New _str_repeat

_str_repeat2.sh

@@ -0,0 +1,56 @@
+. "$AGLIB"/_char_repeat.sh
+
+# Fancy string repeat
+# Note: we use = to compare numbers for speed
+_str_repeat2() {
+  s="$1"
+  count="$2" # _test_parse_int ?
+  method="$3"
+  strlen=${#1}
+  # TODO Auto-select method based on string length and count
+  # Use awk when $count > 500 ? (only for long strings?) UNLESS printf is a builtin, then use printf-seq until ~800 or more?
+  # Use awk when $strlen > ?
+  # Most important aspect is to reduce iterations as much as possible to reduce builtin/external command calls
+  if [ -z "$method" ]; then
+     if [ $strlen = 1 ]; then method=char
+   else return 4
+     fi
+  fi
+  [ ${4-0} = 1 ] && >&2 printf 'Source: "%s"\nLength: %d\nCount: %d\nMethod: %s\n' "$s" $strlen $count "$method"
+  case "$method" in
+          char) _char_repeat "$1" $count; echo ;;
+    printf-seq) for i in $(seq $count); do printf %s "$s"; done ;;
+        static) case $count in # TODO Source a seperate file of compiled case statement from 2-100+
+                  2) printf %s "$s$s" ;;
+                  3) printf %s "$s$s$s" ;;
+                  4) printf %s "$s$s$s$s" ;;
+                  5) printf %s "$s$s$s$s$s" ;;
+                  6) printf %s "$s$s$s$s$s$s" ;;
+                  7) printf %s "$s$s$s$s$s$s$s" ;;
+                  8) printf %s "$s$s$s$s$s$s$s$s" ;;
+                  9) printf %s "$s$s$s$s$s$s$s$s$s" ;;
+                 10) printf %s "$s$s$s$s$s$s$s$s$s$s" ;;
+                 11) printf %s "$s$s$s$s$s$s$s$s$s$s$s" ;;
+                 12) printf %s "$s$s$s$s$s$s$s$s$s$s$s$s" ;;
+                 13) printf %s "$s$s$s$s$s$s$s$s$s$s$s$s$s" ;;
+                 14) printf %s "$s$s$s$s$s$s$s$s$s$s$s$s$s$s" ;;
+                 15) printf %s "$s$s$s$s$s$s$s$s$s$s$s$s$s$s$s" ;;
+                 16) printf %s "$s$s$s$s$s$s$s$s$s$s$s$s$s$s$s$s" ;;
+                 17) printf %s "$s$s$s$s$s$s$s$s$s$s$s$s$s$s$s$s$s" ;;
+                 18) printf %s "$s$s$s$s$s$s$s$s$s$s$s$s$s$s$s$s$s$s" ;;
+                 19) printf %s "$s$s$s$s$s$s$s$s$s$s$s$s$s$s$s$s$s$s$s" ;;
+                 20) printf %s "$s$s$s$s$s$s$s$s$s$s$s$s$s$s$s$s$s$s$s$s" ;;
+                esac
+              ;;
+          eval) t=; until [ ${#t} -eq $(($count * 2)) ]; do t="$t\$s"; done; eval "printf %s \"$t\"" ;;
+       awk-var) awk -v s="$s" 'BEGIN{for(i=0;i<'${count}';i++)printf "%s",s}' ;;
+      awk-pipe) awk 'BEGIN{for(i=0;i<'${count}';i++)printf "%s",s}' ;;
+    concat-seq) t=; for i in $(seq $count); do t="$t$s"; done; printf %s "$t" ;;
+    perl-slurp) perl -0777pe '$_=$_ x '$count ;;
+             *) return 3;;
+  esac
+}
+
+# TODO String builder using a fifo
+# Background process (awk? dd?) reading from pipe, building a string efficiently in memory, then dumping it back to pipe
+# to be read back into main context variable somehow?