xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
NAME
xv - interactive image display for the X Window System
SYNTAX
xv [options] [filename [filename...]]
NOTE
This man page is merely the text portion of the (PostScript) xv docs,
without the figures, and without the appendicies. As such, it is highly
recommended that you get your hands on the *real* documentation.
DESCRIPTION
xv is an X11 program that displays images in the GIF, JPEG, TIFF, PBM,
PGM, PPM, X11 bitmap, Utah Raster Toolkit RLE, PDS/VICAR, Sun Rasterfile,
and PM formats on 1-, 2-, 4-, 6-, 8-, 16-, 24-, and 32-bit X displays.
xv will also read compress-ed versions of these files.
SECTION 1: OVERVIEW
xv version 2.10 lets you do a large number of things (many of them
actually useful), including, but not limited to, the following:
⊕ display an image in a window on the screen
⊕ display an image on the root window, in a variety of styles
⊕ grab any rectangular portion of the screen and turn it into an image
⊕ arbitrarily stretch or compress the image
⊕ rotate the image in 90-degree steps
⊕ flip the image around the horizontal or vertical axes
⊕ crop a rectangular portion of the image
⊕ magnify any portion of the image by any amount, up to the size of the
screen
⊕ determine pixel values and x,y coordinates in the image
⊕ adjust image brightness and contrast with a gamma correction function
⊕ apply different gamma functions to the Red, Green, and Blue color
components, to correct for non-linear color response
⊕ adjust global image saturation
⊕ perform global hue remapping
10/89 Page 1
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
⊕ perform histogram equalization
⊕ edit an image's colormap
⊕ reduce the number of colors in an image
⊕ dither in color and b&w
⊕ smooth an image
⊕ crop off solid borders automatically
⊕ convert image formats
⊕ generate Encapsulated PostScript
Unfortunately the Automatic Checkbook Balancing Module still isn't
completely debugged, and is not included in this distribution.
SECTION 2: STARTING XV
Note: unless explicitly stated otherwise, the term click means "click
with the Left mouse button."
Start the program up by typing 'xv'. After a short delay, a window will
appear with the default image (the xv logo, credits and revision date)
displayed in it. If you change the size of the window (using whatever
method your window manager provides), the image will be automatically
stretched to fit the window.
Section 2.1: Displaying Pixel Values
Clicking (and optionally dragging) the Left mouse button inside this
window will display pixel information in the following format:
196, 137 = 191,121,209 (287 42 81 HSV)
The first pair of numbers (196,137) are the x and y positions of the
cursor, in image coordinates. These numbers remain the same regardless
of any image resizing, or cropping. For example, if you click on the eye
of the fish on the right side of the default image, you'll get
(approximately) 251,129 regardless of the size of the displayed image.
This allows you to zoom in for precise measurements.
The first triplet of numbers (191,121,209) are the RGB values of the
selected pixel. The components will have integer values in the range 0-
255. The values displayed are prior to any HSV/RGB modification, but
after any colormap changes. See "Section 5: The Color Editor" for
details.
The second triplet of numbers (287 42 81) are the HSV values of the
selected pixel. The first component will have integer values in the
Page 2 10/89
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
range 0-359, and the second and third components will have integer values
in the range 0-100. The values displayed are prior to any HSV/RGB
modification, but after any colormap changes. See "Section 5: The Color
Editor" for details. Also, see "Appendix D: RGB and HSV Colorspaces" for
more information about what these numbers mean.
Note: If you actually want to measure some pixels, it will probably help
to crop to a small region of your image, and expand that region to the
point where you can see the individual pixels.
This string is automatically copied to your X server's cut buffer
whenever you measure pixel values. This lets you easily feed this
information to another program, useful if you're doing manual feature
extraction, or something. Try it: measure a pixel's value, and then go
click your Middle mouse button in an xterm window.
Section 2.2: Cropping
Bring up the xv controls window by typing the '?' key or clicking the
Right mouse button inside the image window.
Clicking and dragging the Middle button of the mouse inside the image
window will allow you to draw a cropping rectangle on the image. If
you're unhappy with the one you've drawn, simply click the Middle button
and draw another. If you'd like the rectangle to go away altogether,
click the Middle button and release it without moving the mouse.
You can determine how large the cropping rectangle is (in image
coordinates) by bringing up the xv info window. Do this by clicking the
Info button in the xv controls window or by typing the 'i' key into any
open xv window.
The xv info window will display, among other things, the current size and
position of the cropping rectangle in terms of image coordinates. For
example, if it says:
114x77 rectangle starting at 119,58
that would mean that the current cropping rectangle is 114 image pixels
wide, 77 image pixels high, and its top-left corner is located 119 image
pixels in from the left edge of the image, and 58 image pixels in from
the top edge. These values will be updated as you drag the cropping
rectangle around.
If you want to set the size or position of the cropping rectangle
precisely, you can use the arrow keys on your keyboard. First, make the
xv info window visible as described above (if it's not already visible).
Second, use the mouse to draw a rough approximation of the cropping
rectangle that you want. You can now use the arrow keys to move the
cropping rectangle around the image. Once you've gotten the top and left
sides of the cropping rectangle precisely where you want them, you can
move the bottom-right corner of the cropping rectangle (only) by holding
10/89 Page 3
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
the <shift> key down while using the arrow keys. Pressing the up arrow
will make the rectangle shorter, and pressing the down arrow will make
the rectangle taller.
Once you have a cropping rectangle that you can live with, you can
proceed with the actual cropping operation. Click the Crop button in the
xv controls window, or type the 'c' key in any open xv window. The image
window will shrink to show only portions of the image that were inside
the cropping rectangle.
Note: if you are running a window manager such as mwm, which decorates
windows with a title bar, resizing regions, and such, it is quite
possible that the aspect ratio of the cropped image will get screwed up.
This is because certain window managers enforce a minimum window size.
If you try to crop to a rectangle that is too small, the window manager
will create the smallest window it can, and the image will be stretched
to fit this window. If this happens, you can press the Aspect button in
the xv controls window, or type the 'a' key in any open xv window. This
will expand the image so that it has the correct aspect ratio again.
You can crop a cropped image by repeating the same steps (drawing a new
cropping rectangle and issuing the Crop command), ad infinitum.
You can return to the original, uncropped image by using the UnCrop
command. Simply click the UnCrop button or type the 'u' key in any open
xv window. Note that using the UnCrop command will turn off image
smoothing (the Smooth) command), due to the potentially long time it can
take to generate a large, smoothed image.
Note that if you try to make the cropping rectangle too small in either
width or height (under 5 screen pixels), it'll just turn itself off. If
you want to crop a very small portion of an image, you'll probably have
to do it in two passes. First, crop to a small (but large enough to
still be enabled) rectangle, expand that region, then crop again.
Section 2.3: Zooming
You can zoom in by a factor of two (or four, eight, etc.) on any
rectangular region of the image by holding down the <ctrl> key on your
keyboard and clicking the Left mouse button in the image. A rectangle
will flash, centered around the cursor position, and the region inside
that rectangle will be doubled in size. The image window should remain
the same size, and in the same position. You can repeat this operation
to zoom in by a factor of four, or eight, or whatever, as many times as
you wish.
You can zoom out by a factor of two (if possible) by holding down the
<ctrl> key on your keyboard and clicking the Right mouse button inside
the image window. You can not zoom out beyond the point where the entire
image fill the window.
Page 4 10/89
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
SECTION 3: THE CONTROL WINDOW
The xv controls window is the central point of control for the program,
hence the name. It provides controls to resize the current image, flip
and rotate it, load and save different files, and bring up the other xv
windows. It can be brought up by clicking the Right mouse button in the
image window, or by typing the '?' key inside any open xv window. Doing
either of these things while the xv controls window is visible will hide
it.
All of the following commands may be executed by either clicking the
appropriate command button, or typing the keyboard equivalent (where
given) into any open xv window.
Section 3.1: Resizing Commands
Note that none of the 'resizing' commands modify the image in any way.
They only affect how the image is displayed. The image remains at its
original size. This allows you to arbitrarily stretch and compact the
image without compounding error caused by earlier resizing. In each
case, the displayed image is recomputed from the original internal image.
Normal (Keyboard equivalent 'n')
Attempts to return the image to its normal size, where one image
pixel maps to one screen pixel. For example, if the image (or the
current cropped portion of the image) has a size of 320x200, this
command will attempt to make the image window be 320 screen pixels
wide by 200 screen pixels high.
This command may fail in two cases. If you're running a window
manager (such as mwm) that enforces a minimum window size, and the
'normal' size is too small, the image may get distorted. See the
note in "Section 2.2: Cropping" for more information.
Also, if the image is larger than the size of your screen, it will
be 'halved' until it fits on the screen. For example, if you try
to display a 1400x900 image on a 1280x1024 screen, the Normal
command will display a 700x450 image.
Max Size (Keyboard equivalent 'm')
This command will make the displayed image be the same size as the
screen. If you are running a window manager that puts up a
titlebar, you'll find that the titlebar is now off the top of the
screen. To get the titlebar back, simply shrink the image to
anything smaller than the size of the screen. The window will be
moved so that the titlebar is once again visible.
Maxpect (Keyboard equivalent 'M')
Makes the image as large as possible, while preserving the aspect
ratio. This avoids the generally unwanted image distortion that
10/89 Page 5
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
Max Size is capable of generating. For example, if you have a
320x200 image, and an 1280x1024 screen, doing the Maxpect command
will result in an image that is 1280x800. Max Size, on the other
hand, would've generated an image of size 1280x1024, which would be
appear 'stretched' vertically.
Dbl Size (Keyboard equivalent '>')
Doubles the current size of the image, with the constraint that
neither axis is allowed to be larger than the screen. For example,
given a 320x200 image and a 1280x1024 screen, the image can be
doubled once (to 640x400), a second time (to 1280x800), but a third
time would make the image 1280x1024. You'll note that on the third
time, the width didn't change at all, since it was already at its
maximum value. Also note that the height wasn't allowed to double
(from 800 to 1600), but was truncated at its maximum value (1024).
Half Size (Keyboard equivalent '<')
Halves the current size of the image, with the constraint that
neither axis is allowed to have a size less than 1 pixel. Also,
you may run into 'minimum size' problems with your window manager.
See the note in "Section 2.2: Cropping" for more information.
Note that the window size is maintained as a pair of integers. As
a result you may see some integer round-off problems. For example,
if you halve a 265x185 image, you'll get a 132x92 image, which is
just fine. However, if you Dbl Size this image, you'll get a
264x184 image, not the 265x185 image you might have expected.
+10% (Keyboard equivalent '.')
Increases the current size of the image by 10%, subject to the
constraint that the image cannot be made larger than the screen
size (in either axis). For example, issuing this command on a
320x200 image will result in a 352x220 image.
-10% (Keyboard equivalent ',')
Decreases the current size of the image by 10%. Neither axis of
the image is allowed to shrink below 1 pixel. Also, you run the
risk of running into 'minimum window size' problems with your
window manager.
It should be noted that the +10% and -10% commands have no concept
of an 'original size'. They simply increase or decrease the
current image size by 10%. As a result, they do not undo each
other. For example, take a 320x200 image. Do a +10% and the image
will be 352x220. If you issue the -10% command now, the image will
be made (352 - 35.2)x(220 - 22), or 316x198.
