Input Validation Rules: DATE, TIME, PATTERN & Filters

Concise guide to Sheerpower's VALID() rules for dates, times, expressions, menus, routines, and powerful FILTER pipelines with RESTORE for multi-step validation.

What You'll Learn

Sheerpower's VALID(text$, vrules$) evaluates input against readable, composable rules like date dmony, required, pattern, and more. You can also pre-process text with filter steps (trim, case-change, replace/remove, etc.) and use restore to validate both the cleaned and the original input in one pass.

Quick Reference

INTEGER / NUMBER / DIGITS / LETTERS / UCASE / LCASE — what the text is made of
ALLOW lo TO hi / ALLOW a, b, c — the values allowed: numbers after integer or number, dates after date, otherwise text
DISALLOW — the same forms, refused
LENGTH n / MINLENGTH n / MAXLENGTH n — exactly, at least, at most n characters
DECIMALS n — at most n decimal places (after number)
DATE DMONY — DD-Mon-YY (e.g., 01-Jan-99)
DATE DMONCY — DD-Mon-CCYY (e.g., 01-Jan-2000)
FULLTIME — CCYYMMDD HHMMSS or YYMMDD HHMMSS
PATTERN — Matches a rule string (see Pattern)
REQUIRED — Non-empty, not just spaces
YES/NO — Only YES, NO, Y, or N (case-insensitive)
VRULES — Validates the rule set itself
PRINTMASK — Checks a print-format mask
EXPRESSION — Legal Sheerpower expression
CODE — Legal Sheerpower code block
MENU — Legal INPUT MENU description
ROUTINE — Routine name exists
FILTER — ltrim, rtrim, trim, ucase, lcase, replace, remove, change; plus restore

Return Values
Most rules return TRUE (1) or FALSE (0). The fulltime rule's documentation below states FALSE (0) as the failure value; confirm in your build. Consistently check your return values when mixing rules.

INTEGER, NUMBER, DIGITS, LETTERS — What the Text Is

integer accepts a whole number, number any number, digits only the characters 0–9, letters only letters; ucase and lcase accept one case only. Rules chain with a semicolon and every one must pass.

// Expect: 1, 0, 1, 0 print valid('42', 'integer') print valid('4.2', 'integer') print valid('4.2', 'number') print valid('abc', 'digits') end

ALLOW, DISALLOW, LENGTH, DECIMALS — The Values Allowed

These are the rules that save a loop around an input: the range a number may fall in, the values a text may take, how long it may be. allow lo to hi is a range and allow a, b, c a list; the two combine (allow 1 to 5, 10). What the values ARE follows the rule before them: after integer or number they are numbers, after a date rule they are dates (and today is a value), with no such rule they are text compared case-blind. disallow takes the same forms and refuses them. length n is an exact length, minlength n and maxlength n the bounds, decimals n the most decimal places a number may have.

// Expect: 1, 0, 0, 1, 0, 1, 0, 1, 0 print valid('50', 'integer; allow 1 to 120') print valid('121', 'integer; allow 1 to 120') print valid('abc', 'integer; allow 1 to 120') print valid('green', 'allow red, green') print valid('13', 'integer; disallow 13') print valid('abcdef', 'length 6') print valid('abcde', 'length 6') print valid('abc', 'minlength 2; maxlength 4') print valid('1.255', 'number; decimals 2') end

On an input field the same words go in the valid option, and the prompt repeats until the answer passes — no loop to write: input 'Age', valid 'integer; allow 1 to 120': age, or in a screen form <valid "required; integer; allow 1 to 120": Age ###>.

DATE DMONY — DD-Mon-YY

Checks if text$ is a legal date in DD-Mon-YY format and returns TRUE (1) or FALSE (0).

// Expect: 1 text$ = '01-Jan-99' vrule$ = 'date dmony' print valid(text$, vrule$) end

DATE DMONCY — DD-Mon-CCYY

Checks if text$ is a legal date in DD-Mon-CCYY format and returns TRUE (1) or FALSE (0).

// Expect: 1 text$ = '01-Jan-2000' vrule$ = 'date dmoncy' print valid(text$, vrule$) end

FULLTIME — Date + Time

Checks if text$ is a legal date/time in either CCYYMMDD HHMMSS or YYMMDD HHMMSS format. Returns TRUE (1) or FALSE (0).

