At-quoted strings carry the sentinel.

2 files changed, 21 insertions, 17 deletions

diff --git a/doc/c1/1-parse.md b/doc/c1/1-parse.md
index babf160..d4c4c2e 100644
--- a/doc/c1/1-parse.md
+++ b/doc/c1/1-parse.md
@@ -176,23 +176,26 @@ is only a *datum* if it adheres to additional constraints as explained below.
 
 Strings can appear *bare* or be quoted in various ways.  A quoted string is in
 fact parsed into a pair value with a rune in the first position to identify the
-quotation variant that was parsed, and the string value in the second position.
-
-    +-----------+----------------------+
-    | Syntax    | Parse output         |
-    +-----------+----------------------+
-    | |bytes|   | (#PQSTR & <STRING>)  |
-    +-----------+----------------------+
-    | "bytes"   | (#DQSTR & <STRING>)  |
-    +-----------+----------------------+
-    | @_bytes_  | (#ATSTR & <STRING>)  |
-    +-----------+----------------------+
+quotation variant that was parsed, and the string value in the second position;
+or, in case of at-quoted strings, a special construct we will look at later.
+
+    +-----------+-----------------------------+
+    | Syntax    | Parse output                |
+    +-----------+-----------------------------+
+    | |bytes|   | (#PQSTR & <STRING>)         |
+    +-----------+-----------------------------+
+    | "bytes"   | (#DQSTR & <STRING>)         |
+    +-----------+-----------------------------+
+    | @_bytes_  | (#ATSTR <BYTE> & <STRING>)  |
+    +-----------+-----------------------------+
 
 The visual token `<STRING>` denotes the actual string, as a Zisp value, in the
-second position of the pair.
+second position of the pair.  The visual token `<BYTE>` stands for an integer
+Zisp value between 0 and 255.
 
 These external representations of strings will be explained in more detail
-further below, including backslash escape sequences allowed within.
+further below, including backslash escape sequences allowed within, and how
+exactly at-quoted strings work.
 
 Strings have a fixed length, counted in bytes.  Each byte can have any value,
 including zero (ASCII NUL).  The parser reads bytes, not Unicode characters; a
diff --git a/src/zisp/io/Parser.zig b/src/zisp/io/Parser.zig
index e29868a..a56a61b 100644
--- a/src/zisp/io/Parser.zig
+++ b/src/zisp/io/Parser.zig
@@ -481,13 +481,14 @@ fn getString(p: *Parser, comptime close: u8) !Value {
 }
 
 fn getAtString(p: *Parser) !Value {
-    const sentinel = try p.readNoEof("at-string");
+    const stop = try p.readNoEof("at-string");
     while (try p.readNoEof2("at-string")) |c| {
-        if (c == sentinel) break;
+        if (c == stop) break;
         try p.addChar(c);
     }
-    const s = try p.getCharsAsString();
-    return p.cons(ATSTR, s);
+    const str = try p.getCharsAsString();
+    const byte = value.fixnum.pack(stop);
+    return p.cons(ATSTR, p.cons(byte, str));
 }
 
 fn skipStringLfEscape(p: *Parser) !u8 {