PDL::Core - fundamental PDL functionality

Methods and functions for type conversions, PDL creation, type conversion, threading etc.

 use PDL::Core;             # Normal routines
 use PDL::Core ':Internal'; # Hairy routines

piddle constructor - creates new piddle from perl scalars/arrays and piddles

 $a = pdl(SCALAR|ARRAY REFERENCE|ARRAY);

 $a = pdl [1..10];             # 1D array
 $a = pdl ([1..10]);           # 1D array
 $a = pdl (1,2,3,4);           # Ditto
 $b = pdl [[1,2,3],[4,5,6]];   # 2D 3x2 array
 $b = pdl 42                   # 0-dimensional scalar
 $c = pdl $a;                  # Make a new copy
 $a = pdl([1,2,3],[4,5,6]);    # 2D
 $a = pdl([[1,2,3],[4,5,6]]);  # 2D

Note the last two are equivalent - a list is automatically converted to a list reference for syntactic convenience. i.e. you can omit the outer CW[]

You can mix and match arrays, array refs, and PDLs in your argument list, and CWpdl will sort them out. You get back a PDL whose last (slowest running) dim runs across the top level of the list you hand in, and whose first (fastest running) dim runs across the deepest level that you supply.

Throwing a PDL into the mix has the same effect as throwing in a list ref:

  pdl(pdl(1,2),[3,4])

is the same as

  pdl([1,2],[3,4]).

All of the dimensions in the list are padded-out with undefval to meet the widest dim in the list, so (e.g.)

  $a = pdl([[1,2,3],[2]])

gives you the same answer as

  $a = pdl([[1,2,3],[2,undef,undef]]);

CWpdl() is a functional synonym for the 'new' constructor, e.g.:

 $x = new PDL [1..10];

In order to control how undefs are handled in converting from perl lists to PDLs, one can set the variable CW$PDL::undefval. For example:

 $foo = [[1,2,undef],[undef,3,4]];
 $PDL::undefval = -999;
 $f = pdl $foo;
 print $f
 [
  [   1    2 -999]
  [-999    3    4]
 ]

CW$PDL::undefval defaults to zero.

Returns a 'null' piddle.

 $x = null;

CWnull() has a special meaning to PDL::PP. It is used to flag a special kind of empty piddle, which can grow to appropriate dimensions to store a result (as opposed to storing a result in an existing piddle).

 perldl> sumover sequence(10,10), $ans=null;p $ans
 [45 145 245 345 445 545 645 745 845 945]

Returns a 'null' piddle.

 $x = PDL->nullcreate($arg)

This is an routine used by many of the threading primitives (i.e. sumover, minimum, etc.) to generate a null piddle for the function's output that will behave properly for derived (or subclassed) PDL objects.

For the above usage: If CW$arg is a PDL, or a derived PDL, then CW$arg->null is returned. If CW$arg is a scalar (i.e. a zero-dimensional PDL) then CW$PDL->null is returned.

 PDL::Derived->nullcreate(1)
   returns PDL::Derived->null.
 PDL->nullcreate($pdlderived)
   returns $pdlderived->null.

Return the number of elements in a piddle

 $n = nelem($piddle); $n = $piddle->nelem;

 $mean = sum($data)/nelem($data);

Return piddle dimensions as a perl list

 @dims = $piddle->dims;  @dims = dims($piddle);

 perldl> p @tmp = dims zeroes 10,3,22
 10 3 22

Returns the number of dimensions in a piddle. Alias for getndims.

Returns the number of dimensions in a piddle

 $ndims = $piddle->getndims;

 perldl> p zeroes(10,3,22)->getndims
 3

Returns the size of the given dimension of a piddle. Alias for getdim.

Returns the size of the given dimension.

 $dim0 = $piddle->getdim(0);

 perldl> p zeroes(10,3,22)->getdim(1)
 3

Negative indices count from the end of the dims array. Indices beyond the end will return a size of 1. This reflects the idea that any pdl is equivalent to an infinitely dimensional array in which only a finite number of dimensions have a size different from one. For example, in that sense a 3D piddle of shape [3,5,2] is equivalent to a [3,5,2,1,1,1,1,1,....] piddle. Accordingly,

  print $a->getdim(1);

will print 1 for most practically encountered piddles.

alternate piddle constructor - ensures arg is a piddle

 $a = topdl(SCALAR|ARRAY REFERENCE|ARRAY);

The difference between pdl() and CWtopdl() is that the latter will just 'fall through' if the argument is already a piddle. It will return a reference and NOT a new copy.

This is particulary useful if you are writing a function which is doing some fiddling with internals and assumes a piddle argument (e.g. for method calls). Using CWtopdl() will ensure nothing breaks if passed with '2'.

