www.digitalmars.com         C & C++   DMDScript  

D - DTL?

reply "Robert M. Münch" <robert.muench robertmuench.de> writes:
Hi, is there anyone working on a STL like library for D, a DTL library?

--
Robert M. Münch
IT & Management Freelancer
Mobile: +49 (0)177 2452 802
Fax   : +49 (0)721 8408 9112
Web   : http://www.robertmuench.de
Jan 10 2003
parent reply Evan McClanahan <evan dontSPAMaltarinteractive.com> writes:
Robert M. Münch wrote:
 Hi, is there anyone working on a STL like library for D, a DTL library?
 
 --
 Robert M. Münch
 IT & Management Freelancer
 Mobile: +49 (0)177 2452 802
 Fax   : +49 (0)721 8408 9112
 Web   : http://www.robertmuench.de

Daniel Yokomiso seems to be working on something like that, but i think that his focus is on mathematical computing, and similar things. I don't know but a quarter of the STL, but I think that many things included there are already in D, other than iterators, which have been said to be forthcoming. Well, there is map, which would be nice. Evan
Jan 10 2003
next sibling parent "Sean L. Palmer" <seanpalmer directvinternet.com> writes:
C++ stl map == D associative array, pretty much

Sean

"Evan McClanahan" <evan dontSPAMaltarinteractive.com> wrote in message
news:avmkat$9hq$1 digitaldaemon.com...
 Robert M. Münch wrote:
 Hi, is there anyone working on a STL like library for D, a DTL library?

 --
 Robert M. Münch
 IT & Management Freelancer
 Mobile: +49 (0)177 2452 802
 Fax   : +49 (0)721 8408 9112
 Web   : http://www.robertmuench.de

Daniel Yokomiso seems to be working on something like that, but i think that his focus is on mathematical computing, and similar things. I don't know but a quarter of the STL, but I think that many things included there are already in D, other than iterators, which have been said to be forthcoming. Well, there is map, which would be nice. Evan

Jan 10 2003
prev sibling parent reply "Robert M. Münch" <robert.muench robertmuench.de> writes:
"Evan McClanahan" <evan dontSPAMaltarinteractive.com> schrieb im Newsbeitrag
news:avmkat$9hq$1 digitaldaemon.com...

 I  don't know but a quarter of the STL, but I think that many things
 included there are already in D, other than iterators, which have been
 said to be forthcoming.

Hi, that's what I thought about too. But I see that all those algortihm stuff still would make sense for D too. Or am I missing something in D? Robert
Jan 11 2003
parent reply "Daniel Yokomiso" <daniel_yokomiso yahoo.com.br> writes:
"Robert M. Münch" <robert.muench robertmuench.de> escreveu na mensagem
news:avoova$1kb9$1 digitaldaemon.com...
 "Evan McClanahan" <evan dontSPAMaltarinteractive.com> schrieb im

 news:avmkat$9hq$1 digitaldaemon.com...

 I  don't know but a quarter of the STL, but I think that many things
 included there are already in D, other than iterators, which have been
 said to be forthcoming.

Hi, that's what I thought about too. But I see that all those algortihm stuff still would make sense for D too. Or am I missing something in D? Robert

Hi, DTL (Deimos Template Library) is my LGPLed project for D. The first release (www.minddrome.com/d/deimos/deimos-0.0.1.zip) included templates for extra array functions, functional programming idioms mostly (like fold, map, filter, etc.) but also some other array utilities. It includes a simple range template class and a range-set template class. Currently I'm working on a basic matrix template and some other numeric templates (rational and big numbers). There's some work to include a bunch of collection classes. If you have something you want first, let me know. And someday (when I have the time) I'll create make cvs project in http://www.opend.org for contributions. My goal is to provide a standard D templates library, something like STL + BOOST. Version 0.0.2 will be released sometime after Walter releases 0.51, there's some compiler and linker bugs I'm waiting to see smashed before continuing my work. Best regards, Daniel Yokomiso. P.S.: IMO DTL should provide enough functionality to attract developers spoiled by languages that provide a huge basic library. Therefore it should include: collections, numerics, xml processors, unicode classes, parser combinators, scientific classes (like units), concurrency and even standard D's AST manipulation classes (making tools writers life's easier). Any help will be appreciated: more tests, code, documentation, benchmarks, critics, etc.. Let's make a better D for everyone "There are perhaps 5% of the population that simply *can't* think. There are another 5% who *can*, and *do*. The remaining 90% *can* think, but *don't*." - Robert A. Heinlein --- Outgoing mail is certified Virus Free. Checked by AVG anti-virus system (http://www.grisoft.com). Version: 6.0.435 / Virus Database: 244 - Release Date: 30/12/2002
Jan 11 2003
next sibling parent reply Burton Radons <loth users.sourceforge.net> writes:
Daniel Yokomiso wrote:
 "Robert M. Münch" <robert.muench robertmuench.de> escreveu na mensagem
 news:avoova$1kb9$1 digitaldaemon.com...
 
