From 55524c8d9115736658384a37388f208660c9cb68 Mon Sep 17 00:00:00 2001
From: Dario Nieuwenhuis <dirbaio@dirbaio.net>
Date: Mon, 28 Oct 2024 19:16:00 +0100
Subject: [PATCH] Hex-encode defmt symbols to avoid compatibility issues.

defmt currently puts json as-is in symbol names, which can contain special
characters not normally found in symbol names like quotes `"`, braces `{}`
and spaces ` `.

This can cause compatibility issues with language features or tools that don't
expect this. For example it breaks `sym` in `asm!`.

This is a *breaking change* of the wire format, so this PR increases the
format numbre. `defmt-decoder` is updated to be able to decode the new format,
while keeping ability to decode older formats.
---
 decoder/src/elf2table/symbol.rs | 29 ++++++++++++++++++++++++++++-
 decoder/src/lib.rs              |  4 ++--
 defmt/src/lib.rs                |  2 +-
 macros/src/construct/symbol.rs  | 13 +++++++++++--
 4 files changed, 42 insertions(+), 6 deletions(-)

diff --git a/decoder/src/elf2table/symbol.rs b/decoder/src/elf2table/symbol.rs
index ce35d4e1..eeb4a416 100644
--- a/decoder/src/elf2table/symbol.rs
+++ b/decoder/src/elf2table/symbol.rs
@@ -1,3 +1,6 @@
+use std::borrow::Cow;
+
+use anyhow::bail;
 use serde::Deserialize;
 
 use crate::Tag;
@@ -38,7 +41,11 @@ pub(super) enum SymbolTag {
 
 impl Symbol {
     pub fn demangle(raw: &str) -> anyhow::Result<Self> {
-        serde_json::from_str(raw)
+        let mut raw = Cow::from(raw);
+        if let Some(s) = raw.strip_prefix("__defmt_") {
+            raw = Cow::from(hex_decode(s)?);
+        }
+        serde_json::from_str(&raw)
             .map_err(|j| anyhow::anyhow!("failed to demangle defmt symbol `{}`: {}", raw, j))
     }
 
@@ -77,3 +84,23 @@ impl Symbol {
         self.crate_name.as_deref()
     }
 }
+
+fn hex_decode_digit(c: u8) -> anyhow::Result<u8> {
+    match c {
+        b'0'..=b'9' => Ok(c - b'0'),
+        b'a'..=b'f' => Ok(c - b'a' + 0xa),
+        _ => bail!("invalid hex char '{c}'"),
+    }
+}
+
+fn hex_decode(s: &str) -> anyhow::Result<String> {
+    let s = s.as_bytes();
+    if s.len() % 2 != 0 {
+        bail!("invalid hex: length must be even")
+    }
+    let mut res = vec![0u8; s.len() / 2];
+    for i in 0..(s.len() / 2) {
+        res[i] = (hex_decode_digit(s[i * 2])? << 4) | hex_decode_digit(s[i * 2 + 1])?;
+    }
+    Ok(String::from_utf8(res)?)
+}
diff --git a/decoder/src/lib.rs b/decoder/src/lib.rs
index affc01e7..9f893a20 100644
--- a/decoder/src/lib.rs
+++ b/decoder/src/lib.rs
@@ -7,10 +7,10 @@
 #![cfg_attr(docsrs, doc(cfg(unstable)))]
 #![doc(html_logo_url = "https://knurling.ferrous-systems.com/knurling_logo_light_text.svg")]
 
-pub const DEFMT_VERSIONS: &[&str] = &["3", "4"];
+pub const DEFMT_VERSIONS: &[&str] = &["3", "4", "5"];
 // To avoid a breaking change, still provide `DEFMT_VERSION`.
 #[deprecated = "Please use DEFMT_VERSIONS instead"]
-pub const DEFMT_VERSION: &str = DEFMT_VERSIONS[1];
+pub const DEFMT_VERSION: &str = DEFMT_VERSIONS[DEFMT_VERSIONS.len() - 1];
 
 mod decoder;
 mod elf2table;
diff --git a/defmt/src/lib.rs b/defmt/src/lib.rs
index abffdf8f..ec1d793e 100644
--- a/defmt/src/lib.rs
+++ b/defmt/src/lib.rs
@@ -29,7 +29,7 @@ extern crate alloc;
 #[used]
 #[cfg_attr(target_os = "macos", link_section = ".defmt,end.VERSION")]
 #[cfg_attr(not(target_os = "macos"), link_section = ".defmt.end")]
-#[export_name = "_defmt_version_ = 4"]
+#[export_name = "_defmt_version_ = 5"]
 static DEFMT_VERSION: u8 = 0;
 
 #[used]
diff --git a/macros/src/construct/symbol.rs b/macros/src/construct/symbol.rs
index 77eae0b2..f668e26e 100644
--- a/macros/src/construct/symbol.rs
+++ b/macros/src/construct/symbol.rs
@@ -52,17 +52,26 @@ impl<'a> Symbol<'a> {
     }
 
     fn mangle(&self) -> String {
-        format!(
+        let json = format!(
             r#"{{"package":"{}","tag":"{}","data":"{}","disambiguator":"{}","crate_name":"{}"}}"#,
             json_escape(&self.package),
             json_escape(&self.tag),
             json_escape(self.data),
             self.disambiguator,
             json_escape(&self.crate_name),
-        )
+        );
+        format!("__defmt_{}", hex_encode(&json))
     }
 }
 
+fn hex_encode(string: &str) -> String {
+    let mut res = String::new();
+    for &b in string.as_bytes() {
+        write!(&mut res, "{b:02x}").unwrap();
+    }
+    res
+}
+
 fn json_escape(string: &str) -> String {
     let mut escaped = String::new();
     for c in string.chars() {