Page 6 10/89
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
4x3 (Keyboard equivalent '4')
Attempts to resize the image so that the ratio of width to height
is equal to 4 to 3. (e.g., 320x240, 400x300, etc.) This is useful
because many images were meant to fill the screen of whatever
system they were generated on, and nearly all video tubes have an
aspect ratio of 4:3. This command will stretch the image so that
things will probably look right on your X display (nearly all of
which, thankfully, have square pixels). This command is
particularly useful for images which have really bizarre sizes
(such as the 600x200 images presumably meant for CGA, and the
640x350 16-color EGA images).
Aspect (Keyboard equivalent 'a')
Applies the 'default aspect ratio' to the image. This is done
automatically when the image is first loaded. Normally, the
default aspect ratio is '1:1', but certain GIF files may have an
aspect ratio encoded in them. You can also set the default aspect
ratio via a command-line argument or an X resource. See 'Section
9: Modifying XV Behavior' for more info. The idea behind this
command is that you'd stretch the image manually (via your window
manager) to roughly the size you'd like, and then use the Aspect
command to fix up the proportions.
Normally Aspect expands one axis of the image to correct the aspect
ratio. If this would result in an image that is larger than the
screen, the Aspect command will instead shrink one of the axes to
correct the aspect ratio.
Section 3.2: Rotate/Flip Commands
Turn CW (Keyboard equivalent 't')
Rotates the image 90 degrees clockwise.
Turn CCW (Keyboard equivalent 'T')
Rotates the image 90 degrees counter-clockwise.
Flip H (Keyboard equivalent 'h')
Flips the image horizontally (around the vertical center-line of
the image).
Flip V (Keyboard equivalent 'v')
Flips the image vertically (around the horizontal center-line of
the image).
Section 3.3: Smoothing Commands
10/89 Page 7
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
Raw (Keyboard equivalent 'r')
Returns the displayed image to its 'raw' state (where each pixel in
the displayed image is as close as possible to the corresponding
pixel in the internal image). In short, it turns off any dithering
or smoothing. When dithering or smoothing haven't been done, this
command is disabled.
Dither (Keyboard equivalent 'd')
Regenerates the displayed image by dithering with the available
colors in an attempt to approximate the original image. This is
only relevant if the color allocation code failed to get all the
colors it wanted. If it did get all the desired colors, the Dither
command will just generate the same display image as the Raw
command. On the other hand, if you didn't get all the desired
colors, the Dither command will try to approximate the missing
colors by dithering with the colors that were obtained. If you're
running xv on a 1-bit display the Dither command will be disabled,
as the image will always be dithered for display.
Smooth (Keyboard equivalent 's')
Smooths out distortion caused by integer round-off when an image is
expanded or shrunk. This is generally a desirable effect, however
it is fairly time-consuming on large images on most current
workstations. As such, by default, it is not done automatically.
See "Section 9: Modifying XV Behavior" for more details.
Section 3.4: Cropping Commands
Crop (Keyboard equivalent 'c')
Crops the image to the current cropping rectangle. This command is
only available when a cropping rectangle has been drawn on the
image. See "Section 2.2: Cropping" for further information.
UnCrop (Keyboard equivalent 'u')
Returns the image to its normal, uncropped state. This command is
only available after the image has been cropped. See "Section 2.2:
Cropping" for further information.
AutoCrop (Keyboard equivalent 'A')
Crops off any constant borders that exist in the image. It will
crop to the smallest rectangle that encloses the 'interesting'
section of the image. It may not always appear to work because of
minor invisible color changes in the image. As such, it works best
on computer-generated images, and not as well on scanned images.
Page 8 10/89
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
Section 3.5: The Display Modes Menu
In addition to displaying an image in a window, xv can also display
images on the root (background) window of your X display. There are a
variety of ways that xv can display an image on the root window. The
Display Modes popup menu lets you select where (and how) xv will display
the image.
Click on the Display Modes button in the xv controls window, and hold the
mouse button down. This will cause the Display Modes menu to pop up.
The current display mode will be shown with a check mark next to it. To
select a new mode, drag the mouse down to the desired mode, and release
the mouse button.
It is not possible for xv to receive button presses or keyboard presses
in the root window. As such, there are several functions that cannot be
used while in a 'root' mode, such as pixel tracking and image cropping.
If you want to do such things, you'll have to temporarily return to
'window' mode, and return to 'root' mode when you're finished. Also,
when you are in a 'root' mode, you will not be able to get rid of the xv
controls window. At best you can iconify it (using your window manager).
(The reason for this is that if you ever got rid of it there'd be no way
to get it back.)
Window
Displays the image in a window. If you were previously in a 'root'
mode, the root window will also be cleared.
Root: Tiled
The image is displayed in the root window. One image is displayed
aligned with the top-left corner of the screen. The image is then
duplicated towards the bottom and right edges of the screen, as
many times as necessary to fill the screen.
Root: Integer Tiled
Similar to Root: Tiled, except that the image is first shrunk so
that its width and height are integer divisors of the screen's
width and height. This keeps the images along the bottom and right
edges of the screen from being Normal, Dbl Size, etc.) will lose
the 'integer'-ness of the image.
Root: Mirrored
Tiles the original image with versions that have been horizontally
flipped, vertically flipped, and both horizontally and vertically
flipped. This gets rid of the sharp dividing lines where tiled
images meet. The effect is quite interesting.
10/89 Page 9
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
Root: Integer Mirrored
Like Root: Mirrored, but also does the integer-ization described
under the Root: Integer Tiled entry.
Root: Center Tiled
Like Root: Tiled, but it positions the images so that one of them
is centered on the screen, and the rest are tiled off in all
directions. Visually pleasing without the image size distortion
associated with Root: Integer Tiled.
Root: Centered
Displays a single image centered in the root window, surrounded by
black.
Root: Centered, Warp
Displays a single image centered in the root window, surrounded by
a black and white 'warp' pattern, which produces some mildly
visually pleasing Moire effects.
Root: Centered, Brick
Displays a single image centered in the root window, surrounded by
a black and white 'brick' pattern.
Note: The three 'centered' modes (Root: Centered, Root: Centered, Warp,
and Root: Centered, Brick, but not Root: Center Tiled) require the
creation of a Pixmap the size of the screen. This can be a fairly large
request for resources, and will fail on a color X terminal with
insufficient memory. They can also require the transmission of
considerably more data than the other 'root' modes. If you're on a
brain-damaged X terminal hanging off a slow network, you should probably
go somewhere else. Barring that, you should certainly avoid the
'centered' modes.
Also note: If you quit xv while displaying an image on the root window,
the image will remain in the root window, and the colors used by the
image will remain allocated. This is generally regarded as correct
behavior. If you decide you want to get rid of the root image to free up
resources, or simply because you're sick of seeing it, the quickest route
is to use run 'xv -clear', which will clear the root window, release any
allocated colors, and exit. Alternately, xsetroot and any other X
program that puts things in the root window should be able to do the
trick as well.
Section 3.6: The 24-bit Conversion Menu
xv can currently only operate on 8-bit images. Whenever you load a 24-
bit image (such as JPEG, TIFF, PPM, etc) it is immediately converted into
Page 10 10/89
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
an 8-bit colormapped image using one of three algorithms.
Fast Converts 24-bit images into 8-bit images by dithering with a fixed
6x6x6 RGB colormap. It is the quickest of the three algorithms,
but also generally produces the worst images. It can also be
selected via the '-quick24' command-line option or X resource.
Slow The default algorithm. Takes about twice as long as the fast
algorithm. Uses the median-cut algorithm to pick a set of 256
colors, and then dithers using these colors. It can be selected
via the '-slow24' command-line option or X resource.
Best By far and away the slowest of the algorithms. It can take up to
ten times as long as the 'slow' algorithm. It uses a cleverer
version of the median-cut algorithm to pick a better set of colors
than the slow algorithm. It does not dither. This might look best
if you're going to be expanding the image by very much, as the
dithering in the other two algorithms becomes very noticable. You
can also select this option via the '-best24' command-line option
or X resource.
Section 3.7: Working With Multiple Files
xv provides a a set of controls that let you conveniently operate on a
list of images. To use the following commands, you'll have to start up
xv with a list of filenames. For example, you could type 'xv *.gif'
(assuming, of course, that you have a bunch of files that end with the
suffix '.gif' in the current directory).
The filenames are listed in a scrollable window. The current selection
is shown in reverse video. If there are more names than will fit in the
window, the scrollbar will be enabled.
Section 3.7.1: Operating a List Window
The scrollbar operates as follows:
⊕ clicking in the top or bottom arrow of the scrollbar scrolls the list
by one line in the appropriate direction. It will continue to scroll
the list as long as you hold the mouse down.
⊕ The thumb (the small white rectangle in the middle of the scrollbar)
shows roughly where in the list you are. You can change your position
in the list by clicking and dragging the thumb to another position in
the scrollbar. The list will scroll around as you move the thumb.
⊕ You can scroll the list up or down a page at a time by clicking in the
grey region between the thumb and the top or bottom arrows.
10/89 Page 11
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
If you click on a name in the list, that name will become highlighted.
You can drag the highlight bar up and down, and the list will scroll
appropriately.
It is also possible to control the list window from the keyboard. In all
cases, you must make sure that the window sees the keypress. Generally,
this means you have to have the cursor inside the window, though your
window manager may also require you to click inside the window first.
⊕ The up and down arrow keys move the highlight bar up and down. If the
bar is at the top or bottom of the window, the list will scroll one
line.
⊕ The page up and page down keys scroll the list up or down a page at a
time.
⊕ Pressing the home key will jump to the beginning of the list.
Pressing the end key will jump to the bottom of the list.
Section 3.7.2: The File Commands
You can directly view any image in the list by double-clicking on its
filename. If xv is unable to load the file (for any of a variety of
reasons), it'll display an error message and put up the default image,
the xv logo.
Next (Keyboard equivalent <space>)
Attempts to load the next file in the list. If it is unable to
load the next file, it will continue down the list until it
successfully loads a file. If it gets to the bottom of the list
without successfully loading a file, it will put up the default
image.
Previous (Keyboard equivalent <backspace>)
Attempts to load the previous file in the list. If it is unable to
load the previous file, it will continue up the list until it
successfully loads a file. If it gets to the top of the list
without successfully loading a file, it will put up the default
image.
Delete (Keyboard equivalent <ctrl-D>)
This command lets delete the currently selected file from the list
(and optionally delete the associated disk file). Note that the
currently selected file is the one with the highlight bar on it.
While this is generally the same as the currently displayed image,
it doesn't have to be.
The Delete command will pop-up a window asking you what you want to
delete. Your choices are:
Page 12 10/89
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
⊕ List Entry, which will remove the highlighted name from the
list. (Keyboard equivalent: the enter key)
⊕ Disk File, which will remove the highlighted name from the list
and also delete the associated disk file. This removes unwanted
images, just like manually typing 'rm <filename>' in another
window. (Keyboard equivalent: <ctrl-D>)
⊕ Cancel, which lets you get out of the Delete command without
actually deleting anything. (Keyboard equivalent: the esc key)
Section 3.7.3: Image Reloading
It is occasionally desirable to reload an image file because the contents
of the file have changed. For example, you could be downloading a file,
and you might want to keep reloading the file to check on the progress of
the download. Or perhaps you have a program that generates images, and
you'd like to view these images without any manual intervention.
XV provides a way to reload an image via an external signal. If you send
the xv process a SIGQUIT signal ('kill -QUIT pid', or currently selected
file. (The one that is currently highlighted in the xv controls window
filename list.) This behavior is exactly the same as hitting '<return>'
in the xv controls window. If xv is currently in a state where hitting
'<return>' in the controls window won't load an image (ie, some pop-up
dialog box is grabbing all such events), then sending this signal wonUt
work either.
