MESSAGE
| DATE | 2011-05-24 |
| FROM | Ruben Safir
|
| SUBJECT | Subject: [NYLXS - HANGOUT] [ruben@mrbrklyn.com: Re: [mrbrklyn@panix.com: Template Argument
|
----- Forwarded message from Ruben Safir -----
Date: Tue, 24 May 2011 12:40:26 -0400
From: Ruben Safir
To: Billy
Cc: Ruben Safir
Subject: Re: [mrbrklyn-at-panix.com: Template Argument type deterination]
User-Agent: Mutt/1.5.20 (2009-06-14)
On Tue, May 24, 2011 at 12:25:09PM -0400, Billy Donahue wrote:
> You didn't provide the constructor of stats::Distribution. If a constructor
> takes a single argument, it will be considered a conversion operator unless
> it is marked 'explicit'. Most style guides including Google's urge
> programmers to mark any single argument constructor as 'explicit', which
> make it ineligible as a conversion operator.
>
> Another thing is that maybe an integer is being used as a pointer or
> something.
> I can't really tell exactly through the levels of indirection.
>
> I'm uneasy because I don't see a single-arg constructor in the 'gdb' ptype
> output.
> But that isn't source code, that's gdb output. Only source code matters.
> The constructor might have been inlined and optimized away, for example.
>
> On Mon, May 23, 2011 at 4:50 PM, Ruben Safir wrote:
>
Ah - thank you for answering. It's driving me crazy and I'm not finding
help in texts or on the net.
So now I have a new concept, a conversion operator. What is a
conversion operator.
I tried to make this a simpler example, but it is really just related,
perhaps not exactly the same.
testm(intval) fails to compile in the below example, as I'd expect since
the param type is stats::Distribution and not an int. But
testme(intval) does compile with the same kind of promotion
http://www.nylxs.com/docs/workshops/test_del3.cpp.html
http://www.nylxs.com/docs/workshops/test_del3.tgz
There is a difference in this example though because the original
behavior was demonstated not by a templated function, but a member of a
templated class.
And I agree the indirection starts to make it very difficult to figure
out or predict that typename T is being sunstituted for as the template
class stacks get deeper.
Ruben
> > ----- Forwarded message from Ruben Safir -----
> >
> > Date: Mon, 23 May 2011 16:35:54 -0400
> > From: Ruben Safir
> > To: hangout-at-nylxs.com
> > Subject: Template Argument type deterination
> > User-Agent: Mutt/1.5.20 (2009-06-14)
> >
> >
> > I'm having trouble understanding why my program is making a change in
> > the
> > argument type of my template class function
> >
> > template < class T >
> > void List::find_value(T val)
> > {
> > cursor() = front();
> > while(cursor() != endd()){
> > if (*(cursor()->value()) == val)
> > return;
> > else{
> > cursor(cursor()->next());
> > }
> > }
> > if(*(endd()->value()) == val)
> > return;
> > else{
> > cursor() = 0; //park the cursor when nothing is found
> > }
> > }
> >
> >
> > It's part of a List template class
> >
> > I call it from a templated fucntion in a different namespace as follows:
> >
> >
> > tally = new chainlist::List >;
> > tallies = new chainlist::List< chainlist::List< stats::Distribution
> > > *>;
> >
> > std::cout << "The standard mean for picking 7 is ==> " <<
> > stats::mean_list(tallies, 7 ) << std::endl;;
> >
> > template
> > float mean_list(chainlist::List<
> > chainlist::List >* > * tallies, T search_val){
> > if(tallies->endd() == 0){
> > std::cout << "Empty List" << std::endl;
> > return 0.0;
> > }
> > int sum = 0;
> > chainlist::List > * dump;
> > tallies->cursor() = tallies->front();
> > while(tallies->cursor() != tallies->endd() ){
> > //tallies->cursor()->value()->find_value(search_val);
> > dump = *(tallies->cursor()->value());
> > //dump->cursor() = dump->endd();
> > //std::cout << "Testing\n" <<
> > //*(dump->cursor()->value()) << std::endl;
> > dump->find_value(search_val);
> > if(dump->cursor() != NULL)
> > sum += dump->cursor()->value()->population();
> > tallies->cursor(tallies->cursor()->next());
> > }
