• dsimcha (20/20) Aug 10 2011 The review of the new std.path is drawing to a close and it looks like
• Steven Schveighoffer (5/9) Aug 10 2011 I just want to get this one straight, *Lars* is the main author, I just ...
• Marco Leise (6/15) Aug 10 2011 Please shed some light on the new std.process! Is this what I expect it ...
• Lars T. Kyllingstad (5/26) Aug 11 2011 Yes, it allows IPC through pipes. Each end of a pipe is an
• Jesse Phillips (4/9) Aug 10 2011 I've got a CSV parser which is close. I just want to get an unofficial
• Andrej Mitrovic (1/1) Aug 10 2011 Where is this new std.process btw.? Is it on github or somewhere?
• Vladimir Panteleev (7/8) Aug 10 2011 I believe this is it:
• Steven Schveighoffer (6/11) Aug 10 2011 Just to warn you, the windows port *does not work* because of a flaw in ...
• Nick Sabalausky (5/25) Aug 10 2011 I hope I'm not venturing too much into bikeshed territiory here, but FWI...
• dsimcha (10/30) Aug 12 2011 Ok, so in addition to the above, we also have Jesse Phillips's CSV
• Jimmy Cao (3/41) Aug 12 2011 That's reasonable. I also think that the Curl wrapper should be reviewe...
• Jonathan M Davis (6/52) Aug 12 2011 std.log has been in the review queue the longest, so I would say that th...
• Johannes Pfau (5/62) Aug 15 2011 bump
• Jesse Phillips (2/14) Aug 15 2011 I have no interest to cutting in line, but I'll mention that I went ahea...
• dsimcha (8/8) Aug 15 2011 I say give Jonas one more day to post the official review request for th...
• Jacob Carlborg (4/12) Aug 15 2011 How bad would it be to review more than one module?
• Jonathan M Davis (6/20) Aug 15 2011 Andrei is against it because it divides people's attention. There are on...
• jdrewsen (6/14) Aug 15 2011 I'm reading the forums - just been waiting for Jose to reply because he
• Jonathan M Davis (4/19) Aug 15 2011 Well, it seems that Jose has disappeared again, and we have enough stuff...

dsimcha <dsimcha yahoo.com> writes:

The review of the new std.path is drawing to a close and it looks like 
another success!  (Congratulations, Lars.)  Lately, though, the queue of 
stuff to review has been getting rather long, admittedly a problem we'd 
like to have.  I want to get a list of stuff that's ready or will be 
ready in short order, so we can prioritize.

I know at least the following are ready or fairly close.  Please list 
anything else:

std.regionallocator (A segmented stack memory allocator, by me.  I'd 
like to fast-track this because it's being used in the GSoC project I'm 
mentoring and getting it into Phobos would simplify things for us.)

std.parallelalgorithm (A small module of parallel implementations of 
various algorithms, by me.  This one can wait because it's pretty niche 
and nothing I know of depends on it.)

std.process (New and much improved, by Steve Schveighoffer and IIRC Lars 
Kylingstad also contributed.  Personally I'd like this to be given a 
high priority b/c the old std.process sucks so much and it's been 
waiting so long for compiler bugs to get fixed, etc.)

std.log  (Logging module by Jose Armando Garcia.)

curl wrapper (By Jonas Drewsen.)

std.variant (Major improvements, by Robert Jacques.)

"Steven Schveighoffer" <schveiguy yahoo.com> writes:

On Wed, 10 Aug 2011 09:22:46 -0400, dsimcha <dsimcha yahoo.com> wrote:

 std.process (New and much improved, by Steve Schveighoffer and IIRC Lars  
 Kylingstad also contributed.  Personally I'd like this to be given a  
 high priority b/c the old std.process sucks so much and it's been  
 waiting so long for compiler bugs to get fixed, etc.)

I just want to get this one straight, *Lars* is the main author, I just  
contributed a Windows implementation for his interface :)

BTW, this one has to be delayed until Walter fixes the DMC runtime...

-Steve

"Marco Leise" <Marco.Leise gmx.de> writes:

Am 10.08.2011, 15:51 Uhr, schrieb Steven Schveighoffer  
<schveiguy yahoo.com>:

 On Wed, 10 Aug 2011 09:22:46 -0400, dsimcha <dsimcha yahoo.com> wrote:

 std.process (New and much improved, by Steve Schveighoffer and IIRC  
 Lars Kylingstad also contributed.  Personally I'd like this to be given  
 a high priority b/c the old std.process sucks so much and it's been  
 waiting so long for compiler bugs to get fixed, etc.)