An idea: You could write a 'clock' program that, once a minute, generates
a really spiffy looking picture of the current time (with color
gradations, 3d extruded numbers, whatever), then sends xv the signal to
reload the generated image.
Section 3.8: The Grab Command
The Grab command works as follows: click on the Grab button in the xv
controls window, or type a <ctrl-G> key in any active xv window (except
for the JPEG, PostScript, and TIFF 'save' dialog boxes). The terminal
will beep once, and the cursor will change to a cross. The screen will
remain frozen until you complete the Grab command.
You can Grab an arbitrary region of the screen by clicking the Left mouse
button and dragging a rectangle in exactly the same way you draw a
cropping rectangle. When you let go of the mouse button, the contents of
this rectangle will be read from the screen and loaded into xv.
Alternately, you can grab the entire contents of a window (including its
frame) by clicking the Middle mouse button anywhere inside the chosen
window. If you click the Middle mouse button somewhere on the root
window, the entire screen will be loaded into xv.
Or, alternately, you can simply abort the Grab command by clicking the
Right mouse button anywhere on the screen.
10/89 Page 13
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
You can use the Grab command for a wide variety of purposes. For
example, you can use it to print the contents of any window (or the whole
screen) by grabbing the appropriate image and then saving it as a
PostScript file.
You can use the Grab command, in conjunction with the Zoom and UnZoom
commands, as an effective replacement for the xmag program.
You can also use the Grab command to pick 'just the right colors' for any
application. Simply start the application in question, Grab the window
into xv, and use the colormap editor to twiddle the colors around to your
heart's content.
Note: The Grab command does not work on Macintoshes running MacX in a
rootless mode, which isn't too surprising, if you think about it...
Section 3.9: Other Commands
Info (Keyboard equivalent 'i')
Opens and closes the xv info window. See "Section 4: The Info
Window" for more details.
ColEdit (Keyboard equivalent 'e')
Opens and closes the xv color editor window. See "Section 5: The
Color Editor" for more details.
Load (Keyboard equivalent <ctrl-L>)
Opens the xv load window. See "Section 6: The Load Window" for
more details.
Save (Keyboard equivalent <ctrl-S>)
Opens the xv save window. See "Section 7: The Save Window" for
more details.
Quit (Keyboard equivalent 'q')
Quits out of the program.
SECTION 4: THE INFO WINDOW
Section 4.1: Overview
xv provides a window to display information about the current image,
color allocation, expansion, cropping, and any error messages. This
window can be opened by issuing the Info command. (Click on the Info
button in the xv controls window, or type 'i' in any open xv window.)
You can close the window by using the Info command while the window is
open. You can also close the window by clicking anywhere inside it.
The top portion of the window displays the program name, revision date,
Page 14 10/89
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
and patchlevel. It also shows the University of Pennsylvania shield, the
GRASP Lab logo, the copyright notice, and of course, the author's name.
Section 4.2: The Fields
The "Filename" field displays the name of the currently loaded file. The
name is displayed without any leading pathname. If there is no currently
loaded image (you're looking at the default image) this field will
display "<none>".
The "Format" field displays information describing what image format the
file is stored in, and how large the file is (in bytes).
The "Resolution" field shows the width and height (in image pixels) of
the loaded image. Note that this does not necessarily have anything to
do with the size of the image currently displayed on your screen. These
numbers do not change as you modify the display image.
The "Cropping" field displays the current state of any cropping activity.
If you are looking at the entire (uncropped) image, and there is no
cropping rectangle drawn, this field will show "<none>". If you draw a
cropping rectangle, or if you are viewing cropped portion of image, this
field will display something like "247x128 rectangle starting at
132,421". See "Section 2.2: Cropping" for more details.
The "Expansion" field gives you information about how the image is
displayed. It will display something like "1.58 x 1.37 (505 x 273)".
This tells you that the current displayed image is 505 pixels wide and
273 pixels high, and that it is 1.58 times wider and 1.37 times higher
than the internal image (which, in this case, had a size of 320x200).
The "Colors" field gives you detailed information on how well (or poorly)
color allocation went. If everything went reasonably well it will
display something like:
Got all 67 desired colors. (66 unique)
This means that 67 entries in the image's colormap were used in the
image, but that only 66 of these colors were different, as far as the X
server was concerned.
See "Appendix E: Color Allocation" for a complete discussion of how
colors are allocated, and what the "Colors" field can tell you.
Note that the fields are filled in as information becomes available. As
such, they can be used as a rough 'progress indicator' when loading
images. When you begin loading, all the fields are cleared. Once the
image has been successfully loaded, the top three fields (Filename,
Format, Resolution) are filled in. Once the colors have been allocated,
and the display image generated, the bottom three fields are shown
(Cropping, Expansion, and Colors).
10/89 Page 15
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
Section 4.3: Status Lines
The bottom two lines in the info window display various error messages,
warnings, and status information. These two lines are also duplicated in
the xv controls window.
The upper line is the more commonly used. It normally displays a one-
line summary of the current image and color allocation success. If an
error occurs, it will be displayed on this line as well.
The lower line is used to display warning messages.
SECTION 5: THE COLOR EDITOR
Section 5.1: Overview
The xv color editor provides a powerful system for manipulating color
images. Since there are many different reasons why a person would want
to modify an image's colors, and many different types of images that may
need modification, there is no one color manipulation tool that would be
'best' for all purposes. Because of this problem, xv gives the user
three different color tools, all of which can be used simultaneously.
⊕ Colormap Editing: This tool lets you arbitrarily modify individual
colormap entries. Useful for modifying the color of captions or other
things that have been added to images. Also works well on images that
have a small number of colors, such as images generated by 'drawing'
or CAD programs. It's also an easy way to spiff up boring 1-bit black
and white images.
⊕ HSV Modification: This tool lets you alter the image globally in the
HSV colorspace. (See "Appendix D: RGB and HSV Colorspaces" for more
info.) Here are examples of the sort of things you can do with this
tool:
⊕ turn all the blues in an image into reds
⊕ change the tint of an image
⊕ change a greyscale image into a mauve-scale image
⊕ increase or decrease the amount of color saturation in an image
⊕ change the overall brightness of an image
⊕ change the overall contrast of an image
⊕ RGB Modification: This tool lets you route the red, green, and blue
color components of an image through independent mapping functions.
The functions can either be the standard gamma function, or any
arbitrary function that can be drawn with straight line segments or a
cubic spline. See "Section 5.3.4: The Intensity Graph" for more info
about graph functions.
Page 16 10/89
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
The major use of the RGB Modification tool is to correct for the
differing color response curves of various color monitors, printers, and
scanners. This is the tool to use when "the image is too red", for
instance.
These three tools are tied together in a fixed order. The Colormap
Editing tool operates on the original colors in the image. The output of
this tool is piped into the HSV Modification tool. Its output is piped
into the RGB Modification tool. The output from the RGB Modification
tool is what actually gets displayed.
In addition there is a collection of buttons that control the xv color
editor as a whole (more or less).
Don't Panic! It's not as complicated as it looks.
Section 5.2: The Colormap Editing Tool
The top portion of this window shows the colormap of the current image.
There are 16 cells across, and up to 16 rows down, for a maximum of 256
color cells. Only cells actually used somewhere in the image are shown
in this array.
The currently selected color cell is shown with a thick border. You can
change the selection by clicking anywhere in the array. If you drag the
mouse through this area, you'll see the dials at the bottom change to
track the current pixel values.
You can also select a color cell by clicking anywhere in the image
window. Whichever pixel value you were on when you let go of the mouse
will become the new selected color cell.
You can define a smoothly gradated range of colors by Left clicking on
the color cell that marks the 'start' of the range, and Middle clicking
on the color cell that marks the 'end' of the range. Intervening color
cells will be interpolated between the colors of the
As an example:
⊕ View the 'default' image by running xv without specifying any
filenames.
⊕ Open the xv color editor window, and Left click on the first color
cell.
⊕ Turn this color cell red by setting the RGB dials to 255,0,0.
⊕ Left click on the 64th color cell (the rightmost color cell in the
last complete row).
10/89 Page 17
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
⊕ Turn this color cell yellow by setting the RGB dials to 255,255,0.
⊕ Middle click on the first color cell. A smooth series of yellowish-
reds will be generated from the 64th color cell to the first color
cell. Note that the 'direction' doesn't matter.
Since certain images will have many colors that are the same, or nearly
the same, it is sometimes convenient to group color cells together.
Grouped color cells all take on the same color, and changing any one of
them affects all of the other colors in the group.
To group color cells together, do the following:
⊕ Hold down the <shift> key.
⊕ Left click on one color cell that you would like to be in the group
⊕ Right click on other color cells that you wish to be in this group.
(Right clicking on cells that are already selected will de-select
them.)
⊕ Release the <shift> key when you're done.
You can create as many groups as you like.
You can use this grouping/ungrouping technique to copy colors from one
color cell to another. Left click on the source color cell, Right click
on the destination color cell, and Right click on the destination color
cell again (to ungroup it).
Section 5.2.1: Using the Dial Controls
At the bottom the Colormap Editing tool are three dials that let you set
the color of the current color cell (or group of cells). By default, the
dials control the Red, Green, and Blue components of the RGB colorspace,
but they can also control the Hue, Saturation, and Value components of
the HSV colorspace. (The RGB/HSV button controls this.)
Regardless of what they control, all dials in xv work the same way.
Clicking on the single arrows increase/decrease the value by 1. Clicking
on the double arrows increase/decrease the value by a larger amount (16
in this case). If you click on one of the arrows, and hold the mouse
button down, the increase/decrease will repeat until you release the
mouse button.
You can also click in the general area of the pointer and simply drag it
to the position you want. The further your mouse cursor is from the
center of the dial, the more precise the control will be. While
dragging, you do not have to keep the cursor inside the dial window.
Page 18 10/89
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
Section 5.2.2: Colormap Editing Commands
ColUndo
Undoes the last change made to the colormap that resulted in a
color cell changing value. This includes grouping and ungrouping
color cells, and changing any of the dials.
Revert
Undoes all color changes. Returns the colormap to its original
state. Destroys any groups that you may have created.
RGB/HSV
Toggles the Colormap Editing dials between editing colors in terms
of Red, Green, and Blue, and editing colors in terms of Hue,
Saturation, and Value.
Grey Turns color images into greyscale images by changing the colormap.
This replaces each color cell with a greyscale representation of
itself. Use the Revert command to restore the colors.
RevVid
This command behaves differently, depending on the setting of the
RGB/HSV mode. (You can tell which mode you're in by the titles on
the dials.)
In RGB mode, each color component is separately 'inverted'. For
example, Yellow (which is composed of full red, full green, and no
blue) would turn to Blue (no red, no green, full blue).
In HSV mode, only the Value (intensity) component is 'inverted'.
The Hue and Saturation components remain the same. In this mode,
bright colors turn to dark versions of the same color. For
example, a Yellow would turn Brown.
Random
Generates a random colormap. This is of questionable usefulness,
but it will occasionally come up with pleasing color combinations
that you never would've come up with yourself. So it stays in. It
works best on images with a small number of colors. Note that it
respects cell groupings, so if your image has a lot of colors, you
can create a few large groups and then use the Random command.
Note: It is HIGHLY RECOMMENDED that if you're using the Colormap Editing
tool, you do NOT use the HSV Modification tool or the RGB Modification
tool as well. If you do, the results can be quite confusing. For
example, you might edit a color cell, and set its color values to produce
a purple. However, because of HSV/RGB Modification further down the
10/89 Page 19
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
line, the actual color displayed on the image (and in the color cell) is
yellow. Very confusing, indeed.
