snowconvert-deploy-tool 0.0.20

Mobilize.Net Database Deploy tool for Snowflake

sc-deploy-db is a cross-platform command line tool for deploying scripts to Snowflake. This tool is designed to help with the deployment of large data warehouses.

A typical challenge when deploying code to snowflake is handling dependendencies. For example if you have a view and it depends on some other views, then your deployment will fail if you do not deploy the dependendencies first.

sc-deploy-db can handle dependendencies between objects. The tool uses a brute-force approach in which if an object fails due to a missing dependendency it will put it a queue and then the deployment of that object it will retried. The tool will keep trying to deploy until it gets to point where no more objects can be deployed.

The tool will also provide deployment logs that will help you to identify and track any deployment issues.

For projects using SnowConvert this deploy tool is highly recommended.

Also a lot of tools have limitations when deploying files that contain several code snippets. A typical error you might get is:

000006 (0A000): Multiple SQL statements in a single API call are not supported; use one API call per statement instead.

This tool has some options that can process the code inside the files, splitting it based on pattern before deployment helping in those situations. Please read more about it the following sections.

However starting on version 0.0.18 the tool can leverate the connector capabilities for running multiple statements. The split functionality has been left for cases where this may still be needed.

Installation

We recommended installing using PYPI

$ pip install snowconvert-deploy-tool --upgrade

Note

If you run this command on MacOS change pip by pip3

You might need to install the python connector for snowflake:

pip install "snowflake-connector-python[pandas]"

Usage

For information about the different parameters or options just run it using the -h option:

$ sc-deploy-db -h

Tool Options

usage: sc-deploy-db [-h] [-A ACCOUNT] [-D DATABASE] [-WH WAREHOUSE] [-R ROLE] [-U USER] [-P PASSWORD] [--AskPassword] [-W WORKSPACE] -I INPATH
                [--activeConn ACTIVECONN] [--authenticator AUTHENTICATOR] [-L LOGPATH] [--SplitBefore SPLITBEFORE] [--SplitAfter SPLITAFTER]
                [--ObjectType [OBJECTTYPE]]

SnowConvertStudio Deployment Script
===================================
This script helps you to deploy a collection of .sql files to a Snowflake Account.
The tool will look for settings like:
- Snowflake Account
- Snowflake Warehouse
- Snowflake Role
- Snowflake Database

If the tool can find a **config_snowsql.ini** file in the current directory or in the **workspace\config_snowsql.ini** location
it will read those parameters from there.
optional arguments:

-h, --help            show this help message and exit
-A ACCOUNT, --Account ACCOUNT
                      Snowflake Account
-D DATABASE, --Database DATABASE
                      Snowflake Database
-S SCHEMA, --Schema SCHEMA
                      Snowflake Initial Schema
-WH WAREHOUSE, --Warehouse WAREHOUSE
                      Snowflake Warehouse
-R ROLE, --Role ROLE  Snowflake Role
-U USER, --User USER  Snowflake User
-P PASSWORD, --Password PASSWORD
                      Password
--AskPassword         If given the tool will prompt for the password
-W WORKSPACE, --Workspace WORKSPACE
                      Path for workspace root. Defaults to current dir
-I INPATH, --InPath INPATH
                      Path for SQL scripts
--activeConn ACTIVECONN
                      When given, it will be used to select connection parameters forn config_snowsql.ini or ~/.snowsql/config
--authenticator AUTHENTICATOR
                      Use the authenticator with you want to use a different authentication mechanism
-L LOGPATH, --LogPath LOGPATH
                      Path for process logs. Defaults to current dir
--SplitBefore SPLITBEFORE
                      Regular expression that can be used to split code in fragments starting **BEFORE** the matching expression
--SplitAfter SPLITAFTER
                      Regular expression that can be used to split code in fragments starting **AFTER** the matching expression
--ObjectType [OBJECTTYPE]
                      Object Type to deploy table,view,procedure,function,macro
--sync-folder-target SYNC_FOLDER_TARGET
                    Target folder where the lastest version of the scripts is kept
--sync-folder-categories SYNC_FOLDER_CATEGORIES
                    It is expected that the workdir will organize code in folders like [table,view,function,macro,procedure]. This
                    parameter is a comma separated list of the categories you would like to sync

This tool assumes :

• that you have a collection of .sql files under a directory.

• that all those .sql files will be deployed to the same database.

• that each file contains only one statement. If your files contain more that one statement then you should use the –SplitBefore or –SplitAfter options.

The tool can also read the connection settings from environment variables.

