Ada Bindings and CI Improvements by julek-wolfssl · Pull Request #9617 · wolfSSL/wolfssl

julek-wolfssl · 2026-01-06T12:24:57Z

Add new crypto support (RSA, SHA, AES, RNG)
CI and Testing Enhancements
- Added GNATprove static analysis and memory tracking to CI.
- Added Ada unit tests using AUnit.
- Integrated Ada AUnit tests and valgrind suppressions for leak checks.
- Updated Ada test harness, reduced boilerplate, and improved suite registration.
Ada Bindings and Examples
- Refactored AES, RSA, RNG, and SHA256 objects to use dynamic allocation.
- Added and improved client-server and cryptographic examples (AES, RSA, SHA256).
- Moved examples to a dedicated sub-directory and updated related build files.
Documentation and Configuration
- Updated README and added documentation for running tests with valgrind and using alr exec.
Code Quality and Maintenance
- Fixed warnings, improved style, and added ownership/postcondition annotations for GNATprove.
- Ensured GNATprove is clean on examples.gpr.

julek-wolfssl · 2026-01-06T14:38:20Z

Retest this please AgentOffline

Copilot

Pull request overview

This pull request adds Ada bindings and cryptographic support (RSA, SHA, AES, RNG) while enhancing CI testing and code quality through GNATprove static analysis, memory tracking, and AUnit tests. The PR also reorganizes examples into a dedicated subdirectory and improves documentation.

Key changes:

Added cryptographic bindings (RSA, SHA256, AES, RNG) with dynamic allocation
Introduced AUnit test framework and valgrind memory checks
Moved examples to examples/ subdirectory with dedicated build configuration
Added GNATprove annotations for ownership checking and memory safety

Reviewed changes

Copilot reviewed 40 out of 46 changed files in this pull request and generated no comments.

Show a summary per file

File	Description
wrapper/Ada/wolfssl.gpr	Removed exclusion of TLS client/server sources (now in examples/)
wrapper/Ada/wolfssl.ads	Added crypto bindings (RSA, SHA256, AES, RNG) with ownership annotations
wrapper/Ada/wolfssl.adb	Implemented crypto bindings with dynamic allocation and exception handling
wrapper/Ada/wolfssl-full_runtime.ads	New package for full runtime features (GNAT.Sockets, PSK callbacks)
wrapper/Ada/wolfssl-full_runtime.adb	Implementation of full runtime features
wrapper/Ada/tls_server.adb	Updated for new API (Read/Write procedures, moved PSK to Full_Runtime)
wrapper/Ada/tls_client.ads	Fixed indentation consistency
wrapper/Ada/tls_client.adb	Updated for new API and moved examples
wrapper/Ada/tests/*	Added AUnit test infrastructure and test cases
wrapper/Ada/examples/*	Moved examples to subdirectory with new build configuration
wrapper/Ada/ada_binding.c	Added C wrapper functions for crypto operations
.github/workflows/ada.yml	Enhanced CI with alire, tests, valgrind, and GNATprove

Comments suppressed due to low confidence (6)

wrapper/Ada/wolfssl.ads:1

The Terminator_Error exception is declared but never documented. Add a comment explaining when this exception is raised and how it should be handled.
wrapper/Ada/wolfssl.ads:1
The comment says 'RSA' but should say 'RNG' since this is the Is_Valid function for RNG_Key_Type.
wrapper/Ada/wolfssl.adb:1
The constant name nul could be more descriptive. Consider renaming to Null_Terminator or Nul_Byte to better indicate its purpose as a null terminator character.
wrapper/Ada/wolfssl.adb:1
Magic number -121212 is used as a default error value in multiple places. Define this as a named constant (e.g., Default_Error_Value) to improve maintainability and clarify its purpose.
wrapper/Ada/wolfssl.adb:1
Magic number -121212 is used as a default error value in multiple places. Define this as a named constant (e.g., Default_Error_Value) to improve maintainability and clarify its purpose.
wrapper/Ada/wolfssl.adb:1
Magic number -121212 is used as a default error value in multiple places. Define this as a named constant (e.g., Default_Error_Value) to improve maintainability and clarify its purpose.

