Hi

I have a little strange problem, probably I am doing something wrong.

I am writing a small parser for an 3d .OBJ format. I am using sscanf for parsing. Most of the results are correct, except one place.

Here is the part of input text file:

g object_1
f 12/934/12 228/1/228 231/936/231
f 228/1/228 72/75/72 229/935/229
f 231/936/231 229/935/229 75/937/75
f 227/2/227 231/936/231 230/10/230
f 72/75/72 695/960/695 1346/3/1346
f 695/960/695 2/1352/2 1345/1328/1345
f 1346/3/1346 1345/1328/1345 73/938/73
f 229/935/229 1346/3/1346 233/939/233
f 654/7/654 232/6/232 75/937/75
f 233/939/233 236/4/236 654/7/654
f 75/937/75 233/939/233 654/7/654
f 644/5/644 236/4/236 233/939/233
f 73/938/73 234/1340/234 644/5/644
f 233/939/233 73/938/73 644/5/644
f 236/4/236 644/5/644 647/940/647
f 644/5/644 234/1340/234 645/859/645
f 647/940/647 645/859/645 646/234/646
f 643/8/643 647/940/647 235/1032/235
f 232/6/232 654/7/654 656/941/656
f 654/7/654 236/4/236 643/8/643
f 656/941/656 643/8/643 655/231/655
f 653/942/653 656/941/656 655/231/655
f 666/15/666 242/943/242 71/9/71
f 230/10/230 244/16/244 666/15/666
f 71/9/71 230/10/230 666/15/666
f 658/12/658 244/16/244 230/10/230
f 75/937/75 232/6/232 658/12/658
f 230/10/230 75/937/75 658/12/658
f 244/16/244 658/12/658 660/11/660


--------------------------
Here is the code I use for reading that :

// 1 Vertex   --  2 Vertex   --  3 Vertex
text.sscanf("f %d/%d/%d %d/%d/%d %d/%d/%d",
&triangles[ arpos ].vertexes[0].vid[0] ,
&triangles[ arpos ].vertexes[0].vid[1] ,
&triangles[ arpos ].vertexes[0].vid[2] ,

&triangles[ arpos ].vertexes[1].vid[0] ,
&triangles[ arpos ].vertexes[1].vid[1] ,
&triangles[ arpos ].vertexes[1].vid[2] ,

&triangles[ arpos ].vertexes[2].vid[0] ,
&triangles[ arpos ].vertexes[2].vid[1] ,
&triangles[ arpos ].vertexes[2].vid[2] );

----
text is char[]
results are just int.

Anyway all the code is parsed pretty well, except for a few places:

12/934/12 228/1/228 231/936/231
228/1/228 72/75/72 229/935/2298 <---- 2298, should be 229 !
231/936/231 229/935/229 75/937/75
227/2/227 231/936/231 230/10/230
72/75/72 695/960/695 1346/3/1346
695/960/695 2/1352/2 1345/1328/1345
1346/3/1346 1345/1328/1345 73/938/73
229/935/229 1346/3/1346 233/939/233
654/7/654 232/6/232 75/937/75
233/939/233 236/4/236 654/7/654
75/937/75 233/939/233 654/7/654
644/5/644 236/4/236 233/939/233
73/938/73 234/1340/234 644/5/644
233/939/233 73/938/73 644/5/644
236/4/236 644/5/644 647/940/647
644/5/644 234/1340/234 645/859/645
647/940/647 645/859/645 646/234/646
643/8/643 647/940/647 235/1032/235
232/6/232 654/7/654 656/941/656
654/7/654 236/4/236 643/8/643
656/941/656 643/8/643 655/231/655
653/942/653 656/941/656 655/231/655
666/15/666 242/943/242 71/9/71
230/10/230 244/16/244 666/15/666
71/9/71 230/10/230 666/15/666
658/12/658 244/16/244 230/10/230
75/937/75 232/6/232 658/12/6583  <---- 6583, should be 658 !
230/10/230 75/937/75 658/12/658
244/16/244 658/12/658 660/11/660

So any ideas, what might be bugging here :) ?

"DMINATOR" <DMINATOR_member@pathlink.com> wrote in message news:e8c4qc$2pji$1@digitaldaemon.com...

> Anyway all the code is parsed pretty well, except for a few places:
>
> 228/1/228 72/75/72 229/935/2298 <---- 2298, should be 229 ! 75/937/75 232/6/232 658/12/6583  <---- 6583, should be 658 !
>
> So any ideas, what might be bugging here :) ?

Hmm.  I tried duplicating the problem, but it seemed to work just fine.

The only thing I can think of is that since sscanf is a C function, it expects strings to be null-terminated, and perhaps your string data might not be.  So try something like

sscanf(toStringz(text), "format string", &all, &the, &outputs);

See if that helps it.

