[Documentation] [TitleIndex] [WordIndex

Note: This tutorial assumes that you have completed the previous tutorials: ROS Tutorials.
(!) Please ask about problems and questions regarding this tutorial on answers.ros.org. Don't forget to include in your question the link to this page, the versions of your OS & ROS, and also add appropriate tags.

Retrieving Files

Description: Demonstrates how to retrieve a file from a URL using the resource_retriever package

Keywords: urdf, resource_retriever, url, curl

Tutorial Level: BEGINNER

Intro

The resource_retriever package lets you download files from any CURL-supported URL, as well as using a special package:// syntax. Here we'll show you how to use it.

Retrieving a File Into Memory

Here we'll show you how to retrieve a file from a URL into memory, which is currently the only supported method of retrieval.

The Code

   1 #include <resource_retriever/retriever.h>
   2 #include <ros/console.h>
   3 
   4 int main(int argc, char** argv)
   5 {
   6   resource_retriever::Retriever r;
   7   resource_retriever::MemoryResource resource;
   8 
   9   try
  10   {
  11      resource = r.get("package://resource_retriever/test/test.txt"); 
  12   }
  13   catch (resource_retriever::Exception& e)
  14   {
  15     ROS_ERROR("Failed to retrieve file: %s", e.what());
  16     return 1;
  17   }
  18 
  19   FILE* f = fopen("out.txt", "w");
  20   fwrite(resource.data.get(), resource.size, 1, f);
  21   fclose(f);
  22   
  23   ROS_INFO("Wrote data from package://resource_retriever/test/test.txt to out.txt");
  24 }

The Code Explained

   1 #include <resource_retriever/retriever.h>
   2 

This is the main include file for resource_retriever. It includes the two main classes, resource_retriver::Retreiver and resource_retriever::MemoryResource.

   6   resource_retriever::Retriever r;
   7   resource_retriever::MemoryResource resource;

Create the resource_retriever::Retriever object, which is the object used to retrieve files. Also create a resource_retriever::MemoryResource object, which is the object type returned from Retriever's get() method.

   9   try
  10   {
  11      resource = r.get("package://resource_retriever/test/test.txt"); 
  12   }
  13   catch (resource_retriever::Exception& e)
  14   {
  15     ROS_ERROR("Failed to retrieve file: %s", e.what());
  16     return 1;
  17   }

Attempt to download the file. Note the package:// syntax, which is ROS-specific. Generic urls like http://www.google.com/ would also work here.

If this fails for any reason, a resource_retriever::Exception will be thrown.

  19   FILE* f = fopen("out.txt", "w");
  20   fwrite(resource.data.get(), resource.size, 1, f);

Write out the downloaded file to disk. MemoryResource contains two members:

MemoryResources can be copied. Once all references to MemoryResource::data have left scope it will automatically be deleted.

Note that we don't do any error checking on the fopen or fwrite here, to simplify the code.


2019-07-13 13:03