You may use the following statements, as you would in the C language:

if (expression) statement

if (expression) statement else statement

for (expression; expression; expression) statement

These statements allow you to pass information along to the user. The compiler prints any string you specify as it would any other internally generated messages.

info string-expression

warning string-expression

error string-expression

The info and warning statements allows compilation to proceed; the error statement causes the compiler to terminate before producing an output file and generates an error status code on exit.

---

Tip: You can build string expressions by using the + operator. Any numeric value expressions are converted to string expressions automatically, and expressions of quoted strings to be concatenated are perfectly legal.

---

Statements may be nested, with each statement successively narrowing the scope. For example, the statement within Example 7-1 has narrowed its scope to the second line of Vertical Back Porch of the first field.

within (field 1,1)
    within each (Vertical Back Porch)
        within (line 2,2)
             ...

In Example 7-1, the following are successively selected:

By successively narrowing the scope, the above figure has selected a time span that starts with the beginning of line 2 of Vertical Back Porch of field 1 and ends with the end of line 2 of Vertical Back Porch of field 1.

A nested scope can never exceed the bounds of a parenting scope. If a nested range expression specifies a range greater than the range of the parenting scope, the scope is simply limited to the parenting scope.

It is possible to narrow a range such that no time remains in the innermost scope. For example, see Example 7-2.

within each (Vertical Back Porch)
    within each (Vertical Front Porch)
        ...

Because Vertical Front Porch and Vertical Back Porch define mutually exclusive time spans within the frame, the innermost range specifies no time span at all and is not a valid scope.

When in the bounds of a board rules selection, three variables become active. Their values depend on the format as specified by the user and by previous signal definitions.

Each range expression as described below sets values uniquely, according to its rules and bounding scope.

