Read course list from MyU

[superstealthysheep]

First, read what the list of courses is. Then, for each course, read

• When the course is (day of week, start/end times)
• Where the course is (Hall, room number) [optional: address]

[superstealthysheep]

First step: reading the classes in a given week. What I've got:

var url = "https://www.myu.umn.edu/psp/psprd/EMPLOYEE/CAMP/s/WEBLIB_IS_DS.ISCRIPT1.FieldFormula.IScript_DrawSection?group=UM_SSS&section=UM_SSS_ACAD_SCHEDULE&pslnk=1&cmd=smartnav&effdt=2023-02-05" // current week (may or may not work for people other than me?? need to test)

// create a (queryable!) DOM element from the url
var el = document.createElement('html'); 
await fetch(url).then(r=>r.text()).then(r => el.innerHTML = r); // idk man, was broken, then the "await" unbroke it

var classMeetings = el.querySelectorAll(".myu_calendar-class") // list of all the classes this week
for (let i = 0; i < classMeetings.length; i++) {

  // defined for convenience/readability(?)
  let meeting = classMeetings[i];
  let classDetails = meeting.querySelector(".myu_calendar-class-details").innerHTML.replace(/\n/g,"").split("<br>") // regex is to get rid of random newlines that are in there for some reason

  let semester = meeting.getAttribute("data-strm") // in format `xyyx', where `yy` is the year and `xx` = `13` is spring, `19` is fall
  let date = meeting.getAttribute("data-fulldate")
  let timeRange = classDetails[1]

  let courseName = meeting.querySelector(".myu_calendar-class-name-color-referencer").innerText
  let meetingType = classDetails[0]
  let room = classDetails[2].replace(/^ | $/g,"") // strip off the leading and trailing space with regex
  // let address = 

  console.log([semester, date, timeRange, courseName, meetingType, room].join(","))
}

Sample console output:

1233,20230206,9:05 - 9:55 AM,CSCI 2041,Lecture,Tate Hall B20
1233,20230206,10:10 - 11:00 AM,MATH 5286H,Lecture,Vincent Hall 1
1233,20230206,1:25 - 2:15 PM,CHEN 3101,Lecture,Amundson Hall B75
1233,20230206,3:35 - 4:25 PM,CSCI 2021,Lecture,Bruininks Hall 330
1233,20230207,9:05 - 9:55 AM,CHEN 3101,Discussion,Akerman Hall 317
1233,20230207,11:15 - 12:05 PM,CSCI 2041,Laboratory,Walter Library B28
1233,20230208,9:05 - 9:55 AM,CSCI 2041,Lecture,Tate Hall B20
1233,20230208,10:10 - 11:00 AM,MATH 5286H,Lecture,Vincent Hall 1
1233,20230208,1:25 - 2:15 PM,CHEN 3101,Lecture,Amundson Hall B75
1233,20230208,2:30 - 3:20 PM,CSCI 2021,Laboratory,Keller Hall 1-250
1233,20230208,3:35 - 4:25 PM,CSCI 2021,Lecture,Bruininks Hall 330
1233,20230209,9:05 - 9:55 AM,CHEN 3101,Discussion,Akerman Hall 317
1233,20230210,9:05 - 9:55 AM,CSCI 2041,Lecture,Tate Hall B20
1233,20230210,10:10 - 11:00 AM,MATH 5286H,Lecture,Vincent Hall 1
1233,20230210,1:25 - 2:15 PM,CHEN 3101,Lecture,Amundson Hall B75
1233,20230210,3:35 - 4:25 PM,CSCI 2021,Lecture,Bruininks Hall 330

Could someone else try this out to see if it works for people other than me?

[superstealthysheep]

Okay, now it returns an array of JSON object things. Also I'm figuring out how to wrap this into a function.

/**
 * Returns a list of json objects containing all the meetings in a given week
 * @param {string} date // a day during the week in question (Let's say the Sunday.), in format "yyyy-mm-dd", WITH DASHES
 * @returns {Array} meetingObjectArray
 */
const weekToJson = async (date) => {
  // appends the date info to our base url
  const baseURL = "https://www.myu.umn.edu/psp/psprd/EMPLOYEE/CAMP/s/WEBLIB_IS_DS.ISCRIPT1.FieldFormula.IScript_DrawSection?group=UM_SSS&section=UM_SSS_ACAD_SCHEDULE&pslnk=1&cmd=smartnav"
  let url = baseURL.concat("&effdt=", date)

  // create a (queryable!) DOM element from the url
  let HTMLText;
  var el = document.createElement('html');
  await fetch(url).then(r => r.text()).then(r => HTMLText = r);
  el.innerHTML = HTMLText;

  var meetingElements = el.querySelectorAll(".myu_calendar-class") // list of all the classes this week as HTML elems
  const meetingObjects = []; // list of json objects holding meeting data
  for (let i = 0; i < meetingElements.length; i++) {

    // defined for convenience/readability(?)
    let meetingEl = meetingElements[i];
    let classDetails = meetingEl.querySelector(".myu_calendar-class-details").innerHTML
                        .replace(/\n/g,"").split("<br>") // regex is to get rid of random newlines that are in there for some reason

    meetingObjects.push({
      "term"        : meetingEl.getAttribute("data-strm"), // in format `xyyx', where `yy` is the year and `xx` = `13` is spring, `19` is fall
      "courseNum"   : meetingEl.getAttribute("data-class-nbr"),
      "date"        : meetingEl.getAttribute("data-fulldate"),
      "meetingType" : classDetails[0],
      "timeRange"   : classDetails[1],
      "room"        : classDetails[2].replace(/^ | $/g,""), // strip off the leading and trailing space with regex
      "courseName"  : meetingEl.querySelector(".myu_calendar-class-name-color-referencer").innerText,
    });
    // console.log([semester, date, timeRange, courseName, meetingType, room].join(","))
  }
  console.log(meetingObjects);
  return meetingObjects;
}

[superstealthysheep]

I've now written 2 functions: one that reads the classes in a given week (as written above), and one that gives more detailed info about a class (e.g. general meeting times, start and end dates) when supplied with a term, course number, and campus. Right now they're living in the branch at https://github.com/samyok/gophergrades/tree/superstealthysheep/scraping.