should work in Firefox and Chrome. This assignment is due Monday August 7
For this next problem we will create a simple photo gallery which allows the user to rotate
through a number of images. Here’s some samples of the webpage in action:
Clicking on the right-hand side of the image will change the picture and the caption to the
next image on our list. When we reach the last image on our list, we will begin again with our
first image. Clicking on the left-hand side of the image will take the user to the previous
image on our list—when we get to the first image, we will wrap to the last image.
The image will always appear in the exact center of the window. When the user resizes the
window the image will be moved to the center of the window. The image will not change size
if the user increases the size of the window. Instead it will always remain centered at its
native 800×600 size. You may hard code in the image size. You may assume that the
window always remains larger than 800×600, so you do not need to worry about the case
where the window is smaller than the image.
The caption will always appear at the bottom of the image. Note that the caption is in white
with a green background #008800 and notice that the image shows through the caption box
because opacity of the caption style property is set to 0.5.
We will be using absolute placement to get things working. Remember that in spite of the
name “absolute”, placement of an item using “absolute” placement is actually relative to the
item it is contained within. This if a div A contains a div B, and both A and B are absolutely
positioned, the left, top, right, or bottom settings of B are actually relative to div A. If we
move div A, div B will also move.
While there are serveral ways to create a gallery with the behavior we want, probably the
most common is to overlay empty divs on top of the image. We will work with four different
We have a parent div which I will refer to as the photoSection. This div contains
everything associated with the image. This includes the actual <img> tag and the
other divs described below. Place this absolutely initially in the middle of the
window when the page loads. Change its position when the user resizes the window.
See below for more information on how to do this.
We have a caption div. The caption div will be a child of the photoSection. If you
absolute position the caption div, it will actually be placed relative to the parent div
in which it is contained. While typically we position an item using left and top, we
can also set its position using bottom. If you set bottom to 0px, the caption div will
be aligned at the bottom of the photoSection.
We will have left and right overlay divs on top of the <img> and the caption div.
You can also place these absolutely. Again as these are contained within
photoSection, they will move with photoSection even though they are absolutely
positioned. Set left to 0px for the left overlay and right to 0px for the right overlay.
Place onclick handlers on the left and right overlays which change the image and
caption either backward or forward as appropriate.
Handling Resizing of the Window
part of the assignment is to give you a bit of practice programming CSS positioned elements
CSS would not give you any practice.
You can cause a function to execute when the window is resized by adding a listener waiting
for the window’s resize event. For example:
will cause the web browser to call the handleResize function when the window is resized.
To get the size of the window use the following:
window.innerWidth and window.innerHeight
Once you get the new window size, use the techniques we learned in class (and found in the
Dynamic Contents handout) to change the location of the photoSection. Don’t forget when
you assign a new top, left, height, or width to an element’s style properties you must include a
“px” at the end of the string.
elem.style.width = 5; // wrong, wrong, wrong
elem.style.width = “5px”; // right
Provided for You
For this part of the assignment I provide you with 6 jpeg images (each 800×600 pixels) and an
HTML file. The HTML file has the correct div structures, but does not do any of the
information you’ll need for the images (including the captions).
There is one important item that is missing from our implementation that you should add,
should you decide to use this on a real website. There is no indication of how this webpage
works. In HCI (Human-Computer Interaction) parlance there is no “affordance” indicating
how someone might manipulate this webpage. If you use this on an actual website you might
consider adding permanently visible right and left buttons (these could go on the caption,
below the entire photoSection, or just outside of the photoSection on the right and left). An
even fancier way to do this is to add right and left indicators on top of the left and right
previous/next image overlays. You can take advantage of the onmouseover and onmouseout
events so that these overlay indicators are only visible when the mouse is moved on top of
their corresponding left or right overlay div.
We will now create a Stanford-specific version of Google Maps. Make sure you’ve
completed the gallery part of the assignment first. It’s much easier and is designed to give
you some practice with techniques you’ll need for Stanford Maps. Here is a screenshot:
Your webpage should support the following functionality:
The map should be enclosed within a map frame (as shown above). The size, color, and
style of the frame is up to you.