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

JacobBarthelmeh

One minor comment about using XMALLOC over malloc. In the future we may need something like the static memory feature so starting out with XMALLOC would help for if using a heap hint later.

Thank you for adding in the test cases. I compiled both the examples and tests and they ran okay with a newer version of gnat, not sure about old versions of gnat.

bash-3.2$ alr run
ⓘ Building tests=0.1.0-dev/tests.gpr...                              
gprbuild: "tests" up to date
✓ Build finished successfully in 0.87 seconds.

OK SHA256('asdf') produces expected hash
OK SHA256('') produces expected hash
OK RSA sign/verify and encrypt/decrypt (rsa_verify_main)
OK AES-CBC encrypt/decrypt roundtrip
OK AES_Free succeeds after Create_AES

Total Tests Run:   5
Successful Tests:  5
Failed Assertions: 0

wrapper/Ada/ada_binding.c

julek-wolfssl · 2026-01-08T11:44:43Z

Pull request overview

...

wrapper/Ada/wolfssl.ads:1

* The `Terminator_Error` exception is declared but never documented. Add a comment explaining when this exception is raised and how it should be handled.
  **wrapper/Ada/wolfssl.ads:1**

* The comment says 'RSA' but should say 'RNG' since this is the `Is_Valid` function for `RNG_Key_Type`.
  **wrapper/Ada/wolfssl.adb:1**

* The constant name `nul` could be more descriptive. Consider renaming to `Null_Terminator` or `Nul_Byte` to better indicate its purpose as a null terminator character.
  **wrapper/Ada/wolfssl.adb:1**

* Magic number -121212 is used as a default error value in multiple places. Define this as a named constant (e.g., `Default_Error_Value`) to improve maintainability and clarify its purpose.
  **wrapper/Ada/wolfssl.adb:1**

* Magic number -121212 is used as a default error value in multiple places. Define this as a named constant (e.g., `Default_Error_Value`) to improve maintainability and clarify its purpose.
  **wrapper/Ada/wolfssl.adb:1**

* Magic number -121212 is used as a default error value in multiple places. Define this as a named constant (e.g., `Default_Error_Value`) to improve maintainability and clarify its purpose.

Fixed.

JacobBarthelmeh · 2026-01-08T16:28:25Z

Retest this please Jenkins

Cannot invoke method getEnvironment() FIPS rolling release

rizlik · 2026-01-08T17:31:00Z

very nice work. I didn't review tests

wrapper/Ada/wolfssl.ads

wrapper/Ada/wolfssl.adb

….Strings and GNAT.Sockets into a child package of the WolfSSL package.

- annotate Is_Valid functions to allow gnatprove to detect leaks

…unused exception, and clean up result initialization in wolfssl.adb

julek-wolfssl · 2026-01-22T09:59:51Z

Retest this please StreamCorruptedException

douzzer · 2026-02-05T18:16:06Z

retest this please
(lost logs/state)

douzzer · 2026-02-05T19:11:43Z

retest this please

ERROR: Checkout failed
hudson.remoting.ProxyException: java.util.concurrent.TimeoutException: Ping started at 1770315636604 hasn't completed by 1770315876604
	at hudson.remoting.PingThread.ping(PingThread.java:135)

dgarske · 2026-02-06T17:55:10Z

Jenkins retest this please. History lost

Copilot

Pull request overview

Copilot reviewed 41 out of 47 changed files in this pull request and generated 5 comments.

