1FITS(3)               User Contributed Perl Documentation              FITS(3)
2
3
4

NAME

6       PDL::IO::FITS -- Simple FITS support for PDL
7

SYNOPSIS

9        use PDL;
10        use PDL::IO::FITS;
11
12        $x = rfits('foo.fits');          # read a FITS file
13        $x->wfits('bar.fits');           # write a FITS file
14

DESCRIPTION

16       This module provides basic FITS support for PDL, in the sense of
17       reading and writing whole FITS files.  (For more complex operations,
18       such as prefiltering rows out of tables or performing operations on the
19       FITS file in-place on disk), you can use the Astro::FITS::CFITSIO
20       module that is available on CPAN.
21
22       Basic FITS image files are supported, along with BINTABLE and IMAGE
23       extensions.  ASCII Table support is planned, as are the HEASARC
24       bintable extensions that are recommended in the 1999 FITS standard.
25
26       Table support is based on hashes and named columns, rather than the
27       less convenient (but slightly more congruent) technique of perl lists
28       of numbered columns.
29
30       The principle interface routines are "rfits" and "wfits", for reading
31       and writing respectively.  FITS headers are returned as perl hashes or
32       (if the module is present) Astro::FITS::Header objects that are tied to
33       perl hashes.  Astro::FITS::Header objects provide convenient access
34       through the tied hash interface, but also allow you to control the card
35       structure in more detail using a separate method interface; see the
36       Astro::FITS::Header documentation for details.
37

AUTHOR

39       Copyright (C) Karl Glazebrook, Craig DeForest, and Doug Burke,
40       1997-2010.  There is no warranty.  You are allowed to redistribute
41       and/or modify this software under certain conditions.  For details, see
42       the file COPYING in the PDL distribution.  If this file is separated
43       from the PDL distribution, the copyright notice should be pasted into
44       in this file.
45

