Now, of course, I could write my regular expression to handle both cases, such as regexp.Compile("[a-zA-Z]")
, but my regular expression is constructed from a string given by the user:
reg, err := regexp.Compile(strings.Replace(s.Name, " ", "[ \\._-]", -1))
Where s.Name
is the name. Which could be something like 'North by Northwest'. Now, the most apparent solution to me would be to walk through each character of s.Name
and write '[nN]' for each letter:
for i := 0; i < len(s.Name); i++ {
if s.Name[i] == " " {
fmt.Fprintf(str, "%s[ \\._-]", str);
} else {
fmt.Fprintf(str, "%s[%s%s]", str, strings.ToLower(s.Name[i]), strings.ToUpper(s.Name[i]))
}
}
But I feel this is a rather non-elegant solution. Speed is not really a concern, but I need to know if there is another way.
You can set a case-insensitive flag as the first item in the regex.
You do this by adding "(?i)"
to the beginning of a regex.
reg, err := regexp.Compile("(?i)"+strings.Replace(s.Name, " ", "[ \\._-]", -1))
For a fixed regex it would look like this.
r := regexp.MustCompile(`(?i)CaSe`)
For more information about flags, search the
regexp/syntax
package documentation
(or the syntax documentation)
for the term "flags".