| Message ID | 20141110111951.GA4185@free.fr (mailing list archive) |
|---|---|
| State | Rejected |
| Delegated to: | Herbert Xu |
| Headers | show |
Stéphane Aulery <saulery@free.fr> wrote: > > --IS0zKkzwUGydFO0o > Content-Type: text/plain; charset=utf-8 > Content-Disposition: inline > Content-Transfer-Encoding: 8bit > > forwarded 501566 dash@vger.kernel.org > stop > ------------------------------------------------------- > > Hello, > > Here is a small patch reported by a user of Debian [1]. Could you please > integrate? Thank you for your help. > > [1] https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=501566 > > Regards, > > -- > Stéphane Aulery > > --IS0zKkzwUGydFO0o > Content-Type: text/x-diff; charset=utf-8 > Content-Disposition: attachment; filename="dash.1.fix-redirection-docs.diff" > > --- dash.1.old 2014-11-10 12:15:29.809398575 +0100 > +++ dash.1.new 2014-11-10 12:17:03.401856125 +0100 > @@ -402,11 +402,11 @@ > .It [n] Ns \*[Lt] file > Redirect standard input (or n) from file. > .It [n1] Ns \*[Lt]& Ns n2 > -Duplicate standard input (or n1) from file descriptor n2. > +Duplicate standard input (or n1) to file descriptor n2. > .It [n] Ns \*[Lt]&- > Close standard input (or n). > .It [n1] Ns \*[Gt]& Ns n2 > -Duplicate standard output (or n1) to n2. > +Duplicate standard output (or n1) from n2. > .It [n] Ns \*[Gt]&- > Close standard output (or n). > .It [n] Ns \*[Lt]\*[Gt] file I'm sorry but this patch looks wrong and the text looks correct to me as is. Cheers,
Le lundi 10 novembre 2014 à 09:24:09, Herbert Xu a écrit : > Stéphane Aulery <saulery@free.fr> wrote: > > > > [1] https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=501566 > > > > I'm sorry but this patch looks wrong and the text looks correct > to me as is. Ok. I passed the information but I'm not really able to judge its pertinence. In this case I close this bug. Thank you for your help.
Hi, On 10 November 2014 14:49, Stéphane Aulery <saulery@free.fr> wrote: > Le lundi 10 novembre 2014 à 09:24:09, Herbert Xu a écrit : >> Stéphane Aulery <saulery@free.fr> wrote: >> > >> > [1] https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=501566 >> > >> >> I'm sorry but this patch looks wrong and the text looks correct >> to me as is. > > Ok. I passed the information but I'm not really able to judge its > pertinence. In this case I close this bug. So, i finally managed to figure out what the problem is. The wording is ambiguous (or maybe just insufficiently clear). "[n1]>&n2 Duplicate standard output (or n1) to n2." How i read it (6 years ago :P) was that meant dup2(n1, n2), because the only actual /duplication/ is of file descriptors. I've just realised that the intended meaning is "redirect n1 to where n2 is currently", i.e. dup2(n2, n1), as expected. Can i suggest a simple fix: "[n1]>&n2 Redirect standard output (or n1) to n2." It doesn't cover the case where n2 is later changed, and how n1 does not follow this, but it's the simplest fix i see. Steve
2014-11-11 14:23:21 +0100, Stephen Shirley: [...] > "[n1]>&n2 Redirect standard output (or n1) to n2." > > It doesn't cover the case where n2 is later changed, and how n1 does > not follow this, but it's the simplest fix i see. [...] That's ambiguous. Maybe: [n1]>&n2 Redirect standard output (or fd n1) to the same "open file description" as on fd n2. [n1]>&n2 Redirect standard output (or fd n1) to the same resource as open on fd n2. [n1]>&n2 Copy fd n2 as stdout (or fd n1) ?