FUNCTIONS

47   rfits()
48       Simple piddle FITS reader.
49
50         $pdl = rfits('file.fits');   # Read a simple FITS image
51
52       Suffix magic:
53
54         $pdl = rfits('file.fits.gz'); # Read a file with gunzip(1)
55         $pdl = rfits('file.fits.Z');  # Read a file with uncompress(1)
56
57         $pdl = rfits('file.fits[2]');    # Read 2nd extension
58         $pdl = rfits('file.fits.gz[3]'); # Read 3rd extension
59         @pdls = rfits('file.fits');      # Read primary data and extensions
60
61       Tilde expansion:
62         #expand leading ~ to home directory (using glob())
63         $pdl = rfits '~/filename.fits';
64
65         $hdr = rfits('file.fits',{data=>0});  # Options hash changes behavior
66
67       In list context, "rfits" reads the primary image and all possible
68       extensions, returning them in the same order that they occurred in the
69       file -- except that, by default, the primary HDU is skipped if it
70       contains no data.  In scalar context, the default is to read the first
71       HDU that contains data. One can read other HDU's by using the [n]
72       syntax.  Using the [0] syntax forces a read of the first HDU,
73       regardless of whether it contains data or no.  Currently recognized
74       extensions are IMAGE and BINTABLE.  (See the addendum on EXTENSIONS for
75       details).
76
77       "rfits" accepts several options that may be passed in as a hash ref if
78       desired:
79
80       bscale (default=1)
81          Determines whether the data are linearly scaled using the
82          BSCALE/BZERO keywords in the FITS header.  To read in the exact data
83          values in the file, set this to 0.
84
85       data (default=1)
86          Determines whether to read the data, or just the header.  If you set
87          this to 0, you will get back the FITS header rather than the data
88          themselves.  (Note that the header is normally returned as the "hdr"
89          field of the returned PDL; this causes it to be returned as a hash
90          ref directly.)
91
92       hdrcpy (default=0)
93          Determines whether the hdrcpy flag is set in the returned PDL.
94          Setting the flag will cause an explicit deep copy of the header
95          whenever you use the returned PDL in an arithmetic or slicing
96          operation.  That is useful in many circumstances but also causes a
97          hit in speed.  When two or more PDLs with hdrcpy set are used in an
98          expression, the result gets the header of the first PDL in the
99          expression.  See hdrcpy for an example.
100
101       expand (default=1)
102          Determines whether auto-expansion of tile-compressed images should
103          happen.  Tile-compressed images are transmitted as binary tables
104          with particular fields ("ZIMAGE") set.  Leaving this alone does what
105          you want most of the time, unpacking such images transparently and
106          returning the data and header as if they were part of a normal IMAGE
107          extension.  Setting "expand" to 0 delivers the binary table, rather
108          than unpacking it into an image.
109
110       afh (default=1)
111          By default rfits uses Astro::FITS::Header tied-hash objects to
112          contain the FITS header information.  This permits explicit control
113          over FITS card information, and conforms well with the FITS
114          specification.  But Astro::FITS::Header objects are about 40-60x
115          more memory intensive than comparable perl hashes, and also use ~10x
116          more CPU to manage.  For jobs where header processing performance is
117          important (e.g. reading just the headers of 1,000 FITS files), set
118          afh to 0 to use the legacy parser and get a large boost in speed.
119
120       FITS image headers are stored in the output PDL and can be retrieved
121       with hdr or gethdr.  The hdrcpy flag of the PDL is set so that the
122       header is copied to derived piddles by default.  (This is inefficient
123       if you are planning to do lots of small operations on the data; clear
124       the flag with "->hcpy(0)" or via the options hash if that's the case.)
125
126       The header is a hash whose keys are the keywords in the FITS header.
127       If you have the "Astro::FITS::Header" module installed, the header is
128       actually a tied hash to a FITS header object, which can give you more
129       control over card order, comment fields, and variable types.  (see
130       Astro::FITS::Header for details).
131
132       The header keywords are converted to uppercase per the FITS standard.
133       Access is case-insensitive on the perl side, provided that
134       Astro::FITS::Header is installed.
135
136       If Astro::FITS::Header is not installed, then a built-in legacy parser
137       is used to generate the header hash.  Keyword-associated comments in
138       the headers are stored under the hash key "<keyword>_COMMENT>".  All
139       HISTORY cards in the header are collected into a single multiline
140       string stored in the "HISTORY" key.  All COMMENT cards are similarly
141       collected under the "COMMENT" key.
142
143       BSCALE/BZERO
144
145       If the BSCALE and/or BZERO keywords are set, they are applied to the
146       image before it is returned.  The returned PDL is promoted as necessary
147       to contain the multiplied values, and the BSCALE and BZERO keywords are
148       deleted from the header for clarity.  If you don't want this type of
149       processing, set 'bscale=>0' in the options hash.
150
151       EXTENSIONS
152
