Roadmap

Toolchain verson 1.0:

  MORPH
    ↓
xslt_filter ←──morphology_data_model    syntax_data_model
    │             (xml_schema)            /  (xml_schema)
    ↓                                    ↙
  morph.xml ────→ [lexer] ───────→ PARSER ────────────→ [parsed_trees] → analysedHebrewBible.xml
                                   ↑    ↑                      │             
                                grammar exception_list         │
                                   ↑     ↗                     │
                              analysed_trees ←─────────────────┘ 
                              [Tree Editor]

Files:

• morph43.wts - MORPH in legacy format
• morph.xsd - morphology data model
• morph.xsl - xslt filter
• morph43.xml - MORPH in xml format
• hebrewSyntax.xsd - syntax data model (wrapping and encapsulating the morph.xsd)
• hebrewSyntax.rules - grammar
• hebrewSyntax.except - exception list of special cases (possibly a database)
• {gn}SyntaxTrees.xml - analysed trees in xml format for use in Tree Editor

As books of the Hebrew Bible are analysed and corrected by hand, they will be archived as "canonical" and used in regression testings, i.e., when changes are made to the grammar, the output will be diff -ed against the "canonical" files.

Versioning

For the time being, given development and radical adjustments are likely in the next three months, the following file naming conventions should be used:

    filename + version (optional) + -YYYYMMDD + extention

    Example: morph43-20050301.xml
             hebrewSyntax-20050301.xsd