"Evan McClanahan" <evan dontSPAMaltarinteractive.com> schrieb im

Newsbeitrag
news:avmkat$9hq$1 digitaldaemon.com...


I  don't know but a quarter of the STL, but I think that many things
included there are already in D, other than iterators, which have been
said to be forthcoming.

Hi, that's what I thought about too. But I see that all those algortihm stuff still would make sense for D too. Or am I missing something in D?

DTL (Deimos Template Library) is my LGPLed project for D. The first release (www.minddrome.com/d/deimos/deimos-0.0.1.zip) included templates for extra array functions, functional programming idioms mostly (like fold, map, filter, etc.) but also some other array utilities. It includes a simple range template class and a range-set template class. Currently I'm working on a basic matrix template and some other numeric templates (rational and big numbers). There's some work to include a bunch of collection classes. If you have something you want first, let me know. And someday (when I have the time) I'll create make cvs project in http://www.opend.org for contributions. My goal is to provide a standard D templates library, something like STL + BOOST. Version 0.0.2 will be released sometime after Walter releases 0.51, there's some compiler and linker bugs I'm waiting to see smashed before continuing my work.

If you pick up DLI (http://www.opend.org/dli/arc/dli-0.1.2.tar.gz), there's a dbignum.d module in the Phobos that you could use, which I stole from Python. Strangely enough, it was written long before D's operator overloading was in place, but the methods of the Bignum struct match their names.
 P.S.: IMO DTL should provide enough functionality to attract developers
 spoiled by languages that provide a huge basic library. Therefore it should
 include: collections, numerics, xml processors, unicode classes, parser
 combinators, scientific classes (like units), concurrency and even standard
 D's AST manipulation classes (making tools writers life's easier). Any help
 will be appreciated: more tests, code, documentation, benchmarks, critics,
 etc.. Let's make a better D for everyone

That's a little closer to the Python standard library then. I might be able to help after the next release of dig, although I might be swamped with offtime work depending.
Jan 11 2003
parent "Daniel Yokomiso" <daniel_yokomiso yahoo.com.br> writes:
Hi,

   Comments embedded.

"Burton Radons" <loth users.sourceforge.net> escreveu na mensagem
news:avqkj8$2iaf$1 digitaldaemon.com...
 Daniel Yokomiso wrote:
 "Robert M. Münch" <robert.muench robertmuench.de> escreveu na mensagem
 news:avoova$1kb9$1 digitaldaemon.com...

"Evan McClanahan" <evan dontSPAMaltarinteractive.com> schrieb im

Newsbeitrag
news:avmkat$9hq$1 digitaldaemon.com...


I  don't know but a quarter of the STL, but I think that many things
included there are already in D, other than iterators, which have been
said to be forthcoming.

Hi, that's what I thought about too. But I see that all those algortihm stuff still would make sense for D too. Or am I missing something in D?

DTL (Deimos Template Library) is my LGPLed project for D. The first release (www.minddrome.com/d/deimos/deimos-0.0.1.zip) included templates


 extra array functions, functional programming idioms mostly (like fold,


 filter, etc.) but also some other array utilities. It includes a simple
 range template class and a range-set template class. Currently I'm


 on a basic matrix template and some other numeric templates (rational


 big numbers). There's some work to include a bunch of collection


 you have something you want first, let me know. And someday (when I have


 time) I'll create make cvs project in http://www.opend.org for
 contributions. My goal is to provide a standard D templates library,
 something like STL + BOOST. Version 0.0.2 will be released sometime


 Walter releases 0.51, there's some compiler and linker bugs I'm waiting


 see smashed before continuing my work.

If you pick up DLI (http://www.opend.org/dli/arc/dli-0.1.2.tar.gz), there's a dbignum.d module in the Phobos that you could use, which I stole from Python. Strangely enough, it was written long before D's operator overloading was in place, but the methods of the Bignum struct match their names.

