- Boost - lists.boost.org

Re: static sized strings (struct hack)
by Bill Seymour 12 Jun '04

12 Jun '04

Rob Stewart: > > From: John Nagle <nagle(a)animats.com> > > > From: "Reece Dunn" <msclrhd(a)hotmail.com> > > > > >> class fs_base > > >> { > > >> size_t len; > > >> szie_t capacity; // needed for buffer-safe operations > > >> CharT str[ 1 ]; > > >> // string manipulation functions > > >> > > >> fs_base( size_t c ): capacity( c ){} > > >> }; > > >> > > >> template< size_t n > > > >> class fixed_string: public fs_base > > >> { > > >> CharT data[ n ]; > > >> fixed_string(): fs_base( n ){} > > >> }; > > > > Are you assuming that "data[n]" physically follows > > "str[1]" in memory, and that "data[n]" can be addressed > > by subscripting "str[1]" out of range? ... > > Oh, right. It was C99 that blessed the struct hack. Oh well. > Well, yes; but the struct hack is not what the code above tries to do. What C99 says is that you can malloc sizeof(fs_base) plus some extra, then address str[2] to str[whatever up to malloc'd size]. Even in C99 (and even if C understood the C++-specific stuff in the above), there is no relationship between fs_base::str and fixed_string::data (except that they don't overlap). --Bill Seymour

2 1

More Math Functions
by Paul A Bristow 12 Jun '04

12 Jun '04

|-----Original Message----- |From: boost-bounces(a)lists.boost.org [mailto:boost-bounces@lists.boost.org] On Behalf |Of Tom Brinkman |Sent: 03 June 2004 00:54 |To: boost(a)lists.boost.org |Subject: [boost] Boost Mathematicians |What is the current status of all Boost mathmatical, |statistical and econometric libraries currently in |development? |The collection of boost math libraries that are |currently included in the library are world-class and |provide a wonderful foundation to build-upon. |However, many areas of advanced statistical and |mathmatical inference are not currently being \addressed by any of the boost libraries. I have been off-line for some time, but have now caught up with this discussion a bit. Strongly support your views and need these things yesterday ;-) Apart from messing about with Microsoft Windows problems, I have been preparing a formal proposal for C and C++ Standard Library TR-2 of a list of Math 'Special' Functions (additional to those from Walter Brown which have been accepted into TR-1). This list will include all the usual statistical suspects - including Student's t, Chi sqr, and F, so that at last it will be possible easily write C++ to provide the probability that means of two series of measurements differ. (The key function is the incomplete beta; it is also the most difficult to write well). Fortunately I have persuaded Stephen Moshier, author of the Cephes function library (in C) to make this available under Boost licence terms. So implementors will have no excuse to hide behind (the reason why these functions were omitted by Walter Brown). It should be possible to produce a working (and I believe rather good) implementation with little more than renaming the functions. This might be submitted for a Boost library. I hope to expose a revised draft to the rigours of Boost and C++.std newsgroup Real Soon Now. Paul Paul A. Bristow Prizet Farmhouse, Kendal UK LA8 8AB +44 1539 561830 07714 33 02 04 pbristow(a)hetp.u-net.com

1 0

Paul Mensonides:
by Vesa Karvonen 12 Jun '04

12 Jun '04

[boost] Re: Some remarks on PP docs Date: Fri, 11 Jun 2004 16:36:19 +0300 Mime-Version: 1.0 Content-Type: text/plain; format=flowed Paul Mensonides: >Chaos, as opposed to Order, intentionally encourages better >understanding of the actual language being used (which is not Chaos, but >the preprocessor) rather than trying to hide it as much as possible. I've discussed this with you earlier, but I'd still like to point out that the intention behind the design of Order has not been to "hide as much as possible" or to discourage "better understanding", but rather to *provide* a language that definitely *is* easy to understand and convenient to use. A person unfamiliar with Boost.PP should be able to learn to use something like Order and use it for productive, even non-trivial, work (which is usually C or C++ code generation) in significantly shorter period of time than it takes to learn how to do similar things with Boost.PP (or Chaos). I also noticed other, IMO, misleading comments on Order [vs Chaos], but I don't have the time to comment on those at the moment. -Vesa Karvonen _________________________________________________________________ The new MSN 8: advanced junk mail protection and 2 months FREE* http://join.msn.com/?page=features/junkmail

