Encoding Chapter 12: Video 339 Customizing Settings If you’re not content with a preset, or if you have additional needs (such as resizing the source), you can click on the Settings button to customize the settings. (You can even save your customizations as a preset of your own.) Figure 12-4 shows the Settings interface. The upper left quadrant allows you to see various sizes of the source material, as well as crop the source prior to encoding. The bottom left quadrant allows you to create cue points that will be embedded in the video during encoding. Between those areas is the video timeline, which you can use to preview the video and set in and out points if you wish to compress only a portion of the video. Figure 12-4. The Settings interface of Adobe Media Encoder (CS5 pictured) The upper right quadrant is a general settings area that allows you to choose the file format, encoding preset, and output directory, as well as show a sum- mary of the settings applied by the chosen preset. The lower right quadrant contains more specific settings including video and audio encoding options, and the ability to apply a blur during encoding. Download from Wow! eBook <www.wowebook.com> Part IV: Sound and Video 340 Components For more information about Adobe Media Encoder, see the Using Adobe Media Encoder CS5 resource at http://help.adobe.com/en_US/mediaencoder/ cs/using/index.html. Information about embedding cue points during encod- ing can be found in the “Encoding and exporting” section of this resource. The companion website has additional information about creating cue points—both during encoding and at runtime through ActionScript when using the FLVPlayback component. Starting the Queue Once a preset is selected, all you need to do is press the Start Queue button. Adobe Media Encoder will encode the file and save it in the location specified in the Output File column. (The default location for the output is the same directory in which the source file resides.) Components Components offer designers and coders alike a chance to speed up the development process by using precreated widgets. Components usually combine ActionScript and assets to make it easier to achieve a specific goal. Components can be dropped onto the stage like a movie clip or button sym- bol and often function with little or no intervention. When effort is required, most components can be configured using the Flash Professional CS5 Properties panel or Flash Professional CS3 or CS4 Components Inspector panel. Most components can also be manipulated with ActionScript, which is what we’ll focus on in this section. Before working with any component, however, Flash Professional users must place the component in the library of the FLA file that will compile to SWF. Simply drag any component from the Components panel to the Library panel, or drag it to the stage and delete it immediately. In this chapter, we’ll work with three different components. First, we’ll add the FLVPlayback component as a prefabricated video player. Next, we’ll add captioning support to the player with the FLVPlaybackCaptioning component. Finally, we’ll add a Button component to satisfy a simple user interface need. Working with the FLVPlayback Component The fastest way to add video to your ActionScript application is by using the FLVPlayback component (Figure 12.5). The component is available in two fla- vors. FLVPlayback is available to Flash Professional users of version CS3 and later, and FLVPlayback 2.5 was introduced with version Flash Professional CS4 and is also available for Flex. N O T E By default, Adobe Media Encoder will start processing the assets in the encod- ing queue after two minutes of idle time. This behavior can be adjusted in the application preferences. Download from Wow! eBook <www.wowebook.com> Components Chapter 12: Video 341 Figure 12-5. The FLVPlayback component simplifies adding video to projects In addition to Flex compatibility, FLVPlayback 2.5 was designed to take advantage of features introduced in version 3.5 of Flash Media Interactive Server—Adobe’s streaming media and real-time communication server software. It improves performance for video on demand and live streaming and supports live DVR functionality (pausing, rewinding, and recording live streams) introduced in FMS 3.5. If you prefer to avoid components—perhaps because you want to design your own player interface, or because components increase the size of your SWF (the FLVPlayback component contributes between 50k and 65k)—we’ll show you how to play video entirely with ActionScript shortly. If you’re open to the use of components, however, FLVPlayback has a few useful benefits. First, you can pick from several preconfigured controllers, or skins, or you