The converter is based upon the SGMLSPM package which processes the output from NSGMLS when it processes a valid sgml document, (hence the effort to ensure that the input documents are valid).
SGMLSPM takes the input from an example which contains all the tags that will be encountered in the whole data set (exemplar.sgml) and generates a processing spec (ebmh_html_transform_spec.pl) on the basis of a processing spec it itself has been given (ebmh_skel.pl). Hence it is strongly recommended (read do it) to edit ebmh_skel.pl not the generated file (ebmh_html_transform.pl).
NGMLS needs to know where to find entities, this is contained in the filecatalog which should be in the converter directory. The DTD declaration is SYSTEM so the DTD and DCL should be in the same directory as the exemplar.
>nsgmls exemplar.sgml | perl d:\sgmlspm\sgmlspl.pl ebmh_skel.pl > ebmh_html_transform_spec.pl |
The data for an issue is held in a single directory.
The input files have an extension of html.
>perl html2sql.pl 9908 |
The data for an issue is held in a single directory.
The cleaned files have an extension of sgml.
>perl sgml2sql.pl 9908 |
Document Dated: Wed Dec 15 0:52:24 1999 by timp
Modify this document
Previous Version