Table 7-2. Range Expressions

Expression Syntax	Description	Start Time	End Time	Polarity
frame	The entire frame. This is provided as a convenience and is not strictly necessary because it is the default range. (See “Default Range”.)	Zero seconds	Duration of the frame.	Positive
field	The field of the frame. The user defines the number of frames in the field via the variable FieldsPerFrame. (See “The General Parameters Section”).	The beginning of Vertical Sync for this field.	The end of Vertical Front Porch for this field.	Positive
line	The whole line boundary. Line durations are defined by the user's format via this formula:     FPS is set by the user via FramesPerSecond and TLPF is set by the user via TotalLinesPerFrame. (These variables are described in “The General Parameters Section”.)	The beginning of the whole line (as calculated in Description).	The end of the whole line (as calculated in Description).	Positive
pixel	The whole pixel boundary. Pixel durations are defined by the user's format via this formula:     FPS is set by the user via FramesPerSecond, TLPF is set by the user via TotalLinesPerFrame, and TPPL is set by the user via TotalPixelsPerLine. (These variables are described in “The General Parameters Section”.) The start point of pixel is calculated from the beginning of the frame, not the beginning of the period; therefore, it is possible to have a nested pixel time range quantized to a point not coincident with its parent time range. The user directly defines the total number of pixels in the frame via this equation (using the previously defined variables):	The beginning of the whole pixel (as calculated in Description).	The beginning of the whole pixel (as calculated in Description).	Positive
Vertical Front Porch	The first of three portions of the vertical blanking interval. One vertical blanking interval exists for each field in the frame. The Vertical Front Porch for the first field may be defined following the active portion of the last field, yet still be associated with the first field. For formats that employ military sync (see definition in Vertical Sync), Vertical Front Porch has a duration equal to zero. There is one Vertical Front Porch in each field.	These choices are possible: 1) If the format, such as NTSC and PAL, has an intra-line pulse (that is, a pulse that does not fall at the beginning of a whole line from the beginning of the frame), time begins at the start of that pulse. 2) If the format has no intra-line pulse, the time at the beginning of the first line following the last active line of the frame.	The time at the leading edge of the vertical sync pulse. If this format contains more than one vertical sync pulse, the end of the Vertical Front Porch is the leading edge of the first pulse	Positive
Vertical Sync	The second of three portions of the vertical blanking interval. This is often considered the true beginning of the frame. A vertical sync pulse is defined to be a pulse that is longer than a horizontal blanking pulse (found on active lines). Three types of vertical sync patterns are possible: 1) Commercial sync (typical for broadcast) that contains some number of equalizing pulses preceding sync pulses (in Vertical Front Porch); Vertical Sync, with serration pulses; equalizing pulses following sync pulses (in Vertical Back Porch). 2) Block sync (such as the Silicon Graphics standard high-resolution format) with a single long pulse that may span several lines. See Chapter 2 for a discussion of block sync 3) Military sync, which has no vertical synchronization pulses and simply has a period of time without active lines. Each field contains exactly one Vertical Sync.	These choices are possible: 1) For commercial and block sync, the time of the leading edge of the first vertical sync pulse. 2) For military sync, the time of the first full line without active picture.	These choices are possible: 1) For commercial sync, the time of the beginning of the first whole line containing no vertical sync pulses (that is, long pulses). 2) For block sync, the beginning of the first line where the long sync pulse ends. 3) For military sync, the beginning of the first line containing active picture.	Negative
Vertical Back Porch	The third of three portions of the vertical blanking interval. For formats that employ military sync (see definition in Vertical Sync), Vertical Back Porch has duration equal zero.	The end of Vertical Sync (see Vertical Sync).	These choices are possible: 1) If the format contains intra-line pulses (see Vertical Front Porch), the time of the first active video. 2) If the format does not contain intra-line pulses, the beginning of the first line containing active picture.	Positive
Field Active	The active portion of this field. It is the time in the field exclusive of the vertical blanking interval. Each field contains exactly one Field Active portion.	The end of Vertical Back Porch (see Vertical Back Porch).	The beginning of Vertical Front Porch (see Vertical Front Porch).	Positive
Horizontal Front Porch	The first of three portions of the horizontal blanking interval containing the dead time preceding horizontal sync. Each line contains exactly one Horizontal Front Porch.	The end of active video in the previous line.	The beginning of the horizontal sync pulse.	Positive
Horizontal Sync	The second of three portions of the horizontal blanking interval defined as the duration of time while the horizontal sync pulse is low. The time marks the point of the 50% crossing of the pulse. Each line contains exactly one Horizontal Sync period.	The falling edge of the sync pulse (50% crossing).	The rising edge of the sync pulse (50% crossing).	Negative
Horizontal Back Porch	The last of three portions of the horizontal blanking interval, containing the dead time following horizontal sync. In composite formats, this is used for color burst. Each line contains exactly one Horizontal Back Porch period.	The end of the horizontal sync pulse.	The beginning of active video.	Positive
Line Active	The active pixels in the line.	The time of the first active element.	The time of the last active element.	Positive
edge-transition-direction edge of board-signal-name	The time of the specified edge. This directive allows you to specify a transition that has a direction of 1) rising (low to high); 2) falling (high to low); or any (matches both rising and falling). The transition refers to a change in the sense of a board signal. The transition must be defined prior to the execution of the range expression in which it is referenced; transitions are defined in: 1) the “set signal” statement (see “The set signal Statement for Transitions”); or 2) in one of the special user signals, defined when the user employs the compiled language (see “The User Signals for Reference”). The polarity corresponds to the sense of the transition. If the transition is a falling edge, the polarity is low; if the transition is rising, the polarity is high.	The time of the transition.	Same as start time.	Can be either. See Description
pulse-transition-direction pulse of board-signal-name	The time of the specified pulse. This directive allows you to specify a transition that has a direction of 1) positive (signal in high state); 2) negative (signal in low state). For a discussion of how transitions are defined, see edge-transition-direction edge of board-signal-name. The polarity corresponds to the sense of the pulse. If the pulse is positive-going, the polarity is positive; if the pulse is negative-going, the polarity is negative.	The time of the transition.	Same as start time	Can be either. See Description
time-expression	The specified duration. This allows you to specify an absolute duration as measured from the beginning of the frame. If you specify an iterative time value that does not comprise exact whole-number units of the frame, the last time expression will be truncated to the time at the end of the frame. Put more simply, you will never get an end time past the end of the frame, even if the whole unit of time would properly set it past the end of the frame.	Consider i to be the number of iterations of the time expressions of duration d. The start time is		Can be either. See Description

The default range is the name given to the range before any range selections are applied and is set to the entire frame. This is equivalent to setting the following range:

within each (frame) ...