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

Copilot · 2026-02-06T18:10:55Z

wrapper/Ada/wolfssl.adb

 with Ada.Unchecked_Conversion;
-pragma Warnings (Off, "* is an internal GNAT unit");
-with GNAT.Sockets.Thin_Common;
-pragma Warnings (On, "* is an internal GNAT unit");
-with Interfaces.C.Extensions;
-with Interfaces.C.Strings;
-with System;
+with WolfSSL;



wrapper/Ada/wolfssl.adb is the body of package WolfSSL, but it has a with WolfSSL; clause. A unit cannot with itself; this will cause a circular dependency / compilation error. Remove the self-with and instead with the actual dependencies needed by the body (or rely on the spec’s withs where appropriate).

Copilot · 2026-02-06T18:10:55Z

wrapper/Ada/tests/src/sha256_bindings_tests.adb

+         AUnit.Assertions.Assert
+           (Hash_Bytes = Expected_Hash,
+            "SHA256('asdf') hash mismatch");


Expected_Hash returned by Hex_Bytes is indexed 0 .. 31, but Hash_Bytes is indexed 1 .. 32. For Interfaces.C.char_array, predefined equality includes bounds, so this comparison will fail even when the bytes match. Normalize bounds (e.g., make both arrays use the same range) before comparing.

Copilot · 2026-02-06T18:10:56Z

wrapper/Ada/tests/src/aes_bindings_tests.adb

+      AUnit.Assertions.Assert
+        (Decoded = Plain,
+         "AES-CBC roundtrip mismatch");


Plain comes from Test_Support.Bytes (0-based Byte_Array), but Decoded is declared Byte_Array (1 .. Plain'Length), so Decoded = Plain will fail due to different bounds even if the data matches. Declare Cipher/Decoded with Plain'Range (or otherwise normalize ranges) so equality is meaningful.

Copilot · 2026-02-06T18:10:56Z

wrapper/Ada/tests/src/support/test_support.adb

+      Last_N : constant Natural := (if S'Length = 0 then 0 else S'Length - 1);
+      B      : WolfSSL.Byte_Array (0 .. WolfSSL.Byte_Index (Last_N));
+      N      : Natural := 0;


For empty input, Last_N is set to 0 and B becomes Byte_Array (0 .. 0), which is a 1-byte array rather than an empty buffer. If callers intend Bytes("") to represent an empty C buffer, consider returning a null range (e.g. 1 .. 0) for the empty-string case.

Copilot · 2026-02-06T18:10:56Z

wrapper/Ada/tests/src/support/test_support.adb

+      declare
+         N      : constant Natural := Len / 2;
+         Last_N : constant Natural := (if N = 0 then 0 else N - 1);
+         B      : WolfSSL.Byte_Array (0 .. WolfSSL.Byte_Index (Last_N));
+         Hi     : Natural;


When Hex is empty, this constructs Byte_Array (0 .. 0) (1 byte) instead of an empty buffer. Consider returning a null range for N = 0 so Hex_Bytes("") is truly empty, and to avoid bound-mismatch surprises in callers.

dgarske · 2026-02-06T18:45:55Z

@julek-wolfssl please squash the 51 commits. Also review the latest copilot findings. Thanks

julek-wolfssl self-assigned this Jan 6, 2026

julek-wolfssl requested a review from JacobBarthelmeh January 6, 2026 12:24

julek-wolfssl force-pushed the ada-testing branch from 7b03d32 to 7fe01ba Compare January 6, 2026 12:41

julek-wolfssl requested a review from rizlik January 6, 2026 14:02

julek-wolfssl assigned JacobBarthelmeh and unassigned julek-wolfssl Jan 6, 2026

JacobBarthelmeh requested a review from Copilot January 7, 2026 21:40

Copilot started reviewing on behalf of JacobBarthelmeh January 7, 2026 21:41 View session

