Lexarithmos is a command-line research application that converts Greek words, phrases, and sentences into numerical representations. It normalizes the input, applies a fixed Greek letter-to-value mapping, calculates the total value, and repeatedly reduces its digits until a single-digit result is reached.

The application can compute a phrase without changing the dataset, insert its analysis into a JSON store, or delete an existing phrase. After each storage change, a modular analysis pipeline rebuilds a second index that groups the stored values with their existing digit permutations.

For the Greek word:

The accents are removed, the text is converted to uppercase, and the supported Greek characters are mapped to their numerical values. The total value 996 is then reduced by repeatedly summing its digits:

9 + 9 + 6 = 24
2 + 4     = 6

flowchart TD
    A["CLI · app/"] --> B["PhraseService · services/"]
    B --> C["Transformer · core/"]
    C --> D["PhraseAnalysis"]
    B --> E["NumberRepository · storage/"]
    E --> F["number_file.json"]
    E --> G["AnalysisPipeline"]
    G --> H["PermutationAnalyzer"]
    H --> I["permutation_file.json"]

The following output was produced by the current command-line application using compute-only mode:

Type your phrase (or 'quit' to exit):
--> πληροφορική

#############################################
# Action list                               #
# -----------                               #
#                                           #
# => Insert: [i] [insert] [2].              #
# => Delete: [d] [delete] [3].              #
# => Enter: Computes only.                  #
#############################################

-->
[Status]: Success
[Action]: compute_only
[Message]: Phrase computed successfully.
[Phrase]: πληροφορική
[Normalized Words]: ['ΠΛΗΡΟΦΟΡΙΚΗ']
[Total Value]: 996
[Subdivisions]: [996, 24, 6]

Lexarithmos uses only the Python standard library; no third-party packages are required.

• Python 3.10 or newer
• Git

git clone https://github.com/3ConstArt3/Lexarithmos.git
cd Lexarithmos
python3 -m app.main

If your environment exposes Python through the python command, use python -m app.main instead.

The current transformation API can also be used directly:

from core.transformer import Transformer

analysis = Transformer().analyze_message("πληροφορική")

print(analysis.normalized_words)  # ['ΠΛΗΡΟΦΟΡΙΚΗ']
print(analysis.total_value)       # 996
print(analysis.subdivisions)      # [996, 24, 6]

Phrase analyses are grouped by their total value in data/number_file.json. The current schema uses sub-divisions for the reduced values and key-phrases for the phrases stored under the same total.

{
    "996": {
        "sub-divisions": [
            24,
            6
        ],
        "key-phrases": [
            "πληροφορική"
        ]
    }
}

The example above is intentionally shortened to one phrase while preserving the current file structure.

{
    "699": {
        "sub-divisions": [
            24,
            6
        ],
        "permutations": {
            "699": ["..."],
            "969": ["..."],
            "996": ["πληροφορική"]
        }
    }
}

Lexarithmos/
├── analysis/
│   ├── permutation_analyzer.py
│   └── pipeline/
│       ├── analyzer.py
│       ├── analysis_pipeline.py
│       └── default_pipeline.py
├── app/
│   ├── cli.py
│   └── main.py
├── config/
│   └── paths.py
├── core/
│   ├── actions.py
│   ├── digit_reducer.py
│   ├── exceptions.py
│   ├── lexarithmos_calculator.py
│   ├── models.py
│   ├── text_normalizer.py
│   └── transformer.py
├── data/
│   ├── number_file.json
│   └── permutation_file.json
├── services/
│   └── phrase_service.py
└── storage/
    ├── json_repository.py
    └── number_repository.py

The architecture separates interaction, domain logic, persistence, and derived analysis. Dependencies can be injected into PhraseService, while additional research analyzers can implement the Analyzer interface and be registered with the pipeline.

• Immutable results: PhraseAnalysis and PhraseResult are frozen data classes.
• Separated responsibilities: normalization, calculation, reduction, storage, and analysis are independent components.
• Safe JSON writes: repository updates are written to a temporary file before replacing the stored dataset.
• Extensible analysis: analyzers share an abstract interface and run through a configurable pipeline.
• No external runtime dependencies: the application relies only on the Python standard library.

The next major direction is a graphical interface that will make the dataset more approachable to non-technical users. A future visual layer could expose phrase transformations and permutation groups without requiring command-line interaction.

Contributions, suggestions, and research-oriented extensions are welcome. You can open an issue or submit a pull request with a clear explanation of the proposed change.

Lexarithmos is available under the MIT License.