 I just want to get this one straight, *Lars* is the main author, I just  
 contributed a Windows implementation for his interface :)

 BTW, this one has to be delayed until Walter fixes the DMC runtime...

 -Steve

Please shed some light on the new std.process! Is this what I expect it  
is? Does it allow IPC through a pipe between parent and child process and  
reading/writing by line as well as single characters? Does it allow  
polling for new data?

"Lars T. Kyllingstad" <public kyllingen.NOSPAMnet> writes:

On Wed, 10 Aug 2011 17:25:44 +0200, Marco Leise wrote:

 Am 10.08.2011, 15:51 Uhr, schrieb Steven Schveighoffer
 <schveiguy yahoo.com>:
 
 On Wed, 10 Aug 2011 09:22:46 -0400, dsimcha <dsimcha yahoo.com> wrote:

 std.process (New and much improved, by Steve Schveighoffer and IIRC
 Lars Kylingstad also contributed.  Personally I'd like this to be
 given a high priority b/c the old std.process sucks so much and it's
 been waiting so long for compiler bugs to get fixed, etc.)

 I just want to get this one straight, *Lars* is the main author, I just
 contributed a Windows implementation for his interface :)

 BTW, this one has to be delayed until Walter fixes the DMC runtime...

 -Steve

 
 Please shed some light on the new std.process! Is this what I expect it
 is? Does it allow IPC through a pipe between parent and child process
 and reading/writing by line as well as single characters? Does it allow
 polling for new data?

Yes, it allows IPC through pipes.  Each end of a pipe is an 
std.stdio.File, and supports all operations of that type.

http://kyllingen.net/code/std-process/phobos-prerelease/std_process.html

-Lars

Jesse Phillips <jessekphillips+d gmail.com> writes:

On Wed, 10 Aug 2011 09:22:46 -0400, dsimcha wrote:

 The review of the new std.path is drawing to a close and it looks like
 another success!  (Congratulations, Lars.)  Lately, though, the queue of
 stuff to review has been getting rather long, admittedly a problem we'd
 like to have.  I want to get a list of stuff that's ready or will be
 ready in short order, so we can prioritize.

I've got a CSV parser which is close. I just want to get an unofficial 
review of the documentation, but I've thought of one thing that I should 
document better.

Andrej Mitrovic <andrej.mitrovich gmail.com> writes:

Where is this new std.process btw.? Is it on github or somewhere?

"Vladimir Panteleev" <vladimir thecybershadow.net> writes:

On Wed, 10 Aug 2011 18:07:53 +0300, Andrej Mitrovic  
<andrej.mitrovich gmail.com> wrote:

 Where is this new std.process btw.? Is it on github or somewhere?

I believe this is it:

https://github.com/kyllingstad/phobos/blob/new-std-process/std/process.d

-- 
Best regards,
  Vladimir                            mailto:vladimir thecybershadow.net

"Steven Schveighoffer" <schveiguy yahoo.com> writes:

On Wed, 10 Aug 2011 11:12:49 -0400, Vladimir Panteleev  
<vladimir thecybershadow.net> wrote:

 On Wed, 10 Aug 2011 18:07:53 +0300, Andrej Mitrovic  
 <andrej.mitrovich gmail.com> wrote:

 Where is this new std.process btw.? Is it on github or somewhere?

 I believe this is it:

 https://github.com/kyllingstad/phobos/blob/new-std-process/std/process.d

Just to warn you, the windows port *does not work* because of a flaw in  
how DMC handles pipes.

Other OSes should work.

-Steve

"Nick Sabalausky" <a a.a> writes:

"dsimcha" <dsimcha yahoo.com> wrote in message 
news:j1u0o8$2gdg$1 digitalmars.com...
 The review of the new std.path is drawing to a close and it looks like 
 another success!  (Congratulations, Lars.)  Lately, though, the queue of 
 stuff to review has been getting rather long, admittedly a problem we'd 
 like to have.  I want to get a list of stuff that's ready or will be ready 
 in short order, so we can prioritize.

 I know at least the following are ready or fairly close.  Please list 
 anything else:

 std.regionallocator (A segmented stack memory allocator, by me.  I'd like 
 to fast-track this because it's being used in the GSoC project I'm 
 mentoring and getting it into Phobos would simplify things for us.)

 std.parallelalgorithm (A small module of parallel implementations of 
 various algorithms, by me.  This one can wait because it's pretty niche 
 and nothing I know of depends on it.)

 std.process (New and much improved, by Steve Schveighoffer and IIRC Lars 
 Kylingstad also contributed.  Personally I'd like this to be given a high 
 priority b/c the old std.process sucks so much and it's been waiting so 
 long for compiler bugs to get fixed, etc.)

 std.log  (Logging module by Jose Armando Garcia.)

 curl wrapper (By Jonas Drewsen.)

 std.variant (Major improvements, by Robert Jacques.)