Note that CWtopdl() is not exported by default (see example below for usage).

 use PDL::Core ':Internal'; # use the internal routines of 
                            # the Core module

 $a = topdl 43;             # $a is piddle with value '43'
 $b = topdl $piddle;        # fall through
 $a = topdl (1,2,3,4);      # Convert 1D array

Internal: Return the numeric value identifying the piddle datatype

 $x = $piddle->get_datatype;

Mainly used for internal routines.

NOTE: get_datatype returns 'just a number' not any special type object, unlike type.

Returns the size of a piddle datatype in bytes.

Note that CWhowbig() is not exported by default (see example below for usage).

 use PDL::Core ':Internal'; # use the internal routines of 
                            # the Core module

 $size = howbig($piddle->get_datatype);

Mainly used for internal routines.

NOTE: NOT a method! This is because get_datatype returns 'just a number' not any special object.

 perldl> p howbig(ushort([1..10])->get_datatype)
 2

Return the internal data for a piddle, as a perl SCALAR ref.

Most piddles hold their internal data in a packed perl string, to take advantage of perl's memory management. This gives you direct access to the string, which is handy when you need to manipulate the binary data directly (e.g. for file I/O). If you modify the string, you'll need to call upd_data afterward, to make sure that the piddle points to the new location of the underlying perl variable.

You shouldn't mess with the SV unless you've called make_physical or something similar. You definitely don't want to do anything to the SV to truncate or deallocate the string, unless you correspondingly call reshape to make the PDL match its new data dimension.

You definitely don't want to use get_dataref unless you know what you are doing (or are trying to find out): you can end up scrozzling memory if you shrink or eliminate the string representation of the variable. Here be dragons.

Update the data pointer in a piddle to match its perl SV.

This is useful if you've been monkeying with the packed string representation of the PDL, which you probably shouldn't be doing anyway. (see get_dataref.)

Returns the piddle thread IDs as a perl list

Note that CWthreadids() is not exported by default (see example below for usage).

 use PDL::Core ':Internal'; # use the internal routines of 
                            # the Core module

 @ids = threadids $piddle;

Turn on/off dataflow

 $x->doflow;  doflow($x);

Whether or not a piddle is indulging in dataflow

 something if $x->flows; $hmm = flows($x);

new piddle constructor method

 $x = PDL->new(SCALAR|ARRAY|ARRAY REF);

 $x = PDL->new(4);
 $y = new PDL [1..10];

Constructs piddle from perl numbers and lists.

Make a physical copy of a piddle

 $new = $old->copy;

Since CW$new = $old just makes a new reference, the CWcopy method is provided to allow real independent copies to be made.

Return an explicit copy of the header of a PDL.

hdr_copy is just a wrapper for the internal routine _hdr_copy, which takes the hash ref itself. That is the routine which is used to make copies of the header during normal operations if the hdrcpy() flag of a PDL is set.

General-purpose deep copies are expensive in perl, so some simple optimization happens:

If the header is a tied array or a blessed hash ref with an associated method called CWcopy, then that ->copy method is called. Otherwise, all elements of the hash are explicitly copied. References are recursively deep copied.

This routine seems to leak memory.

Return a piddle which is the same as the argument except that all threadids have been removed.

 $y = $x->unwind;

Make sure the data portion of a piddle can be accessed from XS code.

 $a->make_physical;
 $a->call_my_xs_method;

Ensures that a piddle gets its own allocated copy of data. This obviously implies that there are certain piddles which do not have their own data. These are so called virtual piddles that make use of the vaffine optimisation (see PDL::Indexing). They do not have their own copy of data but instead store only access information to some (or all) of another piddle's data.

Note: this function should not be used unless absolutely neccessary since otherwise memory requirements might be severly increased. Instead of writing your own XS code with the need to call CWmake_physical you might want to consider using the PDL preprocessor (see PDL::PP) which can be used to transparently access virtual piddles without the need to physicalise them (though there are exceptions).

Insert a 'dummy dimension' of given length (defaults to 1)

No relation to the 'Dungeon Dimensions' in Discworld!

Negative positions specify relative to last dimension, i.e. CWdummy(-1) appends one dimension at end, CWdummy(-2) inserts a dummy dimension in front of the last dim, etc.

If you specify a dimension position larger than the existing dimension list of your PDL, the PDL gets automagically padded with extra dummy dimensions so that you get the dim you asked for, in the slot you asked for. This could cause you trouble if, for example, you ask for CW$a->dummy(5000,1) because CW$a will get 5,000 dimensions, each of rank 1.

Because padding at the beginning of the dimension list moves existing dimensions from slot to slot, it's considered unsafe, so automagic padding doesn't work for large negative indices only for large positive indices.

