### Install Image Credential Helper Example

Source: https://github.com/nvidia/nvcf/blob/main/src/compute-plane-services/image-credential-helper/examples/README.md

Run this script to install the example. Ensure you are in the project root directory.

```shell
./examples/run.sh
```

--------------------------------

### Local Development Setup with Make

Source: https://github.com/nvidia/nvcf/blob/main/src/control-plane-services/nats-auth-callout/README.md

Install dependencies and Air, then start development with hot reloading for automatic restarts on file changes.

```shell
make go-install-deps
go install github.com/air-verse/air@latest

make dev
```

--------------------------------

### Install Objx

Source: https://github.com/nvidia/nvcf/blob/main/src/compute-plane-services/nvca/vendor/github.com/stretchr/objx/README.md

Use 'go get' to install the Objx package.

```bash
go get github.com/stretchr/objx
```

--------------------------------

### Example Helm Chart Installation

Source: https://github.com/nvidia/nvcf/blob/main/deploy/helm/nats/README.md

An example of how to install the NATS Helm chart with a specific release name, namespace, and values file.

```bash
helm install nvcf-nats . \
  --namespace nats-system \
  --create-namespace \
  --values values.yaml
```

--------------------------------

### Installation Example: SQLite (Default)

Source: https://github.com/nvidia/nvcf/blob/main/src/control-plane-services/nats-auth-callout/deploy/README.md

Install the Helm chart with default settings, which uses SQLite as the database provider.

```bash
helm install my-release ./deploy
```

--------------------------------

### Install sigs.k8s.io/yaml

Source: https://github.com/nvidia/nvcf/blob/main/src/libraries/go/lib/vendor/sigs.k8s.io/yaml/README.md

Install the library using the go get command.

```bash
$ go get sigs.k8s.io/yaml
```

--------------------------------

### Install Mergo

Source: https://github.com/nvidia/nvcf/blob/main/src/compute-plane-services/image-credential-helper/vendor/dario.cat/mergo/README.md

Install the Mergo library using the go get command. This is the first step before using it in your Go code.

```go
go get dario.cat/mergo
```

--------------------------------

### Install Afero Package

Source: https://github.com/nvidia/nvcf/blob/main/src/compute-plane-services/nvca/vendor/github.com/spf13/afero/README.md

Use go get to install the Afero package. This command fetches and installs the specified package and its dependencies.

```bash
go get github.com/spf13/afero
```

--------------------------------

### Install Locafero

Source: https://github.com/nvidia/nvcf/blob/main/src/compute-plane-services/nvca/vendor/github.com/sagikazarmark/locafero/README.md

Use this command to install the Locafero library using go get.

```shell
go get github.com/sagikazarmark/locafero
```

--------------------------------

### Install lo Library

Source: https://github.com/nvidia/nvcf/blob/main/src/compute-plane-services/nvca/vendor/github.com/samber/lo/README.md

Install the lo library using go get. This library is v1 and follows SemVer strictly.

```bash
go get github.com/samber/lo@v1
```

--------------------------------

### Install Mergo

Source: https://github.com/nvidia/nvcf/blob/main/src/compute-plane-services/nvca/vendor/github.com/imdario/mergo/README.md

Install the Mergo library using go get.

```bash
go get github.com/imdario/mergo
```

--------------------------------

### Install Control Plane and Register Compute Planes

Source: https://github.com/nvidia/nvcf/blob/main/ai-tooling/user/skills/nvcf-self-managed-cli/examples/multi-cluster.md

This snippet shows the initial setup for a multi-cluster environment. First, it installs the control plane, then it iterates through a list of contexts to register and install each compute plane.

```sh
# 1. Bring up the control plane (one-time, on the control-plane cluster):
KUBECONFIG=cp.yaml nvcf-cli self-hosted install --control-plane | kubectl apply -f -
nvcf-cli self-hosted check --control-plane --wait 5m

# 2. Register + install each compute plane:
for CTX in admin@gpu-east-1 admin@gpu-west-1 admin@gpu-eu-1; do
  NAME=$(echo "$CTX" | cut -d@ -f2)
  nvcf-cli self-hosted up \
    --cluster-name=$NAME \
    --compute-plane-context=$CTX \
    --icms-url=https://icms.nvcf.example.com \
    --token=$NVCF_ADMIN_JWT \
    --non-interactive \
    --json
done

# 3. Verify all compute planes registered:
nvcf-cli cluster list --json | jq '.clusters[] | {name, region, lastHeartbeat}'
```

--------------------------------

### Example Dockerfile

Source: https://github.com/nvidia/nvcf/blob/main/docs/user/container-functions.md