153       Sometimes a FITS file contains only extensions and a stub header in the
154       first header/data unit ("primary HDU").  In scalar context, you
155       normally only get back the primary HDU -- but in this special case, you
156       get back the first extension HDU.  You can force a read of the primary
157       HDU by adding a '[0]' suffix to the file name.
158
159       BINTABLE EXTENSIONS
160
161       Binary tables are handled. Currently only the following PDL datatypes
162       are supported: byte, short, ushort, long, float, and double. At present
163       ushort() data is written as a long rather than as a short with
164       TSCAL/ZERO; this may change.
165
166       The return value for a binary table is a hash ref containing the names
167       of the columns in the table (in UPPER CASE as per the FITS standard).
168       Each element of the hash contains a PDL (for numerical values) or a
169       perl list (for string values).  The PDL's 0th dimension runs across
170       rows; the 1st dimension runs across the repeat index within the row
171       (for rows with more than one value).  (Note that this is different from
172       standard threading order - but it allows Least Surprise to work when
173       adding more complicated objects such as collections of numbers (via the
174       repeat count) or variable length arrays.)
175
176       Thus, if your table contains a column named "FOO" with type "5D", the
177       expression
178
179         $x->{FOO}->((2))
180
181       returns a 5-element double-precision PDL containing the values of FOO
182       from the third row of the table.
183
184       The header of the table itself is parsed as with a normal FITS HDU, and
185       is returned in the element 'hdr' of the returned hash.  You can use
186       that to preserve the original column order or access the table at a low
187       level, if you like.
188
189       Scaling and zero-point adjustment are performed as with BSCALE/BZERO:
190       the appropriate keywords are deleted from the as-returned header.  To
191       avoid this behavior, set 'bscale=>0' in the options hash.
192
193       As appropriate, TSCAL/ZERO and TUNIT are copied into each column-PDL's
194       header as BSCALE/BZERO and BUNIT.
195
196       The main hash also contains the element 'tbl', which is set to 'binary'
197       to distinguish it from an ASCII table.
198
199       Because different columns in the table might have identical names in a
200       FITS file, the binary table reader practices collision avoidance.  If
201       you have multiple columns named "FOO", then the first one encountered
202       (numerically) gets the name "FOO", the next one gets "FOO_1", and the
203       next "FOO_2", etc.  The appropriate TTYPEn fields in the header are
204       changed to match the renamed column fields.
205
206       Columns with no name are assigned the name "COL_<n>", where <n> starts
207       at 1 and increments for each no-name column found.
208
209       Variable-length arrays are supported for reading.  They are unpacked
210       into PDLs that appear exactly the same as the output for fixed-length
211       rows, except that each row is padded to the maximum length given in the
212       extra characters -- e.g. a row with TFORM of 1PB(300) will yield an
213       NAXIS2x300 output field in the final hash.   The padding uses the TNULn
214       keyword for the column, or 0 if TNULn is not present.  The output hash
215       also gets an additional field, "len_<name>", that contains the number
216       of elements in each table row.
217
218       TILE-COMPRESSED IMAGES
219
220       CFITSIO and several large projects (including NASA's Solar Dynamics
221       Observatory) now support an unofficial extension to FITS that stores
222       images as a collection of individually compressed tiles within a
223       BINTABLE extension.  These images are automagically uncompressed by
224       default, and delivered as if they were normal image files.  You can
225       override this behavior by supplying the "expand" key in the options
226       hash.
227
228       Currently, only Rice compression is supported, though there is a
229       framework in place for adding other compression schemes.
230
231       BAD VALUE HANDLING
232
233       If a FITS file contains the "BLANK" keyword (and has "BITPIX > 0"), the
234       piddle will have its bad flag set, and those elements which equal the
235       "BLANK" value will be set bad.  For "BITPIX < 0", any NaN's are
236       converted to bad (if necessary).
237
238   rfitshdr()
239       Read only the header of a FITS file or an extension within it.
240
241       This is syntactic sugar for the "data=>0" option to rfits.
242
243       See rfits for details on header handling.  rfitshdr() runs the same
244       code to read the header, but returns it rather than reading in a data
245       structure as well.
246
247   wfits()
248       Simple PDL FITS writer
249
250         wfits $pdl, 'filename.fits', [$BITPIX], [$COMPRESSION_OPTIONS];
251         wfits $hash, 'filename.fits', [$OPTIONS];
252         $pdl->wfits('foo.fits',-32);
253
254       Suffix magic:
255
256         # Automatically compress through pipe to gzip
257         wfits $pdl, 'filename.fits.gz';
258         # Automatically compress through pipe to compress
259         wfits $pdl, 'filename.fits.Z';
260
261       Tilde expansion:
262         #expand leading ~ to home directory (using glob())
263         wfits $pdl, '~/filename.fits';
264
265       ·  Ordinary (PDL) data handling:
266
267          If the first argument is a PDL, then the PDL is written out as an
268          ordinary FITS file with a single Header/Data Unit of data.
269
270          $BITPIX is then optional and coerces the output data type according
271          to the standard FITS convention for the BITPIX field (with positive
272          values representing integer types and negative values representing
273          floating-point types).
274
275          If $pdl has a FITS header attached to it (actually, any hash that
276          contains a "SIMPLE=>T" keyword), then that FITS header is written
277          out to the file.  The image dimension tags are adjusted to the
278          actual dataset.  If there's a mismatch between the dimensions of the
279          data and the dimensions in the FITS header, then the header gets
280          corrected and a warning is printed.
281
282          If $pdl is a slice of another PDL with a FITS header already present
283          (and header copying enabled), then you must be careful.  "wfits"
284          will remove any extraneous "NAXISn" keywords (per the FITS
285          standard), and also remove the other keywords associated with that
286          axis: "CTYPEn", "CRPIXn", "CRVALn", "CDELTn", and "CROTAn".  This
287          may cause confusion if the slice is NOT out of the last dimension:
288          "wfits($x(:,(0),:),'file.fits');" and you would be best off
289          adjusting the header yourself before calling "wfits".
290
291          You can tile-compress images according to the CFITSIO extension to
292          the FITS standard, by adding an option hash to the arguments:
293
294          compress
295             This can be either unity, in which case Rice compression is used,
296             or a (case-insensitive) string matching the CFITSIO compression
297             type names.  Currently supported compression algorithms are:
298
299             ·  RICE_1 - linear Rice compression
300
301                This uses limited-symbol-length Rice compression, which works
302                well on low entropy image data (where most pixels differ from
303                their neighbors by much less than the dynamic range of the
304                image).
305
306          tilesize (default "[-1,1]")
307             This specifies the dimension of the compression tiles, in pixels.
308             You can hand in a PDL, a scalar, or an array ref. If you specify
309             fewer dimensions than exist in the image, the last dim is
310             repeated - so "32" yields 32x32 pixel tiles in a 2-D image.  A
311             dim of -1 in any dimension duplicates the image size, so the
312             default "[-1,1]" causes compression along individual rows.
313
314          tilesize (RICE_1 only; default 32)
315             For RICE_1, BLOCKSIZE indicates the number of pixel samples to
316             use for each compression block within the compression algorithm.
317             The blocksize is independent of the tile dimensions.  For RICE
318             compression the pixels from each tile are arranged in normal
319             pixel order (early dims fastest) and compressed as a linear
320             stream.
321
322       ·  Table handling:
323
324          If you feed in a hash ref instead of a PDL, then the hash ref is
325          written out as a binary table extension.  The hash ref keys are
326          treated as column names, and their values are treated as the data to
327          be put in each column.
328
329          For numeric information, the hash values should contain PDLs.  The
330          0th dim of the PDL runs across rows, and higher dims are written as
331          multi-value entries in the table (e.g. a 7x5 PDL will yield a single
332          named column with 7 rows and 5 numerical entries per row, in a
333          binary table).  Note that this is slightly different from the usual
334          concept of threading, in which dimension 1 runs across rows.
335
336          ASCII tables only allow one entry per column in each row, so if you
337          plan to write an ASCII table then all of the values of $hash should
338          have at most one dim.
339
340          All of the columns' 0 dims must agree in the threading sense. That
341          is to say, the 0th dimension of all of the values of $hash should be
342          the same (indicating that all columns have the same number of rows).
343          As an exception, if the 0th dim of any of the values is 1, or if
344          that value is a PDL scalar (with 0 dims), then that value is
345          "threaded" over -- copied into all rows.
346
347          Data dimensions higher than 2 are preserved in binary tables, via
348          the TDIMn field (e.g. a 7x5x3 PDL is stored internally as seven rows
349          with 15 numerical entries per row, and reconstituted as a 7x5x3 PDL
350          on read).
351
352          Non-PDL Perl scalars are treated as strings, even if they contain
353          numerical values.  For example, a list ref containing 7 values is
354          treated as 7 rows containing one string each.  There is no such
355          thing as a multi-string column in FITS tables, so any nonscalar
356          values in the list are stringified before being written.  For
357          example, if you pass in a perl list of 7 PDLs, each PDL will be
358          stringified before being written, just as if you printed it to the
359          screen.  This is probably not what you want -- you should use glue
360          to connect the separate PDLs into a single one.  (e.g.
361          "$x->glue(1,$y,$c)->mv(1,0)")
362
363          The column names are case-insensitive, but by convention the keys of