 $y = $x->dummy($position[,$dimsize]);

 perldl> p sequence(3)->dummy(0,3)
 [
  [0 0 0]
  [1 1 1]
  [2 2 2]
 ]

 perldl> p sequence(3)->dummy(3,2)
 [
  [
   [0 1 2]
  ]
  [
   [0 1 2]
  ]
 ]

 perldl> p sequence(3)->dummy(-3,2)
 For safety, <pos> < -(dims+1) is not allowed in dummy, allowed min=-2.

clumps several dimensions into one large dimension

If called with one argument CW$n clumps the first CW$n dimensions into one. For example, if CW$a has dimensions CW(5,3,4) then after

 $b = $a->clump(2);   # Clump 2 first dimensions

the variable CW$b will have dimensions CW(15,4) and the element CW$b->at(7,3) refers to the element CW$a->at(1,2,3).

Use CWclump(-1) to flatten a piddle. The method flat is provided as a convenient alias.

Clumping with a negative dimension in general leaves that many dimensions behind e.g. clump(-2) clumps all of the first few dimensions into a single one, leaving a 2-D piddle.

If CWclump is called with an index list with more than one element it is treated as a list of dimensions that should be clumped together into one. The resulting clumped dim is placed at the position of the lowest index in the list. This convention ensures that CWclump does the expected thing in the usual cases. The following example demonstrates typical usage:

  $a = sequence 2,3,3,3,5; # 5D piddle
  $c = $a->clump(1..3);    # clump all the dims 1 to 3 into one
  print $c->info;          # resulting 3D piddle has clumped dim at pos 1
 PDL: Double D [2,27,5]

define functions that support threading at the perl level

 thread_define 'tline(a(n);b(n))', over {
  line $_[0], $_[1]; # make line compliant with threading
 };

CWthread_define provides some support for threading (see PDL::Indexing) at the perl level. It allows you to do things for which you normally would have resorted to PDL::PP (see PDL::PP); however, it is most useful to wrap existing perl functions so that the new routine supports PDL threading.

CWthread_define is used to define new threading aware functions. Its first argument is a symbolic repesentation of the new function to be defined. The string is composed of the name of the new function followed by its signature (see PDL::Indexing and PDL::PP) in parentheses. The second argument is a subroutine that will be called with the slices of the actual runtime arguments as specified by its signature. Correct dimension sizes and minimal number of dimensions for all arguments will be checked (assuming the rules of PDL threading, see PDL::Indexing).

The actual work is done by the CWsignature class which parses the signature string, does runtime dimension checks and the routine CWthreadover that generates the loop over all appropriate slices of pdl arguments and creates pdls as needed.

Similar to CWpp_def and its CWOtherPars option it is possible to define the new function so that it accepts normal perl args as well as piddles. You do this by using the CWNOtherPars parameter in the signature. The number of CWNOtherPars specified will be passed unaltered into the subroutine given as the second argument of CWthread_define. Let's illustrate this with an example:

 PDL::thread_define 'triangles(inda();indb();indc()), NOtherPars => 2',
  PDL::over {
    ${$_[3]} .= $_[4].join(',',map {$_->at} @_[0..2]).",-1,\n";
  };

This defines a function CWtriangles that takes 3 piddles as input plus 2 arguments which are passed into the routine unaltered. This routine is used to collect lists of indices into a perl scalar that is passed by reference. Each line is preceded by a prefix passed as CW$_[4]. Here is typical usage:

 $txt = '';
 triangles(pdl(1,2,3),pdl(1),pdl(0),\$txt," "x10);
 print $txt;

resulting in the following output

 1,1,0,-1,
 2,1,0,-1,
 3,1,0,-1,

which is used in PDL::Graphics::TriD::VRML to generate VRML output.

Currently, this is probably not much more than a POP (proof of principle) but is hoped to be useful enough for some real life work.

Check PDL::PP for the format of the signature. Currently, the CW[t] qualifier and all type qualifiers are ignored.

Use explicit threading over specified dimensions (see also PDL::Indexing)

 $b = $a->thread($dim,[$dim1,...])

 $a = zeroes 3,4,5;
 $b = $a->thread(2,0);

Same as PDL::thread1, i.e. uses thread id 1.

Returns the multidimensional diagonal over the specified dimensions.

 $d = $x->diagonal(dim1, dim2,...)

 perldl> $a = zeroes(3,3,3);
 perldl> ($b = $a->diagonal(0,1))++;
 perldl> p $a
 [
  [
   [1 0 0]
   [0 1 0]
   [0 0 1]
  ]
  [
   [1 0 0]
   [0 1 0]
   [0 0 1]
  ]
  [
   [1 0 0]
   [0 1 0]
   [0 0 1]
  ]
 ]

Explicit threading over specified dims using thread id 1.

