Merge pull request #30 from akhildevelops/main

Zig contribution and Benchmarking on Github actions runner

.github/workflows/benchmark.yml

@@ -0,0 +1,38 @@
+name: Benchmark
+
+on:
+    pull_request:
+      branches:
+          - main
+    push:
+      branches:
+          - main
+
+jobs:
+    run-benchmarks:
+        runs-on:  ubuntu-22.04
+        steps:
+            - name: Checkout
+              uses: actions/checkout@v4
+            - name: Install  Rust
+              uses: dtolnay/rust-toolchain@stable
+            - name: Install Go
+              uses: actions/setup-go@v4
+              with:
+                go-version: '1.21.1'
+            - name: Install crystal
+              uses: crystal-lang/install-crystal@v1
+            - name: Install zig
+              uses: goto-bus-stop/setup-zig@v2
+            - name: Install Julia
+              uses: julia-actions/setup-julia@v1
+            - name: Install Vlang
+              uses: vlang/[email protected]
+            - name: Install jq
+              run: sudo apt-get install -y jq
+            - name: Install odin
+              run: wget -O odin.zip https://github.com/odin-lang/Odin/releases/download/dev-2023-08/odin-ubuntu-amd64-dev-2023-08.zip && mkdir odin_exe && unzip odin.zip -d odin_exe && echo -e 'export PATH=$(pwd)/odin_exe:"$PATH"\n' >> ~/.bashrc && chmod +x odin_exe/odin
+            - name: Install hyperfine for benchmarking
+              run: cargo install hyperfine
+            - name: Run Benchmark
+              run: ./run.sh all

.gitignore

@@ -16,3 +16,8 @@ ols.json
 odin/related
 .dart_tool/
 v/related
+
+# Zig
+zig-cache
+zig/main
+zig/main.o

go.work

@@ -1,5 +1,4 @@
 go 1.21.1
-
 use (
 	./go
 	./go_con

go/go.mod

@@ -1,7 +1,5 @@
 module g.io/related
-
 go 1.21.1
-
 require github.com/goccy/go-json v0.10.2
 
 require github.com/ugurcsen/gods-generic v0.10.4

go_con/go.mod

@@ -1,5 +1,3 @@
 module g.io/related_concurrent
-
 go 1.21.1
-
 require github.com/goccy/go-json v0.10.2

run.sh

@@ -108,6 +108,19 @@ run_crystal() {
 
 }
 
+run_zig() {
+    echo "Running Zig" &&
+        cd ./zig &&
+        zig build-exe -lc -O ReleaseFast main.zig
+        if [ $HYPER == 1 ]; then
+            command hyperfine -r 10 -w 3 --show-output "./main"
+        else
+            command time -f '%es %Mk' ./main
+        fi
+
+    check_output "related_posts_zig.json"
+}
+
 run_julia_v1() {
     echo "Running Julia v1" &&
         cd ./julia &&
@@ -207,6 +220,10 @@ elif [ "$first_arg" = "cr" ]; then
 
     run_crystal
 
+elif [ "$first_arg" = "zig" ]; then
+
+    run_zig
+
 elif [ "$first_arg" = "jul1" ]; then
 
     run_julia_v1
@@ -229,16 +246,22 @@ elif [ "$first_arg" = "v" ]; then
 
 elif [ "$first_arg" = "all" ]; then
 
-    echo "running all" &&
-        run_go &&
-        run_go_concurrent &&
-        run_rust &&
-        run_rust_rayon &&
-        run_python &&
-        run_python_np &&
-        run_crystal &&
-        run_julia_v1 &&
-        run_julia_v2
+    echo -e "Running all\n" &&
+        run_go || echo -e "\n" &&
+        run_go_concurrent || echo -e "\n" &&
+        run_rust || echo -e "\n" &&
+        run_rust_rayon || echo -e "\n" &&
+        run_python || echo -e "\n" &&
+        run_python_np || echo -e "\n" &&
+        run_crystal || echo -e "\n" &&
+        run_zig || echo -e "\n" &&
+        run_julia_v1 || echo -e "\n" &&
+        run_julia_v2 || echo -e "\n" &&
+        run_odin || echo -e "\n" &&
+        run_jq || echo -e "\n" &&
+        run_vlang || echo -e "\n" &&
+        echo -e "Finished running all\n"
+
 