On Tue, Nov 11, 2014 at 02:23:21PM +0100, Stephen Shirley wrote: > > "[n1]>&n2 Redirect standard output (or n1) to n2." > > It doesn't cover the case where n2 is later changed, and how n1 does > not follow this, but it's the simplest fix i see. Yes that sounds much better. Please send a patch. Thanks,
Hello, Le mardi 11 novembre 2014 à 11:08:05, Herbert Xu a écrit : > On Tue, Nov 11, 2014 at 02:23:21PM +0100, Stephen Shirley wrote: > > > > "[n1]>&n2 Redirect standard output (or n1) to n2." > > > > It doesn't cover the case where n2 is later changed, and how n1 does > > not follow this, but it's the simplest fix i see. > > Yes that sounds much better. Please send a patch. Here is one. Stephen, the correction of the first line is abandoned? [n1]<&n2 Duplicate standard input (or n1) from file descriptor n2. Regards,
Stephane Chazelas <stephane.chazelas@gmail.com> wrote: > 2014-11-11 14:23:21 +0100, Stephen Shirley: > [...] >> "[n1]>&n2 Redirect standard output (or n1) to n2." >> >> It doesn't cover the case where n2 is later changed, and how n1 does >> not follow this, but it's the simplest fix i see. > [...] > > That's ambiguous. > > Maybe: > > [n1]>&n2 Redirect standard output (or fd n1) to the same "open > file description" as on fd n2. > > [n1]>&n2 Redirect standard output (or fd n1) to the same > resource as open on fd n2. > > [n1]>&n2 Copy fd n2 as stdout (or fd n1) > > ? Stephane, please keep the cc list in future list replies as otherwise people may miss your messages since not everyone is subscribed and even those subscribed may filter out messages not addressed to them directly. Thanks,
Hello Stephen, Le mardi 11 novembre 2014 à 02:23:21, Stephen Shirley a écrit : > On 10 November 2014 14:49, Stéphane Aulery <saulery@free.fr> wrote: > > Le lundi 10 novembre 2014 à 09:24:09, Herbert Xu a écrit : > >> Stéphane Aulery <saulery@free.fr> wrote: > >> > > >> > [1] https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=501566 > >> > >> I'm sorry but this patch looks wrong and the text looks correct > >> to me as is. > > > > Ok. I passed the information but I'm not really able to judge its > > pertinence. In this case I close this bug. > > So, i finally managed to figure out what the problem is. The wording > is ambiguous (or maybe just insufficiently clear). > > "[n1]>&n2 Duplicate standard output (or n1) to n2." Herbert rejected both corrections. Your new proposal is better but incomplete. Stéphane Chazelas tried to complete it. Moreover you have not made a new proposal for the first correction. As I am not dash user, I do not have a good knowledge of the matter. Can you complete your answer / choice, please? Thank you in advance for your help. Original version: 1) [n1]<&n2 Duplicate standard input (or n1) from file descriptor n2. 2) [n1]>&n2 Duplicate standard output (or n1) to n2. Version on Debian bug tracker and refused by H. Xu: 1) [n1]<&n2 Duplicate standard input (or n1) to file descriptor n2. 2) [n1]>&n2 Duplicate standard output (or n1) from n2. Second version of the patch refused by H. Xu: 2) [n1]>&n2 Redirect standard output (or n1) to n2. Stéphane Chazelas proposals: 1a) [n1]>&n2 Redirect standard output (or fd n1) to the same "open file description" as on fd n2. 1b) [n1]>&n2 Redirect standard output (or fd n1) to the same resource as open on fd n2. 1c) [n1]>&n2 Copy fd n2 as stdout (or fd n1) Regards,
Hello Stephen and Herbert,
Stéphane Chazelas just made a complete proposal:
1) [n1]<&n2 Redirect standard input (or fd n1) from the same
resource as open on fd n2.
2b) [n1]>&n2 Redirect standard output (or fd n1) to
the same resource as open on fd n2.
What do you think about?
Regards,
2014-11-19 00:46:03 +0100, Stéphane Aulery: > Hello Stephen and Herbert, > > Stéphane Chazelas just made a complete proposal: > > > 1) [n1]<&n2 Redirect standard input (or fd n1) from the same > resource as open on fd n2. > > 2b) [n1]>&n2 Redirect standard output (or fd n1) to > the same resource as open on fd n2. > > > What do you think about? [...] A final touch (add "currently"). [n1]<&n2 Redirect standard input (or fd n1) from the same resource as currently open on fd n2. [n1]>&n2 Redirect standard output (or fd n1) to the same resource as currently open on fd n2.
--- dash.1.old 2014-11-10 12:15:29.809398575 +0100 +++ dash.1.new 2014-11-10 12:17:03.401856125 +0100 @@ -402,11 +402,11 @@ .It [n] Ns \*[Lt] file Redirect standard input (or n) from file. .It [n1] Ns \*[Lt]& Ns n2 -Duplicate standard input (or n1) from file descriptor n2. +Duplicate standard input (or n1) to file descriptor n2. .It [n] Ns \*[Lt]&- Close standard input (or n). .It [n1] Ns \*[Gt]& Ns n2 -Duplicate standard output (or n1) to n2. +Duplicate standard output (or n1) from n2. .It [n] Ns \*[Gt]&- Close standard output (or n). .It [n] Ns \*[Lt]\*[Gt] file