 $xx = $x->thread1(3,1)

 Wibble

Convenience function interfacing to PDL::Slices::threadI.

Explicit threading over specified dims using thread id 2.

 $xx = $x->thread2(3,1)

 Wibble

Convenience function interfacing to PDL::Slices::threadI.

Explicit threading over specified dims using thread id 3.

 $xx = $x->thread3(3,1)

 Wibble

Convenience function interfacing to PDL::Slices::threadI.

sever any links of this piddle to parent piddles

In PDL it is possible for a piddle to be just another view into another piddle's data. In that case we call this piddle a virtual piddle and the original piddle owning the data its parent. In other languages these alternate views sometimes run by names such as alias or smart reference.

Typical functions that return such piddles are CWslice, CWxchg, CWindex, etc. Sometimes, however, you would like to separate the virtual piddle from its parent's data and just give it a life of its own (so that manipulation of its data doesn't change the parent). This is simply achieved by using CWsever. For example,

   $a = $pdl->index(pdl(0,3,7))->sever;
   $a++;       # important: $pdl is not modified!

In many (but not all) circumstances it acts therefore similar to copy. However, in general performance is better with CWsever and secondly, CWsever doesn't lead to futile copying when used on piddles that already have their own data. On the other hand, if you really want to make sure to work on a copy of a piddle use copy.

   $a = zeroes(2);
   $a->sever;   # NOOP since $a is already its own boss!

Again note: CWsever is not the same as copy! For example,

   $a = zeroes(1); # $a does not have a parent, i.e. it is not a slice etc
   $b = $a->sever; # $b is now pointing to the same piddle as $a
   $b++;
   print $a;
 [1]

but

   $a = zeroes(1);
   $b = $a->copy; # $b is now pointing to a new piddle
   $b++;
   print $a;
 [0]

Return formatted information about a piddle.

 $x->info($format_string);

 print $x->info("Type: %T Dim: %-15D State: %S");

Returns a string with info about a piddle. Takes an optional argument to specify the format of information a la sprintf. Format specifiers are in the form CW%<width><letter> where the width is optional and the letter is one of

T

Type

D

Formatted Dimensions

F

Dataflow status

S

Some internal flags (P=physical,V=Vaffine,C=changed,B=may contain bad data)

C

Class of this piddle, i.e. CWref $pdl

A

Address of the piddle struct as a unique identifier

M

Calculated memory consumption of this piddle's data area

test for approximately equal values (relaxed CW==)

  # ok if all corresponding values in
  # piddles are within 1e-8 of each other
  print "ok\n" if all approx $a, $b, 1e-8;

CWapprox is a relaxed form of the CW== operator and often more appropriate for floating point types (CWfloat and CWdouble).

Usage:

  $res = approx $a, $b [, $eps]

The optional parameter CW$eps is remembered across invocations and initially set to 1e-6, e.g.

  approx $a, $b;         # last $eps used (1e-6 initially)
  approx $a, $b, 1e-10;  # 1e-10
  approx $a, $b;         # also 1e-10

Convenience interface to slice, allowing easier inclusion of dimensions in perl code.

 $a = $x->mslice(...);

 # below is the same as $x->slice("5:7,:,3:4:2")
 $a = $x->mslice([5,7],X,[3,4,2]);

Internally used interface to slice and dice that is the runtime part of the PDL::NiceSlice implementation.

 $a = $x->nslice(...);

 # below is the same as $x->slice("5:7,:,3:4:2")
 $a = $x->nslice([5,7],X,[3,4,2]);

It implements a superset of mslice's features. Should probably not be used in your scripts. Rather resort to the PDL::NiceSlice interface.

Flag a piddle so that the next operation is done 'in place'

 somefunc($x->inplace); somefunc(inplace $x);

In most cases one likes to use the syntax CW$y = f($x), however in many case the operation CWf() can be done correctly 'in place', i.e. without making a new copy of the data for output. To make it easy to use this, we write CWf() in such a way that it operates in-place, and use CWinplace to hint that a new copy should be disabled. This also makes for clear syntax.

Obviously this will not work for all functions, and if in doubt see the function's documentation. However one can assume this is true for all elemental functions (i.e. those which just operate array element by array element like CWlog10).

 perldl> $x = xvals zeroes 10;
 perldl> log10(inplace $x)
 perldl> p $x
 [      -Inf 0    0.30103 0.47712125 0.60205999    0.69897
 0.77815125 0.84509804 0.90308999 0.95424251]

Test the in-place flag on a piddle

  $out = ($in->is_inplace) ? $in : zeroes($in);
  $in->set_inplace(0)

Provides access to the inplace hint flag, within the perl millieu. That way functions you write can be inplace aware... If given an argument the inplace flag will be set or unset depending on the value at the same time. Can be used for shortcut tests that delete the inplace flag while testing:

  $out = ($in->is_inplace(0)) ? $in : zeroes($in); # test & unset!

