Skip to content

DDS QoS XML Generator

A tool for automatically generating XML files for DDS QoS test cases. It produces efficient test cases using Pairwise testing, covering combinations of 16 parameters with minimal redundancy.

Table of Contents

Download

Download XML Generator Package (ZIP)

Parameter List

Overview

Parameter Pub Values Pub Value Types Sub Values Sub Value Types Constraint Combinations
ENTITY_FACTORY 2 True, False 2 True, False pub=sub 2
DATA_EXISTS 2 exists, not_exists 2 exists, not_exists pub=sub 2
RELIABILITY 2 RELIABLE, BEST_EFFORT 2 RELIABLE, BEST_EFFORT Independent 4
DURABILITY 4 TRANSIENT_LOCAL, VOLATILE, TRANSIENT, PERSISTENT 4 TRANSIENT_LOCAL, VOLATILE, TRANSIENT, PERSISTENT Independent 16
DEADLINE 3 PP, 2×PP, DURATION_INFINITY 3 PP, 2×PP, DURATION_INFINITY Independent 9
LIVELINESS 6 (AUTOMATIC, 0.5×PP), (AUTOMATIC, PP), (AUTOMATIC, 2×PP), (AUTOMATIC, ∞), (MANUAL_BY_PARTICIPANT, None), (MANUAL_BY_TOPIC, None) 6 (same) Independent 36
HISTORY 5 (KEEP_ALL, None), (KEEP_LAST, 1), (KEEP_LAST, <(RTT/PP)+2), (KEEP_LAST, =(RTT/PP)+2), (KEEP_LAST, >(RTT/PP)+2) 5 (same) Independent 25
RESOURCE_LIMITS_MAX_SAMPLES_PER_INSTANCE 4 1, <(RTT/PP)+2, =(RTT/PP)+2, >(RTT/PP)+2 4 (same) Independent 16
RESOURCE_LIMITS_MAX_SAMPLES 2 1, (RTT/PP+3)×PP 2 (same) Independent 4
LIFESPAN 2 0.5×RTT, DURATION_INFINITY 2 (same) Independent 4
OWNERSHIP 2 SHARED, EXCLUSIVE 2 (same) Independent 4
DESTINATION_ORDER 2 BY_RECEPTION_TIMESTAMP, BY_SOURCE_TIMESTAMP 2 (same) Independent 4
WRITER_DATA_LIFECYCLE 2 True, False 0 N/A (pub only) pub only 2
READER_DATA_LIFECYCLE_NO_WRITER 0 N/A (sub only) 2 0, 3 sub only 2
READER_DATA_LIFECYCLE_DISPOSED 0 N/A (sub only) 2 0, 3 sub only 2

Parameter Details

1. ENTITY_FACTORY

  • Constraint: pub and sub must always have the same value.
  • Combinations: (True, True), (False, False)

2. DATA_EXISTS

  • Description: Combined parameter for PARTITION, USER_DATA, GROUP_DATA, TOPIC_DATA.
  • Constraint: pub and sub must always have the same value.
  • Combinations: (exists, exists), (not_exists, not_exists)

3. RELIABILITY

  • Independent: pub and sub can be set independently.
  • Combinations: 2 × 2 = 4

4. DURABILITY

  • Independent: pub and sub can be set independently.
  • Combinations: 4 × 4 = 16

5. DEADLINE

  • Values:
  • PP: D < 2×PP case
  • 2×PP: D ≥ 2×PP case
  • DURATION_INFINITY: infinite
  • Independent: pub and sub can be set independently.
  • Combinations: 3 × 3 = 9

6. LIVELINESS

  • Values:
  • (AUTOMATIC, 0.5×PP): lease_duration < 2×PP
  • (AUTOMATIC, PP): lease_duration < 2×PP
  • (AUTOMATIC, 2×PP): lease_duration ≥ 2×PP
  • (AUTOMATIC, DURATION_INFINITY): infinite
  • (MANUAL_BY_PARTICIPANT, None): manual type
  • (MANUAL_BY_TOPIC, None): manual type
  • Independent: pub and sub can be set independently.
  • Combinations: 6 × 6 = 36

7. HISTORY

  • Values:
  • (KEEP_ALL, None): KEEP_ALL
  • (KEEP_LAST, 1): depth=1
  • (KEEP_LAST, &lt;(RTT/PP)+2): depth < (RTT/PP)+2
  • (KEEP_LAST, =(RTT/PP)+2): depth = (RTT/PP)+2
  • (KEEP_LAST, &gt;(RTT/PP)+2): depth > (RTT/PP)+2
  • Independent: pub and sub can be set independently.
  • Combinations: 5 × 5 = 25

8. RESOURCE_LIMITS_MAX_SAMPLES_PER_INSTANCE

  • Values: 1, &lt;(RTT/PP)+2, =(RTT/PP)+2, &gt;(RTT/PP)+2
  • Independent: pub and sub can be set independently.
  • Combinations: 4 × 4 = 16

9. RESOURCE_LIMITS_MAX_SAMPLES

  • Values: 1, (RTT/PP+3)×PP
  • Independent: pub and sub can be set independently.
  • Combinations: 2 × 2 = 4

10. LIFESPAN

  • Values: 0.5×RTT, DURATION_INFINITY
  • Independent: pub and sub can be set independently.
  • Combinations: 2 × 2 = 4

11. OWNERSHIP

  • Values: SHARED, EXCLUSIVE
  • Independent: pub and sub can be set independently.
  • Combinations: 2 × 2 = 4

12. DESTINATION_ORDER

  • Values: BY_RECEPTION_TIMESTAMP, BY_SOURCE_TIMESTAMP
  • Independent: pub and sub can be set independently.
  • Combinations: 2 × 2 = 4

13. WRITER_DATA_LIFECYCLE

  • Description: Applied to Publisher only.
  • Values: True, False
  • Combinations: 2

14. READER_DATA_LIFECYCLE_NO_WRITER

  • Description: Applied to Subscriber only.
  • Values: 0, 3
  • Combinations: 2

15. READER_DATA_LIFECYCLE_DISPOSED

  • Description: Applied to Subscriber only.
  • Values: 0, 3
  • Constraint: Must match READER_DATA_LIFECYCLE_NO_WRITER value.
  • Combinations: 2

Pairwise Test Case Generation

What is Pairwise Testing?

