From a7638eeac0006f2baafd1ccadea3fee9fad42038 Mon Sep 17 00:00:00 2001
From: "Vriezen, E.C. (Emma)" <e.vriezen@student.ru.nl>
Date: Mon, 16 Dec 2019 21:38:19 +0100
Subject: [PATCH] Fixed all of the wrong classifications, by making sure that
shadow words are never matched (as mistakes) to source words that appear
after the next correctly matched words.
---
umbra/mistake_finder.py | 11 ++++++++---
umbra/words.py | 27 ++++++++++++++++++++++++++-
2 files changed, 34 insertions(+), 4 deletions(-)
diff --git a/umbra/mistake_finder.py b/umbra/mistake_finder.py
index 31b9e1ee..a1c59e30 100644
--- a/umbra/mistake_finder.py
+++ b/umbra/mistake_finder.py
@@ -30,8 +30,6 @@ class MistakeFinder:
# Loop over the shadow:
for index, word in enumerate(self._shadow):
if not word.correct and word.mistake is None:
- if word.word == "geen":
- print("boo")
self._determine_mistake(index, word)
# Loop over the source:
for word in self._source:
@@ -131,8 +129,15 @@ class MistakeFinder:
src_index = 0
else:
src_index = self._source.index(self._shadow[last_shd_index].source)
+ next_shd_index = self._shadow.find_next_matched_shadow(index)
+ if next_shd_index < 0:
+ src_end_index = len(self._source) - 1
+ else:
+ src_end_index = self._source.index(
+ self._shadow[next_shd_index].source)
while self._source[src_index].get_difference(shd_word) > 0 and\
- not form_mistake and src_index < len(self._source) - 1:
+ not form_mistake and src_index < len(self._source) and\
+ src_index <= src_end_index:
src_word = self._source[src_index]
form_mistake = self.form_related(src_word, shd_word)
if form_mistake:
diff --git a/umbra/words.py b/umbra/words.py
index 6629ddb4..345d7229 100644
--- a/umbra/words.py
+++ b/umbra/words.py
@@ -248,6 +248,27 @@ class Sentence(list):
return index
return -1
+ def find_next_matched_shadow(self, index):
+ """Find the index of the next shadowed word.
+
+ Args:
+ index: the index after which should be sought for a shadow word
+ that is matched with a source word.
+
+ Returns:
+ last_index: the index of the next matched shadow. This is -1
+ if there is no matches shadow after or on the specified index, or
+ if the index was out of bounds.
+ """
+ if 0 <= index < len(self):
+ while index < len(self):
+ if self[index].has_source():
+ return index
+ index += 1
+ else:
+ index = -1
+ return index
+
def find_previous_anchor(self, index):
"""Find the index of the last anchor before specified index.
@@ -264,7 +285,9 @@ class Sentence(list):
if self[index].is_anchor():
return index
index -= 1
- return -1
+ else:
+ index = -1
+ return index
def find_last_matched_shadow(self, index):
"""Find the index of the last shadowed word.
@@ -283,6 +306,8 @@ class Sentence(list):
if self[index].has_source():
return index
index -= 1
+ else:
+ index = -1
return index
def __str__(self):
--
GitLab