522 commented on issue #1220: How to set list's offsets correctly

Posted by GitBox <gi...@apache.org>.
coderex2522 commented on issue #1220:
URL: https://github.com/apache/orc/issues/1220#issuecomment-1220202314

   For List type, the length of the offset array is one greater than the number of rows. Because the number of elements in the current row is confirmed by offsets[row] - offsets[row - 1](row number>= 1). So your example code needs to add the following two lines.
   void write_orc()
   {
       using namespace orc;
   
       ORC_UNIQUE_PTR<OutputStream> outStream = writeLocalFile("test-file.orc");
       ORC_UNIQUE_PTR<Type> schema(
           Type::buildTypeFromString("struct<id:int,list1:array<string>>"));
       WriterOptions options;
       ORC_UNIQUE_PTR<Writer> writer = createWriter(*schema, outStream.get(), options);
   
       std::unique_ptr<Writer> writer = createWriter(*type, stream.get(), options);
   
       uint64_t batch_size = 1024, row_count = 2048;
   
       std::unique_ptr<ColumnVectorBatch> batch =
           writer->createRowBatch(row_count);
       StructVectorBatch &root_batch =
           dynamic_cast<StructVectorBatch &>(*batch.get());
       LongVectorBatch &id_batch =
           dynamic_cast<LongVectorBatch &>(*struct_batch.fields[0]);
       ListVectorBatch &list_batch =
           dynamic_cast<ListVectorBatch &>(*struct_batch.fields[1]);
       StringVectorBatch &str_batch =
           dynamic_cast<StringVectorBatch &>(*list_batch.elements.get());
       
       std::vector<std::string> vs{"str1", "str2"};
   
       char **data         = str_batch.data.data();
       int64_t *offsets    = list_batch.offsets.data();
       uint64_t offset     = 0, rows = 0;
       for (size_t i = 0; i < row_count; ++i) {
           offsets[rows] = static_cast<int64_t>(offset);
   
           id_batch.data[rows] = articles[i]->get_id();
   
           for (auto &s : vs)
           {
               data[offset] = &s[0];
               str_batch.length[offset++] = s.size();
           }
   
           rows++;
           if (rows == batch_size) 
           {
               offsets[rows] = offset; // new line
               root_batch.numElements = rows;
               id_batch.numElements   = rows;
               list_batch.numElements = rows;
   
               writer->add(*batch);
               rows = 0;
               offset = 0;
           }
       }
   
       if (rows != 0) 
       {
           offsets[rows] = offset; // new line
           root_batch.numElements = rows;
           id_batch.numElements   = rows;
           list_batch.numElements = rows;
   
           writer->add(*batch);
           rows = 0;
           offset = 0;
       }
   
       writer->close();
   }


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: issues-unsubscribe@orc.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


[GitHub] [orc] dongjoon-hyun closed issue #1220: How to set list's offsets correctly

Posted by GitBox <gi...@apache.org>.
dongjoon-hyun closed issue #1220: How to set list's offsets correctly
URL: https://github.com/apache/orc/issues/1220


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: issues-unsubscribe@orc.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


[GitHub] [orc] dongjoon-hyun commented on issue #1220: How to set list's offsets correctly

Posted by GitBox <gi...@apache.org>.
dongjoon-hyun commented on issue #1220:
URL: https://github.com/apache/orc/issues/1220#issuecomment-1220477980

   Thank you, @coderex2522 .


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: issues-unsubscribe@orc.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org