can use the component without a skin and create your own custom control- ler. This lets the component handle all the heavy lifting in the video display area, but allows you to control playback with your own code and your own design. (We’ll show you a very simple implementation of this approach later in the chapter.) More importantly, the code in the FLVPlayback component takes care of some important behind-the-scenes tasks that you would have to recreate. For example, the component will automatically determine if you’re using a streaming server by parsing the URL of the video source. If so, it will then N O T E You may notice in further research that Flash Media Interactive Server is typically abbreviated as FMS. This is because “interactive” was added to the product’s name only in recent versions. Download from Wow! eBook <www.wowebook.com> Part IV: Sound and Video 342 Components handle the necessary initial communication with the streaming server for you so you don’t have to script those connections yourself. We advise starting out with the FLVPlayback component, even if you choose to create your own controller. Then you can move on to coding your own player to replace the component after you’re comfortable with the relevant classes. Scripting the component The following example, found in the video_comp.fla source file, demonstrates the minimum code necessary to play a video. 1 import fl.video.FLVPlayback; 2 3 var vid:FLVPlayback = new FLVPlayback(); 4 vid.source = "nero.flv"; 5 addChild(vid); Note in line 1 that the FLVPlayback class must be imported even in the time- line because it’s not part of the flash package. (Most component classes are found in the fl package and are not automatically part of the Flash Player to keep the player size small.) Line 3 types the instance variable and instantiates the component. Line 4 populates the source property (telling the component which video to play), and line 5 adds the instance to the display list. Skinning the component To add controls simply, we can use a skin that ships with the component. FLVPlayback skins are external SWFs that are loaded at runtime. Therefore, to add a skin with ActionScript, you must know the path to the skin. Fortunately, Flash Professional users can take advantage of the fact that Flash will move your chosen skin to the same directory as your SWF when you test your file. To choose a skin, save your FLA, or create a new temporary file and save that, to the directory you’re using for your project. Temporarily drag the component from your file’s library to the stage and select it. Flash Professional CS5 users can then look in the Component Parameters section of the Properties panel (shown in Figure 12-6) to customize the component. Flash Professional CS3 and CS4 users will need to open the Component Inspector panel to see the same content. Next, click on the UI element next to the skin option. (In Flash Professional CS5 there’s a pencil button, while other versions show a magnifying glass button after clicking on the field.) This will open a dialog box that allows you to preview all the available skins, collected into groups that display the con- troller under your video or over your video. Flash Professional CS5 users will also see an additional grouping of new skins called Minima. You can choose which functionality to include in your controller by looking at the name of the skin and previewing its appearance when displayed in the dialog box. N O T E If you test the video_comp.fla file, you may wonder how it can play without any instruction to do so. The component has an autoplay property that’s set to true by default. Figure 12-6. The Component Parameters section of the Properties panel (CS5 pictured) N O T E Users of Flex Framework authoring tools can download the component, skins, sample files, component source, and several additional related tools at http://www.adobe.com/products/flash- mediaserver/tool_downloads/. Download from Wow! eBook <www.wowebook.com> Full-Screen Video Chapter 12: Video 343 For the following exercise, found in the video_comp_skin.fla source file, choose the SkinUnderAllNoFullNoCaption skin and test your movie. Don’t worry about the fact that it won’t work. After all, you didn’t select a video source. All that matters is that Flash copies the skin to the same directory in which you published your SWF. You should see SkinUnderAllNoFullNoCaption. swf in that directory. (If not, be sure you save the FLA you were using for this task and retest.) Once you have your skin in place, you can remove the FLVPlayback component from the stage, or discard any temporary file you created. Once the skin is in place, all that remains is to add one or more of the fol- lowing three lines to your existing script. Line 6 specifies your skin choice, and the optional lines 7 and 8 specify the color and alpha of the skin. Now, when you test your movie, you’ll see a skin that you can use to control video playback. 