That's great! I plan to write a complete bignum library, following standards (http://www2.hursley.ibm.com/decimal/ for example), with real, integer, complex and rational bignums. But this module will be a starting point (I was planning to stole Sather bignums, they're report to be very fast).
 P.S.: IMO DTL should provide enough functionality to attract developers
 spoiled by languages that provide a huge basic library. Therefore it


 include: collections, numerics, xml processors, unicode classes, parser
 combinators, scientific classes (like units), concurrency and even


 D's AST manipulation classes (making tools writers life's easier). Any


 will be appreciated: more tests, code, documentation, benchmarks,


 etc.. Let's make a better D for everyone

That's a little closer to the Python standard library then. I might be able to help after the next release of dig, although I might be swamped with offtime work depending.

It would be wonderful. I know you're creating some vector handling code in dig (OpenGL stuff I think). If we provide a single module for vectors and matrices, usable in numerical and graphic libraries, D's users would have less trouble to work with it. Best regards, Daniel Yokomiso. "All the water in the world cannot sink the smallest ship unless it gets inside." - Boyd K. Packer --- Outgoing mail is certified Virus Free. Checked by AVG anti-virus system (http://www.grisoft.com). Version: 6.0.435 / Virus Database: 244 - Release Date: 30/12/2002
Jan 12 2003
prev sibling parent reply "Walter" <walter digitalmars.com> writes:
"Daniel Yokomiso" <daniel_yokomiso yahoo.com.br> wrote in message
news:avqbiv$2dat$1 digitaldaemon.com...
 Version 0.0.2 will be released sometime after
 Walter releases 0.51, there's some compiler and linker bugs I'm waiting to
 see smashed before continuing my work.

I understand. I put D on the shelf for the last month to do a major upgrade to DMC++. I plan to get back on D now and address some of the outstanding issues.
Jan 12 2003
parent reply "Daniel Yokomiso" <daniel_yokomiso yahoo.com.br> writes:
"Walter" <walter digitalmars.com> escreveu na mensagem
news:avt1r8$v5e$2 digitaldaemon.com...
 "Daniel Yokomiso" <daniel_yokomiso yahoo.com.br> wrote in message
 news:avqbiv$2dat$1 digitaldaemon.com...
 Version 0.0.2 will be released sometime after
 Walter releases 0.51, there's some compiler and linker bugs I'm waiting


 see smashed before continuing my work.

I understand. I put D on the shelf for the last month to do a major

 to DMC++. I plan to get back on D now and address some of the outstanding
 issues.

Take your time. D is great, I enjoy helping it get better. It's just that I'm getting some bugs hard to nail down. My codebase is getting larger and I'm getting some misterious illegal access errors only when some of my modules get linked, but not when others get linked. I don't want to give you a "It crashes!" bug report, or a "Download my codebase and track down a nasty bug in more than 10000 lines of D code for me". --- Outgoing mail is certified Virus Free. Checked by AVG anti-virus system (http://www.grisoft.com). Version: 6.0.435 / Virus Database: 244 - Release Date: 31/12/2002
Jan 12 2003
parent reply "Walter" <walter digitalmars.com> writes:
"Daniel Yokomiso" <daniel_yokomiso yahoo.com.br> wrote in message
news:avt7mv$15qv$2 digitaldaemon.com...
 Take your time. D is great, I enjoy helping it get better. It's just that
 I'm getting some bugs hard to nail down. My codebase is getting larger and
 I'm getting some misterious illegal access errors only when some of my
 modules get linked, but not when others get linked. I don't want to give

 a "It crashes!" bug report, or a "Download my codebase and track down a
 nasty bug in more than 10000 lines of D code for me".

There are several bugs in DMD that happen when multiple files are compiled on the same command line. Try splitting them into separate lines, i.e. for: dmd foo bar replace with: dmd -c foo dmd -c bar dmd foo.obj bar.obj
Jan 23 2003
parent reply "Daniel Yokomiso" <daniel_yokomiso yahoo.com.br> writes:
"Walter" <walter digitalmars.com> escreveu na mensagem
news:b0oh2c$2f4i$1 digitaldaemon.com...
 "Daniel Yokomiso" <daniel_yokomiso yahoo.com.br> wrote in message
 news:avt7mv$15qv$2 digitaldaemon.com...
 Take your time. D is great, I enjoy helping it get better. It's just


 I'm getting some bugs hard to nail down. My codebase is getting larger


 I'm getting some misterious illegal access errors only when some of my
 modules get linked, but not when others get linked. I don't want to give

 a "It crashes!" bug report, or a "Download my codebase and track down a
 nasty bug in more than 10000 lines of D code for me".

