echo "words words words ⚑" echo """ <html> <head> </head>\n\n <body> </body> </html> """ proc re(s: string): string = s echo r".""." echo re"\b[a-z]++\b"
$ nim c -r strings.nim words words words ⚑ <html> <head> <head/>\n\n <body> <body/> <html/> .". \b[a-z]++\b
There are several types of string literals:
- Quoted Strings: Created by wrapping the body in triple quotes, they never interpret escape codes
- Raw Strings: created by prefixing the string with an
r. They do not interpret escape sequences, except for
"", which is interpreted as
". This means that
r"\b[a-z]\b"is interpreted as
\b[a-z]\binstead of failing to compile with a syntax error.
- Proc Strings: raw strings, but the method name that prefixes the string is called, so that
Strings are null-terminated, so that
cstring("foo") requires zero copying. However, you should be careful that the lifetime of the cstring does not exceed the lifetime of the string it is based upon.
Strings can also almost be thought of as
seq[char] with respect to assignment semantics. See seqs
A note about unicode
Unicode symbols are allowed in strings, but are not treated in any special way, so if you want count glyphs or uppercase unicode symbols, you must use the
Strings are generally considered to be encoded as UTF-8, so because of unicode’s backwards compatibility, can be treated exactly as ASCII, with all values above 127 ignored.