I hope I'm not venturing too much into bikeshed territiory here, but FWIW, 
I've been really, really looking forward to std.process and the curl 
wrapper.

dsimcha <dsimcha yahoo.com> writes:

On 8/10/2011 9:22 AM, dsimcha wrote:
 The review of the new std.path is drawing to a close and it looks like
 another success! (Congratulations, Lars.) Lately, though, the queue of
 stuff to review has been getting rather long, admittedly a problem we'd
 like to have. I want to get a list of stuff that's ready or will be
 ready in short order, so we can prioritize.

 I know at least the following are ready or fairly close. Please list
 anything else:

 std.regionallocator (A segmented stack memory allocator, by me. I'd like
 to fast-track this because it's being used in the GSoC project I'm
 mentoring and getting it into Phobos would simplify things for us.)

 std.parallelalgorithm (A small module of parallel implementations of
 various algorithms, by me. This one can wait because it's pretty niche
 and nothing I know of depends on it.)

 std.process (New and much improved, by Steve Schveighoffer and IIRC Lars
 Kylingstad also contributed. Personally I'd like this to be given a high
 priority b/c the old std.process sucks so much and it's been waiting so
 long for compiler bugs to get fixed, etc.)

 std.log (Logging module by Jose Armando Garcia.)

 curl wrapper (By Jonas Drewsen.)

 std.variant (Major improvements, by Robert Jacques.)

Ok, so in addition to the above, we also have Jesse Phillips's CSV 
parser.  The review for std.path has been successfully completed, so 
let's get another iron in the fire.

As far as stuff that's completely ready to go (no caveats, need for 
polishing, etc.), I know std.regionallocator and the curl wrapper are. 
If there's anything else, please speak up.  I'm willing to postpone 
std.regionallocator if the curl wrapper is ready to be reviewed right 
now, since I think the curl wrapper is more universally useful. 
Therefore, IMHO next in line should be the curl wrapper.

Jimmy Cao <jcao219 gmail.com> writes:

On Fri, Aug 12, 2011 at 7:53 AM, dsimcha <dsimcha yahoo.com> wrote:

 On 8/10/2011 9:22 AM, dsimcha wrote:

 The review of the new std.path is drawing to a close and it looks like
 another success! (Congratulations, Lars.) Lately, though, the queue of
 stuff to review has been getting rather long, admittedly a problem we'd
 like to have. I want to get a list of stuff that's ready or will be
 ready in short order, so we can prioritize.

 I know at least the following are ready or fairly close. Please list
 anything else:

 std.regionallocator (A segmented stack memory allocator, by me. I'd like
 to fast-track this because it's being used in the GSoC project I'm
 mentoring and getting it into Phobos would simplify things for us.)

 std.parallelalgorithm (A small module of parallel implementations of
 various algorithms, by me. This one can wait because it's pretty niche
 and nothing I know of depends on it.)

 std.process (New and much improved, by Steve Schveighoffer and IIRC Lars
 Kylingstad also contributed. Personally I'd like this to be given a high
 priority b/c the old std.process sucks so much and it's been waiting so
 long for compiler bugs to get fixed, etc.)

 std.log (Logging module by Jose Armando Garcia.)

 curl wrapper (By Jonas Drewsen.)

 std.variant (Major improvements, by Robert Jacques.)

 Ok, so in addition to the above, we also have Jesse Phillips's CSV parser.
  The review for std.path has been successfully completed, so let's get
 another iron in the fire.

 As far as stuff that's completely ready to go (no caveats, need for
 polishing, etc.), I know std.regionallocator and the curl wrapper are. If
 there's anything else, please speak up.  I'm willing to postpone
 std.regionallocator if the curl wrapper is ready to be reviewed right now,
 since I think the curl wrapper is more universally useful. Therefore, IMHO
 next in line should be the curl wrapper.

That's reasonable.  I also think that the Curl wrapper should be reviewed
next.

"Jonathan M Davis" <jmdavisProg gmx.com> writes:

On Friday, August 12, 2011 10:48 Jimmy Cao wrote:
 On Fri, Aug 12, 2011 at 7:53 AM, dsimcha <dsimcha yahoo.com> wrote:
 On 8/10/2011 9:22 AM, dsimcha wrote:
 The review of the new std.path is drawing to a close and it looks like
 another success! (Congratulations, Lars.) Lately, though, the queue of
 stuff to review has been getting rather long, admittedly a problem we'd
 like to have. I want to get a list of stuff that's ready or will be
 ready in short order, so we can prioritize.
 