6 vid.skin = "SkinUnderAllNoFullNoCaption.swf"; 7 vid.skinBackgroundColor = 0x003366; 8 vid.skinBackgroundAlpha = 0.75; If you don’t want to store the skin file in the same directory as your main SWF (for example, if you want to store multiple skins in a directory), you can specify another path for the skin property. Also, remember that the skin you see is an external SWF that’s loaded at runtime. Therefore, just like the video file, the skin must be deployed with your main SWF and HTML files. Full-Screen Video One of the most entertaining Flash video features is true full-screen video— video that occupies the entire screen, rather than a maximized browser or player window, hiding other computer operating system interface elements for a fully immersive experience. Both the FLVPlayback component and pure ActionScript can launch into full-screen mode, both of which we’ll cover. Before we get to implementation, however, we need to cover two preliminary steps. The first step is to start with optimal source material for final assets. This includes the highest quality source, the largest size your interface will allow, and careful attention during encoding. Beyond those common sense sug- gestions, you’ll probably want to experiment with such encoding options as different bitrates and deinterlacing your content if you’re using a DV source. Deinterlacing is the process of converting the two fields of a DV source (which are like video frames but each contain half the horizontal lines and are dis- played twice as fast) into the frames used by the FLV format. One common artifact that is more pronounced when working with interlaced source mate- rial is jagged lines visible along sharp edges in your videos. Deinterlacing the source during encoding significantly reduces this effect. Download from Wow! eBook <www.wowebook.com> Part IV: Sound and Video 344 Captions The second step is to instruct Flash Player to allow the switch to full-screen display. If you think about it for a moment, you certainly don’t want the deci- sion to switch to full-screen mode left in the hands of content creators. If that were the case, every Flash advertisement would take over your screen, leaving you no control. Instead, the developer must make the feature possible, and the user must be responsible for switching back and forth between normal and full-screen modes. To enable the feature, you must add the allowFullScreen parameter, with a value of true, to the file’s host HTML file. One way to do this is to add this parameter manually to the object and embed tags, as seen in the following excerpt. <object> <param name="allowFullScreen" value="true" /> <embed allowfullscreen="true" /> </object> Flash Professional users can also use the quick and easy solution (par- ticularly handy during testing) of choosing the “Flash Only – Allow Full Screen” publishing template in the Publish Settings dialog (File →Publish Settings →HTML→Template). After adding support for full-screen video in your HTML host file, you’re ready to enable the full-screen button in the FLVPlayback component. To do so, choose any skin that supports full screen, such as SkinUnderAll or SkinOverPlayFullscreen, to cite two examples. These and other skins add the Full Screen button shown in Figure 12-7. The following change to line 6 of the previous example, found in the video_ comp_skin_full.fla source file, changes the skin to one that supports full screen mode. 6 vid.skin = "SkinUnderAllNoCaption.swf"; Once you have a video and have supported full screen mode in your host HTML file and skin, you can test your file in a browser. Full screen mode will not work when testing within Flash Professional, so Flash users can select the default Publish Preview command, File →Publish Preview→HTML. Pressing the Full Screen button in the skin will switch to full-screen mode, and you can press the Escape key to return to normal mode. Later in this chapter, we’ll show you how to add full-screen playback using your own ActionScript. Captions Captions, also referred to in some contexts as subtitles, consist of text that is displayed synchronously during video playback. Captions are useful for providing alternate language tracks to bring your video to a wider audience. Captions are also appreciated by the deaf and hearing impaired, as they provide Figure 12-7. The Full Screen button used by select FLVPlayback skins (color and alpha may differ) N O T E If HTML is not available for Flash Professional users, go to the File →Publish Settings menu dialog and add