Bug #55502 for PerlIO-subfile: binmode() fix

Fri Mar 12 12:02:49 2010 SAFRONOV [...] cpan.org - Ticket created

Subject:

binmode() fix

Only in PerlIO-subfile-0.08-my: Makefile.old diff -u PerlIO-subfile-0.08/subfile.xs PerlIO-subfile-0.08-my/subfile.xs Apply this patch to address binmode(FH) problem. --- PerlIO-subfile-0.08/subfile.xs 2004-05-14 00:39:29.000000000 +0400 +++ PerlIO-subfile-0.08-my/subfile.xs 2009-04-01 00:17:26.000000000 +0400 @@ -296,7 +296,7 @@ sizeof(PerlIO_funcs), "subfile", sizeof(PerlIOSubfile), - PERLIO_K_BUFFERED, + PERLIO_K_BUFFERED|PERLIO_K_RAW, PerlIOSubfile_pushed, PerlIOBase_noop_ok, PerlIOBuf_open, Common subdirectories: PerlIO-subfile-0.08/t and PerlIO-subfile-0.08-my/t

Wed Jun 21 15:56:24 2017 nwc10+please+use+perlbug+for+perl+queries [...] colon.colondot.net - Correspondence added

On Fri Mar 12 12:02:49 2010, SAFRONOV wrote: Show quoted text

> Only in PerlIO-subfile-0.08-my: Makefile.old > diff -u PerlIO-subfile-0.08/subfile.xs PerlIO-subfile-0.08-my/subfile.xs > Apply this patch to address binmode(FH) problem. > > --- PerlIO-subfile-0.08/subfile.xs 2004-05-14 00:39:29.000000000 +0400 > +++ PerlIO-subfile-0.08-my/subfile.xs 2009-04-01 00:17:26.000000000 +0400

Sorry for the extremely long delay in replying - this ticket got overlooked. What would a regression test for this bug look like? Something like a test file that contains a "\r\n" sequence on disk, where the test code reads this, *and* correctly does binmode on the handle it has with the subfile layer, but the test fails (without this patch) because the data read in is "\n"? Or make sure the test file has something after the "\r\n" on disk, and what should be read in ends with the "\r\n", but without the patch what one gets ends with "\n" and a byte of trailing "garbage"? Nicholas Clark

Wed Jun 21 15:56:24 2017 The RT System itself - Status changed from 'new' to 'open'

Thu Jun 22 04:02:53 2017 vlad [...] yandex.ru - Correspondence added

Subject:	Re: [rt.cpan.org #55502] binmode() fix
Date:	Thu, 22 Jun 2017 11:00:58 +0300
To:	"bug-PerlIO-subfile [...] rt.cpan.org" <bug-perlio-subfile [...] rt.cpan.org>, "SAFRONOV [...] cpan.org" <safronov [...] cpan.org>
From:	Vlad. <vlad [...] yandex.ru>

Ouch.. I need to recall the where exactly I hit that bug.

--
V.

21.06.2017, 22:56, "Nicholas Clark via RT" <bug-perlio-subfile@rt.cpan.org>:

Show quoted text

<URL: https://rt.cpan.org/Ticket/Display.html?id=55502 >

On Fri Mar 12 12:02:49 2010, SAFRONOV wrote:

Only in PerlIO-subfile-0.08-my: Makefile.old
diff -u PerlIO-subfile-0.08/subfile.xs PerlIO-subfile-0.08-my/subfile.xs
Apply this patch to address binmode(FH) problem.

--- PerlIO-subfile-0.08/subfile.xs 2004-05-14 00:39:29.000000000 +0400
+++ PerlIO-subfile-0.08-my/subfile.xs 2009-04-01 00:17:26.000000000 +0400

Sorry for the extremely long delay in replying - this ticket got overlooked.

What would a regression test for this bug look like? Something like a test file that contains a "\r\n" sequence on disk, where the test code reads this, *and* correctly does binmode on the handle it has with the subfile layer, but the test fails (without this patch) because the data read in is "\n"?

Or make sure the test file has something after the "\r\n" on disk, and what should be read in ends with the "\r\n", but without the patch what one gets ends with "\n" and a byte of trailing "garbage"?

Nicholas Clark

Thu Jun 22 05:52:27 2017 vlad [...] yandex.ru - Correspondence added

Subject:	Re: [rt.cpan.org #55502] binmode() fix
Date:	Thu, 22 Jun 2017 12:52:06 +0300
To:	"bug-PerlIO-subfile [...] rt.cpan.org" <bug-perlio-subfile [...] rt.cpan.org>, "SAFRONOV [...] cpan.org" <safronov [...] cpan.org>
From:	Vlad. <vlad [...] yandex.ru>

I think this one is correct:

"Something like a test file that contains a "\r\n" sequence on disk, where the test code reads this, *and* correctly does binmode on the handle it has with the subfile layer, but the test fails (without this patch) because the data read in is "\n"?"

--
V.

22.06.2017, 11:03, "vlad@yandex.ru via RT" <bug-perlio-subfile@rt.cpan.org>:

Show quoted text

<URL: https://rt.cpan.org/Ticket/Display.html?id=55502 >

Ouch.. I need to recall the where exactly I hit that bug. --
V. 21.06.2017, 22:56, "Nicholas Clark via RT" <bug-perlio-subfile@rt.cpan.org>:

  <URL: https://rt.cpan.org/Ticket/Display.html?id=55502 >

  On Fri Mar 12 12:02:49 2010, SAFRONOV wrote:

    Only in PerlIO-subfile-0.08-my: Makefile.old
    diff -u PerlIO-subfile-0.08/subfile.xs
    PerlIO-subfile-0.08-my/subfile.xs
    Apply this patch to address binmode(FH) problem.

    --- PerlIO-subfile-0.08/subfile.xs 2004-05-14 00:39:29.000000000 +0400
    +++ PerlIO-subfile-0.08-my/subfile.xs 2009-04-01 00:17:26.000000000
    +0400

  Sorry for the extremely long delay in replying - this ticket got
  overlooked.

  What would a regression test for this bug look like? Something like a test
  file that contains a "\r\n" sequence on disk, where the test code reads
  this, *and* correctly does binmode on the handle it has with the subfile
  layer, but the test fails (without this patch) because the data read in is
  "\n"?

  Or make sure the test file has something after the "\r\n" on disk, and what
  should be read in ends with the "\r\n", but without the patch what one gets
  ends with "\n" and a byte of trailing "garbage"?

  Nicholas Clark