 I know at least the following are ready or fairly close. Please list
 anything else:
 
 std.regionallocator (A segmented stack memory allocator, by me. I'd like
 to fast-track this because it's being used in the GSoC project I'm
 mentoring and getting it into Phobos would simplify things for us.)
 
 std.parallelalgorithm (A small module of parallel implementations of
 various algorithms, by me. This one can wait because it's pretty niche
 and nothing I know of depends on it.)
 
 std.process (New and much improved, by Steve Schveighoffer and IIRC Lars
 Kylingstad also contributed. Personally I'd like this to be given a high
 priority b/c the old std.process sucks so much and it's been waiting so
 long for compiler bugs to get fixed, etc.)
 
 std.log (Logging module by Jose Armando Garcia.)
 
 curl wrapper (By Jonas Drewsen.)
 
 std.variant (Major improvements, by Robert Jacques.)

 
 Ok, so in addition to the above, we also have Jesse Phillips's CSV
 parser.
 
 The review for std.path has been successfully completed, so let's get
 
 another iron in the fire.
 
 As far as stuff that's completely ready to go (no caveats, need for
 polishing, etc.), I know std.regionallocator and the curl wrapper are. If
 there's anything else, please speak up. I'm willing to postpone
 std.regionallocator if the curl wrapper is ready to be reviewed right
 now, since I think the curl wrapper is more universally useful.
 Therefore, IMHO next in line should be the curl wrapper.

 
 That's reasonable. I also think that the Curl wrapper should be reviewed
 next.

std.log has been in the review queue the longest, so I would say that that 
should be next except that when I posted a question about it a couple of days 
ago, he never responded. So, we should probably just review the curl wrapper 
next and get on with it.

- Jonathan M Davis

Johannes Pfau <spam example.com> writes:

Jonathan M Davis wrote:
On Friday, August 12, 2011 10:48 Jimmy Cao wrote:
 On Fri, Aug 12, 2011 at 7:53 AM, dsimcha <dsimcha yahoo.com> wrote:
 On 8/10/2011 9:22 AM, dsimcha wrote:
 The review of the new std.path is drawing to a close and it looks
 like another success! (Congratulations, Lars.) Lately, though,
 the queue of stuff to review has been getting rather long,
 admittedly a problem we'd like to have. I want to get a list of
 stuff that's ready or will be ready in short order, so we can
 prioritize.
 
 I know at least the following are ready or fairly close. Please
 list anything else:
 
 std.regionallocator (A segmented stack memory allocator, by me.
 I'd like to fast-track this because it's being used in the GSoC
 project I'm mentoring and getting it into Phobos would simplify
 things for us.)
 
 std.parallelalgorithm (A small module of parallel implementations
 of various algorithms, by me. This one can wait because it's
 pretty niche and nothing I know of depends on it.)
 
 std.process (New and much improved, by Steve Schveighoffer and
 IIRC Lars Kylingstad also contributed. Personally I'd like this
 to be given a high priority b/c the old std.process sucks so much
 and it's been waiting so long for compiler bugs to get fixed,
 etc.)
 
 std.log (Logging module by Jose Armando Garcia.)
 
 curl wrapper (By Jonas Drewsen.)
 
 std.variant (Major improvements, by Robert Jacques.)

 
 Ok, so in addition to the above, we also have Jesse Phillips's CSV
 parser.
 
 The review for std.path has been successfully completed, so let's
 get
 
 another iron in the fire.
 
 As far as stuff that's completely ready to go (no caveats, need for
 polishing, etc.), I know std.regionallocator and the curl wrapper
 are. If there's anything else, please speak up. I'm willing to
 postpone std.regionallocator if the curl wrapper is ready to be
 reviewed right now, since I think the curl wrapper is more
 universally useful. Therefore, IMHO next in line should be the
 curl wrapper.

 
 That's reasonable. I also think that the Curl wrapper should be
 reviewed next.

std.log has been in the review queue the longest, so I would say that
that should be next except that when I posted a question about it a
couple of days ago, he never responded. So, we should probably just
review the curl wrapper next and get on with it.

- Jonathan M Davis

bump
(Just so that this doesn't get lost)

-- 
Johannes Pfau

Jesse Phillips <jessekphillips+D gmail.com> writes:

Johannes Pfau Wrote:

std.log has been in the review queue the longest, so I would say that
that should be next except that when I posted a question about it a
couple of days ago, he never responded. So, we should probably just
review the curl wrapper next and get on with it.