Although not harmful, it is not necessary to use this construct.

Nested ranges always find the beginning and end of the whole unit for which they are defined. Examine this example, where Vertical Front Porch contains 10.5 lines and starts in the middle of a defined line of video. Watch what happens when the following statements are used:

within each (Vertical Front Porch)
    within each (line)
        ...

In the above example, the statement within each (Vertical Front Porch) would be repeated exactly 10 times, corresponding to the number of whole units (in this case, lines) contained within Vertical Front Porch. Because Vertical Front Porch begins on a half-line boundary, the first whole unit (line) begins half a line past the beginning of Vertical Front Porch.

In other words, nested ranges always round forward in time to the next whole unit.

The user sync signal is high when the signal is at blanking level, low when at sync level. Within the rules or the line-based language, you may refer to user sync.

The user blank signal is high when the signal is enabled (that is, the signal is high when blanking is enabled, low when blanking is disabled and pixels are displayed). Within the rules or the line-based language, you may refer to user blank.

The user blank signal aligns itself on whole lines, presuming a normal frame layout. This works correctly with the half lines of commercial sync and with the whole lines of block sync and military sync. If the source video format does not have its active area aligned with whole lines, you should not use the user blank signal; instead, reference the user sync signal directly (see “User Sync Signal”).

The user active signal is high when the signal is enabled (that is, high when pixels are displayed). The distinction between user blank and user active is on those lines when active takes only one-half the line (for example, first lines of one field of NTSC and PAL): on those half lines, the duration of user active is the full line, while the duration of user blank is half-line. With this difference between the signals, those hardware platforms that must always fetch an entire line when requesting pixels may use both signals to create the distinction; those that can request partial lines may do so by referencing only the blank signal. Within the rules or the line-based language, you may refer to user active.

These variables get their values from the attributes the user specifies for each field. The attributes and their definitions are in Table 5-3.

Each variable is an array with each element corresponding to a field. The array is zero-based, so the first field is in array element [0], the second in element [1], and so on. How many fields in the format? You can use the variable FieldsPerFrame (see Table 5-2) to find out.

These are useful only after frame analysis takes place, so you can use them in either rules definitions or post-processing operations.

It is often necessary to create a signal with the same shape as the sync signal provided by the user. Because the compiler provides the user sync signal as a predefined set of transitions in the database, you can easily copy the signal. You can read about the user sync signal in “User Sync Signal”.

within each (edge of user sync) {
    set signal "CSYNC" CorrespondingPolarity at BeginTime;
}

To copy user sync to a signal named CSYNC, this example uses the each edge range expression; for information, see edge-transition-direction edge of board-signal-name in Table 7-2. For each edge of the user sync signal in the database, the range expression causes within to execute the set signal statement. The CorrespondingPolarity value copies the polarity of the user sync signal, while BeginTime sets the transition to the same time as the user sync signal (see “How You Can Use Range Information” for information on these variables).

It is unusual to use another true signal with no changes at all (why would a hardware designer require two identical signals?). More often, some changes are needed to make a signal operate properly.

within each (rising edge of "IMPREG_LD") {
    set signal "XMAPREG_LD" high at BeginTime;
}
within each (falling edge of "IMPREG_LD") {
    set signal "XMAPREG_LD" low at BeginTime - 0.1 usec;
}
within each (positive pulse of "IMPREG_LD") {
    time pulseDuration;

    pulseDuration = (EndTime - BeginTime);

    if (pulseDuration < 1.0 usec)
        error "Insufficient IMPREG_LD pulse == " + pulseDuration;
}

The each rising edge range expression is similar to the copy operation shown in “Copying User Sync”, yet this expression discriminates by copying only half the edges (only the rising edges); for information, see edge-transition-direction edge of board-signal-name in Table 7-2. The time and polarity of XMAPREG_LD transitions match those of IMPREG_LD.