Sets up an NVIDIA-optimized base container, installs Python requirements, copies model code, and configures the run command.

```dockerfile
FROM nvcr.io/nvidia/cuda:12.1.1-devel-ubuntu22.04
RUN apt-get update && apt-get install -y \
    git \
    python3 \
    python3-pip \
    python-is-python3 \
    libsm6 \
    libxext6 \
    libxrender-dev \
    curl \
    && rm -rf /var/lib/apt/lists/*
WORKDIR /workspace/

# Install requirements file
COPY requirements.txt requirements.txt
RUN pip install --no-cache-dir --upgrade pip
RUN pip install --no-cache-dir -r requirements.txt
ENV DEBIAN_FRONTEND=noninteractive

# Copy model source code and weights
COPY model_weights /models
COPY model_source .
COPY run.py .

# Set run command to start PyTriton to serve the model
CMD python3 run.py
```

--------------------------------

### Check Prerequisites with NVCF CLI

Source: https://github.com/nvidia/nvcf/blob/main/docs/user/quickstart.md

Before running the quickstart installation, verify that all prerequisites are met using this command.

```bash
nvcf-cli self-hosted check --pre
```

--------------------------------

### Install Viper

Source: https://github.com/nvidia/nvcf/blob/main/src/compute-plane-services/nvca/vendor/github.com/spf13/viper/README.md

Install the Viper library using go get. Viper uses Go Modules for dependency management.

```shell
go get github.com/spf13/viper
```

--------------------------------

### Update Examples

Source: https://github.com/nvidia/nvcf/blob/main/src/compute-plane-services/byoo-otel-collector/AGENTS.md

Regenerates example configurations after modifying templates. Always verify generated examples before committing.

```bash
make update-examples
```

--------------------------------

### Setup

Source: https://github.com/nvidia/nvcf/blob/main/src/compute-plane-services/nvca/vendor/github.com/urfave/cli/v2/godoc-current.txt

Performs initialization to prepare the application for running or inspection.

```APIDOC
## App.Setup()

### Description
The `Setup` method runs essential initialization code to ensure that all internal data structures are correctly set up and ready for the `Run` method or for inspection prior to execution. This method is called internally by `Run` and will return early if the setup has already been completed.

### Signature
```go
func (a *App) Setup()
```
```

--------------------------------

### Install Protobuf Compiler

Source: https://github.com/nvidia/nvcf/blob/main/src/libraries/go/lib/vendor/github.com/go-kit/kit/transport/grpc/README.md

Installs the protobuf compiler from source using Homebrew. Ensure autoconf, automake, and libtool are installed first.

```shell
brew install autoconf automake libtool
git clone https://github.com/google/protobuf
cd protobuf
./autogen.sh ; ./configure ; make ; make install
```

--------------------------------

### Full Example: Basic Mux Server

Source: https://github.com/nvidia/nvcf/blob/main/src/compute-plane-services/nvca/vendor/github.com/gorilla/mux/README.md

A complete, runnable example of a small web server using Gorilla Mux. It demonstrates setting up a router and handling a basic route.

```Go
package main

import (
    "net/http"
    "log"
    "github.com/gorilla/mux"
)

func YourHandler(w http.ResponseWriter, r *http.Request) {
    w.Write([]byte("Gorilla!\n"))
}

func main() {
    r := mux.NewRouter()
    // Routes consist of a path and a handler function.
    r.HandleFunc("/", YourHandler)

    // Bind to a port and pass our router in
    log.Fatal(http.ListenAndServe(":8000", r))
}
```

--------------------------------

### Setup

Source: https://github.com/nvidia/nvcf/blob/main/src/libraries/go/lib/vendor/github.com/urfave/cli/v2/godoc-current.txt

Performs initialization to prepare data structures for `Run` or inspection before `Run` is called.

```APIDOC
## Setup

### Description
Runs initialization code to ensure all data structures are ready for `Run` or inspection prior to `Run`. It is internally called by `Run`, but will return early if setup has already happened.

### Signature
`func (a *App) Setup()`
```

--------------------------------

### Install go.uber.org/atomic

Source: https://github.com/nvidia/nvcf/blob/main/src/libraries/go/lib/vendor/go.uber.org/atomic/README.md

Install the atomic package using go get.

```shell
$ go get -u go.uber.org/atomic@v1
```

--------------------------------

### Example Environment Variables Configuration

Source: https://github.com/nvidia/nvcf/blob/main/src/control-plane-services/nats-auth-callout/docs/CONFIGURATION.md

