jt Command-Line Tool

The jt tool is a powerful command-line utility for manipulating tree structures. It supports pattern matching, transformations, and multiple export formats. Version 1.0 has been completely rewritten with the modern AlgoTree API.

Installation

The jt tool is automatically installed with AlgoTree:

pip install AlgoTree
jt --version  # Should show 1.0.0

Basic Usage

# Read from file
jt tree.json

# Read from stdin
cat tree.json | jt

# Pretty print
jt tree.json --pretty

# Get statistics
jt tree.json --stats

Input Formats

jt can parse multiple input formats:

JSON Format

echo '{"name": "root", "children": [{"name": "child"}]}' | jt

DSL Format (Indent-based)

echo "root
  child1
  child2" | jt --input-format dsl

DSL Format (Visual)

cat << EOF | jt --input-format dsl
root
├── child1
└── child2
EOF

DSL Format (S-expression)

echo "(root (child1) (child2))" | jt --input-format dsl

Output Formats

JSON Output (default)

jt tree.json  # Default is JSON
jt tree.json --output json --indent 4  # Custom indent

Pretty Tree Output

jt tree.json --pretty
# Or
jt tree.json --output pretty

DSL Output

# Indent format (default)
jt tree.json --output dsl

# Visual format
jt tree.json --output dsl --dsl-format visual

# S-expression format
jt tree.json --output dsl --dsl-format sexpr

Filtering and Queries

Filter by Level

# Nodes deeper than level 2
jt tree.json --filter "level > 2"

# Nodes at level 1 or 0
jt tree.json --filter "level < 2"

Filter by Name

# Nodes containing "test" in name
jt tree.json --filter "name contains test"

# Nodes with specific attribute
jt tree.json --filter "type = 'folder'"

Custom Filter Expressions

# Lambda expressions
jt tree.json --filter "n.payload.get('size', 0) > 100"

Tree Navigation

Get Specific Node

jt tree.json --get node_name

Children and Parents

# Get children of a node
jt tree.json --children node_name

# Get parent of a node
jt tree.json --parent node_name

# Get siblings
jt tree.json --siblings node_name

Ancestors and Descendants

# All ancestors up to root
jt tree.json --ancestors node_name

# All descendants (subtree)
jt tree.json --descendants node_name

# Path from root to node
jt tree.json --path node_name

Tree Transformations

Map Operations

# Double all size values
jt tree.json --map "{'size': n.payload.get('size', 0) * 2}"

# Add computed field
jt tree.json --map "{'depth': n.level}"

Prune Operations

# Remove leaf nodes
jt tree.json --prune "n.is_leaf"

# Remove nodes with size < 10
jt tree.json --prune "n.payload.get('size', 0) < 10"

Tree Analysis

Basic Statistics

# Tree size (node count)
jt tree.json --size

# Tree height
jt tree.json --height

# All statistics
jt tree.json --stats

Leaf Nodes

# List all leaf nodes
jt tree.json --leaves

Format Conversion

JSON to DSL

# To indent format
jt tree.json --output dsl

# To visual format
jt tree.json --output dsl --dsl-format visual

DSL to JSON

cat tree.dsl | jt --input-format dsl --output json

Pretty Print Any Format

# Pretty print JSON
jt tree.json --pretty

# Pretty print DSL
cat tree.dsl | jt --input-format dsl --pretty

Advanced Examples

Pipeline Processing

# Filter, transform, and output
jt data.json \
  --filter "level > 1" \
  --map "{'value': n.payload.get('value', 0) * 2}" \
  --output pretty

Complex Queries

# Find all nodes with specific properties
jt org.json --filter "n.payload.get('dept') == 'Engineering' and n.level > 2"

Tree Building from Multiple Sources

# Combine multiple trees
(echo '{"name": "root", "children": ['; \
 jt tree1.json --get subtree1; echo ','; \
 jt tree2.json --get subtree2; \
 echo ']}') | jt --pretty

Comparison with v0.8

The v1.0 jt tool has been completely rewritten with:

  • New features: DSL support, fluent operations, better filtering

  • Better performance: Uses modern Node class internally

  • Cleaner output: Improved formatting and error messages

  • More formats: Visual tree, S-expressions, multiple DSL styles

  • Chainable operations: Filter, map, and prune in one command

Migration from v0.8

Most v0.8 commands still work, but some have been renamed:

  • --subtree → --descendants

  • --node-stats → --stats

  • --find-path → --path

Tips and Tricks

  1. Use stdin for pipelines:

    curl -s api.example.com/tree | jt --pretty

  2. Combine with jq for JSON processing:

    jt tree.json | jq '.children[0]'

  3. Save common filters as aliases:

    alias jt-leaves='jt --leaves'
alias jt-deep='jt --filter "level > 2"'

  4. Use for directory trees:

    find . -type d | python -c "..." | jt --pretty

See Also