RDKEMW-17573 : Triggering a issue type with more than 35 chars in RRD is leading RDM crash

[Abhinavpv28]

Reason For change : Restricing the number of characters allocated for dynamic issue category to 34, since commonutilities has a restriction on postfield, beyond this length causing a crash in rdm

[github-actions]

Code Coverage Summary

                               Total:|83.5%   5890|97.0%  1531|    -      0

[Copilot]

Pull request overview

This PR updates processIssueTypeInStaticProfile() in the remote debugger event processing flow by adding an inline comment near the fallback path that checks the installed package when an issue isn’t found in the static JSON profile.

Changes:

• Added an inline note/TODO comment before calling processIssueTypeInInstalledPackage() when the issue is not present in the static profile.

[Copilot]

@copilot please work on the comment added in the PR

Addressed in commit efa4d22: removed the unclear inline note from src/rrdEventProcess.c as requested by the PR review comment. No UI changes were involved in this update.

[rdkcmf-jenkins]

b'## Copyright scan failure
Commit: 899b304
Report detail: https://gist.github.com/rdkcmf-jenkins/e04e078f7dae01113094a4e3ba839f2b'

[github-actions]

Code Coverage Summary

                               Total:|83.2%   5917|96.8%  1534|    -      0

[Copilot]

Pull request overview

Copilot reviewed 2 out of 2 changed files in this pull request and generated 4 comments.

[github-actions]

Code Coverage Summary

                               Total:|84.1%   6235|97.3%  1660|    -      0

[github-actions]

Code Coverage Summary

                               Total:|84.1%   6235|97.3%  1660|    -      0

[github-actions]

Code Coverage Summary

                               Total:|84.1%   6235|97.3%  1660|    -      0

[Copilot]

Pull request overview

Copilot reviewed 2 out of 2 changed files in this pull request and generated 3 comments.

[nhanasi]

@Abhinavpv28 - can we add L1/L2 for above scenario change

[Abhinavpv28]

@copilot add gtest for new changes

[Abhinavpv28]

@copilot add gtest for new changes

[Abhinavpv28]

@copilot add gtest for new changes

[github-actions]

Code Coverage Summary

                               Total:|84.1%   6235|97.3%  1660|    -      0

[Abhinavpv28]

@copilot
add gtest for new changes

[Abhinavpv28]

@copilot add gtest testcases for new change

[Copilot]

Pull request overview

Copilot reviewed 3 out of 3 changed files in this pull request and generated 2 comments.

Comments suppressed due to low confidence (4)

src/unittest/rrdUnitTestRunner.cpp:1900

• These cacheData nodes no longer initialize the prev pointer. remove_item() dereferences cache->prev (rrdCacheprevnode = cache->prev), so leaving it uninitialized can cause undefined behavior/crashes in the test. Initialize prev to NULL for both node and node_dummy.

            }
            free(cacheDataNode);
            cacheDataNode = next;
        }
    }
};

TEST_F(RemoveItemTest, HandlesCacheNull)
{
    remove_item(NULL);

src/unittest/rrdUnitTestRunner.cpp:1967

• The expected token count/values here don’t match issueTypeSplitter() behavior in rrdEventProcess.c, which only splits on the delimiter and does not remove special characters. For input "a@,b,&,cd,ef" it will return 5 tokens (including "a@" and "&"). Update the assertions or sanitize the input before calling issueTypeSplitter().

{
    char str[] = "a,b.c";
    removeSpecialCharacterfromIssueTypeList(str);

    EXPECT_STREQ(str, "a,b.c");
}

TEST(RemoveSpecialCharacterfromIssueTypeListTest, HandlesStringWithoutCommaAndDot)
{

src/unittest/rrdUnitTestRunner.cpp:2090

• data_buf rbuf is no longer zero-initialized, leaving fields like appendMode/dsEvent uninitialized. processIssueTypeEvent() reads appendMode and may take different branches nondeterministically. Prefer data_buf rbuf = {}; (or call RRD_data_buff_init) and free rbuf.mdata after the call to avoid leaks.

    EXPECT_STREQ(suffix, "");
}

TEST(SplitIssueTypeTest, BaseTruncatedWhenTooSmall)
{
    /* "abc_suffix": suffix "_suffix" is 7 chars (≤ 9) → accepted.

src/unittest/rrdUnitTestRunner.cpp:1219

• This test’s setup can’t actually exercise an allocation-failure path: RRDGetProfileStringLength() now enforces RRD_DYNAMIC_PROFILE_MAX_LENGTH via strnlen() and returns -1 for long Node values, so RRDRdmManagerDownloadRequest() will bail out before attempting a large calloc. Update the test to assert the max-length rejection, or introduce a deterministic fault-injection mechanism if you want to cover allocation failures.

    RRDRdmManagerDownloadRequest(NULL, buff.jsonPath, &buff, false);
    free(issuestructNode.Node);
    free(issuestructNode.subNode);
    free(buff.jsonPath);
    free(buff.mdata);
}

TEST_F(RRDRdmManagerDownloadRequestTest, HandlesParamStringAllocFail)
{
    // Simulate calloc failure by passing mSGLength=0 (should skip allocation)
    issueNodeData issuestructNode;
    issuestructNode.Node = strdup("MainNode");

[github-actions]

Code Coverage Summary

                               Total:|84.3%   6289|97.3%  1676|    -      0