Shows how to configure the service using environment variables with the specified prefix.

```bash
# Environment Variables: NVCF_NATS_AUTH_CALLOUT_SERVICE_SERVER_PORT=8080 NVCF_NATS_AUTH_CALLOUT_SERVICE_SERVICE_NAME=my-service NVCF_NATS_AUTH_CALLOUT_SERVICE_SECRETS_FILE_PATH=secrets.json
```

--------------------------------

### Install mapstructure

Source: https://github.com/nvidia/nvcf/blob/main/src/compute-plane-services/nvca/vendor/github.com/go-viper/mapstructure/v2/README.md

Use go get to install the mapstructure library. This command fetches and installs the specified package.

```shell
go get github.com/go-viper/mapstructure/v2
```

--------------------------------

### Example File Read

Source: https://github.com/nvidia/nvcf/blob/main/src/compute-plane-services/ess-agent/docs/templating-language.md

Demonstrates reading the content of a specific local file.

```Go
{{ file "/path/to/my/file" }}
```

--------------------------------

### Install json-iterator/go

Source: https://github.com/nvidia/nvcf/blob/main/src/compute-plane-services/image-credential-helper/vendor/github.com/json-iterator/go/README.md

Install the json-iterator/go library using the go get command.

```bash
go get github.com/json-iterator/go
```

--------------------------------

### Clone and Setup Repository

Source: https://github.com/nvidia/nvcf/blob/main/src/libraries/go/lib/vendor/go.uber.org/automaxprocs/CONTRIBUTING.md

Steps to fork, clone the repository, and set up remote upstream for contributing.

```bash
mkdir -p $GOPATH/src/go.uber.org
cd $GOPATH/src/go.uber.org
git clone git@github.com:your_github_username/automaxprocs.git
cd automaxprocs
git remote add upstream https://github.com/uber-go/automaxprocs.git
git fetch upstream
```

--------------------------------

### Install automemlimit

Source: https://github.com/nvidia/nvcf/blob/main/src/libraries/go/lib/vendor/github.com/KimMachineGun/automemlimit/README.md

Install the latest version of the automemlimit library using go get.

```shell
go get github.com/KimMachineGun/automemlimit@latest
```

--------------------------------

### Install conc Package

Source: https://github.com/nvidia/nvcf/blob/main/src/compute-plane-services/nvca/vendor/github.com/sourcegraph/conc/README.md

Install the conc package using the go get command.

```sh
go get github.com/sourcegraph/conc
```

--------------------------------

### Example Configuration File

Source: https://github.com/nvidia/nvcf/blob/main/examples/function-samples/helmchart-samples/multi-node-helm-function-test/README.md

Example of a populated config.env file with NVCF credentials.

```bash
KEY="<nvcf-cli-generated-api-key>"
FUNCTION_ID="ce460ed1-6f17-4bdc-ad6b-00a569fc780d"
GATEWAY_ADDR="a1b2c3d4.us-east-1.elb.amazonaws.com"
```

--------------------------------

### Install Ray Serve Sample (CPU, ARM64)

Source: https://github.com/nvidia/nvcf/blob/main/examples/function-samples/helmchart-samples/ray-serve-sample/README.md

Installs the Ray Serve sample Helm chart for CPU-only testing on Apple Silicon (ARM64). Uses the aarch64 image tag and sets memory resources.

```bash
helm install ray-serve-sample ./ray-serve \
  --set gpu.count=0 \
  --set image.tag=2.40.0-py310-aarch64 \
  --set resources.requests.memory=2Gi \
  --set resources.limits.memory=4Gi
```

--------------------------------

### Example Initial Keyspace Migrations

Source: https://github.com/nvidia/nvcf/blob/main/migrations/cassandra/README.md

Illustrates the conventional sequence for initializing a new keyspace, including creating the keyspace, roles, and canonical schema.

```text
01_init_keyspace.up.sql - creates the keyspace with NetworkTopologyStrategy (uses {{ $replicaCount }} template variable, substituted upstream).
02_init_roles.up.sql - creates the application role and grants, with the login password supplied by ${SERVICE_ROLE_PASSWORD}.
03_init_tables.up.sql - the canonical schema (UDTs, tables, indexes) for the keyspace.
```

--------------------------------

### Install go.yaml.in/yaml/v3

Source: https://github.com/nvidia/nvcf/blob/main/src/compute-plane-services/image-credential-helper/vendor/go.yaml.in/yaml/v3/README.md

Install the YAML package for Go using the go get command.

```bash
go get go.yaml.in/yaml/v3
```

--------------------------------

### Build and Run with Make

Source: https://github.com/nvidia/nvcf/blob/main/src/control-plane-services/nats-auth-callout/README.md

Build the server binary and run it using the Makefile. Also includes running directly with Go.

```shell
make build

./nvcf-nats-auth-callout-service server

# Or run directly with Go
make run
```

