First code import

Makefile

@@ -0,0 +1,12 @@
+DEST=/var/gemini/fosdem/
+
+all: updateinstall
+
+updateinstall: update install
+
+update: 
+	wget --quiet --output-document schedule.xml https://fosdem.org/2021/schedule/xml
+	./schedule2gemtext.py
+
+install:
+	rsync -a -v -p --exclude=Makefile --exclude="*~" *.gmi ${DEST}

README.md

@@ -1,3 +1,24 @@
 # fosdem2gemini
 
-Convert the FOSDEM schedule to gemtext (text/gemini) format, to be served with Gemini.
+Convert the [FOSDEM](https://fosdem.org/) schedule to gemtext
+(`text/gemini`) format, to be served via
+[Gemini](https://en.wikipedia.org/wiki/Gemini_(protocol)).
+
+## Installation
+
+Download the program. Be sure you have the
+[lxml](https://lxml.de/library. Edit sample-index.gmi to suit your
+needs and rename it index.gmi. Then run the program periodically, for
+instance through cron:
+
+```
+# FOSDEM
+26 1,5,9,13,17,21 * * *  (cd /path/to/FOSDEM; make all)
+```
+
+## Technical details 
+
+The FOSDEM schedule is publically available (from the schedule
+management system Pentabarf) as a XML file (whch we can retrieve with,
+for instance, `wget https://fosdem.org/2021/schedule/xml`). The use of
+a structured format allows us to convert it easily.

sample-index.gmi

@@ -0,0 +1,14 @@
+# FOSDEM schedule
+
+A non-official Gemini version of FOSDEM schedule for the 2021 "COVID edition".
+
+=> allevents.gmi All events, sorted by date and time
+=> allrooms.gmi All "rooms", with their events
+
+=> https://fosdem.org/ FOSDEM official page
+=> https://fosdem.org/2021/schedule/ FOSDEM official schedule
+
+The schedule here is automatically produced from the FOSDEM "Pentabarf" files.
+
+This service is maintained by Stéphane Bortzmeyer <stephane+fosdem@bortzmeyer.org>.
+

