Configuration

iec-checker can be configured with a JSON file named iec_checker.json. When no configuration file is present, built-in defaults are used and the analyzer behaves identically to previous versions.

Config file discovery

iec-checker looks for iec_checker.json starting from the current working directory and walking up to the filesystem root. The first file found wins. You can also pass an explicit path:

iec_checker --config path/to/iec_checker.json src/

Generating a default config

iec_checker --generate-config

This writes iec_checker.json with all default values to the current directory. You can then edit the parts you need.

Inspecting the effective config

iec_checker --dump-config

Prints the merged configuration (defaults + config file + CLI overrides) as JSON and exits. Useful for debugging.

CLI precedence

When the same option is set in both the config file and on the command line, CLI arguments always win.

For example, if iec_checker.json sets "format": "plain" in the output section, but you run:

iec_checker -o json src/

The output will be JSON.

Reference

detectors

Control which checks run.

Field	Type	Default	Description
disabled	string[]	[]	Detector IDs to skip. All other detectors run normally.
enabled	string[]	[]	Detector IDs to run exclusively. Mutually exclusive with disabled.

Use iec_checker --list-checks to see all available detector IDs.

Example -- disable two checks:

{
  "detectors": {
    "disabled": ["PLCOPEN-L17", "PLCOPEN-L10"]
  }
}

Available detector IDs

Registered detectors: PLCOPEN-CP1, PLCOPEN-CP2, PLCOPEN-CP3, PLCOPEN-CP4, PLCOPEN-CP6, PLCOPEN-CP8, PLCOPEN-CP9, PLCOPEN-CP13, PLCOPEN-CP25, PLCOPEN-CP28, PLCOPEN-L10, PLCOPEN-L17, PLCOPEN-N1, PLCOPEN-N2, PLCOPEN-N3, PLCOPEN-N4, PLCOPEN-N5, PLCOPEN-N6, PLCOPEN-N8, PLCOPEN-N9, PLCOPEN-N10.

Built-in analysis passes: DeclarationAnalysis, UnusedVariable, UseDefine.

thresholds

Tune complexity and size limits.

Field	Type	Default	Description
mccabe_complexity	int	15	McCabe cyclomatic complexity threshold (PLCOPEN-CP9)
statements_count	int	25	Maximum statements per POU (PLCOPEN-CP9)
max_string_length	int	4096	Maximum size of STRING / WSTRING data types

output

Field	Type	Default	Description
format	string	"plain"	Output format: "plain" or "json"
color	bool	true	Enable ANSI colors in plain output

input

Field	Type	Default	Description
format	string	"st"	Input format: "st", "xml", or "selxml"
merge	bool	false	Merge all input files before analysis
exclude_paths	string[]	[]	Glob patterns for files/directories to skip during traversal

Example -- exclude test and vendor directories:

{
  "input": {
    "exclude_paths": ["test/*", "vendor/*"]
  }
}

analysis

Field	Type	Default	Description
dump	bool	false	Create JSON dump files
verbose	bool	false	Show additional informational output

naming_conventions

Configure the PLCOpen-N family of naming-convention detectors. Every field in this section is opt-in — an empty map, a null case style, or a zero length bound disables the corresponding check and leaves the detector a no-op.

Field	Type	Default	Used by	Description
type_prefixes	{ string: string }	{}	PLCOPEN-N2	Map from elementary IEC 61131-3 type name to required identifier prefix (e.g. "INT": "i").
case	object	{}	PLCOPEN-N4	Per-kind case style. See below.
min_length	int	0	PLCOPEN-N6	Minimum identifier length. 0 disables the lower-bound check.
max_length	int	0	PLCOPEN-N6	Maximum identifier length. 0 disables the upper-bound check.
udt_prefixes	{ string: string }	{}	PLCOPEN-N10	Map from derived-type kind (STRUCT, ENUM, ARRAY, SUBRANGE, REF, ALIAS, FUNCTION_BLOCK) to required prefix.

case sub-section

Field	Type	Default	Description
variable	string | null	null	Required style for regular variable declarations (VAR, VAR_INPUT, etc.).
constant	string | null	null	Required style for declarations with the CONSTANT qualifier.
pou	string | null	null	Required style for function / function block / program / class / interface names.
type	string | null	null	Required style for user-defined type names (entries inside TYPE .. END_TYPE).

Recognized case styles:

• "UpperCamelCase" — e.g. StartMotor
• "lowerCamelCase" — e.g. startMotor
• "UPPER_SNAKE_CASE" — e.g. MAX_SPEED
• "lower_snake_case" — e.g. max_speed

Any other value (or a missing key) leaves the corresponding check off.

Example -- Hungarian prefixes, lower camel case for variables, upper snake case for constants, UDT prefixes by kind:

{
  "detectors": {
    "enabled": ["PLCOPEN-N2", "PLCOPEN-N4", "PLCOPEN-N6", "PLCOPEN-N10"]
  },
  "naming_conventions": {
    "type_prefixes": { "INT": "i", "BOOL": "x", "REAL": "r" },
    "case": {
      "variable": "lowerCamelCase",
      "constant": "UPPER_SNAKE_CASE",
      "pou":      "UpperCamelCase",
      "type":     "UpperCamelCase"
    },
    "min_length": 3,
    "max_length": 25,
    "udt_prefixes": {
      "STRUCT":         "ST_",
      "ENUM":           "E_",
      "FUNCTION_BLOCK": "FB_"
    }
  }
}

PLCOPEN-N1, PLCOPEN-N3, PLCOPEN-N5, PLCOPEN-N8 and PLCOPEN-N9 do not read from this section — they are always on when enabled and have nothing to tune.

Full example

{
  "detectors": {
    "disabled": ["PLCOPEN-L17", "PLCOPEN-L10"]
  },
  "thresholds": {
    "mccabe_complexity": 20,
    "statements_count": 50
  },
  "input": {
    "exclude_paths": ["test/*"]
  },
  "naming_conventions": {
    "type_prefixes": { "INT": "i", "BOOL": "x", "REAL": "r" },
    "case": {
      "variable": "lowerCamelCase",
      "pou":      "UpperCamelCase"
    },
    "min_length": 3,
    "max_length": 25,
    "udt_prefixes": {
      "ENUM":           "E_",
      "FUNCTION_BLOCK": "FB_"
    }
  }
}

Only the fields you want to override need to be present. Everything else falls back to defaults.