Fwd: RC 1.34.1 bjam build error
Hi Could somebody confirm this/provide a fix? Thanks Thomas Thomas Witt witt@acm.org Begin forwarded message:
From: "Jon Olesberg" <jonathon-olesberg@uiowa.edu> Date: July 12, 2007 6:38:32 AM PDT To: <witt@acm.org> Subject: FW: RC 1.34.1 bjam build error
Thomas,
The issue I reported to you (attached below) persists in RC2. Also, all of the references to the Boost version in the Getting Started docs refer to 1_34_0.
Jon Olesberg
-----Original Message----- From: Jon Olesberg [mailto:jonathon-olesberg@uiowa.edu] Sent: Wednesday, June 27, 2007 5:36 PM To: 'witt@acm.org' Subject: RC 1.34.1 bjam build error
Thomas,
In trying out the release candidate, I noticed a problem with the batch file that builds the bjam executable on windows. If I execute build.bat from the tools\jam\src directory, I get an error:
--------------------------------------------------------------------- T:\boost-1_34_1\boost\tools\jam\src>.\build.bat Setting environment for using Microsoft Visual Studio 2005 x86 tools. \Microsoft was unexpected at this time. ---------------------------------------------------------------------
However, if I specify msvc, things work fine:
--------------------------------------------------------------------- T:\boost-1_34_1\boost\tools\jam\src>.\build.bat msvc ### ### Using 'msvc' toolset. ###
T:\boost-1_34_1\boost\tools\jam\src>rd /S /Q bootstrap
T:\boost-1_34_1\boost\tools\jam\src>md bootstrap
... and so on... ---------------------------------------------------------------------
This problem is fixed in CVS head.
Jon
-- Jonathon Olesberg Optical Science and Technology Center University of Iowa http://www.ostc.uiowa.edu/~olesberg
Thomas Witt wrote:
Hi
Could somebody confirm this/provide a fix?
Well there are two fixes: a) Merge the *experimental* bjam from HEAD to RC. b) Use the precompiled bjam, but then we already strongly suggest this. -- -- Grafik - Don't Assume Anything -- Redshift Software, Inc. - http://redshift-software.com -- rrivera/acm.org - grafik/redshift-software.com -- 102708583/icq - grafikrobot/aim - grafikrobot/yahoo
Rene Rivera wrote:
Thomas Witt wrote:
Hi
Could somebody confirm this/provide a fix?
Well there are two fixes:
a) Merge the *experimental* bjam from HEAD to RC. b) Use the precompiled bjam, but then we already strongly suggest this.
I missed an option: c) Whomever can replicate the problem with the bjam in 1.34. Check if the problem exists in the 3.1.14 release of bjam. If it's not a problem there we can put that version in the Boost release. -- -- Grafik - Don't Assume Anything -- Redshift Software, Inc. - http://redshift-software.com -- rrivera/acm.org - grafik/redshift-software.com -- 102708583/icq - grafikrobot/aim - grafikrobot/yahoo
The problem exists in 3.1.14 also: --------------------------------------------------------------------- T:\boost_1_34_1\tools\jam\src>build.bat Setting environment for using Microsoft Visual Studio 2005 x86 tools. \Microsoft was unexpected at this time. T:\boost_1_34_1\tools\jam\src>build.bat msvc ### ### Using 'msvc' toolset. ### T:\boost_1_34_1\tools\jam\src>rd /S /Q bootstrap ... builds successfully ... --------------------------------------------------------------------- I point this out as an installation/build issue, albeit a minor one. It should not delay 1.34.1. Jon Olesberg Rene Rivera wrote:
Thomas Witt wrote:
Hi
Could somebody confirm this/provide a fix?
Well there are two fixes:
a) Merge the *experimental* bjam from HEAD to RC. b) Use the precompiled bjam, but then we already strongly suggest this.
I missed an option: c) Whomever can replicate the problem with the bjam in 1.34. Check if the problem exists in the 3.1.14 release of bjam. If it's not a problem there we can put that version in the Boost release.
From: Thomas Witt
Hi
Could somebody confirm this/provide a fix?
How about "deny"? I've just downloaded rc2, gone to the relevent directory, and here is my output: D:\boost_1_34_1\tools\jam\src>.\build.bat Setting environment for using Microsoft Visual Studio 2005 x86 tools. ### ### Using 'vc8' toolset. ### <BIG snip> Generating Code... ...updated 2 targets... D:\boost_1_34_1\tools\jam\src> Works for me. I'm using Visual Studio 2005 WITHOUT service pack 1. I have VS installed to C:\Program Files\. I'm running Windows XP SP2. My environment contains: ALLUSERSPROFILE=C:\Documents and Settings\All Users APPDATA=C:\Documents and Settings\mbonner\Application Data CLIENTNAME=Console CommonProgramFiles=C:\Program Files\Common Files COMPUTERNAME=<censored> ComSpec=C:\WINDOWS\system32\cmd.exe FP_NO_HOST_CHECK=NO HOMEDRIVE=C: HOMEPATH=\Documents and Settings\mbonner HTTP_proxy=<censored> LOGONSERVER=<censored> NUMBER_OF_PROCESSORS=2 OS=Windows_NT Path=C:\Program Files\Tcl\bin;C:\Perl\bin\;C:\WINDOWS\system32;C:\WINDOWS;C:\WINDOWS\Sys tem32\Wbem;C:\Program Files\UltraEdit;C:\Program Files\Perforce;C:\PROGRA~1\ATT\Graphviz\bin; PATHEXT=.COM;.EXE;.BAT;.CMD;.VBS;.VBE;.JS;.JSE;.WSF;.WSH;.tcl PROCESSOR_ARCHITECTURE=x86 PROCESSOR_IDENTIFIER=x86 Family 15 Model 4 Stepping 9, GenuineIntel PROCESSOR_LEVEL=15 PROCESSOR_REVISION=0409 ProgramFiles=C:\Program Files PROMPT=$P$G SESSIONNAME=Console SystemDrive=C: SystemRoot=C:\WINDOWS TEMP=C:\DOCUME~1\mbonner\LOCALS~1\Temp TMP=C:\DOCUME~1\mbonner\LOCALS~1\Temp USERDNSDOMAIN=<censored> USERDOMAIN=<censored> USERNAME=mbonner USERPROFILE=C:\Documents and Settings\mbonner VS80COMNTOOLS=C:\Program Files\Microsoft Visual Studio 8\Common7\Tools\ windir=C:\WINDOWS __COMPAT_LAYER=EnableNXShowUI (I've removed some details of our internal network structure). Any more information that could help track this down?
Thanks
Thomas
Thomas Witt witt@acm.org
Begin forwarded message:
From: "Jon Olesberg" <jonathon-olesberg@uiowa.edu> Date: July 12, 2007 6:38:32 AM PDT To: <witt@acm.org> Subject: FW: RC 1.34.1 bjam build error
Thomas,
The issue I reported to you (attached below) persists in RC2. Also, all of the references to the Boost version in the Getting Started docs refer to 1_34_0.
Jon Olesberg
-----Original Message----- From: Jon Olesberg [mailto:jonathon-olesberg@uiowa.edu] Sent: Wednesday, June 27, 2007 5:36 PM To: 'witt@acm.org' Subject: RC 1.34.1 bjam build error
Thomas,
In trying out the release candidate, I noticed a problem with the batch file that builds the bjam executable on windows. If I execute build.bat from the tools\jam\src directory, I get an error:
--------------------------------------------------------------------- T:\boost-1_34_1\boost\tools\jam\src>.\build.bat Setting environment for using Microsoft Visual Studio 2005 x86 tools. \Microsoft was unexpected at this time. ---------------------------------------------------------------------
However, if I specify msvc, things work fine:
--------------------------------------------------------------------- T:\boost-1_34_1\boost\tools\jam\src>.\build.bat msvc ### ### Using 'msvc' toolset. ###
T:\boost-1_34_1\boost\tools\jam\src>rd /S /Q bootstrap
T:\boost-1_34_1\boost\tools\jam\src>md bootstrap
... and so on... ---------------------------------------------------------------------
This problem is fixed in CVS head.
Jon
-- Jonathon Olesberg Optical Science and Technology Center University of Iowa http://www.ostc.uiowa.edu/~olesberg
_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
SPAM and Anti-Virus Checked by the FROG Group. http://www.frog.co.uk
server MX2 241
-- Martin Bonner Project Leader PI SHURLOK LTD Telephone: +44 1223 441434 / 203894 (direct) Fax: +44 1223 203999 Email: martin.bonner@pi-shurlok.com www.pi-shurlok.com disclaimer
on Thu Jul 12 2007, "Martin Bonner" <Martin.Bonner-AT-pi-shurlok.com> wrote:
Hi
Could somebody confirm this/provide a fix?
First of all, just FYI, building bjam is Rene's bailiwick, not mine. Secondly, this discussion ought to be attached to a Trac ticket.
(I've removed some details of our internal network structure). Any more information that could help track this down?
We could get the user's environment dump? -- Dave Abrahams Boost Consulting http://www.boost-consulting.com The Astoria Seminar ==> http://www.astoriaseminar.com
David Abrahams wrote:
on Thu Jul 12 2007, "Martin Bonner" <Martin.Bonner-AT-pi-shurlok.com> wrote:
Hi
Could somebody confirm this/provide a fix?
First of all, just FYI, building bjam is Rene's bailiwick, not mine.
I was aware of that. This is what I thought is your bailiwick (BTW what is a bailiwick?):
Also, all of the references to the Boost version in the Getting Started docs refer to 1_34_0.
Thomas -- Thomas Witt witt@acm.org
on Thu Jul 12 2007, Thomas Witt <witt-AT-acm.org> wrote:
David Abrahams wrote:
on Thu Jul 12 2007, "Martin Bonner" <Martin.Bonner-AT-pi-shurlok.com> wrote:
Hi
Could somebody confirm this/provide a fix?
First of all, just FYI, building bjam is Rene's bailiwick, not mine.
I was aware of that.
This is what I thought is your bailiwick (BTW what is a bailiwick?):
Also, all of the references to the Boost version in the Getting Started docs refer to 1_34_0.
Oh! Well, I can fix most of that, but part of it has to be done by -- or in coordination with -- the release manager. The last line of http://boost.cvs.sourceforge.net/boost/boost/more/getting_started/detail/release-variables.rst?view=markup&pathrev=RC_1_34_0 needs to be updated and the docs need to be regenerated (install docutils, go to more/getting_started, remove *.html in case docutils.jam isn't tracing dependencies quite right, invoke bjam, and cvs commit) before the release. -- Dave Abrahams Boost Consulting http://www.boost-consulting.com The Astoria Seminar ==> http://www.astoriaseminar.com
Dave, David Abrahams wrote:
on Thu Jul 12 2007, Thomas Witt <witt-AT-acm.org> wrote:
Oh! Well, I can fix most of that, but part of it has to be done by -- or in coordination with -- the release manager.
The last line of http://boost.cvs.sourceforge.net/boost/boost/more/getting_started/detail/release-variables.rst?view=markup&pathrev=RC_1_34_0 needs to be updated and the docs need to be regenerated (install docutils, go to more/getting_started, remove *.html in case docutils.jam isn't tracing dependencies quite right, invoke bjam, and cvs commit) before the release.
I would be thankful if you could regenerate the docs. I don't have a working docutils install on my machine. Thomas -- Thomas Witt witt@acm.org
on Thu Jul 12 2007, Thomas Witt <witt-AT-acm.org> wrote:
Dave,
David Abrahams wrote:
on Thu Jul 12 2007, Thomas Witt <witt-AT-acm.org> wrote:
Oh! Well, I can fix most of that, but part of it has to be done by -- or in coordination with -- the release manager.
The last line of http://boost.cvs.sourceforge.net/boost/boost/more/getting_started/detail/release-variables.rst?view=markup&pathrev=RC_1_34_0 needs to be updated and the docs need to be regenerated (install docutils, go to more/getting_started, remove *.html in case docutils.jam isn't tracing dependencies quite right, invoke bjam, and cvs commit) before the release.
I would be thankful if you could regenerate the docs. I don't have a working docutils install on my machine.
I'd be happy to. The point is that the release package needs to exist, or at least we need to know what its URL will be, before that can be done. -- Dave Abrahams Boost Consulting http://www.boost-consulting.com The Astoria Seminar ==> http://www.astoriaseminar.com
David Abrahams wrote:
on Thu Jul 12 2007, Thomas Witt <witt-AT-acm.org> wrote:
I would be thankful if you could regenerate the docs. I don't have a working docutils install on my machine.
I'd be happy to. The point is that the release package needs to exist, or at least we need to know what its URL will be, before that can be done.
Sorry for being dense, but why wasn't that a problem for 1.34.0 and is a problem now? What exactly do you need? the naming convention will be the same as in 1.34 the packages will be hosted on SF and the available packages will be the same (.tar.gz .tar.bz .exe). I am surely missing something here but I can't figure it out. Thomas -- Thomas Witt witt@acm.org
on Fri Jul 13 2007, Thomas Witt <witt-AT-acm.org> wrote:
David Abrahams wrote:
on Thu Jul 12 2007, Thomas Witt <witt-AT-acm.org> wrote:
I would be thankful if you could regenerate the docs. I don't have a working docutils install on my machine.
I'd be happy to. The point is that the release package needs to exist, or at least we need to know what its URL will be, before that can be done.
Sorry for being dense, but why wasn't that a problem for 1.34.0
I don't know As you can see from the rev 1.1.2.1 checkin comment at http://boost.cvs.sourceforge.net/boost/boost/more/getting_started/detail/release-variables.rst?view=log&pathrev=RC_1_34_0, the link was updated for the release.
and is a problem now? What exactly do you need?
A link analogous to http://sourceforge.net/project/showfiles.php?group_id=7586&package_id=8041, but for 1.34.1
the naming convention will be the same as in 1.34 the packages will be hosted on SF and the available packages will be the same (.tar.gz .tar.bz .exe). I am surely missing something here but I can't figure it out.
Click the link above and you'll see. -- Dave Abrahams Boost Consulting http://www.boost-consulting.com The Astoria Seminar ==> http://www.astoriaseminar.com
David Abrahams wrote:
on Fri Jul 13 2007, Thomas Witt <witt-AT-acm.org> wrote:
and is a problem now? What exactly do you need?
A link analogous to http://sourceforge.net/project/showfiles.php?group_id=7586&package_id=8041, but for 1.34.1
That link is generic to the "boost" package. SO it will work for whatever the current release is. -- -- Grafik - Don't Assume Anything -- Redshift Software, Inc. - http://redshift-software.com -- rrivera/acm.org - grafik/redshift-software.com -- 102708583/icq - grafikrobot/aim - grafikrobot/yahoo
Dave, David Abrahams wrote:
on Fri Jul 13 2007, Thomas Witt <witt-AT-acm.org> wrote:
I don't know As you can see from the rev 1.1.2.1 checkin comment at http://boost.cvs.sourceforge.net/boost/boost/more/getting_started/detail/release-variables.rst?view=log&pathrev=RC_1_34_0, the link was updated for the release.
Yes. But IIUC the html pages still have to be rebuilt. As of now all I do with respect to documentation generation is run bjam in doc/. I was assuming (probably for no good reason) you generate the html from the r(e)st.
and is a problem now? What exactly do you need?
A link analogous to http://sourceforge.net/project/showfiles.php?group_id=7586&package_id=8041, but for 1.34.1
As Rene said the link is version agnostic. Thomas -- Thomas Witt witt@acm.org
on Fri Jul 13 2007, Thomas Witt <witt-AT-acm.org> wrote:
Dave,
David Abrahams wrote:
on Fri Jul 13 2007, Thomas Witt <witt-AT-acm.org> wrote:
I don't know As you can see from the rev 1.1.2.1 checkin comment at http://boost.cvs.sourceforge.net/boost/boost/more/getting_started/detail/release-variables.rst?view=log&pathrev=RC_1_34_0, the link was updated for the release.
Yes. But IIUC the html pages still have to be rebuilt.
Yes, but I've done that already.
As of now all I do with respect to documentation generation is run bjam in doc/. I was assuming (probably for no good reason) you generate the html from the r(e)st.
Of course I do. I sure don't convert the ReST to HTML by hand! I do it by running bjam in more/getting_started/
and is a problem now? What exactly do you need?
A link analogous to http://sourceforge.net/project/showfiles.php?group_id=7586&package_id=8041, but for 1.34.1
As Rene said the link is version agnostic.
Then we're all set. -- Dave Abrahams Boost Consulting http://www.boost-consulting.com The Astoria Seminar ==> http://www.astoriaseminar.com
Thomas Witt wrote:
David Abrahams wrote:
on Thu Jul 12 2007, "Martin Bonner" <Martin.Bonner-AT-pi-shurlok.com> wrote:
Hi
Could somebody confirm this/provide a fix?
First of all, just FYI, building bjam is Rene's bailiwick, not mine.
I was aware of that.
This is what I thought is your bailiwick (BTW what is a bailiwick?):
Bailiwick is generally used in a metaphorical <http://en.wikipedia.org/wiki/Metaphor> sense, to indicate a sphere of authority, experience, activity, study, or interest.
Also, all of the references to the Boost version in the Getting Started docs refer to 1_34_0.
Thomas
Regards Richard Day
participants (6)
-
David Abrahams -
Jon Olesberg -
Martin Bonner -
Rene Rivera -
Richard Day -
Thomas Witt