classification.zig to use srf parser

src/models/classification.zig

@@ -10,6 +10,7 @@
 ///   symbol::02315N600,asset_class::International Developed,pct:num:20
 ///   symbol::02315N600,asset_class::Bonds,pct:num:15
 const std = @import("std");
+const srf = @import("srf");
 
 /// A single classification entry for a symbol.
 pub const ClassificationEntry = struct {
@@ -41,7 +42,7 @@ pub const ClassificationMap = struct {
 };
 
 /// Parse a metadata SRF file into a ClassificationMap.
-/// Each line is: symbol::<SYM>,sector::<S>,geo::<G>,asset_class::<A>,pct:num:<P>
+/// Each record has: symbol::<SYM>,sector::<S>,geo::<G>,asset_class::<A>,pct:num:<P>
 /// All fields except symbol are optional. pct defaults to 100.
 pub fn parseClassificationFile(allocator: std.mem.Allocator, data: []const u8) !ClassificationMap {
     var entries = std.ArrayList(ClassificationEntry).empty;
@@ -55,36 +56,32 @@ pub fn parseClassificationFile(allocator: std.mem.Allocator, data: []const u8) !
         entries.deinit(allocator);
     }
 
-    var line_iter = std.mem.splitScalar(u8, data, '\n');
+    var reader = std.Io.Reader.fixed(data);
-    while (line_iter.next()) |line| {
+    const parsed = srf.parse(&reader, allocator, .{ .alloc_strings = false }) catch return error.InvalidData;
-        const trimmed = std.mem.trim(u8, line, &std.ascii.whitespace);
+    defer parsed.deinit();
-        if (trimmed.len == 0 or trimmed[0] == '#') continue;
-        if (std.mem.startsWith(u8, trimmed, "#!")) continue;
 
-        // Parse comma-separated key::value pairs
+    for (parsed.records.items) |record| {
         var symbol: ?[]const u8 = null;
         var sector: ?[]const u8 = null;
         var geo: ?[]const u8 = null;
         var asset_class: ?[]const u8 = null;
         var pct: f64 = 100.0;
 
-        var field_iter = std.mem.splitScalar(u8, trimmed, ',');
+        for (record.fields) |field| {
-        while (field_iter.next()) |field| {
+            if (std.mem.eql(u8, field.key, "symbol")) {
-            const f = std.mem.trim(u8, field, &std.ascii.whitespace);
+                if (field.value) |v| symbol = strVal(v);
-            if (std.mem.startsWith(u8, f, "symbol::")) {
+            } else if (std.mem.eql(u8, field.key, "sector")) {
-                symbol = f["symbol::".len..];
+                if (field.value) |v| sector = strVal(v);
-            } else if (std.mem.startsWith(u8, f, "sector::")) {
+            } else if (std.mem.eql(u8, field.key, "geo")) {
-                sector = f["sector::".len..];
+                if (field.value) |v| geo = strVal(v);
-            } else if (std.mem.startsWith(u8, f, "geo::")) {
+            } else if (std.mem.eql(u8, field.key, "asset_class")) {
-                geo = f["geo::".len..];
+                if (field.value) |v| asset_class = strVal(v);
-            } else if (std.mem.startsWith(u8, f, "asset_class::")) {
+            } else if (std.mem.eql(u8, field.key, "pct")) {
-                asset_class = f["asset_class::".len..];
+                if (field.value) |v| pct = numVal(v);
-            } else if (std.mem.startsWith(u8, f, "pct:num:")) {
-                pct = std.fmt.parseFloat(f64, f["pct:num:".len..]) catch 100.0;
             }
         }
 
-        const sym = symbol orelse continue; // skip lines without symbol
+        const sym = symbol orelse continue;
         try entries.append(allocator, .{
             .symbol = try allocator.dupe(u8, sym),
             .sector = if (sector) |s| try allocator.dupe(u8, s) else null,
@@ -100,6 +97,20 @@ pub fn parseClassificationFile(allocator: std.mem.Allocator, data: []const u8) !
     };
 }
 
+fn strVal(v: srf.Value) ?[]const u8 {
+    return switch (v) {
+        .string => |s| s,
+        else => null,
+    };
+}
+
+fn numVal(v: srf.Value) f64 {
+    return switch (v) {
+        .number => |n| n,
+        else => 100.0,
+    };
+}
+
 test "parse classification file" {
     const data =
         \\#!srfv1