First Word
January 25, 2019
We have a simple exercise today, inspired a co-worker. Where I work, we have a reporting tool that permits a “hook” to the underlying SQL in some places. My co-worker asked me how to write an SQL statement that extracts the first word (a maximal sequence of non-spaces) from the beginning of a string (assume there are no leading spaces). For instance, given the string “abcdefg hijklmnop qrs tuv wxyz” the first word is “abcdefg”. Here’s the SQL expression, wrapped in a select statement, with &&STR representing the string:
select substr('&&STR', 1, instr('&&STR', ' ') - 1) from dual
Your task is to write a program to extract the first word from a string. When you are finished, you are welcome to read or run a suggested solution, or to post your own solution or discuss the exercise in the comments below.
Extra points for implementing the Unicode default word breaking algorithm: https://unicode.org/reports/tr29/#Word_Boundaries
Two solutions in golang.
package main
import (
"fmt"
"regexp"
)
func main() {
s1 := "abcdefg hijklmnop qrs tuv wxyz"
s2 := "~hola caracola de piazzolla"
fmt.Println(firstWordWithLoop(s1))
fmt.Println(firstWordWithRegexp(s1))
fmt.Println()
fmt.Println(firstWordWithLoop(s2))
fmt.Println(firstWordWithRegexp(s2))
}
func firstWordWithLoop(str string) string {
word := ""
for _, char := range str {
if char == ‘ ‘ {
break
}
word += string(char)
}
return word
}
func firstWordWithRegexp(str string) string {
return regexp.MustCompile(`[^ ]+`).FindString(str)
}
Here’s a solution in C.
#include <stdio.h> #include <stdlib.h> int main(int argc, char* argv[]) { if (argc != 2) { fprintf(stderr, "Usage: %s STR\n", argv[0]); return EXIT_FAILURE; } char* str = argv[1]; while (1) { char c = *(str++); if (c == ' ' || c == '\0') break; printf("%c", c); } printf("\n"); return EXIT_SUCCESS; }Example Usage:
AWK version
$ echo "abcdefg hijklmnop qrs tuv wxyz" | awk ‘{ print $1 }’
abcdefg
—
Klong version
(-1)_((a?" ")@0)#a::"abcdefg hijklmnop qrs tuv wxyz"
"abcdef"
a
"abcdefg hijklmnop qrs tuv wxyz"
a?" "
[7 17 21 25]
—
MUMPS version
YDB>w $p("abcdefg hijklmnop qrs tuv wxyz"," ")
abcdefg
That Unicode algorithm looks a bit complicated, here’s a simple Unicode-friendly solution using Python str.isspace():
def firstword(s): start = -1 for i,c in enumerate(s): if start < 0: if not c.isspace(): start = i; elif c.isspace(): return s[start:i] return None if start < 0 else s[start:] assert(firstword("") is None) assert(firstword(" ") is None) assert(firstword("foo") == "foo") assert(firstword(" foo") == "foo") assert(firstword("foo ") == "foo")Python solution:
firstWord = lambda x: x.lstrip(" ").split(" ")[0]