amclain has quit [Read error: Connection reset by peer]
amclain has joined #ponylang
jemc has quit [Ping timeout: 260 seconds]
c355e3b has quit [Quit: Connection closed for inactivity]
jemc has joined #ponylang
graaff has joined #ponylang
graaff has quit [Quit: Leaving]
amclain has quit [Quit: Leaving]
jemc has quit [Ping timeout: 258 seconds]
dinfuehr has quit [Ping timeout: 252 seconds]
dinfuehr has joined #ponylang
mrkishi has quit [Ping timeout: 265 seconds]
_andre has joined #ponylang
c355e3b has joined #ponylang
mrkishi has joined #ponylang
[CBR]Unspoken has quit [Ping timeout: 260 seconds]
jemc has joined #ponylang
[CBR]Unspoken has joined #ponylang
mrkishi has quit [Ping timeout: 265 seconds]
amclain has joined #ponylang
mrkishi has joined #ponylang
_andre has quit [Quit: leaving]
mrkishi has quit [Ping timeout: 244 seconds]
Praetonus has joined #ponylang
<andreaa>
is there a way to pass PCRE flags to Regex? (say 'gm' :)
mrkishi has joined #ponylang
<jemc>
andreaa: I notice that `Regex.replace` has a `global: Bool` parameter that affects the options passed to PCRE
<jemc>
other than that, I don't see many ways to affect the options
<jemc>
andreaa: I think it could be argued that the Regex class should be redesigned a bit to leverage the `collections.Flags` type to handle setting arbitrary PCRE flags
<andreaa>
jemc, Flags seems nice! is it essentially a collection of or-ed ints?
<andreaa>
jemc, it seems easy enough I could try to do that myself :D
<jemc>
somewhat recently we decided to adopt an RFC-based process for changes like this, to try to gather more community input for standard library design decisions
<andreaa>
I'm hoping to get away with a default argument also for Regex, so that the default behaviour should not change, but will definitely read that first
<jemc>
in some certain cases we may do an API change without an RFC if the current behaviour is significantly "wrong" in some way that makes it more "bug-like" than "feature-like"
<jemc>
in the case of your split question, it's possible that it could fall into the "API bug" category if the current behaviour is *too* surprising
<jemc>
however, if not, we might add the "turn into rfc" label to the ticket and explain why
<jemc>
I haven't looked at the issue in detail yet