--------------------------------

### Install Ray Serve Sample (CPU, AMD64)

Source: https://github.com/nvidia/nvcf/blob/main/examples/function-samples/helmchart-samples/ray-serve-sample/README.md

Installs the Ray Serve sample Helm chart for CPU-only testing on AMD64 architecture. Disables GPU requests.

```bash
helm install ray-serve-sample ./ray-serve \
  --set gpu.count=0 \
  --set image.tag=2.40.0-py310
```

--------------------------------

### Install go.uber.org/multierr

Source: https://github.com/nvidia/nvcf/blob/main/src/compute-plane-services/nvca/vendor/go.uber.org/multierr/README.md

Install the latest version of the multierr library using the go get command.

```bash
go get -u go.uber.org/multierr@latest
```

--------------------------------

### Instantiation with Options

Source: https://github.com/nvidia/nvcf/blob/main/src/compute-plane-services/nvca/vendor/go.opentelemetry.io/otel/CONTRIBUTING.md

Demonstrates how to instantiate a type using a variadic Option parameter. Any required parameters can precede the options.

```go
func NewT(options ...Option) T {…}
```

--------------------------------

### Install Cobra Library

Source: https://github.com/nvidia/nvcf/blob/main/src/compute-plane-services/nvca/vendor/github.com/spf13/cobra/README.md

Install the latest version of the Cobra library using go get.

```bash
go get -u github.com/spf13/cobra@latest
```

--------------------------------

### Example CLI Configuration Flags

Source: https://github.com/nvidia/nvcf/blob/main/src/control-plane-services/nats-auth-callout/docs/CONFIGURATION.md

Demonstrates how to set core service configuration options using command-line flags.

```bash
# CLI Flags: --port 8080 --service-name my-service --config config.yaml --secrets-file secrets.json
```

--------------------------------

### Install Blackfriday v2

Source: https://github.com/nvidia/nvcf/blob/main/src/libraries/go/lib/vendor/github.com/russross/blackfriday/v2/README.md

Install the Blackfriday v2 package using go get in module mode.

```go
go get github.com/russross/blackfriday/v2
```

--------------------------------

### Install and Import YAML Package

Source: https://github.com/nvidia/nvcf/blob/main/src/compute-plane-services/image-credential-helper/vendor/sigs.k8s.io/yaml/README.md

Install the package using go get and import it into your Go project.

```bash
$ go get sigs.k8s.io/yaml
```

```go
import "sigs.k8s.io/yaml"
```

--------------------------------

### Example requirements.txt

Source: https://github.com/nvidia/nvcf/blob/main/docs/user/container-functions.md

Lists Python dependencies for your model, including nvidia-pytriton.

```text
--extra-index-url https://pypi.ngc.nvidia.com
opencv-python-headless
pycocotools
matplotlib
torch==2.1.0
nvidia-pytriton==0.3.0
numpy
```

--------------------------------

### Basic Command Line Usage

Source: https://github.com/nvidia/nvcf/blob/main/src/control-plane-services/nats-auth-callout/docs/CONFIGURATION.md

Start the service with basic configuration, including port and service name, using command-line flags.

```bash
# Basic usage
./nvcf-nats-auth-callout-service server --port 9090 --service-name my-service
```

--------------------------------

### NVCF CLI Fresh Setup Workflow

Source: https://github.com/nvidia/nvcf/blob/main/src/clis/nvcf-cli/USAGE-GUIDE.md

Demonstrates the recommended setup workflow where the CLI generates and manages tokens automatically in the state file.

```bash
# 1. User gets CLI and runs init
#  Token generated and saved to state file
# DEBUG: Successfully generated admin token, expires: 2025-10-15 14:16:21
```

```bash
# 2. User runs commands - tokens work automatically
./nvcf-cli list functions --debug
#  DEBUG: NVCF_TOKEN loaded from: state
#  Uses token from state file seamlessly
```

--------------------------------

### Helm Validation Error Examples

Source: https://github.com/nvidia/nvcf/blob/main/src/control-plane-services/nats-auth-callout/deploy/README.md

Examples of common validation errors encountered during Helm chart installation for invalid configurations.

```bash
Error: When database.provider is 'postgres', either postgres.existingSecret must be set OR both postgres.username and postgres.password must be provided
```

```bash
Error: Unsupported database provider 'mysql'. Supported providers are: postgres, postgresql, sqlite, sqlite3
```

```bash
Error: When metrics.enabled is true, metrics.port must be specified
```

```bash
Error: When metrics.enabled is false, metrics.serviceMonitor.enabled cannot be true. ServiceMonitor requires metrics to be enabled.
```

```bash
Error: When metrics.enabled is false, dashboard.enabled cannot be true. Dashboard requires metrics to be enabled.
```