Set the in-place flag on a piddle

  $out = ($in->is_inplace) ? $in : zeroes($in);
  $in->set_inplace(0);

Provides access to the inplace hint flag, within the perl millieu. Useful mainly for turning it OFF, as inplace turns it ON more conveniently.

Return back either the argument pdl or a copy of it depending on whether it be flagged in-place or no. Handy for building inplace-aware functions.

Internal method: create piddle by specification

This is the argument processing method called by zeroes and some other functions which constructs piddles from argument lists of the form:

 [type], $nx, $ny, $nz,...

For CW$nx, CW$ny, etc. 0 and 1D piddles are allowed. Giving those has the same effect as if saying CW$arg->list, e.g.

   1, pdl(5,2), 4

is equivalent to

   1, 5, 2, 4

Note, however, that in all functions using CWnew_from_specification calling CWfunc $piddle will probably not do what you want. So to play safe use (e.g. with zeroes)

  $pdl = zeroes $dimpdl->list;

Calling

  $pdl = zeroes $dimpdl;

will rather be equivalent to

  $pdl = zeroes $dimpdl->dims;

However,

  $pdl = zeroes ushort, $dimpdl;

will again do what you intended since it is interpreted as if you had said

  $pdl = zeroes ushort, $dimpdl->list;

This is unfortunate and confusing but no good solution seems obvious that would not break existing scripts.

Test whether a piddle is empty

 print "The piddle has zero dimension\n" if $pdl->isempty;

This function returns 1 if the piddle has zero elements. This is useful in particular when using the indexing function which. In the case of no match to a specified criterion, the returned piddle has zero dimension.

 perldl> $a=sequence(1)
 perldl> $i=which($a < -1)
 perldl> print "I found no matches!\n" if ($a->isempty);

Note that having zero elements is rather different from the concept of being a null piddle, see the PDL::FAQ and PDL::Indexing manpages for discussions of this.

construct a zero filled piddle from dimension list or template piddle.

Various forms of usage,

(i) by specification or (ii) by template piddle:

 # usage type (i):
 $a = zeroes([type], $nx, $ny, $nz,...);
 $a = PDL->zeroes([type], $nx, $ny, $nz,...);
 $a = $pdl->zeroes([type], $nx, $ny, $nz,...);
 # usage type (ii):
 $a = zeroes $b;
 $a = $b->zeroes
 zeroes inplace $a;     # Equivalent to   $a .= 0;
 $a->inplace->zeroes;   #  ""

 perldl> $z = zeroes 4,3
 perldl> p $z
 [
  [0 0 0 0]
  [0 0 0 0]
  [0 0 0 0]
 ]
 perldl> $z = zeroes ushort, 3,2 # Create ushort array
 [ushort() etc. with no arg returns a PDL::Types token]

See also new_from_specification for details on using piddles in the dimensions list.

construct a one filled piddle

 $a = ones([type], $nx, $ny, $nz,...);
 etc. (see 'zeroes')

 see zeroes() and add one

See also new_from_specification for details on using piddles in the dimensions list.

Change the shape (i.e. dimensions) of a piddle, preserving contents.

 $x->reshape(N); reshape($x, NEWDIMS);

The data elements are preserved, obviously they will wrap differently and get truncated if the new array is shorter. If the new array is longer it will be zero-padded.

***Potential incompatibility with earlier versions of PDL**** If the list of CWNEWDIMS is empty CWreshape will just drop all dimensions of size 1 (preserving the number of elements):

  $a = sequence(3,4,5);
  $b = $a(1,3);
  $b->reshape();
  print $b->info;
 PDL: Double D [5]

Dimensions of size 1 will also be dropped if CWreshape is invoked with the argument -1:

  $b = $a->reshape(-1);

As opposed to CWreshape without arguments, CWreshape(-1) preserves dataflow:

  $a = ones(2,1,2);
  $b = $a(0)->reshape(-1);
  $b++;
  print $a;
 [
  [
   [2 1]
  ]
  [
   [2 1]
  ]
 ]

Note: an explicit copy of slices is generally forced - this is the only way (for now) of stopping a crash if CW$x is a slice. Important: Physical piddles are changed inplace!

 perldl> $x = sequence(1)
 perldl> reshape $x,3,4; p $x
 [
  [0 1 2]
  [3 4 5]
  [6 7 8]
  [9 0 0]
 ]
 perldl> reshape $x,5; p $x
 [0 1 2 3 4]

eliminate all singleton dimensions (dims of size 1)

 $b = $a(0,0)->squeeze;

Alias for CWreshape(-1). Removes all singleton dimensions and preserves dataflow. A more concise interface is provided by PDL::NiceSlice via modifiers:

 use PDL::NiceSlice;
 $b = $a(0,0;-); # same as $a(0,0)->squeeze

flatten a piddle (alias for CW$pdl-clump(-1)>)

  $srt = $pdl->flat->qsort;

Useful method to make a 1D piddle from an arbitrarily sized input piddle. Data flows back and forth as usual with slicing routines. Falls through if argument already <= 1D.

Generic datatype conversion function

 $y = convert($x, $newtype);

 $y = convert $x, long
 $y = convert $x, ushort

CW$newtype is a type number, for convenience they are returned by CWlong() etc when called without arguments.

byte|short|ushort|long|longlong|float|double convert shorthands

 $y = double $x; $y = ushort [1..10];
 # all of byte|short|ushort|long|float|double behave similarly

When called with a piddle argument, they convert to the specific datatype.

When called with a numeric or list / listref argument they construct a new piddle. This is a convenience to avoid having to be long-winded and say CW$x = long(pdl(4))

Thus one can say:

 $a = float(1,2,3,4);           # 1D
 $a = float([1,2,3],[4,5,6]);   # 2D
 $a = float([[1,2,3],[4,5,6]]); # 2D

Note the last two are equivalent - a list is automatically converted to a list reference for syntactic convenience. i.e. you can omit the outer CW[]

When called with no arguments return a special type token. This allows syntactical sugar like:

 $x = ones byte, 1000,1000;

This example creates a large piddle directly as byte datatype in order to save memory.

In order to control how undefs are handled in converting from perl lists to PDLs, one can set the variable CW$PDL::undefval; see the function pdl() for more details.

 perldl> p $x=sqrt float [1..10]
 [1 1.41421 1.73205 2 2.23607 2.44949 2.64575 2.82843 3 3.16228]
 perldl> p byte $x
 [1 1 1 2 2 2 2 2 3 3]

Convert to byte datatype - see 'Datatype_conversions'

Convert to short datatype - see 'Datatype_conversions'

Convert to ushort datatype - see 'Datatype_conversions'

Convert to long datatype - see 'Datatype_conversions'

Convert to longlong datatype - see 'Datatype_conversions'

Convert to float datatype - see 'Datatype_conversions'

Convert to double datatype - see 'Datatype_conversions'

return the type of a piddle as a blessed type object

A convenience function for use with the piddle constructors, e.g.

 $b = PDL->zeroes($a->type,$a->dims,3);
 die "must be float" unless $a->type == float;

See also the discussion of the CWPDL::Type class in PDL::Types. Note that the CWPDL::Type objects have overloaded comparison and stringify operators so that you can compare and print types:

 $a = $a->float if $a->type < float;
 $t = $a->type; print "Type is $t\";

Convert piddle to perl list

 @tmp = list $x;

Obviously this is grossly inefficient for the large datasets PDL is designed to handle. This was provided as a get out while PDL matured. It should now be mostly superseded by superior constructs, such as PP/threading. However it is still occasionally useful and is provied for backwards compatibility.

 for (list $x) {
   # Do something on each value...
 }

list converts any bad values into the string 'BAD'.

Convert piddle indices to perl list

 @tmp = listindices $x;

CW@tmp now contains the values CW0..nelem($x).

Obviously this is grossly inefficient for the large datasets PDL is designed to handle. This was provided as a get out while PDL matured. It should now be mostly superseded by superior constructs, such as PP/threading. However it is still occasionally useful and is provied for backwards compatibility.

 for $i (listindices $x) {
   # Do something on each value...
 }

Set a single value inside a piddle

 set $piddle, @position, $value

CW@position is a coordinate list, of size equal to the number of dimensions in the piddle. Occasionally useful, mainly provided for backwards compatibility as superseded by use of slice and assigment operator CW.=.

 perldl> $x = sequence 3,4
 perldl> set $x, 2,1,99
 perldl> p $x
 [
  [ 0  1  2]
  [ 3  4 99]
  [ 6  7  8]
  [ 9 10 11]
 ]

Returns a single value inside a piddle as perl scalar.

 $z = at($piddle, @position); $z=$piddle->at(@position);

CW@position is a coordinate list, of size equal to the number of dimensions in the piddle. Occasionally useful in a general context, quite useful too inside PDL internals.

 perldl> $x = sequence 3,4
 perldl> p $x->at(1,2)
 7

return a single value from a piddle as a scalar

  $val = $a(1)->sclr;
  $val = sclr inner($a,$b);

The CWsclr method is useful to turn a piddle into a normal Perl scalar. Its main advantage over using CWat for this purpose is the fact that you do not need to worry if the piddle is 0D, 1D or higher dimensional. Using CWat you have to supply the correct number of zeroes, e.g.

