Slice fails when ANSI string and ordinary string are mixed

[kjirou]

Hello, the following behaviors looked like a bug.

node
> chalk = require('chalk')
> sliceAnsi = require('slice-ansi')
> str = 'a' + chalk.red('bc') + 'd'
'a\u001b[31mbc\u001b[39md'
> sliceAnsi(str, 0, 1) === 'a'
false  # I expected true
> sliceAnsi(str, 1, 2) === chalk.red('b')
true
> sliceAnsi(str, 2, 3) === chalk.red('c')
false  # I expected true
> sliceAnsi(str, 3, 4) === 'd'
false  # I expected true

As a supplement, the example described in the README appeared to be working properly.

node
> str2 = 'The quick brown ' + chalk.red('fox jumped over ') + 'the lazy ' + chalk.green('dog and then ran away with the unicorn.');
'The quick brown \u001b[31mfox jumped over \u001b[39mthe lazy \u001b[32mdog and then ran away with the unicorn.\u001b[39m'
> sliceAnsi(str2, 20, 30) === chalk.red('jumped ove')
true

Thanks.

Environments

• node v6.8.0
• chalk v2.3.0
• slice-ansi v1.0.0

---

IssueHunt Summary

#### [ tiagodanin](https://issuehunt.io/u/tiagodanin) has been rewarded. ### Backers (Total: $60.00) - [ issuehunt](https://issuehunt.io/u/issuehunt) ($60.00) ### Submitted pull Requests - [#25 [WIP] Get all ansi escape sequences](https://issuehunt.io/r/chalk/slice-ansi/pull/25) --- ### Tips - Checkout the [Issuehunt explorer](https://issuehunt.io/r/chalk/slice-ansi/) to discover more funded issues. - Need some help from other developers? [Add your repositories](https://issuehunt.io/r/new) on IssueHunt to raise funds. --- IssueHunt has been backed by the following sponsors. [Become a sponsor](https://issuehunt.io/membership/members)

[sindresorhus]

Would you be able to submit a pull request with a failing test? :)

[kjirou]

@sindresorhus Pull-requested!

[kjirou]

Pull-requireted again with #16

[IssueHuntBot]

@issuehunt has funded $60.00 to this issue. See it on IssueHunt

[sindresorhus]

// @stroncium This one has a bounty too, in case you're interested ;)

[stroncium]

Haha, don't think of me as so materialistic. I actually don't think I will be able to even use bounties except funding some more issues(don't think paypal will allow me to accept any money due to my country of residence).

Anyhow, I actually wrote all the code for this issue already. The problem is the code is much bigger than whole this library as it required implementing first proper escape sequence tokenizer, then a virtual machine to process SGR commands, compare states and a compiler of state changes to commands which introduce them, plus I'd like to also implement a few other escape sequences which would make sense in terms of string manipulations. I also reckon this code would allow to implement a much simpler and correct version of wrap-ansi too. So, I'm thinking what to do with this code now.

[issuehunt-oss[bot]]

@sindresorhus has rewarded $54.00 to @tiagodanin. See it on IssueHunt

• :moneybag: Total deposit: $60.00
• :tada: Repository reward(0%): $0.00
• :wrench: Service fee(10%): $6.00