--------------------------------

### Install blackfriday-tool

Source: https://github.com/nvidia/nvcf/blob/main/src/libraries/go/lib/vendor/github.com/russross/blackfriday/v2/README.md

Download and install the blackfriday-tool command-line utility.

```go
go get github.com/russross/blackfriday-tool
```

--------------------------------

### Install and Run Local Go Doc Site

Source: https://github.com/nvidia/nvcf/blob/main/src/compute-plane-services/nvca/vendor/go.opentelemetry.io/otel/CONTRIBUTING.md

Installs the pkgsite tool and runs a local Go documentation site. This is useful for previewing package documentation locally.

```sh
go install golang.org/x/pkgsite/cmd/pkgsite@latest
pkgsite
```

--------------------------------

### Install Gorilla Mux

Source: https://github.com/nvidia/nvcf/blob/main/src/compute-plane-services/nvca/vendor/github.com/gorilla/mux/README.md

Install the gorilla/mux package using the go get command. Ensure your Go toolchain is correctly configured.

```shell
go get -u github.com/gorilla/mux
```

--------------------------------

### Complete Workflow Example

Source: https://github.com/nvidia/nvcf/blob/main/src/clis/nvcf-cli/README.md

A step-by-step example of a complete workflow using the NCF CLI, from initialization to cleanup.

```bash
# 1. Initialize with fresh admin token
nvcf-cli init

# 2. Generate API key for invocations
nvcf-cli api-key generate --description "Demo key"

# 3. Create function (context automatically saved)
nvcf-cli create --input-file examples/create-function.json

# 4. Deploy function (uses saved context automatically)
nvcf-cli deploy

# 5. Invoke function (uses saved context and API key automatically)
nvcf-cli invoke --request-body '{"message": "hello world"}'

# 6. Check everything is working
nvcf-cli status

# 7. Clean up when done
nvcf-cli undeploy  # New undeploy command!
nvcf-cli delete    # Uses saved context
```

--------------------------------

### Install NVCF Single-Cluster

Source: https://github.com/nvidia/nvcf/blob/main/ai-tooling/user/skills/nvcf-self-managed-cli/SKILL.md

Installs both the control plane and compute plane on the current Kubernetes cluster context. Use for initial setup.

```bash
nvcf-cli self-hosted up --cluster-name=ncp-local
```

--------------------------------

### Generate Documentation with Custom Configuration

Source: https://github.com/nvidia/nvcf/blob/main/src/compute-plane-services/byoo-otel-collector/generator/README.md

This example shows how to generate documentation with a custom configuration file and a custom output filename. It also specifies a custom source template folder.

```bash
uv run -m generator \
  ---document-output-dir doc \
  ---template-output-dir gen \
  --config ./source-config.yaml \
  --document-filename METRICS.md \
  --source-template-folder internal/otelconfig/templates
```

--------------------------------

### Starting and Stopping gocron Scheduler

Source: https://github.com/nvidia/nvcf/blob/main/src/libraries/go/lib/vendor/github.com/go-co-op/gocron/README.md

Provides examples of how to start the gocron scheduler asynchronously or by blocking the current execution path, and how to stop the scheduler.

```golang
// you can start running the scheduler in two different ways:
// starts the scheduler asynchronously
s.StartAsync()
// starts the scheduler and blocks current execution path
s.StartBlocking()

// stop the running scheduler in two different ways:
// stop the scheduler
s.Stop()

// stop the scheduler and notify the `StartBlocking()` to exit
s.StopBlockingChan()
```

--------------------------------

### Install Mock Server Dependencies

Source: https://github.com/nvidia/nvcf/blob/main/src/control-plane-services/function-autoscaler/README.md

Installs the necessary Python dependencies for the mock NVCF API server. This is required before manually starting the server.

```bash
cd local_env/mock_server
pip install -r requirements.txt
```

--------------------------------

### Logging and Compression Handler Example

Source: https://github.com/nvidia/nvcf/blob/main/src/compute-plane-services/nvca/vendor/github.com/gorilla/handlers/README.md

Demonstrates how to use LoggingHandler for logging requests to stdout and CompressHandler for gzipping responses with net/http.

```go
import (
    "net/http"
    "github.com/gorilla/handlers"
)

func main() {
    r := http.NewServeMux()

    // Only log requests to our admin dashboard to stdout
    r.Handle("/admin", handlers.LoggingHandler(os.Stdout, http.HandlerFunc(ShowAdminDashboard)))
    r.HandleFunc("/", ShowIndex)

    // Wrap our server with our gzip handler to gzip compress all responses.
    http.ListenAndServe(":8000", handlers.CompressHandler(r))
}
```

