Thanks for FileSystem. I was thinking that it would be great to do another pass on the comments. I would like to start to convert the example in comment to use the new >>> convention.
I do not know how to push fixes using the new process. I will discuss with marcus to know if he knows. Once I know we will focus on the FileSystem fixes. Stef On Sun, Jun 18, 2017 at 5:10 PM, Alistair Grant <[email protected]> wrote: > Hi Stef, > > On Sat, Jun 17, 2017 at 08:48:36AM +0200, Stephane Ducasse wrote: >> Dear Pharoers >> >> there are many things that we would like to achieve to offer you a much >> better >> system but we cannot clone ourselves. >> >> Please let us know if you want to participate building better tools >> around for example Iceberg. >> >> Stef > > As there haven't been any other replies yet... > > I've been using Iceberg for a few months without any problems. I've > just started trying out Esteban's suggested PR process for Pharo 7 and > am starting to see issues. > > I've got some FileSystem patches which I would like to get in, but once > they're done I will hopefully have a chance to look at Iceberg. > > Just FYI, the FileSystem patches are: > > - FileRefence>>moveTo: works across disks > - FileReference>>/ is more intuitive, i.e. accept directory deliminators > in the paramter, and make path canonicalisation consistent (also > below). > - FileReference>>= considers files equal if they point to the same file > / directory. > - Path canonicalisation has been tightened up (there's still some room > for improvement). > - isSymlink works, and all attributes from the libc stat() / lstat() > calls are accessible (about 11 new attributes). > > And file attribute operations and directory enumarating are slightly > faster than before. This is a fairly major patch, introducing a new VM > plugin (which is why I waited for Pharo 7). > > And of course, thanks to everyone who has helped make Pharo what it is > today. > > Cheers, > Alistair >
