Hi Niklas,<div><br></div><div>The function writeFile takes a FilePath. We could fork base or tell everyone to use writeFile2, but in practice everyone will keep using writeFile, and this String for FilePath. This approach is the only thing we could figure that made sense.</div><div><br></div><div>Henning: we do not propose normalisation on initialisation. For ASCII strings fromFilePath . toFilePath will be id. It might also be for unicode on some/all platforms. Of course, you can write your own FilePath creator that does normalisation on construction. <span></span></div><div><br></div><div>Thanks, Neil <br><br>On Saturday, 27 June 2015, Niklas Larsson <<a href="javascript:_e(%7B%7D,'cvml','metaniklas@gmail.com');" target="_blank">metaniklas@gmail.com</a>> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div><div><div style="font-family:Calibri,sans-serif;font-size:11pt">Hi!<br><br>Instead of trying to minimally patch the existing API and still breaking loads of code, why not make a new API that doesn't have to compromise and depreciate the old one?<br><br>Niklas </div></div><div dir="ltr"><hr><span style="font-family:Calibri,sans-serif;font-size:11pt;font-weight:bold">Från: </span><span style="font-family:Calibri,sans-serif;font-size:11pt"><a>Herbert Valerio Riedel</a></span><br><span style="font-family:Calibri,sans-serif;font-size:11pt;font-weight:bold">Skickat: </span><span style="font-family:Calibri,sans-serif;font-size:11pt">‎2015-‎06-‎26 18:09</span><br><span style="font-family:Calibri,sans-serif;font-size:11pt;font-weight:bold">Till: </span><span style="font-family:Calibri,sans-serif;font-size:11pt"><a>libraries@haskell.org</a>; <a>ghc-devs@haskell.org</a></span><br><span style="font-family:Calibri,sans-serif;font-size:11pt;font-weight:bold">Ämne: </span><span style="font-family:Calibri,sans-serif;font-size:11pt">Abstract FilePath Proposal</span><br><br></div>-----BEGIN PGP SIGNED MESSAGE-----<br>Hash: SHA1<br><br>Hello *,<br><br>What?<br>=====<br><br>We (see From: & CC: headers) propose, plain and simple, to turn the<br>currently defined type-synonym<br><br>  type FilePath = String<br><br>into an abstract/opaque data type instead.<br><br>Why/How/When?<br>=============<br><br>For details (including motivation and a suggested transition scheme)<br>please consult<br><br>  <a href="https://ghc.haskell.org/trac/ghc/wiki/Proposal/AbstractFilePath" target="_blank">https://ghc.haskell.org/trac/ghc/wiki/Proposal/AbstractFilePath</a><br><br><br><br>Suggested discussion period: 4 weeks<br>-----BEGIN PGP SIGNATURE-----<br>Version: GnuPG v1<br><br>iQIcBAEBAgAGBQJVjXkZAAoJELo8uj/+IrV0WXUP/0romoKazwLbQpaMAKgCNZon<br>BsY8Di44w6rkbdBXoky0xZooII8LJJyQfexH0BLRYEVLZFy0+LB8XzpPt8Ekg526<br>YlY4x0qFm9oiJbJDMqHUnb6z6Lr2KxzBcV37drTPbltUA+HB49DUVkkPbvHimpL2<br>28SIyhAr4fN6fLpGcFAkv6Rcs0mkvnTp7vsC0HNyshmGi6qQ+C+eB4mklQzWOPcn<br>koHZ2wtI8AJmyTdHKcXKAIFM0r+xl4MJ5445IvDjvIuGXZCzybXMw9Ss/4wSG3VN<br>qSIJVEDGZXrBCc12fPxPEB0Bqx9MIVytjplXKIo8rFrk93h3at9t9kDM26z+9PZ5<br>KYnEdjRKF4KL4j+3xqJDOEJT15GVRbGRRzb9A8xH0YIQ0S3Q3pt1PAfla1Hss75+<br>NRQgfowZYryL9dfCkAj2XNfdQ+pUk25N3bNig11se+zjk2JO77QRM0u3GOYZ9+CU<br>tSlwhtIMF32xnjgQyWE5yBBiEg3/Y+S+809tVaPseUEzkQJXMGq5TFxBrN6bj1Vm<br>awr6QghThKjeoRwky5bmFn/gept/lbYN6VV5B6gNznGP5xgFrmvVtmjbQJBRMYCv<br>aEUnrYqxkkbIddJjD5gl771/LWH4M2F1yBgJjfiZw2paEVAXKxEr327LsbOQaPdb<br>HjIPRrJbVK9AABo4AZ/Y<br>=lg0o<br>-----END PGP SIGNATURE-----<br>_______________________________________________<br>ghc-devs mailing list<br><a>ghc-devs@haskell.org</a><br><a href="http://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-devs" target="_blank">http://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-devs</a><br></div></blockquote></div>