Section 5.3: The HSV Modification Tool
There are four separate controls in the HSV Modification tool. At the
top of the window are a pair of circular controls that handle hue
remapping. Lower down is a circular control that maps 'white' (and
greys) to a specified color. There is a dial control that lets you
saturate/desaturate the colors of the current information. Finally, at
the bottom there is a graph window that lets you modify intensity values
via an arbitrary remapping function.
Section 5.3.1: Hue Remapping Controls
These two dials are used to define a source and a destination range of
hue values. Every hue in the source range (defined in the From dial)
gets mapped to the value of the corresponding point in the destination
range (defined in the To dial).
Each dial has a pair of radial lines with handles at their ends. Between
the two lines an arc is drawn with an arrow at one end. The wedge drawn
by these lines and the arc defines a range of values (in degrees). The
direction of the arc (clockwise, or counter-clockwise) determines the
direction of this range of values (increasing or decreasing).
Distributed around the dial are tick marks and the letters 'R', 'Y', 'G',
'C', 'B', and 'M'. These letters stand for the colors Red, Yellow,
Green, Cyan, Blue, and Magenta, and they show where these colors appear
on the circle.
The range is shown numerically below the control. By default the range
is '330, 30 CW'. This means that a range of values [330, 331, 332, ...
359, 0, 1, ... 28, 29, 30] has been defined. Note that (being a circle)
it wraps back to 0 after 359.
The range can be changed in many different ways. You can click on the
'handles' at the end of the radial lines and move them around. If you
click inside the dial, but not on one of the handles, you'll be able to
drag the range around as a single object. There are also 5 buttons below
the dial that let you rotate the range, flip the direction of the range,
and increase/decrease the size of the range while keeping it centered
around the same value.
In its default state, the To dial is set to the same range as the From
dial. When the two dials are set to the same range, they are effectively
'turned off', and ignored.
An example of hue remapping:
Page 20 10/89
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
⊕ As a simple example of the sort of things you can do with the hue
remapping control, we'll change the background color of the default
(xv logo) image without changing any other colors in the image. Since
the background is composed of a gradient of 64 colors, you would not
want to do this with the Colormap Editing tool. It would take
forever.
⊕ First, get the default image up on the screen by running 'xv' without
giving any filenames. Open up the xv color editor window via the
ColEdit command.
⊕ Next, click the mouse in the image window and drag it around. You'll
see that all the background pixels have the same Hue component value
(240).
⊕ To remap this hue, simply adjust the From dial so that its range
includes this Hue value. The background should change from 'blue' to
a reddish color, assuming the To dial is still set to its default
range (centered around 'R'). If more than the background changed
color, you can shrink the From range so that it covers fewer colors.
In fact, it's possible to shrink the range to the point where it only
covers only a single value.
Note that the values printed when you are tracking pixel values in the
image are the values before the HSV Modification tool is applied. For
example, the background of the default image will still claim to be blue,
regardless of what color you may have changed it to. This is so that you
know what Hue value you will need to remap if you want to change its
color again.
If you press the Reset button that is located near the hue remapping
controls, it will effectively disable the hue remapping by setting the To
range equal to the From range.
Below the hue remapping controls are a group of 'radio buttons'. You can
have up to six different hue remappings happening simultaneously. Higher
numbered mappings take precedence over lower number mappings.
An example of multiple hue remappings:
⊕ Draw a From range that is a complete circle. The easiest way to do
this is to draw a range that is nearly a full circle, then click and
hold down the 'increase range' button located below the From range
dial until the range stops getting bigger.
⊕ Copy this range to the To range by pressing the Reset button.
⊕ Rotate the To range slightly, by either clicking and dragging anywhere
in the To range dial, or by using the 'rotate clockwise' and 'rotate
counter-clockwise' buttons located below the To range.
10/89 Page 21
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
⊕ You've just built yourself what is effectively a tint control.
⊕ Now, suppose, you'd like to adjust the background color of your
(tint-modified) image, without affecting anything else. Clicking on
the background in the image window reveals that the background still
has an (original) hue of 240. To modify this hue without affecting
anything else, we'll need a second hue remapping.
⊕ Click on the 2 radio button. The dials will change to some other
default setting. As before, set the From range to encompass the value
240, preferably as 'tightly' as possible, and set the To range to
produce the desired background color.
Note that the six hue remappings are not 'cascaded'. The output of one
remapping is not fed as input into any of the other hue remappings. The
hue remappings always operate on the hue values in the original image.
In this example, if remapping #1 adds 32 to all hue values, thereby
mapping the blue background (value 240) into a purple-blue (value 272),
remapping #2 still sees the background at 240, and can remap it to
anything it likes. Similarly, in the same example, if remapping #1 has
mapped a green-blue color (value 208) into blue (value 240), remapping #2
will not map this into another color. As far as remapping #2 is
concerned, that green-blue is still green-blue.
If it seems complicated, I'm sorry. It is.
Section 5.3.2: The White Remapping Control
In the HSV colorspace, 'white' (including black, and all the greys in
between) has no Hue or Saturation components. As such, it is not
possible to use the hue remapping controls to change the color of white
pixels in the image, since they have no 'color' to change.
The white remapping control provides a way to add Hue and Saturation
components to all the whites in the image. It consists of a movable
point in a color dial. The angle of the dot from the center of the dial
determines the Hue component. The distance of the dot from the center of
the dial determines the Saturation component. The further the dot is
from the center of the dial, the more saturated the color will be.
You can control the white remapping control in several ways. You can
click on the handle and drag it around with the mouse. There are also
four buttons provided under the dial. One pair allows you to rotate the
handle clockwise and counter-clockwise without changing its distance from
the center. The other pair of buttons lets you change the distance
between the handle and the center without changing the angle.
The current Hue and Saturation values provided by the control is
displayed below the dial. The first number is the Hue component, in
degrees, and the second is the Saturation component, as a percentage.
There is also a checkbox that will let you turn off the white remapping
Page 22 10/89
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
control. This lets you quickly compare your modified 'white' with the
original white. You can also effectively disable the white remapping
control by putting the handle back in the center of the control. The
easiest way to do this is to click and hold the 'move towards center'
button until the saturation gets down to 0%.
Example:
⊕ Press the Grey control in the Colormap Editing tool. This turns all
the colors in the image into shades of grey.
⊕ Drag the handle in the white remapping control halfway down towards
the 'R' mark. The Hue and Saturation values should be roughly 0-
degrees and 50%. The image should now be displayed in shades of pink.
Section 5.3.3: The Saturation Control
The saturation control lets you globally increase or decrease the color
saturation of the image. In effect, it is much like the 'color' control
on most color televisions.
The saturation control is a dial that operates exactly like the dials
described in "Section 5.2.1 Using the Dial Controls". In short, you can
click and hold down any of the four buttons in the bottom of the control
to increase or decrease the control's value. You can also click on the
dial itself and move the pointer around directly.
The saturation control has values that range from '-100%' to '+100%'. At
its default setting of '0%', the saturation control has no effect on the
image. As the values increase, the colors become more saturated, up to
'+100%' where every color is fully saturated. Likewise, as values
decrease, the colors become desaturated. At '-100%', every color will
become a completely desaturated (i.e., a shade of grey). Note that this
control is applied after the the White Remapping control, so if you
'greyify' the image by completely desaturating it, you will not be able
to color it using the White Remapping control.
Unless you're trying for some special effects, the useful range of this
control is probably '+/-20%'. Also note that the control will have no
effect on shades of grey, as they have no color to saturate.
Section 5.3.4: The Intensity Graph
The intensity graph allows you to change the brightness of the image,
change the contrast of the image, and get some unique effects.
The intensity graph is a function that lets you remap intensity values
(the Value component in HSV Colorspace) into other intensity values. The
input and output values of this function both range from 0 to 255. The
input values range along the x axis of this graph (the horizontal). For
every input value (point along the x axis) there is a unique output value
determined by the height of the graph at that point. In the graph's
10/89 Page 23
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
default state, the function is a straight line from bottom-left to top-
right. In this case, each input value produces an equivalent output
value, and the graph has no effect.
There are a number of 'handles' along the graph. These provide your
major means of interacting with the graph. You can move them around
arbitrarily, subject to these two constraints: the handles at the far
left and far right of the graph can only be moved vertically, and handles
must remain between their neighboring handles for the graph to remain a
proper function.
The handles are normally connected by a spline curve. To see this, move
one of the handles by clicking and dragging it. (Note that the x,y
position of the current handle is displayed while the mouse button is
held down.) The function will remain a smoothly curved line that passes
through all the handles. You can change this behavior by putting the
function into 'lines' mode. Press the 'lines' button (the second button
down from the top). The function will change to a series of line
segments that connect the handles. Press the 'spline' button (the top
button) to go back to 'spline' mode.
The next two buttons let you add or delete handles. The 'add handle'
button will insert a handle into the largest 'gap' in the function. The
'delete handle' button will remove a handle from the smallest 'gap' in
the function. You can have as little as 2 handles, or as many as 16.
Note that as the number of handles gets large, the spline will start
getting out of control. You may wish to switch to 'lines' mode in this
case.
The 'Reset' button puts everything back on a straight line connecting
bottom-left to top-right (a 1:1 function). It does not change the number
of handles, nor does it change the x-positions of the handles.
The 'Gam' button lets you set the function curve by entering a single
number. The function is set equal to the gamma function:
Y = 255 * (I/255) ^ (1/g)
where I is the input value (0-255), g is the gamma value, and Y is the
computed result.
Gamma values (for our purposes) can range between 0 and 10000, non-
inclusive.
⊕ A gamma value of '1.00' results in the normal 1:1 straight line.
⊕ Gamma values of less than 1.00 but greater than 0.00 result in
'exponential' curves, which will dim the image.
⊕ Gamma values greater than 1.00 result in 'logarithmic' curves, which
will brighten the image. Try it and see.
Page 24 10/89
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
There is a shortcut for the 'Gam' button. Type 'g' while the mouse is
inside the graph window.
Also, touching any of the handles after a 'Gam' command will put the
graph back into its 'normal' mode. (Either 'spline' or 'lines' depending
on which of the top two buttons is turned on.)
Generally, whenever you move a graph handle and let go of it, the image
will be redrawn to show you the effects of what you've done. This can be
time-consuming if you intend to move many points around. You can
temporarily prevent the redisplay of the image by holding down a <shift>
key. Continue to hold the <shift> key down while you move the handles to
the new position. Release the <shift> key when you're done, and the
image will be redisplayed.
Section 5.4: The RGB Modification Tool
The RGB Modification tool is a collection of three graph windows, each of
which operate on one of the components of the RGB colorspace. This tool
lets you perform global color-correction on the image by boosting or
cutting the values of one or more of the RGB color components. You can
use this to correct for color screens that are 'too blue', or for color
printers that produce 'brownish' output, or whatever.
The graphs work exactly as explained in "Section 5.3.4: The Intensity
Graph".
Neat Trick: In addition to color-correction, you can use the RGB
modification tool to add color to images that didn't have color to begin
with. For instance, you can 'pseudo-color' a greyscale image.
An example of pseudo-coloring:
⊕ Adjust the Red graph so that there is a strong red presence on the
right side of the graph, and none on the left, or in the middle.
⊕ Adjust the Green graph so that there is a strong green presence in the
middle of the graph, and none on the left or right.
⊕ Adjust the Blue graph so that there is a strong blue presence on the
left side of the graph, and none on the left, or in the middle.
