Coursebook parser building/room bug #48

[SoggyRhino]

This is to fix the bug where meeting locations were wrong. I tested it the best I could and it I haven't found any problems. The logic mostly the same as the original just less strict with the format of the meeting fields.

Get Meetings

• Separate meetings based on html divs rather than regex
  • This helps the parser not overrun into the next meeting to find locations
• Meeting Fields are parsed using Regex Find All String
  • Handles Missing or partial information
• Location is parsed from the <a> at the bottom of the div
  • Uri is just the href of the <a>
    • If the link doesn't point to a valid building leave location blank
  • Building and number is just the text of the <a> split by space
• Potential issues:
  • Theoretically we could have a meeting that is all blank since none of the fields are required
    • Still technically correct since the meeting (on coursebook) would just be filled with placeholder data
  • Should meeting days be null or []
    • The json has fields that are null but I thought [] made more sense

Get Academic session (This wasn't part of the bug)

• The container is essential a map where the key is a <b> and the value next text
• Switch on the key and set the field of the session based on the value

Parse Time or Panic Helper Method

• we parse time the same way 4 times in sectionParser so it made sense to make a helper

Everything Else

• There are some minor changes to accommodate the passing of *goquery.Selection instead of strings
  • Basically just adding a .text to a few methods

I'm not sure why the go.mod changed but that is what was created when I ran the build.bat.

[jpahm]

This looks great, thank you! I think you handled the potential issues well.