  $a = sequence(1);
  $b = $a->slice('4');
  print $b->sclr; # no problem
  print $b->at(); # error: needs at least one zero

CWsclr is generally used when a Perl scalar is required instead of a one-element piddle. If the input is a multielement piddle the first value is returned as a Perl scalar. You can optionally switch on checks to ensure that the input piddle has only one element:

  PDL->sclr({Check => 'warn'}); # carp if called with multi-el pdls
  PDL->sclr({Check => 'barf'}); # croak if called with multi-el pdls

are the commands to switch on warnings or raise an error if a multielement piddle is passed as input. Note that these options can only be set when CWsclr is called as a class method (see example above). Use

  PDL->sclr({Check=>0});

to switch these checks off again (default setting); When called as a class method the resulting check mode is returned (0: no checking, 1: warn, 2: barf).

concatentate piddles to N+1 dimensional piddle

Takes a list of N piddles of same shape as argument, returns a single piddle of dimension N+1

 perldl> $x = cat ones(3,3),zeroes(3,3),rvals(3,3); p $x
 [
  [
   [1 1 1]
   [1 1 1]
   [1 1 1]
  ]
  [
   [0 0 0]
   [0 0 0]
   [0 0 0]
  ]
  [
   [1 1 1]
   [1 0 1]
   [1 1 1]
  ]
 ]

The output piddle is set bad if any input piddles have their bad flag set.

Opposite of 'cat' :). Split N dim piddle to list of N-1 dim piddles

Takes a single N-dimensional piddle and splits it into a list of N-1 dimensional piddles. The breakup is done along the last dimension. Note the dataflown connection is still preserved by default, e.g.:

 perldl> $p = ones 3,3,3
 perldl> ($a,$b,$c) = dog $p
 perldl> $b++; p $p
 [
  [
   [1 1 1]
   [1 1 1]
   [1 1 1]
  ]
  [
   [2 2 2]
   [2 2 2]
   [2 2 2]
  ]
  [
   [1 1 1]
   [1 1 1]
   [1 1 1]
  ]
 ]

 Break => 1   Break dataflow connection (new copy)

The output piddles are set bad if the original piddle has its bad flag set.

Standard error reporting routine for PDL.

CWbarf() is the routine PDL modules should call to report errors. This is because CWbarf() will report the error as coming from the correct line in the module user's script rather than in the PDL module.

It does this magic by unwinding the stack frames until it reaches a package NOT beginning with CW"PDL::". If you DO want it to report errors in some module PDL::Foo (e.g. when debugging PDL::Foo) then set the variable CW$PDL::Foo::Debugging=1.

Additionally if you set the variable CW$PDL::Debugging=1 you will get a COMPLETE stack trace back up to the top level package.

Finally CWbarf() will try and report usage information from the PDL documentation database if the error message is of the form 'Usage: func'.

Remember CWbarf() is your friend. *Use* it!

At the perl level:

 barf("User has too low an IQ!");

In C or XS code:

 barf("You have made %d errors", count);

Note: this is one of the few functions ALWAYS exported by PDL::Core

Retrieve header information from a piddle

 $pdl=rfits('file.fits');
 $h=$pdl->gethdr;
 print "Number of pixels in the X-direction=$$h{NAXIS1}\n";

The CWgethdr function retrieves whatever header information is contained within a piddle. The header can be set with sethdr and is always a hash reference or undef.

CWgethdr returns undef if the piddle has not yet had a header defined; compare with CWhdr and CWfhdr, which are guaranteed to return a defined value.

Note that gethdr() works by reference: you can modify the header in-place once it has been retrieved:

  $a  = rfits($filename);
  $ah = $a->gethdr();
  $ah->{FILENAME} = $filename;

It is also important to realise that in most cases the header is not automatically copied when you copy the piddle. See hdrcpy to enable automatic header copying.

Here's another example: a wrapper around rcols that allows your piddle to remember the file it was read from and the columns could be easily written (here assuming that no regexp is needed, extensions are left as an exercise for the reader)

 sub ext_rcols {
    my ($file, @columns)=@_;
    my $header={};
    $$header{File}=$file;
    $$header{Columns}=\@columns;

    @piddles=rcols $file, @columns;
    foreach (@piddles) { $_->sethdr($header); } 
    return @piddles;
 }

Retrieve or set header information from a piddle

 $pdl->hdr->{CDELT1} = 1;

The CWhdr function allows convenient access to the header of a piddle. Unlike CWgethdr it is guaranteed to return a defined value, so you can use it in a hash dereference as in the example. If the header does not yet exist, it gets autogenerated as an empty hash.