--------------------------------

### Start Development Environment

Source: https://github.com/nvidia/nvcf/blob/main/src/control-plane-services/nats-auth-callout/local-dev/README.md

Use this command to start the local development environment. It can be run with observability profiles for tracing and metrics.

```bash
# Start development environment (basic)
devspace dev

# Start development with observability profiles
devspace dev -p tracing
devspace dev -p metrics  
devspace dev -p obs
```

--------------------------------

### Rendered Varnish Configuration Example

Source: https://github.com/nvidia/nvcf/blob/main/src/compute-plane-services/ess-agent/examples/varnish.md

An example of a Varnish configuration file rendered by Consul Template, showing backend definitions and load balancing setup.

```text
import directors;

backend consul_consul {
    .host = "104.131.109.106";
    .port = "8300";"
}

sub vcl_init {
  new bar = directors.round_robin();

  bar.add_backend(consul_consul);
}

sub vcl_recv {
  set req.backend_hint = bar.backend();
}
```

--------------------------------

### Install and Use Crypt CLI for Remote Config

Source: https://github.com/nvidia/nvcf/blob/main/src/libraries/go/lib/vendor/github.com/spf13/viper/README.md

Steps to install the crypt CLI tool and use it to set and get configuration values from a key/value store.

```bash
$ go get github.com/sagikazarmark/crypt/bin/crypt
$ crypt set -plaintext /config/hugo.json /Users/hugo/settings/config.json
```

```bash
$ crypt get -plaintext /config/hugo.json
```

--------------------------------

### Building Cloud-Agnostic Applications with Afero

Source: https://github.com/nvidia/nvcf/blob/main/src/compute-plane-services/nvca/vendor/github.com/spf13/afero/README.md

Demonstrates how to write applications that can work with any Afero-compatible filesystem, including local disk, cloud storage, and in-memory implementations, by abstracting filesystem interactions.

```Go
type DocumentProcessor struct {
    fs afero.Fs
}

func NewDocumentProcessor(fs afero.Fs) *DocumentProcessor {
    return &DocumentProcessor{fs: fs}
}

func (p *DocumentProcessor) Process(inputPath, outputPath string) error {
    content, err := afero.ReadFile(p.fs, inputPath)
    if err != nil {
        return err
    }
    
    processed := processContent(content)
    return afero.WriteFile(p.fs, outputPath, processed, 0644)
}

// Use with local filesystem
processor := NewDocumentProcessor(afero.NewOsFs())

// Use with Google Cloud Storage
processor := NewDocumentProcessor(gcsFS)

// Use with in-memory filesystem for testing
processor := NewDocumentProcessor(afero.NewMemMapFs())
```

--------------------------------

### Start Consul Agent with Service Configuration

Source: https://github.com/nvidia/nvcf/blob/main/src/compute-plane-services/ess-agent/examples/nginx-connect-proxy.md

Starts the Consul agent in development mode with configurations for 'ingress' and 'webserver' services. Ensure Consul is installed and accessible.

```shell
$ consul agent -dev -log-level=info -config-file=consul-services.json
```

--------------------------------

### Download and Prepare golang-migrate Binaries

Source: https://github.com/nvidia/nvcf/blob/main/migrations/cassandra/README.md

Download the golang-migrate binary for amd64 and arm64 architectures, place them in the 'files/' directory, and make them executable.

```bash
# amd64
curl -L https://github.com/golang-migrate/migrate/releases/download/v4.18.2/migrate.linux-amd64.tar.gz \
  | tar xz -C files/
mv files/migrate files/migrate-amd64
chmod +x files/migrate-amd64

# arm64 (run on / for an arm64 build)
curl -L https://github.com/golang-migrate/migrate/releases/download/v4.18.2/migrate.linux-arm64.tar.gz \
  | tar xz -C files/
mv files/migrate files/migrate-arm64
chmod +x files/migrate-arm64
```

--------------------------------

### Install NVCF Helm Chart

Source: https://github.com/nvidia/nvcf/blob/main/docs/user/standalone-deployment.md

Example command to install or upgrade an NVCF Helm chart using `helm upgrade --install`. This command specifies the release name, namespace, chart source, version, and a values file.

```bash
helm upgrade --install api -n nvcf \
  oci://<your-registry>/<your-repo>/helm-nvcf-api --version <version> \
  -f values.yaml
```

