- Boost - lists.boost.org

Re: [boost] Boost Units library preview
by Eric Lemings 22 Aug '06

22 Aug '06

> -----Original Message----- > From: boost-bounces(a)lists.boost.org > [mailto:boost-bounces@lists.boost.org] On Behalf Of Andy Little > Sent: Sunday, August 20, 2006 9:34 AM > To: boost(a)lists.boost.org > Subject: Re: [boost] Boost Units library preview > > ... > > Having had correspondence from various people, I am now > fairly sure that the Americans have a different concept of a > unit than (at least) Europeans. Well I don't presume to speak for all Americans but I define a "unit" as a standard of measure. Plain and simple. How do Europeans define it? Eric.

3 2

[Boost-bugs] [ boost-Support Requests-1544410 ] Regex: multi-processor scalability, user-defined allocators
by SourceForge.net 22 Aug '06

22 Aug '06

Support Requests item #1544410, was opened at 2006-08-22 11:19 Message generated for change (Tracker Item Submitted) made by Item Submitter You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=207586&aid=1544410&group_… Please note that this message will contain a full copy of the comment thread, including the initial issue submission, for this request, not just the latest update. Category: regex Group: None Status: Open Resolution: None Priority: 5 Submitted By: Roy (r_emek) Assigned to: John Maddock (johnmaddock) Summary: Regex: multi-processor scalability, user-defined allocators Initial Comment: I'm using boost-regex library, and I'm running into multi-threading scalability issues. When tested on a machine with several CPUs, my performance tests show improvement of about 60-70% when moving from a single thread to 2 threads. Question 1: Is anyone aware of this problem? Is there a known solution? >From some testing we've done, it seems that the cause of the scalability issue is problematic implementation of std:allocator. This is a known issue on some operating systems (e.g., Solaris 8). Some of the boost::regex classes (e.g., match_results) accept a user-defined allocator. However, as far as I can understand there's no way to completely override the use of std::allocator. Question 2: is there a way to completely prevent boost / regex from using std::allocator? If no, can such a capability be added? (One solution to this problem would be global overriding the 'new' and 'delete' operators - but for various reasons this cannot be done in the application I'm developing). ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=207586&aid=1544410&group_… ------------------------------------------------------------------------- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 _______________________________________________ Boost-bugs mailing list Boost-bugs(a)lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/boost-bugs

1 0

Visual Studio 2003 SP1 is now available...
by Kevin Wheatley 22 Aug '06

22 Aug '06

Available from http://msdn.microsoft.com/downloads/ Direct Link http://www.microsoft.com/downloads/details.aspx?FamilyID=69D2219F-CE82-46A5… 'release notes' http://www.microsoft.com/downloads/details.aspx?FamilyID=DC455C5D-2C99-43CC… There isn't much in terms of a list of bugs fixed at the moment, but I'm sure they will correct that. Lots more fun for everyone Kevin -- | Kevin Wheatley, Cinesite (Europe) Ltd | Nobody thinks this | | Senior Technology | My employer for certain | | And Network Systems Architect | Not even myself |

5 6

Re: [boost] email client weirdness (was: Boost Units library preview)
by Eric Lemings 22 Aug '06

22 Aug '06

> -----Original Message----- > From: boost-bounces(a)lists.boost.org > [mailto:boost-bounces@lists.boost.org] On Behalf Of Janek Kozicki > Sent: Monday, August 21, 2006 2:24 PM > To: boost(a)lists.boost.org > Subject: Re: [boost] email client weirdness (was: Boost Units > library preview) > > Eric Lemings said: (by the date of Mon, 21 Aug 2006 > 13:39:49 -0600) > > > Doh! This should be addition, not multiplication. > > > > m + f; > > > > Eric. > > > Eric, please check your email client configuration with > respect to 'replying to posts' on this mailing lists. All > posts from you do not stack correctly with other posts to > form a thread. They all seem to be just a reply to the > toplevel post. That makes reading them a bit confusing. Unfortunately, I have no idea what the problem is or how to fix it. Would appreciate any help. Eric.

2 2

[SoC] Tree
by Bernhard Reiter 21 Aug '06

21 Aug '06

Hello to all, as time's almost over for this year's Summer of Code, I'll join my fellow SoCers in presenting the resulting projects. I've been working on what was entitled a "Generic tree container" in the original Boost SoC suggestions wiki; but having discovered that most people were looking much more for some unifying concepts dealing with the similarities found in the vast diversity of existing tree designs, I sought to expand the scope of my work towards e.g. all kinds of search trees that my original proposal hadn't covered. I believe that I've distilled a couple of quite powerful concepts from the resources I reviewed; but as you might guess, this took more time than anticipated so that I can only present a very incomplete implementation of what I came up with. The code can be obtained via svn from https://www.boost-consulting.com:8443/svn/main/boost/soc/2006/tree/trunk or viewed in a browser at https://boost-consulting.com:8443/trac/soc/browser/boost/soc/2006/tree/trunk At the moment, there is only an unbalanced binary tree (including a "cursor" class), a couple of traversal and search algorithms, plus a "searcher" adaptor. For more information about the individual components, please refer to the documentation at http://www.unet.univie.ac.at/~a0057324/php/soc2006/doc/html/ Again, please keep in mind that neither code nor documentation are complete enough to be regarded a proper tree framework library -- which I intend them to be eventually, however. For now, I'll again join other SoCers, this time trying to catch some sun; I'll be leaving tomorrow (22nd) morning (CEST) for about a week of vacations. After that, I'll have another month until my university classes start again, which coincides with the deadline for TR2 submission. Now I know the following might seem a little ambitious, but: given the amount of literature and newsgroup/mailing list posts dealing with trees and C++, there seems to have been interest in an STL tree component about since the standard came out. So I'd like to try to refine especially the conceptual documentation parts first when I come back, implement what's still missing, and, with the help of some more experienced souls than I am, submit a proposal for TR2... (Or am I just dreaming?) Any comments very welcome. Grateful for an awesome experience, Bernhard