HTML as a publishable format. Download from Wow! eBook <www.wowebook.com> Captions Chapter 12: Video 345 a much needed accessible alternative for audio tracks when it comes to dialog and descriptive audio services. Captions help satisfy requirements imposed by the United States Rehabilitation Act of 1973, Section 508, which establishes accessibility man- dates for content developed for government use, or financed by federal funds. Many private entities, particularly those serving the educational markets, also require accessible content. As the demand for this requirement increases, cap- tions will play an increasingly more important role in digital video. Using the FLVPlaybackCaptioning Component Flash supports captioning via the FLVPlaybackCaptioning component, when used in conjunction with the FLVPlayback component. Adding the FLVPlaybackCaptioning component to the stage at authoring time, or dynamically at runtime with ActionScript, opens the door for caption use. The simplest way to display captions is to use the FLVPlayback component. In fact, with only one FLVPlayback instance on the stage the captioning component will automatically detect the playback component, and use its internal text element for caption display. You can also manually specify any FLVPlayback component as the target for the captions (in case you require more than one at any given time), or even your own target for the captions (in the event that you want to use another text element—perhaps integrated into your interface, rather than the video). To use the FLVPlayback, you’ll need to choose any skin that supports cap- tions, such as SkinUnderAll or SkinOverPlayCaption, among others. These skins feature the Captions button shown in Figure 12-8. The following edit to line 6 of the previous example, found in the video_ comp_skin_full_captions.fla source file, uses a skin that supports all skinned features, including captions. 6 vid.skin = "SkinUnderAll.swf"; Once the FLVPlayback component is configured to display captions, we must add the FLVPlaybackCaptioning component to the stage. The following code continues the example first by importing the component class in line 10, and instantiating the component in line 11. Line 12 assigns the caption file for loading at runtime (which we’ll discuss in a moment), and line 13 adds the component to the display list. 10 import fl.video.FLVPlaybackCaptioning; 11 var cap:FLVPlaybackCaptioning = new FLVPlaybackCaptioning(); 12 cap.source = "nero_timed_text.xml"; 13 addChild(cap); Note that we’re not placing the component at a particular location on the stage. Although it appears as a small rectangle when dragged to the stage in Figure 12-8. The Captions button used by select FLVPlayback skins (color and alpha may differ) N O T E As with the FLVPlayback component, Flash Professional users must have the component in the library of their FLA to instantiate it with ActionScript. See the “Working with the FLVPlayback Component” section of this chapter for more information. Download from Wow! eBook <www.wowebook.com> Part IV: Sound and Video 346 Captions authoring mode, this is merely to simplify selecting the component. At run- time, it will be invisible, so its position is irrelevant. Now both components are ready to display captions, so we need to create the caption file. You can create a captioned video in two ways. You can embed the caption data in the video using cue points. Embedding means they’ll always be with the video, but it also means that you have to reencode the video just to edit the text. A far more flexible option is to load a caption file at runtime. This approach also allows you to switch caption files dynamically—ideal for offering subtitles in multiple languages, a task we’ll look at later in the chap- ter. First, however, we need to know how to format the captions. Creating Captions with Timed Text To create a caption file to load at runtime, you need to write an XML (Extensible Markup Langauge, discussed in Chapter 14) file using the World Wide Web Consortium (W3C) Timed Text Markup Language (TTML or, familiarly, TT)—also sometimes referred to by its format name, Distribution Format Exchange Profile (DFXP). We’ll cover a portion of Timed Text features here, but you can learn more about the language by visiting the W3C page at http://www.w3.org/AudioVideo/TT/. More importantly, you can learn about the subset of features supported by the FLVPlaybackCaptioning component from Adobe’s ActionScript 3.0 Language and Components Reference at http:// www.adobe.com/livedocs/flash/9.0/ActionScriptLangRefV3/TimedTextTags. html. Several tools can create Timed Text files, including the pair listed in the adjacent note. However, you can also write your own Timed Text files. The example XML that follows is an edited excerpt of the nero_timed_text.xml source file provided in this chapter’s source archive. (For brevity, two cap- tions are shown and minor edits have been made to use all features from the source file.) 1 <?xml version="1.0" encoding="UTF-8"?