kevingurney opened a new pull request, #45: URL: https://github.com/apache/arrow-experiments/pull/45
### Overview Now that the MATLAB interface supports [reading the Arrow IPC Stream format from a byte array (`uint8`)](#45274), we can now easily consume Arrow IPC Streams over HTTP and load them into an `arrow.tabular.Table` in memory. This PR adds a simple HTTP `GET` MATLAB Client example. This was tested against the Python server. We would be happy to do more testing as needed. @ianmcook - would you like to take a look at this PR? Thank you! ### Example Output from `client.m` ```shell $ matlab -batch client Reading Arrow IPC Stream from http://localhost:8008... DONE ✔ --------------- Results --------------- Time (s): 6.28 Num Bytes: 3207031800 Num Rows:100000000 Num Columns:4 ``` ### Notes 1. We included fairly detailed comments in the example code for completeness. However, if we prefer to keep the `client.m` script shorter, we are happy to remove these. 2. The average time it takes to run the client in MATLAB (not including MATLAB startup time) is about 5.5-6s. This example is particularly well optimized but does work as expected. the `README.md` instructs users to run `client` in `-batch` mode which can result in some additional delay due to MATLAB startup. The script should run just fine if MATLAB is started up fully and then the script is executed, as well. We just thought it would be easier to directly run the script in `-batch` mode from the command line. 3. We included some diagnostic output such as number of bytes received and number of rows. However, if there is more information that we want to include, we are happy to add that. 4. Thanks @sgilmore10 for your help with this pull request! -- 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: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
