Python ConfigParser: read literal percent signs without hidden interpolation

Operation contract

A controlled export configuration contains a literal percent sign. With interpolation disabled, the application reads those bytes as text. A second configuration repeats the same option with different letter case; strict parsing rejects that duplicate before it can replace the first value.

Failure boundary

Disabling interpolation does not validate business values, protect secrets, or decide which configuration file is trusted. ConfigParser stores string values; parse and range-check each field before publishing settings to workers.

Working program

import configparser

settings = configparser.ConfigParser(interpolation=None, strict=True)
settings.read_string("[export]\npattern = paid%complete\nretention_days = 47\n")
print("pattern", settings["export"]["pattern"])
print("retention", settings.getint("export", "retention_days"))

duplicate = configparser.ConfigParser(interpolation=None, strict=True)
try:
    duplicate.read_string("[export]\nRetention = 47\nretention = 73\n")
except configparser.DuplicateOptionError:
    print("case_duplicate_rejected", True)

Output

pattern paid%complete
retention 47
case_duplicate_rejected True

Costs and limits

Configuration parsing allocates values proportional to the source text. The input file still needs an ownership and size policy before parsing.

Common Mistakes

• Default interpolation treats percent syntax as special.
• Option names are case-insensitive by default, which affects duplicate detection.
• getint converts text but does not enforce a domain range.

Connected lessons

• Python tomllib: parse configuration and validate its schema
• Python configuration precedence: merge, validate, then publish
• Python assert versus input validation: optimized execution changes the contract

Test this contract.