Rewrite some Enum.reduce when they can be expressed with a standard library functions - #1
Open
katafrakt wants to merge 3 commits into
Open
Rewrite some Enum.reduce when they can be expressed with a standard library functions#1katafrakt wants to merge 3 commits into
katafrakt wants to merge 3 commits into
Conversation
katafrakt
force-pushed
the
rewrite-reduces
branch
from
July 30, 2026 13:00
b89f13d to
7635bc9
Compare
s3cur3
approved these changes
Aug 3, 2026
Comment on lines
+380
to
+383
| case reduce_rewrite(init, reducer) do | ||
| :sum -> {{:., dm, [{:__aliases__, am, [:Enum]}, :sum]}, m, [enum]} | ||
| {module, name, new_reducer} -> {{:., dm, [{:__aliases__, am, module}, name]}, m, [enum, new_reducer]} | ||
| :none -> node |
Comment on lines
+787
to
+794
| with true <- sum_product_by_available?(), | ||
| true <- literal?(init, identity), | ||
| {:ok, expr} <- accumulator_partner(a, b, acc), | ||
| false <- references_var?(expr, acc), | ||
| false <- same_variable?(expr, item), | ||
| false <- literal_constant?(expr) do | ||
| name = if op == :+, do: :sum_by, else: :product_by | ||
| {{[:Enum], name}, rebuild.(expr)} |
| # `Enum.reduce(enum, 0, fn item, acc -> acc + expr end)` => `Enum.sum_by(enum, fn item -> expr end)` | ||
| # `Enum.reduce(enum, 1, fn item, acc -> acc * expr end)` => `Enum.product_by(enum, fn item -> expr end)` | ||
| defp classify_reduce(init, item, acc, _stmts, {op, _, [a, b]}, rebuild) when op in [:+, :*] do | ||
| identity = if op == :+, do: 0, else: 1 |
There was a problem hiding this comment.
We could potentially combine the two if statements within this function:
Suggested change
| identity = if op == :+, do: 0, else: 1 | |
| {identity, name} = if op == :+, do: {0, :sum_by}, else: {1, :product_by} |
|
|
||
| defp simplify_reduce(_init, _reducer), do: nil | ||
|
|
||
| defp reduce_rewrite(init, reducer) do |
There was a problem hiding this comment.
Thoughts on pulling out the ~200 lines of reduce classifier code out into a new helper module? It'd kind of be nice from an organizational perspective, but I probably wouldn't bother trying to test it in isolation (that is, apart from the actual rewriting pipeline).
Comment on lines
+1927
to
+1930
| expected = | ||
| "a |> Enum.filter(&((&1 > 0 and &1 < 10) && &1 != 5)) |> bar()" | ||
| |> Code.format_string!() | ||
| |> IO.iodata_to_binary() |
Comment on lines
+1592
to
+1594
| test "leaves the bare acc + item case for the Enum.sum rewrite" do | ||
| assert_style("Enum.reduce(items, 0, fn item, acc -> acc + item end)", "Enum.sum(items)") | ||
| end |
There was a problem hiding this comment.
We can probably remove this one, since it's covered by another existing test, right?
Comment on lines
+1548
to
+1552
| assert_style( | ||
| "Enum.reduce(items, 0, fn item, acc -> acc + item.count end)", | ||
| "Enum.sum_by(items, fn item -> item.count end)" | ||
| ) | ||
| end |
There was a problem hiding this comment.
Can the expression be anything? For instance, do we rewrite this:
invoice
|> Map.get("items", [])
|> Enum.reduce(0, fn item, acc -> acc + (item["amount"] || 0) end)
Comment on lines
+1657
to
+1669
| test "leaves a frequency counter whose value reads the accumulator" do | ||
| assert_style("Enum.reduce(items, %{}, fn item, acc -> Map.put(acc, item.k, Map.get(acc, item.k, 0) + 1) end)") | ||
| end | ||
|
|
||
| test "leaves a conditional Map.put (a filter/reject shape)" do | ||
| assert_style( | ||
| "Enum.reduce(items, %{}, fn item, acc -> if item.ok?, do: Map.put(acc, item.k, item.v), else: acc end)" | ||
| ) | ||
| end | ||
|
|
||
| test "leaves a non-empty initial map untouched" do | ||
| assert_style("Enum.reduce(items, %{a: 1}, fn item, acc -> Map.put(acc, item.k, item.v) end)") | ||
| end |
Comment on lines
+1684
to
+1686
| test "leaves the unconditional acc + 1 form" do | ||
| assert_style("Enum.reduce(items, 0, fn item, acc -> acc + 1 end)") | ||
| end |
There was a problem hiding this comment.
This one could be rewritten to Enum.count(items), right? (I agree if the added value were anything but 1, though, we wouldn't want to touch it.)
No need to make that happen here—might just be a nice followup for some point in the future.
Comment on lines
+1718
to
+1724
| test "leaves the [expr | acc] prepend form (needs an external Enum.reverse)" do | ||
| assert_style("Enum.reduce(items, [], fn item, acc -> [item.name | acc] end)") | ||
| end | ||
|
|
||
| test "leaves a multi-element append (a flat_map shape)" do | ||
| assert_style("Enum.reduce(items, [], fn item, acc -> acc ++ [item.a, item.b] end)") | ||
| end |
There was a problem hiding this comment.
These are nice for documentation of future followup opportunities. 😄
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
No description provided.