[arxhr007]

Summary

A valid substring placeholder with an end index, {1~start,end}, is printed as is when the start index is outside of the string. The same placeholder without the end index prints nothing, and so do {1~10,10} and {1~10,5}, which are covered by the existing tests.

With 12345 67890 (11 characters):

Format	Before	After
{1~20}	`` (empty)	`` (empty)
{1~11,12}	{1~11,12}	`` (empty)
{1~20,30}	{1~20,30}	`` (empty)
{1~-50,3}	{1~-50,3}	`` (empty)
{1~-50,-1}	{1~-50,-1}	`` (empty)

The cause is that the ,end part was only parsed inside if (start >= 0 && start < length). When the start was outside, the parser pointer stayed on the ,, and the check for trailing characters then took the valid placeholder for an invalid one. The existing tests for invalid syntax ({1~-,1}, {1~0,,1}, {1~0,1,} and so on) still print the placeholder as is.

Related issue (required for new logos for new distros)

Not a logo. No issue, it is a small fix in the same code as #2669.

Changes

• src/common/impl/format.c: parse the optional end index first and check the whole range before using it. A valid range outside of the string now gives an empty result.
• tests/format.c: four cases for a start index outside of the string with an end index. The first fails on the old code.
• CHANGELOG.md: bugfix entry.

Screenshots

Not a visual change.

Checklist

• I have tested my changes locally.

Built with -DBUILD_TESTS=On -DENABLE_ASAN=On and -fsanitize=undefined; all 19 tests pass.

[CarterLi]

Conflict

[arxhr007]

Rebased onto dev. The conflict was with #2696: both touch the ~ substring parsing and the changelog. I kept the range check from this PR and made it use the new strToInt32(). The new cases in tests/format.c fail on current dev and pass with this change; all tests pass with ASan and UBSan.