lint + static analysis: clang-tidy and cppcheck fixes#5594
Conversation
…po diagnostics during normalization
…lude/lib in normalize-clang-tidy.py
- MySQL_Thread.h: wrap negative hostgroup constants in parens - PgSQL_Thread.h: wrap negative hostgroup constants in parens - proxysql_admin.h: wrap expression multipliers in parens - proxysql_debug.h: wrap macro args in ASSERT_SQLITE* macros - proxysql_macros.h: wrap macro args in strdup_null, free_null, itostr, Copy4B - proxysql_structs.h: wrap macro args in IS_PGTRACKED_VAR_OPTION_* macros
Use static_cast<int> instead of C-style cast for clarity and to avoid unsigned->signed narrowing warnings while preserving the -1 sentinel.
Replace __CLASS_* and ____CLASS_* patterns with PROXYSQL_*_H to comply with the C++ standard where identifiers starting with __ are reserved in the global namespace. Also fixes: - ____CLASS_STANDARD_MYSQL_THREAD_H -> PROXYSQL_STANDARD_MYSQL_THREAD_H - ____CLASS_STANDARD_PGSQL_THREAD_H -> PROXYSQL_STANDARD_PGSQL_THREAD_H - __CLASS_PROXYSQL_POLL -> PROXYSQL_POLL_H (added _H for consistency)
Rename ROS__SIZE -> ROS_SIZE_, RLS__SIZE -> RLS_SIZE_, etc. to avoid the reserved identifier pattern (double underscore at start or in name).
Rename __active -> active_, __frontend -> frontend_, __backend -> backend_ in authentication and hostgroups classes to comply with C++ reserved identifier rules. Includes updates to: - MySQL_Authentication.hpp - ClickHouse_Authentication.hpp - PgSQL_Authentication.h - MySQL_HostGroups_Manager.h - Base_HostGroups_Manager.h - Corresponding .cpp files
performance-enum-size fixes: - p_mysql_error_type: add uint8_t underlying type (2 values) - p_hg_dyn_gauge: add uint8_t underlying type (5 values) - p_hg_dyn_counter: add uint8_t underlying type (9 values) bugprone-reserved-identifier fixes: - Rename __size -> SIZE_ in metric enums to avoid reserved identifier (names starting with __ are reserved in C++) - Update all std::array size references using these enum values
…hostgroups performance-enum-size fixes: - p_mysql_error_type: add uint8_t underlying type (2 values) - p_pgsql_error_type: add uint8_t underlying type (2 values) - p_hg_dyn_gauge: add uint8_t underlying type (5 values) - p_hg_dyn_counter: add uint8_t underlying type (9 values) - PgSQL_p_hg_dyn_gauge: add uint8_t underlying type (5 values) - PgSQL_p_hg_dyn_counter: add uint8_t underlying type (9 values) bugprone-reserved-identifier fixes: - Rename __size -> SIZE_ in all metric enums to avoid reserved identifier - Update std::array size references to use new SIZE_ enum values Applies same fixes as MySQL_HostGroups_Manager.h to: - Base_HostGroups_Manager.h (base class) - PgSQL_HostGroups_Manager.h (PostgreSQL implementation)
…, admin, thread, debug performance-enum-size fixes: - p_qc_counter, p_qc_gauge: add uint8_t (7 and 2 values) - p_admin_counter, p_admin_gauge, p_admin_dyn_counter, p_admin_dyn_gauge: add uint8_t - p_debug_dyn_counter: add uint8_t (2 values) bugprone-reserved-identifier fixes: - Rename __size -> SIZE_ in all metric enums - Update std::array size references across multiple files - Update prometheus_helpers.h template parameter names and comments Files changed: - Query_Cache.h - proxysql_admin.h - MySQL_Thread.h - proxysql_debug.h - prometheus_helpers.h (template changes) - debug.cpp (usage update)
bugprone-reserved-identifier fixes: - gen_utils.h: rename __size parameter to 'sz' in PtrArray and PtrSizeArray constructors (parameter names starting with __ are reserved) - prometheus_helpers.h: update documentation comments to reference '::SIZE_' instead of '::__size' to match the actual enum values
modernize-use-nullptr: replace NULL with nullptr for type-safe null pointer constants. This is a C++11 modernization that improves type safety. Files changed: - Base_HostGroups_Manager.h - ClickHouse_Authentication.hpp - MySQL_HostGroups_Manager.h - MySQL_LDAP_Authentication.hpp - MySQL_Monitor.hpp - MySQL_Session.h - PgSQL_Error_Helper.h - PgSQL_HostGroups_Manager.h - PgSQL_Session.h
Remove lint output files from git tracking - these are generated files that should not be committed. Keep only SUPPRESSIONS.md and .gitkeep.
- Rename __HGM_TABLES_SIZE -> HGM_TABLES_SIZE_ (Base/MySQL/PgSQL HGM) - Rename __TYPE_SIZE -> TYPE_SIZE_ (Base/MySQL/PgSQL HGM) - Fix include guard defines: __PROXYSQL_* -> PROXYSQL_* - Fix include guard comments: __CLASS_* -> PROXYSQL_*_H (29 files) - Add uint8_t underlying type to small enums
- Rename __HGM_TABLES_SIZE -> HGM_TABLES_SIZE_ (Base/MySQL/PgSQL HGM) - Rename __TYPE_SIZE -> TYPE_SIZE_ (Base/MySQL/PgSQL HGM) - Fix include guard defines: __PROXYSQL_* -> PROXYSQL_* - Fix include guard comments: __CLASS_* -> PROXYSQL_*_H (29 files) - Add uint8_t underlying type to small enums
…era_Info, AWS_Aurora_Info, MySQL_Errors_stats)
- Fix invalidPrintfArgType: %d->%u for hid, %u->%d for use_ssl - Fix nullPointerRedundantCheck: replace assert with continue - Add deleted copy ctor/operator= to KillArgs and MySQL_Session - Fix invalidFunctionArg: add query_len > 3 check before memcpy
- Fix wrongPrintfScanfArgNum: remove extra args in 4 sprintf calls - Fix nullPointer: add GloMTH check, add null checks for strdup args - All high-risk cppcheck warnings now resolved
…unds - Fix invalidLifetime: look up pointer from map after insertion, not before - Fix arrayIndexOutOfBounds: add bounds check before array access
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Organization UI Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (1)
✅ Files skipped from review due to trivial changes (1)
📜 Recent review details⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (5)
📝 WalkthroughWalkthroughAdds repository-wide static-analysis tooling and CI workflow: top-level Changes
Sequence Diagram(s)sequenceDiagram
participant Dev as Developer
participant Make as Makefile ("make lint")
participant ScriptGen as scripts/lint/generate-compile-commands.sh
participant Bear as bear
participant RunLocal as scripts/lint/run-local.sh
participant ClangTidy as clang-tidy
participant Cppcheck as cppcheck
participant Normalize as normalize scripts
participant LintDir as lint/ (artifacts)
Dev->>Make: run `make lint`
Make->>ScriptGen: run generate-compile-commands.sh
ScriptGen->>Bear: wrap build to produce compile_commands.json
Bear-->>ScriptGen: compile_commands.json
ScriptGen-->>Make: done
Make->>RunLocal: run run-local.sh
RunLocal->>ClangTidy: run clang-tidy per file (--export-fixes)
RunLocal->>Cppcheck: run cppcheck --project=compile_commands.json
ClangTidy->>Normalize: export YAMLs -> normalize-clang-tidy.py
Cppcheck->>Normalize: raw output -> normalize-cppcheck.py
Normalize->>LintDir: write normalized `clang-tidy.txt`, `cppcheck.txt`
LintDir-->>Dev: artifacts for review/CI upload
Estimated code review effort🎯 4 (Complex) | ⏱️ ~60 minutes Possibly related PRs
Poem
🚥 Pre-merge checks | ✅ 2 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (2 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches📝 Generate docstrings
🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
There was a problem hiding this comment.
Code Review
This pull request introduces a static analysis framework using clang-tidy and cppcheck, incorporating configuration files, implementation plans, and build system integration. The changes feature extensive codebase modernization, including the adoption of nullptr, updated header guards, and the use of initialization lists. Furthermore, several bugs were resolved, such as a use-after-move error, potential null pointer dereferences, and incorrect arguments in sprintf and fprintf calls. Feedback indicates that an unreachable logic block in the run-local.sh script should be removed.
| else | ||
| # Merge all per-file stderr clang-tidy outputs if present | ||
| if ls ${TMPD:-/dev/null}/ct-*.stderr >/dev/null 2>&1; then | ||
| cat ${TMPD}/ct-*.stderr > lint/clang-tidy-raw.txt || true | ||
| python3 scripts/lint/normalize-clang-tidy.py lint/clang-tidy-raw.txt > lint/clang-tidy.txt || true | ||
| else | ||
| # Nothing found | ||
| : | ||
| fi | ||
| fi |
There was a problem hiding this comment.
This else block appears to be unreachable. The Python script at lines 53-66 always creates lint/clang-tidy-fixes.yaml, so the condition [ -f lint/clang-tidy-fixes.yaml ] on line 78 will always evaluate to true. The logic for handling cases with no diagnostics should be handled within the normalize-clang-tidy.py script when parsing the (potentially empty) YAML file. This else block can be removed.
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 1
Note
Due to the large number of review comments, Critical severity comments were prioritized as inline comments.
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (7)
include/proxysql_restapi.h (1)
43-43:⚠️ Potential issue | 🟡 Minor
#endifcomment is outdated and inconsistent.Line 43 references
__PROXYSQL_RESTAPI_H__, but the current guard macro isPROXYSQL_RESTAPI_H. Please align the trailing comment with the actual guard used.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/proxysql_restapi.h` at line 43, The trailing comment on the include guard is inconsistent: change the outdated "__PROXYSQL_RESTAPI_H__" reference to match the actual guard macro "PROXYSQL_RESTAPI_H" used in the file; update the comment after the `#endif` to read "// PROXYSQL_RESTAPI_H" so the closing comment matches the opening guard (refer to the guard macro PROXYSQL_RESTAPI_H in the file).include/proxysql_sslkeylog.h (1)
121-121:⚠️ Potential issue | 🟡 MinorClosing guard comment no longer matches the defined macro.
Line 121 still references
__PROXYSQL_SSLKEYLOG_Hwhile the active guard isPROXYSQL_SSLKEYLOG_H.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/proxysql_sslkeylog.h` at line 121, The closing header-guard comment is inconsistent: the active macro is PROXYSQL_SSLKEYLOG_H but the trailing comment on the `#endif` uses __PROXYSQL_SSLKEYLOG_H; update the `#endif` comment to match the defined macro (PROXYSQL_SSLKEYLOG_H) so the guard name is consistent (look for the `#ifndef/`#define pair and the final `#endif` in the proxysql_sslkeylog header).include/LLM_Bridge.h (1)
337-337:⚠️ Potential issue | 🟡 MinorFix stale
#endifcomment to match the active guard.Line 337 still references
__CLASS_LLM_BRIDGE_Hwhile the active macro at Line 28 isPROXYSQL_LLM_BRIDGE_H, which makes maintenance/error-tracing harder.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/LLM_Bridge.h` at line 337, The closing preprocessor comment is stale: replace the comment on the final `#endif` so it matches the active include guard macro PROXYSQL_LLM_BRIDGE_H (not __CLASS_LLM_BRIDGE_H); update the comment after the `#endif` in include/LLM_Bridge.h to read // PROXYSQL_LLM_BRIDGE_H to keep the guard name consistent with the opening `#ifndef/`#define for PROXYSQL_LLM_BRIDGE_H.include/proxysql_coredump.h (1)
27-27:⚠️ Potential issue | 🟡 MinorNormalize trailing
#endifcomment to the actual macro.Line 27 still references
__PROXYSQL_COREDUMP_H, which is inconsistent with the current defined guard.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/proxysql_coredump.h` at line 27, The trailing endif comment currently reads "#endif // __PROXYSQL_COREDUMP_H" but doesn't match the header guard actually declared at the top of the file; update the comment on the closing `#endif` to exactly match the macro used in the `#ifndef/`#define guard (replace __PROXYSQL_COREDUMP_H with the actual guard macro name present at the top of this header) so the closing comment reflects the correct symbol.include/MySQL_Protocol.h (1)
145-151:⚠️ Potential issue | 🟠 MajorInitialize
dump_pktin the constructor under#ifdef DEBUG.Under
#ifdef DEBUG,dump_pktis declared but not initialized by the constructor orinit()method. Multiple methods read this member withif (dump_pkt) {...}checks throughout the codebase. The explicit assignment inBase_Thread.cpp::138happens only after the object is already in use, creating undefined behavior for any reads before that point.Suggested fix
MySQL_Protocol() - : userinfo(nullptr), sess(nullptr), myds(nullptr), current_PreStmt(nullptr) + : userinfo(nullptr), sess(nullptr), myds(nullptr) +#ifdef DEBUG + , dump_pkt(false) +#endif + , current_PreStmt(nullptr) { sent_auth_plugin_id = AUTH_MYSQL_NATIVE_PASSWORD; auth_plugin_id = AUTH_UNKNOWN_PLUGIN; prot_status=0; more_data_needed = false; }🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/MySQL_Protocol.h` around lines 145 - 151, The MySQL_Protocol constructor (and/or init()) must explicitly initialize the DEBUG-only member dump_pkt to a safe default to avoid undefined reads; update the MySQL_Protocol() initializer or body to set dump_pkt = nullptr (or false as appropriate) under `#ifdef` DEBUG so that any code checking "if (dump_pkt) { ... }" before Base_Thread.cpp later assigns it will behave deterministically; ensure the initialization is present in the same places where other members (userinfo, sess, myds, current_PreStmt, sent_auth_plugin_id, auth_plugin_id, prot_status, more_data_needed) are set.include/MySQL_LDAP_Authentication.hpp (1)
4-11:⚠️ Potential issue | 🟡 MinorRename enum member
__SIZEtoSIZE_to comply with reserved identifier guidelines.The enum sentinel
__SIZEat line 9 uses a reserved identifier (double underscore prefix) and should be renamed toSIZE_per the PR objectives and C++ naming conventions for constants. The member is unused in the codebase, making this refactoring safe.Suggested fix
struct LDAP_USER_FIELD_IDX { enum index { USERNAME = 0, FRONTEND_CONNECTIONS = 1, FRONTED_MAX_CONNECTIONS = 2, - __SIZE + SIZE_ }; };🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/MySQL_LDAP_Authentication.hpp` around lines 4 - 11, The enum sentinel in LDAP_USER_FIELD_IDX::index uses a reserved identifier __SIZE; rename that member to SIZE_ (replace __SIZE with SIZE_) across the enum declaration and any references to it so the identifier no longer begins with double underscore; update LDAP_USER_FIELD_IDX and any code that might reference LDAP_USER_FIELD_IDX::index::__SIZE to LDAP_USER_FIELD_IDX::index::SIZE_.lib/MySQL_Authentication.cpp (1)
17-29:⚠️ Potential issue | 🔴 CriticalBug: Duplicate
ad.passwordfree block causes potential double-free.Lines 26-29 duplicate the exact same logic as lines 18-21, freeing
ad.passwordtwice. This will cause undefined behavior (double-free) ifad.passwordis non-null.🐛 Proposed fix: Remove duplicate block
void free_account_details(account_details_t& ad) { if (ad.password) { free(ad.password); ad.password = nullptr; } if (ad.sha1_pass) { free(ad.sha1_pass); ad.sha1_pass=NULL; } - if (ad.password) { - free(ad.password); - ad.password = nullptr; - } if (ad.clear_text_password[PASSWORD_TYPE::PRIMARY]) {🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@lib/MySQL_Authentication.cpp` around lines 17 - 29, The function free_account_details contains a duplicate block that frees ad.password twice; remove the second duplicate if-check and free so ad.password is only freed once and then set to nullptr, leaving the ad.sha1_pass free logic intact; update the function free_account_details (operating on account_details_t and its members ad.password and ad.sha1_pass) to ensure each pointer is freed exactly once and nulled afterwards.
🟠 Major comments (36)
include/StatCounters.h-1-2 (1)
1-2:⚠️ Potential issue | 🟠 MajorGuard rename breaks mandated include-guard convention.
The new
PROXYSQL_STAT_COUNTERS_Hguard does not follow the required__CLASS_*_Hpattern forinclude/headers.As per coding guidelines,
include/**/*.{h,hpp}: "Include guards must follow the pattern#ifndef_CLASS*_H".Also applies to: 74-74
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/StatCounters.h` around lines 1 - 2, The include guard in include/StatCounters.h currently uses PROXYSQL_STAT_COUNTERS_H which violates the mandated pattern; rename the guard to follow the required __CLASS_*_H convention (e.g., use __CLASS_STATCOUNTERS_H or __CLASS_STAT_COUNTERS_H consistently) and update both the `#ifndef` and corresponding `#define` and the matching `#endif` comment if present; ensure the symbol used matches elsewhere in the file (replace PROXYSQL_STAT_COUNTERS_H wherever referenced) so the header conforms to the include/**/*.{h,hpp} guideline.include/AI_Vector_Storage.h-1-2 (1)
1-2:⚠️ Potential issue | 🟠 MajorHeader guard format conflicts with include-directory rule.
PROXYSQL_AI_VECTOR_STORAGE_Hdoes not match the required__CLASS_*_Hinclude-guard convention for headers underinclude/.As per coding guidelines,
include/**/*.{h,hpp}: "Include guards must follow the pattern#ifndef_CLASS*_H".🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/AI_Vector_Storage.h` around lines 1 - 2, The header guard macro PROXYSQL_AI_VECTOR_STORAGE_H does not follow the required include-guard convention for files under include/; update both occurrences of the guard (the `#ifndef` and matching `#define`) in include/AI_Vector_Storage.h to the mandated pattern using the __CLASS_*_H style (e.g., replace PROXYSQL_AI_VECTOR_STORAGE_H with an appropriately named macro like __CLASS_AI_VECTOR_STORAGE_H) and ensure the corresponding `#endif` comment matches the new macro.include/MySQL_Query_Cache.h-1-2 (1)
1-2:⚠️ Potential issue | 🟠 MajorInclude guard does not match required repository pattern.
PROXYSQL_MYSQL_QUERY_CACHE_Hconflicts with the enforced__CLASS_*_Hformat for headers ininclude/.As per coding guidelines,
include/**/*.{h,hpp}: "Include guards must follow the pattern#ifndef_CLASS*_H".Also applies to: 26-26
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/MySQL_Query_Cache.h` around lines 1 - 2, The include guard macro in this header uses PROXYSQL_MYSQL_QUERY_CACHE_H which violates the repo pattern; rename the guard to follow `#ifndef` __CLASS_*_H (e.g., __CLASS_MYSQL_QUERY_CACHE_H) and update the matching `#define` and trailing `#endif` comment to the new macro; ensure the symbol references in this file (MySQL_Query_Cache.h) are consistently replaced so the ifndef/define/endif trio uses __CLASS_MYSQL_QUERY_CACHE_H.include/PgSQL_Protocol.h-1-2 (1)
1-2:⚠️ Potential issue | 🟠 MajorInclude guard rename violates include-header convention.
PROXYSQL_PGSQL_PROTOCOL_Hdoes not match the required__CLASS_*_Hpattern for files underinclude/.As per coding guidelines,
include/**/*.{h,hpp}: "Include guards must follow the pattern#ifndef_CLASS*_H".🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/PgSQL_Protocol.h` around lines 1 - 2, The include guard PROXYSQL_PGSQL_PROTOCOL_H in PgSQL_Protocol.h violates the project's include-header convention; rename the guard to follow the __CLASS_*_H pattern (for example __CLASS_PGSQL_PROTOCOL_H) by replacing the current `#ifndef` and `#define` macros and update the matching `#endif` comment if present so the header now uses __CLASS_PGSQL_PROTOCOL_H consistently throughout the file.include/mysql_connection.h-1-2 (1)
1-2:⚠️ Potential issue | 🟠 MajorInclude guard naming violates repository include-header rule.
This guard was renamed to
PROXYSQL_MYSQL_CONNECTION_H, but the include-path guideline requires the__CLASS_*_Hpattern.As per coding guidelines,
include/**/*.{h,hpp}: "Include guards must follow the pattern#ifndef_CLASS*_H".Also applies to: 279-279
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/mysql_connection.h` around lines 1 - 2, The include guard in include/mysql_connection.h currently uses PROXYSQL_MYSQL_CONNECTION_H which violates the repo rule; rename the guard everywhere to the required pattern (e.g. __CLASS_MYSQL_CONNECTION_H) so the `#ifndef`, matching `#define`, and the trailing `#endif` comment use the new symbol; update references to PROXYSQL_MYSQL_CONNECTION_H in this file to __CLASS_MYSQL_CONNECTION_H to conform with the include-header guideline.include/mysql_backend.h-1-2 (1)
1-2:⚠️ Potential issue | 🟠 MajorHeader guard naming is out of policy for
include/headers.The new
PROXYSQL_MYSQL_BACKEND_Hguard should follow the required__CLASS_*_Hconvention.As per coding guidelines,
include/**/*.{h,hpp}: "Include guards must follow the pattern#ifndef_CLASS*_H".Also applies to: 25-25
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/mysql_backend.h` around lines 1 - 2, The header guard macro PROXYSQL_MYSQL_BACKEND_H does not follow the required include pattern; rename the guard to the __CLASS_*_H style (e.g., __CLASS_MYSQL_BACKEND_H) by updating the `#ifndef` and matching `#define` at the top and any matching `#endif` comment at the bottom so the guard consistently uses __CLASS_MYSQL_BACKEND_H instead of PROXYSQL_MYSQL_BACKEND_H (ensure you update all occurrences in the file, including any trailing comment after `#endif`).include/proxysql_config.h-1-2 (1)
1-2:⚠️ Potential issue | 🟠 MajorInclude guard no longer complies with required pattern.
PROXYSQL_CONFIG_Hdoes not satisfy the__CLASS_*_Hinclude-guard rule defined for headers ininclude/.As per coding guidelines,
include/**/*.{h,hpp}: "Include guards must follow the pattern#ifndef_CLASS*_H".🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/proxysql_config.h` around lines 1 - 2, The include guard PROXYSQL_CONFIG_H doesn't follow the required pattern; replace both occurrences of PROXYSQL_CONFIG_H in the header's guards with the mandated pattern __CLASS_PROXYSQL_CONFIG_H (update the `#ifndef` and matching `#define` and ensure the trailing `#endif` comment, if any, references the new symbol) so the header conforms to the include/ include-guard rule.include/LLM_Bridge.h-28-29 (1)
28-29:⚠️ Potential issue | 🟠 MajorHeader guard format violates include-path policy.
Line 28 and Line 29 use
PROXYSQL_LLM_BRIDGE_H; this header path requires the__CLASS_*_Hinclude-guard format.As per coding guidelines,
include/**/*.{h,hpp}: Include guards must follow the pattern#ifndef __CLASS_*_H.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/LLM_Bridge.h` around lines 28 - 29, The header guard macro PROXYSQL_LLM_BRIDGE_H in LLM_Bridge.h violates the project's include-path policy; replace it with the required pattern using the __CLASS_*_H format (e.g., change both the `#ifndef` and `#define` from PROXYSQL_LLM_BRIDGE_H to a guard following __CLASS_<UNIQUE>_H and update the corresponding `#endif` comment if present) so the guard matches the coding guideline for include/**/*.{h,hpp}.include/proxysql_restapi.h-1-2 (1)
1-2:⚠️ Potential issue | 🟠 MajorUse
__CLASS_*_Hinclude guards for files underinclude/.Line 1 and Line 2 currently define
PROXYSQL_RESTAPI_H; this does not match the repository guard policy for this path.As per coding guidelines,
include/**/*.{h,hpp}: Include guards must follow the pattern#ifndef __CLASS_*_H.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/proxysql_restapi.h` around lines 1 - 2, The current include guard symbol PROXYSQL_RESTAPI_H does not follow the repository policy; change the guards to the required pattern by replacing PROXYSQL_RESTAPI_H with __CLASS_PROXYSQL_RESTAPI_H in the header (update both the `#ifndef` and `#define` at the top and the matching `#endif` comment if present) so the file uses `#ifndef` __CLASS_PROXYSQL_RESTAPI_H / `#define` __CLASS_PROXYSQL_RESTAPI_H and the corresponding `#endif`.include/PgSQL_Backend.h-1-2 (1)
1-2:⚠️ Potential issue | 🟠 MajorUse the repository-required include guard pattern.
Lines 1-2 use
PROXYSQL_PGSQL_BACKEND_H, but this path is underinclude/and the required guard format is__CLASS_*_H. Please switch this guard back to the mandated pattern for consistency and policy compliance.As per coding guidelines,
include/**/*.{h,hpp}: Include guards must follow the pattern#ifndef __CLASS_*_H.Also applies to: 25-25
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/PgSQL_Backend.h` around lines 1 - 2, Change the non-conforming include guard PROXYSQL_PGSQL_BACKEND_H to the mandated pattern by replacing the `#ifndef` and `#define` with a guard using the __CLASS_*_H format (e.g., __CLASS_PGSQL_BACKEND_H) and update the matching `#endif` comment/identifier accordingly so both the opening and closing guard symbols match; modify the macros surrounding the PgSQL_Backend declarations (the existing PROXYSQL_PGSQL_BACKEND_H symbols and the corresponding `#endif`) to the new __CLASS_PGSQL_BACKEND_H macro.include/PgSQL_Connection.h-1-2 (1)
1-2:⚠️ Potential issue | 🟠 MajorPlease revert to the mandated include-guard format for
include/headers.Line 1 and Line 2 use
PROXYSQL_PGSQL_CONNECTION_H, but the repository rule for this path requires guards in__CLASS_*_Hformat.As per coding guidelines,
include/**/*.{h,hpp}: Include guards must follow the pattern#ifndef __CLASS_*_H.Also applies to: 743-743
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/PgSQL_Connection.h` around lines 1 - 2, Replace the current include guard macro PROXYSQL_PGSQL_CONNECTION_H with the mandated pattern __CLASS_PgSQL_Connection_H in this header; update both the `#ifndef` and `#define` lines to use __CLASS_PgSQL_Connection_H and adjust the matching `#endif` comment (if present) to reference __CLASS_PgSQL_Connection_H so the guard names are consistent.include/PgSQL_Data_Stream.h-1-2 (1)
1-2:⚠️ Potential issue | 🟠 MajorInclude guard naming is non-compliant for headers under
include/.Line 1 and Line 2 use
PROXYSQL_PGSQL_DATA_STREAM_H; the project rule for headers in this path requires__CLASS_*_Hguards.As per coding guidelines,
include/**/*.{h,hpp}: Include guards must follow the pattern#ifndef __CLASS_*_H.Also applies to: 272-272
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/PgSQL_Data_Stream.h` around lines 1 - 2, The include guard macro PROXYSQL_PGSQL_DATA_STREAM_H does not follow the project's include path convention; replace it with the required pattern by renaming the guard to __CLASS_PGSQL_DATA_STREAM_H (update both the `#ifndef/`#define at the top and the corresponding `#endif` comment at the bottom) so the header uses __CLASS_PGSQL_DATA_STREAM_H consistently; locate the guards in the file (symbols: PROXYSQL_PGSQL_DATA_STREAM_H, the opening `#ifndef/`#define and the closing `#endif`) and update them to the new macro.include/proxysql_sslkeylog.h-17-18 (1)
17-18:⚠️ Potential issue | 🟠 MajorInclude guard style does not follow required project convention.
Line 17 and Line 18 use
PROXYSQL_SSLKEYLOG_H; headers ininclude/are required to use the__CLASS_*_Hpattern.As per coding guidelines,
include/**/*.{h,hpp}: Include guards must follow the pattern#ifndef __CLASS_*_H.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/proxysql_sslkeylog.h` around lines 17 - 18, The include guard in include/proxysql_sslkeylog.h uses PROXYSQL_SSLKEYLOG_H but must follow the project convention using the __CLASS_*_H pattern; update the guard macro (both the `#ifndef` and `#define` and the corresponding `#endif` comment) to use the required pattern (e.g., replace PROXYSQL_SSLKEYLOG_H with __CLASS_PROXYSQL_SSLKEYLOG_H) so the header's include guard matches the __CLASS_*_H naming convention throughout (ensure you update the matching identifier wherever referenced in this file).include/sqlite3db.h-1-2 (1)
1-2:⚠️ Potential issue | 🟠 MajorInclude guard naming does not comply with current repository rule.
Line 1 and Line 2 define
PROXYSQL_SQLITE3DB_H; this path requires include guards in__CLASS_*_Hformat.As per coding guidelines,
include/**/*.{h,hpp}: Include guards must follow the pattern#ifndef __CLASS_*_H.Also applies to: 250-250
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/sqlite3db.h` around lines 1 - 2, The current include guard macro PROXYSQL_SQLITE3DB_H does not follow the repository pattern; change the guard to the required format by replacing PROXYSQL_SQLITE3DB_H with __CLASS_SQLITE3DB_H in both the `#ifndef` and `#define` at the top of include/sqlite3db.h, and update the matching `#endif` comment at the bottom (if present) to reference __CLASS_SQLITE3DB_H so the pair remains consistent; ensure you also update any repeated occurrences of the old macro in the same file.include/proxysql_coredump.h-1-2 (1)
1-2:⚠️ Potential issue | 🟠 MajorGuard macro naming violates include-header policy.
Line 1 and Line 2 define
PROXYSQL_COREDUMP_H; for headers ininclude/, guard macros are required to follow__CLASS_*_H.As per coding guidelines,
include/**/*.{h,hpp}: Include guards must follow the pattern#ifndef __CLASS_*_H.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/proxysql_coredump.h` around lines 1 - 2, The include guard macro PROXYSQL_COREDUMP_H does not follow the required pattern; rename the guard to the policy-compliant form (e.g., __CLASS_PROXYSQL_COREDUMP_H) and update the matching `#ifndef`, `#define` and closing `#endif` comment to use the new macro; ensure all three occurrences in include/proxysql_coredump.h reference __CLASS_PROXYSQL_COREDUMP_H so the header aligns with the include/**/*.{h,hpp} guard naming convention.include/Command_Counter.h-1-2 (1)
1-2: 🛠️ Refactor suggestion | 🟠 MajorHeader guard format should match the repository include policy.
Line 1, Line 2, and Line 63 should use the
__CLASS_*_Hpattern for headers underinclude/.As per coding guidelines
include/**/*.{h,hpp}: Include guards must follow the pattern#ifndef__CLASS_*_H.Also applies to: 63-63
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/Command_Counter.h` around lines 1 - 2, The include guard in Command_Counter.h uses PROXYSQL_COMMAND_COUNTER_H but must follow the repo pattern; replace the current `#ifndef/`#define/#endif with the __CLASS_*_H form (use __CLASS_COMMAND_COUNTER_H) so the guard and the trailing `#endif` comment match; update the symbol references in this header (e.g., the opening `#ifndef/`#define and the closing `#endif` comment) to __CLASS_COMMAND_COUNTER_H to comply with include/**/*.{h,hpp} guidelines.include/Anomaly_Detector.h-35-36 (1)
35-36: 🛠️ Refactor suggestion | 🟠 MajorPlease keep include guard naming consistent with the
include/rule set.Line 35 and Line 36 should use the
__CLASS_*_Hguard format mandated for headers in this directory.As per coding guidelines
include/**/*.{h,hpp}: Include guards must follow the pattern#ifndef__CLASS_*_H.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/Anomaly_Detector.h` around lines 35 - 36, The include guard in Anomaly_Detector.h uses PROXYSQL_ANOMALY_DETECTOR_H but must follow the project pattern __CLASS_*_H; update the guard around the file (the `#ifndef/`#define at the top and the matching `#endif` at the bottom) to use __CLASS_ANOMALY_DETECTOR_H so it conforms with the include/ rule set and matches the class/name in Anomaly_Detector.h.include/MySQL_Data_Stream.h-1-2 (1)
1-2: 🛠️ Refactor suggestion | 🟠 MajorAlign include guard with the repository’s include-header convention.
Line 1 and Line 2 changed the guard to
PROXYSQL_MYSQL_DATA_STREAM_H; Line 290 mirrors it. The current repo rule for headers underinclude/expects__CLASS_*_H.Proposed fix
-#ifndef PROXYSQL_MYSQL_DATA_STREAM_H -#define PROXYSQL_MYSQL_DATA_STREAM_H +#ifndef __CLASS_MYSQL_DATA_STREAM_H +#define __CLASS_MYSQL_DATA_STREAM_H ... -#endif /* PROXYSQL_MYSQL_DATA_STREAM_H */ +#endif /* __CLASS_MYSQL_DATA_STREAM_H */As per coding guidelines
include/**/*.{h,hpp}: Include guards must follow the pattern#ifndef__CLASS_*_H.Also applies to: 290-290
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/MySQL_Data_Stream.h` around lines 1 - 2, The include guard macro PROXYSQL_MYSQL_DATA_STREAM_H does not follow the repository convention; replace both occurrences of that macro (the opening `#ifndef/`#define and the matching `#endif` identifier at the bottom) with a guard following the __CLASS_*_H pattern used for headers under include/ (e.g., change PROXYSQL_MYSQL_DATA_STREAM_H to the repository's __CLASS_<SOMETHING>_H style name that matches this header's class/filename), ensuring the same identifier is used in the `#ifndef`, `#define`, and trailing `#endif` comment.include/PgSQL_Query_Cache.h-1-2 (1)
1-2: 🛠️ Refactor suggestion | 🟠 MajorUse the required
__CLASS_*_Hinclude-guard pattern here too.Line 1, Line 2, and Line 22 currently use
PROXYSQL_PGSQL_QUERY_CACHE_H, which diverges from the include-header guard convention configured for this repository.As per coding guidelines
include/**/*.{h,hpp}: Include guards must follow the pattern#ifndef__CLASS_*_H.Also applies to: 22-22
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/PgSQL_Query_Cache.h` around lines 1 - 2, The include guard symbol PROXYSQL_PGSQL_QUERY_CACHE_H does not follow the repository pattern; replace it with __CLASS_PgSQL_Query_Cache_H in the `#ifndef` and `#define` at the top (and the matching `#endif` at the bottom), ensuring the symbol exactly matches the file/class name casing used in PgSQL_Query_Cache.h and update any trailing comment on the `#endif` if present to use the same symbol.include/PgSQL_Query_Processor.h-1-2 (1)
1-2: 🛠️ Refactor suggestion | 🟠 MajorGuard naming needs to follow the
__CLASS_*_Hrule forinclude/headers.Line 1, Line 2, and Line 45 currently use a
PROXYSQL_*guard, which is out of policy for this path pattern.As per coding guidelines
include/**/*.{h,hpp}: Include guards must follow the pattern#ifndef__CLASS_*_H.Also applies to: 45-45
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/PgSQL_Query_Processor.h` around lines 1 - 2, The header guard macro PROXYSQL_PGSQL_QUERY_PROCESSOR_H used in PgSQL_Query_Processor.h does not follow the include/ header policy; replace it with the required pattern (e.g. __CLASS_PGSQL_QUERY_PROCESSOR_H) by updating the `#ifndef` and matching `#define` at the top and the corresponding `#endif` comment at the bottom (and any duplicate/secondary occurrence around the previous line 45) so the guard name consistently uses the __CLASS_*_H format.include/log_utils.h-1-2 (1)
1-2: 🛠️ Refactor suggestion | 🟠 MajorPlease revert this include guard to the mandated
__CLASS_*_Hformat.Line 1, Line 2, and Line 221 currently use
PROXYSQL_LOG_UTILS_H, which does not match the configured include guard rule for headers ininclude/.As per coding guidelines
include/**/*.{h,hpp}: Include guards must follow the pattern#ifndef__CLASS_*_H.Also applies to: 221-221
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/log_utils.h` around lines 1 - 2, The include guard macro PROXYSQL_LOG_UTILS_H does not follow the mandated pattern; update the guard in include/log_utils.h by replacing PROXYSQL_LOG_UTILS_H with the configured format (e.g., __CLASS_LOG_UTILS_H) in the `#ifndef` and `#define` at the top and the matching `#endif` comment at the bottom (the occurrence around line 221); ensure all three occurrences use the exact same new macro name so the guard remains consistent.include/proxysql_utils.h-1-2 (1)
1-2: 🛠️ Refactor suggestion | 🟠 MajorGuard macro naming is out of policy for
include/headers.Line 1 and Line 2 should follow the
__CLASS_*_Hguard format used by the repo rules forinclude/**/*.{h,hpp}.As per coding guidelines
include/**/*.{h,hpp}: Include guards must follow the pattern#ifndef__CLASS_*_H.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/proxysql_utils.h` around lines 1 - 2, The header guard macros in include/proxysql_utils.h use PROXYSQL_UTILS_H which violates the repo policy; rename the guard to follow the __CLASS_*_H pattern (e.g., change `#ifndef` PROXYSQL_UTILS_H and `#define` PROXYSQL_UTILS_H to `#ifndef` __CLASS_PROXYSQL_UTILS_H and `#define` __CLASS_PROXYSQL_UTILS_H) and update the corresponding `#endif` comment if present so the opening and closing macros match exactly.include/PgSQL_Error_Helper.h-1-2 (1)
1-2: 🛠️ Refactor suggestion | 🟠 MajorInclude guard should follow the repository
__CLASS_*_Hpattern.Line 1, Line 2, and Line 672 currently use
PROXYSQL_PGSQL_ERROR_HELPER_H, which diverges from the include-header guard requirement configured for this repo.As per coding guidelines
include/**/*.{h,hpp}: Include guards must follow the pattern#ifndef__CLASS_*_H.Also applies to: 672-672
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/PgSQL_Error_Helper.h` around lines 1 - 2, Replace the current include guard macro PROXYSQL_PGSQL_ERROR_HELPER_H with the repo-required pattern __CLASS_PGSQL_ERROR_HELPER_H throughout the header (start guard, define, and the trailing `#endif` comment), i.e., update the `#ifndef`, `#define` and the closing `#endif` comment in include/PgSQL_Error_Helper.h to use __CLASS_PGSQL_ERROR_HELPER_H so it matches the repository include guard convention.include/query_digest_topk.h-1-2 (1)
1-2:⚠️ Potential issue | 🟠 MajorThis include guard format conflicts with the configured header rule.
Line 1 and Line 2 should follow the
__CLASS_*_Hnaming scheme required for headers ininclude/.As per coding guidelines "Include guards must follow the pattern
#ifndef_CLASS*_H".🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/query_digest_topk.h` around lines 1 - 2, The include guard in include/query_digest_topk.h uses PROXYSQL_QUERY_DIGEST_TOPK_H which violates the required pattern; rename the guard to follow the __CLASS_*_H scheme (e.g., __CLASS_QUERY_DIGEST_TOPK_H) and update the matching `#ifndef` and `#define` macros (and the trailing `#endif` comment if present) so the header uses the new __CLASS_QUERY_DIGEST_TOPK_H identifier consistently.include/PgSQL_Thread.h-1-3 (1)
1-3:⚠️ Potential issue | 🟠 MajorHeader guard naming is out of policy for
include/headers.Line 1 and Line 2 should use the
__CLASS_*_Hinclude-guard pattern expected by the repository rules.As per coding guidelines "Include guards must follow the pattern
#ifndef_CLASS*_H".🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/PgSQL_Thread.h` around lines 1 - 3, The header guards in include/PgSQL_Thread.h currently use PROXYSQL_PGSQL_THREAD_H / PROXYSQL_STANDARD_PGSQL_THREAD_H and must be changed to follow the repository pattern; replace both `#ifndef` and `#define` with a single guard named __CLASS_PGSQL_THREAD_H (and update the matching `#endif` comment if present) so the file uses the required __CLASS_*_H include-guard convention for the PgSQL_Thread header.include/MySQL_Set_Stmt_Parser.h-1-2 (1)
1-2:⚠️ Potential issue | 🟠 MajorRevert this include guard to the expected
__CLASS_*_Hform.Line 1 and Line 2 no longer match the include-guard naming convention enforced for headers under
include/.As per coding guidelines "Include guards must follow the pattern
#ifndef_CLASS*_H".🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/MySQL_Set_Stmt_Parser.h` around lines 1 - 2, The include guard uses PROXYSQL_MYSQL_SET_STMT_PARSER_H but must follow the project pattern `#ifndef` __CLASS_*_H; change both the `#ifndef` and `#define` to use __CLASS_MYSQL_SET_STMT_PARSER_H (and update the trailing `#endif` comment if present) in the MySQL_Set_Stmt_Parser.h header so the guard matches the required __CLASS_*_H naming convention.include/ProxySQL_Poll.h-1-2 (1)
1-2:⚠️ Potential issue | 🟠 MajorPlease keep this include guard in the expected
__CLASS_*_Hformat.Line 1 and Line 2 currently use a guard naming style that conflicts with the configured rule for headers under
include/.As per coding guidelines "Include guards must follow the pattern
#ifndef_CLASS*_H".🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/ProxySQL_Poll.h` around lines 1 - 2, Replace the existing include guard macros PROXYSQL_POLL_H with the required pattern by renaming both occurrences to __CLASS_PROXYSQL_POLL_H (update the `#ifndef` and matching `#define` and any trailing `#endif` comment if present) so the header uses the mandated __CLASS_*_H format; ensure the macro name matches exactly in all places within ProxySQL_Poll.h.include/AI_Features_Manager.h-30-31 (1)
30-31:⚠️ Potential issue | 🟠 MajorUse the repository-required include-guard pattern.
Line 30 and Line 31 use
PROXYSQL_AI_FEATURES_MANAGER_H, but this header path is configured to require the__CLASS_*_Hguard format.As per coding guidelines "Include guards must follow the pattern
#ifndef_CLASS*_H".🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/AI_Features_Manager.h` around lines 30 - 31, The include guard macro PROXYSQL_AI_FEATURES_MANAGER_H in AI_Features_Manager.h doesn't follow the required pattern; replace the guard with the repository standard (e.g., __CLASS_AI_FEATURES_MANAGER_H) by changing the `#ifndef` and `#define` to use __CLASS_AI_FEATURES_MANAGER_H and update the matching `#endif` comment to reference the same macro, ensuring the three occurrences (ifndef, define, endif) are consistent.include/PgSQL_Set_Stmt_Parser.h-1-2 (1)
1-2:⚠️ Potential issue | 🟠 MajorKeep include guards consistent with project convention.
Line 1 and Line 2 use
PROXYSQL_*naming, but this directory’s rule requires__CLASS_*_Hguard names.As per coding guidelines "Include guards must follow the pattern
#ifndef_CLASS*_H".🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/PgSQL_Set_Stmt_Parser.h` around lines 1 - 2, The include guard in PgSQL_Set_Stmt_Parser.h uses PROXYSQL_PGSQL_SET_STMT_PARSER_H but must follow the project pattern `#ifndef` __CLASS_*_H; update the guard to use __CLASS_PgSQL_Set_Stmt_Parser_H (replace the existing `#ifndef` and `#define`) and adjust the trailing `#endif` comment if present to match the new guard; ensure references to the header’s guard correspond to the class/header name PgSQL_Set_Stmt_Parser.include/MCP_Thread.h-1-2 (1)
1-2:⚠️ Potential issue | 🟠 MajorGuard naming here should follow the repository include-header convention.
Line 1 and Line 2 should use the
__CLASS_*_Hform for this path.As per coding guidelines "Include guards must follow the pattern
#ifndef_CLASS*_H".🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/MCP_Thread.h` around lines 1 - 2, The include guard currently uses PROXYSQL_MCP_THREAD_H but must follow the repository pattern `#ifndef` __CLASS_*_H; replace the guard macro in the top two lines with __CLASS_MCP_THREAD_H (and update the matching `#endif` comment at the end of the file to reference __CLASS_MCP_THREAD_H) so the macros used in MCP_Thread.h follow the __CLASS_*_H convention.include/GenAI_Thread.h-1-2 (1)
1-2:⚠️ Potential issue | 🟠 MajorThis include-guard rename breaks the configured naming rule for
include/.Line 1 and Line 2 should stay in the
__CLASS_*_Hguard pattern.As per coding guidelines "Include guards must follow the pattern
#ifndef_CLASS*_H".🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/GenAI_Thread.h` around lines 1 - 2, The include guard in include/GenAI_Thread.h uses PROXYSQL_GENAI_THREAD_H which violates the project rule; replace the two lines using that guard with the project pattern by renaming them to the __CLASS_*_H form (e.g., change both `#ifndef` PROXYSQL_GENAI_THREAD_H and `#define` PROXYSQL_GENAI_THREAD_H to `#ifndef` __CLASS_GENAI_THREAD_H and `#define` __CLASS_GENAI_THREAD_H) and update the corresponding closing `#endif` comment (if present) to match the new __CLASS_GENAI_THREAD_H symbol.include/query_processor.h-187-196 (1)
187-196:⚠️ Potential issue | 🟠 MajorConstructor initialization is dead code—remove or align with
init()method.The constructor initializer list (lines 189-193) sets fields to specific values (e.g.,
reconnect(true),retries(3),cache_ttl(0)), butinit()immediately overwrites all these fields with-1sentinels. Every instantiation in the codebase follows the pattern:new Query_Processor_Output(); obj->init();Even
MCP_Query_Processor_Outputexplicitly callsinit()in its constructor. The constructor initialization serves no purpose and should be removed or aligned withinit()to avoid future confusion.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/query_processor.h` around lines 187 - 196, The constructor Query_Processor_Output currently sets member defaults that are immediately overwritten by init(); either remove the redundant initializer list and leave the constructor empty or have the constructor call init() so there is a single source of truth; update Query_Processor_Output() to either (a) drop the long initializer list and implement an empty body, or (b) call this->init() from the constructor body (and ensure MCP_Query_Processor_Output still calls init() only if needed), keeping init() as the canonical initializer for all fields.include/Query_Cache.h-1-2 (1)
1-2:⚠️ Potential issue | 🟠 MajorInclude guard naming deviates from repo header-guard convention.
Line 1, Line 2, and Line 219 use
PROXYSQL_QUERY_CACHE_H, but this file path matches the include-guard rule requiring__CLASS_*_H.Suggested fix
-#ifndef PROXYSQL_QUERY_CACHE_H -#define PROXYSQL_QUERY_CACHE_H +#ifndef __CLASS_QUERY_CACHE_H +#define __CLASS_QUERY_CACHE_H ... -#endif /* PROXYSQL_QUERY_CACHE_H */ +#endif /* __CLASS_QUERY_CACHE_H */As per coding guidelines,
include/**/*.{h,hpp}: Include guards must follow the pattern#ifndef__CLASS_*_H.Also applies to: 219-219
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/Query_Cache.h` around lines 1 - 2, The include guard macro PROXYSQL_QUERY_CACHE_H in Query_Cache.h deviates from the repo convention; replace it with the required pattern (e.g., __CLASS_QUERY_CACHE_H) everywhere it appears (the opening `#ifndef/`#define and the closing `#endif` comment) so the guard uses the __CLASS_*_H form; update the macro name consistently in the header (look for the symbols PROXYSQL_QUERY_CACHE_H and the matching `#endif`) to prevent mismatched guards.include/MySQL_Thread.h-1-3 (1)
1-3:⚠️ Potential issue | 🟠 MajorHeader guards do not match the configured include-guard pattern.
Line 1-3 and Line 765 switched to
PROXYSQL_*guards, but this header is underinclude/and should follow the__CLASS_*_Hpattern.Suggested fix
-#ifndef PROXYSQL_MYSQL_THREAD_H -#define PROXYSQL_MYSQL_THREAD_H -#define PROXYSQL_STANDARD_MYSQL_THREAD_H +#ifndef __CLASS_MYSQL_THREAD_H +#define __CLASS_MYSQL_THREAD_H +#define ____CLASS_STANDARD_MYSQL_THREAD_H ... -#endif /* PROXYSQL_MYSQL_THREAD_H */ +#endif /* __CLASS_MYSQL_THREAD_H */As per coding guidelines,
include/**/*.{h,hpp}: Include guards must follow the pattern#ifndef__CLASS_*_H.Also applies to: 765-765
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/MySQL_Thread.h` around lines 1 - 3, The header guard macros at the top and bottom use PROXYSQL_MYSQL_THREAD_H and PROXYSQL_STANDARD_MYSQL_THREAD_H but must follow the include/ file pattern __CLASS_*_H; replace those PROXYSQL_* macros with a single consistent guard matching the class/header name (e.g., __CLASS_MYSQL_THREAD_H) and update both the `#ifndef/`#define and the matching `#endif` macro reference (the existing PROXYSQL_MYSQL_THREAD_H / PROXYSQL_STANDARD_MYSQL_THREAD_H identifiers) so the macro names are identical and conform to the __CLASS_*_H pattern used for files under include/.include/proxysql_macros.h-7-7 (1)
7-7: 🛠️ Refactor suggestion | 🟠 MajorRemove local
itostrredefinitions or update to match the safer header version.Three files locally redefine
itostrwithout the protective parentheses present ininclude/proxysql_macros.h:7:
lib/MySQL_HostGroups_Manager.cpp:25lib/PgSQL_HostGroups_Manager.cpp:26lib/Base_HostGroups_Manager.cpp:25All three use
{ __s=char_malloc(32); sprintf(__s, "%lld", __i); }instead of the safer{ (__s)=char_malloc(32); sprintf(__s, "%lld", (__i)); }. Remove these local redefinitions to use the header version or update all to match the safer implementation.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/proxysql_macros.h` at line 7, The local macro redefinitions of itostr in MySQL_HostGroups_Manager.cpp, PgSQL_HostGroups_Manager.cpp and Base_HostGroups_Manager.cpp are unsafe and differ from the header's safer form; remove the local definitions so the header's itostr is used, or update each local macro to exactly match the header version (ensure parentheses around both __s and __i as in ({ (__s)=char_malloc(32); sprintf(__s, "%lld", (__i)); })). Locate the macro name itostr in those files, delete the local `#define`, or replace it with the header-equivalent definition, and then rebuild to confirm no warnings or redefinition conflicts.include/proxysql_admin.h-1-2 (1)
1-2:⚠️ Potential issue | 🟠 MajorResolve include-guard policy mismatch.
This guard rename is consistent with the lint cleanup intent, but it conflicts with the current repository guard rule for headers under
include/. Please align either the code or the guideline to avoid recurring churn.As per coding guidelines
include/**/*.{h,hpp}: Include guards must follow the pattern#ifndef__CLASS_*_H.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/proxysql_admin.h` around lines 1 - 2, The header's include guard macro PROXYSQL_ADMIN_H doesn't follow the project's include/ header naming policy; change the guard to match the required pattern (e.g. __CLASS_PROXYSQL_ADMIN_H) by updating the `#ifndef`, `#define` and matching `#endif` macro to the new name used in include_guard checking; ensure the new macro is unique and all three places (ifndef, define, endif comment if present) reference the same __CLASS_*_H style macro (for the header represented by PROXYSQL_ADMIN).
🧹 Nitpick comments (9)
include/gen_utils.h (1)
159-166: Update parameter name inPtrSizeArray::PtrSizeArrayimplementation for consistency.The constructor declaration at
include/gen_utils.h:246usesszas the parameter name, but the definition atlib/gen_utils.cpp:167still uses__size. Update the implementation to match the declaration for consistency.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/gen_utils.h` around lines 159 - 166, The PtrSizeArray constructor implementation uses a different parameter name (__size) than its declaration (sz); update the implementation of PtrSizeArray::PtrSizeArray in lib/gen_utils.cpp to use the same parameter name (sz) as the declaration so signatures match and avoid confusion, and ensure any uses of the old name inside the function are renamed to sz (references to PtrSizeArray, PtrSizeArray::PtrSizeArray).lib/MySQL_Session.cpp (1)
3344-3352: Either dropendor render it.These paths still trim to
253chars and computeend = "...", but the format string no longer has a placeholder for it. Low priority, but the truncation marker is now dead code and the message no longer shows that the query was shortened.Also applies to: 3517-3525, 5389-5397
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@lib/MySQL_Session.cpp` around lines 3344 - 3352, The message-building code computes an ellipsis in variable end when a query is truncated but never uses it, resulting in a dead truncation marker and an incorrect buffer size; update the Session logging in MySQL_Session.cpp (around the sprintf that builds buf using err_msg, nqn and end) to either remove the unused end variable and its length from the malloc calculation or include a "%s" placeholder for end in err_msg and pass end to sprintf, and while here replace malloc+sprintf with a bounds-safe allocation/formatting (e.g., compute required length including the ellipsis and use snprintf) so the truncated query and the "..." marker are correctly reflected and no buffer overflow can occur (apply same fix at the other similar sites referenced).lib/MySQL_Thread.cpp (1)
6120-6131: Align PostgreSQL thread metrics to useSIZE_sentinel naming.The MySQL_Thread code has been updated to use
p_th_counter::SIZE_andp_th_gauge::SIZE_(compliant with UPPER_SNAKE_CASE), butlib/PgSQL_Thread.cppstill references__sizeat lines 5487 and 5520. Update these comparisons to match the MySQL naming convention and coding guidelines.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@lib/MySQL_Thread.cpp` around lines 6120 - 6131, The PostgreSQL thread code still compares index/sentinel against the old symbol `__size`; update those comparisons to use the new sentinel names `p_th_counter::SIZE_` and `p_th_gauge::SIZE_` (matching the MySQL change) wherever `__size` is referenced (e.g., the checks at the locations that currently read like `if (m_idx != __size)` and similar for gauge indices); ensure you replace both counter and gauge comparisons with the corresponding `p_th_counter::SIZE_` and `p_th_gauge::SIZE_` symbols so the naming and behavior match the MySQL_Thread pattern.scripts/lint/normalize-cppcheck.py (1)
13-16: Use a context manager for file reads.
open(path).read()leaves lifecycle management implicit. Preferwith open(...)for deterministic close behavior.Suggested patch
-if path and os.path.exists(path): - data = open(path).read() +if path and os.path.exists(path): + with open(path, encoding="utf-8") as f: + data = f.read() else: data = sys.stdin.read()🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@scripts/lint/normalize-cppcheck.py` around lines 13 - 16, Replace the implicit open() call with a context manager so the file is deterministically closed: in the branch that reads from path (the code using variables path and data) use a with-open block (opening in read mode) to read into data, leaving the else branch that reads from sys.stdin unchanged; ensure the variable names (path, data) and behavior remain the same.include/MySQL_Authentication.hpp (1)
1-2: Align include guard with repository guard convention.This header guard no longer matches the repository’s include-guard pattern for files under
include/.Suggested patch
-#ifndef PROXYSQL_MYSQL_AUTHENTICATION_H -#define PROXYSQL_MYSQL_AUTHENTICATION_H +#ifndef __CLASS_MYSQL_AUTHENTICATION_H +#define __CLASS_MYSQL_AUTHENTICATION_HAs per coding guidelines "
include/**/*.{h,hpp}: Include guards must follow the pattern#ifndef_CLASS*_H".🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@include/MySQL_Authentication.hpp` around lines 1 - 2, The include guard macro in MySQL_Authentication.hpp (currently PROXYSQL_MYSQL_AUTHENTICATION_H) does not follow the repository convention; change it to the required pattern using the __CLASS_*_H format (e.g., align with the file/class name so the macro becomes something like __CLASS_MYSQL_AUTHENTICATION_H), updating both the `#ifndef` and `#define` (and corresponding `#endif` comment if present) to match the new macro; locate the guard at the top of include/MySQL_Authentication.hpp and update any occurrences to the new __CLASS_*_H token so it follows the repository include-guard convention.scripts/lint/normalize-clang-tidy.py (1)
130-131: Rename ambiguous variablelfor clarity.The variable name
lis easily confused with the number1. Consider usinglineordiaginstead.📝 Suggested fix
-for l in sorted(diagnostics): - print(l) +for line in sorted(diagnostics): + print(line)🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@scripts/lint/normalize-clang-tidy.py` around lines 130 - 131, The loop variable `l` in the for-loop iterating over `sorted(diagnostics)` is ambiguous; rename it to a clear identifier like `diag` or `line` throughout the loop (e.g., change `for l in sorted(diagnostics): print(l)` to `for diag in sorted(diagnostics): print(diag)`) and update any other references in the same scope to use the new name so readability is improved and `l` isn’t confused with the digit one.scripts/lint/run-local.sh (2)
39-39: Minor: Simplify redundantechoaround command substitution.The
echo "$(pwd)"is equivalent to justpwd. The git command output doesn't need echo wrapping either.📝 Suggested fix
- REPO_ROOT=$(git rev-parse --show-toplevel 2>/dev/null || echo "$(pwd)") + REPO_ROOT=$(git rev-parse --show-toplevel 2>/dev/null || pwd)🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@scripts/lint/run-local.sh` at line 39, The REPO_ROOT assignment uses a redundant echo wrapper; update the fallback to use pwd directly so REPO_ROOT=$(git rev-parse --show-toplevel 2>/dev/null || pwd) (replace the current use of echo "$(pwd)" and keep the existing command substitution and error redirection intact).
82-84: Quote glob patterns to prevent word splitting issues.The
${TMPD}variable should be quoted to handle paths with spaces correctly.📝 Suggested fix
- if ls ${TMPD:-/dev/null}/ct-*.stderr >/dev/null 2>&1; then - cat ${TMPD}/ct-*.stderr > lint/clang-tidy-raw.txt || true + if ls "${TMPD:-/dev/null}"/ct-*.stderr >/dev/null 2>&1; then + cat "${TMPD}"/ct-*.stderr > lint/clang-tidy-raw.txt || true🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@scripts/lint/run-local.sh` around lines 82 - 84, The shell glob uses ${TMPD} unquoted which breaks on paths with spaces; update the checks and commands that reference ${TMPD} and the ct-*.stderr pattern (the ls condition, the cat into lint/clang-tidy-raw.txt, and the call to python3 scripts/lint/normalize-clang-tidy.py) to quote the variable and glob (e.g. use "${TMPD:-/dev/null}/ct-*.stderr" and "${TMPD}/ct-*.stderr") so word-splitting is prevented while preserving the same redirection and fallback behavior.docs/superpowers/plans/2026-04-10-cppcheck-fixes.md (1)
174-181: Minor: Add language specifier to fenced code block.The suppression policy code block should specify a language (e.g.,
markdownortext) for consistency with other documentation.📝 Suggested fix
-``` +```text Lint suppressions policy - Prefer fixing issues over suppressing.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@docs/superpowers/plans/2026-04-10-cppcheck-fixes.md` around lines 174 - 181, The fenced code block containing the "Lint suppressions policy" text lacks a language specifier; update that block to use a language tag (e.g., `text` or `markdown`) so it matches other docs and renders consistently, ensuring the opening fence becomes ```text (or ```markdown) and the closing fence remains ```; locate the block by searching for the literal lines "Lint suppressions policy" or the nearby snippet shown in the diff and update the fence accordingly.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
Run ID: d394ad47-f14a-4e0f-8f02-03a69a2aa438
📒 Files selected for processing (77)
.clang-tidy.gitignoreMakefilecompile_commands.events.jsondocs/superpowers/plans/2026-04-09-static-analysis-plan.mddocs/superpowers/plans/2026-04-10-cppcheck-fixes.mddocs/superpowers/specs/2026-04-09-linting-design.mdinclude/AI_Features_Manager.hinclude/AI_Vector_Storage.hinclude/Anomaly_Detector.hinclude/Base_HostGroups_Manager.hinclude/ClickHouse_Authentication.hppinclude/Command_Counter.hinclude/GenAI_Thread.hinclude/LLM_Bridge.hinclude/MCP_Thread.hinclude/MySQL_Authentication.hppinclude/MySQL_Data_Stream.hinclude/MySQL_HostGroups_Manager.hinclude/MySQL_LDAP_Authentication.hppinclude/MySQL_Logger.hppinclude/MySQL_Monitor.hppinclude/MySQL_PreparedStatement.hinclude/MySQL_Protocol.hinclude/MySQL_Query_Cache.hinclude/MySQL_Query_Processor.hinclude/MySQL_Session.hinclude/MySQL_Set_Stmt_Parser.hinclude/MySQL_Thread.hinclude/PgSQL_Authentication.hinclude/PgSQL_Backend.hinclude/PgSQL_Connection.hinclude/PgSQL_Data_Stream.hinclude/PgSQL_Error_Helper.hinclude/PgSQL_HostGroups_Manager.hinclude/PgSQL_Logger.hppinclude/PgSQL_Protocol.hinclude/PgSQL_Query_Cache.hinclude/PgSQL_Query_Processor.hinclude/PgSQL_Session.hinclude/PgSQL_Set_Stmt_Parser.hinclude/PgSQL_Thread.hinclude/ProxySQL_Cluster.hppinclude/ProxySQL_Poll.hinclude/Query_Cache.hinclude/Servers_SslParams.hinclude/StatCounters.hinclude/gen_utils.hinclude/log_utils.hinclude/mysql_backend.hinclude/mysql_connection.hinclude/prometheus_helpers.hinclude/proxysql_admin.hinclude/proxysql_config.hinclude/proxysql_coredump.hinclude/proxysql_debug.hinclude/proxysql_macros.hinclude/proxysql_restapi.hinclude/proxysql_sslkeylog.hinclude/proxysql_structs.hinclude/proxysql_utils.hinclude/query_digest_topk.hinclude/query_processor.hinclude/sqlite3db.hlib/ClickHouse_Authentication.cpplib/MySQL_Authentication.cpplib/MySQL_HostGroups_Manager.cpplib/MySQL_Protocol.cpplib/MySQL_Session.cpplib/MySQL_Thread.cpplib/PgSQL_Authentication.cpplib/ProxySQL_Admin.cpplib/debug.cppscripts/lint/generate-compile-commands.shscripts/lint/normalize-clang-tidy.pyscripts/lint/normalize-cppcheck.pyscripts/lint/run-local.sh
| username=u ? strdup(u) : nullptr; | ||
| password=p ? strdup(p) : nullptr; | ||
| hostname=h ? strdup(h) : nullptr; |
There was a problem hiding this comment.
Don’t make hostname nullable without hardening the worker path.
This constructor now allows hostname == nullptr, but kill_query_thread() still treats it as a mandatory C string at Lines 266, 271, 313, 319, and 321. That turns the old constructor-time failure into a later null-deref/UB path in the kill worker. Either normalize hostname here to a safe value for all call paths, or make the worker consistently guard and use get_host_address()/safe logging before shipping this.
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@lib/MySQL_Session.cpp` around lines 205 - 207, The constructor currently
allows hostname to be nullptr (hostname = h ? strdup(h) : nullptr;) which later
causes null-derefs in kill_query_thread(); change the constructor to normalize
hostname to a non-null safe string (e.g., strdup("") or strdup("unknown")
instead of nullptr) so all code paths can assume a valid C string, and update
any associated cleanup accordingly; alternatively, if you prefer defensive
checks, modify kill_query_thread() to guard uses of hostname (or call
get_host_address() / safe logging helpers) before dereferencing so no UB occurs.
Ensure the fix references the hostname field initialization and the
kill_query_thread() usages.
- Fix double-free bug in free_account_details() (lib/MySQL_Authentication.cpp) - Fix trailing #endif comments in 5 files to match guard names - Rename __SIZE -> SIZE_ in LDAP_USER_FIELD_IDX enum - Remove dead code: Query_Processor_Output ctor now calls init()
There was a problem hiding this comment.
Actionable comments posted: 1
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (3)
lib/MySQL_Authentication.cpp (3)
814-827:⚠️ Potential issue | 🟡 MinorThe documented column order and these indices disagree.
Lines 798-800 say the resultset includes
activebeforeuse_ssl, but this lambda hardcodesactive_ = trueand still readsuse_sslfromrow[2]. Please make the mapping and the comment match; otherwise every field fromuse_sslonward is one edit away from drifting silently.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@lib/MySQL_Authentication.cpp` around lines 814 - 827, The lambda create_account_details currently hardcodes acc_details->active_ = true and reads use_ssl from row[2], which misaligns with the documented column order (active before use_ssl); fix by reading active_ from the correct column (e.g., acc_details->active_ = strcmp(row[2], "1") == 0) and shift all subsequent column reads one index higher (use_ssl from row[3], default_hostgroup via atoi(row[4]), default_schema from row[5], schema_locked from row[6], transaction_persistent from row[7], fast_forward from row[8], backend_ from row[9], frontend_ from row[10]), keep the null checks (row[n] ? row[n] : "") and boolean strcmp patterns, and/or update the surrounding comment/doc to reflect the final column ordering consistently.
221-254:⚠️ Potential issue | 🟠 MajorInitialize struct fields for new accounts.
The
new_adpath (line 222) allocates but does not initializenum_connections_used_addl_pass,frontend_, andbackend_. These are mutated later (lines 430 and 466 fornum_connections_used_addl_pass), leaving them in indeterminate state on first use.Minimal fix
new_ad=true; ad->sha1_pass=NULL; ad->num_connections_used=0; + ad->num_connections_used_addl_pass=0; ad->clear_text_password[0] = NULL; ad->clear_text_password[1] = NULL; + ad->frontend_=false; + ad->backend_=false; // FIXME: if the password is a clear text password, automatically generate sha1_pass and clear_text_password }🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@lib/MySQL_Authentication.cpp` around lines 221 - 254, The new account allocation path (when new_ad is set) fails to initialize account_details_t fields that are later mutated: set ad->num_connections_used_addl_pass = 0 and initialize the connection pointers ad->frontend_ = NULL and ad->backend_ = NULL in the same block where new_ad is created (alongside ad->num_connections_used and ad->sha1_pass). This ensures those fields have defined values before later use (see account_details_t and the new_ad allocation block).
337-362:⚠️ Potential issue | 🟠 MajorUse
calloc()instead ofmalloc()to zero-initializeaccount_details_tallocations.The struct uses C++ default member initializers (
= nullptr), which only apply to stack ornew-allocated objects. Withmalloc(), the_complete == falsebranch (lines 337-349) leavessha1_passandclear_text_password[]uninitialized. Whilefree_account_details()checks before freeing, garbage pointers will likely trigger invalid free operations. The second loop at lines 369–387 already explicitly initializes these fields, so apply the fix to both locations for consistency.Fix
- account_details_t *ad=(account_details_t *)malloc(sizeof(account_details_t)); + account_details_t *ad=(account_details_t *)calloc(1, sizeof(account_details_t));🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@lib/MySQL_Authentication.cpp` around lines 337 - 362, The allocation of account_details_t using malloc leaves members like sha1_pass and clear_text_password[] uninitialized (causing unsafe frees); replace malloc(sizeof(account_details_t)) with calloc(1, sizeof(account_details_t)) for the ad allocations (both places where account_details_t *ad is created) so the struct is zero-initialized, and keep the existing explicit assignments for other fields; ensure this fixes fields used by free_account_details() such as sha1_pass and clear_text_password.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@include/MySQL_LDAP_Authentication.hpp`:
- Line 5: The header is not self-sufficient: types used by enum index (uint8_t,
uint64_t), the std::unique_ptr members, and the SQLite3_result references are
undeclared; add `#include` <cstdint> and `#include` <memory> at the top and either
include the appropriate SQLite header or add a forward declaration like "struct
SQLite3_result;" so symbols referenced in MySQL_LDAP_Authentication.hpp (enum
index, uint8_t, uint64_t, std::unique_ptr<...>, and SQLite3_result) are defined
when this header is parsed standalone.
---
Outside diff comments:
In `@lib/MySQL_Authentication.cpp`:
- Around line 814-827: The lambda create_account_details currently hardcodes
acc_details->active_ = true and reads use_ssl from row[2], which misaligns with
the documented column order (active before use_ssl); fix by reading active_ from
the correct column (e.g., acc_details->active_ = strcmp(row[2], "1") == 0) and
shift all subsequent column reads one index higher (use_ssl from row[3],
default_hostgroup via atoi(row[4]), default_schema from row[5], schema_locked
from row[6], transaction_persistent from row[7], fast_forward from row[8],
backend_ from row[9], frontend_ from row[10]), keep the null checks (row[n] ?
row[n] : "") and boolean strcmp patterns, and/or update the surrounding
comment/doc to reflect the final column ordering consistently.
- Around line 221-254: The new account allocation path (when new_ad is set)
fails to initialize account_details_t fields that are later mutated: set
ad->num_connections_used_addl_pass = 0 and initialize the connection pointers
ad->frontend_ = NULL and ad->backend_ = NULL in the same block where new_ad is
created (alongside ad->num_connections_used and ad->sha1_pass). This ensures
those fields have defined values before later use (see account_details_t and the
new_ad allocation block).
- Around line 337-362: The allocation of account_details_t using malloc leaves
members like sha1_pass and clear_text_password[] uninitialized (causing unsafe
frees); replace malloc(sizeof(account_details_t)) with calloc(1,
sizeof(account_details_t)) for the ad allocations (both places where
account_details_t *ad is created) so the struct is zero-initialized, and keep
the existing explicit assignments for other fields; ensure this fixes fields
used by free_account_details() such as sha1_pass and clear_text_password.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
Run ID: 4545da87-b745-4258-8452-40d880cc9a19
📒 Files selected for processing (7)
include/LLM_Bridge.hinclude/MySQL_LDAP_Authentication.hppinclude/proxysql_coredump.hinclude/proxysql_restapi.hinclude/proxysql_sslkeylog.hinclude/query_processor.hlib/MySQL_Authentication.cpp
✅ Files skipped from review due to trivial changes (4)
- include/proxysql_restapi.h
- include/proxysql_coredump.h
- include/proxysql_sslkeylog.h
- include/LLM_Bridge.h
🚧 Files skipped from review as they are similar to previous changes (1)
- include/query_processor.h
📜 Review details
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (5)
- GitHub Check: CI-builds / builds (debian12,-dbg)
- GitHub Check: CI-builds / builds (ubuntu22,-tap)
- GitHub Check: CI-builds / builds (ubuntu24,-tap-genai-gcov)
- GitHub Check: run / trigger
- GitHub Check: claude-review
🧰 Additional context used
📓 Path-based instructions (4)
**/*.{cpp,h,hpp}
📄 CodeRabbit inference engine (CLAUDE.md)
**/*.{cpp,h,hpp}: Class names must use PascalCase with protocol prefixes (MySQL_, PgSQL_, ProxySQL_)
Member variables must use snake_case
Constants and macros must use UPPER_SNAKE_CASE
C++17 is required; use conditional compilation via#ifdefPROXYSQLGENAI,#ifdefPROXYSQL31, etc. for feature flags
Use pthread mutexes for synchronization and std::atomic<> for counters
Files:
lib/MySQL_Authentication.cppinclude/MySQL_LDAP_Authentication.hpp
**/*.cpp
📄 CodeRabbit inference engine (CLAUDE.md)
Use RAII for resource management and jemalloc for memory allocation
Files:
lib/MySQL_Authentication.cpp
lib/**/*.cpp
📄 CodeRabbit inference engine (CLAUDE.md)
One class per file is the typical convention
Files:
lib/MySQL_Authentication.cpp
include/**/*.{h,hpp}
📄 CodeRabbit inference engine (CLAUDE.md)
Include guards must follow the pattern
#ifndef_CLASS*_H
Files:
include/MySQL_LDAP_Authentication.hpp
🧠 Learnings (3)
📓 Common learnings
Learnt from: yuji-hatakeyama
Repo: sysown/proxysql PR: 5307
File: test/tap/tests/reg_test_5306-show_warnings_with_comment-t.cpp:39-48
Timestamp: 2026-01-20T09:34:27.165Z
Learning: In ProxySQL test files (test/tap/tests/), resource leaks (such as not calling `mysql_close()` on early return paths) are not typically fixed because test processes are short-lived and the OS frees resources on process exit. This is a common pattern across the test suite.
Learnt from: CR
Repo: sysown/proxysql PR: 0
File: CLAUDE.md:0-0
Timestamp: 2026-03-22T14:38:16.093Z
Learning: Applies to **/*.{cpp,h,hpp} : C++17 is required; use conditional compilation via `#ifdef` PROXYSQLGENAI, `#ifdef` PROXYSQL31, etc. for feature flags
📚 Learning: 2026-03-22T14:38:16.093Z
Learnt from: CR
Repo: sysown/proxysql PR: 0
File: CLAUDE.md:0-0
Timestamp: 2026-03-22T14:38:16.093Z
Learning: Applies to **/*.{cpp,h,hpp} : C++17 is required; use conditional compilation via `#ifdef` PROXYSQLGENAI, `#ifdef` PROXYSQL31, etc. for feature flags
Applied to files:
include/MySQL_LDAP_Authentication.hpp
📚 Learning: 2026-03-22T14:38:16.093Z
Learnt from: CR
Repo: sysown/proxysql PR: 0
File: CLAUDE.md:0-0
Timestamp: 2026-03-22T14:38:16.093Z
Learning: Admin interface uses SQLite3 backend for SQL-based configuration and schema versions are tracked in ProxySQL_Admin_Tables_Definitions.h
Applied to files:
include/MySQL_LDAP_Authentication.hpp
🪛 Clang (14.0.6)
include/MySQL_LDAP_Authentication.hpp
[error] 5-5: unknown type name 'uint8_t'
(clang-diagnostic-error)
[error] 24-24: use of undeclared identifier 'std'
(clang-diagnostic-error)
[error] 24-24: use of undeclared identifier 'SQLite3_result'
(clang-diagnostic-error)
[error] 34-34: unknown type name 'SQLite3_result'
(clang-diagnostic-error)
[error] 35-35: unknown type name 'SQLite3_result'
(clang-diagnostic-error)
[error] 36-36: unknown type name 'SQLite3_result'
(clang-diagnostic-error)
[error] 37-37: unknown type name 'uint64_t'
(clang-diagnostic-error)
[error] 38-38: unknown type name 'SQLite3_result'
(clang-diagnostic-error)
🪛 Cppcheck (2.20.0)
lib/MySQL_Authentication.cpp
[warning] 254-254: If memory allocation fails, then there is a possible null pointer dereference
(nullPointerOutOfMemory)
[warning] 361-361: If memory allocation fails, then there is a possible null pointer dereference
(nullPointerOutOfMemory)
[warning] 362-362: If memory allocation fails, then there is a possible null pointer dereference
(nullPointerOutOfMemory)
[warning] 386-386: If memory allocation fails, then there is a possible null pointer dereference
(nullPointerOutOfMemory)
[warning] 387-387: If memory allocation fails, then there is a possible null pointer dereference
(nullPointerOutOfMemory)
🔇 Additional comments (2)
include/MySQL_LDAP_Authentication.hpp (1)
20-22:nullptrmigration is clean and consistent.These replacements improve type safety and readability; no concerns on this segment.
Also applies to: 29-31, 35-36, 38-38
lib/MySQL_Authentication.cpp (1)
78-118: Consistent rename toactive_.This rename lines up with
include/MySQL_Authentication.hppand the sibling auth backends, so the inactive-mark/sweep flow stays coherent.
|
|
||
| struct LDAP_USER_FIELD_IDX { | ||
| enum index { | ||
| enum index : uint8_t { |
There was a problem hiding this comment.
🧩 Analysis chain
🏁 Script executed:
cat -n include/MySQL_LDAP_Authentication.hppRepository: sysown/proxysql
Length of output: 2097
Add required includes and forward declarations for header self-sufficiency.
This header uses uint8_t, uint64_t, std::unique_ptr<>, and SQLite3_result without the necessary includes or declarations:
uint8_tanduint64_t(line 5, 37) require#include <cstdint>std::unique_ptr<>(line 24) requires#include <memory>SQLite3_result(lines 24, 34, 35, 36, 38) requires a forward declaration or include
Without these, the header will fail to parse standalone and break compilation.
Proposed fix
`#ifndef` CLASS_MYSQL_LDAP_AUTHENTICATION_H
`#define` CLASS_MYSQL_LDAP_AUTHENTICATION_H
+#include <cstdint>
+#include <memory>
+
+class SQLite3_result;
+
struct LDAP_USER_FIELD_IDX {
enum index : uint8_t {
USERNAME = 0,🧰 Tools
🪛 Clang (14.0.6)
[error] 5-5: unknown type name 'uint8_t'
(clang-diagnostic-error)
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@include/MySQL_LDAP_Authentication.hpp` at line 5, The header is not
self-sufficient: types used by enum index (uint8_t, uint64_t), the
std::unique_ptr members, and the SQLite3_result references are undeclared; add
`#include` <cstdint> and `#include` <memory> at the top and either include the
appropriate SQLite header or add a forward declaration like "struct
SQLite3_result;" so symbols referenced in MySQL_LDAP_Authentication.hpp (enum
index, uint8_t, uint64_t, std::unique_ptr<...>, and SQLite3_result) are defined
when this header is parsed standalone.
There was a problem hiding this comment.
Actionable comments posted: 3
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@lib/MySQL_HostGroups_Manager.cpp`:
- Line 1880: The debug fprintf prints the latency label "max_latency_ms" but
uses mysrvc->max_latency_us * 1000 (microseconds multiplied), yielding values
1000x too large; update the print expression for mysrvc->max_latency_us in the
fprintf call (the line that references mysrvc->myhgc->hid, mysrvc->address, ...
mysrvc->max_latency_us*1000, mysrvc->comment) to convert microseconds to
milliseconds by dividing by 1000 instead of multiplying so the printed
"max_latency_ms" value matches the rest of the file.
- Around line 3581-3583: The code silently skips when host_server_mapping is
null; before the continue add a warning log using the same logging facility used
elsewhere in this file (e.g., MXB_WARNING / MXS_WARNING or the local logger)
that reports the unexpected null mapping and any available context (host name,
host id, server identifier or the variables in scope) so topology drift is
visible; keep the continue but ensure the log message clearly indicates the
function/loop (mention host_server_mapping) and relevant identifiers to aid
diagnosis.
- Around line 4997-4999: Change the declaration of q from a non-const pointer to
a const char* (e.g., const char* q = "DELETE ...") so the string literal is
treated as const in this C++17 code path; update any uses (snprintf call using
q, qsz, query, _hostname, _port, info->offline_hostgroup,
info->backup_writer_hostgroup, info->writer_hostgroup) accordingly to avoid
discarding qualifiers.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
Run ID: 152094ce-a319-4d99-aa6e-e8978820986f
📒 Files selected for processing (2)
include/gen_utils.hlib/MySQL_HostGroups_Manager.cpp
🚧 Files skipped from review as they are similar to previous changes (1)
- include/gen_utils.h
📜 Review details
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (5)
- GitHub Check: CI-builds / builds (ubuntu22,-tap)
- GitHub Check: CI-builds / builds (debian12,-dbg)
- GitHub Check: CI-builds / builds (ubuntu24,-tap-genai-gcov)
- GitHub Check: run / trigger
- GitHub Check: claude-review
🧰 Additional context used
📓 Path-based instructions (3)
**/*.{cpp,h,hpp}
📄 CodeRabbit inference engine (CLAUDE.md)
**/*.{cpp,h,hpp}: Class names must use PascalCase with protocol prefixes (MySQL_, PgSQL_, ProxySQL_)
Member variables must use snake_case
Constants and macros must use UPPER_SNAKE_CASE
C++17 is required; use conditional compilation via#ifdefPROXYSQLGENAI,#ifdefPROXYSQL31, etc. for feature flags
Use pthread mutexes for synchronization and std::atomic<> for counters
Files:
lib/MySQL_HostGroups_Manager.cpp
**/*.cpp
📄 CodeRabbit inference engine (CLAUDE.md)
Use RAII for resource management and jemalloc for memory allocation
Files:
lib/MySQL_HostGroups_Manager.cpp
lib/**/*.cpp
📄 CodeRabbit inference engine (CLAUDE.md)
One class per file is the typical convention
Files:
lib/MySQL_HostGroups_Manager.cpp
🧠 Learnings (1)
📚 Learning: 2026-01-20T09:34:27.165Z
Learnt from: yuji-hatakeyama
Repo: sysown/proxysql PR: 5307
File: test/tap/tests/reg_test_5306-show_warnings_with_comment-t.cpp:39-48
Timestamp: 2026-01-20T09:34:27.165Z
Learning: In ProxySQL test files (test/tap/tests/), resource leaks (such as not calling `mysql_close()` on early return paths) are not typically fixed because test processes are short-lived and the OS frees resources on process exit. This is a common pattern across the test suite.
Applied to files:
lib/MySQL_HostGroups_Manager.cpp
🔇 Additional comments (1)
lib/MySQL_HostGroups_Manager.cpp (1)
1054-1054: Good fix on the post-insert lookup.Fetching the pointer after insertion avoids keeping a raw pointer from a moved
unique_ptr, which was the real hazard in this path.
| break; | ||
| } | ||
| fprintf(stderr,"HID: %d , address: %s , port: %d , gtid_port: %d , weight: %ld , status: %s , max_connections: %ld , max_replication_lag: %u , use_ssl: %u , max_latency_ms: %u , comment: %s\n", mysrvc->myhgc->hid, mysrvc->address, mysrvc->port, mysrvc->gtid_port, mysrvc->weight, st, mysrvc->max_connections, mysrvc->max_replication_lag, mysrvc->use_ssl, mysrvc->max_latency_us*1000, mysrvc->comment); | ||
| fprintf(stderr,"HID: %u , address: %s , port: %d , gtid_port: %d , weight: %ld , status: %s , max_connections: %ld , max_replication_lag: %u , use_ssl: %d , max_latency_ms: %u , comment: %s\n", mysrvc->myhgc->hid, mysrvc->address, mysrvc->port, mysrvc->gtid_port, mysrvc->weight, st, mysrvc->max_connections, mysrvc->max_replication_lag, mysrvc->use_ssl, mysrvc->max_latency_us*1000, mysrvc->comment); |
There was a problem hiding this comment.
Fix the latency unit in this debug dump.
The label says max_latency_ms, but this prints mysrvc->max_latency_us * 1000. Everywhere else in the file converts max_latency_us to milliseconds with / 1000, so this log will report values far too large and make troubleshooting misleading.
Suggested fix
- fprintf(stderr,"HID: %u , address: %s , port: %d , gtid_port: %d , weight: %ld , status: %s , max_connections: %ld , max_replication_lag: %u , use_ssl: %d , max_latency_ms: %u , comment: %s\n", mysrvc->myhgc->hid, mysrvc->address, mysrvc->port, mysrvc->gtid_port, mysrvc->weight, st, mysrvc->max_connections, mysrvc->max_replication_lag, mysrvc->use_ssl, mysrvc->max_latency_us*1000, mysrvc->comment);
+ fprintf(stderr,"HID: %u , address: %s , port: %d , gtid_port: %d , weight: %ld , status: %s , max_connections: %ld , max_replication_lag: %u , use_ssl: %d , max_latency_ms: %u , comment: %s\n", mysrvc->myhgc->hid, mysrvc->address, mysrvc->port, mysrvc->gtid_port, mysrvc->weight, st, mysrvc->max_connections, mysrvc->max_replication_lag, mysrvc->use_ssl, mysrvc->max_latency_us/1000, mysrvc->comment);📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
| fprintf(stderr,"HID: %u , address: %s , port: %d , gtid_port: %d , weight: %ld , status: %s , max_connections: %ld , max_replication_lag: %u , use_ssl: %d , max_latency_ms: %u , comment: %s\n", mysrvc->myhgc->hid, mysrvc->address, mysrvc->port, mysrvc->gtid_port, mysrvc->weight, st, mysrvc->max_connections, mysrvc->max_replication_lag, mysrvc->use_ssl, mysrvc->max_latency_us*1000, mysrvc->comment); | |
| fprintf(stderr,"HID: %u , address: %s , port: %d , gtid_port: %d , weight: %ld , status: %s , max_connections: %ld , max_replication_lag: %u , use_ssl: %d , max_latency_ms: %u , comment: %s\n", mysrvc->myhgc->hid, mysrvc->address, mysrvc->port, mysrvc->gtid_port, mysrvc->weight, st, mysrvc->max_connections, mysrvc->max_replication_lag, mysrvc->use_ssl, mysrvc->max_latency_us/1000, mysrvc->comment); |
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@lib/MySQL_HostGroups_Manager.cpp` at line 1880, The debug fprintf prints the
latency label "max_latency_ms" but uses mysrvc->max_latency_us * 1000
(microseconds multiplied), yielding values 1000x too large; update the print
expression for mysrvc->max_latency_us in the fprintf call (the line that
references mysrvc->myhgc->hid, mysrvc->address, ... mysrvc->max_latency_us*1000,
mysrvc->comment) to convert microseconds to milliseconds by dividing by 1000
instead of multiplying so the printed "max_latency_ms" value matches the rest of
the file.
| if (!host_server_mapping) { | ||
| continue; | ||
| } |
There was a problem hiding this comment.
Don’t silently swallow a null mapping here.
If host_server_mapping is unexpectedly null, skipping the server with no log leaves topology drift invisible. Please emit at least a warning before continue so this state is diagnosable.
Suggested fix
if (!host_server_mapping) {
+ proxy_warning("Server mapping unexpectedly null for %s:%d\n", hostname.c_str(), port);
continue;
}📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
| if (!host_server_mapping) { | |
| continue; | |
| } | |
| if (!host_server_mapping) { | |
| proxy_warning("Server mapping unexpectedly null for %s:%d\n", hostname.c_str(), port); | |
| continue; | |
| } |
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@lib/MySQL_HostGroups_Manager.cpp` around lines 3581 - 3583, The code silently
skips when host_server_mapping is null; before the continue add a warning log
using the same logging facility used elsewhere in this file (e.g., MXB_WARNING /
MXS_WARNING or the local logger) that reports the unexpected null mapping and
any available context (host name, host id, server identifier or the variables in
scope) so topology drift is visible; keep the continue but ensure the log
message clearly indicates the function/loop (mention host_server_mapping) and
relevant identifiers to aid diagnosis.
| q="DELETE FROM mysql_servers_incoming WHERE hostname='%s' AND port=%d AND hostgroup_id in (%d, %d, %d)"; | ||
| //query=(char *)malloc(strlen(q)+strlen(_hostname)+64); | ||
| sprintf(query,q,_hostname,_port, info->offline_hostgroup, info->backup_writer_hostgroup, info->writer_hostgroup, info->writer_hostgroup); | ||
| snprintf(query, qsz, q, _hostname, _port, info->offline_hostgroup, info->backup_writer_hostgroup, info->writer_hostgroup); |
There was a problem hiding this comment.
🧩 Analysis chain
🌐 Web query:
In C++17, can a string literal be assigned to char*, or must it be stored in const char*?
💡 Result:
In C++17, an ordinary string literal like "hello" has type const char[N] (an array of const char) and decays to const char* in most expressions, so it must be stored in const char* (or std::string, etc.). (en.cppreference.com)
Assigning it to char* (e.g., char* p = "hello";) is ill-formed in standard C++ (many compilers may still accept it as a non-standard extension with a deprecation warning). (en.cppreference.com)
Citations:
- 1: https://en.cppreference.com/w/cpp/language/string_literal.html?utm_source=openai
- 2: https://en.cppreference.com/w/cpp/language/string_literal.html?utm_source=openai
🏁 Script executed:
# First, check the actual code at lines 4997-4999
sed -n '4995,5005p' lib/MySQL_HostGroups_Manager.cpp | cat -nRepository: sysown/proxysql
Length of output: 769
🏁 Script executed:
# Search for the declaration of q in the function context around line 4997
sed -n '4950,5010p' lib/MySQL_HostGroups_Manager.cpp | head -80Repository: sysown/proxysql
Length of output: 2899
Keep the SQL template const in this C++17 path.
Line 4997 assigns a string literal to char* q. In C++17, string literals have type const char*, so this reintroduces a non-const-literal conversion. Change the assignment to use const char*:
- q="DELETE FROM mysql_servers_incoming WHERE hostname='%s' AND port=%d AND hostgroup_id in (%d, %d, %d)";
+ const char* q = "DELETE FROM mysql_servers_incoming WHERE hostname='%s' AND port=%d AND hostgroup_id in (%d, %d, %d)";
snprintf(query, qsz, q, _hostname, _port, info->offline_hostgroup, info->backup_writer_hostgroup, info->writer_hostgroup);📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
| q="DELETE FROM mysql_servers_incoming WHERE hostname='%s' AND port=%d AND hostgroup_id in (%d, %d, %d)"; | |
| //query=(char *)malloc(strlen(q)+strlen(_hostname)+64); | |
| sprintf(query,q,_hostname,_port, info->offline_hostgroup, info->backup_writer_hostgroup, info->writer_hostgroup, info->writer_hostgroup); | |
| snprintf(query, qsz, q, _hostname, _port, info->offline_hostgroup, info->backup_writer_hostgroup, info->writer_hostgroup); | |
| const char* q = "DELETE FROM mysql_servers_incoming WHERE hostname='%s' AND port=%d AND hostgroup_id in (%d, %d, %d)"; | |
| //query=(char *)malloc(strlen(q)+strlen(_hostname)+64); | |
| snprintf(query, qsz, q, _hostname, _port, info->offline_hostgroup, info->backup_writer_hostgroup, info->writer_hostgroup); |
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@lib/MySQL_HostGroups_Manager.cpp` around lines 4997 - 4999, Change the
declaration of q from a non-const pointer to a const char* (e.g., const char* q
= "DELETE ...") so the string literal is treated as const in this C++17 code
path; update any uses (snprintf call using q, qsz, query, _hostname, _port,
info->offline_hostgroup, info->backup_writer_hostgroup, info->writer_hostgroup)
accordingly to avoid discarding qualifiers.
|
1 participant
Summary
This PR applies clang-tidy and cppcheck static analysis fixes to improve code quality and fix real bugs in ProxySQL.
Changes Overview
__CLASS_*→PROXYSQL_*_H__var→var_,__size→SIZE_Detailed Fixes
1. Clang-Tidy Fixes (Reserved Identifiers)
Include guards (37 files):
#ifndef __CLASS_*→#ifndef PROXYSQL_*_H#endifcommentsMember variables:
__active→active___user→user___sess→sess_Enum sentinels:
__size→SIZE_(for Prometheus metric enums)ROS__SIZE→ROS_SIZE_Constant names:
__HGM_TABLES_SIZE→HGM_TABLES_SIZE___TYPE_SIZE→TYPE_SIZE_NULL → nullptr (32 fixes in headers)
2. Cppcheck Fixes (Real Bugs Fixed)
Memory Issues
gen_utils.h): Fixedreallocthat could lose original pointer on failureMySQL_HostGroups_Manager.cpp): Fixed dangling pointer to moved unique_ptrUninitialized Variables (~40 fixes)
Query_Processor_Output: 22 member variablesMySQL_Protocol: 4 member variablesQuery_Info: 4 member variablesMySQL_Query_Processor_Output: 2 member variablesstmt_execute_metadata_t: 2 member variablesBuffer/Format Bugs
%d/%umismatchesNull Pointer Bugs
3. Performance Improvements
Files Changed
Key Files Modified
include/*.h(37 files) - Include guards, nullptr, enum fixeslib/MySQL_HostGroups_Manager.cpp- Memory safety, printf bugslib/MySQL_Session.cpp- Null pointers, array bounds, copy opslib/MySQL_Authentication.cpp- Member initializationinclude/query_processor.h- Uninitialized membersinclude/gen_utils.h- Memory leak fixscripts/lint/- New lint scripts addedNew Lint Infrastructure
Added comprehensive lint scripts (from earlier PR):
scripts/lint/run-local.sh- Local lint runnerscripts/lint/normalize-clang-tidy.py- Normalize clang-tidy outputscripts/lint/normalize-cppcheck.py- Normalize cppcheck outputscripts/lint/generate-compile-commands.sh- Generate compile commandsTesting
make debugBefore/After Cppcheck
Notes
Summary by CodeRabbit
New Features
Documentation
Refactor
Chores