You now have a transformation that will take greyscale images and display
them in pseudo-color, using a 'temperature' color scheme. Neato!
Section 5.5: The Color Editor Controls
These buttons provide general control over the whole xv color editor
window. You can display the image with or without color modification,
save and recall presets, and undo/redo changes. Also, convenience
controls are given for performing some of the most common operations on
the Intensity graph.
10/89 Page 25
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
Apply (Keyboard equivalent 'p')
Displays the image using the current HSV and RGB Modifications.
Also turns the 'Display with HSV/RGB mods' checkbox on. (See
below.)
This is only useful when the 'Auto-apply HSV/RGB mods' checkbox is
off.
NoMod Displays the image without any HSV or RGB Modifications. Also
turns the 'Display with HSV/RGB mods' checkbox off.
Reset (Keyboard equivalent 'R')
Resets all HSV and RGB controls to their default settings. Doesn't
affect the Colormap Editing tool.
Undo Undoes the last change to the HSV or RGB controls.
It may be helpful to think of xv as maintaining a series of 32
'snapshots' of the HSV and RGB controls. You are normally looking
at the last frame in this series. The Undo control moves you
backwards in the series.
Redo Only available after you've hit Undo. Moves you forward in the
'snapshot' series described above. Note that if you have hit Undo
a few times (i.e., you're now looking at some frame in the middle
of the series), and you change an HSV or RGB control, all
subsequent frames in the series are thrown away, and the current
state becomes that last frame in the series.
1,2,3,4
Pressing any of these buttons recalls a preset (a complete set of
values for the HSV and RGB controls).
Set Used in conjunction with the Reset,1,2,3,4 buttons to store the
current settings of the HSV and RGB controls into a preset. To do
so, press the Set button, and then press one of the Reset,1,2,3,4
buttons. The current HSV and RGB control settings will be stored
in that preset, as long as xv continues running. The values will
be lost when the program exits. It is also possible to save these
values permanently. See the CutRes button (below) and "Section 9:
Modifying XV Behavior" for more details.
CutRes
Copies the current settings of the HSV and RGB controls, as text,
into the X server's cut buffer. You can then use a text editor to
Page 26 10/89
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
paste these values into your '.Xdefaults' (or '.Xresources') file.
This lets you save the current settings 'permanently'. See
"Section 9: Modifying XV Behavior" for more details.
Close This button closes the xv color editor window.
Brite Brightens the image by moving all the handles in the Intensity
graph up by a constant amount.
Dim Darkens the image by moving all the handles in the Intensity graph
down by a constant amount.
Sharp Increases the contrast of the image by moving handles on the left
side of the Intensity graph down, and handles on the right side up.
Dull Decreases the contrast of the image by moving handles on the left
side of the Intensity graph up, and handles on the right side down.
Norm (Keyboard equivalent 'N')
Normalizes the image so that the darkest pixels in the image are
given an intensity of '0', and the brightest pixels in the image
are given an intensity of '255'. Intermediate colors are
interpolated accordingly. This forces the image to have the full
(maximum) dynamic range.
HistEQ (Keyboard equivalent 'H')
Runs a histogram equalization algorithm on the currently displayed
region of the image. This is, if you're cropped, it will only run
the algorithm on the cropped section. Note, however, that the only
modification it makes to the image is to generate a bizarre
corrective Intensity curve. As such, if you HistEQ a section of
the image, then UnCrop, the rest of the image will probably not be
what you'd want. Also note that the histogram curve will 'go away'
if you touch any of the handles in the Intensity graph window, just
like a 'gamma' curve would.
The 'Display with HSV/RGB mods' checkbox tells you whether or you're
looking at a modified image (checked) or the 'raw', unmodified image
(unchecked). The Apply and NoMod buttons change the setting of this
checkbox, and you can also change the checkbox directly by clicking on
it.
The 'Auto-apply HSV/RGB mods' checkbox controls whether or not the
program regenerates and redisplays the image after each change to an HSV
10/89 Page 27
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
or RGB control. By default, this checkbox is turned on, so that you can
easily see the results of your modifications. However, in the case that
you want to make a large number of changes at once, it might be
preferable to turn automatic redisplay off for a while, to speed things
up.
The 'Auto-reset on new image' checkbox controls whether or not the HSV
and RGB controls are Reset back to their default values whenever a new
image is loaded up. By default, this is also turned on, as when you're
playing with the HSV/RGB controls, you probably only want to affect the
current image, and not all subsequently loaded images as well.
SECTION 6: THE LOAD WINDOW
The xv load window lets you load and view images interactively, without
specifying them on the command line when you start xv.
The load window shows the contents of the current directory in a
scrolling window. The files will be sorted alphabetically, with all the
directories (and symbolic links to directories, if your operating system
supports them) displayed first.
This list window operates in the same way that the one in the xv controls
window works. (See "Section 3.7.1: Operating a List Window" for
details.) In short, you can operate the scroll bar, drag the highlight
bar around the window, and use the up-arrow, down-arrow, Home, End, Page
Up, and Page Down keys on your keyboard.
Whenever you click on a name in the list (or otherwise change the
position of the highlight bar), the name of the highlighted file is
copied to the "Load file" text entry region, located below the list
window. Pressing the Ok button (or typing <return>) will cause the
program to attempt to load the specified file. If the load attempt is
successful, the load window will disappear, and the new image will be
displayed. Otherwise, an error message will be displayed, and the load
window will remain visible.
The Browse checkbox overrides this behavior, and keeps the load window
visible until it is explicitly closed via the Cancel button. This is
handy if you're using xv to 'wander around a directory tree', and plan to
be using the Load command quite often.
If the image is successfully loaded, its name will be added to the xv
controls window list. This will let you quickly reload it later without
have to go through the xv load window again.
You can also load a file by double-clicking on its name in the file list.
If the specified filename begins with a '!' or '|' character, the
filename will be interpeted as a shell command to run. The leading '!'
or '|' gets stripped off, and the rest of the line is fed to the default
system shell. The command is expected to generate an image in one of the
Page 28 10/89
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
formats that xv recognizes. This image is expected to be the stdout of
the specified command. If the command returns non-zero, it is assumed
that the command failed, and no image is loaded. You can pipe multiple
commands together. For example, loading "! xwd | xwdtopnm" would run xwd
to generate a window dump, pipe that to xwdtopnm to convert it to a PPM
file, which in turn is piped to xv.
If the specified file is a directory, xv will figure that out and
(instead of loading it) will 'cd' to that directory, and display its
contents in the list window.
Above the list window is a pop-up menu button, much like the Display
Modes button in the xv controls window. It normally displays the name of
the current directory. If you click this button, and hold the mouse
down, the complete path will be shown, one directory per line. You can
go 'up' the directory tree any number of levels, all the way up to the
root directory, by simply selecting a directory name in this list.
For those who prefer the direct approach, you can simply type file or
directory names in the "Load file" text entry region. If you type a
directory name and hit <return>, xv will 'cd' to that directory and
display its contents in the list window. If you type a file name and hit
<return>, xv will attempt to load the file. You can enter relative paths
(relative to the currently displayed directory), absolute paths, and even
paths that begin with a '~'.
The "Load file" text entry region supports a number of emacs-like editing
keys.
Ctrl-F moves the cursor forward one character
Ctrl-B moves the cursor backward one character
Ctrl-A moves the cursor to the beginning of the line
Ctrl-E moves the cursor to the end of the line
Ctrl-D deletes the character to the right of the cursor
Ctrl-U clears the entire line
Ctrl-K clears from the cursor position to the end of the line.
If the filename is so long that it cannot be completely displayed in the
text entry region, a thick line will appear on the left or right side (or
both sides) of the region to show that "there's more over this way".
Pressing the Rescan button will rescan the current directory. While the
contents of the current directory are read each time the load window is
opened, it is perfectly possible (given a multitasking operating system)
that some other program may add, delete, or rename files in the current
directory. XV would not know if this happened. The Rescan button gives
10/89 Page 29
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
you an easy way of 'kicking' xv into looking again.
SECTION 7: THE SAVE WINDOW
The xv save window lets you write images back to disk, presumably after
you've modified them. You can write images back many different formats,
not just the original format. Warning! Images are saved as they are
currently shown (at the current size, with the current color
modification, rotation, cropping, etc. applied). The only exception to
this rule is if you are on a 1-bit B/W display. The fact that images
have to be stippled in black and white in order to be displayed on such a
screen doesn't count as 'modification', and the file won't be saved 'as
displayed'. You can manipulate and save full-color images on such a
display, even if you can't really see them.
For the most part, the xv save window operates exactly like the xv load
window. (See "Section 6: The Load Window" for details.) Only the
differences are listed here.
When the window is opened, it should have the filename of the currently
loaded file already entered into the text entry region. If you change
directories, or click on a file name in the list window, this name will
be cleared and replaced with the new name.
At the bottom of the window are a list of possible formats in which you
can save the file. If you click on one of these formats, and your
filename has a recognized suffix (i.e., '.gif', '.GIF', '.pbm', etc.),
the suffix portion of your filename will be replaced with the new,
appropriate suffix for the selected format.
You can pipe output from xv to other programs by using the xv save
window. If the first character of the specified filename is '!' or '|',
the rest of the filename is interpreted as a command to pipe the output
to, in the currently selected image format. A fine use for this feature
is directly printing images to a PostScript printer by selecting
'PostScript' in the formats list, and typing something like "| lpr" as
the filename. In this case, xv will create a temporary file, write the
PostScript to that file, and cat the contents of that file to the entered
command. XV will wait for the command to complete. If the command
completed successfully, the xv save window will disappear. If the
command was unsuccessful, the window will remain visible. In any event,
the temporary file will be deleted.
There is a 'Save at normal size' checkbox. Normally, when you save an
image, it will be saved at the current expansion (ie, one screen pixel
will map to one image pixel in the saved file. Sometimes, however, it is
desirable to save an image at its original size. This is most relevant
when you're viewing images larger than your screen. By default, xv will
automatically shrink images so that they fit on the screen. If you save
these images, you'll find that you've lost a lot of data, that maybe you
wanted to keep. That's what this checkbox is here for. Note: certain
operations, such as Smooth and Dither only affect the displayed image.
Page 30 10/89
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
If you choose to save an image at its normal size, these effects will not
be in the saved image.
At the bottom right side of the window there is a list of possible
'Color' variations to save. Most file formats support different 'sub-
formats' for 24-bit color, 8-bit greyscale, 1-bit B/W stippled, etc. Not
all of them do. Likewise, not all 'Color' choices are available in all
formats.
In general, the 'Color' choices do the following:
Full Color
Saves the image as currently shown with all color modifications,
cropping, rotation, flipping, resizing, and smoothing. The image
will be saved with all of its colors, even if you weren't able to
display them all on your screen. For example, you can load a color
image on a 1-bit B/W display, modify it, and write it back. The
saved image will still be full color, even though all you could see
on your screen was some B/W-dithered nightmare.
Greyscale
Like Full Color, but saves the image in a greyscale format.
B/W Dithered
Like Full Color, but before saving the image xv generates a 1-bit-
per-pixel, black-and-white dithered version of the image, and saves
that, instead.
Reduced Color
Saves the image as currently shown, with all color modifications,
cropping, rotation, flipping, resizing, and smoothing. The image
will be saved as shown on the screen, with as many or few colors as
xv was able to use on the display. The major purpose of this is to
allow special effects (color reduction) to be saved, in conjunction
with the '-ncols' command line option. You will probably never
need to use this.
