How can I match text up to the next delimiter without crossing it?

Use a character class that excludes the delimiter, then match the delimiter explicitly: ([^;]*);. The parentheses capture the field, [^;]* consumes zero or more characters that are not semicolons, and the final ; matches the delimiter itself. This avoids the problem with a broad wildcard such as .*;: .* can consume semicolons, so the match may extend beyond the delimiter you meant to stop at. [S1]

For example, applying ([^;]*); at the start of red;blue; captures red and matches the first semicolon. To locate a field specifically at the beginning of a line, use ^([^;]*);, with ^ marking the start in this pattern. Without a start anchor or other positioning logic, where the match is found depends on the matching method and surrounding pattern. [S1]

With *, an empty field is allowed. For ;blue, ([^;]*); matches the first semicolon and captures an empty string. Adjacent delimiters can represent an empty field too: in red;;blue, starting at the second field’s position, the pattern captures an empty value before the second semicolon.

If you need at least one character in the field, replace * with +: ([^;]+);. That version does not match an empty field. In either version, the explicit semicolon means the complete pattern cannot match a field without a following semicolon. For instance, red alone does not satisfy ([^;]*);. [S1] By contrast, [^;]* by itself can match an empty string even when no delimiter exists; it does not establish that a delimiter was present. [S2]

Assume each input below is one field and matching begins at its first character. These examples use Python-style re syntax; the character-class pattern also has the described behavior in JavaScript regex syntax. [S1] [S2]

1. Input name;rest with ^([^;]*); → match name;, capture name.
2. Input ;rest with ^([^;]*); → match ;, capture an empty string.
3. Input name with ^([^;]*); → no match, because there is no semicolon.
4. Input name;rest with ^([^;]+); → match name;, capture name.
5. Input ;rest with ^([^;]+); → no match, because + requires at least one non-semicolon character. [S1]

The ^ makes these examples start at the beginning of the input. If the field begins at a known position within a larger string, arrange the surrounding pattern or extraction logic to start there instead. A pattern that can search elsewhere may find a later field, which might not be the one you intended. [S1]

Start with this template, replacing ; with your chosen single-character delimiter:

^([^;]*);

Use it when an empty field is acceptable and the delimiter must be present. For nonempty fields, use ^([^;]+);. The whole match includes the delimiter, while the captured field is in the parenthesized group. [S1] [S2]

Before relying on it, try ordinary, empty, and missing-delimiter inputs like the examples above. Confirm where the field starts and whether your delimiter is a single character or a more complex separator. These patterns address a single semicolon delimiter; they do not establish how to parse quoted or escaped delimiters inside a field. If those occur in your data, use a pattern or parser appropriate to that format rather than assuming this simple pattern handles them.