DMINATOR wrote:
> Hi
> 
> I have a little strange problem, probably I am doing something wrong.
> 
> I am writing a small parser for an 3d .OBJ format. I am using sscanf for
> parsing. Most of the results are correct, except one place.
> 
> Here is the part of input text file:
> 
> g object_1
> f 12/934/12 228/1/228 231/936/231
> f 228/1/228 72/75/72 229/935/229
> f 231/936/231 229/935/229 75/937/75
> f 227/2/227 231/936/231 230/10/230
> f 72/75/72 695/960/695 1346/3/1346
> f 695/960/695 2/1352/2 1345/1328/1345
> f 1346/3/1346 1345/1328/1345 73/938/73
> f 229/935/229 1346/3/1346 233/939/233
> f 654/7/654 232/6/232 75/937/75
> f 233/939/233 236/4/236 654/7/654
> f 75/937/75 233/939/233 654/7/654
> f 644/5/644 236/4/236 233/939/233
> f 73/938/73 234/1340/234 644/5/644
> f 233/939/233 73/938/73 644/5/644
> f 236/4/236 644/5/644 647/940/647
> f 644/5/644 234/1340/234 645/859/645
> f 647/940/647 645/859/645 646/234/646
> f 643/8/643 647/940/647 235/1032/235
> f 232/6/232 654/7/654 656/941/656
> f 654/7/654 236/4/236 643/8/643
> f 656/941/656 643/8/643 655/231/655
> f 653/942/653 656/941/656 655/231/655
> f 666/15/666 242/943/242 71/9/71
> f 230/10/230 244/16/244 666/15/666
> f 71/9/71 230/10/230 666/15/666
> f 658/12/658 244/16/244 230/10/230
> f 75/937/75 232/6/232 658/12/658
> f 230/10/230 75/937/75 658/12/658
> f 244/16/244 658/12/658 660/11/660
> 
> 
> --------------------------
> Here is the code I use for reading that :
> 

Here is my function for reading OBJ faces:

alias char[] string;

void read_face( string data, Group target )
{
  if ( target is null ) throw new Exception("Face data with no group found!" );
  string[] info = split( data );
  if ( info.length != 3 ) throw new Exception("Malformed face data:" ~ data );

  face* f = target.addface();
  void  add( int which, string vi, string ci )
  {
    f.v[which] = atoi(vi)-1;
    f.c[which] = atoi(ci)-1;
  }

  void add_info( int index )
  {
    string s = info[index];
    string[] arr = split( s, "/" );
    if ( arr.length<2 ) throw new Exception("Malformed face data group:"~ s );
    else add( index, arr[0], arr[1] );
  }

  add_info( 0 );
  add_info( 1 );
  add_info( 2 );
}

-DavidM

On Wed, 05 Jul 2006 14:22:02 +0300, David Medlock <noone@nowhere.com> wrote:

> DMINATOR wrote:
>> Hi
>>  I have a little strange problem, probably I am doing something wrong.
>>  I am writing a small parser for an 3d .OBJ format. I am using sscanf for
>> parsing. Most of the results are correct, except one place.
>>  Here is the part of input text file:
>>  g object_1
>> f 12/934/12 228/1/228 231/936/231
>> f 228/1/228 72/75/72 229/935/229
>> f 231/936/231 229/935/229 75/937/75
>> f 227/2/227 231/936/231 230/10/230
>> f 72/75/72 695/960/695 1346/3/1346
>> f 695/960/695 2/1352/2 1345/1328/1345
>> f 1346/3/1346 1345/1328/1345 73/938/73
>> f 229/935/229 1346/3/1346 233/939/233
>> f 654/7/654 232/6/232 75/937/75
>> f 233/939/233 236/4/236 654/7/654
>> f 75/937/75 233/939/233 654/7/654
>> f 644/5/644 236/4/236 233/939/233
>> f 73/938/73 234/1340/234 644/5/644
>> f 233/939/233 73/938/73 644/5/644
>> f 236/4/236 644/5/644 647/940/647
>> f 644/5/644 234/1340/234 645/859/645
>> f 647/940/647 645/859/645 646/234/646
>> f 643/8/643 647/940/647 235/1032/235
>> f 232/6/232 654/7/654 656/941/656
>> f 654/7/654 236/4/236 643/8/643
>> f 656/941/656 643/8/643 655/231/655
>> f 653/942/653 656/941/656 655/231/655
>> f 666/15/666 242/943/242 71/9/71
>> f 230/10/230 244/16/244 666/15/666
>> f 71/9/71 230/10/230 666/15/666
>> f 658/12/658 244/16/244 230/10/230
>> f 75/937/75 232/6/232 658/12/658
>> f 230/10/230 75/937/75 658/12/658
>> f 244/16/244 658/12/658 660/11/660
>>   --------------------------
>> Here is the code I use for reading that :
>>
>
> Here is my function for reading OBJ faces:
>
> alias char[] string;
>
> void read_face( string data, Group target )
> {
>    if ( target is null ) throw new Exception("Face data with no group found!" );
>    string[] info = split( data );
>    if ( info.length != 3 ) throw new Exception("Malformed face data:" ~ data );
>
>    face* f = target.addface();
>    void  add( int which, string vi, string ci )
>    {
>      f.v[which] = atoi(vi)-1;
>      f.c[which] = atoi(ci)-1;
>    }
>
>    void add_info( int index )
>    {
>      string s = info[index];
>      string[] arr = split( s, "/" );
>      if ( arr.length<2 ) throw new Exception("Malformed face data group:"~ s );
>      else add( index, arr[0], arr[1] );
>    }
>
>    add_info( 0 );
>    add_info( 1 );
>    add_info( 2 );
> }
>
> -DavidM
>


Thank you for reply.

