[jayhemnani9910]

renderInlineAsText skips code_inline tokens, so a code span in an image description vanishes from the alt attribute:

>>> from markdown_it import MarkdownIt
>>> MarkdownIt("commonmark").render("![a `code` b](x.png)")
'<p><img src="x.png" alt="a  b" /></p>\n'

markdown-it 15 gives alt="a code b", and the CommonMark spec keeps the plain text of the description. Upstream's renderInlineAsText handles code_inline together with text; this does the same.

I ran markdown-it 15.0.3 and markdown-it-py 4.2.0 (commonmark preset) over about 3000 Markdown files from real projects; this was the only difference in the output.

Tests: new case in tests/test_port/fixtures/issue-fixes.md, which fails without the change. pytest with the linkify and plugins extras: 1033 passed.

(Add a CHANGELOG line under Unreleased once the PR number is known, like the other entries.)