2 1

Re: Some remarks on PP docs
by Vesa Karvonen 12 Jun '04

12 Jun '04

David Abrahams: >I'm writing an appendix on the PP lib for the MPL book today. I think >that'll demonstrate that it's possible to provide a reasonable gentle >and high-level introduction without exposing too many ugly details. That would be great! >- Fundamental PP Abstractions > - Macros > - Non function > - Function macros > - Tokens > - Sequences of tokens not including un-parenthesized commas Data structures are an important topic and their representation isn't fundamentally PP *library* specific. You might want to present the information on the representation of data structures (and simple macros like a projection operator for tuples and some sequences operations) here when you discuss fundamental PP abstractions. I would also probably discuss how to implement primitives like IF and WHILE using preprocessor macros at this point. The purpose being to show that the preprocessor is capable of significant computation. I would also emphasize that the purpose of explaining the techniques is not that people would roll their own versions, but simply to deepen their understanding of the preprocessor. >Well, if you really have a conforming C99 preprocessor, you could use >Paul M's Chaos, which I think is both easier to use and *way* more >efficient than the current Boost PP. >From what I know, Chaos is essentially an extended Boost.PP (meaning that Chaos is essentially a combinator library), with a single recursion backend (enabled by stricter requirements on preproprecessor conformance), which makes it more orthogonal. However, Chaos still shares many problems with Boost.PP, the main issues being that code will be highly verbose and very intricate. The verbosity and intricacy probably do not matter that much when one only performs simple repetition using the canned library macros. -Vesa Karvonen _________________________________________________________________ Protect your PC - get McAfee.com VirusScan Online http://clinic.mcafee.com/clinic/ibuy/campaign.asp?cid=3963

3 5

Some remarks on PP docs
by David Abrahams 12 Jun '04

12 Jun '04

techniques.html: It is usually okay to use a standard macro name like BOOST_PP_DEF for this kind of code because the macro is both defined and undefined in the immediate site of its use. I was really confused by the use of the word "standard". This would be clearer: It is usually okay to reuse a single generic macro name like BOOST_PP_DEF throughout your code because the macro is both defined and undefined very close to its point of use. Although I don't know why you'd want to advocate that over using a descriptive name. ------ A clearer divider between examples would be a *huge* help visually. An <hr> would be enough, for example. Example - Use BOOST_PP_EMPTY as an unused parameter in local macro instantiations. #define BOOST_PP_DEF(cv) /* ... */ \ template<class base> \ cv() typename implement_subscript_using_begin_subscript<base>::value_type& \ implement_subscript_using_begin_subscript<base>::operator[](index_type i) cv() { \ return base::begin()[i]; \ } \ /**/ ?? Where is BOOST_PP_EMPTY used in this example? I think I know what you mean, but really, you need to show it! How: BOOST_PP_EMPTY() expands to nothing and can be used as an unused parameter. Note: BOOST_PP_EMPTY with the () never gets expanded. This sounds self-contradictory. Just above you said it expands (to nothing)! The () is necessary to invoke a function-like macro. No duh; I think stating the obvious that way only adds confusion. Caveat: You cannot safely use concatenation while using BOOST_PP_EMPTY(). That's way too vague and general. I certainly can use concatenation and BOOST_PP_EMPTY "together", for many reasonable definitions of that word! I think you need to be more specific, or strike it. ------ The BOOST_PP_ENUM_PARAMS example could benefit greatly from a side-by-side format in a table showing the expansion on the right. In fact, that would help all the examples, but in particular if you're just getting acquainted with the library at this point you may not have any idea what some of those macros do. You're advocating that people use them here, but unless the reader has some idea of what they generate, it's lost on her. Showing not-even-correct details of BOOST_PP_REPEAT's implementation doesn't help here. I'd strike it. Furthermore, BOOST_PP_ENUM_PARAMS and its variations use BOOST_PP_REPEAT. BOOST_PP_COMMA_IF(I) expands to a comma if I != 0. BOOST_PP_INC(I) essentially expands to "I+1," and BOOST_PP_DEC(I) essentially expands to "I-1.". I'd strike that too. You haven't said a word about BOOST_PP_INC/DEC yet, so the reader is left wondering why you're giving their implementation details. ----- Example - Use BOOST_PP_REPEAT and a token look-up function to eliminate categorical repetition. Is this really better than iterating over a SEQ of categories in any way other than efficiency? ----- Example - Use BOOST_PP_REPEAT to avoid O(n*n) repetition. It's unclear to me how that example avoids O(n*n) repetition. It certainly seems to have that complexity, and that appears to be intrinsic to the code it generates. Maybe you mean Example - Use BOOST_PP_REPEAT for O(n*n) repetition. Or something? ----- BOOST_PP_IF enables convenient generation of lists using BOOST_PP_REPEAT. This statement, without an example, is too vague to be helpful. ----- Implementation details of BOOST_PP_IF don't help: How: BOOST_PP_IF is defined for the entire repeat range (psuedo code): #define BOOST_PP_IF(c, THEN, ELSE) BOOST_PP_IF ## c(THEN, ELSE) #define BOOST_PP_IF0(THEN, ELSE) ELSE #define BOOST_PP_IF1(THEN, ELSE) THEN #define BOOST_PP_IF1(THEN, ELSE) THEN ^ especially since they appear to be wrong. I think the point is that BOOST_PP_IF works on first arguments other than 0 or 1. If so, you should just say that it works on numeric arguments up to N (whatever N is). --------- Example: Use arithmetic, logical, and comparison operations when necessary. With no commentary, and no generated code, this example is really useless. I mean, of course you'd use arithmetic, logical, and comparison operations "when neccessary". -- Dave Abrahams Boost Consulting http://www.boost-consulting.com

