---json
{
    "aliases": [
        { "path": ":docs:resource:script" }
    ],
    "page_id": "txy3nxmj1horoj8tgrry7"
}
---
====== Data Resource - Runtime Data Resource ======


===== About =====
A ''runtime  data resource'' is a special type of data resource that represents:
  * a [[docs:resource:resource|data resource]] known as the ''executable resource''
  * that should be executed against a [[docs:connection:connection|connection]]




They are called [[docs:resource:content#runtime|runtime]] [[resource|data resources]] because these data resources are created at and by the ''execution''.
===== Example =====

  * A `create.sql` [[:docs:resource:sql_ddl|DDL file]] located in the [[:docs:connection:cd|current directory]] that should be executed against [[:docs:system:sqlite:sqlite|sqlite]]
<code>
(create.sql@cd)@sqlite
</code>
  * A `script.sh` Shell [[:docs:resource:script|script]] located in the [[:docs:connection:cd|current directory]] that should be executed at the [[:docs:connection:tmp|temporary runtime directory]]
<code>
(script.sh@cd)@tmp
</code>






===== Type =====

^ Execution Locality ^ Description ^ List ^ 
| [[request|Remote request]] | Request/Response to a [[:docs:service:service|service]] | [[:docs:resource:sql_request|SQL Request]] - database request \\ [[:docs:system:http|HTTP request]] - web server request |
| [[:docs:system:local|Local]] | Operating System Command | [[command#script|Script command]] - Textual computer language file \\ [[command#binary|Binary command]] - Language Interpreter, Compiled Machine code instructions |



===== Data Uri =====

==== Default Execution ====

A executable is represented with the following [[:docs:resource:data_uri|data uri]]
<code bash>
(executablePath@executableConnection)@executionConnection
</code>
where:
   * ''executablePath@executableConnection'' is the [[:docs:resource:data_uri|data uri]] of the ''executable resource''. ie:
     * the resource containing the code
     * or an executable binary
   * ''executionConnection'' is the connection where the ''execution resource'' is executed.


==== Nested Execution ====

We even support deep nested execution resource when the execution of an executable returns a another executable.

<code javascript>
((executableResourcePath@executionResourceConnection)@executionResourceConnection)@executionConnection
</code>

For instance:
  * Execution of a SQL file located in a online zip archive against MySQL: [[:howto:mysql:sample_schema|]]
<code javascript>
((archive/world-sql--archive-entry.yml@howto)@tmp)@mysql
</code>
  * [[docs:resource:sql_request|Execution of SQL scripts]] against ''@postgres''
      * where the scripts are retrieved from a [[:docs:resource:sql_select|select statement]] where:
        * the ''select'' is stored in the ''select_statement.sql'' [[docs:resource:sql_file|sql file]] located in the [[:docs:connection:cd|current directory]]
        * executed against ''@sqlite''
<code javascript>
((select_statement.sql@cd)@sqlite)@postgres
</code>







===== Runtime Data Selectors =====
''Runtime resources'' can be selected with a [[docs:flow:runtime_selector|executable selector]].

===== Execution =====


''runtime resources'' are automatically executed at access/runtime.

For instance:
  * a [[docs:resource:sql_select|sql select]] would be automatically executed and would return a result set.
  * a [[:docs:resource:script|script]] would be automatically executed and return its standard output (as default)


Every [[:docs:op:op|operation]] will execute them when accessing them
  * [[:docs:op:print|print]]
  * [[:docs:op:copy|copy]]
  * ...

ie:
  * If you print it ''2'' times, you will execute it ''2'' times.
  * If you print and copy it, you will execute it ''2'' times.





Example of execution:
  * with the [[:docs:tabul:data:print|print command]] for a [[:docs:resource:sql_select|SQL Query]]
<code bash>
tabul data print '(select.sql@cd)@sqlite'
# The quotes are mandatory in bash because parenthesis are a bash token (ie subshell)
</code>
  * with the [[:docs:tabul:data:copy|copy command]] for a series of [[:docs:resource:sql_dml|SQL DML statement]] where the execution result is stored in the ''result.log'' file in the [[:docs:connection:tmp|temporary directory]].
<code bash>
tabul data copy '(create_dml.sql@cd)@sqlite'  result.log@tmp
# The quotes are mandatory in bash because parenthesis are a bash token (ie subshell)
</code>

===== Execute Operation =====

With the [[:docs:op:execute|execute operation]], you can perform execution in batch and get back the exit status.


Example: 
To execute:
  * all [[:docs:resource:sql_file|sql files]] [[:howto:tabul:glob|beginning with query]] 
  * located in the [[:docs:connection:cd|current directory (cd)]] 
  * against the [[:howto:sqlite:howto_connection|sqlite howto connection]]
you would execute this [[:docs:tabul:data:execute|tabul data execute command]]
<code bash>
tabul data execute '(query*.sql@cd)@sqlite'
# The quotes are mandatory in bash because parenthesis are a bash token (ie subshell)
</code>

===== Attributes =====

==== Logical Name ====
The default [[:docs:resource:logical_name|logical name]] is the logical name of the ''execution resource''.


==== Executable ====

The executable is the file found in the first part of the [[#data uri|runtime data uri]].

For instance, in this [[#data uri|runtime data uri]]
<code>
(query.sql@cd)@sqlite
</code>
the executable is: ''query.sql@cd'' 

''query.sql@cd''  defines:
  * the ''query.sql'' [[:docs:resource:sql_file|sql file]]
  * in the [[:docs:connection:cd|current directory (cd)]]
as ''executable'' 
  
===== FAQ =====

==== Does a runtime resource exists ====

^ Runtime created from ^ Exists ^
| A data uri | the ''executable'' exists |
| Dynamically with a pipeline operation | true |