Format notes:
GIF While xv can read both the GIF87a and GIF89a formats, it will only
write GIF87a. This is in keeping with the GIF89 specification,
which states that if you don't need any of the features added in
GIF89 (which xv doesn't), you should continue to write GIF87, for
greater compatibility with old GIF87-only readers.
Since GIF only supports one format (up to 8 bits per pixel, with a
colormap), there will be no size difference between a Full Color
10/89 Page 31
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
and a Greyscale image. A B/W Dithered image, on the other hand,
will be considerably smaller.
PM Full Color images are saved in the 3-plane, 1-band, PM_C format.
Greyscale and B/W Dithered images are both saved in the 1-plane,
1-band, PM_C format. As such, there is no size advantage to saving
in the B/W Dithered format.
PBM (raw)
Full Color images are saved in PPM format. Greyscale images are
saved in PGM format. B/W Dithered images are saved in PBM format.
Each of these formats are tailored to the data that they save, so
PPM images are larger than PGM images, which are in turn larger
than PBM images.
In the raw variation of the PBM formats, the header information is
written in plain ASCII text, and the image data is written as
binary data. This is the more popular of the two dialects of PBM.
PBM (ascii)
Like PBM (raw), only the image data is written as ASCII text. As
such, images written in this format will be several times larger
than images written in PBM (raw). This is a pretty good format for
interchange between systems because it is easy to parse. Also,
since they are pure, printable ASCII text, images saved in this
format can be mailed, without going through a uuencode-like
program.
Note that xv-produced PBM files may break some PBM readers that do
not correctly parse comments. If your PBM reader cannot parse
comments, you can easily edit the PBM file and remove the comment
lines. A comment is everything from a "#" character to the end of
the line.
X11 Bitmap
Saves files in the format used by the bitmap program, which is part
of the standard X11 distribution. Since bitmap files are
inherently 1-bit per pixel, you can only select the B/W Dithered
option for this format.
Sun Rasterfile
Full/Reduced Color images are stored in a 24-bit RGB format,
Greyscale images are stored in an 8-bit greyscale format, and B/W
Dithered images are stored in a 1-bit B/W format.
Page 32 10/89
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
PostScript
Full/Reduced Color images are stored in a 24-bit RGB format,
Greyscale images are stored in an 8-bit greyscale format, and B/W
Dithered images are stored in a 1-bit B/W format.
XV writes Encapsulated PostScript, so you can incorporate xv-
generated PostScript into many desktop-publishing programs. XV
also prepends some color-to-greyscale code, so even if your printer
doesn't support color, you can still print 'color' PostScript
images. These images will be three times larger (in file size)
than their greyscale counterparts, so it's a good idea to save
Greyscale PostScript, unless you know you may be printing the file
on a color printer at some point.
Also, you should probably never need to generate B/W Dithered
PostScript, as every PostScript printer I've ever heard of can
print greyscale images. The only valid cases I can think of are:
A) doing it for a special effect, and B) doing it to generate a
much smaller (roughly 1/8th the size) PostScript file.
Note: When you try to save a PostScript file, the xv postscript
window will pop up to let you specify how you want the image
printed. (See "Section 8: The PostScript Window", for details.)
JPEG XV writes files in the JFIF format created by the Independent JPEG
Group. Full/Reduced Color images are written in a 24-bit RGB
format, and Greyscale images are written in an 8-bit greyscale
format. B/W Dithered images should not be used, as they will
probably wind up being larger than Greyscale versions of the same
images, due to the way JPEG works. Note: You cannot write Reduced
Color JPEG files. If you attempt to, a Full Color JPEG file will
be saved.
When you save in the JPEG format, a dialog box will pop up and ask
you for a quality setting. '75%' is the default value, and really,
it's a fine value. You shouldn't have to change it unless you're
specifically trying to trade off quality for compression, or vice
versa. The useful range of values is 5%-95%.
TIFF Full/Reduced Color images are written in a 24-bit RGB format,
Greyscale images are written in an 8-bit greyscale format, and B/W
Dithered images are written in a 1-bit B/W format.
When you save in the TIFF format, a dialog box will pop up and ask
you which type of image compression it should use. None, LZW, and
PackBits compression types are available for Full/Reduced Color,
Greyscale, and B/W Dithered modes. In addition, there are two B/W
Dithered-only algorithms, CCITT Group3 and CCITT Group4.
10/89 Page 33
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
SECTION 8: THE POSTSCRIPT WINDOW
The xv postscript window lets you describe how your image should look
when printed. You can set the paper size and the image size, position
the image on the paper, and print in 'portrait' or 'landscape' mode.
The majority of the xv postscript window is taken up by a window that
shows a white rectangle (the page) with a black rectangle (the image)
positioned on it. You can position the image rectangle anywhere on the
page. The only constraint is that the center of the image (where the two
diagonal lines meet) must remain on the page. Only the portion of the
image that is on the page will actually be printed.
The image can be (roughly) positioned on the page by clicking in the
image rectangle and dragging it around. As you move the image, the "Top"
and "Left" position displays will show the size of the top and left
margins (the distance between the top-left corner of the page and the
top-left corner of the image).
You'll note that you have limited placement resolution with the mouse.
If you want to fine-position the image, you can use the arrow keys to
move the image around. The arrow keys will move the image in .001"
increments. You can hold them down, and they will auto-repeat. You can
also hold a <shift> key down while using the arrow keys. This will move
the image in .01" increments.
You can change the size of the printed image by adjusting the "Width" or
"Height" dials. Normally, the dials are locked together, to keep the
aspect ratio of the image constant. You can unlock the dials by turning
the off the checkbox located below the dials. As you change the dials,
the size of the image (when printed) is displayed below, in inches and in
millimeters. The current resolution of the image is also displayed
below. The "Resolution" numbers tell you how many image pixels will be
printed per inch.
Located below the 'page' rectangle are a set of radio buttons that let
you specify the current paper size (8.5" x 11", 8.5" x 14", 11" x 17",
A4, B5, 4"x5", and 35mm), and orientation (Portrait and Landscape).
The Center button will center the image on the page. The Maxpect button
will make the image as large as possible (maintaining half-inch margins
on all sides) without changing the aspect ratio.
There are a pair of small buttons located next to the "Left" and "Top"
displays. Clicking the "Left" one will cycle between displaying the
"Left" margin, the "Right" margin, and the "Center X" position (the
distance from the left edge of the paper to the center of the image).
Clicking the "Top" display's button will cycle between displaying the
size of the "Top" margin, the size of the "Bottom" margin, and the
"Center Y" position (the distance from the top edge of the paper to the
center of the image).
Page 34 10/89
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
At the top of the xv postscript window are a pair of checkboxes. The
"preview" checkbox lets you specify whether or not to include a b/w
preview of the image in the PostScript file. Certain desktop publishing
programs may make use of such a preview.
The 'compress' checkbox lets you specify whether or not to generate
compressed 8-bit PostScript. This is particularly handy if you're
generating color PostScript as color PostScript files are normally three
times larger than their greyscale counterparts. Compression can shrink
these color PostsScript files by a factor of 4:1. It has a lesser effect
on greyscale images. It should be noted, however, that compressed
PostScript files may take 2-3 times longer to print than uncompressed
PostScript files. However, if you are connected to your laser printer via
a slow 9600 baud serial line, the decreased transmission time due to
compressed data may more than make up for the increased execution time.
You'll have to decide for yourself.
Click the "Ok" button when you're finished with the xv postscript window.
If everything is successful, the xv postscript and the xv save window
will both close. If xv was unable to write the PostScript file, the xv
postscript window will close, but the xv save window will remain open, to
give you a chance to enter a different filename.
SECTION 9: MODIFYING XV BEHAVIOR
XV supports literally dozens of command line options and X11 resources.
Fortunately, it is doubtful that you'll ever need to use more than a
small few. The rest are provided mainly for that 'one special case'
application of xv...
Section 9.1: Command Line Options Overview
If you start xv with the command 'xv -help', the current list of options
will be displayed:
xv [-] [-2xlimit] [-aspect w:h] [-bg color] [-black color]
[-bw width] [-cegeometry geom] [-cemap] [-cgeometry geom]
[-clear] [-cmap] [-cursor char#] [-DEBUG level]
[-display disp] [-dither] [-expand exp] [-fg color]
[-fixed] [-geometry geom] [-help] [-hi color] [-hsv]
[-igeometry geom] [-imap] [-keeparound] [-lo color] [-max]
[-maxpect] [-mono] [-ncols nc] [-nglobal] [-ninstall]
[-nopos] [-noqcheck] [-owncmap] [-perfect] [-quit] [-rbg color]
[-rfg color] [-rgb] [-rmode mode] [-root] [-rw] [-slow24]
[-smooth] [-visual type] [-wait seconds] [-white color]
[-wloop] [-noresetroot] [-browse] [-nostat] [-best24] [-quick24]
[-cecmap] [-crop] [-rv] [-nolimits] [-loadclear] [filename ...]
Section 9.2: General Options
10/89 Page 35
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
-help Print usage instructions, listing the current available command-
line options. Any unrecognized option will do this as well.
-display disp
Specifies the display that xv should attempt to connect to. If you
don't specify a display, xv will use the environment variable
$DISPLAY.
-fg color (Resource name: foreground string)
Sets the foreground color used by the windows.
-bg color (Resource name: background string)
Sets the background color used by the windows.
-hi color (Resource name: highlight string)
Sets the highlight color used for the top-left edges of the control
buttons.
-lo color (Resource name: lowlight string)
Sets the lowlight color used for the bottom-right edges of the
control buttons, and also the background of some windows.
-bw bwidth (Resource name: borderWidth integer)
Sets the width of the border on the windows. Your window manager
may choose to ignore this, however.
Section 9.3: Image Sizing Options
-geometry geom (Resource name: geometry string)
Lets you specify the size and placement of the 'image' window.
It's most useful when you only specify a position, and let xv
choose the size. If you specify a size as well, xv will create a
window of that size, unless -fixed is specified. The geom argument
is in the form of a normal X geometry string (e.g. "300x240" or
"+10+10" or "400x300+10+10").
-fixed (Resource name: fixed boolean)
Only used in conjunction with the -geometry option. If you specify
a window size with the -geometry option, xv will normally stretch
the picture to exactly that size. This is not always desirable, as
it may seriously distort the aspect ratio of the picture.
Specifying the -fixed option corrects this behavior by instructing
xv to use the specified geometry size as a maximum window size. It
will, however, preserve the original aspect ratio of the picture.
Page 36 10/89
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
For example, if you give a rectangular geometry of '320x240', and
you try to display a square picture with a size of '256x256', the
window opened will actually be '240x240', which is the largest
square that still fits in the '320x240' rectangle that was
specified.
-expand exp (Resource name: expand floating-point)
Lets you specify an initial expansion or compression factor for the
picture. You can specify floating-point values. Values larger
than zero multiply the picture's dimensions by the given factor.
(i.e., an expand factor of '3' will make a 320x200 image display as
960x600).
Factors less than zero are treated as reciprocals. (i.e., an
expand factor of '-4' makes the picture 1/4th its normal size.).
'0' is not a valid expansion factor.
-aspect w:h (Resource name: aspect string)
Lets you set an initial aspect ratio, and also sets the value used
by the Aspect control. The aspect ratio of nearly every X display
(and, in fact, any civilized graphics display) is 1:1. What this
means is that pixels appear to be 'square'. A 100 pixel wide by
100 pixel high box will appear on the screen as a square.
