Module text 

Text helpers shared by the engines: markup stripping, previews, whole-word matching, tag balance and serialized sizes.

Statics

WHITESPACE_TO_COLLAPSE 🔒

Whitespace that isn’t a single space; collapsing only these leaves most text borrowed.

Functions

collapse_whitespace 🔒

is_inside_character_reference

Whether byte offset at falls after the & of a character reference such as & and at or before its ;.

is_inside_tag

Whether byte offset at of html falls inside a tag (<a href="…">).

is_word_char 🔒

Letters, digits and _, non-ASCII ones included. main-frontend’s glossary highlighting (textParsing.tsx) uses the same set; change both together.

open_tags_at

Inline elements still open at byte offset at, outermost first.

plain_text

The visible text of a rich-text string: tags removed, common entities decoded, whitespace collapsed to single spaces.

preview

Cuts text to at most max_chars characters at a word boundary, adding an ellipsis when cut.

serialized_len

The size of value as JSON, in bytes, without building the JSON.

whole_word_matches

Byte ranges of the occurrences of needle in text that don’t start or end inside a word. An end of needle that isn’t a word character (as in “C++”) needs no word boundary.