364          $hash should normally be ALL CAPS, containing only digits, capital
365          letters, hyphens, and underscores.  If you include other characters,
366          then case is smashed to ALL CAPS, whitespace is converted to
367          underscores, and unrecognized characters are ignored -- so if you
368          include the key "Au Purity (%)", it will be written to the file as a
369          column that is named "AU_PURITY".  Since this is not guaranteed to
370          produce unique column names, subsequent columns by the same name are
371          disambiguated by the addition of numbers.
372
373          You can specify the use of variable-length rows in the output,
374          saving space in the file.  To specify variable length rows for a
375          column named "FOO", you can include a separate key "len_FOO" in the
376          hash to be written.  The key's value should be a PDL containing the
377          number of actual samples in each row.  The result is a FITS P-type
378          variable length column that, upon read with "rfits()", will restore
379          to a field named FOO and a corresponding field named "len_FOO".
380          Invalid data in the final PDL consist of a padding value (which
381          defaults to 0 but which you may set by including a TNULL field in
382          the hdr specificaion).  Variable length arrays must be 2-D PDLs,
383          with the variable length in the 1 dimension.
384
385          Two further special keys, 'hdr' and 'tbl', can contain meta-
386          information about the type of table you want to write.  You may
387          override them by including an $OPTIONS hash with a 'hdr' and/or
388          'tbl' key.
389
390          The 'tbl' key, if it exists, must contain either 'ASCII' or 'binary'
391          (case-insensitive), indicating whether to write an ascii or binary
392          table.  The default is binary. [ASCII table writing is planned but
393          does not yet exist].
394
395          You can specify the format of the table quite specifically with the
396          'hdr' key or option field.  If it exists, then the 'hdr' key should
397          contain fields appropriate to the table extension being used.  Any
398          field information that you don't specify will be filled in
399          automatically, so (for example) you can specify that a particular
400          column name goes in a particular position, but allow "wfits" to
401          arrange the other columns in the usual alphabetical order into any
402          unused slots that you leave behind.  The "TFORMn", "TFIELDS",
403          "PCOUNT", "GCOUNT", "NAXIS", and "NAXISn" keywords are ignored:
404          their values are calculated based on the hash that you supply.  Any
405          other fields are passed into the final FITS header verbatim.
406
407          As an example, the following
408
409            $x = long(1,2,4);
410            $y = double(1,2,4);
411            wfits { 'COLA'=>$x, 'COLB'=>$y }, "table1.fits";
412
413          will create a binary FITS table called table1.fits which contains
414          two columns called "COLA" and "COLB". The order of the columns is
415          controlled by setting the "TTYPEn" keywords in the header array, so
416
417            $h = { 'TTYPE1'=>'Y', 'TTYPE2'=>'X' };
418            wfits { 'X'=>$x, 'Y'=>$y, hdr=>$h }, "table2.fits";
419
420          creates table2.fits where the first column is called "Y" and the
421          second column is "X".
422
423       ·  multi-value handling
424
425          If you feed in a perl list rather than a PDL or a hash, then each
426          element is written out as a separate HDU in the FITS file.  Each
427          element of the list must be a PDL or a hash. [This is not
428          implemented yet but should be soon!]
429
430       ·  DEVEL NOTES
431
432          ASCII tables are not yet handled but should be.
433
434          Binary tables currently only handle one vector (up to 1-D array) per
435          table entry; the standard allows more, and should be fully
436          implemented.  This means that PDL::Complex piddles currently can not
437          be written to disk.
438
439          Handling multidim arrays implies that perl multidim lists should
440          also be handled.
441
442       For integer types (ie "BITPIX > 0"), the "BLANK" keyword is set to the
443       bad value.  For floating-point types, the bad value is converted to NaN
444       (if necessary) before writing.
445
446   fits_field_cmp
447       fits_field_cmp
448
449       Sorting comparison routine that makes proper sense of the digits at the
450       end of some FITS header fields.  Sort your hash keys using
451       "fits_field_cmp" and you will get (e.g.) your "TTYPE" fields in the
452       correct order even if there are 140 of them.
453
454       This is a standard kludgey perl comparison sub -- it uses the magical
455       $a and $b variables, rather than normal argument passing.
456
457   _rows()
458       Return the number of rows in a variable for table entry
459
460       You feed in a PDL or a list ref, and you get back the 0th dimension.
461
462   _prep_table()
463       Accept a hash ref containing a table, and return a header describing
464       the table and a string to be written out as the table, or barf.
465
466       You can indicate whether the table should be binary or ascii.  The
467       default is binary; it can be overridden by the "tbl" field of the hash
468       (if present) or by parameter.
469
470
471
472perl v5.30.2                      2020-04-02                           FITS(3)
Impressum