1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
|
.help wfits May97 dataio
.ih
NAME
wfits -- convert individual IRAF image files to FITS image data
.ih
USAGE
wfits iraf_files fits_files
.ih
PARAMETERS
.ls iraf_files
The input IRAF image file(s), e.g. "image.imh" or "*.imh".
.le
.ls fits_files
The output FITS files.
Magnetic tape output is assumed if the first two characters of fits_files
are "mt", otherwise the disk output is assumed. Tape output will begin
at the file number specified in fits_files, e.g. file 5 if fits_files =
"mtb1600[5]", and the data in file 5 and succeeding files will be overwritten.
If no tape file number is specified in fits_files, the newtape parameter
is queried, and tape output will begin at BOT (beginning of tape) if
newtape = yes, otherwise at EOT (end of tape, after the double EOF).
Requesting a tape write at EOT on a blank tape may cause tape runaway.
In the case of disk output, fits_files may be either a file name template
or a root filename. In the former case there must be one output FITS file
name for every input image. In the latter case fits_files is a root name
and a sequence number will be appended to fits_files if the number of
input images > 1.
.le
.ls newtape
Boolean parameter specifying whether an output tape is blank or already
contains data. Newtape is requested only if no tape file number is specified in
fits_files, e.g. fits_files = "mtb1600".
.le
.ls bscale
The FITS bscale parameter, defined as p = i * bscale + bzero, where
p and i are the physical and tape data values respectively.
The bscale parameter is only requested if the scale switch is on
and the autoscale switch is off.
.le
.ls bzero
The FITS bzero parameter (see bscale for a definition).
Bzero is only requested if the scale switch is on and the autoscale
switch is off.
.le
.ls fextn = "fits"
The output fits file extension. If fextn is defined, an extension of
the form ".fextn", e.g. ".fits" is added to the output fits file name.
Fextn should be chosen to be compatible with one of the permitted fits
kernel extensions.
.le
.ls extensions = no
By default wfits writes each input image to a separate disk or tape FITS
file. If \fIextensions\fR is "yes", then wfits will write all the images in
the input image list to a single disk or tape FITS file using the FITS
standard IMAGE extension to write images other than the first. Extension
numbering is 0 indexed. The first image will be written to extension 1 if
\fIglobal_header\fR is "yes", or to extension 0 if \fIglobal_hdr\fR is "no".
.le
.ls global_hdr = yes
Write a short global header to the 0th extension of the output FITS file
if \fIextensions\fR is "yes".
.le
.ls make_image = yes
By default wfits writes the FITS image(s) to the output destination.
If the make_image switch is turned off, wfits prints the FITS headers
on the standard output and no output file is created. In this way the
output FITS headers can be examined before actually writing a FITS tape.
.le
.ls long_header = no
If this switch is set the full FITS header will be printed on the standard
output for each IRAF image converted.
.le
.ls short_header = yes
If this switch is set only a short header, listing the files processed and
their dimensions will be printed on the standard output.
The long_header switch must be turned off.
.le
.ls bitpix = 0
A bitpix of 8, 16, or 32 will produce either an unsigned byte,
twos-complement 16 bit integer, or twos-complement 32 bit integer FITS
image. If bitpix is -32 or
-64 IEEE real or double precision floating point FITS images are produced.
If bitpix is set to 0 (the default), wfits will choose one of 8,
16, 32, -32 or -64 based on the data type of the IRAF image.
For example a short integer and real image will default to bitpix 16 and
-32 respectively.
Users should be wary or overriding the default value of bitpix as loss
of precision in their data may result. In this case wfits will issue a
warning message and an estimate of the maximum loss of precision to be
expected.
.le
.ls blocking_factor = 0
The tape blocking factor for FITS.
Wfits normally writes \fIblocking_factor\fR * 2880 byte records,
where \fIblocking_factor\fR is an integer from 1 to 10.
If \fIblocking_factor\fR = 0, wfits uses the default FITS blocking
factor specified for the device by the "fb" parameter in the
file dev$tapecap, or 1 if the "fb" parameter is not present. For
devices which support variable block sizes, e.g. 9-track tapes, exabytes
and dats, "fb" is normally set to 10.
The user may override this value by setting \fIblocking_factor\fR
>= 1 or <= 10. If the device does not support variable block sizes, e.g.
various types of cartridge drives, blocks of the size defined for the
device by the "bs" parameter in the dev$tapecap file are written
and \fIblocking_factor\fR is ignored.
