markdownLinkOpen top-level constant
String
const markdownLinkOpen
Matches only the opening of a markdown link: the [text] label followed
by optional whitespace and the ( that begins the destination.
The destination itself is intentionally not captured here. The previous
single-regex form \[([^\[\]]*)\]\((.*?)\) was fragile in two ways:
- the non-greedy
(.*?)stopped at the first), so URLs containing balanced parentheses (e.g./Primer_(film)) were truncated, and - the captured group swallowed any whitespace padding inside the
parentheses (
( url )), which then desynced the length-based end index and left)/spaces behind informat()output.
Instead we match just the opener and let MarkdownLinksExtractor scan the
destination structurally (leading/trailing whitespace, optional <...>
wrapping, balanced parens, and the true closing )). The \s* between ]
and ( also tolerates a gap such as [text] (url); a false match is
prevented downstream because the destination must still parse as a valid URL.
Implementation
const markdownLinkOpen = r'\[([^\[\]]*)\]\s*\(';