Documentation Index
Fetch the complete documentation index at: https://mintlify.com/acdeveloper-sci/forti4d/llms.txt
Use this file to discover all available pages before exploring further.
Forti4D is a pure-Python package with a single third-party dependency (loguru), so installation is straightforward on any platform. The sections below cover the standard package install, a source install for contributors, and the optional Graphviz setup needed for call graph rendering.
Requirements
- Python 3.8 or later — Forti4D is tested on Python 3.8+ on Windows, Linux, and macOS.
- Graphviz — optional; only required if you want to render the
.dot files produced by the visual_graph pipeline step into images (PNG, SVG, PDF).
loguru is the only third-party Python dependency. It is declared in Forti4D’s package metadata and is installed automatically — you do not need to install it separately.
Install from PyPI
Both commands install the forti4d CLI command and the forti4d Python package.
Verify the Installation
After installing, run forti4d --list to confirm the CLI is working and to see all 19 pipeline steps:
Expected output:
Step Module Description
──────────────────────────────────────────────────────────────────────────────────────────
inventory forti4d.analyzers.inventory Build unit inventory from source files
dependencies forti4d.analyzers.dependencies Build call graph and compute Fan-In/Fan-Out
profiler forti4d.analyzers.profiler Classify statements and produce audit/ DEBUG files
blocks block_analysis (in-process, batch) Block topology analysis (one file per source, to output/)
structure_analysis forti4d.analyzers.structure_analysis Classify files by architectural role
cross_analysis forti4d.analyzers.cross_analysis Assign migration strategy per unit
executive_summary forti4d.analyzers.executive_summary Generate executive summary
complexity forti4d.analyzers.complexity Compute McCabe cyclomatic complexity
common_blocks forti4d.analyzers.common_blocks Detect COMMON block coupling
symbols forti4d.analyzers.symbols Extract variable/parameter/implicit symbols per unit
derived_types forti4d.analyzers.derived_types Extract derived TYPE definitions and their components
equivalences forti4d.analyzers.equivalences Detect EQUIVALENCE aliasing groups (union-find)
reachability forti4d.analyzers.reachability Dead code detection from entry points
sloc forti4d.analyzers.sloc Precise SLOC count per unit
clones forti4d.analyzers.clones Detect identical/similar/diverged duplicate units
consolidate forti4d.analyzers.consolidate Join all reports into report_consolidated.csv
visual_graph forti4d.analyzers.visual_graph Generate call graph DOT files
prioritization forti4d.analyzers.prioritization Compute composite risk score and rank units for migration
html_report forti4d.analyzers.html_report Generate self-contained HTML report
FORT_SRC → tests/fixtures/
FORT_OUT → results/
--list also prints the active FORT_SRC and FORT_OUT values so you can confirm your environment is configured correctly before running a full analysis.
Install from Source (Contributors)
If you want to contribute to Forti4D or run the test suite, install in editable mode with the dev extras. Start by cloning the repository:
git clone https://github.com/acdeveloper-sci/forti4d.git
cd forti4d
uv pip install -e ".[dev]"
dev extra installs pytest>=7 for running the test suite.
You can also run the pipeline without installing at all:
# with uv
uv run forti4d --project /path/to/fortran/source --output out/
# with Python directly
python -m forti4d.pipeline --project /path/to/fortran/source --output out/
Graphviz (Optional)
Graphviz is only needed if you want to convert the .dot files produced by the visual_graph step into rendered images. If you skip this, visual_graph still runs and writes .dot files — you simply cannot render them locally without Graphviz installed.
macOS
Ubuntu / Debian
Windows
sudo apt-get install graphviz
Download and run the installer from graphviz.org/download, then ensure the bin/ directory is on your PATH. PATH:
