[1.36.0] Libraries with unmerged changed - more accurate results.
Below is a more accurate list of headers and libraries whose header files appear to differ between the trunk and branches/release. The most likely causes are: * Changes that should be merged into branches/release for 1.36.0. If that is the case, developers need to merge these changes right away! * Work in process that should *not* be merged into branches/release for 1.36.0. Please let me know that that is the case so I can stop worrying:-) * Although it seems to be working much better, there still may be issues with the script used to generate the report. Please let me know so the script can be improved. Thanks, --Beman Headers: any.hpp concept_check.hpp enable_shared_from_this.hpp intrusive_ptr.hpp lexical_cast.hpp make_shared.hpp memory_order.hpp operators.hpp pointer_cast.hpp pointer_to_other.hpp rational.hpp scoped_array.hpp scoped_ptr.hpp shared_ptr.hpp smart_cast.hpp static_assert.hpp static_warning.hpp swap.hpp tokenizer.hpp weak_ptr.hpp Libraries: asio bimap circular_buffer config date_time detail filesystem function function_types graph interprocess intrusive iostreams iterator math mpl numeric optional parameter pool program_options property_tree python regex serialization signals spirit statechart system test tr1 type_traits units utility
on Wed Jul 09 2008, Beman Dawes <bdawes-AT-acm.org> wrote:
Below is a more accurate list of headers and libraries whose header files appear to differ between the trunk and branches/release.
The most likely causes are:
* Changes that should be merged into branches/release for 1.36.0. If that is the case, developers need to merge these changes right away!
* Work in process that should *not* be merged into branches/release for 1.36.0. Please let me know that that is the case so I can stop worrying:-)
* Although it seems to be working much better, there still may be issues with the script used to generate the report. Please let me know so the script can be improved.
What does it mean if my library appears in the list but none of its headers show up?
Headers:
any.hpp concept_check.hpp enable_shared_from_this.hpp intrusive_ptr.hpp lexical_cast.hpp make_shared.hpp memory_order.hpp operators.hpp pointer_cast.hpp pointer_to_other.hpp rational.hpp scoped_array.hpp scoped_ptr.hpp shared_ptr.hpp smart_cast.hpp static_assert.hpp static_warning.hpp swap.hpp
The above is a new feature of the utility lib, not for release in 1.36
tokenizer.hpp weak_ptr.hpp
Libraries:
iterator parameter python utility
What, specifically, are we concerned about with these libs? -- Dave Abrahams BoostPro Computing http://www.boostpro.com
David Abrahams wrote:
on Wed Jul 09 2008, Beman Dawes <bdawes-AT-acm.org> wrote:
Below is a more accurate list of headers and libraries whose header files appear to differ between the trunk and branches/release. ... What does it mean if my library appears in the list but none of its headers show up?
To keep the size of down, only the library names are given. The header names are only given for the headers in boost-root/boost. See below for how to get more detail.
...
swap.hpp
The above is a new feature of the utility lib, not for release in 1.36
OK.
...
Libraries:
iterator parameter python utility
What, specifically, are we concerned about with these libs?
See trunk/tools/release/diff_trunk_release.bat for a little Windows command line tool: @echo off if not %1$==$ goto usage_ok echo Usage: diff_trunk_release library-name goto done :usage_ok svn diff http://svn.boost.org/svn/boost/branches/release/boost/%1 http://svn.boost.org/svn/boost/trunk/boost/%1 svn diff http://svn.boost.org/svn/boost/branches/release/libs/%1 http://svn.boost.org/svn/boost/trunk/libs/%1 :done Sorry, but I haven't had time to create a bash version. If you run it like this: D:\boost\trunk\tools\release>diff_trunk_release parameter | grep "Index: " | sed 's/Index: //' Output is: macros.hpp aux_/cast.hpp aux_/set.hpp aux_/python/invoker.hpp aux_/tagged_argument.hpp aux_/arg_list.hpp aux_/default.hpp preprocessor.hpp doc/html/reference.html doc/html/python.html doc/html/index.html doc/reference.rst doc/index.rst If you run "diff_trunk_release parameter", you will get an actual diff of the differences. I've run it for each of the four libraries you mentioned, and all appear to have unmerged changes. Of course I have no way of knowing if these should go in 1.36.0 or be held for 1.37.0. --Beman
AMDG Beman Dawes wrote:
:usage_ok svn diff http://svn.boost.org/svn/boost/branches/release/boost/%1 http://svn.boost.org/svn/boost/trunk/boost/%1 svn diff http://svn.boost.org/svn/boost/branches/release/libs/%1 http://svn.boost.org/svn/boost/trunk/libs/%1
:done
Would it be faster to use --summarize? In Christ, Steven Watanabe
Steven Watanabe wrote:
AMDG
Beman Dawes wrote:
:usage_ok svn diff http://svn.boost.org/svn/boost/branches/release/boost/%1 http://svn.boost.org/svn/boost/trunk/boost/%1 svn diff http://svn.boost.org/svn/boost/branches/release/libs/%1 http://svn.boost.org/svn/boost/trunk/libs/%1
:done
Would it be faster to use --summarize?
Yes. Thanks for the suggestion! --Beman
on Mon Jul 07 2008, "Jamie Ricks (RTI)" <Jamie.Ricks-AT-tradingtechnologies.com> wrote:
Hi David,
We were thinking about doing your course with 20 people. At this point we don't think we have more than 20 people who could keep up with your class.
Hi Jamie, For what it's worth, we can tailor the classes to any level, but if you have 20 people who are ready for the advanced stuff it's probably a good idea to train them separately anyway. We'll send you an invoice hopefully later today.
Let me know if you need more information!
Jamie Ricks Trading Technologies 222 S. Riverside Plaza Suite 1100 Chicago, IL 60606 Desk: 312-476-1109 Cell: 312-730-7373 Fax: 312-476-1001
-----Original Message----- From: David Abrahams [mailto:dave@boostpro.com] Sent: Wednesday, July 02, 2008 6:16 PM To: Jamie Ricks (RTI) Subject: Re: Request for quote
Jamie Ricks (RTI) wrote:
Hi Dave,
Thanks for the information. Are you available the first week of December? We would love to have you come out for three days sometime that week.
At this point I'm wanting to submit the purchase order for your trip out here. Is it possible to get a draft invoice with estimated costs for everything? After I get the approval from our accounting department, I'll get you in touch with a few of our developer to discuss how best to spend your time here.
Hi Jamie,
Are we doing a course for 30-40 people with two instructors, as suggested below? I only ask because the answer affects the amount on our invoice.
Thanks, Dave! Hope you're having a great day!
Jamie Ricks Trading Technologies 222 S. Riverside Plaza Suite 1100 Chicago, IL 60606 Desk: 312-476-1109 Cell: 312-730-7373 Fax: 312-476-1001
-----Original Message----- From: David Abrahams [mailto:dave@boostpro.com] Sent: Monday, June 30, 2008 9:39 AM To: Jamie Ricks (RTI) Cc: dave@boost-consulting.com Subject: Re: Request for quote
Jamie Ricks (RTI) wrote:
Hello Mr. Abrahams,
We are a software development company focused on electronic trading of futures contracts. We are interested in having you here for a seminar. We are interested in possibly 2-3 topics over a few days or a week. Do you have a standard lecture package that you offer?
Hi Jamie,
It was good talking with you on the phone today. Just reiterating here for the record, we have a fairly wide variety of courses, but we always customize them to the needs and interests of the students. Given the recent interests of your students, our one-week course on C++ Template Metaprogramming might be an appropriate choice. Also popular this year have been our course called "Higher-Level C++ with the Boost libraries," and our course on Generic Programming. I've attached descriptive outlines of all three.
Can you please reply and let me know what your availability and cost is? We were thinking this November would be a good time to have you here.
November is a good time for me; there are really no constraints that month, although the week of the 3rd and the week of the 17th are the best.
Our cost is $6000 per instructor-day plus expenses and a travel time allowance. Expenses include all costs associated with presenting the course, such as travel, lodging, meals, and printed materials such as handouts. The travel time allowance for Chicago is 1/2 day each way at our standard offsite rate of $200/hr, which comes to $1600 per instructor. Note that these courses can be spread across two visits necessary -- you'll pay the travel allowance twice but it might work better for your peoples' schedules. Given the class sizes you're talking about (30-40 students), I suggest two instructors. Twenty-five people is already at the upper limit for what one person can handle.
Regards,
-- Dave Abrahams BoostPro Computing http://www.boostpro.com
-- Dave Abrahams BoostPro Computing http://www.boostpro.com
on Thu Jul 10 2008, Beman Dawes <bdawes-AT-acm.org> wrote:
Libraries:
iterator parameter python utility
Done for parameter and python. I believe that utility is merely swap support which is a new feature and too late for 1.35.0 I'll check out iterator in a few hours. -- Dave Abrahams BoostPro Computing http://www.boostpro.com
on Thu Jul 10 2008, Beman Dawes <bdawes-AT-acm.org> wrote:
Libraries:
iterator
This one is now taken care of. Thanks for your patience. -- Dave Abrahams BoostPro Computing http://www.boostpro.com
On 10 Jul 2008, at 03:43, Beman Dawes wrote:
Below is a more accurate list of headers and libraries whose header files appear to differ between the trunk and branches/release.
serialization
The remaining diffs are in the Jamfiles and due to the fact that Boost.Config differs on the trunk and release branch Matthias
Beman Dawes wrote:
Below is a more accurate list of headers and libraries whose header files appear to differ between the trunk and branches/release.
OK two questions: 1) Is there an easy way to make a directory in the Release branch identical to the Trunk? 2) If not what was the revision ID of the last release? Thanks, John.
John Maddock wrote:
Beman Dawes wrote:
Below is a more accurate list of headers and libraries whose header files appear to differ between the trunk and branches/release.
OK two questions:
1) Is there an easy way to make a directory in the Release branch identical to the Trunk?
Yes... I mentioned that use case long ago in the last release cycle. Just svn merge from release/HEAD to trunk/HEAD into release/HEAD: svn merge https://svn.boost.org/svn/boost/branches/release https://svn.boost.org/svn/boost/trunk /some/path/to/boost/release (IIRC) Equivalent instructions for tsvn apply... See the window snapshot attached. Of course adjust to use some subpath of the boost tree ;-) -- -- Grafik - Don't Assume Anything -- Redshift Software, Inc. - http://redshift-software.com -- rrivera/acm.org (msn) - grafik/redshift-software.com -- 102708583/icq - grafikrobot/aim,yahoo,skype,efnet,gmail
Beman Dawes <bdawes <at> acm.org> writes:
intrusive_ptr.hpp
I did a full test run on the release branch last night, and got a huge number of failures due to a missing reset() function in intrusive_ptr (e.g. http://tinyurl.com/67wnex). I guess this needs to be merged to fix the problem? Thanks, Richard Webb
Richard Webb <richard.webb@boldonjames.com> writes:
Beman Dawes <bdawes <at> acm.org> writes:
intrusive_ptr.hpp
I did a full test run on the release branch last night, and got a huge number of failures due to a missing reset() function in intrusive_ptr (e.g. http://tinyurl.com/67wnex). I guess this needs to be merged to fix the problem?
Oops. I thought I'd changed the thread code to pass "0" explicitly when calling reset(). I guess I didn't check it in before I merged to the release branch. If intrusive_ptr is going to be merged, I'll leave the code as-is. Anthony -- Anthony Williams | Just Software Solutions Ltd Custom Software Development | http://www.justsoftwaresolutions.co.uk Registered in England, Company Number 5478976. Registered Office: 15 Carrallack Mews, St Just, Cornwall, TR19 7UL
Richard Webb:
Beman Dawes <bdawes <at> acm.org> writes:
intrusive_ptr.hpp
I did a full test run on the release branch last night, and got a huge number of failures due to a missing reset() function in intrusive_ptr (e.g. http://tinyurl.com/67wnex). I guess this needs to be merged to fix the problem?
Should be merged now.
Peter Dimov wrote:
Richard Webb:
Beman Dawes <bdawes <at> acm.org> writes:
intrusive_ptr.hpp
I did a full test run on the release branch last night, and got a huge number of failures due to a missing reset() function in intrusive_ptr (e.g. http://tinyurl.com/67wnex). I guess this needs to be merged to fix the problem?
Should be merged now.
Thanks, Peter! --Beman
shared_ptr.hpp
What a pain. This has diverged from the release version considerably and contains diffs of all kinds - some are to be merged, some are not to be merged, some can be merged but are somewhat risky, and others have no other purpose apart from making the job more difficult. Add to that that the committee changed the atomic interface to free functions just to make the current (ready and tested) implementation non-release-able. The two options I see is (1) ignore everything and release as-is, (2) let me try and merge as much as I can and see how it goes, then optionally fall back to (1) if things don't go well.
The recent sets of merges have fixed some of the VC9 test failures i was seeing on the release branch, but have caused some more :( 1) There are some Parameter related failures due to a missing #include <boost/type_traits/is_convertible.hpp> in boost/parameter/aux_/tagged_argument.hpp that has been fixed in Trunk, but not merged. (e.g. http://tinyurl.com/69amkf) 2) Theres a lot of failures relating to is_void in boost/function/function_template.hpp (e.g. http://tinyurl.com/57gb65) This looks to be down to a missing #include <boost/type_traits/is_void.hpp> in boost/function/detail/prologue.hpp Thanks, Richard Webb
config
Boost.Config changes should be merged, IMO. I haven't inspected the exact differences, but changes to the config machinery are (nearly) always made in response to a specific library problem. We may not see it in the test results since they are run on relatively conforming compilers (as far as I can see).
Peter Dimov wrote:
config
Boost.Config changes should be merged, IMO. I haven't inspected the exact differences, but changes to the config machinery are (nearly) always made in response to a specific library problem. We may not see it in the test results since they are run on relatively conforming compilers (as far as I can see).
Merged this morning, I'm slowly getting round to merging the rest of my stuff... John.
On Jul 15, 2008, at 2:25 PM, John Maddock wrote:
Peter Dimov wrote:
config
Boost.Config changes should be merged, IMO. I haven't inspected the exact differences, but changes to the config machinery are (nearly) always made in response to a specific library problem. We may not see it in the test results since they are run on relatively conforming compilers (as far as I can see).
Merged this morning, I'm slowly getting round to merging the rest of my stuff...
Did you update the serialization Jamfiles as well to use the new path or should we merge the changes to those from the trunk again? Matthias
Matthias Troyer wrote:
Did you update the serialization Jamfiles as well to use the new path or should we merge the changes to those from the trunk again?
You bet, status/Jamfile would be broken otherwise :-) John.
participants (9)
-
Anthony Williams -
Beman Dawes -
David Abrahams -
John Maddock -
Matthias Troyer -
Peter Dimov -
Rene Rivera -
Richard Webb -
Steven Watanabe