Pairwise testing generates a minimal set of test cases that cover every pair of parameter values at least once. This reduces the total number of cases while maintaining good coverage.

Generation Process

  1. Collect parameters

  2. Select only parameters that have pub or sub values among the 16 parameters.

  3. Generate combinations per parameter

  4. _get_param_values() builds pub/sub combinations for each parameter.

  5. Combination count is determined by constraints.

  6. Apply Pairwise algorithm

  7. With allpairspy (optional): Uses an optimized algorithm.

  8. Without allpairspy: Built-in simple Pairwise implementation:

    • Generate all parameter pairs: C(16,2) = 120 pairs
    • Build value combinations for each pair
    • Create test cases (default values + current pair values)
    • Remove duplicates

  9. Apply constraints

  10. READER_DATA_LIFECYCLE_NO_WRITER and READER_DATA_LIFECYCLE_DISPOSED must have the same value.
  11. Exclude combinations that violate this.

Example

Parameters A, B, C with 2, 3, and 2 values respectively:

  • Full combinations: 2 × 3 × 2 = 12
  • Pairwise cases: Minimal set covering all pairs (A–B, A–C, B–C) at least once (about 6–8 cases)

This yields effective test coverage with far fewer cases than exhaustive combination.

Final Case Count

Theoretical Full Combination Count

If all parameters were combined independently:

2 × 2 × 4 × 16 × 9 × 36 × 25 × 16 × 4 × 4 × 4 × 4 × 2 × 2 × 2
≈ 271 trillion

Actual Generated Case Count

With the Pairwise algorithm:

  • Generated cases: about 853 (for PP=0.1, RTT=0.2)
  • Efficiency: about 0.0000003% of full combination

How the Count is Computed

  1. Per-parameter combination count
  2. Independent: pub count × sub count

  3. Constrained: pub count (pub=sub) or sub count (pub-only/sub-only)

  4. Pairwise algorithm

  5. Build minimal set covering every parameter pair at least once.

  6. Actual count depends on parameter interactions.

  7. Constraint filtering

  8. Remove cases that violate READER_DATA_LIFECYCLE constraints.

Sample Run (PP=0.1, RTT=0.2)

Total: 853 test cases
XML files: 1,706 (853 cases × 2 files)
  - pub_qos_case_00001.xml ~ pub_qos_case_00853.xml
  - sub_qos_case_00001.xml ~ sub_qos_case_00853.xml

How to Run

Requirements

  • Python 3.6 or higher
  • Required: 
    pip install xml
  • Optional (optimized Pairwise): 
    pip install allpairspy

    The tool works without allpairspy using the built-in algorithm.

Project Structure

XML Generator/
├── xml_generator.py      # Main script
├── xml/
│   ├── pub_qos.xml       # Publisher template
│   └── sub_qos.xml       # Subscriber template
└── output/               # Generated XML files

Running the Generator

Option 1: Interactive

cd "XML Generator"
python3 xml_generator.py

You will be prompted for:

  • PP (Publication Period): in seconds
  • RTT (Round Trip Time): in seconds

Option 2: From code

from xml_generator import XMLGenerator

generator = XMLGenerator(pp=0.1, rtt=0.2)
generator.generate_all_test_cases(output_dir='output')

Option 3: Non-interactive (script)

cd "XML Generator"
python3 xml_generator.py <<< $'0.1\n0.2\n'

Sample Output

============================================================
DDS QoS XML Generator
============================================================
Enter PP (Publication Period) in seconds: 0.1
Enter RTT (Round Trip Time) in seconds: 0.2

Input values:
  PP: 0.1 s
  RTT: 0.2 s
  RTT/PP: 2.00
Generating test case combinations...
Total: 853 test cases

Generating XML files...
Progress: 100/853 (11%)
Progress: 200/853 (23%)
...
Progress: 800/853 (93%)

Done! 853 test cases generated.
Output directory: output/

Output Files

Generated XML files are written to output/:

  • Publisher: pub_qos_case_00001.xml ~ pub_qos_case_00853.xml
  • Subscriber: sub_qos_case_00001.xml ~ sub_qos_case_00853.xml

Each file contains the DDS QoS profile with parameter values for that test case.

Notes

  • PP: Must be greater than 0.
  • Overwrite: Existing files with the same name are overwritten.
  • Directory: output/ is created automatically if it does not exist.

Reference

Dynamic Value Calculation

Some parameter values depend on PP and RTT:

  • DEADLINE: PP, 2×PP
  • LIVELINESS: 0.5×PP, PP, 2×PP
  • HISTORY: based on (RTT/PP)+2
  • RESOURCE_LIMITS: based on (RTT/PP)+2
  • LIFESPAN: 0.5×RTT

Case count and values may change with different PP and RTT.

Constraints Summary

Constraint Description
ENTITY_FACTORY, DATA_EXISTS pub and sub must be equal
READER_DATA_LIFECYCLE NO_WRITER and DISPOSED must be equal
WRITER_DATA_LIFECYCLE Publisher only (sub is None)
READER_DATA_LIFECYCLE_* Subscriber only (pub is None)

License

This project is a tool for generating DDS QoS test cases.