4 6

RE: [boost] Re: Documentation
by Reece Dunn 11 Jun '04

11 Jun '04

David Abrahams wrote: >"Reece Dunn" <msclrhd(a)hotmail.com> writes: > > David Abrahams wrote: > >>"Reece Dunn" <msclrhd(a)hotmail.com> writes: > >> > David B. Held wrote: > > It should be fairly easy to adapt the ReST/SpiritDoc/whatever > > translation phase to generate BoostBook instead of HTML. > >Yeah, especially since docutils (ReST) already contains a DocBook >generating backend. It can also be done with XSLT from the ReST XML >generator. But ReST doesn't contain ways to represent many of the >C++-specific logical elements of BoostBook, IIUC. These could always be added in by the author after the BoostBook files have been generated, if you are planning to migrate from ReST/SpiritDocs to BoostBook. This was the idea, to give the user a tool/toolset to help migrate to the BoostBook format. > >> > As far as I understand it, the BoostBook tool is relatively new and > >> > has a learning curve, so it will take some time to migrate the docs >to > >> > this format. > >> > >>And it requires writing in XML :-( or translating to it :-|. > > > > XML is essentially a stricter version of HTML. > >I thought the main thing was that it is extensible. It is. You can have any element and attribute structure you want; this is how it is extensible. What I was referring to is that XML emposes stricter formatting rules than HTML, requiring attributes to be quoted and all tags to have corresponding closing tags (the open/close tag being a special case). This might be new to people familiar only with HTML and so may make mistakes, resulting in an invalid XML file. Automating this conversion process will help ensure a valid XML file to work from. > > The main problem I see is porting the existing HTML-based docs to > > BoostBook. The others can be generated using a BoostBook generator > > (not sure how easy it will be for each format). > >There's no advantage to translating into BoostBook, AFAICT, unless >we're going to stick in its special logical elements. We can always >just use a common .css if you just want a similar look. I thought one of the ideas behind BoostBook was to have a documentation format that will eventually become Boost-wide. If this is the case, then having tools like what I'm suggesting will help speed up the migration process. If not, then what you are suggesting is also fine. Regards, Reece _________________________________________________________________ It's fast, it's easy and it's free. Get MSN Messenger today! http://www.msn.co.uk/messenger