There are several bugs in DMD that happen when multiple files are compiled on the same command line. Try splitting them into separate lines, i.e.

     dmd foo bar
 replace with:
     dmd -c foo
     dmd -c bar
     dmd foo.obj bar.obj

There's any easy way to do this using dmd's make? Right now I define some macros with the source and object file names, to simplify my target definition. With 17 modules already (I try to keep unrelated code in independent modules) would be difficult to change my targets. Also I'm not proficient with make, there's any manual/tutorial for it? The documentation seems pretty scarce and I was spoiled by the Java compiler, so I'm not familiar with other make utilities (besides Ant). --- Outgoing mail is certified Virus Free. Checked by AVG anti-virus system (http://www.grisoft.com). Version: 6.0.443 / Virus Database: 248 - Release Date: 10/1/2003
Jan 23 2003
next sibling parent reply "Christopher J. Sequeira" <csequeir __mit.edu> writes:
I didn't pick any particular link, but here's a Google search:

http://www.google.com/search?hl=en&ie=UTF-8&oe=UTF-8&q=gnu+make+tutorial

I'm pretty sure there's a GNU make for Win32, probably available from 
the CygWin or MinGW sets of code.  DJGPP should also come with "make", 
or you could at least download the GNU make source code and build it 
with DJGPP or Cygwin.  Cheers!


Daniel Yokomiso wrote:
 
 There's any easy way to do this using dmd's make? Right now I define some
 macros with the source and object file names, to simplify my target
 definition. With 17 modules already (I try to keep unrelated code in
 independent modules) would be difficult to change my targets. Also I'm not
 proficient with make, there's any manual/tutorial for it? The documentation
 seems pretty scarce and I was spoiled by the Java compiler, so I'm not
 familiar with other make utilities (besides Ant).

Christopher J. Sequeira '05 csequeir __mit.edu (remove hyphens for email)
Jan 23 2003
parent "Daniel Yokomiso" <daniel_yokomiso yahoo.com.br> writes:
"Christopher J. Sequeira" <csequeir __mit.edu> escreveu na mensagem
news:b0pdh5$2vuq$1 digitaldaemon.com...
 I didn't pick any particular link, but here's a Google search:

 http://www.google.com/search?hl=en&ie=UTF-8&oe=UTF-8&q=gnu+make+tutorial

 I'm pretty sure there's a GNU make for Win32, probably available from
 the CygWin or MinGW sets of code.  DJGPP should also come with "make",
 or you could at least download the GNU make source code and build it
 with DJGPP or Cygwin.  Cheers!

don't need to start a cygwin session just to compile and run my tests. --- Outgoing mail is certified Virus Free. Checked by AVG anti-virus system (http://www.grisoft.com). Version: 6.0.443 / Virus Database: 248 - Release Date: 10/1/2003
Jan 23 2003
prev sibling parent reply "Walter" <walter digitalmars.com> writes:
"Daniel Yokomiso" <daniel_yokomiso yahoo.com.br> wrote in message
news:b0osdn$2lh4$1 digitaldaemon.com...
 There's any easy way to do this using dmd's make? Right now I define some
 macros with the source and object file names, to simplify my target
 definition. With 17 modules already (I try to keep unrelated code in
 independent modules) would be difficult to change my targets. Also I'm not
 proficient with make, there's any manual/tutorial for it? The

 seems pretty scarce and I was spoiled by the Java compiler, so I'm not
 familiar with other make utilities (besides Ant).

Take a look at \dmd\src\phobos\makefile, that should be all you need.
Jan 23 2003
parent "Daniel Yokomiso" <daniel_yokomiso yahoo.com.br> writes:
"Walter" <walter digitalmars.com> escreveu na mensagem
news:b0qdah$hlm$1 digitaldaemon.com...
 "Daniel Yokomiso" <daniel_yokomiso yahoo.com.br> wrote in message
 news:b0osdn$2lh4$1 digitaldaemon.com...
 There's any easy way to do this using dmd's make? Right now I define


 macros with the source and object file names, to simplify my target
 definition. With 17 modules already (I try to keep unrelated code in
 independent modules) would be difficult to change my targets. Also I'm


 proficient with make, there's any manual/tutorial for it? The

 seems pretty scarce and I was spoiled by the Java compiler, so I'm not
 familiar with other make utilities (besides Ant).

Take a look at \dmd\src\phobos\makefile, that should be all you need.

Jan 24 2003