Unfortunately, this is not the case with some screens and
digitizers. The -aspect option lets you stretch the picture so
that the picture appears correctly on your display. Unlike the
other size-related options, this one doesn't care what the size of
the overall picture is. It operates on a pixel-by-pixel basis,
stretching each image pixel slightly, in either width or height,
depending on the ratio.
Aspect ratios greater than '1:1' (e.g., '4:3') make the picture
wider than normal. Aspect ratios less than '1:1' (e.g. '2:3') make
the picture taller than normal. (Useful aspect ratio: A 512x480
image that was supposed to fill a standard 4x3 video screen
(produced by many video digitizers) should be displayed with an
aspect ratio of '5:4')
Section 9.4: Color Allocation Options
-ncols nc (Resource name: ncols integer)
Sets the maximum number of colors that xv will use. Normally, this
is set to 'as many as it can get'. However, you can set this to
smaller values for interesting effect. Most notably, if you set it
to '0', it will display the picture by dithering with 'black' and
'white'. (The actual colors used can be set by the -black and
-white options, below.)
10/89 Page 37
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
-nglobal (Resource name: nglobal boolean)
Adjusts the way the program behaves when it is unable to get all
the colors it requested. Normally, it will search the display's
default colormap, and 'borrow' any colors it deems appropriate.
These borrowed colors are, however, not owned by xv, and as such,
can changed without xv's permission, or knowledge. If this
happens, the displayed picture will change, in a less-than-
desirable direction. If you specify the -nglobal option, xv will
not use 'global' colors. It will only use colors that it
successfully allocated, which makes it immune to any color changes.
It should be noted that 'use global colors' is the default because
color changes aren't generally a problem if you are only using xv
to display a picture for a short time. Color changes only really
become a problem if you use xv to display a picture that you will
be keeping around for a while, while you go and do some other work
(such as using xv to display a background). In such cases you will
want to specify -nglobal. Note: using the -ncols or -root options
automatically turn on -nglobal.
-rw (Resource name: rwColor boolean)
Tells xv to use read/write color cells. Normally, xv allocates
colors read-only, which allows it to share colors with other
programs. If you use read/write color cells, no other program can
use the colors that xv is using, and vice-versa. The only reason
you'd do such a thing is that using read/write color cells allows
the Apply function in the xv color editor window to operate much
faster.
-perfect (Resource name: perfect boolean)
Makes xv try 'extra hard' to get all the colors it wants. In
particular, when -perfect is specified, xv will allocate and
install its own colormap if (and only if) it was unable to allocate
all the desired colors. This option is not allowed in conjunction
with the -root option.
-owncmap (Resource name: ownCmap boolean)
Like '-perfect', only this option forces xv to always allocate and
install its own colormap, thereby leaving the default colormap
untouched.
-cecmap (Resource name: ceditColorMap boolean)
Specifies whether xv installs the image's colormap in the xv color
editor window, as well as in the image's window. By default, the
program does install the colormap in the color editor window,
however this can occasionally make the color editor window
unreadable. (This option only apples when the '-perfect' or '-
owncmap' options create their own colormaps.)
Page 38 10/89
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
-ninstall (Resource name: ninstall boolean)
Prevents xv from 'installing' its own colormap, when the -perfect
or -owncmap options are in effect. Instead of installing the
colormap, it will merely 'ask the window manager, nicely' to take
care of it. This is the correct way to install a colormap (i.e.,
ask the WM to do it), unfortunately, it doesn't actually seem to
work in many window managers, so the default behavior is for xv to
handle installation itself. However, this has been seen to annoy
one window manager (dxwm), so this option is provided if your WM
doesn't like programs installing their own colormaps.
Section 9.5: 24-bit Conversion Options
The following options only come into play if you are using xv to display
24-bit RGB data (PPM files, color PM files, JPEG files, the output of
bggen, etc.). They have no effect whatsoever on how GIF pictures or 8-
bit greyscale images are displayed.
-quick24 (Resource name: quick24 boolean)
Forces xv to use the 'quick' 24-bit to 8-bit conversion algorithm.
This algorithm dithers the picture using a fixed set of colors that
span the entire RGB colorspace. In versions of xv prior to 2.10,
this was the default algorithm. It no longer is.
-slow24 (Resource name: slow24 boolean)
Specifies that the 'slow' 24-bit to 8-bit conversion algorithm is
to be used by the program. This algorithm uses a version of
Heckbert's median cut algorithm to pick the 'best' colors on a
per-image basis, and dithers with those. This is the current
default conversion algorithm.
Advantages: The -slow24 algorithm often produces better looking
pictures than the -quick24 algorithm.
Disadvantages: The -slow24 algorithm is about half the speed of the
-quick24 algorithm. Also, since the colors are chosen on a per-
image basis, it can't be used to display multiple images
simultaneously, as each image will almost certainly want a
different set of 256 colors. The -quick24 algorithm, however, uses
the same exact colors for all images, so it can display many images
simultaneously, without running out of colors.
-best24 (Resource name: best24 boolean)
Forces xv to use the same algorithm used in the program ppmquant,
written by Jef Poskanzer. This algorithm also uses a version of
Heckbert's median cut algorithm, but is capable of picking 'better'
colors than the -slow24 algorithm, and it doesn't dither.
Advantages: Generally produces slightly better images than the
10/89 Page 39
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
-slow24 algorithm. Also, the images are undithered, so they look
better when expanded.
Disadvantages: Much slower than the -slow24 algorithm. Like, 5 to
10 times slower. The images produced aren't that much better than
those produced by the -slow24 algorithm.
-noqcheck (Resource name: noqcheck boolean)
Turns off a 'quick check' that is normally made. Normally, before
running either of the 24-bit to 8-bit conversion algorithms, xv
determines whether the picture to be displayed has more than 256
unique colors in it. If the picture doesn't, it will treat the
picture as an 8-bit colormapped image (i.e., GIF), and won't run
either of the conversion algorithms.
Advantages: The pictures will be displayed 'perfectly', whereas if
they went through either of the conversion algorithms, they'd be
dithered.
Disadvantages: Often uses a lot of colors, which limits the ability
to view multiple images at once. (See the -slow24 option above for
further info about color sharing.)
Section 9.6: Root Window Options
xv has the ability to display images on the root window of an X display,
rather than opening its own window (the default behavior). When using
the root window, the program is somewhat limited, because the program
cannot receive input events (key press and mouse clicks) from the root
window. As a result, you cannot track pixel values, or crop, nor can you
use keyboard commands while the mouse is in the root window.
-root (Resource name: <none>)
Directs xv to display images in the root window, instead of opening
its own window. Exactly how the images will be displayed in the
root window is determined by the setting of the -rmode option.
-rmode mode (Resource name: rootMode integer)
Determines how images are to be displayed on the root window, when
-root has been specified. You can find the current list of 'modes'
by using a mode value of '-1'. XV will complain, and show a list
of valid modes. The current list at of the time of this writing
is:
0: tiling
1: integer tiling
2: mirrored tiling
3: integer mirrored tiling
4: centered tiling
Page 40 10/89
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
5: centered on a solid background
6: centered on a 'warp' background
7: centered on a 'brick' background
The default mode is '0'. See "Section 3.5: The Display Modes Menu"
for a description of the different display modes.
-noresetroot (Resource name: resetroot)
Normally, when changing from root to window display mode, the root
is set to the standard X crosshatch pattern. Using -noresetroot or
setting resetroot to false will force the root to remain. This is
useful when comparing between different tiled images.
-rfg color (Resource name: rootForeground string)
Sets the 'foreground' color used in some of the root display modes.
-rbg color (Resource name: rootBackground string)
Sets the 'background' color used in some of the root display modes.
-max (Resource name: <none>)
Makes xv automatically stretch the image to the full size of the
screen. This is mostly useful when you want xv to display a
background. While you could just as well specify the dimensions of
your display ('-geom 1152x900' for example), the -max option is
display-independent. If you suddenly decide to start working on a
1280x1024 display (ferinstance) the same command will still work.
Note: If you specify -max when you aren't using -root, the behavior
is slightly different. The image will be made as large as possible
while still preserving the normal aspect ratio.
-maxpect (Resource name: <none>)
Makes the image as large as possible while preserving the aspect
ratio.
-quit (Resource name: <none>)
Makes xv display the (first) specified file and exit, without any
user intervention. Since images displayed on the root window
remain there until explicitly cleared, this is very useful for
having xv display background images on the root window in some sort
of start-up script. Needless to say, this is only useful if you
are using -root.
-clear (Resource name: <none>)
Clears the root window of any extraneous xv images. Note: it is
not necessary to do an 'xv -clear' before displaying another
10/89 Page 41
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
picture in the root window. xv will detect that there's an old
image in the root window and automatically clear it out (and free
the associated colors).
Section 9.7: Window Options
XV currently consists of three main windows, plus one window for the
actual image. These three windows (the xv controls window, the xv info
window, and the xv color editor window) may be automatically mapped and
positioned when the program starts.
-cmap (Resource name: ctrlMap boolean)
Maps the xv controls window.
-cgeom geom (Resource name: ctrlGeometry string)
Sets the initial geometry of the xv controls window. Note: only
the position information is used. The window is of fixed size.
-imap (Resource name: infoMap boolean)
Maps the xv info window.
-igeom geom (Resource name: infoGeometry string)
Sets the initial geometry of the xv info window. Note: only the
position information is used. The window is of fixed size.
-cemap (Resource name: ceditMap boolean)
Maps the xv color editor window.
-cegeom geom (Resource name: ceditGeometry string)
Sets the initial geometry of the xv color editor window. Note:
only the position information is used. The window is of fixed
size.
-nopos (Resource name: nopos boolean)
Turns off the 'default' positioning of the various xv windows.
Every time you open a window, you will be asked to position it.
(Assuming your window manager asks you such things. mwm, for
instance doesn't seem to ask)
Section 9.8: Miscellaneous Options
-mono (Resource name: mono boolean)
Forces the image to be displayed as a greyscale. This is most
useful when you are using certain greyscale X displays. While xv
attempts to determine if it's running on a greyscale display, many
Page 42 10/89
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
X displays lie, and claim to be able to do color. (This is often
because they have color graphics boards hooked up to b/w monitors.
The computer, of course, has no way of knowing what type of monitor
is attached.) On these displays, if you don't specify -mono, what
you will see is a greyscale representation of one of the RGB
outputs of the system. (For example, you'll see the 'red' output
on our greyscale Sun 3/60s.) The -mono option corrects this
behavior.
-white color (Resource name: white string)
Specifies the 'white' color used when the picture is b/w stippled.
(When '-ncols 0' has been specified.)
-black color (Resource name: black string)
Specifies the 'black' color used when the picture is b/w stippled.
(When '-ncols 0' has been specified.)
Try something like: 'xv -ncols 0 -bl red -wh yellow <filename>' for
some interesting, late-'60s-style psychodelia effects.
-wait secs (Resource name: <none>)
Turns on a 'slide-show' feature. Normally, if you specify multiple
input files, xv will display the first one, and wait for you to
give the Next command (or whatever). The -wait option makes xv
wait the specified number of seconds, and then go on to the next
picture, without any user intervention. The program still accepts
commands, so it's possible to 'abort' the current picture without
waiting the full specified time by using the Next command.
-wloop (Resource name: <none>)
Normally, when running a slide-show with the -wait option, xv will
terminate after displaying the last image. If you also specify the
-wloop option, the program will loop back to the first image and
continue the slide-show until the user issues the Quit command.
-rgb (Resource name: hsvMode boolean)
Specifies that, by default, the colormap editing dials in the xv
color editor window should be in RGB mode. This is the normal
default behavior.