Allow the user to click and drag on the map to change the map area which is currently
visible within the map frame.
Set the cursor to the move cursor when the user begins a drag operation. Return the
cursor back to the standard arrow cursor when the drag operation is complete. You can
set the cursor using the cursor style property.1
When the user double-clicks on the map move the point double-clicked to the center of
the view area.
The map frame should resize as the window is resized. As the window is enlarged, the
map area should enlarge, as the window size is reduced the map area should be reduced.
The map frame should maintain fixed margins on all sides. These margins should be
maintained as the window is resized. The exact margin sizes are up to you—choose
something which you find aesthetically pleasing.
The map should support zooming in and out. As the map zooms, the point in the center
of the view area should stay fixed. In other words, if the map is centered on the Gates
Computer Science building, the Gates building should stay in the center regardless of the
zoom level. We’ll discuss how zooming works in more detail below.
Provide controls to scroll left, right, up, and down. Left or right scrolling should cause
the window to scroll 1/2 of the total width currently visible. Scrolling up or down should
cause the window to scroll 1/2 of the total height currently visible.
To keep things simple, if the user scrolls off the edges of the map it’s okay to go ahead
and let them keep scrolling even if it means you can no longer see the map.
Creating the Map and Map Frame
The map itself is represented by a GIF file. Create a <div> for the map frame. Put an <img>
tag for the map within the map frame’s div. Set the map frame’s overflow style property to
hidden. The overflow property controls what happens when the contents of an element do not
fit within the containing element. Setting the overflow property to hidden tells the web
browser to hide any sections of the contents which overflow the map frame.
Resizing the Map Frame
Use the same technique used for the photo gallery to execute code when the window is
resized. In this case you’ll want to resize the map frame. You may assume that the user never
resizes the window so small that there is not space for the map, map frame, and navigation
Dragging the Map
You should be able support dragging the map around via careful use of mousedown,
mousemove, and mouseup events. However, a couple of points will be helpful.
These next two paragraphs are very important. If you ignore these you may be in for great
frustration. In your mousedown handler if you want to move the map, you must call
preventDefault() on the event object. Otherwise the standard default behavior supporting a
drag and drop to a separate window will run. You do not want this to happen.
1 You’ll only have complete control the cursor while it’s over your map or map frame. When the mouse
moves over other HTML elements, they will temporarily override the cursor based on their own style
settings, even if you set the cursor on the <body> tag. If you experiment with Google Maps, you’ll
discover this is the same behavior they have.
Place your event handlers on the document, not map (or whatever specific element you are
trying to drag). Placing the handler on the actual element can cause problems if the user
moves the mouse faster than you can move the element (if the mouse moves outside of the
element you will lose mousemove and mouseup events). In theory you can place the handlers
on the <body> tag. However, in my experience you’re better off putting it on the document
itself. You can do this programmatically like this:
As previously noted, you need to provide controls for zooming in and out and
for scrolling in each of the four cardinal directions (up, down, left, right).
The easiest way to provide controls is simply creating push buttons using
<input> tags inside an HTML <form>. You can see a closeup of my controls
If you want to get fancier, you can replace the buttons with images (or use the
<button> tag to combine a pushbutton with an image). If you’re feeling more
ambitious you can move the navigation buttons onto the map (actually this is not that
difficult). You can also create a zoom slider control as on Google Maps.
We support zooming by having four separate GIFs—“map-xl.gif”, “map-l.gif”, “map-m.gif”,
and “map-s.gif”. We swap back and forth between the GIFs as the user zooms in and out.
You’ll need to do some careful arithmetic combined with setting the position of the GIFs to
make sure that the same map point stays in the middle of the viewing area as the user zooms.
Don’t forget to preload the map images for smooth operation when the user first starts
If you want to get some extra practice, here are some things you can try adding:
Add a text field and allow the user to enter in building names. Center on the building or
if you want to get fancy add a flag or map pin.
The default application can instantly jump from one location to another in response to
double clicks or scroll requests. A more gradual scrolling may help the user by allowing
them to see the relationship between the original location and the new. Support smooth
scrolling when the user double clicks or clicks on a scroll button.
Our map is from the Stanford Maps and Records Department. The original Map was from: