6.13 Regexp Demos

6.13.1 Search for regexp

The following example is a simple program that enters a regexp from the keyboard, followed by another test string, trying to find the regexp in the test string. If found, the registers are printed on the screen, and the corresponding matched substrings:

Program 6: searchDemo.bet

ORIGIN '~beta/basiclib/regexp'; --- program: descriptor --- (# do loop: (# regex: ^text; do 'Search for: ' -> puttext; getLine -> regex[]; (if regex.length = 0 then leave loop if); loop: (# string: ^text; do 'Search for: '->puttext; regex[]->puttext; ' in: ' -> puttext; getline -> string[]; (if string.length = 0 then leave loop if); 0 -> string.pos; regex[] -> string.regexp_search (# noMatch::< (# do '\tNo match' -> putline #) do (for i: regexp_numberOfRegisters repeat (if (i-1->regs.start)>=0 then '\tregister '->puttext; i-1 -> putint; ': '->puttext; i-1->regs.start->putint; ' to '->puttext; i-1->regs.end->putint; ': '->puttext; ((i-1->regs.start)+1, i-1->regs.end)->string.sub->putline if) for); #); restart loop #); restart loop #) #)

Output from the demo might be:

Search for: 1\(234\)5 Search for: 1\(234\)5 in: dfjdfhdf1235jfdkj12345ghdf register 0: 17 to 22: 12345 register 1: 18 to 21: 234 Search for: 1\(234\)5 in: 12345 register 0: 0 to 5: 12345 register 1: 1 to 4: 234 Search for: 1\(234\)5 in: 123 No match Search for: 1\(234\)5 in: Search for: \(22\).*\1 Search for: \(22\).*\1 in: xxx22yyy2zzz No match Search for: \(22\).*\1 in: xxx22yyy22zzz register 0: 3 to 10: 22yyy22 register 1: 3 to 5: 22

6.13.2 Replace with a regexp

The following example is nearly identical to the previous, except that it allows for the specification of a replacement string:

Program 7: replaceDemo.bet

ORIGIN '~beta/basiclib/regexp'; --- program: descriptor --- (# do loop: (# regex: ^text; do 'Replace: ' -> puttext; getLine -> regex[]; (if regex.length = 0 then leave loop if); loop: (# string: ^text; do 'Replace: '->puttext; regex[]->puttext; ' in: ' -> puttext; getline -> string[]; (if string.length = 0 then leave loop if); loop: (# t, replacement: ^text do string.copy -> t[]; 0 -> t.pos; 'Replace: '->puttext; regex[]->puttext; ' in: ' -> puttext; string[]->puttext; ' with: '->puttext; getline -> replacement[]; (if replacement.length = 0 then leave loop if); (regex[], replacement[]) -> t.regexp_replace (# noMatch::< (# do '\tNo match'->screen.putline #) #); 'Replace: '->puttext; regex[]->puttext; ' in: ' -> puttext; string[]->puttext; ' with: '->puttext; replacement[]->puttext; ' gives: '->puttext; t[] -> putline; restart loop #); restart loop #); restart loop #) #)

Output from the demo might be:

Replace: [Ww]ord Replace: [Ww]ord in: This Word is in capital Replace: [Ww]ord in: This Word is in capital with: sentence Replace: [Ww]ord in: This Word is in capital with: sentence gives: This sentence is in capital