 elif [ "$first_arg" = "clean" ]; then
 
@@ -251,10 +274,12 @@ elif [ "$first_arg" = "clean" ]; then
         cd .. &&
         cd rust_rayon && cargo clean &&
         cd .. &&
+        cd zig && rm -f main main.o &&
+        cd ..
         rm -f related_*.json
 
 else
 
-    echo "Valid args: go | go_con | rust | rust_ray | py | numpy | cr | odin | jq | jul1 | jul2 | v | all | clean. Unknown argument: $first_arg"
+    echo "Valid args: go | go_con | rust | rust_ray | py | numpy | cr | zig | odin | jq | jul1 | jul2 | v | all | clean. Unknown argument: $first_arg"
 
 fi

zig/main.zig

@@ -0,0 +1,62 @@
+const std = @import("std");
+var allocator = std.heap.c_allocator;
+const post = struct { _id: []const u8, title: []const u8, tags: [][]const u8 };
+const posts = []post;
+const constop_posts = struct { _id: *const []const u8, tags: *const [][]const u8, related: []*post };
+const stdout = std.io.getStdOut().writer();
+
+fn lessthan(context: void, lhs: usize, rhs: usize) bool {
+    _ = context;
+    return lhs < rhs;
+}
+
+pub fn main() !void {
+    const file = try std.fs.cwd().openFile("../posts.json", .{});
+    defer file.close();
+    const arr_posts = std.ArrayList(usize);
+    var map = std.StringHashMap(arr_posts).init(allocator);
+    defer map.deinit();
+    var json_reader = std.json.reader(allocator, file.reader());
+    defer json_reader.deinit();
+    const parsed = try std.json.parseFromTokenSource(posts, allocator, &json_reader, .{});
+    defer parsed.deinit();
+
+    const start = try std.time.Instant.now();
+
+    for (parsed.value, 0..) |post_ele, i| {
+        for (post_ele.tags) |tag| {
+            var g_p = try map.getOrPut(tag);
+            if (g_p.found_existing) {
+                try g_p.value_ptr.*.append(i);
+            } else {
+                var temp = arr_posts.init(allocator);
+                try temp.append(i);
+                g_p.value_ptr.* = temp;
+            }
+        }
+    }
+    var op = std.ArrayList(constop_posts).init(allocator);
+    for (0..parsed.value.len) |post_index| {
+        var top_5 = std.ArrayList(*post).init(allocator);
+        var rel_posts_tags: [6000]usize = [_]usize{0} ** 6000;
+        for (parsed.value[post_index].tags) |tag| {
+            const tag_posts = map.getPtr(tag).?;
+            for (0..tag_posts.*.items.len) |i_t| {
+                if (tag_posts.*.items[i_t] != post_index) rel_posts_tags[tag_posts.*.items[i_t]] += 1;
+            }
+        }
+        var n: usize = 5;
+        while (n > 0) : (n -= 1) {
+            const index = std.sort.argMax(usize, &rel_posts_tags, {}, lessthan).?;
+            try top_5.append(&parsed.value[index]);
+            rel_posts_tags[index] = 0;
+        }
+
+        try op.append(.{ ._id = &parsed.value[post_index]._id, .tags = &parsed.value[post_index].tags, .related = try top_5.toOwnedSlice() });
+    }
+    const end = try std.time.Instant.now();
+    try stdout.print("Processing time (w/o IO): {d}ms\n", .{@divFloor(end.since(start), std.time.ns_per_ms)});
+    const op_file = try std.fs.cwd().createFile("../related_posts_zig.json", .{});
+    defer op_file.close();
+    try std.json.stringify(try op.toOwnedSlice(), .{}, op_file.writer());
+}