This learning resources about the generation of an abstract syntax tree from a given document. The learning resource will address the syntactical analysis MediaWiki documents, e.g. in Wikiversity.
History of Learning Resouce
Learning resource was generated in the context of passes for MarkDown in Markup languages like HTML. Abstract Syntax Trees (AST) are generated when parsers analyse the source code in a specific programming language. The tree structures represent nested environments in a programming language. As an introduction to Compilers Theory we will create an Abstract Syntax Tree (AST) for text document.
Decomposition of a Document
In a first step we decompose a document into tree nodes. A tree consists of connected tree nodes. A tree is a specific graph in which die nodes (vertices) may have attributes. In our context is
- (Document Node) the root node of the tree is the "document" node. The document node contains the information about the document which was parsed. If we parse the document in Wiki markdown, we would add references and time stamps and version id of the source document, so that JSON file contains the information about the origin of the document. This can be compared with a citation in scientific paper where a certain part of the text in an article refers to source of the cited content.
- (Content Element List) is basically a array of content elements. An array is required because the order of the content elements must be preserved.
- (Content Element) Can be anything from text block to media elements like video, audio or images.
Nested structures can be created by using a "Content Element List" as "Content Element".
- "Sections" a specific "Content Element Lists" that contain a "Title" and "Content Element List" of "paragrapha",
- "Paragraph"
- "Images"
- "Audio"
- "Video"
- "Quiz"
- "Mathematical Expression" - Type-Attribute: inline/block
- ...
Learning Tasks
- Define a JSON structure for Wikiversity Quizzes?
- How can you export a Wiki source for a quiz from a JSON?
- What are the benefits and drawbacks of a JSON format for a Wikiversity quiz instead of having the source code available directly as a text file?
- Analyse the concept of a JSON Editor and create Quiz editor that is able to generate
- a quiz for Wikiversity and
- a quiz for the OpenSource library SurveyJS[1]
See also
References
- ↑ SurveyJS GitHub Respository (2021) URL: https://github.com/surveyjs/survey-library (Accessed 2021/01/22)