only show unique entries in history search

Author: KristofferC

stdlib/REPL/src/History/resumablefiltering.jl

@@ -258,16 +258,19 @@ end
 Incrementally filter `candidates[1:idx]` in reverse order.
 
 Pushes matches onto `out` until either `maxtime` is exceeded or `maxresults`
-collected, then returns the new resume index.
+collected, then returns the new resume index. Only unique entries (by content)
+are added to avoid showing duplicate history items.
 """
 function filterchunkrev!(out::Vector{HistEntry}, candidates::DenseVector{HistEntry},
                          spec::FilterSpec, idx::Int = length(candidates);
                          maxtime::Float64 = Inf, maxresults::Int = length(candidates))
+    seen = Set(e.content for e in out)
     batchsize = clamp(length(candidates) ÷ 512, 10, 1000)
     for batch in Iterators.partition(idx:-1:1, batchsize)
         time() > maxtime && break
         for outer idx in batch
             entry = candidates[idx]
+            entry.content ∈ seen && continue
             if !isempty(spec.modes)
                 entry.mode ∈ spec.modes || continue
             end
@@ -293,6 +296,7 @@ function filterchunkrev!(out::Vector{HistEntry}, candidates::DenseVector{HistEnt
                 end
             end
             matchfail && continue
+            push!(seen, entry.content)
             pushfirst!(out, entry)
             length(out) == maxresults && break
         end

stdlib/REPL/test/history.jl

@@ -344,6 +344,29 @@ end
                 @test filterchunkrev!(results, entries, spec) == 0
                 @test results == entries[3:6]
             end
+            @testset "Uniqueness" begin
+                empty!(results)
+                # Create entries with duplicate content
+                dup_entries = [
+                    HistEntry(:julia, now(UTC), "println(\"hello\")", 1),
+                    HistEntry(:julia, now(UTC), "cos(2π)", 2),
+                    HistEntry(:julia, now(UTC), "println(\"hello\")", 3),  # duplicate
+                    HistEntry(:julia, now(UTC), "sin(π)", 4),
+                    HistEntry(:julia, now(UTC), "cos(2π)", 5),  # duplicate
+                    HistEntry(:julia, now(UTC), "println(\"hello\")", 6),  # duplicate
+                    HistEntry(:julia, now(UTC), "tan(π/4)", 7),
+                ]
+                cset = ConditionSet("")  # Match all
+                spec = FilterSpec(cset)
+                @test filterchunkrev!(results, dup_entries, spec) == 0
+                # Should only get unique entries
+                # Since we iterate in reverse (7->1), we keep the most recent occurrence of each unique content
+                @test length(results) == 4
+                @test results[1] == dup_entries[4]  # sin(π)
+                @test results[2] == dup_entries[5]  # cos(2π) - most recent
+                @test results[3] == dup_entries[6]  # println("hello") - most recent
+                @test results[4] == dup_entries[7]  # tan(π/4)
+            end
         end
         @testset "Strictness comparison" begin
             c1 = ConditionSet("hello world")