- Jonathan M Davis

 
 bump
 (Just so that this doesn't get lost)
 
 -- 
 Johannes Pfau

I have no interest to cutting in line, but I'll mention that I went ahead and
made the CSV module exactly how I think it should be and created a branch to
include it. So I say it is ready for a true ridicule upon request.

dsimcha <dsimcha yahoo.com> writes:

I say give Jonas one more day to post the official review request for the Curl
wrapper, and if he doesn't then we can move CSV parser ahead.

1.  We've got an embarrassment of riches with how long the review queue is
getting.  It's a shame to have so many bubbles in it.

2.  If Jonas is busy/out of town/otherwise not reading the forums, then this
isn't
a good time to review the curl wrapper anyhow.

3.  We'd only be pushing the Curl wrapper back one spot, to behind the CSV
parser,
so it's not like we'd be severely devaluing it.

Jacob Carlborg <doob me.com> writes:

On 2011-08-15 20:19, dsimcha wrote:
 I say give Jonas one more day to post the official review request for the Curl
 wrapper, and if he doesn't then we can move CSV parser ahead.

 1.  We've got an embarrassment of riches with how long the review queue is
 getting.  It's a shame to have so many bubbles in it.

 2.  If Jonas is busy/out of town/otherwise not reading the forums, then this
isn't
 a good time to review the curl wrapper anyhow.

 3.  We'd only be pushing the Curl wrapper back one spot, to behind the CSV
parser,
 so it's not like we'd be severely devaluing it.

How bad would it be to review more than one module?

-- 
/Jacob Carlborg

Jonathan M Davis <jmdavisProg gmx.com> writes:

On Monday, August 15, 2011 20:52:46 Jacob Carlborg wrote:
 On 2011-08-15 20:19, dsimcha wrote:
 I say give Jonas one more day to post the official review request for
 the Curl wrapper, and if he doesn't then we can move CSV parser ahead.
 
 1.  We've got an embarrassment of riches with how long the review queue
 is getting.  It's a shame to have so many bubbles in it.
 
 2.  If Jonas is busy/out of town/otherwise not reading the forums, then
 this isn't a good time to review the curl wrapper anyhow.
 
 3.  We'd only be pushing the Curl wrapper back one spot, to behind the
 CSV parser, so it's not like we'd be severely devaluing it.

 
 How bad would it be to review more than one module?

Andrei is against it because it divides people's attention. There are only so 
many eyeballs, and if you review multiple items at once, one or both suffer 
from not getting as much review as they would have were they to be reviewed 
separately.

- Jonathan M Davis

jdrewsen <jdrewsen nospam.com> writes:

Den 15-08-2011 20:19, dsimcha skrev:
 I say give Jonas one more day to post the official review request for the Curl
 wrapper, and if he doesn't then we can move CSV parser ahead.

 1.  We've got an embarrassment of riches with how long the review queue is
 getting.  It's a shame to have so many bubbles in it.

 2.  If Jonas is busy/out of town/otherwise not reading the forums, then this
isn't
 a good time to review the curl wrapper anyhow.

I'm reading the forums - just been waiting for Jose to reply because he 
is before the curl wrapper in the review queue. But I guess your post 
indicates that it is probably ok to go ahead with the curl review.

I will post the request tomorrow since I going to bed in soon.

/Jonas

 3.  We'd only be pushing the Curl wrapper back one spot, to behind the CSV
parser,
 so it's not like we'd be severely devaluing it.

"Jonathan M Davis" <jmdavisProg gmx.com> writes:

On Monday, August 15, 2011 13:34 jdrewsen wrote:
 Den 15-08-2011 20:19, dsimcha skrev:
 I say give Jonas one more day to post the official review request for the
 Curl wrapper, and if he doesn't then we can move CSV parser ahead.
 
 1. We've got an embarrassment of riches with how long the review queue
 is getting. It's a shame to have so many bubbles in it.
 
 2. If Jonas is busy/out of town/otherwise not reading the forums, then
 this isn't a good time to review the curl wrapper anyhow.

 
 I'm reading the forums - just been waiting for Jose to reply because he
 is before the curl wrapper in the review queue. But I guess your post
 indicates that it is probably ok to go ahead with the curl review.
 
 I will post the request tomorrow since I going to bed in soon.

Well, it seems that Jose has disappeared again, and we have enough stuff in 
the review queue that there's no reason to wait for him.

- Jonathan M Davis