The current version, DESpecs-XML.tex are specs for creating the body of XML documents according to the TEI P5 Guidelines. The TEI header as well as the facsimile part are left out since they can be attached in post-production, filled with the appropriate metadata. See the body2tei script in the postprocessing directory.
The data entry company uses the attached DTD file (in schema directory) for checking. For internal purposes, there is also a Relax NG file (converted from the DTD). A ODD file is not yet available.
The schema is maintained in the DTD file schema/mpiwg-erfassung.dtd. The only reason for using a DTD is that it is required by the data entry company. Internally, the RelaxNG versions are preferred.
The file mpiwg-erfassung.rnc is a direct conversion of the DTD file by running
trang mpiwg-erfassung.dtd mpiwg-erfassung.rnc
With this file, the texts returned by the data entry firm can be checked before performing the postprocessing steps (details below).
The file schema/mpiwg_tei_schema.rnc is using on the RNC file for data entry as a module. However, slight changes have to be made in order to successfully validate postprocessed files:
- The
divstructure has been introduced, that is the body element has onlydivchildren. - The
pbelement gets the attributefacsfor linking purposes. - The
startstatement has to be removed, since the file is called as a submodule.
If changes to the main DTD are necessary, the workflow documented in the sections above has to be re-applied.
Scripts exist for postprocessing
Converts text bodies typed after the conventions of the XML DE Specs to TEI documents. Specifically, it adds the header and facsimile part (using data from index.meta) and introduces a div structure in brute-force manner.
This converts a TEI document to LaTeX, a side-by-side view of facsimile and transcription.
The previous version, DESpecs.tex were specs to create an XML-like markup which was lateron converted to proper XML through a workflow. The use of these instructions is discouraged and only kept here for archival purposes. There is also a legacy version for Chinese texts, DESpecs_chinese.tex, which would need to be translated to a TEI version, if need be.