parser: Sequencia de cerrada de ATX Heading

cli.zig

@@ -4,7 +4,7 @@ const aaronsw = @import("main.zig");
 pub fn main() !void {
     var gpa = std.heap.GeneralPurposeAllocator(.{}){};
     defer _ = gpa.deinit();
-    const allocator = &gpa.allocator;
+    const allocator = gpa.allocator();
 
     const str = try std.io.getStdIn().reader().readAllAlloc(allocator, 10241024);
     defer allocator.free(str);

commonmark-spec

@@ -1 +1 @@
-Subproject commit 499ebbad90163881f51498c4c620652d0c66fb2e
+Subproject commit 3e26d562b76deaa634676b86ebdeeb2ed0e419d2

parser.zig

@@ -37,20 +37,20 @@ const Document = struct {
 };
 const Paragraph = struct {
     text: []u8,
-    pub fn deinit(self: *Paragraph, allocator: *std.mem.Allocator) void {
+    pub fn deinit(self: *Paragraph, allocator: std.mem.Allocator) void {
         allocator.free(self.text);
     }
 };
 const Heading = struct {
     level: u8,
     text: []u8,
-    pub fn deinit(self: *Heading, allocator: *std.mem.Allocator) void {
+    pub fn deinit(self: *Heading, allocator: std.mem.Allocator) void {
         allocator.free(self.text);
     }
 };
 const BlockQuote = struct {
     children: Children,
-    pub fn deinit(self: *BlockQuote, _: *std.mem.Allocator) void {
+    pub fn deinit(self: *BlockQuote, _: std.mem.Allocator) void {
         deinitChildren(&self.children);
     }
 };
@@ -61,13 +61,13 @@ const List = struct {
     kind: Kind,
     tight: bool,
     children: Children,
-    pub fn deinit(self: *List, _: *std.mem.Allocator) void {
+    pub fn deinit(self: *List, _: std.mem.Allocator) void {
         deinitChildren(&self.children);
     }
 };
 const ListItem = struct {
     children: Children,
-    pub fn deinit(self: *ListItem, _: *std.mem.Allocator) void {
+    pub fn deinit(self: *ListItem, _: std.mem.Allocator) void {
         deinitChildren(&self.children);
     }
 };
@@ -79,7 +79,7 @@ fn arrayListAppend(comptime T: type, list: *std.ArrayList(T), item: T) !*T {
 }
 
 fn reallocAppend(
-    allocator: *std.mem.Allocator,
+    allocator: std.mem.Allocator,
     alloc_str: []u8,
     append_str: []const u8,
 ) ![]u8 {
@@ -131,15 +131,79 @@ fn checkIfBlockStillMatches(line: *[]const u8, block: Block) bool {
         },
     };
 }
-fn getHeadingLevel(line: []const u8) u8 {
-    var level: u8 = 0;
-    for (line) |char| {
-        if (level == 6) break;
-        if (char == '#') level += 1 else break;
+
+const ATXHeading = struct {
+    spaces: u2,
+    level: u3,
+    closing_sequence: usize,
+    closing_sequence_spaces: usize,
+
+    pub fn content(self: ATXHeading, line: []const u8) []const u8 {
+        // It's possible it might not have a space after the heading as it can
+        // be an empty heading.
+        const has_space: u1 = if (line.len > self.spaces + self.level) 1 else 0;
+        return line[self.spaces + self.level + has_space .. line.len - self.closing_sequence - self.closing_sequence_spaces];
     }
-    return level;
+};
+
+fn parseATXHeading(line: []const u8) ?ATXHeading {
+    var spaces: u2 = 0;
+    for (line) |char| {
+        if (char == ' ' and spaces != 3)
+            spaces += 1
+        else if (char == '#')
+            break
+        else
+            return null;
+    }
+    var level: u3 = 0;
+    for (line[spaces..]) |char| {
+        if (char == '#' and level != 6)
+            level += 1
+        else if (char == ' ' or char == '\t' or char == '\n')
+            break
+        else
+            return null;
+    }
+    if (level == 0) return null;
+    var heading = ATXHeading{
+        .level = level,
+        .spaces = spaces,
+        .closing_sequence = 0,
+        .closing_sequence_spaces = 0,
+    };
+    closing_sequence_block: {
+        const rest = heading.content(line);
+        var closing_sequence: usize = 0;
+        var closing_sequence_spaces: usize = 0;
+        if (rest.len == 0) break :closing_sequence_block;
+        var i: usize = rest.len - 1;
+        while (i >= 0) : (i -= 1) {
+            const char = rest[i];
+            if (char == ' ')
+                closing_sequence_spaces += 1
+            else if (char == '#')
+                break
+            else
+                break :closing_sequence_block;
+            if (i == 0) break;
+        }
+        while (i >= 0) : (i -= 1) {
+            const char = rest[i];
+            if (char == '#')
+                closing_sequence += 1
+            else if (char == ' ' or char == '\t')
+                break
+            else
+                break :closing_sequence_block;
+            if (i == 0) break;
+        }
+        heading.closing_sequence = closing_sequence;
+        heading.closing_sequence_spaces = closing_sequence_spaces;
+    }
+    return heading;
 }
-fn checkIfBlockStarts(allocator: *std.mem.Allocator, line: *[]const u8) !?Block {
+fn checkIfBlockStarts(allocator: std.mem.Allocator, line: *[]const u8) !?Block {
     if (std.mem.startsWith(u8, line.*, ">")) {
         line.* = line.*[1..];
         const indent = detectIndent(line.*);
@@ -152,15 +216,11 @@ fn checkIfBlockStarts(allocator: *std.mem.Allocator, line: *[]const u8) !?Block
             .kind = .{ .bullet = '*' },
             .tight = false,
         } };
-    } else if (std.mem.startsWith(u8, line.*, "#")) {
-        const level = getHeadingLevel(line.*);
-        if (line.*[level..].len != 0) {
-            if (line.*[level] != ' ' or line.*[level] != '\t') return null;
-            line.* = line.*[level + 1 ..];
-        }
+    } else if (parseATXHeading(line.*)) |heading| {
+        line.* = heading.content(line.*);
         return Block{ .heading = .{
             .text = try allocator.alloc(u8, 0),
-            .level = level,
+            .level = heading.level,
         } };
     }
     return null;
@@ -193,7 +253,7 @@ fn verifyStack(stack: Stack) void {
     }
 }
 
-pub fn parse(allocator: *std.mem.Allocator, str: []const u8) !Document {
+pub fn parse(allocator: std.mem.Allocator, str: []const u8) !Document {
     var doc_block = Block{ .document = Document{ .children = Children.init(allocator) } };
     errdefer doc_block.document.deinit();
 

readme.md

@@ -19,7 +19,7 @@ Ver [cli.zig](cli.zig)
 
 ## CommonMark
 
-Actualmente: `114 passed, 538 failed, 0 errored, 0 skipped`
+Actualmente: `127 passed, 525 failed, 0 errored, 0 skipped`
 
 Para correr los tests:
 

readme_en.md

@@ -17,7 +17,7 @@ See [cli.zig](cli.zig)
 
 ## CommonMark
 
-Currently: `114 passed, 538 failed, 0 errored, 0 skipped`
+Currently: `127 passed, 525 failed, 0 errored, 0 skipped`
 
 To run the tests: