)]}'
{"src/openvpn/crypto.c":[{"author":{"_account_id":1000003,"name":"Arne Schwabe","display_name":"Arne Schwabe","email":"arne-openvpn@rfc2549.org","username":"plaisthos"},"change_message_id":"d0cd678794178aa1b6e3fc155d90f334d327d63c","unresolved":true,"context_lines":[{"line_number":1474,"context_line":"                    if (hb_index \u003d\u003d 2)"},{"line_number":1475,"context_line":"                    {"},{"line_number":1476,"context_line":"                        uint8_t u;"},{"line_number":1477,"context_line":"                        ASSERT(sscanf((const char *)hex_byte, \"%\" SCNx8, \u0026u) \u003d\u003d 1);"},{"line_number":1478,"context_line":"                        *out++ \u003d u;"},{"line_number":1479,"context_line":"                        hb_index \u003d 0;"},{"line_number":1480,"context_line":"                        if (++count \u003d\u003d keylen)"}],"source_content_type":"text/x-csrc","patch_set":2,"id":"3dd58ae2_6f546134","side":"PARENT","line":1477,"updated":"2026-07-01 11:30:47.000000000","message":"I am confused about this one. Does that mean we have a toolchain that defines SCNx8 as hh but then cannot work with that?","commit_id":"712f3d61488a47e80577112489e9e9f98a32aac6"},{"author":{"_account_id":1000006,"name":"Heiko Hund","display_name":"Heiko Hund","email":"heiko@ist.eigentlich.net","username":"d12fk"},"change_message_id":"92e74a94dc30056ce3b352ddc180d6efb16d3cb2","unresolved":false,"context_lines":[{"line_number":1474,"context_line":"                    if (hb_index \u003d\u003d 2)"},{"line_number":1475,"context_line":"                    {"},{"line_number":1476,"context_line":"                        uint8_t u;"},{"line_number":1477,"context_line":"                        ASSERT(sscanf((const char *)hex_byte, \"%\" SCNx8, \u0026u) \u003d\u003d 1);"},{"line_number":1478,"context_line":"                        *out++ \u003d u;"},{"line_number":1479,"context_line":"                        hb_index \u003d 0;"},{"line_number":1480,"context_line":"                        if (++count \u003d\u003d keylen)"}],"source_content_type":"text/x-csrc","patch_set":2,"id":"a49a3e2c_5f20a54c","side":"PARENT","line":1477,"in_reply_to":"3dd58ae2_6f546134","updated":"2026-07-02 22:09:34.000000000","message":"Certainly true for MinGW on 26.04. There is a uLibc, that supports %hhx. Unfortunately MinGW isn\u0027t linked with it, but MSVCRT, which doesn\u0027t support %hhx. That is true for Ubuntu. Debian also provides MinGW with uLibc. Ubunutu doesn\u0027t have those packages for some reason. So the situation is that the headers are shared (it seems) between both libc implementations, while the compiler is hard liked to one of them. Which one the headers do not seem to care about much.\n\nThere seems to be some compile time flag, which makes MinGW use the glibc scanf/printf format string parser, which does support C99 as well, but when you have to recompile going uLibc is the saner approach.\n\nEven more sane is to skip on use of unsupported C99 features until this mess is resolved in all distros. uLibc seems to be the standard today already with MSVC.","commit_id":"712f3d61488a47e80577112489e9e9f98a32aac6"},{"author":{"_account_id":1000001,"name":"Frank Lichtenheld","display_name":"Frank Lichtenheld","email":"frank@lichtenheld.com","username":"flichtenheld","status":"OpenVPN Inc."},"change_message_id":"ea99ce7faad4d6654d370f1d0e53b219ec175ba3","unresolved":false,"context_lines":[{"line_number":1474,"context_line":"                    if (hb_index \u003d\u003d 2)"},{"line_number":1475,"context_line":"                    {"},{"line_number":1476,"context_line":"                        uint8_t u;"},{"line_number":1477,"context_line":"                        ASSERT(sscanf((const char *)hex_byte, \"%\" SCNx8, \u0026u) \u003d\u003d 1);"},{"line_number":1478,"context_line":"                        *out++ \u003d u;"},{"line_number":1479,"context_line":"                        hb_index \u003d 0;"},{"line_number":1480,"context_line":"                        if (++count \u003d\u003d keylen)"}],"source_content_type":"text/x-csrc","patch_set":2,"id":"7e2808a9_614d1954","side":"PARENT","line":1477,"in_reply_to":"a49a3e2c_5f20a54c","updated":"2026-07-03 11:53:04.000000000","message":"After researching this myself a bit I agree with both the sentiment that this is a mess and with Heiko\u0027s assessment that a work-around is currently the easiest solution.","commit_id":"712f3d61488a47e80577112489e9e9f98a32aac6"}]}