-hsv (Resource name: hsvMode boolean)
Specifies that, by default, the colormap editing dials in the xv
color editor window should be in HSV mode.
10/89 Page 43
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
-dither (Resouce name: autoDither boolean)
When specified, tells xv to automatically issue a Dither command
whenever an image is first displayed. Useful on displays with
limited color capabilities (4-bit and 6-bit displays.)
-smooth (Resource name: autoSmooth boolean)
When specified, tells xv to automatically issue a Smooth command
whenever an image is first displayed. This is useful when you are
using one of the image sizing options (such as '-expand' or '-
max').
-crop (Resource name: autoCrop boolean)
When specified, tells xv to automatically issue an AutoCrop command
whenever an image is first displayed.
-visual vistype (Resource name: visual string)
Normally, xv uses the default visual model provided by your X
server. You can override this by explicitly selecting a visual to
use. Valid types are StaticGray, StaticColor, TrueColor,
GrayScale, PseudoColor,
and DirectColor. Not all of these are necessarily provided on any
given X display. Run xdpyinfo on your display to find out what
visual types are supported.
-cursor curs (Resource name: cursor integer)
Specifies an alternate cursor to use in the image window (instead
of the normal 'cross' cursor). curs values are obtained by finding
the character number of a cursor you like in the 'cursor' font.
(Run 'xfd -fn cursor' to display the cursor font.) For example, a
curs value of '56' corresponds to the (singularly useless) 'Gumby'
cursor.
-keeparound (Resource name: keepAround boolean)
The '-keeparound' option is now poorly named, as I've changed the
default behavior. Now, if you Delete the last file in the xv
controls list, nothing unexpected will happen. If you specify the
'-keeparound' option, which toggles the '-keeparound' flag off, the
program will automatically exit as a convenience. It should
probably be renamed '-goaway' or something...
-2xlimit (Resource name: 2xlimit boolean)
By default, xv prevents the image window from ever getting larger
than the screen. Unfortunately, because of this, if you load an
image that is larger than your screen, the image will be shrunk
until it fits on your screen. Some folks find this undesirable
behavior. Specifying the -2xlimit option doubles the size
Page 44 10/89
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
limitations. The image window will be kept from getting larger
than 2x the width and height of your screen.
Just in case you're wondering why there's any size limitations:
it's fairly easy to accidentally ask for a huge image to be
generated. Simply crop a section of the image, zoom so you can see
the individual pixels, and uncrop. If there were no size
limitations, the (expanded many times) image could be huge, and
might crash your X server. At the very least, it would take a long
period of time, and freeze your X server during part of it.
Generally undesirable behavior.
-nolimits (Resource name: nolimites boolean)
For the truly daring, this turns off all limitations on the maximum
size of an image window. (Well, there's still an X-imposed maximum
size of 64k by 64k, but that really shouldn't be a problem.)
Warning: as mentioned above, it is fairly easy to accidentally
generate a huge image when you do an UnCrop command, and you may
well crash xv, your X server, the host machine, or all three. Use
At Your Own Risk!!!
-rv (Resource name: reverse)
Makes xv display a 'negative' of the loaded image. White becomes
black, and black becomes white. Color images will have
'interesting' effects, as the RGB components are individually
reversed. For example, red (255,0,0) will become cyan (0,255,255),
yellow will become blue, and so on.
-DEBUG level (Resource name: <none>)
Turns on some debugging information. You shouldn't need this. If
everything worked perfectly, I wouldn't need this either.
-browse (Resource name: browseMode boolean)
Prevents the xv load window from being closed whenever you
successfully load a file. This makes 'browsing' a directory
somewhat more pleasant.
-nostat (Resource name: nostat boolean)
Turns off the stat() call is is performed for each file whenever
you change directories in the xv load and xv save windows. This is
useful if you're on a machine with lots of remote files mounted on
it, and you find the directory switching to be too slow.
-loadclear (Resource name: clearOnLoad boolean)
If you were on a PseudoColor display, xv used to automatically
clear the image window (or the root window, if you were in a root
10/89 Page 45
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
mode), whenever you loaded a new image. This was to prevent the
potentially annoying/confusing 'rainbow' effect that happens when
colormap entries are freed and reallocated with different colors.
Ths has changed. By default, xv no longer clears the image/root
window. This is for two reasons: I've decided that the rainbow
effect is semi-entertaining, in that it gives you something to look
at while the next image is being loaded. Secondly, if you are
viewing a series of images that have the same colors in them, it's
possible for xv to animate them (by using the '-wait' command line
option), albeit no faster than one frame every 1-2 seconds. For
example, you can go get the satellite radar images from
vmd.cso.uiuc.edu (in the directory 'wx'), run 'xv -wait 0 -wloop
SA*', and voila! Just like the evening news!
- Specifying '-' all by itself tells xv to take its input from stdin,
rather than from a file. This lets you put xv on the end of a Un*x
pipe.
Section 9.9: Color Editor Resources
You can set default values for all of the HSV and RGB modification
controls in the xv color editor window via X resources. The easiest way
to explain this is with an example.
⊕ Start xv and put it in the background by typing 'xv &'.
⊕ Type the command 'cat >foo' in an active xterm window
⊕ Bring the xv color editor window up.
⊕ Issue the Cut Resources command.
⊕ Click your Middle mouse button in the xterm window. A set of resource
lines describing the current state of the xv color editor controls
will be 'pasted' into the window.
⊕ You could type '<ctrl-D>' in the xterm to complete the cat command,
edit this file, and put it in your .Xdefaults/.Xresources file.
The lines generated by Cut Resources will look like the following:
xv.default.huemap1: 330 30 CW 330 30 CW
xv.default.huemap2: 30 90 CW 30 90 CW
xv.default.huemap3: 90 150 CW 90 150 CW
xv.default.huemap4: 150 210 CW 150 210 CW
xv.default.huemap5: 210 270 CW 210 270 CW
xv.default.huemap6: 270 330 CW 270 330 CW
xv.default.whtmap: 0 0 1
xv.default.satval: 0
xv.default.igraf: S 4 : 0,0 : 64,64 : 192,192 : 254,254
Page 46 10/89
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
xv.default.rgraf: S 4 : 0,0 : 64,64 : 192,192 : 254,254
xv.default.ggraf: S 4 : 0,0 : 64,64 : 192,192 : 254,254
xv.default.bgraf: S 4 : 0,0 : 64,64 : 192,192 : 254,254
These lines completely describe one state of the xv color editor
controls. There are five different states that you can specify via X
resources. The 'default' state (as shown) holds the settings used
whenever the program is first started, and whenever the Reset command is
used. You can also store settings in one of the four xv presets
(accessed via the '1'-'4' buttons in the xv color editor) by changing the
string 'default' in the above lines to 'preset1', 'preset2', 'preset3',
or 'preset4' respectively.
There are four types of resource described in these lines: huemap,
whtmap, satval, and graf.
Section 9.9.1: Huemap Resources
The huemap resources describe the state of the hue remapping dials.
There are six huemap resources per state of the xv color editor. These
huemap resources are numbered 'huemap1', 'huemap2', ... 'huemap6', and
correspond to the '1'-'6' radio buttons under the hue remapping dials.
Each huemap resources takes six parameters:
1. The 'starting' angle of the From range, in degrees (integer).
2. The 'ending' angle of the From range, in degrees (integer).
3. The direction of the From range. Either 'cw' (clockwise) or 'ccw'
(counter-clockwise).
4. The 'starting' angle of the To range, in degrees (integer).
5. The 'ending' angle of the To range, in degrees (integer).
6. The direction of the To range. Either 'cw' or 'ccw'.
Section 9.9.2: Whtmap Resources
The whtmap resource describes the state of the white remapping control.
There is one whtmap resource per state of the xv color editor controls.
The whtmap resource takes three parameters:
1. The hue to remap 'white' to, in degrees (integer).
2. The saturation to give to the remapped 'white', in percent (integer).
3. A boolean specifying whether the white remapping control is enabled.
If '1', the control is enabled. If '0', the control is disabled.
10/89 Page 47
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
Section 9.9.3: Satval Resource
The satval resource describes the value of the Saturation dial. There is
one satval resource per state. The satval resource takes a single
integer value, in the range +/-100, which specifies how much to add or
subtract to overall image color saturation.
Section 9.9.4: Graf Resources
The graf resources describe the state of the four 'graph' windows in the
xv color editor window (Intensity, Red, Green, and Blue). The graf
resources can be in one of two formats, 'gamma' and 'spline/line'.
In 'gamma' format, the graf resource takes two parameters:
1. The letter 'G', specifying 'gamma' mode
2. A single floating point number specifying the gamma value.
In 'spline/line' mode, the graf resource takes a variable number of
parameters:
1. The letter 'S' specifying 'spline' mode, or the letter 'L' specifying
'line' mode.
2. An integer number indicating the number of handles (control points)
that this graph window will have. (Must be in the range 2-16,
inclusive.)
3. For each handle, there will be a ':', and the x and y positions of the
handle, separated by a comma. The x and y positions can be in the
range 0-255 inclusive.
Section 9.9.5: Other Resources
Also, there are the boolean resources 'autoApply', 'displayMods', and
'autoReset', which control the initial settings of the three checkboxes
in the xv color editor window.
There are also boolean resources 'saveNormal', 'pspreview', and
'pscompress' which control the initial settings of the checkboxes in the
xv save and xv postscript windows.
LIMITATIONS
xv will NOT work on displays that aren't 1-, 2-, 4-, 6-, 8-, 16- 24-, or
32-bits deep. Luckily, that should still cover nearly every display out
there. It may not work on certain 6- or 24-bit displays.
It also only displays the first image in GIF files that have multiple
images in them.
Page 48 10/89
xv(l) Rev: 2.10(Feb 26, 1992) xv(l)
As for PM pictures, this program only displays 1-plane PM_I pictures, or
1-, 3-, or 4-plane PM_C pictures.
PM FORMAT
The PM format is a file format that we use at the GRASP Lab for our image
processing work. If you aren't at Penn, you are unlikely to ever run
into a PM-format file, so don't worry about it. Please ignore all
references to PM.
The 4-, 6-, 16-, 24-, and 32-bit code has not been extensively tested.
(A 4-bit MicroVax GPX system, a 6-bit HP 9000/320, a 16-bit Sony 3710,
and a 24-bit HP 9000/350, respectively. The 32-bit code hasn't actually
been tested at all.) You won't be able to do '-ncols 0' on a 6-, 16-,
24-, or 32-bit display, not that you should want to.
AUTHORS
John Bradley - bradley@cis.upenn.edu
GIF reading code based on gif2ras.c, by Patrick J. Naughton
(naughton@wind.sun.com)
GIF writing code essentially unchanged from code written by Michael
Maudlin (mlm@cs.cmu.edu).
SUN Rasterfile i/o code written by Dave Heath (heath@cs.jhu.edu)
JPEG interface code written by Markus Baur (s_baur@iravcl.ira.uka.de)
JPEG i/o code provided by the Independent JPEG Group.
TIFF i/o code and interface code written by Sam Leffler (sam@sgi.com)
Portions of 'ppmquant' snarfed for the '-best24' algorithm. ppmquant (and
the rest of the pbmplus package) was written by Jef Poskanzer.
(jef@well.sf.ca.us)
fsQuick code written and supplied by David B. Rosen (rosen@cns.bu.edu).
This code is a very fast implementation of Floyd-Steinberg dithering for
1-bit b/w displays.
10/89 Page 49