> 2 <tt xmlns="http://www.w3.org/2006/04/ttaf1" 3 xmlns:tts="http://www.w3.org/2006/04/ttaf1#styling"> 4 <head> 5 <styling> 6 <style id="1" 7 tts:textAlign="center" 8 tts:fontFamily="_sans" 9 tts:fontSize="18" 10 tts:fontWeight="bold" 11 tts:color="#FFFF00FF" /> 12 <style id="2" tts:backgroundColor="#00000000" /> 13 <style id="3" tts:backgroundColor="#000000FF" /> 14 <style id="trans" style="1 2" /> 15 <style id="opaq" style="1 3" /> 16 </styling> 17 </head> 18 <body> 19 <div> 20 <p begin="00:00:05.00" dur="00:00:04.00" style="opaq"> N O T E MAGpie is a free captioning tool devel- oped by accessibility leaders at the National Center for Accessible Media (NCAM). For more information, see http://ncam.wgbh.org/invent_build/ web_multimedia/tools-guidelines/mag- pie. You can find the Manitu Group’s Captionate at http://www.captionate. com, and Adobe’s Flash Developer Center features a tutorial on using Captionate with the FLVPlayback and FLVPlaybackCaptioning components (http://www.adobe.com/devnet/flash/ articles/video_captionate.html). Download from Wow! eBook <www.wowebook.com> Captions Chapter 12: Video 347 21 Nero is a Lionfish<br /> (<span tts:fontStyle="italic"> 22 Pterois volitans</span>), 23 </p> 24 <p begin="00:00:09.00" dur="00:00:02.00" style="trans"> 25 in his reef aquarium. 26 </p> 27 </div> 28 </body> 29 </tt> We’ll discuss custom XML solutions in Chapter 14, but Timed Text is a predefined format so conforming to its specification is pretty straightfor- ward. We’ll occasionally point out things that we’ll cover in greater detail in Chapter 14, but you should feel comfortable simply editing an existing Timed Text file until you gain a little experience with XML. Lines 1 through 3 include two default tags used to validate the file. The first tag (also called a node) is <?xml ?> and is the XML declaration tag. We’ll discuss this in Chapter 14 but, essentially, it declares the version of XML in use and the character encoding used when writing the document. The second tag, <tt>, is the document’s root node. All XML documents must have a root node that encloses all other nodes, and we’ll discuss this further in Chapter 14, as well. Be sure to see the accompanying note describing the use of attributes in this tag. N O T E The ActionScript 3.0 Language and Components Reference entry “Timed Text Tags”, found at http://www.adobe.com/livedocs/flash/9.0/ActionScriptLangRefV3/ TimedTextTags.html, specifies that all attributes of the <tt> tag are ignored. However, this is not the case if you style your captions. If you omit the xmlns attri- bute, your captions will not be styled, and if you omit the xmlns:tts attribute, the use of the tts namespace in styles will result in errors. When using styles, consider both of these attributes required. A <head> tag (spanning lines 4 through 17) is optional, but we recommend its use because it makes styling your captions much easier. Within it, a <styl- ing> tag (spanning lines 5 through 16) is also optional but necessary if you intend to create styles. Styles are Cascading Style Sheet (CSS) entities for the Timed Text document and are itemized in lines 6 through 15. You can have as many styles as you like, but each must have a unique id attribute. The style attributes that are actually responsible for the formatting are very similar to CSS properties, but are preceded by the tts: prefix. It’s possible to assign multiple styles directly by their alphanumeric id, but it’s also possible to manage formatting efficiently by creating new styles con- sisting of other styles. Take a look at the styles in our example. We wanted to achieve two looks for our captions: one with a black background, for use over light areas of video, and one with a transparent background, to allow more of the video to show through the text. N O T E Character encoding just maps text characters to specific codes (usually numeric), so that software responsible for parsing the text know which char- acter to use based on a given code. It’s a way of bringing platform, hardware, and software neutrality to the process of rendering text. We recommend using UTF-8, which includes a wide range of characters, such as those used in differ- ent languages around the world. For more information about character encoding, see http://en.wikipedia.org/ wiki/Character_encoding. For more information about UTF-8, see http:// en.wikipedia.org/wiki/UTF-8. N O T E Be sure to consult the “Timed Text Tags” ActionScript 3.0 Language and Components Reference resource, men- tioned earlier in this section, for a com- plete list of supported and unsupported properties. Here are a few noteworthy mentions: • fontFamily supports device fonts, as seen in our example. • fontSize supports only the first size found; supports absolute and rela- tive sizes but not percentages. • li neHeight, pa dding, and over- flow , although potentially useful for captions, are among several options that are not supported. Download from Wow! eBook <www.wowebook.com> Part IV: Sound and Video 348 Captions Style 1 consists of all styling attributes common to both treatments, which means that the background alpha information appears in other styles. Styles 2 and 3 itemize only the background color and specify transparent and opaque, respectively. The Timed Text format uses #RRGGBBAA color nota- tion, where AA is alpha. However, the ActionScript components support only opaque and transparent settings. All zeros will be seen as transparent, but any value other than zero will be opaque. We’ve used the opposite of zero for alpha, FF, to remind us that this is opaque. The resulting value of #000000FF is, therefore, an opaque black background. Once you’ve created these individual styles, you can then apply more than one style at a time. It’s possible to do so at the caption level by using syntax like id="1 2", but it’s also possible to create a new style the same way. For example, you can create a new style combining styles 1 and 2 and, because the style names can be alphanumeric, you can give it a descriptive name. We’ve done this in lines 14 and 15, specifying that trans is centered, sans-serif, 18-point, bold, yellow text on a transparent (because it uses styles 1 and 2), and opaq shares the same font attributes but is atop an opaque background because it uses styles 1 and 3. A <body> tag (lines 18 through 28) is required and is used to hold all the caption data. Within the body tag, one <div> tag (lines 19 through 27) is required, and paragraph tags <p> are required for each line of caption (lines 20 through 26). The ActionScript documentation doesn’t say that <div> is required but nei- ther <p> nor <span> tags can appear in the <body> tag. Similarly, the documen- tation says zero or more paragraph tags are supported, but we didn’t find a logical way of applying time or style attributes to individual captions without them. For example, <span> tags (lines 21 and 22) are supported, but not in the <body> or <div> tags. Therefore, we suggest you consider <div> and <p> tags required. For each caption (in our case, in each <p> tag), a begin attribute is required to set the time of the caption. The attributes dur (duration) and end (the time at which the caption should end) are optional. If omitted, the caption will remain onscreen until the next caption appears. Time can be specified in full clock format (HH:MM:SS.m, where m is milliseconds), partial clock format (MM:SS.m or SS.m), or offset time (with units, such as “1s” for one second). Frames are not supported as a measure of time. Now that you know how to create a Timed Text file, you can run the previous source file, video_comp_skin_full_captions.fla, (discussed in the “Captions” section of this chapter) which makes use of the nero_timed_text.xml caption source file. N O T E If you’ve spent some time with Chapter 9 in this book, you may recall that the color notation that included alpha was specified as 0xAARRGGBB. The dif- ference between this BitmapData color notation, and the #RRGGBBAA used with Timed Text, can lead to confusion. If you see an unpredictable color behind your caption text, check to see if you’ve used the wrong format. N O T E In our main Timed Text example, we used full clock format for clarity and consistency, even when the duration matched the time at which the next caption appeared. However, you can simplify this by using partial clock for- mat, and omitting any duration or end attributes when the caption is to remain on screen until replaced. As an illustra- tion, we have formatted our Spanish- language example this way, which we’ll discuss shortly. Download from Wow! eBook <www.wowebook.com> . <body> 19 <div> 20 < ;p begin=" ;00 :00 :05 .00 " dur=" ;00 :00 :04 .00 " style="opaq"> N O T E MAGpie is a free captioning tool devel- oped by accessibility leaders. features supported by the FLVPlaybackCaptioning component from Adobe’s ActionScript 3. 0 Language and Components Reference at http:// www.adobe.com/livedocs/flash/9 .0/ ActionScriptLangRefV3/TimedTextTags. html <www.wowebook.com> Captions Chapter 12: Video 34 7 21 Nero is a Lionfish<br /> (<span tts:fontStyle="italic"> 22 Pterois volitans</span>), 23 < /p& gt; 24 < ;p begin=" ;00 :00 :09 .00 "