From 174c4eceabcc55a925f8cc03947834976171fb1d Mon Sep 17 00:00:00 2001
From: Belissimo_T <bela.tollking@gmail.com>
Date: Tue, 26 Sep 2023 23:54:24 +0200
Subject: [PATCH] improve additional info generation

---
 backend/lesson_info.py    | 26 ++++++++++++++------------
 backend/plan_processor.py |  2 +-
 backend/vplan_utils.py    | 10 +++++++++-
 3 files changed, 24 insertions(+), 14 deletions(-)

diff --git a/backend/lesson_info.py b/backend/lesson_info.py
index 177ef43..bc5a69b 100644
--- a/backend/lesson_info.py
+++ b/backend/lesson_info.py
@@ -8,7 +8,7 @@
 import typing
 
 from .vplan_utils import periods_to_block_label, parse_periods, _parse_form_pattern, ParsedForm, parsed_forms_to_str, \
-    MajorMinorParsedForm, AlphanumParsedForm, forms_to_str
+    MajorMinorParsedForm, AlphanumParsedForm, forms_to_str, _loose_parse_form_pattern
 from . import models
 
 
@@ -751,7 +751,7 @@ def process_additional_info_line(text: str, parsed_existing_forms: list[ParsedFo
     text = re.sub(r"(?<=\w)/ ", "/", text.strip())
 
     funcs = (
-        lambda s: add_fuzzy_teachers(s, teacher_abbreviation_by_surname, date),
+        lambda s: add_fuzzy_teacher_links(s, teacher_abbreviation_by_surname, date),
         lambda s: add_fuzzy_form_links(s, parsed_existing_forms, date)
     )
 
@@ -772,10 +772,13 @@ def add_fuzzy_form_links(text: str, parsed_existing_forms: list[ParsedForm], dat
                          ) -> list[LessonInfoTextSegment]:
     segments = []
 
-    prev = 0
-    for match in re.finditer(_parse_form_pattern, text):
-        segments.append(LessonInfoTextSegment(text[prev:match.start()]))
-        prev = match.end()
+    i = 0
+    while i < len(text):
+        match = re.search(_loose_parse_form_pattern, text[i:])
+
+        if match is None:
+            segments.append(LessonInfoTextSegment(text=text[i:]))
+            break
 
         parsed_forms = ParsedForm.from_form_match(match)
 
@@ -798,23 +801,22 @@ def add_fuzzy_form_links(text: str, parsed_existing_forms: list[ParsedForm], dat
                 matched_forms.append(form_match)
 
         if matched_forms:
+            segments.append(LessonInfoTextSegment(text[i:i+match.start()]))
             segments.append(
                 LessonInfoTextSegment(
                     parsed_forms_to_str(matched_forms),
                     link=LessonInfoTextSegmentLink("forms", [f.to_str() for f in matched_forms], date, None)
                 )
             )
+            i += match.end()
         else:
-            segments[-1].text += match.group()
-
-    segments.append(LessonInfoTextSegment(
-        text=text[prev:]
-    ))
+            segments.append(LessonInfoTextSegment(text[i]))
+            i += 1
 
     return segments
 
 
-def add_fuzzy_teachers(text: str, teacher_abbreviation_by_surname: dict[str, str], date: datetime.date):
+def add_fuzzy_teacher_links(text: str, teacher_abbreviation_by_surname: dict[str, str], date: datetime.date):
     abbreviations = set(teacher_abbreviation_by_surname.values())
 
     segments = []
diff --git a/backend/plan_processor.py b/backend/plan_processor.py
index 67fb5d9..57349fd 100644
--- a/backend/plan_processor.py
+++ b/backend/plan_processor.py
@@ -14,7 +14,7 @@
 
 
 class PlanProcessor:
-    VERSION = "79"
+    VERSION = "81"
 
     def __init__(self, cache: Cache, school_number: str, *, logger: logging.Logger):
         self._logger = logger
diff --git a/backend/vplan_utils.py b/backend/vplan_utils.py
index 555bc95..2131969 100644
--- a/backend/vplan_utils.py
+++ b/backend/vplan_utils.py
@@ -14,7 +14,15 @@
     r"(?P<sep>[/.] |[^A-Za-zÄÖÜäöüß0-9() \n]?)"
     r"(?P<minor>(?:\d{1,2}[A-Za-zÄÖÜäöüß]?|[A-Za-zÄÖÜäöüß]+?)(?:,(?:\d{1,2}[A-Za-zÄÖÜäöüß]?|[A-Za-zÄÖÜäöüß]+?))*)"
     r"|(?P<alpha>\d{1,2})"
-    r")(?![^\s,])"
+    r")(?![^\s,:])"
+)
+_loose_parse_form_pattern = re.compile(
+    r"(?<!\S)(?:"
+    r"(?P<major>\d{1,2}(?!\d)|[A-Za-zÄÖÜäöüß]+(?![A-Za-zÄÖÜäöüß]))"
+    r"(?P<sep>[/.]? |[^A-Za-zÄÖÜäöüß0-9() \n]?)"
+    r"(?P<minor>(?:\d{1,2}[A-Za-zÄÖÜäöüß]?|[A-Za-zÄÖÜäöüß]+?)(?:,(?:\d{1,2}[A-Za-zÄÖÜäöüß]?|[A-Za-zÄÖÜäöüß]+?))*)"
+    r"|(?P<alpha>\d{1,2})"
+    r")(?![^\s,:])"
 )