schedule2gemtext.py

@@ -0,0 +1,235 @@
+#!/usr/bin/python3
+
+# https://fosdem.org/2021/schedule/xml
+FILE = "schedule.xml"
+INDEX = "index.gmi"
+ALLEVENTS = "allevents.gmi"
+ALLROOMS = "allrooms.gmi"
+
+import xml.etree.ElementTree as ET
+import sys
+
+# https://lxml.de/
+from lxml.html.clean import Cleaner
+from lxml.html.soupparser import fromstring
+
+def html2gemini(html):
+    result = ""
+    for child in html:
+        if child.tag == "p":
+            if child.text != None:
+                result += child.text
+        elif child.tag == "div":
+            result += html2gemini(child)
+        elif child.tag == "hr":
+            result += "\n"
+        elif child.tag == "h1":
+            result += "# %s" % child.text
+        elif child.tag == "h2":
+            result += "## %s" % child.text
+        elif child.tag == "h3":
+            result += "### %s" % child.text
+        elif child.tag == "ul" or child.tag == "ol":
+            result += html2gemini(child)
+        elif child.tag == "li":
+            if child.text != None:
+                result += """
+* %s
+""" % child.text
+        elif child.tag == "code":
+            result += child.text
+        elif child.tag == "pre":
+            result += """
+```
+%s
+```
+""" % html2gemini(child)
+        else:
+            print("Unknown tag %s" % child.tag, file=sys.stderr)
+        if child.tag == "div" or child.tag == "p" or child.tag == "ul" or child.tag == "ol":
+            result += """
+            
+"""
+    return result                     
+
+def cleanroom(name):
+    components = name.split(".")
+    first_word = components[1][0].upper() + components[1][1:]
+    if len(components) == 2:
+        return first_word 
+    elif len(components) == 3:
+        return first_word + " " + components[2]
+    else:
+        return first_word + " " + " ".join(components[2:])
+
+def typeroom(name):
+    components = name.split(".")
+    if components[0] == "S":
+        return "Stand"
+    elif components[0] in ("D", "F", "M", "B"):
+        return "Room"
+    elif components[0] == "K":
+        return "Keynote"
+    elif components[0] == "I": # Infodesk
+        return ""
+    elif components[0] == "L": # Lightning talks
+        return ""
+    else:
+        return "Unknown"
+
+def matrixroom(name):
+    ptype = ""
+    if typeroom(name) == "Stand":
+        ptype = "-stand"
+    return "#%s%s:fosdem.org" % (cleanroom(name).lower(), ptype)
+
+def shortroom(name):
+    components = name.split(".")
+    return "".join(components).lower()
+
+def events_of(room):
+    result = []
+    for event in events:
+        if events[event]["room"] == room:
+            result.append(event)
+    return result
+
+tree = ET.parse(FILE)
+root = tree.getroot()
+days = {}
+rooms = {}
+events = {}
+# https://lxml.de/api/lxml.html.clean.Cleaner-class.html
+html_cleaner = Cleaner(scripts=True, javascript=True, embedded=True,
+                       meta=True, page_structure=True, links=True,
+                       remove_unknown_tags=True,
+                       frames=True,annoying_tags=True,add_nofollow=True,
+                       style=False)
+for day in root.findall("day"):
+    days[day.attrib["index"]] = {"date": day.attrib["date"]}
+    for room in day.findall("room"):
+        rooms[room.attrib["name"]] = {}
+        for event in room.findall("event"):
+            speakers = ""
+            first = True
+            for speaker in event.findall("persons/person"):
+                if not first:
+                    speakers += " and "
+                speakers += speaker.text
+                first = False
+            id = event.attrib["id"]
+            if event.find("abstract").text is None:
+                abstract = None
+            else:
+                abstract = html_cleaner.clean_html(event.find("abstract").text)
+            events[id] = {"room": room.attrib["name"],
+                          "day": day.attrib["index"],
+                          "start": event.find("start").text,
+                          "duration": event.find("duration").text,
+                          "title": event.find("title").text,
+                          "subtitle": event.find("subtitle").text,
+                          "type": event.find("type").text,
+                          "abstract": abstract,
+                          "speakers": speakers
+            }
+            eventfile = open("event-%s.gmi" % event.attrib["id"], "w")
+            if events[id]["subtitle"] is None:
+                subtitle = ""
+            else:
+                subtitle = """
+%s
+""" % events[id]["subtitle"]
+            if events[id]["abstract"] is None:
+                abstract = ""
+            else:
+                abstract_root = fromstring(events[id]["abstract"])
+                abstract = """
+%s
+""" % html2gemini(abstract_root)
+            if events[id]["speakers"] == "":
+                speakers = ""
+            else:
+                speakers = events[id]["speakers"]
+            print("""# FOSDEM event "%s"
+
+%s
+Type %s
+%s
+Starts on day %s (%s) at %s (Brussels time, UTC+1) in room %s (duration %s)
+Matrix room %s
+%s
+
+=> . FOSDEM schedule page
+""" % (events[id]["title"], speakers, events[id]["type"],
+           subtitle, events[id]["day"], days[events[id]["day"]]["date"],
+           events[id]["start"], cleanroom(events[id]["room"]), events[id]["duration"],
+           matrixroom(events[id]["room"]),
+           abstract),
+                      file=eventfile)
+            eventfile.close()
+sortedevents = sorted(events.keys(),
+                      key=lambda x: "%s %s" % (events[x]["day"], events[x]["start"]),
+                      reverse=False)
+allevents = open(ALLEVENTS, "w")
+print("""# All events at FOSDEM
+""", file=allevents)
+day = None
+for event in sortedevents:
+    if typeroom(events[event]["room"]) not in ["Room", "Keynote"]:    
+        continue
+    if events[event]["day"] != day:
+        day = events[event]["day"]
+        print("""## Day %s (%s)
+
+""" % (events[event]["day"], days[events[event]["day"]]["date"]), file=allevents)
+    if events[event]["subtitle"] is None:
+        subtitle = ""
+    else:
+        subtitle = """%s
+""" % events[event]["subtitle"]
+    if events[event]["speakers"] == "":
+        speakers = ""
+    else:
+        speakers = "By %s" % events[event]["speakers"]
+    print("""### %s
+%s
+%s
+Starts at %s (Brussels time, UTC+1) in room %s
+
+=> event-%s.gmi Details on the event
+
+""" % (events[event]["title"], speakers,
+       subtitle,
+       events[event]["start"], cleanroom(events[event]["room"]), event),
+          file=allevents)
+print("""
+
+=> . FOSDEM schedule page
+""", file=allevents)
+allevents.close()
+sortedrooms = sorted(rooms.keys(),
+                      key=lambda x: cleanroom(x),
+                      reverse=False)
+allrooms = open(ALLROOMS, "w")
+print("""# All rooms at FOSDEM
+""", file=allrooms)
+day = None
+for room in sortedrooms:
+    print("""## %s "%s"
+
+""" % (typeroom(room), cleanroom(room)),
+          file=allrooms)
+    room_events = events_of(room)
+    for event in room_events:
+        print("""=> event-%s.gmi %s""" % (event, events[event]["title"]), file=allrooms)
+    print("""
+
+Matrix room : %s
+
+=> https://fosdem.org/2021/schedule/room/%s/ Description of this room on FOSDEM site
+""" % (matrixroom(room), shortroom(room)), file=allrooms)
+print("""
+=> . FOSDEM schedule page
+""", file=allrooms)
+allrooms.close()
+