.le
.ls scale = yes
If the scale switch is set, the IRAF image will be scaled before output.
Two types of scaling are available. The scaling parameters bscale and
bzero may be entered by the user (autoscale = no), or the program can
calculate the appropriate bscale and bzero factors (autoscale = yes).
If the scale switch is turned off, the IRAF image data is converted
directly to integers of the specified bitpix with possible loss of
precision.
.le
.ls autoscale = yes
If the autoscale switch is set, wfits calculates the appropriate bscale and
bzero factors based on the IRAF image data type, and the maximum and minimum
values of the data.
.le
.ih
DESCRIPTION
IRAF data is read from disk and written to the specified destination,
either disk or magnetic tape. The FITS header may optionally be printed
on the standard output as either a full listing or a short description,
with or without creating an output image file. If a the default value
of bitpix (default = 0) is entered, wfits will select the appropriate
bitpix value based on the precision of the IRAF data. Otherwise the
user value is used and loss of precision is possible. Two data scaling
options are available. In autoscale mode wfits calculates the appropriate
scaling factors based on the maximum and minimum data values in the
IRAF image and the FITS bits per pixel. Alternatively the scaling factors
can be entered directly. If no scaling is requested the IRAF data values
will be converted directly to FITS integers or floating point values
with possible loss of precision.
.ih
EXAMPLES
1. Convert a list of IRAF image files to a list of FITS image files on a blank
magnetic tape, allowing wfits to select the appropriate bitpix
and scaling parameters.
.nf
cl> wfits iraf_file* mtb1600[1]
.fi
2. Convert a list of IRAF image files to FITS image files on disk,
allowing wfits to select the appropriate bitpix and scaling parameters.
In the first example below the images specified by the template are written
to files fits001, fits002, etc. In the second the list of input images
specified one per line in the text file imlist are written to the
files specified one per line in the text file fitslist.
.nf
cl> wfits iraf_file* fits
cl> wfits @imlist @fitslist
.fi
3. Convert an IRAF image file to a 32 bits per pixel FITS file with no
scaling and append to a tape already containing data.
.nf
cl> wfits iraf_file mtb1600[EOT] bi=32 sc-
.fi
4. Convert an IRAF image to a 16 bit FITS image on disk, and specify
bscale and bzero explicitly in the process.
.nf
cl> wfits iraf_file fits_file bi=16 au- bs=4.0 bz=0.0
.fi
5. Print the FITS headers on the standard output.
.nf
cl> wfits iraf_file* ma-
.fi
6. Create a disk file called headers containing the FITS headers for a list
of IRAF image files.
.nf
cl> wfits iraf_file* ma- > headers
.fi
7. Write a FITS tape with 14400 bytes per record (5 2880 FITS records per
tape block) on a 9-track tape.
.nf
cl> wfits images* mtb[1] block=5
.fi
8. Write a FITS Exabyte tape with a blocking factor of 1 (1 2880 FITS record
per block). Note that wfits will normally by default write a 28000 (
10 2880 FITS logical records per block) byte record.
.nf
cl> wfits images* mtb[1] block=1
.fi
9. Write a list of images to a single tape file using the FITS standard
extension IMAGE. Users who are planning on reading their data with
local FITS readers should check that those local readers support the
FITS IMAGE extension before selecting this option.
.nf
cl> wfits *.imh mtb[1] block=1 extensions+
.fi
10. Repeat the previous example but do not write a global header.
.nf
cl> wfits *.imh mtb[1] block=1 extensions+ global-
.fi
.ih
BUGS
WFITS does not attempt to recover from write errors. When an error is
detected, WFITS issues an error message and attempts to write a double
EOF at the end of the last good record. In this case the last file on
the tape will be a partial file. IF WFITS is not successful in writing
the double EOF, the message "Cannot close magtape file (name)" will be
issued. Problems occur as some drives permit the double EOF to be
written after the physical end of tape and some do not. Similarly
some drives can read a double EOF after end of tape and some cannot. Depending
on operating system and device driver, an attempt to read or write past
end of tape may or may not be distinguishable from a normal write error.
Blank pixel values are not correctly handled.
Attempting to write at EOT on a blank tape will at best result in numerous
error messages being issued and at worst result in tape runaway depending
on the driver.
.ih
SEE ALSO
rfits, reblock, fits kernel
.endhelp
|