test harness: Fix off-by-one discrepancy between observed and expected.

Fixes issue #195.

tests/README.md

@@ -5,6 +5,13 @@ Utility scripts for testing zsh-syntax-highlighting highlighters.
 
 The tests expect the highlighter directory to contain a `test-data` directory with test data files. See the [main highlighter](../highlighters/main/test-data) for examples.
 
+Each test should define the array parameter `$expected_region_highlight`.
+The value of that parameter is a list of `"$i $j $style"` strings.
+Each string specifies the highlighting that `$BUFFER[$i,$j]` should have;
+that is, `$i` and `$j` specify a range, 1-indexed, inclusive of both endpoints.
+
+_Note_: `$region_highlight` uses the same `"$i $j $style"` syntax but interprets the indexes differently.
+
 
 highlighting test
 -----------------

tests/test-highlighting.zsh

@@ -84,7 +84,10 @@ for data_file in ${0:h:h}/highlighters/$1/test-data/*; do
       observed_result=()
       for i in {1..${#region_highlight}}; do
         highlight_zone=${(z)region_highlight[$i]}
-        for j in {$highlight_zone[1]..$highlight_zone[2]}; do
+        integer start=$highlight_zone[1] end=$highlight_zone[2]
+        (( --end )) # region_highlight ranges are half-open
+        (( ++start, ++end )) # region_highlight is 0-indexed; expected_region_highlight is 1-indexed
+        for j in {$start..$end}; do
           observed_result[$j]=$highlight_zone[3]
         done
       done