worlddevelopment / exercise_database Goto Github PK

Definitions

Indexless := no index numbering scheme, i.e. if a number occurs then it is either content or denoting a hierarchy in a markup and not a series. => numbers are explicit (no regular expression) => can only have an implicit ordering.
indexed := with index numbering scheme (i.e. explicite order)

Generic := filter by an expression (regex|wildcard|...)
Specific := explicit := filter by explicit content (repeating phrase)
Raw content := markup content
Content := plain text.content, i.e. the visual content like information text, media, ...

Content part declarations

• [Content phrase filter] (matching all specific content parts within all hierarchy levels mixed)

  • generic|regex|wildcard

    • can match index (have an explicit series order)

      • all|mixed

        building[ ]*([\\d][ .->_]*)+:
        Solution[ ]*([\\d][ .->_]*)+:
        Exercise[ ]*([\\d][ .->_]*)+:
        Teacher[ ]*([\\d][ .->_]*)+:
        Text[ ]*([\\d][ .->_]*)+:
        

      • numbers+special chars only (have an explicit series order, hierarchy e.g. 1.1, 1.2, 2.1, ...)

        ([\\d].)+
        ([\\d])+
        

        Note: Number based index may need filtering of false positives due to numbers occuring in the content parts, too.)

    • guaranteed indexless

      Exercise[ ]*: // generic|regex|wildcard
      ...
      

  • specific|explicit (guaranteed indexless)

    • all|mixed

      Exercise:
      Teacher:
      Teacher (Physics):
      Teacher Physics:
      Structures :
      Structures:
      Text:
      Mission to achieve: // phrase is a sentence|clause
      

    • numbers+special chars only (e.g. 1.1:, 1.1:, 1.1:, ...)

      3:
      1.2:
      800:
      

• [Raw content | Markup phrase filter]

  • specific|explicit, match only indexless (no order; numbers denote hierarchy depth; Matches only within one hierarchy level)

    #
    ##
    ---
    ===
    h1
    h2
    h:p level='1'
    header level="1"
    header level="2"
    ...
    

  • generic, can match index (Matching all series within all hierachy levels in one pass! [1])
    Note: This is the default case for XML base file formats. It requires keeping track of hierarchy depth counting in code because a node has no number attached! It can however have a style attached denoting depth.

    #+
    header[\\d]*
    h[\\d]*
    section
    

• [Mixed: Markup & Content phrase filters]

  • specific, mostly indexless (matching all within one hierarchy level with a content filter:)

    #Breaking:
    # Tex
    # general information
    ## specific information
    

  • generic, match only indexless (Matching all series within all hierachy levels in one pass! [1])
    Note: For all XML base file formats this merged pattern is easier to achieve via postprocessing the respective content part's head after employing the generic, indexless filter.

    #+(^[<][\\w][>])*[Tt]ext
    #+[ ]?[Gg]eneral information
    

[1] Only of limited use as higher level elements have no content part if following strict sectioning. what remains is only the declaration unless there is summary|description content between e.g. 1. and its subsection 1.1 .

Purpose

They are essential for the worlddevelopment civilization editor, open bookkeeper bot, ...

It could be named in a generic or functional way:
e.g.

• core
• section_search
• ...

If it is renamed to section_search, then specific Java files like Aufgaben_DB.java must be moved into another package if it is not generic or does not belong to section_search, else it could be renamed to Core.java.

If the package is renamed to core, then the library functionality could be isolated and called section_search.
Put the project onto the build path.
The source import path then were www.section_search.core.*.

A public domain prefix is employed as at.ciry does lose part of its meaning and international.worlddevelopment is long nor does a claim for worlddevelopment.int|ernational exist.

A part contains Content as well as Raw content (markup), thus Content part is too specific. Also simply part is shorter.

Also requires a database update.
Also requires to rename Aufgaben_DB to PartDB.

The design needs to be extended to support non-text-based markup content part declaration detection.
This is easily possible by extending the Declaration ID and employing node traversal for XML markup format.

TODO JSON is either treated as text-based or it is parsed into a tree via Virtual Time Machine functionality.