Re: [racket-dev] build: schemef.h -> schemex.h

2013-03-10 Thread Eli Barzilay
If you do that, then you can just as well go with `git archive', which will get you just the sources instead of a full clone. 30 minutes ago, Jon Rafkind wrote: > Yes, I'll make that change. > > On 03/10/2013 07:09 PM, Robby Findler wrote: > > Seems better to blow everything away and re-"gi

Re: [racket-dev] build: schemef.h -> schemex.h

2013-03-10 Thread Jon Rafkind
Yes, I'll make that change. On 03/10/2013 07:09 PM, Robby Findler wrote: > Seems better to blow everything away and re-"git clone" for an automated > build machine. > > Robby > > > On Sun, Mar 10, 2013 at 7:56 PM, Jon Rafkind > wrote: > > On 03/10/2013 06:14 PM, E

Re: [racket-dev] build: schemef.h -> schemex.h

2013-03-10 Thread Robby Findler
Seems better to blow everything away and re-"git clone" for an automated build machine. Robby On Sun, Mar 10, 2013 at 7:56 PM, Jon Rafkind wrote: > On 03/10/2013 06:14 PM, Eli Barzilay wrote: > > 20 minutes ago, Jon Rafkind wrote: > >> The ubuntu ppa build has been failing for a while, and I f

Re: [racket-dev] build: schemef.h -> schemex.h

2013-03-10 Thread Jon Rafkind
On 03/10/2013 06:14 PM, Eli Barzilay wrote: > 20 minutes ago, Jon Rafkind wrote: >> The ubuntu ppa build has been failing for a while, and I finally >> tracked it down to this issue: src/racket/src/schemef.h has a newer >> timestamp than src/racket/src/schemex.h on a fresh git checkout from >> gith

Re: [racket-dev] build: schemef.h -> schemex.h

2013-03-10 Thread Eli Barzilay
20 minutes ago, Jon Rafkind wrote: > The ubuntu ppa build has been failing for a while, and I finally > tracked it down to this issue: src/racket/src/schemef.h has a newer > timestamp than src/racket/src/schemex.h on a fresh git checkout from > github. This shouldn't be right. Git doesn't track m

[racket-dev] build: schemef.h -> schemex.h

2013-03-10 Thread Jon Rafkind
The ubuntu ppa build has been failing for a while, and I finally tracked it down to this issue: src/racket/src/schemef.h has a newer timestamp than src/racket/src/schemex.h on a fresh git checkout from github. This results in the make system trying to execute the following make[6]: Entering dir

[racket-dev] take/drop argument order

2013-03-10 Thread Eli Barzilay
(Sorry for the subject change, I wanted to separate it from the new *f things.) So it seems to me that most people agree that the current order for these things is bad, since I didn't hear from anyone who *likes* the current order. (Feel free to mail me off-list if you do.) Assuming this, here's

Re: [racket-dev] take, drop, split-at, etc -- and argument order, again

2013-03-10 Thread Eli Barzilay
9 hours ago, Laurent wrote: > Just a side question: > Would it be a bad idea to make a separate library that uses the > normal names without f, so that people can write > #lang racket > (require racket/flist) > > ? > or even use `prefix-in' if they prefer? Something that comes up from time to tim

Re: [racket-dev] take, drop, split-at, etc -- and argument order, again

2013-03-10 Thread Matthias Felleisen
I like this idea a lot. If you start from racket/base, you can then 'graduate' to a saner Racket gradually. -- Matthias On Mar 10, 2013, at 7:07 AM, Laurent wrote: > Just a side question: > Would it be a bad idea to make a separate library that uses the normal names > without f, so that peo

Re: [racket-dev] take, drop, split-at, etc -- and argument order, again

2013-03-10 Thread Laurent
Just a side question: Would it be a bad idea to make a separate library that uses the normal names without f, so that people can write #lang racket (require racket/flist) ? or even use `prefix-in' if they prefer? Laurent On Sat, Mar 9, 2013 at 4:28 PM, Eli Barzilay wrote: > Two days ago, As