Home
LNK APIs
The LNK programming interface includes language-specific methods and objects that provide common functionality across multiple environments.
C API
The C API enables applications to invoke maps and to develop adapters to use with HCL Link.
Custom adapters
The custom adapters interface makes it possible for users to develop adapters with as much functionality as those provided with the product.
Passing data to and from maps
Data is passed between adapters and the engine through stream objects. From the adapter perspective, there is no need to create these stream objects; they are already created when either the Get or Put method is called.
Sending data to a map
Use a Get method to return (send) data to the engine through stream objects. There are a number of different ways to accomplish this task. The method chosen depends on the manner in which the resource returns data to the adapter.

LNK APIs
The LNK programming interface includes language-specific methods and objects that provide common functionality across multiple environments.
- Link programming interface overview
  The Link programming interface is an object-oriented approach that enables applications to invoke HCL Link maps, thus removing reliance on command lines.
- C API
  The C API enables applications to invoke maps and to develop adapters to use with HCL Link.
  - C API interface methods overview
    The map, card, and adapter object properties define the properties that can be set for each object. Use the C API interface methods to GET or SET the adapter object properties.
  - Object methods
  - Loading, running, and controlling maps
    There are various methods you can use to load, run, and control maps when you use APIs with HCL Link.
  - Custom adapters
    The custom adapters interface makes it possible for users to develop adapters with as much functionality as those provided with the product.
    - Developing custom adapters
      The custom adapters interface makes it possible for users to develop adapters with as much functionality as those provided with the product.
    - Adapter registration and the library virtual table
      The first step required to develop an adapter is to add the adapter to the adapters.xml registration file.
    - Initializing the adapter
      InitializeLibrary provides an entry point by which any global structures, critical sections, or lists that are required by the adapter can be initialized.
    - Point to adapter methods
      After your adapter is loaded, you call the virtual table (vtable) structure to obtain pointers to the methods.
    - Adapter registration file
      The adapters.xml registration file contains a single tag for each adapter, including a number of attributes to provide information about the adapter.
    - Vtable adapter library
    - Methods provided by adapter
    - Additional adapter methods
      There are a number of additional methods that are part of the HIP Programming Interface library that are of particular relevance to adapter developers.
    - Passing data to and from maps
      Data is passed between adapters and the engine through stream objects. From the adapter perspective, there is no need to create these stream objects; they are already created when either the Get or Put method is called.
      - Receiving data from a map
        Use the Put method to receive data from a map through stream objects. There are two different ways of accomplishing this. The method chosen depends on the manner in which the data has to be sent to the resource.
      - Sending data to a map
        Use a Get method to return (send) data to the engine through stream objects. There are a number of different ways to accomplish this task. The method chosen depends on the manner in which the resource returns data to the adapter.
    - Threads, connections, and transactions overview
      It is important to understand how threads, connections, and transactions interact.
- Python API
  Use the Python API to create, edit, and run maps and schemas from a Python program. For details, see the documentation installed in install_dir/python/api/hip/docs/html/index.html. Examples of the Python API are installed in install_dir/python/api/examples. Documentation for the Python API examples is available in install_dir/python/api/docs/html/examples.html.
- Java API
  The Java API interfaces enable you to create, edit, and run maps and schemas from a Java™ program.
- Java Tools
  The Java™ Tools provide objects and JNI utilities that you can use to get information about your map instances.

Sending data to a map

Use a Get method to return (send) data to the engine through stream objects. There are a number of different ways to accomplish this task. The method chosen depends on the manner in which the resource returns data to the adapter.

A memory buffer is returned from the resource that can be passed to the engine.
In this scenario, only a single call to mpiStreamWrite is required:
```
mpiStreamWrite (hStream, pData, nSizeOfData);
```
This will create a page, copy the data to the page and append the page to the stream.
The adapter allocates a single buffer that is then populated by the resource.
If the adapter is receiving data into a buffer that it has allocated, the buffer should be allocated through the stream interface as shown in the following example:
```
 /* Create a page */      
mpiStreamNewPage (hStream, &hPage, nSizeOfBuffer);

/* Get the address of the buffer */
mpiStreamPageGetInfo (hPage, &pData, &nSizeOfBuffer);

/* Get the data into the buffer */
MyGetData (pData, &ActualDataSize);

/* Add the page to the stream */ 
mpiStreamWritePageEx (hStream, hPage, nActualDataSize); 
```
This will create a page and append the page to the stream. This is the preferable way to handle this kind of resource because it requires no memory copies or reallocations.
The adapter gets data in pieces from the resource.
This is identical to the above scenario except that it is allocated repeatedly. Each call to mpiStreamWritePage appends a page to the stream.