Update sdigest.h

[ggalgoczi]

Fix snprintf size argument in sdigest.h to avoid _FORTIFY_SOURCE abort in eic_cuda container as described in #213

[github-actions]

Cpp-linter Review

Used clang-format v20.1.2

Click here for the full clang-format patch

diff --git a/sysrap/sdigest.h b/sysrap/sdigest.h
index 0ac1147..e7275fc 100644
--- a/sysrap/sdigest.h
+++ b/sysrap/sdigest.h
@@ -312 +312,2 @@ inline std::string sdigest::DescRaw( unsigned char* digest16 )
-    for (int n = 0; n < 16; ++n) std::snprintf( &buf[2*n], 3, "%02x", (unsigned int)digest16[n]) ;
+    for (int n = 0; n < 16; ++n)
+        std::snprintf(&buf[2 * n], 3, "%02x", (unsigned int)digest16[n]);
@@ -326 +327,2 @@ inline std::string sdigest::Finalize(MD5_CTX& c) // static
-    for (int n = 0; n < 16; ++n) std::snprintf( &buf[2*n], 3, "%02x", (unsigned int)digest[n]) ;
+    for (int n = 0; n < 16; ++n)
+        std::snprintf(&buf[2 * n], 3, "%02x", (unsigned int)digest[n]);

Have any feedback or feature suggestions? Share it here.

outdated suggestion

[Copilot]

Pull request overview

This PR fixes an _FORTIFY_SOURCE runtime abort by correcting the std::snprintf size argument used when formatting MD5 digests into a 32-character hex string in the header-only sdigest implementation (per issue #213).

Changes:

• Adjust sdigest::DescRaw() to use a per-iteration buffer size of 3 bytes (2 hex chars + null) instead of the full buffer length.
• Adjust sdigest::Finalize() similarly to avoid passing an oversized snprintf length for sub-slices of the output buffer.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[plexoos]

LGTM. So, do you intend to propose this upstream?

[ggalgoczi]

Yes, I think it is useful for everybody.

[plexoos]

Okay, looking forward to an upstream PR for this change. Otherwise, it may be overwritten in the next sync.

[ggalgoczi]

Opened PR upstream: simoncblyth/opticks#10