However, note the each falling edge range expression's set signal statement. The low transition of XMAPREG_LD is advanced by 0.1 usec; the new transition precedes the time of IMPREG_LD by that amount of time.

The last range expression, each positive pulse, performs error checking and has no set signal statement associated with it; see pulse-transition-direction pulse of board-signal-name in Table 7-2. This example shows checking the duration of a pulse (delimited by BeginTime and EndTime). If the pulse is too short, the compiler will issue an error message generated by the error statement; see “Diagnostic Statements”.

The example in “Copying Another Signal With Changes” showed calculation of a variable for detecting an error state. It is possible to calculate exported variables for later use as well (see “Variables”).

{
    exported double HBackPorchClampStart;
    exported double HBackPorchClampLength;

    within (field active 1,1) {
        within (line 2,2) {
            time LineStartTime;

            LineStartTime = BeginTime;

            within each (Horizontal Back Porch) {
                time tBackPorchStart;
                time tBackPorchLength;

                tBackPorchStart = BeginTime - LineStartTime + 2 clocks;

                HBackPorchClampLength = sec(HorizontalBackPorch)/2;
                HBackPorchClampStart = 
                    sec(tBackPorchStart) + (HBackPorchClampLength/2);
            }
        }
    }
}

The code in Example 7-5 shows the definition and calculation of two variables: HBackPorchClampStart and HBackPorchClampLength. You can see the declaration of the variables is the first item within a compound statement delimited by curly brackets ({ }). Syntax requires you to declare items only at the beginning of a block.

Example 7-5 shows use of nested within statements. The innermost loop is defined as the horizontal back porch area of second line of the first field's active region of this format.

The exported length variable is calculated by determining the duration (in seconds) of the back porch using the sec function. The exported start variable also uses the tBackPorchStart variable, the calculated time position of the relative start location of horizontal back porch within a line.

You can use if, for and while statements to alter control flow to execute different sections of code. You can see their use in Example 7-6.

{
    int nEdgeCnt = 0;

    /* check for serrations */
    within (Vertical Sync 1, 1) {
        within each (edge of user sync) {
            nEdgeCnt++;
        }
    }

    if (nEdgeCnt > 2) {
        /* we have serrations, so don't create transitions */
        signal "PLL_PHASE" initial state = high;
    } else {
        /* mask out vertical from the PLL's phase detector */
        signal "PLL_PHASE" initial state = low;
        within each (Vertical Sync) {
            set signal "PLL_PHASE" low  at BeginTime - 2.5H 
                ignoring wrap error;
            set signal "PLL_PHASE" high at EndTime + 2.5H
                ignoring wrap error;
        }
    }
}

The code in Example 7-6 performs different operations based on whether the user's sync signal contained serration pulses (see “The Vertical Interval”).

To determine whether the format has serrations, this code fragment counts the number of transitions (using the each edge range expression) in the first vertical sync region. Note that the variable nEdgeCnt is defined at the beginning of a compound statement because syntax requires variables be declared there; if this variable were not needed, the extra curly brackets ({ }) at the beginning and end of the example would probably not be needed.

If the format does have serrations, no signal transitions are defined—only the initial state is set. However, if the sync signal does have serration pulses, the two set signal statements are executed every vertical sync. The difference of 2.5 lines (2.5H) from the beginning and end of the vertical sync region puts the transitions beyond the beginning and end of the frame; therefore, the ignoring wrap error statement is needed to keep the compiler from issuing a warning for this circumstance (see “The set signal Statement for Transitions”).