Parser für den Uni-Kalender

#!/usr/bin/python3

# Input:
# u.txt  -> list of calendar urls to process

# Output:
# m.ics  -> calendar containing mandatory events
# v.ics  -> calendar containing voluntary and duplicate events
# e.txt  -> calendar as a human readable string
# d.html -> diff of changes to calendar since last run

from datetime import datetime
import urllib.request
import difflib

def parse_ics(text):
    # Lines before and after the block of events
    pre, post = [], []
    # Events: (m)andatory, (v)oluntary, (d)uplicate; Calender as (h)uman readable string
    m, v, d, h = [], [], [], []
    # Current (b)uffer
    b = []
    # Event details
    start, end, summary, location = None, None, None, None

    # Booleans describing event type
    # NOTE: is_duplicate in this context refers to events such as bank holidays, which appear
    #       in all input calendars, but should only be added to the output calendars once
    is_mandatory = False
    is_duplicate = False

    for line in text.split('\r\n'):
        if line == "BEGIN:VEVENT":
            # Save lines before first event into pre
            # NOTE: The buffer is not updated yet, so the line "BEGIN:VEVENT" will not part of pre
            if not pre:
                pre = b
                b = []

        # Generate pseudo UID from start, end and summary
        if line.startswith("UID"):
            uid = '{:s}{:s}{:s}'.format(
                    start.strftime("%Y-%m-%d%H:%M:%S"),
                    end.strftime("%H:%M:%S"),
                    summary
                    ).encode("utf-8").hex()
            line = "UID:{:s}".format(uid)

        # Append current line to buffer
        b.append(line)

        # Parse event into variables
        if line.startswith("DTSTART"):
            date = line.split(':')[1]
            start = datetime.strptime(date, '%Y%m%dT%H%M%S')
        if line.startswith("DTEND"):
            date = line.split(':')[1]
            end = datetime.strptime(date, '%Y%m%dT%H%M%S')
        if line.startswith("SUMMARY"):
            summary = line.split(':')[1].replace('\\', '')
        if line.startswith("LOCATION"):
            location = line.split(':')[1].replace('\\', '')

        if line == "END:VEVENT":
            # Append buffer to list depending on type of event
            if is_duplicate:
                d.extend(b)
            elif is_mandatory:
                # Add a notification to mandatory events
                b.insert(-1, "BEGIN:VALARM")
                b.insert(-1, "TRIGGER:-PT30M")
                b.insert(-1, "ACTION:DISPLAY")
                b.insert(-1, "END:VALARM")
                m.extend(b)
            else:
                v.extend(b)

            # Generate human readable string and append to h
            if not is_duplicate:
                h.append('{:s} - {:s} {:s} ({:s})'.format(
                    start.strftime("%Y-%m-%d %H:%M:%S"),
                    end.strftime("%H:%M:%S"),
                    summary,
                    location
                ))

            # Clear event details; Reset booleans; Clear buffer
            start, end, summary, location = None, None, None, None
            is_mandatory = False
            is_duplicate = False
            b = []

        # Set is_mandatory and is_duplicate booleans depending on event summary
        if summary:
            # Mandatory event summaries end with an asterisk
            if summary.endswith("*"):
                is_mandatory = True
            # Bank holidays and similar events
            if summary == "Feiertag" or summary == "Studieninformationstag":
                is_duplicate = True

    # Save lines after last event into post
    post = b
    return pre, post, m, v, d, h

# Combined lists m, v, h from different calendars
ma, va, ha = [], [], []

# Read list of calendar urls
with open("u.txt", "r") as f:
    l = f.read().splitlines()

# Iterate over calendar urls
for url in l:
    # Download calendar
    response = urllib.request.urlopen(url)
    data = response.read()
    text = data.decode('utf-8')

    # Call parse_ics to process response
    pre, post, m, v, d, h = parse_ics(text)

    # Append events to combined lists
    # NOTE: While the lists m, v and h get appended to a combined list, the variables 
    #       pre, post and d will be overwritten on every iteration and only contain
    #       the content from the last calendar when exiting the loop
    ma.extend(m)
    va.extend(v)
    ha.extend(h)

# ma, va, ha contain the combined lists m, v, h from every calendar
# pre, post, d contain the content from the last processed calendar

# Sort lines for human readable string
ha.sort()

# Save mandatory events to m.ics
with open("m.ics", "w") as f:
    f.write('\n'.join(pre + ma + post))

# Save voluntary event and duplicates to v.ics
with open("v.ics", "w") as f:
    f.write('\n'.join(pre + va + d + post))

# e.txt contains the human readable string from the last run
# If it exists we can compare it to the current result to detect changes
try:
    # Load old e.txt
    with open("e.txt", "r") as f:
        l = f.read().splitlines()

    # Compare and print diff to stdout
    d = difflib.unified_diff(l, ha, "Old", datetime.today().strftime('%Y-%m-%d'), n=0)
    print('\n'.join(d), end="")

    # Compare and save diff to d.html
    d = difflib.HtmlDiff().make_file(l, ha, "Old", datetime.today().strftime('%Y-%m-%d'))
    with open("d.html", "w") as f:
        f.write(d)
except IOError:
    print('\n'.join(ha))

# Save human readable string to e.txt
with open("e.txt", "w") as f:
    f.write('\n'.join(ha))

--- Old

+++ 2023-03-05

@@ -257 +257,2 @@

-2023-05-15 09:00:00 - 10:30:00 V Neuro Hirnnerven (HS Kopf; INF 400)
+2023-05-15 09:00:00 - 09:30:00 V Neuro Einführung (HS Kopf; INF 400)
+2023-05-15 09:45:00 - 10:30:00 V Neuro Hirnnerven (HS Kopf; INF 400)