> > dump = *(tallies->cursor()->value());
> > dump->find_value(search_val);
> > if(dump->cursor() != 0)
> > sum += dump->cursor()->value()->population();
> > float tot = (float) sum/(float)(tallies->size());
> > std::cout << "Mean " << tot << " sum " << sum << " size " <<
> > tallies->size() << std::endl;
> > return tot;
> >
> > }
> >
> > Now T is an integer for mean_list, and the first argument is of type
> >
> > chainlist::List< chainlist::List >* > * when
> > called and the second argument is of type int
> >
> > I then pass int search_val to method chainlist::List::find_value(T)
> > where T = stats::Distribution (where T is an int)
> > on this line:
> >
> > dump->find_value(search_val);
> >
> > and that's where the weirdness happens. I have a type mismatch since
> > I'm sending by value an integer instead of a stats::Distribution
> >
> > Instead the program calls the stat::Distribution constructor using 7,
> > the value of search_val, as an argument redrawing search_val as type
> > to a Distribution
> >
> > :in gdb
> >
> >
> > 312 void List::find_value(T val)
> > 313 {
> > 314 cursor() = front();
> > 315 while(cursor() != endd()){
> > 316 if (*(cursor()->value()) == val)
> > 317 return;
> > 318 else{
> > 319 cursor(cursor()->next());
> > 320 }
> > 321 }
> > 322 if(*(endd()->value()) == val)
> > 323 return;
> > 324 else{
> > 325 cursor() = 0; //park the cursor when nothing is found
> > 326 }
> > 327 }
> >
> > chainlist::List >::find_value (this=0x804f008,
> > val=...) at linklist.h:314
> > (gdb) ptype val
> > type = class stats::Distribution {
> > public:
> > chainlist::List > *tally;
> > private:
> > int freq;
> > int occurances;
> >
> > public:
> > void Distribution(int, int);
> > void Distribution(void);
> > int description(void) const;
> > int population(void) const;
> > void increase_occ(void);
> > void descrease_occ(void);
> > int operator()(void);
> > bool operator==(stats::Distribution &);
> > bool operator<(stats::Distribution &);
> > float stddev(chainlist::List > *);
> > }
> >
> >
> > What rule for Template argument matching allows for this kind of magic?
> >
> >
> > ----- End forwarded message -----
> >
> > --
> > http://www.mrbrklyn.com - Interesting Stuff
> > http://www.nylxs.com - Leadership Development in Free Software
> >
> > So many immigrant groups have swept through our town that Brooklyn, like
> > Atlantis, reaches mythological proportions in the mind of the world - RI
> > Safir 1998
> >
> > http://fairuse.nylxs.com DRM is THEFT - We are the STAKEHOLDERS - RI
> > Safir 2002
> >
> > "Yeah - I write Free Software...so SUE ME"
> >
> > "The tremendous problem we face is that we are becoming sharecroppers to
> > our own cultural heritage -- we need the ability to participate in our own
> > society."
> >
> > "> I'm an engineer. I choose the best tool for the job, politics be
> > damned.<
> > You must be a stupid engineer then, because politcs and technology have
> > been attached at the hip since the 1st dynasty in Ancient Egypt. I guess
> > you missed that one."
> >
> > ? Copyright for the Digital Millennium
> >
>
>
>
> --
> ?n??uop ?ll?q
--
http://www.mrbrklyn.com - Interesting Stuff
http://www.nylxs.com - Leadership Development in Free Software
So many immigrant groups have swept through our town that Brooklyn, like Atlantis, reaches mythological proportions in the mind of the world - RI Safir 1998
http://fairuse.nylxs.com DRM is THEFT - We are the STAKEHOLDERS - RI Safir 2002
"Yeah - I write Free Software...so SUE ME"
"The tremendous problem we face is that we are becoming sharecroppers to our own cultural heritage -- we need the ability to participate in our own society."
"> I'm an engineer. I choose the best tool for the job, politics be damned.<
You must be a stupid engineer then, because politcs and technology have been attached at the hip since the 1st dynasty in Ancient Egypt. I guess you missed that one."
? Copyright for the Digital Millennium
----- End forwarded message -----
|
|
