[SwastikTripathi]

An ini file saved as "UTF-8 with BOM" (Notepad's default on older Windows, and an option in most editors) loses its first section header:

ini.decode('[section]\nkey=value\n')
// before: { '[section]': true, key: 'value' }
// after:  { section: { key: 'value' } }

fs.readFileSync(file, 'utf8') keeps the BOM, so the first line is [section], which doesn't match the ^\[ section pattern. It then falls through to the key branch, and unsafe() trims the BOM (JS trim() treats U+FEFF as whitespace), which is why it comes back as a key named [section]. Every key until the next header then ends up at the top level. Keys and comments on the first line are not affected for the same reason, only section headers.

decode now drops a leading U+FEFF before splitting lines. test/bom.js covers a BOM before a section, before CRLF sections, and before a top-level key; it fails without the change. npm test passes, with lint and 100% coverage.