Automating ZDK CLI commands using shell scripting

Hello everyone,

Welcome back to Kaizen. ZDK CLI's command-line nature allows you to easily incorporate it within a shell script, to achieve any custom action that you want to execute. It allows you to execute multiple ZDK CLI commands seamlessly, enhancing productivity.

In this week's Kaizen post, we will discuss a shell script designed to automate the process of pushing locally modified metadata to our CRM environment. Before executing the push operation, ensure that the local metadata is up-to-date and conflicts (same file has been modified in both the local ZDK environment and the remote UI environment) are resolved. To achieve this, we will perform the following steps for files in conflict where manual intervention is required:

Retrieve the latest metadata from Zoho CRM to our local environment using zdk org:pull command
Address conflicts that may arise during the pull operation using zdk org:pull:resolve command
Update our local directory with the newly pulled file using zdk org:pull:update command
Push changes from your local environment to the CRM using zdk org:push command
Update local directory and check status of push command using zdk org:push:result command

This script automates the sequential execution of these commands.

Prerequisites

ZDK CLI Version 1.0.2-beta

To update ZDK CLI, execute the following command in your terminal

npm install -g @zohocrm/zdk-cli

This will initiate the installation process for the latest version.

Crafting the Shell Script

1.Create a new shell script file zdk_automation.sh in your ZDK project's root directory.

#!/bin/bash

touch stdout.txt

#This script is used to automate ZDK CLI conflict resolve flow

conflict_flag=false

# Clear stdout.txt at the start

> stdout.txt

#Step 1: Initial Push Command

echo "====Start of the script ====" >> stdout.txt

echo "**Executing the initial zdk org:push -j" >> stdout.txt

json_content=$(zdk org:push -j)

echo "Output $json_content" >> stdout.txt

next_command=$(echo "$json_content" | jq -r '.next_command_sequence // empty')

echo "next_command after initial push is $next_command">> stdout.txt

# Check if the JSON is valid

if echo "$json_content" | jq empty 2>/dev/null; then

#Step 2: Conflict Detection

message=$(echo "$json_content" | jq -r '.message')

echo "message: $message" >>stdout.txt

if [[ "$message" == *"Merge Conflicts!"* ]]; then

#Step 3: Resolving Conflicts

conflict_flag=true

# Pull changes to the local repository and store the output in a variable

echo "**Executing zdk org:pull -j" >> stdout.txt

json_content=$(zdk org:pull -j)

echo "Output $json_content" >> stdout.txt

if echo "$json_content" | jq empty 2>/dev/null; then

message=$(echo "$json_content" | jq -r '.result')

echo "message $message" >>stdout.txt

IFS=$'\n' read -r -d '' -a filenames <<< "$message"

# Iterate over each filename and execute the resolve command

for filename in "${filenames[@]}"; do

echo "**Executing zdk org:pull:resolve for $filename" >> stdout.txt

# Executing zdk org:pull:resolve

json_content=$(zdk org:pull:resolve -f "$filename" -j)

echo "Output $json_content" >> stdout.txt

done

echo "***Executing zdk org:pull:update" >> stdout.txt

json_content=$(zdk org:pull:update -j)

echo "Output $json_content" >> stdout.txt

#Step 4: Second Push Attempt after conflict

if [ "$conflict_flag" = true ]; then

# Push changes to the org and store the output in a variable again

echo "***Executing: zdk org:push -j after conflict resolution" >> stdout.txt

json_content=$(zdk org:push -j)

echo "Output $json_content" >> stdout.txt

# Extract the line containing "next_command_sequence" from the push output

next_command=$(echo "$json_content" | jq -r '.next_command_sequence')

echo "Next command $next_command " >> stdout.txt

#Step 5: Push result command

# Execute the next command if it exists

counter=0 # Initialize a counter

if [ -n "$next_command" ]; then

while true; do

# Execute the command stored in next_command and capture the output

echo "***Executing: $next_command -j" >> stdout.txt

output=$(eval "$next_command -j")

echo "Output $output" >>stdout.txt

# Check if the output contains "Process in progress"

if echo "$output" | grep -q "Process in progress"; then

echo "Push process in progress. Waiting for 5 seconds..." >> stdout.txt

sleep 5 # Wait for 5 seconds

else

# If not found, break the loop and handle the output

echo "Command executed successfully or completed." >> stdout.txt

break

# Increment the counter

counter=$((counter + 1))

# Break the loop after 10 iterations

if [ "$counter" -ge 10 ]; then

echo "Maximum number of attempts(10) reached. Exiting the loop." >> stdout.txt

echo "Try executing $next_command after some time" >> stdout.txt

break

done

Explanation of the script

Step 1: Initial Push Command

The script starts by executing the zdk org:push -j command.
The commands executed using -j option ensures that the output is in a machine-readable JSON format, which facilitates easier parsing and processing.
The conflict_flag is set to false.