// Expect: 1 then 1 text$ = '20000122 010101' text2$ = '990122 010101' vrule$ = 'fulltime' print valid(text$, vrule$) print valid(text2$, vrule$) end

PATTERN — Custom Rule Match

Matches text$ against a rule string. See Section 6.5.4 PATTERN(str_expr1, str_expr2) for details.

REQUIRED — Non-blank

Returns TRUE if text$ is not null and not only spaces.

// Expect: 1 then 0 text$ = 'a b c d' text2$ = ' ' vrule$ = 'required' print valid(text$, vrule$) print valid(text2$, vrule$) end

YES/NO — Strict Yes/No

Returns TRUE if text$ is one of: YES, NO, Y, N.

// Expect: 1, 1, 0 text$ = 'yes' text2$ = 'n' text3$ = 'a' vrule$ = 'yes/no' print valid(text$, vrule$) print valid(text2$, vrule$) print valid(text3$, vrule$) end

VRULES — Validate Rule Set

Checks if a vrules$ string itself is a legal set of validation rules.

// Expect: 1 print valid('integer', 'vrules') end

PRINTMASK — Format Mask

Validates that text$ is a legal print mask.

// Expect: 1 text_str$ = '##.##' vrule$ = 'printmask' print valid(text_str$, vrule$) end

EXPRESSION — Sheerpower Expression

Returns TRUE if text$ is a legal Sheerpower expression.

// Expect: 1 then 0 text_str$ = 'total = a% + 30' text_str2$ = '##~-###' vrule$ = 'expression' print valid(text_str$, vrule$) print valid(text_str2$, vrule$) end

CODE — Sheerpower Code

Returns TRUE if text$ is syntactically legal Sheerpower code.

// Expect: false path taken a$ = "print 'hello, ' name$" if not valid(a$, 'code') then print 'false' end if end

MENU — INPUT MENU Description

Validates a menu description string for use with INPUT MENU.

// Expect: 1 text_str$ = '%multi,a,b,c' vrule$ = 'menu' print valid(text_str$, vrule$) end

ROUTINE — Existence Check

TRUE if the named routine exists; FALSE otherwise.

// Expect: 1 then 0 text$ = 'do_totals' text2$ = 'test_nos' vrule$ = 'routine' print valid(text$, vrule$) print valid(text2$, vrule$) end routine do_totals: private mytotal, desc$ mytotal = 15 desc$ = 'Test Totals' print desc$; mytotal end routine

FILTER — Clean, Transform, Validate

Filters modify text$ before validation. Chain multiple steps with semicolons. Supported filters:

ltrim$ — remove leading spaces, control characters, and DEL
rtrim$ — remove trailing spaces, control characters, and DEL
trim$ — remove leading and trailing spaces, control characters, and DEL
ucase$ — uppercase
lcase$ — lowercase
replace — replace chars: old=new, old2=new2, ...
remove — remove listed chars from text$
change — complex remap (see CHANGE$)
restore — restore original (pre-filter) text for later validations in the same rule chain

Examples

FILTER + REMOVE / REPLACE / TRIM

// Expect: true, true, true text$ = 'abcd10' vrules$ = 'filter remove "10"; letters' text2$ = 'ab1cd1' vrules2$ = "filter replace '1'='e'; letters" text3$ = ' 1234 ' vrules3$ = "filter trim; number" if valid(text$, vrules$) then print 'true' if valid(text2$, vrules2$) then print 'true' if valid(text3$, vrules3$) then print 'true' end

FILTER + RESTORE

Use restore when you need a filtered check and then a second check on the original text later in the chain.

// Expect: true text$ = "123" vrule$ = "filter replace '1'='a'; restore; number" if valid(text$, vrule$) then print 'true' end

Design Tip
Problem: Real-world text is messy—extra spaces, odd casing, stray characters.
Solution: Standardize with filter (trim/case/remap), then restore to re-validate the untouched original when needed.
Efficiency: A single valid() call can handle a pipeline of filters and checks—fewer passes, clearer intent.
Takeaway: Treat validation as a readable pipeline: filter ...; restore; rule; rule; ...

Putting It Together

Combine rules with semicolons to form pipelines. Validate and transform in one statement, and lean on vrules to sanity-check complex rule strings at load time.

(Show/Hide Sheerpower Validation Rules Takeaways)

Hide Description

a$ b$ x y

Enter or modify the code below, and then click on RUN

Looking for the full power of Sheerpower?
Check out the Sheerpower website. Free to download. Free to use.