Die Rolling Program
Help Files (old)
We would like to provide a "die rolling" program for Reality Fault
that has a number of features:
- Algebraic-style operation and precedence
- Multiple die rolls on a single command line
- Gameing-specific die types, tests, and results
- Easy, per-environment configuration
- Easy customization of operators, precedence, and die-types
The components of this program should be organized in a fashion
that makes it easy to reuse those components in other programs.
- Check input for conditions that provide guidance for the major
mode of operation, i.e. roll dice, configure environment, report
- Check the input for obvious, "non-action" requests:
#help, #version, bad flags, etc.
- For each argument:
- Make sure each is a valid dSPEC
- Execute the dSPEC
- Format the result
- Report the result to the correct people
Should each dSPEC be checked for validity before any are executed?
Are any results reported before each dSPEC is validated?
When/where should conflicting die-operators be detected, and
what should be done if conflicting die-operators are present in
the same dSPEC?
Other things the library should be good for
These arguments can be part of a list of dSPEC arguments
#v, #!v turn on/off verbose reporting.
#q, #!q turn on/off quiet reporting
#t, #!t turn on/off testing
read/write configuration information
display results (a default reporting function)
execute? (the basic "please turn this dSPEC into a result")
Not actually operators, per se, these tokens are used to
over-ride precedence. Frex:
- Standard Math
+ - / * %
The only one of these that may not be 100% familiarly to one
and all is the modulo operator. This returns the remainder of a
division operation. Frex:
10%3="three with one left over"=1
< > =
These tokens are used to compare the values.
Frex, 1<2 returns 1, 1>2 returns 0, 1=0 returns "What were you
d a s f
Return the sum of X numbers, each from 1 to Y (default 1d6)
Return the sum of X numbers, skewed toward the "average" on
the die range from:
(1+A)/2 to (Y+A)/2
(Are we worried about rounding in any of this? Solving
this for d6 gives a flat range from 2.25 to 4.75.)
This formula preserves the average, and eliminates the
possiblity of rolls at either end of the nominal range. --
For a d6 (1-6, avg=3.5) the range becomes (2-5, avg=3.5).
I will point out that the formula we are using does not
exactly simulate the d6 "averaging die" we were using as an
example. In our formula, each of the numbers from 2-5
occurs equally often. On the averaging die, a 3 or 4 occurs
twice as often as a 2 or 5. I don't think this is important
for our needs, but it is something to be aware of.
Returns a list of X numbers, each from 1 to Y (default 1d6)
This default isn't really useful as a sequence, is it?
Unless someone can suggest a useful default sequence, I'm
not inclined to change it, though.
Mostly we were thinking these operator would be used with the
Boolean operators for rolling "Storyteller" type successes.
Frex, trying to roll 5d10, examining each to see if it is
greater than or equal to a target number of 7, and reporting
the number of rolls that met the criteria might be 5x[d10>7]
- [ ]
Used to specify a dSPEC as an argument to another operator.
Frex, 2d3d would be "Roll 2d3. Roll that many d6 and read
the total." 2d[3d] would be "roll 3d6. Roll 2 dice of that
2d[3d] -> 2d -> 2d10 -> 11
10d[4d3] -> 10d -> 10d8 -> 45
Returns X rolls of a die of Y sides. Frex, 3x6 would expand
to: (1d6 1d6 1d6), which is no different than 3d6. Again,
the non-useful default would be 1x6.
What would 3x6+1 return? 1d6+1 three times? (Hm. No.)
(1d6 1d6 1d6+1) Maybe. (1d6 1d6 1d6 +1) Probably.
Does this operator essentially expand an expression into a
dSPEC, determine the results and then the total?
d10>7 d10>7 d10>7 d10>7 d10>7 ->
5>7 4>7 6>7 3>7 7>7 ->
0 0 0 0 1 ->
This would mean that 5x[d10>7]+1 (given the same rolls)
would be 2. So, 3x6+1 ->
[d6 d6 d6]+1
[3 4 5]+1
n k i
Send output to various users, in various formats. (lib-argo?)
Command Line Processing
LSEDIT "Array" Library
roll program parses, hands to HSR wrapper
decided that 5 DC = 1.5d6
rolls 1d6+1d3, gets BODY
rolls 1d6-1+1, gets STUN MULTIPLE
"TOTAL_STUN TOTAL_BODY" "BODY_ON_D6 BODY_ON_D3 MULT"
Rolling "Target Numbers" Samples
The GM says:
Roll 5d versus a target of 7
Using only simplest operators
roll d10 d10 d10 d10 d10
Any result of 7+ is a success.
Using "Boolean" operators
Reported results are the number of successes.
Using a "Repeat" operator
Reported results are the number of successes.
Using a "Storyteller" wrapper
roll 5d vs 7
roll 5 vs 7
roll 5v <- Wrapper defined default of 7
Using a "Property Parsing" roller