Copilot AI reviewed Jan 7, 2026

View reviewed changes

JacobBarthelmeh reviewed Jan 7, 2026

View reviewed changes

wrapper/Ada/ada_binding.c Outdated Show resolved Hide resolved

JacobBarthelmeh assigned julek-wolfssl and unassigned JacobBarthelmeh Jan 7, 2026

julek-wolfssl requested a review from JacobBarthelmeh January 8, 2026 10:53

julek-wolfssl assigned JacobBarthelmeh and unassigned julek-wolfssl Jan 8, 2026

JacobBarthelmeh previously approved these changes Jan 8, 2026

View reviewed changes

rizlik reviewed Jan 8, 2026

View reviewed changes

wrapper/Ada/wolfssl.ads Outdated Show resolved Hide resolved

wrapper/Ada/wolfssl.ads Outdated Show resolved Hide resolved

wrapper/Ada/wolfssl.ads Show resolved Hide resolved

wrapper/Ada/wolfssl.adb Outdated Show resolved Hide resolved

wrapper/Ada/wolfssl.adb Outdated Show resolved Hide resolved

julek-wolfssl assigned julek-wolfssl and unassigned JacobBarthelmeh Jan 9, 2026

julek-wolfssl dismissed JacobBarthelmeh’s stale review via 1621908 January 20, 2026 15:45

julek-wolfssl force-pushed the ada-testing branch from 5ee8987 to 1621908 Compare January 20, 2026 15:45

julek-wolfssl requested a review from rizlik January 20, 2026 15:45

julek-wolfssl assigned rizlik and unassigned julek-wolfssl Jan 20, 2026

joakim-strandberg added 2 commits January 22, 2026 10:10

Refactored Ada binding. Separated code that depends upon Interfaces.C…

caa3d0a

….Strings and GNAT.Sockets into a child package of the WolfSSL package.

Added Ada binding for the SHA256 implementation in the Wolfssl library.

053bfaa

julek-wolfssl added 17 commits January 22, 2026 10:10

Ada: update readme

597e633

Ada: fix the client-server examples

972ee56

Ada: add examples to ada.yml

039e08b

Ada: leverage gnatprove to track memory

d8511f8

- annotate Is_Valid functions to allow gnatprove to detect leaks

Ada: fix length proofs in gnatprove

56ac0a6

Ada: add gnatprove to CI

b0fea88

Ada: update editorconfig

6f11509

Ada: update gitignore to match others

9a07720

Ada: fix default.gpr

ac2c7e5

Ada: build default.gpr in CI

515c8e6

Ada: fix tls_client call in CI

7ac2700

Ada: use XMALLOC and XFREE for dynamic allocation

56d6d28

Ada wrapper: update .gitignore tracking, clarify RNG comment, remove …

b533ac8

…unused exception, and clean up result initialization in wolfssl.adb

Ada: RNG_Key_Type => RNG_Type

22b98bc

Ada: Use nul for clarity

23a937f

Ada: Remove Text from Finalize_SHA256

c36e33b

Ada: fix typo

af9086a

julek-wolfssl force-pushed the ada-testing branch from 1621908 to af9086a Compare January 22, 2026 09:10

rizlik approved these changes Feb 2, 2026

View reviewed changes

julek-wolfssl assigned wolfSSL-Bot and unassigned rizlik Feb 3, 2026

dgarske requested a review from Copilot February 6, 2026 17:56

Copilot started reviewing on behalf of dgarske February 6, 2026 17:58 View session

Copilot AI reviewed Feb 6, 2026

View reviewed changes

dgarske assigned julek-wolfssl Feb 6, 2026

Conversation

julek-wolfssl commented Jan 6, 2026

Uh oh!

julek-wolfssl commented Jan 6, 2026

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

JacobBarthelmeh left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

julek-wolfssl commented Jan 8, 2026

Pull request overview

Uh oh!

JacobBarthelmeh commented Jan 8, 2026

Uh oh!

rizlik commented Jan 8, 2026

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

julek-wolfssl commented Jan 22, 2026

Uh oh!

douzzer commented Feb 5, 2026

Uh oh!

douzzer commented Feb 5, 2026

Uh oh!

dgarske commented Feb 6, 2026

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Copilot AI Feb 6, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 6, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 6, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 6, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 6, 2026

Choose a reason for hiding this comment

Uh oh!

dgarske commented Feb 6, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

7 participants