1 0

Thread lib (good reason for static)
by Peter Danford 11 Jun '04

11 Jun '04

[This is from an email about the impact of not having a static Threads lib available anymore. I wanted to post this here to see if anyone else has similar experiences since I have only seen posts about the problem of distributing RTL dlls so far - not computation time impact.] Hello, I was reading some some of the posts about your thread implementation and the removal of the static RTL option. I understand the technical reason for this, but I wanted to submit another compelling reason (for me) for the need for static RTL linkage even if this means loosing some of the features under win32. Some of the software I develop deals with different types of analysis (such as scientific and financial). The runs can take as much as 20 hours to complete. If I do nothing but change to link with dynamic RTL, the run time is increased on average of 35%. This means and additional 7 hours of run time! So dynamic RTL is really undesirable. Do you see any way of making a more limited static Boost.threads available in the next Boost release? [Above I say "limited" because some of the features of the Thread lib must be implemented via dll to work under win32.] Peter --------------------------------- Do you Yahoo!? Friends. Fun. Try the all-new Yahoo! Messenger

10 46

RE: [boost] Re: Documentation
by Reece Dunn 11 Jun '04

11 Jun '04

David Abrahams wrote: >"Reece Dunn" <msclrhd(a)hotmail.com> writes: > > David B. Held wrote: > >>If we're going to demand uniformity in licensing, should we also > >>demand more uniformity in documentation? > > > > I agree on this. At the moment, we have the docs written in HTML, > > those in BoostBook format (like Boost.Any and the BBv2 docs) and the > > Spirit docs which have their own format. > >And others; we used ReST for the iterator library. It should be fairly easy to adapt the ReST/SpiritDoc/whatever translation phase to generate BoostBook instead of HTML. > > As far as I understand it, the BoostBook tool is relatively new and > > has a learning curve, so it will take some time to migrate the docs to > > this format. > >And it requires writing in XML :-( or translating to it :-|. XML is essentially a stricter version of HTML. The main problem I see is porting the existing HTML-based docs to BoostBook. The others can be generated using a BoostBook generator (not sure how easy it will be for each format). I have a Spirit-based HTML parser that does basic correction (not as good as HTMLTidy) that is event based (like SAX for XML) and can be used to generate XHTML. It should be possible to adapt this to generate a BoostBook document that can be used as the basis for migrated docs. The SpiritDoc processor tool is freely available and should be easy to migrate to BoostBook output. Other generators can be produced for ReST, etc. NOTE: The output of these should be seen as a basis for the actual docs as they won't be able to properly detect things like the licence info, author name or library name/category; these will need to be edited by the document provider. (It all depends on how this information is encoded in the base text. If there is a markup pattern, then it should be possible to identify this and extract the information.) Let me know if I should take a look into seeing how feasible this is. Having a common document format will help maintain a consistent navigation/look. Also, it should be possible to generate frames as an alternate look by editing the BoostBook stylesheets and with a little wizardry it should be possible ot turn this on or off, depending on your preferences like it is currently possible to select Spirit or DocBook navigation styles. Regards, Reece _________________________________________________________________ Want to block unwanted pop-ups? Download the free MSN Toolbar now! http://toolbar.msn.co.uk/

2 1

feature request: file lock
by Jordi Vilar 11 Jun '04

11 Jun '04

Hi, I propose the inclussion of the following set of private members to the boost::filesystem::fstream class: private: void do_lock(); bool do_trylock(); bool do_timedlock(const xtime& xt); void do_unlock(); void do_lock(cv_state& state); void do_unlock(cv_state& state); this will enable additional classes: namespace boost { namespace filesystem { class scooped_lock; class scooped_try_lock; class scoped_timed_lock; } } with an interface like boost::thread::scooped_lock and sccoped_try_lock. What do you think about this? If you want to proceed and need help, please let me know, I can provide an implementation for Windows (NT and above) and posix (I'm currently working on this functionality). Cordially, Jordi Vilar

1 0

program_options and optional
by Neal D. Becker 11 Jun '04

11 Jun '04

Seems natural to use boost::optional with program_options. Has anyone tried this? For example, consider this: int main (int argc, char** argv) { namespace po = boost::program_options; optional<double> x; po::options_description desc("Allowed options"); desc.add_options() ("help", "produce help message") ("x", po::value<double>()->notifier (var(x)=_1)) ; To set the optional<double> x, this uses boost::lambda. Is there a more direct approach?

2 1