redpanda using docker-compose

This commit is contained in:
priyanshu111299
2025-10-06 19:52:00 +00:00
parent e003bc130a
commit ff812ab021
108 changed files with 7902 additions and 0 deletions

View File

@@ -0,0 +1,73 @@
= Modify the Wasm Transform in the Quickstart
This directory contains the Go source code (`transform.go`) for the data transform that is used in the Redpanda Self-Managed quickstart.
If you're following the quickstart, you *do not* need to modify or rebuild this code. The Docker Compose configuration automatically deploys a pre-built transform called `regex.wasm`.
However, if you want to customize the data transform logic, continue reading.
== Why customize the transform?
- **Custom filtering**: Filter by a different regex or apply multiple conditions.
- **Data manipulation**: Transform records before writing them out. For example, redacting sensitive data or combining fields.
- **Extended functionality**: Add advanced logging, error handling, or multi-topic routing.
== Prerequisites
You need the following:
- At least Go 1.20 installed.
+
[source,bash]
----
go version
----
- The Redpanda CLI (`rpk`) installed.
- A running Redpanda cluster. If you're using the local quickstart with Docker Compose, ensure the cluster is up and running. Or, point `rpk` to another Redpanda environment.
== Modify and deploy your transform
. Open link:transform.go[transform.go] and make your changes. For example:
+
--
- Change the regex logic to handle different use cases.
- Add environment variables to control new features.
- Extend the `doRegexFilter()` function to manipulate records.
--
. Compile your Go code into a `.wasm` file:
+
[source,bash]
----
rpk transform build
----
+
This command compiles your Go source and produces a `.wasm` file that you can deploy to Redpanda.
. Deploy the new transform.
+
If your Docker Compose setup already has a service to deploy the transform, you can restart that service.
+
Otherwise, you can deploy your updated `.wasm` manually using `rpk transform deploy`.
. Produce messages into the input topic. For example:
+
[source,bash]
----
echo '{"key":"alice@university.edu","value":"test message"}' | rpk topic produce logins
----
. Consume from the output topic. For example:
+
[source,bash]
----
rpk topic consume edu-filtered-domains --num 1
----
== Suggested reading
- link:https://docs.redpanda.com/current/reference/rpk/[Redpanda `rpk` CLI Reference^].
- link:https://docs.redpanda.com/current/develop/data-transforms/build/[Develop Data Transforms^].
- https://golang.org/ref/mod[Go Modules^] for managing dependencies and builds in Go.
- https://docs.docker.com/compose/[Docker Compose^] for customizing your environment.

View File

@@ -0,0 +1,5 @@
module regex
go 1.20
require github.com/redpanda-data/redpanda/src/transform-sdk/go/transform v1.1.0

View File

@@ -0,0 +1,2 @@
github.com/redpanda-data/redpanda/src/transform-sdk/go/transform v1.1.0 h1:KxgHJZsHsrT3YX7DMpu/vJN4TZN3KFm1jzrCFLyOepA=
github.com/redpanda-data/redpanda/src/transform-sdk/go/transform v1.1.0/go.mod h1:QGgiwwf/BIsD1b7EiyQ/Apzw+RLSpasRDdpOCiefQFQ=

Binary file not shown.

