5 Document-level Functions

A TEI document is a TEI element struct that represents the root TEI element of a document. TEI document values implement the instance info interface for bibliographic information.

5.1 Reading & Writing TEI Documents

High-level clients should use valid-xml-file? or directory-validate-xml to validate file before calling file->tei-document due to the current limitations on the validation performed by any-tei-xexpr/c.

This function uses read-xexpr/standardized to parse the raw XML into x-expressions consistently and without information loss.

Currently, file->tei-document should usually be used instead of read-tei-document, as it cooperates more easily with the validation needs documented under file->tei-document and any-tei-xexpr/c.

This function uses read-xexpr/standardized to parse the raw XML into x-expressions consistently and without information loss.

Use write-tei-document rather than other methods for writing XML: write-tei-document uses write-xexpr/standardized to generate consistent output and includes an appropriate prelude.

The resulting string is not the XML representation of doc: it is formated for uses that expect unstructured plain text.

When include-header? is non-false (the default), the resulting string will begin with a header which includes, for example, the title and other information about the corresponding instance. When include-header? is #false, only the content will be included, which is sometimes preferable if the plain text form is intended for further processing by computer.

5.2 Paragraph Inference

A value of 'todo means that paragraph-guessing has not been performed and should be done as soon as possible. A value of 'skip means that paragraph-guessing has been intentionally postponed, perhaps because the current strategies have not proven effective for doc.

The values 'line-breaks, 'blank-lines, and 'done all mean that paragraph-guessing has been completed successfully: 'line-breaks and 'blank-lines indicate the strategy by which paragraphs were infered, whereas 'done is a legacy value indicating that paragraph-guessing was performed before this library began recording which strategy was used.