Merge pull request #300 from aslafy-z/patch-1

fix: upgrading nltk dependency to fix wordnet/punkt download
srbhr · Jan 8, 2025 · 367b9dd · 367b9dd
2 parents baa9e02 + 7e60d1a
commit 367b9dd
Show file tree

Hide file tree

Showing 4 changed files with 12 additions and 8 deletions.
diff --git a/requirements.txt b/requirements.txt
@@ -42,7 +42,7 @@ mdurl==0.1.2
 more-itertools==9.1.0
 murmurhash==1.0.9
 networkx==3.1
-nltk==3.9
+nltk==3.9.1
 numpy==1.25.1
 packaging==23.1
 pandas==2.0.3
@@ -113,4 +113,4 @@ reportlab==3.6.13
 easygui==0.98.3
 xhtml2pdf==0.2.11
 fastembed~=0.2.2
-qdrant-client==1.9.0
+qdrant-client==1.9.0
diff --git a/streamlit_app.py b/streamlit_app.py
@@ -28,9 +28,9 @@
 config_path = os.path.join(cwd, "scripts", "similarity")
 
 try:
-    nltk.data.find("tokenizers/punkt")
+    nltk.data.find("tokenizers/punkt_tab")
 except LookupError:
-    nltk.download("punkt")
+    nltk.download("punkt_tab")
 
 parameters.SHOW_LABEL_SEPARATOR = False
 parameters.BORDER_RADIUS = 3

diff --git a/streamlit_interactive.py b/streamlit_interactive.py
@@ -31,11 +31,11 @@
 cwd = find_path("Resume-Matcher")
 config_path = os.path.join(cwd, "scripts", "similarity")
 
-# Check if NLTK punkt data is available, if not, download it
+# Check if NLTK punkt_tab data is available, if not, download it
 try:
-    nltk.data.find("tokenizers/punkt")
+    nltk.data.find("tokenizers/punkt_tab")
 except LookupError:
-    nltk.download("punkt")
+    nltk.download("punkt_tab")
 
 # Set some visualization parameters using the annotated_text library
 parameters.SHOW_LABEL_SEPARATOR = False

diff --git a/streamlit_second.py b/streamlit_second.py
@@ -20,7 +20,11 @@
     initial_sidebar_state="auto",
 )
 
-nltk.download("punkt")
+# Check if NLTK punkt_tab data is available, if not, download it
+try:
+    nltk.data.find("tokenizers/punkt_tab")
+except LookupError:
+    nltk.download("punkt_tab")
 
 parameters.SHOW_LABEL_SEPARATOR = False
 parameters.BORDER_RADIUS = 3