View File

@@ -0,0 +1,122 @@
package main
// This data transform filters records based on a customizable regex pattern.
// If a record's key or value
// (determined by an environment variable) matches the specified regex,
// the record is forwarded to the output.
// Otherwise, it is dropped.
//
// Usage:
// 1. Provide the following environment variables in your Docker or configuration setup:
// - PATTERN : (required) a regular expression that determines what you want to match.
// - MATCH_VALUE : (optional) a boolean to decide whether to check the record value. If false,
// the record key is checked. Default is false.
//
// Example environment variables:
// PATTERN=".*\\.edu$"
// MATCH_VALUE="true"
//
// Logs:
// This transform logs information about each record and whether it matched.
// The logs appear in the _redpanda.transform_logs topic, so you can debug how your records are being processed.
//
// Build instructions:
// go mod tidy
// rpk transform build
//
// For more details on building transforms with the Redpanda SDK, see:
// https://docs.redpanda.com/current/develop/data-transforms
//
import (
"log"
"os"
"regexp"
"strings"
"github.com/redpanda-data/redpanda/src/transform-sdk/go/transform"
)
var (
re *regexp.Regexp
checkValue bool
)
func isTrueVar(v string) bool {
switch strings.ToLower(v) {
case "yes", "ok", "1", "true":
return true
default:
return false
}
}
// The main() function runs only once at startup. It performs all initialization steps:
// - Reads and compiles the regex pattern.
// - Determines whether to match on the key or value.
// - Registers the doRegexFilter() function to process records.
func main() {
// Set logging preferences, including timestamp and UTC time.
log.SetPrefix("[regex-transform] ")
log.SetFlags(log.Ldate | log.Ltime | log.LUTC | log.Lmicroseconds)
// Start logging the transformation process
log.Println("Starting transform...")
// Read the PATTERN environment variable to get the regex pattern.
pattern, ok := os.LookupEnv("PATTERN")
if !ok {
log.Fatal("Missing PATTERN environment variable")
}
// Log the regex pattern being used.
log.Printf("Using PATTERN: %q\n", pattern)
// Compile the regex pattern for later use.
re = regexp.MustCompile(pattern)
// Read the MATCH_VALUE environment variable to determine whether to check the record's value.
mk, ok := os.LookupEnv("MATCH_VALUE")
checkValue = ok && isTrueVar(mk)
log.Printf("MATCH_VALUE set to: %t\n", checkValue)
log.Println("Initialization complete, waiting for records...")
// Listen for records to be written, calling doRegexFilter() for each record.
transform.OnRecordWritten(doRegexFilter)
}
// The doRegexFilter() function executes each time a new record is written.
// It checks whether the record's key or value (based on MATCH_VALUE) matches the compiled regex.
// If it matches, the record is forwarded, if not, it's dropped.
func doRegexFilter(e transform.WriteEvent, w transform.RecordWriter) error {
// This stores the data to be checked (either the key or value).
var dataToCheck []byte
// Depending on the MATCH_VALUE environment variable, decide whether to check the record's key or value.
if checkValue {
// Use the value of the record if MATCH_VALUE is true.
dataToCheck = e.Record().Value
log.Printf("Checking record value: %s\n", string(dataToCheck))
} else {
// Use the key of the record if MATCH_VALUE is false.
dataToCheck = e.Record().Key
log.Printf("Checking record key: %s\n", string(dataToCheck))
}
// If there is no key or value to check, log and skip the record.
if dataToCheck == nil {
log.Println("Record has no key/value to check, skipping.")
return nil
}
// Check if the data matches the regex pattern.
pass := re.Match(dataToCheck)
if pass {
// If the record matches the pattern, log and write the record to the output topic.
log.Printf("Record matched pattern, passing through. Key: %s, Value: %s\n", string(e.Record().Key), string(e.Record().Value))
return w.Write(e.Record())
} else {
// If the record does not match the pattern, log and drop the record.
log.Printf("Record did not match pattern, dropping. Key: %s, Value: %s\n", string(e.Record().Key), string(e.Record().Value))
// Do not write the record if it doesn't match the pattern.
return nil
}
}

View File

@@ -0,0 +1,33 @@
# Transform metadata used by the rpk transform build command.
# This metadata file tells rpk:
# 1) The transforms display name, which also becomes the base for the .wasm file name.
# 2) A brief description of what it does.
# 3) Defaults for environment variables.
# 4) Input and output topics (if you want to define them here rather than in the deploy command).
# Human-readable name of the transform. rpk transform build uses this for the generated .wasm file.
name: regex
description: |
Filters the input topic to records that only match a regular expression.
Regular expressions are implemented using Go's regexp library, which uses the syntax of RE2.
See the RE2 wiki for allowed syntax: https://github.com/google/re2/wiki/Syntax
Environment variables:
- PATTERN: The regular expression that will match against records (required).
- MATCH_VALUE: By default, the regex matches keys, but if set to "true", the regex matches values.
# By default, no input topic is set here. (You can set it in your deploy command if preferred.)
input-topic: ""
# By default, no output topic is set here. (You can set it in your deploy command if preferred.)
output-topic: ""
# Indicates the specific TinyGo environment used to compile your transform.
language: tinygo-no-goroutines
env:
# The PATTERN variable must be provided at deploy time.
# Example: --var=PATTERN=".*@example.com"
PATTERN: '<required>'