Shivraj wrote:
> Would you know when was the [New Persian] word firangi used for the
> first time?
In the Middle Ages, certainly *after* the Crusades.
> In Iran I thought a similar word represents light violet color. In
> Hindi "rang" is color and firang could have something to do with
> color.
I guess, you must have actually in mind the New Persian word <fi:roza-rang> 'of turquoise colour' (= azure/blue)? AFAIK, there's no term "firang" meaning 'a colour' in Persian. In this language, <fi:roza> means 'a turquoise' (basic meaning: 'victorious'; the stone is, indeed, said to bring its wearer good luck), and <rang> means 'colour'. The Hindi adjective <fi:rozi:> 'violet-blue' and the noun <fi:roza:> 'a violet-blue gem' are loans from Persian.
You say that in Hindi "rang" (actually <ra~g>) means 'dye, colour', and that is true, yet the phonetically nearly identical New Persian term <rang> is not a loan from Hindi. The two words -- the one Indic, the other Iranian -- evolved independently starting from a Proto-Indo-Iranian root *raj-/rag- 'to colour, become red', which resulted, both in Proto-Indic and Proto-Iranian, into *ranga- and *ra:ga- 'dye, colour'.
Indic:
The Sanskrit verbal root ra~j- is first attested in the Atharvaveda; the derived participle <rakta> 'dyed, red', as well as the derived noun <ra:ga> '(red) colour, passion, love', are attested from the Brahmana period onwards; <ranga> 'dye, colour' (whence Hindi <ra~g>) and <ra:ga> 'colour, redness' are first attested in the Mahabharata.
Iranian:
Words deriving from this root (according to Bailey, probably both in the forms *ranga- and *ra:ga-) are attested starting from the Middle Iranian period only.
Zoaroastrian Pahlavi, New Persian <rang> 'colour'
Sogdian <rnk> 'colour'
Khwarezmian <rxtk> 'red' (compare with Sanskrit <rakta>)
Khotanese Saka <rra:Sa> 'dark-coloured'
Kind regards,
Francesco
PS More later on your other, stimulating questions about the Varangians and the Longobardic term <weregang>.