ERASE

The ERASE statement is used to clear all data from an array.

Syntax

ERASE arrayName [, arrayName2 ...]

ERASE parentArray(index).subordinateNestedArray

Description

• You do not have to include the array brackets in an ordinary ERASE call, but you must specify the parent index to clear a subordinary array.
• Multiple arrays can be erased using commas between the array names for ordinary arrays only.
• Static arrays will always keep their dimensions and bounds, only the values in the array (contents) are cleared as follows:
  • All numerical array elements become zero(0).
  • All chars of fixed length STRING array elements are filled with CHR$(0).
  • Variable length STRINGs will be emptied ("").
• Dynamic arrays otherwise are completly destroyed by ERASE, i.e. all its contents is gone and the memory associated with the array is freed.
  • Dynamic arrays must be REDIMed if they are referenced after being erased.
  • You may reduce the memory impact of your program by erasing no longer needed dynamic arrays.

Notes on the second Syntax

QB64-PE v4.5.0 extends ERASE support for arrays nested inside TYPE definitions. Besides the traditional use on ordinary arrays, ERASE may now also be used on an individual nested array member, or on a parent array/variable containing such members. Erasing the parent also clears all subordinate nested arrays. Since arrays inside TYPEs are always static arrays, their dimensions and bounds remain unchanged, only the contents is reset.

Availability

• 
  all
• 
  all
• 
• 
  yes
• 
  yes
• 
  yes

• Since arrays in TYPEs were implemented in QB64-PE v4.5.0, this statement was adapted to properly handle it (see Example 2).

Examples

Example 1

The classical use on ordinary arrays.

DIM NumArr(5) AS INTEGER DIM FixArr(5) AS STRING * 3 DIM VarArr(5) AS STRING NumArr(1) = 123 FixArr(1) = "ABC" VarArr(1) = "XYZ" PRINT "Numeric before ERASE: "; NumArr(1) PRINT "FixStr before ERASE: "; FixArr(1), "LEN = "; LEN(FixArr(1)) PRINT "VarStr before ERASE: "; VarArr(1), "LEN = "; LEN(VarArr(1)) PRINT ERASE NumArr, FixArr, VarArr PRINT "Numeric after ERASE: "; NumArr(1) PRINT "FixStr after ERASE: "; FixArr(1), "LEN = "; LEN(FixArr(1)) PRINT "VarStr after ERASE: "; VarArr(1), "LEN = "; LEN(VarArr(1)) PRINT PRINT "Note that the lenght of the fixed length STRING is unchanged," PRINT "it's now just filled with CHR$(0) chars, which don't print."

---

Example 2

The new usage on subordinary and nested arrays.

TYPE TestType aaa(2) AS DOUBLE END TYPE DIM a(1) AS TestType DIM x AS INTEGER DIM y AS INTEGER PRINT "Generating values..." FOR y = 0 TO 1 FOR x = 0 TO 2 a(y).aaa(x) = (x + 1) * (y + 1) PRINT a(y).aaa(x); NEXT x NEXT y PRINT PRINT PRINT "Erasing only one nested member array: a(0).aaa" ERASE a(0).aaa FOR y = 0 TO 1 FOR x = 0 TO 2 PRINT a(y).aaa(x); NEXT x NEXT y PRINT PRINT PRINT "Generating values again..." FOR y = 0 TO 1 FOR x = 0 TO 2 a(y).aaa(x) = (x + 1) * (y + 1) NEXT x NEXT y PRINT "Erasing the parent array: a" ERASE a FOR y = 0 TO 1 FOR x = 0 TO 2 PRINT a(y).aaa(x); NEXT x NEXT y PRINT

Example by Petr

Explanation In this example, aaa() is a array member inside TestType, and a() is an array of TestType elements. ERASE a(0).aaa clears only that nested member array. ERASE a clears the parent array and all nested array members it contains. This extends the original use of ERASE. Since arrays inside TYPEs are always static arrays, their dimensions and bounds remain unchanged, only the contents is reset.

See also

• Featured in our "Keyword of the Day" series
• DIM, REDIM, CLEAR
• $STATIC, $DYNAMIC
• Arrays, TYPE

---

Navigation: Main Page with Articles and Tutorials Keyword Reference - Alphabetical Keyword Reference - By usage Report a broken link