architecture.txt

AMI2-SVG2XML Architecture
=========================

SVG2XML can currently:
  * process a list of PDFs with DocumentAnalyzer
  * split each PDF (managed by PDFAnalyzer) into Pages
  * split each Page (managed by PageAnalyzer) into Chunks (components of a page), 
  * analyze SVG primitives (<text> , <path>, <image>) in a chunk 
     via TextAnalyzer, PathAnalyzer, ImageAnalyzer.
     These use AbstractContainers to hold the content as it is analyzed.
  * more specialized Containers are needed for complex content
    such as styled and subscripted text (ScriptContainer).
  * some chunks become more specialized (MixedAnalyzer = Text+path+image) and acquire
    structure and purpose FigureAnalyzer, TableAnalyzer 
  * packages paths, text, figure, table are used for specialist analysis
  
 Each Analyzer communicates with its container:
  * PDFAnalyzer contains DocumentListAnalyzer
  * PageAnalyzer contains PDFAnalyzer 
  * PageChunkAnalyzer contains PageAnalyzer.
  This each pageChunk "knows" which PDFAnalyzer it is from and can access resources
    such as directories
    
 PDFAnalyzer uses
  * PDFAnalyzerIO for its directories, etc
  * PDFAnalyzerOptions for control of operations
  
 PageAnalyzer uses
  * PageIO for directories and options (this is primarily for analysing single pages)