--------------------------------

### Install Example Dashboards

Source: https://github.com/nvidia/nvcf/blob/main/docs/user/example-dashboards.md

Install the nvcf-example-dashboards Helm chart to configure Grafana with pre-built dashboards for various NVCF services. Ensure you use the latest version from NGC.

```bash
helm upgrade \
  --install nvcf-example-dashboards \
  oci://nvcr.io/0833294136851237/nvcf-ncp-staging/nvcf-example-dashboards \
  --version 1.6.0 \
  --namespace observability \
  --create-namespace
```

--------------------------------

### Example: Initial Keyspace Creation

Source: https://github.com/nvidia/nvcf/blob/main/migrations/cassandra/keyspaces/README.md

This migration script creates the keyspace using NetworkTopologyStrategy. It utilizes a template variable for the replica count.

```sql
CREATE KEYSPACE {{ $replicaCount }} WITH replication = {'class': 'NetworkTopologyStrategy', 'replication_factor': 3};

```

--------------------------------

### Define a WebService and a GET Route

Source: https://github.com/nvidia/nvcf/blob/main/src/compute-plane-services/image-credential-helper/vendor/github.com/emicklei/go-restful/v3/README.md

Example of defining a WebService with a path, supported content types, and a GET route for retrieving a user. It includes route documentation and parameter definition.

```go
ws := new(restful.WebService)
ws.
	Path("/users").
	Consumes(restful.MIME_XML, restful.MIME_JSON).
	Produces(restful.MIME_JSON, restful.MIME_XML)

ws.Route(ws.GET("/{user-id}").To(u.findUser).
	Doc("get a user").
	Param(ws.PathParameter("user-id", "identifier of the user").DataType("string")).
	Writes(User{}))		
...
	
func (u UserResource) findUser(request *restful.Request, response *restful.Response) {
	id := request.PathParameter("user-id")
	...
}
```

--------------------------------

### Install Make

Source: https://github.com/nvidia/nvcf/blob/main/src/control-plane-services/nats-auth-callout/README.md

Provides commands to install Make on Ubuntu/Debian, macOS, or Windows.

```bash
# Install Make:

# - Ubuntu/Debian: sudo apt-get install make
# - macOS: brew install make or install Xcode command line tools
# - Windows: Use WSL or install Make for Windows
```

--------------------------------

### Get and Set unsafe.Pointer Value with reflect2

Source: https://github.com/nvidia/nvcf/blob/main/src/compute-plane-services/image-credential-helper/vendor/github.com/modern-go/reflect2/README.md

This example shows how to get and set a value using unsafe.Pointer with reflect2. Similar to interface{} manipulation, use the pointer to the type for setting values.

```go
valType := reflect2.TypeOf(1)
i := 1
j := 10
valType.UnsafeSet(unsafe.Pointer(&i), unsafe.Pointer(&j))
// i will be 10
```

--------------------------------

### Install NATS Go Client

Source: https://github.com/nvidia/nvcf/blob/main/src/compute-plane-services/nvca/vendor/github.com/nats-io/nats.go/README.md

Install the latest released or a specific version of the NATS Go client using go get. Ensure you are using a compatible NATS Server version.

```bash
# To get the latest released Go client:
go get github.com/nats-io/nats.go@latest

# To get a specific version:
go get github.com/nats-io/nats.go@v1.47.0

# Note that the latest major version for NATS Server is v2:
go get github.com/nats-io/nats-server/v2@latest
```

--------------------------------

### Install JSON-Patch v4

Source: https://github.com/nvidia/nvcf/blob/main/src/compute-plane-services/image-credential-helper/vendor/gopkg.in/evanphx/json-patch.v4/README.md

Use this command to get version 4 of the jsonpatch library for Go.

```bash
go get -u gopkg.in/evanphx/json-patch.v4
```

--------------------------------

### Setup Environment Variables and Build CLI

Source: https://github.com/nvidia/nvcf/blob/main/src/clis/nvcf-cli/test/e2e/README.md

Sets up necessary environment variables and builds the nvcf-cli binary for testing. Assumes execution from the monorepo root.

```bash
# From the monorepo root
export STACK_PATH=/path/to/nvcf-self-managed-stack
export NVCF_TOKEN=$(jq -r .token ~/.nvcf-cli.state)

# Build the CLI under test
bazel build //src/clis/nvcf-cli:nvcf-cli
cp "$(bazel cquery --output=files //src/clis/nvcf-cli:nvcf-cli)" /tmp/nvcf-cli
```

--------------------------------

### Install Latest JSON-Patch

Source: https://github.com/nvidia/nvcf/blob/main/src/compute-plane-services/image-credential-helper/vendor/gopkg.in/evanphx/json-patch.v4/README.md

Use this command to get the latest version of the jsonpatch library for Go.

```bash
go get -u github.com/evanphx/json-patch/v5
```