3 2

Re: [boost] Boost Units library preview
by Eric Lemings 21 Aug '06

21 Aug '06

> -----Original Message----- > From: boost-bounces(a)lists.boost.org > [mailto:boost-bounces@lists.boost.org] On Behalf Of Dave Steffen > Sent: Monday, August 21, 2006 4:49 PM > To: boost(a)lists.boost.org > Subject: Re: [boost] Boost Units library preview > ... > > Comment from the back row: Some people will want automatic > conversions. Some people will want a compilation failure. > Both are perfectly reasonable behaviors, and which one is > appropriate is project specific. > > If (either) units library fails to provide both > capabilities, it will be unusable for some category of projects. > > I don't know how to control this, nor what the interface > should look like, but I am certain that a good units library > will be able to go both ways. Ah! Finally, some good design criticisms. I will certainly take that into account in future implementations. Eric.

1 0

Re: [boost] Boost Units library preview
by Eric Lemings 21 Aug '06

21 Aug '06

> -----Original Message----- > From: boost-bounces(a)lists.boost.org > [mailto:boost-bounces@lists.boost.org] On Behalf Of Deane Yang > Sent: Monday, August 21, 2006 4:28 PM > To: boost(a)lists.boost.org > Subject: Re: [boost] Boost Units library preview > ... > > My main objection to this is that I am pretty sure that a > quantity/units library that does not do implicit unit > conversion is a *lot* easier to implement than one that does. > And it's hard for me to believe that a library that does > automatic unit conversion doesn't have inside of it a simpler > non-automatic unit conversion library. So implementing > explicit unit conversion on top of an implicit conversion > library seems, well, backwards. Not really. The implementation that I used is that, internally, all arithmetic values are stored in a standard unit (typically the SI unit). Only when an arithmetic value is assigned to or retrieved from the object is a conversion operation actually required. All other arithmetic operations can therefore be oblivious to conversion operations. So doing the conversions up front actually simplifies the implementation. Eric.

1 0

Re: [boost] Boost Units library preview
by Eric Lemings 21 Aug '06

21 Aug '06

> -----Original Message----- > From: Eric Lemings > Sent: Monday, August 21, 2006 10:44 AM > To: 'boost(a)lists.boost.org' > Subject: RE: [boost] Boost Units library preview > > > Darn email client. Concatenated my line breaks. > dimension - an "abstract" type > quantity - an "abstract" object whose "type" is a dimension > unit - a type "derived" from a particular dimension > measure - an arithmetic object whose "type" is a unit

2 1

Re: [boost] Boost Units library preview
by Eric Lemings 21 Aug '06

21 Aug '06

> -----Original Message----- > From: boost-bounces(a)lists.boost.org > [mailto:boost-bounces@lists.boost.org] On Behalf Of Andy Little > Sent: Sunday, August 20, 2006 9:34 AM > To: boost(a)lists.boost.org > Subject: Re: [boost] Boost Units library preview > > ... > Having had correspondence from various people, I am now > fairly sure that the Americans have a different concept of a > unit than (at least) Europeans. I define a "unit" as a standard of measure. Nothing more, nothing less. How else can one define a "unit"? Eric.

1 0

Re: [boost] Boost Units library preview
by Eric Lemings 21 Aug '06

21 Aug '06

> -----Original Message----- > From: boost-bounces(a)lists.boost.org > [mailto:boost-bounces@lists.boost.org] On Behalf Of Matthias Troyer > Sent: Sunday, August 20, 2006 6:29 AM > To: boost(a)lists.boost.org > Subject: Re: [boost] Boost Units library preview > > ... > Wait, if I multiply quantities (which have units) then also the > result will have a unit. E.g. if I divide 10 meters by 2 > seconds then > I get 5 m/s . I wouldn't call that anonymous since I know the > unit: m/s This brings up a good question. Say you have two objects: meters m = 1; feet f = 2; What is the type (or unit) of the following expression? m * f; Is it feet? meters? Some other type? If there is a rule (in dimensional analysis), please let me know. Meantime, I define the resulting type as not as a unit at all but as the dimension of the unit: void f (length l) { ... } f (m * f); This is not to say that the user can not specify the desired units: rods r = m * f; This is possible since a quantity of a compatible dimension can be implicitly converted back into a measure of the appropriate unit (in theory at least). Eric.

1 0