I still haven't solved that sscanf problem event with C strings.
I tried to replicate the problem with a new project, and it worked fine. Still I used exactly the same code !

Can anyone take a look at the code, here is the link for the projects http://adogames.com/programming/problem.zip (One of them uses - with problem uses Derelict )

On Thu, 06 Jul 2006 21:17:09 +0300, DMINATOR <kna_int@mail.ru> wrote:

> On Wed, 05 Jul 2006 14:22:02 +0300, David Medlock <noone@nowhere.com> wrote:
>
>> DMINATOR wrote:
>>> Hi
>>>  I have a little strange problem, probably I am doing something wrong.
>>>  I am writing a small parser for an 3d .OBJ format. I am using sscanf for
>>> parsing. Most of the results are correct, except one place.
>>>  Here is the part of input text file:
>>>  g object_1
>>> f 12/934/12 228/1/228 231/936/231
>>> f 228/1/228 72/75/72 229/935/229
>>> f 231/936/231 229/935/229 75/937/75
>>> f 227/2/227 231/936/231 230/10/230
>>> f 72/75/72 695/960/695 1346/3/1346
>>> f 695/960/695 2/1352/2 1345/1328/1345
>>> f 1346/3/1346 1345/1328/1345 73/938/73
>>> f 229/935/229 1346/3/1346 233/939/233
>>> f 654/7/654 232/6/232 75/937/75
>>> f 233/939/233 236/4/236 654/7/654
>>> f 75/937/75 233/939/233 654/7/654
>>> f 644/5/644 236/4/236 233/939/233
>>> f 73/938/73 234/1340/234 644/5/644
>>> f 233/939/233 73/938/73 644/5/644
>>> f 236/4/236 644/5/644 647/940/647
>>> f 644/5/644 234/1340/234 645/859/645
>>> f 647/940/647 645/859/645 646/234/646
>>> f 643/8/643 647/940/647 235/1032/235
>>> f 232/6/232 654/7/654 656/941/656
>>> f 654/7/654 236/4/236 643/8/643
>>> f 656/941/656 643/8/643 655/231/655
>>> f 653/942/653 656/941/656 655/231/655
>>> f 666/15/666 242/943/242 71/9/71
>>> f 230/10/230 244/16/244 666/15/666
>>> f 71/9/71 230/10/230 666/15/666
>>> f 658/12/658 244/16/244 230/10/230
>>> f 75/937/75 232/6/232 658/12/658
>>> f 230/10/230 75/937/75 658/12/658
>>> f 244/16/244 658/12/658 660/11/660
>>>   --------------------------
>>> Here is the code I use for reading that :
>>>
>>
>> Here is my function for reading OBJ faces:
>>
>> alias char[] string;
>>
>> void read_face( string data, Group target )
>> {
>>    if ( target is null ) throw new Exception("Face data with no group found!" );
>>    string[] info = split( data );
>>    if ( info.length != 3 ) throw new Exception("Malformed face data:" ~ data );
>>
>>    face* f = target.addface();
>>    void  add( int which, string vi, string ci )
>>    {
>>      f.v[which] = atoi(vi)-1;
>>      f.c[which] = atoi(ci)-1;
>>    }
>>
>>    void add_info( int index )
>>    {
>>      string s = info[index];
>>      string[] arr = split( s, "/" );
>>      if ( arr.length<2 ) throw new Exception("Malformed face data group:"~ s );
>>      else add( index, arr[0], arr[1] );
>>    }
>>
>>    add_info( 0 );
>>    add_info( 1 );
>>    add_info( 2 );
>> }
>>
>> -DavidM
>>
>
>
> Thank you for reply.
>
> I still haven't solved that sscanf problem event with C strings.
> I tried to replicate the problem with a new project, and it worked fine. Still I used exactly the same code !
>
> Can anyone take a look at the code, here is the link for the projects http://adogames.com/programming/problem.zip (One of them uses - with problem uses Derelict )

Sorry that was http://adogames.com/programing/problem.zip

P.S. Is there any D sscanf analog ?

"DMINATOR" <kna_int@mail.ru> wrote in message
news:op.tb95hmzvfa6o3s@kna-x0tr7o23p0o.infonet.ee...
On Thu, 06 Jul 2006 21:17:09 +0300, DMINATOR <kna_int@mail.ru> wrote:

P.S. Is there any D sscanf analog ?

There is, but for some reason, it's not a separate function.  It's defined in std.stream.Stream, and is the method .readf().  If your input is a Stream (such as a std.stream.File), you can use readf to extract data from it.