Script to parse COMSATS Lahore Timetable from PDF

timetableParser.py

import argparse
import calendar
import re

import pdfplumber


class TableCell:
    def __init__(self, rect, text):
        self.left = rect[0]
        self.top = rect[1]
        self.width = rect[2] - rect[0]
        self.height = rect[3] - rect[1]
        self.text = text


class TimeTableSlot:
    TEXT_RE = re.compile(r"(\d)\n(\d{1,2}:\d{1,2})\s+-\s+(\d{1,2}:\d{1,2})", re.MULTILINE)

    def __init__(self, cell):
        self.left = cell.left
        self.slot, self.startTime, self.endTime = re.match(self.TEXT_RE, cell.text.strip()).groups()


class TimeTableLecture:
    def __init__(self, cell, slots, weekday_name):
        self.weekday = self.weekday_from_name(weekday_name)
        data = cell.text.split("\n")

        self.has_double_slot = False
        if cell.width > 200:
            self.has_double_slot = True

        for index, slot in enumerate(slots):
            if int(slot.left) == int(cell.left):
                self.slots = [slot]

                if self.has_double_slot:
                    self.slots.append(slots[index + 1])

        if len(data) == 3:
            self.teacher, self.subject, self.room = data
        elif len(data) == 2:
            self.subject, self.room = data
        else:
            raise NotImplementedError("Multi line teacher/subject name not handled yet. Cell text is: \n%s" % cell.text)

    @staticmethod
    def weekday_from_name(weekday_name):
        weekdays = ["mo", "tu", "we", "th", "fr", "sa", "su"]
        if weekday_name.lower() in weekdays:
            return weekdays.index(weekday_name.lower())

    @property
    def weekday_name(self):
        return calendar.day_abbr[self.weekday]


class TimeTable:
    def __init__(self, page):
        self.page = page
        self.class_name = self._get_class_name()

    def parse_timetable(self):
        self._parse_timetable_cells()
        self._parse_timetable_slots()
        self._parse_timetable_slots()
        self._parse_lectures()

    def _text_in_rect(self, rect):
        return self.page.crop(rect).extract_text()

    def _get_class_name(self):
        return self._text_in_rect((70, 30, self.page.width, 60))

    def _parse_timetable_cells(self):
        table = self.page.find_tables()[0]
        self._rows = []

        for row in table.rows[1:]:
            cells = []
            for cell in row.cells:
                if not cell:
                    continue
                cells.append(TableCell(cell, self._text_in_rect(cell)))
            self._rows.append(cells)

    def _parse_timetable_slots(self):
        self.slots = []
        for cell in self._rows[0]:
            if cell.text:
                self.slots.append(TimeTableSlot(cell))

    def _parse_lectures(self):
        self.lectures = []
        last_day = ""

        for row in self._rows[1:]:
            for cell in row:
                if not cell.text:
                    continue
                elif len(cell.text.strip()) == 2:
                    last_day = cell.text.strip()
                else:
                    self.lectures.append(TimeTableLecture(cell, self.slots, last_day))


def print_time_table(timetable):
    for lec in timetable.lectures:
        print(f"{lec.weekday_name} - {lec.slots[0].slot} - {lec.subject} ({lec.room})")


if __name__ == "__main__":
    parser = argparse.ArgumentParser()
    parser.add_argument("-f", "--file", required=True, help="PDF file to parse")
    parser.add_argument("-c", "--classes", help='Classes name like (FA18-BCS-C)', nargs="+")
    parser.add_argument("-p", "--pages", help='Pages to parse', nargs="+", type=int)
    args = vars(parser.parse_args())

    timetables = []

    with pdfplumber.open(args["file"]) as pdf:
        if args["classes"]:
            for page in pdf.pages:
                timetable = TimeTable(page)

                for klass in args["classes"]:
                    if klass in timetable.class_name:
                        timetable.parse_timetable()
                        timetables.append(timetable)

        if args["pages"]:
            for page in args["pages"]:
                timetable = TimeTable(pdf.pages[page - 1])
                timetable.parse_timetable()
                timetables.append(timetable)

    for timetable in timetables:
        print("\n", timetable.class_name)
        print_time_table(timetable)

Running the script

# install dependencies
sudo pip3 install pdfplumber

# running parser with the class name
python3 timetableParser.py -f sample.pdf -c FA18-BCS-C

# running parser with the page number
python3 timetableParser.py -f sample.pdf -p 137

Features and Limitations

• Can handle complex timetable like:
  

output:

 FA16-CHE-B (Semester 7)
Mon - 4 - Process Control (N-1)
Mon - 5 - Petrochemical Engineering (G1 N-1)
Mon - 6 - Polymer Engineering (N-2)
Mon - 5 - Waste Management (G2 N-2)
Mon - 5 - Industrial Energy Systems (G3 D-117)
Tue - 3 - Process Modelling and Simulation (N-2)
Tue - 4 - Chemical Engineering Plant Design (D-114)
Tue - 5 - Lab-Process Modelling and Simulation (PC Lab-Pharmacy)
Tue - 6 - Process Instrumentation (N-1)
Wed - 3 - Lab-Process Control (G1 Che Lab-13)
Wed - 5 - Petrochemical Engineering (G1 N-1)
Wed - 6 - Process Modelling and Simulation (PC Lab-Pharmacy)
Wed - 1 - Lab-Process Control (G2 Che Lab-13)
Wed - 5 - Waste Management (G2 N-2)
Wed - 5 - Industrial Energy Systems (G3 D-117)
Thu - 3 - Polymer Engineering (N-12)
Thu - 5 - Lab-Process Modelling and Simulation (PC Lab-Pharmacy)
Thu - 6 - Chemical Engineering Plant Design (N-3)
Fri - 3 - Process Instrumentation (D-117)
Fri - 5 - Process Control (N-2)

• Can't handle multi-line text for teacher and subject name yet. Like: