XDocument: Cannot identify and manipulate XText nodes without breaking whitespace formatting

[SetTrend]

Description

.NET Runtime System.Xml.Linq.XDocument Issue

It is not possible to identify and manipulate XText nodes without breaking the original whitespace formatting.

1. Using XDocument.Parse(String),

   … XText nodes can correctly be identified, but whitespace formatting is broken, i.e., whitespace is unsolicitedly added.

2. Using XDocument.Parse(String, LoadOptions.PreserveWhitespace),

   … significant XText nodes can not correctly be identified and segregated from insignificant whitespace XText nodes.

Reproduction Steps

See this MRE:

https://github.com/SetTrend/XDocument-MRE

Expected behavior

It should be possible to identify and manipulate significant XText nodes without breaking the original whitespace formatting.

Actual behavior

See MRE.

Configuration

.NET 9.0 on Windows 10x64.

[dotnet-policy-service]

Tagging subscribers to this area: @dotnet/area-system-xml
See info in area-owners.md if you want to be subscribed.