--------------------------------

### Example: Role Creation and Privilege Grant

Source: https://github.com/nvidia/nvcf/blob/main/migrations/cassandra/keyspaces/README.md

This script initializes application roles, grants necessary privileges, and sets the service login password using an environment variable.

```sql
CREATE ROLE service WITH PASSWORD = '${SERVICE_ROLE_PASSWORD}' NOSUPERUSER INHERIT CREATEROLE;
GRANT ALL ON KEYSPACE "api_keys_api" TO service;

```

--------------------------------

### Example run.py for PyTriton Server

Source: https://github.com/nvidia/nvcf/blob/main/docs/user/container-functions.md

Defines a PyTriton model with an inference function and server setup.

```python
import numpy as np
from pytriton.model_config import ModelConfig, Tensor
from pytriton.triton import Triton, TritonConfig
import time
....
class PyTritonServer:
    """triton server for timed_sleeper"""

    def __init__(self):
        # basically need to accept image, mask(PIL Images), prompt, negative_prompt(str), seed(int)
        self.model_name = "timed_sleeper"

    def _infer_fn(self, requests):
        responses = []
        for req in requests:
            req_data = req.data
            sleep_duration = numpy_array_to_variable(req_data.get("sleep_duration"))
            # deal with header dict keys being lowerscale
            request_parameters_dict = uppercase_keys(req.parameters)
            time.sleep(sleep_duration)
            responses.append({"sleep_duration": np.array([sleep_duration])})

        return responses

    def run(self):
        """run triton server"""
        with Triton(
            config=TritonConfig(
                http_header_forward_pattern="NVCF-*",  # this is required
                http_port=8000,
                grpc_port=8001,
                metrics_port=8002,
            )
        ) as triton:
            triton.bind(
                model_name="timed_sleeper",
                infer_func=self._infer_fn,
                inputs=[
                    Tensor(name="sleep_duration", dtype=np.uint32, shape=(1,)),
                ],
                outputs=[Tensor(name="sleep_duration", dtype=np.uint32, shape=(1,))],
                config=ModelConfig(batching=False),
            )
            triton.serve()
if __name__ == "__main__":
    server = PyTritonServer()
    server.run()
```

--------------------------------

### Uninstall Helm Releases

Source: https://github.com/nvidia/nvcf/blob/main/docs/user/example-dashboards.md

Remove the installed Helm releases for both the example dashboards and the observability reference stack.

```bash
helm uninstall nvcf-example-dashboards -n observability
helm uninstall observability -n observability
```

--------------------------------

### Basic Object Store Operations

Source: https://github.com/nvidia/nvcf/blob/main/src/compute-plane-services/nvca/vendor/github.com/nats-io/nats.go/jetstream/README.md

Demonstrates creating an object store, putting objects as strings, bytes, or files, retrieving objects, and deleting objects and the store.

```go
js, _ := jetstream.New(nc)
ctx := context.Background()

// Create a new bucket. Bucket name is required and has to be unique within a JetStream account.
os, _ := js.CreateObjectStore(ctx, jetstream.ObjectStoreConfig{Bucket: "configs"})

config1 := bytes.NewBufferString("first config")
// Put an object in a bucket. Put expects an object metadata and a reader
// to read the object data from.
os.Put(ctx, jetstream.ObjectMeta{Name: "config-1"}, config1)

// Objects can also be created using various helper methods

// 1. As raw strings
os.PutString(ctx, "config-2", "second config")

// 2. As raw bytes
os.PutBytes(ctx, "config-3", []byte("third config"))

// 3. As a file
os.PutFile(ctx, "config-4.txt")

// Get an object
// Get returns a reader and object info
// Similar to Put, Get can also be used with helper methods
// to retrieve object data as a string, bytes or to save it to a file
object, _ := os.Get(ctx, "config-1")
data, _ := io.ReadAll(object)
info, _ := object.Info()

// Prints `configs.config-1 -> "first config"`
fmt.Printf("%s.%s -> %q\n", info.Bucket, info.Name, string(data))

// Delete an object.
// Delete will remove object data from stream, but object metadata will be kept
// with a delete marker.
os.Delete(ctx, "config-1")

// getting a deleted object will return an error
_, err := os.Get(ctx, "config-1")
fmt.Println(err) // prints `nats: object not found`

// A bucket can be deleted once it is no longer needed
js.DeleteObjectStore(ctx, "configs")
```

--------------------------------

### Install Dependencies

Source: https://github.com/nvidia/nvcf/blob/main/src/compute-plane-services/byoo-otel-collector/generator/README.md

Run this command to synchronize and download project dependencies using uv.

```bash
uv sync
```