using Scribe to extract text patterns from Word

[SMckenna]

Is it true that Scribe functions (such as ScribePatternMatchAll) are intended to operate on text extracted from a file into a Filemaker text field rather than directly on the file itself (eg held in a container field) ?  For example, if I use ScribeFileAsText to bring a word document content into a Filemaker text field, the function ScribePatternMatchAll works as expected. However, if I attempt to make ScribePatternMatchAll work on a file loaded into a Filemaker container field, the function doesn't work properly (it sometimes finds none, one or a few instances of a word pattern for example when there are actually numerous instances of that word pattern  in the file).

Any help would be much appreciated.