Note that this is usually but not always What You Want. If you want to use a tied Astro::FITS::Header hash, for example, you should either construct it yourself and use CWsethdr to put it into the piddle, or use fhdr instead. (Note that you should be able to write out the FITS file successfully regardless of whether your PDL has a tied FITS header object or a vanilla hash).

Retrieve or set FITS header information from a piddle

 $pdl->fhdr->{CDELT1} = 1;

The CWfhdr function allows convenient access to the header of a piddle. Unlike CWgethdr it is guaranteed to return a defined value, so you can use it in a hash dereference as in the example. If the header does not yet exist, it gets autogenerated as a tied Astro::FITS::Header hash.

Astro::FITS::Header tied hashes are better at matching the behavior of FITS headers than are regular hashes. In particular, the hash keys are CAsE INsEnSItiVE, unlike normal hash keys. See Astro::FITS::Header for details.

If you do not have Astro::FITS::Header installed, you get back a normal hash instead of a tied object.

Set header information of a piddle

 $pdl = zeroes(100,100);
 $h = {NAXIS=>2, NAXIS1=>100, NAXIS=>100, COMMENT=>"Sample FITS-style header"};
 # add a FILENAME field to the header
 $$h{FILENAME} = 'file.fits';
 $pdl->sethdr( $h );

The CWsethdr function sets the header information for a piddle. You must feed in a hash ref or undef, and the header field of the PDL is set to be a new ref to the same hash (or undefined).

The hash ref requirement is a speed bump put in place since the normal use of headers is to store fits header information and the like. Of course, if you want you can hang whatever ugly old data structure you want off of the header, but that makes life more complex.

Remember that the hash is not copied the header is made into a ref that points to the same underlying data. To get a real copy without making any assumptions about the underlying data structure, you can use one of the following:

  use PDL::IO::Dumper;
  $pdl->sethdr( deep_copy($h) );

(which is slow but general), or

  $pdl->sethdr( PDL::_hdr_copy($h) )

(which uses the built-in sleazy deep copier), or (if you know that all the elements happen to be scalars):

  { my %a = %$h;
    $pdl->sethdr(\%a);
  }

which is considerably faster but just copies the top level.

The CWsethdr function must be given a hash reference or undef. For further information on the header, see gethdr, hdr, fhdr and hdrcpy.

switch on/off/examine automatic header copying

 print "hdrs will be copied" if $a->hdrcpy;
 $a->hdrcpy(1);       # switch on automatic header copying 
 $b = $a->sumover;    # and $b will inherit $a's hdr
 $a->hdrcpy(0);       # and now make $a non-infectious again

CWhdrcpy without an argument just returns the current setting of the flag. See also hcpy which returns its PDL argument (and so is useful in method-call pipelines).

Normally, the optional header of a piddle is not copied automatically in pdl operations. Switching on the hdrcpy flag using the CWhdrcpy method will enable automatic hdr copying. Note that an actual deep copy gets made, which is rather processor-inefficient so avoid using header copying in tight loops!

Most PDLs have the CWhdrcpy flag cleared by default; however, some routines (notably rfits) set it by default where that makes more sense.

The CWhdrcpy flag is viral: if you set it for a PDL, then derived PDLs will get copies of the header and will also have their CWhdrcpy flags set. For example:

  $a = xvals(50,50);
  $a->hdrcpy(1);
  $a->hdr->{FOO} = "bar";
  $b = $a++;
  $c = $b++;
  print $b->hdr->{FOO}, " - ", $c->hdr->{FOO}, "\n";
  $b->hdr->{FOO} = "baz";
  print $a->hdr->{FOO}, " - ", $b->hdr->{FOO}, " - ", $c->hdr->{FOO}, "\n";

will print:

  bar - bar
  bar - baz - bar

Switch on/off automatic header copying, with PDL pass-through

  $a = rfits('foo.fits')->hcpy(0);
  $a = rfits('foo.fits')->hcpy(1);

CWhcpy sets or clears the hdrcpy flag of a PDL, and returns the PDL itself. That makes it convenient for inline use in expressions.

Copyright (C) Karl Glazebrook (kgb@aaoepp.aao.gov.au), Tuomas J. Lukka, (lukka@husc.harvard.edu) and Christian Soeller (c.soeller@auckland.ac.nz) 1997. Modified, Craig DeForest (deforest@boulder.swri.edu) 2002. All rights reserved. There is no warranty. You are allowed to redistribute this software / documentation under certain conditions. For details, see the file COPYING in the PDL distribution. If this file is separated from the PDL distribution, the copyright notice should be included in the file.