[Add] AttributeValueXHTML.ExtractNormalizedTextFromValue for plain text with normalized whitespace; fixes #92 - #102

Prerequisites

• I have written a descriptive pull-request title
• I have verified that there are no overlapping pull-requests open
• I have verified that I am following the ReqIFSharp code style guidelines
• I have provided test coverage for my change (where applicable)

Description

Fixes #92.

AttributeValueXHTML.ExtractUnformattedTextFromValue() joins the text nodes with a space, so its result has extra spaces around inline elements and keeps the line breaks and indentation of the XHTML source (brew a cup in under 30 seconds .). This adds a second method and leaves the existing one unchanged.

New: ExtractNormalizedTextFromValue(bool keepLineBreaks = false)

• Text is joined as it is written, so no space is added around inline elements: the example from the issue gives brew a cup in under 30 seconds.
• Outside pre, runs of space, tab, carriage return and line feed are collapsed to one space, and each line is trimmed.
• Block elements (p, div, li, headings, blockquote, dl/dt/dd, table, tr, ...) and br separate lines; table cells are separated by a space. Lines are joined with a space, or with \n when keepLineBreaks is true.
• The text of pre is kept as written (spaces, tabs, indentation, one line per source line); blank lines at its start and end are left out, blank lines inside it are kept only with keepLineBreaks.
• Non-breaking spaces (U+00A0) are kept: not collapsed, replaced or trimmed.
• XML entities and character references are expanded; HTML named entities such as   and é (which no DTD declares in ReqIF) are decoded with WebUtility.HtmlDecode; an unknown entity throws an XmlException with its line and position.
• The value is read as an XML fragment, so several top-level elements, plain text and undeclared prefixes work, and error positions are those in the value. DTD processing is prohibited and no resolver is used.

These rules for pre, non-breaking spaces and entities were decided during the review.

ExtractUnformattedTextFromValue is unchanged; a test pins its output for the example. Note that it still drops HTML named entities silently (café gives caf).

Tests (AttributeValueXHTMLTestFixture)

The issue example; empty values; source line breaks and indentation; every block and cell element with and without keepLineBreaks, including upper-case and unprefixed names; whitespace between inline elements and xml:space="preserve"; non-breaking spaces; pre (CRLF and lone CR, indentation, blank lines, br, inline elements, nested and adjacent blocks, empty pre); several top-level elements, plain text, CDATA, comments and processing instructions; attribute values (also with an entity) not in the text; references and HTML entities; unknown entities and malformed values with their exact line and column; DOCTYPE rejected; every XHTML value of reqifsharpgenerated.reqif.

dotnet build -c Release: 0 warnings; dotnet test: 620 passed. The README quick-start mentions the new method.

🤖 Generated with Claude Code