The following environment variables are recognized by this tool (notice that the tool also recognizes SNOWSQL CLI Environment Variables):

Environmental Variables

Variable Name	Description
SNOW_USER or SNOWSQL_USER	The username that will be used for the connection
SNOW_PASSWORD or SNOWSQL_PWD	The password that will be used for the connection
SNOW_ROLE or SNOWSQL_ROLE	The snowflake role that will used for the connection
SNOW_ACCOUNT or SNOWSQL_ACCOUNT	The snowflake accountname that will used for the connection
SNOW_WAREHOUSE or SNOWSQL_WAREHOUSE	The warehouse to use when running the sql
SNOW_DATABASE or SNOW_DATABASE	The database to use when running the sql

If you are a SNOWSQL CLI user, this tool can use you configuration settings, using the –activeConn connectionName parameter will search for the [connections.connectionName] section in your config file.

Note

If your files contains several statements you can use the –SplitBefore or –SplitAfter argument, as explained below, so the tool will try to split the statements prior to execution.

Examples

We recommend to have a folder structure like::

+ code
   + procs
     proc1.sql
   + tables
     table1.sql
     + folder1
         table2.sql

If that is the case you can deploy then by running::

sc-deploy-db -A my_sf_account -WH my_wh -U user -P password -I code

If you want to use another authentication like Azure AD you can do::

sc-deploy-db -A my_sf_account -WH my_wh -U user -I code --authenticator externalbrowser

A recommended approach is that you setup a bash shell script, for example config.sh with contents like::

export SNOW_ACCOUNT="demo.us-east-1"
export SNOW_WAREHOUSE="DEMO_WH"
export SNOW_ROLE="DEMO_FULL_ROLE"
export SNOW_DATABASE="DEMODB"
echo "Reading User and Password. When you type values wont be displayed"
read -s -p "User: "     SNOW_USER
echo ""
read -s -p "Password: " SNOW_PASSWORD
echo ""
export SNOW_USER
export SNOW_PASSWORD

You can then run the script like: source config.sh. After that you can just run sc-deploy-db -I folder-to-deploy

Files with multiple statements

If your files have multiple statements, it will cause some failures as the snowflake Python API does not allow multiple statements on a single call.:

000006 (0A000): Multiple SQL statements in a single API call are not supported; use one API call per statement instead.

In order to handle that, you give a tool a regular expression that can be used to split the file contents. This pattern could be used to split before using –SplitBefore pattern or to split after the pattern –SplitAfter pattern.

Let’s see some examples.

If you have a file with contents like:

CREATE OR REPLACE SEQUENCE SEQ1
START WITH 1
INCREMENT BY 1;

/* <sc-table> TABLE1 </sc-table> */
CREATE TABLE TABLE1 (
    COL1 VARCHAR
);

You can use an argument like –SplitAfter ‘;’ that will create a fragment from the file anytime a ; is found.:

sc-deploy-db -A my_sf_account -WH my_wh -U user -P password -I code  --SplitAfter ';'

If you have a file with statements like::

/* <sc-table> TABLE2 </sc-table> */
CREATE TABLE OR REPLACE TABLE1 (
    COL1 VARCHAR
);

/* <sc-table> TABLE2 </sc-table> */
CREATE TABLE TABLE2 (
    COL1 VARCHAR
);

You can use an argument like –SplitBefore ‘CREATE (OR REPLACE)?’. That will create a fragment each time a CREATE or CREATE OR REPLACE fragment is found;

sc-deploy-db -A my_sf_account -WH my_wh -U user -P password -I code  --SplitBefore 'CREATE (OR REPLACE)?'

You can also use something like:

sc-deploy-db -A my_sf_account -WH my_wh -U user -P password -I code  --SplitBefore '\/\*[^\*]*\*\/'

To split before a block comment

Folder Syncronization

A very common practice when using SnowConvert is to organize your files on folders per category [table,view,procedure,macro,function] and per schema. This makes it easier for team collaboration and progress tracking.

Another recommended practice is to have unstabilized code on a work directory and then run the sc-deploy-db, the tool will generate execution logs with summaries of the found errors.

Data Engineers should work on removing the errors found and re-run the sc-deploy-db.

At some point you might need to sync your progress on another folder. A common practice is that you will have a Target folder, where you are supposed to have only the files that have been successfully deployed.

To ease that task the deploy tool provides a folder sync command. This command assumes that you have an structure like::

- WorkDir
    - group1
        -table
            -schema1
               table1.sql
               table2.sql
            -schema2
               table3.sql
               table4.sql
        -view
            -schema1
               view1.sql
            -schema4
               view5.sql
        -function
            -schema2
               function1.sql
               function2.sql
        -procedure
            -schema1
              proc1.sql
              proc2.sql

For example to syncronize tables and views this command should be executed as:

sc-deploy-db -I WorkDir --sync-folder-target WorkDir/group1 --sync-folder-categories "table,view"

The tool will perform queries agains the information_schema tables. It will assume that the file name matches the object name.

Reporting issues and feedback

If you encounter any bugs with the tool please file an issue in the Issues section of our GitHub repo.

License

sc-deploy-db is licensed under the MIT license.