Conversation
|
R: @apilloud @melap |
apilloud
left a comment
There was a problem hiding this comment.
Few minor comments, but LGTM.
|
|
||
| ## Overview | ||
|
|
||
| Starting with version 2.6.0, Beam SQL includes an interactive shell, called the Beam SQL shell. The shell allows you to write pipelines as SQL queries without needing the Java SDK. By default, Beam uses the `DirectRunner` to execute the queries as Beam pipelines. |
There was a problem hiding this comment.
nit: Should we make this 2.7.0? These instructions won't work in 2.6.0.
There was a problem hiding this comment.
These instructions should work as of late June. 2.6.0 RC was only cut last week.
There was a problem hiding this comment.
Removed the instructions for :run, replaced with installDist. Works locally
src/documentation/dsls/sql/shell.md
Outdated
|
|
||
| The `CREATE TABLE` statement registers the CSV file as a table in Beam SQL and specifies the table's schema. This statement does not directly create a persistent physical table; it only describes the source/sink to Beam SQL so that you can use the table in the queries that read data and write data. | ||
|
|
||
| _For more information about `CREATE TABLE` syntax and supported table types, see the [CREATE TABLE reference page](https://beam.apache.org/documentation/dsls/sql/create-table/)._ |
src/documentation/dsls/sql/shell.md
Outdated
| ``` | ||
|
|
||
|
|
||
| _For more information about `SELECT` syntax, see the [SELECT reference page](https://beam.apache.org/documentation/dsls/sql/select/)._ |
src/documentation/dsls/sql/shell.md
Outdated
|
|
||
| ## Quickstart | ||
|
|
||
| To use Beam SQL shell, you must first clone the [Beam SDK repository](https://github.com/apache/beam). Then, from the root of the repository clone, execute the Gradle task to run the shell ([source code](https://github.com/apache/beam/blob/master/sdks/java/extensions/sql/shell/build.gradle#L39)) by running the following command: |
There was a problem hiding this comment.
Linking to line numbers in master is a bad idea. Can you use a release tag or commit?
There was a problem hiding this comment.
That will become stale quick as well. Do you think linking just to the file is better?
There was a problem hiding this comment.
Just linking to the file would probably be good. The target audience of the SQL shell probably doesn't care about the code, so just dropping the links might be a good option too.
src/documentation/dsls/sql/shell.md
Outdated
|
|
||
| By default, Beam uses the `DirectRunner` to run the pipeline on the machine where you're executing the commands. If you want to run the pipeline with a different runner, you must perform two steps: | ||
|
|
||
| 1. Make sure the SQL shell includes the desired runner. Add the corresponding project id to the `-Pbeam.sql.shell.bundled` parameter of the Gradle invocation ([source code](https://github.com/apache/beam/blob/master/sdks/java/extensions/sql/shell/build.gradle#L26), [project ids](https://github.com/apache/beam/blob/master/settings.gradle)). For example, use the following command to include Flink runner: |
There was a problem hiding this comment.
Same comment about line numbers in master.
238c7f1 to
c42c415
Compare
|
@melap , 2.6.0 is out, can we merge this? Thank you |
|
@asfgit merge |
Add documentation section for Beam SQL Shell
http://apache-beam-website-pull-requests.storage.googleapis.com/527/documentation/dsls/sql/shell/index.html