Skip to main content

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.

All path and parallelism settings in Forti4D flow through a single central module — config.py. It defines three configurable values: the Fortran source directory, the output directory, and the worker count for parallel steps. Each value is resolved at runtime using a consistent priority chain: an explicit CLI flag always wins, followed by the corresponding environment variable, and finally a hardcoded default. This means you can configure a project once via environment variables for a whole session, or override everything on a single command line without touching your shell.

​
Configuration Reference

Environment VariableCLI FlagDefaultDescription
FORT_SRC--project DIRtests/fixtures/Path to the directory containing the Fortran source files to analyze. Resolved to a pathlib.Path at startup.
FORT_OUT--output DIRresults/Root directory where all output files are written — CSVs, audit files, block topology files, DOT graphs, the HTML report, and the run log.
FORT_WORKERS--workers N1Number of parallel worker processes for the steps that support per-file parallelism (inventory, profiler, blocks). Must be a positive integer; non-integer values fall back silently to 1.

​
Priority Order

Each setting is resolved by config.resolve_paths() and config.resolve_workers() at call time using this fixed precedence:
1

Explicit CLI flag

Values passed directly to --project, --output, or --workers always take highest priority and override everything else.
2

Environment variable

If no CLI flag is provided, Forti4D reads FORT_SRC, FORT_OUT, or FORT_WORKERS from the current environment.
3

Hardcoded default

If neither a CLI flag nor an environment variable is present, the built-in defaults are used: tests/fixtures/ for source, results/ for output, and 1 for workers.
The --project and --output flags are the most explicit and portable way to configure a run. They require no shell setup and work identically across platforms: 
# Specify both source and output directories
forti4d --project /path/to/fortran --output /path/to/output/

# Enable parallel processing on top of explicit paths
forti4d --project /path/to/fortran --output /path/to/output/ --workers 4
When the pipeline runs, it mirrors --project and --output into FORT_SRC and FORT_OUT in the process environment, so every analyzer automatically picks up the correct paths without needing separate configuration.

​
Setting via Environment Variables

Environment variables are useful for one-off runs or when integrating Forti4D into shell scripts and CI pipelines: 
# One-off run against a specific project
FORT_SRC=/path/to/fortran FORT_OUT=results/ forti4d

# Set workers alongside path overrides
FORT_SRC=/path/to/fortran FORT_OUT=results/ FORT_WORKERS=8 forti4d

# Export variables for a whole shell session
export FORT_SRC=/home/user/legacy-project
export FORT_OUT=/home/user/analysis-results
forti4d          # uses the exported values
forti4d --list   # also uses them — shows active paths
For individual analyzer scripts invoked outside the pipeline, the same variables apply: 
FORT_SRC=/path/to/project FORT_OUT=results/ python3 -m forti4d.analyzers.inventory

​
Output Directory Creation

FORT_OUT (the results directory) is created automatically by the pipeline before any step runs. You do not need to create it manually. The call is equivalent to Path(results_dir).mkdir(parents=True, exist_ok=True), so nested paths like deep/nested/output/ work without pre-creating intermediate directories. Subdirectories required by individual steps — audit/ and blocks/ — are also created on demand by the steps that write to them.

​
Output Directory Layout

After a complete pipeline run, the results directory contains the following structure:

audit/

Per-file *_DEBUG.csv files produced by the profiler step. Each file contains a line-by-line statement classification for one source file. Required by the blocks step and several downstream analyzers.

blocks/

Per-file *_blocks.txt block topology reports produced by the blocks step. One file per source file processed during the profiler step.

graph_*.dot

Graphviz DOT files generated by the visual_graph step. Includes a full call graph, a simplified call graph, and one DOT file per entry point. Render with dot -Tsvg graph_full.dot -o graph_full.svg.

report.html

Self-contained HTML report produced by the html_report step. Contains a filterable and sortable unit table with all collected metrics. No external assets required — open directly in a browser.
The root of the results directory also holds all CSV report files (such as report_consolidated.csv, report_prioritization.csv, inventory_report.csv, and others), the executive summary (PROJECT_SUMMARY.md), and the run log (forti4d.log).
PathProduced byDescription
<FORT_OUT>/audit/profilerPer-file DEBUG CSV files, one per source file
<FORT_OUT>/blocks/blocksPer-file block topology text files
<FORT_OUT>/graph_*.dotvisual_graphGraphviz call graph DOT files
<FORT_OUT>/report.htmlhtml_reportSelf-contained HTML report
<FORT_OUT>/forti4d.logPipeline startupFull DEBUG-level run log
When running through the forti4d CLI, FORT_SRC and FORT_OUT are set automatically from --project and --output before any step executes. Individual analyzer scripts invoked outside the pipeline rely on the environment variables or the config.py defaults directly.

Build docs developers (and LLMs) love

Get started for freeTalk to us