Internal

Basic Structure

Internally, Plate have 3 steps :

• Get API Document
• Convert markdown to HTML
• Highlight code according to programming languages

1. Get API Document

• Read markdown from API Document based on markdown format using API_DOC_PATH and API_DOC_INDEX_PATH in config.json.
• Sort by index.json of API_DOC_INDEX_PATH.

2. Convert markdown to HTML

• Convert markdown to HTML using markdown python module
• Use markdown extensions : - fence_codes : markdown code block to html pre tag - tables : markdown table syntax to html table tag

def conv_md_to_html(md_text):
    import markdown
    return markdown.markdown(md_text, extensions=["fenced_code", "tables"])

3. Highlight code according to programming languages

• Use pygemnts for highlighting codes
• Support various programming languages and markup.

from pygments import highlight
from pygments.lexers import PythonLexer
from pygments.lexers import JavaLexer
from pygments.formatters import HtmlFormatter

highlighted = highlight(code, PythonLexer(), HtmlFormatter())
highlighted = highlight(code, JavaLexer(), HtmlFormatter())

Realtime monitoring API Document

For monitoring the modification of API Documents, use wachdog . After the server start, the watchdog start and monitor all documents in API_DOC_PATH of config.json . When the server stop, also the watchdog stop. In this process, use method watchdocs.watch_api_doc.start_watch and watchdocs.watch_api_doc.stop_watch .

If raise any modification fo files, run watchdocs.document_trace_handler.on_modified method. In this method, enqueue a event about the modification of any file to document_trace_queue . It is a instance of DocumentTraceQueue singleton class.

And then, receive new request from user, Plate check document_trace_queue whether a event exist or not. If any event in queue, Plate load all API Documents and convert to html.