Step 2: Conflict Detection

The script checks the output for any merge conflicts.
If conflicts are detected, it executes the zdk org:pull -j command to retrieve the latest changes from the organization. The conflict_flag is set to true.
If conflicts are not present, the script proceeds to Step 5: Executing the push:result command.

Step 3: Resolving Conflicts

If conflicts are present, the script extracts a list of filenames associated with these conflicts.
It iterates through this list and executes the zdk org:pull:resolve -j command for each filename, allowing for manual resolution of conflicts as needed.
This opens a three-way editor in the browser where the conflicts can be resolved.

Step 4: Second Push Attempt after conflict

Once all conflicts are resolved, the script attempts to push any changes back to Zoho CRM by executing the zdk org:push -j command.

Step 5: Executing push:result command

This command should be followed by zdk org:push:result --jobid {id} command (indicated by a next_command_sequence) which displays the result of the push process with the mentioned job id and updates the local directory.
During this process, if the result is shown as "Process in progress", a subsequent push:result command is required. In this case, the script waits for five seconds before trying again to ensure that operations complete successfully.

2.Make the script executable

chmod +x zdk_automation.sh

3.Run your script

./zdk_automation.sh & tail -f stdout.txt

Logging Actions: All actions taken by the script are logged in stdout.txt. This serves as a record for later reference and troubleshooting.

Using ZDK CLI with shell script ensures that your ZDK CLI operations are automated daily, minimizing manual intervention while effectively managing conflicts.

For the current beta release, ZDK CLI is exclusively available for Sandbox environment and is not operational in Production environments.

We hope you found this post useful. We will meet you next week with another interesting topic!

If you have any questions, let us know in the comment section.

Happy Exploring!

Access your files securely from anywhere

Zoho Developer Community

New to Zoho LandingPage?

Access Zoho LandingPage

Zoho LandingPage Resources

New to Zoho Survey?

Access Zoho Survey

Latest Feature Updates

Explore our Pricing & Plans

Zoho Survey Resources

Android Mobile Surveys

Insurvey Blogs

Insight Blogs

Tips & Tricks

New to Zoho CRM Plus?

Deliver unforgettable customer experiences

ACCESS ZOHO CRM PLUS

New to Zoho CRM Plus?

Deliver unforgettable customer experiences

ACCESS ZOHO CRM PLUS

New to Zoho Marketing Plus?

Everything you need to run your marketing

ACCESS ZOHO MARKETING PLUS

New to Zoho Marketing Plus?

Everything you need to run your marketing

ACCESS ZOHO MARKETING PLUS

New to Bigin?

Signup Now

Access Bigin

Latest Feature Updates

Zoho Desk Resources

Desk Community Learning Series
Digest
Functions
Meetups
Kbase
Resources
Glossary
Desk Marketplace
MVP Corner
Word of the Day

Tout forum

France ZUGs

Zoho SalesIQ Resources

Topic Participants
Mable Mary M P

Sticky Posts
Kaizen#162 : Add Tags and Open Pre-filled Email Drafts via Client Scripts
Hello everyone! Welcome to another informative Kaizen post. In this post, let us see how to accomplish the following using Client Script. 1. How to auto-tag a record based on field update? 2. How to Open a pre-filled email draft This post will provide
Client Script | Update - Introducing Commands in Client Script!
Have you ever wished you could trigger Client Script from contexts other than just the supported pages and events? Have you ever wanted to leverage the advantage of Client Script at your finger tip? Discover the power of Client Script - Commands! Commands
Extension Pointers - JS SDK series #4: Managing data from a widget using ZOHO.CRM.HTTP
Handling and managing the data between two applications is a major factor for an efficient business. There are a variety of ways to handle data between Zoho CRM and other third-party applications. You can use deluge CRM tasks, create connectors, use API
Kaizen #152 - Client Script Support for the new Canvas Record Forms
Hello everyone! Have you ever wanted to trigger actions on click of a canvas button, icon, or text mandatory forms in Create/Edit and Clone Pages? Have you ever wanted to control how elements behave on the new Canvas Record Forms? This can be achieved
Extension pointers for integrating Zoho CRM with Zoho products #8: Upload and manage Zoho Workdrive folders and files from within Zoho CRM
Keeping records on your customers and business prospects is essential for tracking data, conducting follow-ups, and running a business smoothly. When you use two separate applications, and store relevant data in each, checking and tracking data becomes