Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
@@ -50,24 +50,57 @@ def extract_text_from_pdf(uploaded_file):
|
|
50 |
st.error(f"PDF extraction error: {str(e)}")
|
51 |
return ""
|
52 |
|
53 |
-
def
|
54 |
-
|
55 |
-
|
|
|
|
|
|
|
56 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
57 |
differ = difflib.Differ()
|
58 |
diff = list(differ.compare(text1.split(), text2.split()))
|
59 |
-
|
60 |
-
|
61 |
-
|
|
|
|
|
62 |
if word.startswith("- "):
|
63 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
64 |
elif word.startswith("+ "):
|
65 |
-
|
66 |
-
|
67 |
-
|
68 |
-
|
69 |
-
|
70 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
71 |
|
72 |
def calculate_similarity(text1, text2):
|
73 |
if not text1.strip() or not text2.strip():
|
@@ -144,14 +177,14 @@ def main():
|
|
144 |
# Update document displays
|
145 |
if uploaded_file1:
|
146 |
doc1_display.text_area("Document 1 Content",
|
147 |
-
|
148 |
-
|
149 |
-
|
150 |
if uploaded_file2:
|
151 |
doc2_display.text_area("Document 2 Content",
|
152 |
-
|
153 |
-
|
154 |
-
|
155 |
|
156 |
if not (uploaded_file1 and uploaded_file2):
|
157 |
st.warning("Please upload both documents to proceed")
|
@@ -159,6 +192,11 @@ def main():
|
|
159 |
|
160 |
# ===== DOCUMENT COMPARISON SECTION =====
|
161 |
st.header("2. Document Comparison")
|
|
|
|
|
|
|
|
|
|
|
162 |
with st.expander("Show Document Differences", expanded=True):
|
163 |
if st.button("Compare Documents"):
|
164 |
with st.spinner("Analyzing documents..."):
|
@@ -167,27 +205,45 @@ def main():
|
|
167 |
return
|
168 |
|
169 |
similarity_score = calculate_similarity(contract_text1, contract_text2)
|
170 |
-
highlighted_diff = highlight_differences(contract_text1, contract_text2)
|
171 |
|
172 |
-
|
173 |
-
|
174 |
-
|
175 |
-
|
176 |
-
|
177 |
-
|
178 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
179 |
if st.session_state.comparison_results:
|
180 |
st.metric("Document Similarity Score",
|
181 |
-
|
182 |
|
183 |
if st.session_state.comparison_results['similarity_score'] < 50:
|
184 |
st.warning("Significant differences detected")
|
185 |
|
186 |
st.markdown("**Visual Difference Highlighting:**")
|
187 |
-
st.
|
188 |
-
|
189 |
-
|
190 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
191 |
|
192 |
# ===== QUESTION ANALYSIS SECTION =====
|
193 |
st.header("3. Clause Analysis")
|
@@ -235,8 +291,8 @@ def main():
|
|
235 |
except Exception as e:
|
236 |
st.session_state.analysis_results = st.session_state.analysis_results or {}
|
237 |
st.session_state.analysis_results['doc2'] = f"Analysis failed: {str(e)}"
|
238 |
-
|
239 |
-
# Display analysis results
|
240 |
if st.session_state.analysis_results:
|
241 |
col1, col2 = st.columns(2)
|
242 |
with col1:
|
@@ -248,4 +304,4 @@ def main():
|
|
248 |
st.success(st.session_state.analysis_results.get('doc2', 'No analysis performed yet'))
|
249 |
|
250 |
if __name__ == "__main__":
|
251 |
-
main()
|
|
|
50 |
st.error(f"PDF extraction error: {str(e)}")
|
51 |
return ""
|
52 |
|
53 |
+
def highlight_differences_side_by_side(text1, text2):
|
54 |
+
differ = difflib.Differ()
|
55 |
+
diff = list(differ.compare(text1.splitlines(keepends=True), text2.splitlines(keepends=True)))
|
56 |
+
|
57 |
+
highlighted_text1 = ""
|
58 |
+
highlighted_text2 = ""
|
59 |
|
60 |
+
for line in diff:
|
61 |
+
if line.startswith("- "):
|
62 |
+
highlighted_text1 += f'<span style="background-color:#ffcccc">{line[2:]}</span>'
|
63 |
+
elif line.startswith("+ "):
|
64 |
+
highlighted_text2 += f'<span style="background-color:#ccffcc">{line[2:]}</span>'
|
65 |
+
elif line.startswith(" "):
|
66 |
+
highlighted_text1 += line[2:]
|
67 |
+
highlighted_text2 += line[2:]
|
68 |
+
|
69 |
+
return highlighted_text1, highlighted_text2
|
70 |
+
|
71 |
+
def highlight_differences_words(text1, text2):
|
72 |
differ = difflib.Differ()
|
73 |
diff = list(differ.compare(text1.split(), text2.split()))
|
74 |
+
|
75 |
+
highlighted_text1 = ""
|
76 |
+
highlighted_text2 = ""
|
77 |
+
|
78 |
+
for i, word in enumerate(diff):
|
79 |
if word.startswith("- "):
|
80 |
+
removed_word = word[2:]
|
81 |
+
highlighted_text1 += f'<span style="background-color:#ffcccc; display: inline-block;">{removed_word}</span>'
|
82 |
+
# Check for corresponding addition to highlight as changed
|
83 |
+
if i + 1 < len(diff) and diff[i + 1].startswith("+ "):
|
84 |
+
added_word = diff[i + 1][2:]
|
85 |
+
highlighted_text2 += f'<span style="background-color:#ffffcc; display: inline-block;">{added_word}</span>'
|
86 |
+
diff[i + 1] = ' '
|
87 |
+
else:
|
88 |
+
highlighted_text2 += " "
|
89 |
elif word.startswith("+ "):
|
90 |
+
added_word = word[2:]
|
91 |
+
highlighted_text2 += f'<span style="background-color:#ccffcc; display: inline-block;">{added_word}</span>'
|
92 |
+
# Check for corresponding removal
|
93 |
+
if i - 1 >= 0 and diff[i - 1].startswith("- "):
|
94 |
+
highlighted_text1 += f'<span style="background-color:#ffffcc; display: inline-block;">{diff[i-1][2:]}</span>'
|
95 |
+
diff[i-1] = ' '
|
96 |
+
else:
|
97 |
+
highlighted_text1 += " "
|
98 |
+
|
99 |
+
elif word.startswith(" "):
|
100 |
+
highlighted_text1 += word[2:] + " "
|
101 |
+
highlighted_text2 += word[2:] + " "
|
102 |
+
|
103 |
+
return highlighted_text1, highlighted_text2
|
104 |
|
105 |
def calculate_similarity(text1, text2):
|
106 |
if not text1.strip() or not text2.strip():
|
|
|
177 |
# Update document displays
|
178 |
if uploaded_file1:
|
179 |
doc1_display.text_area("Document 1 Content",
|
180 |
+
value=contract_text1,
|
181 |
+
height=200,
|
182 |
+
key="area1")
|
183 |
if uploaded_file2:
|
184 |
doc2_display.text_area("Document 2 Content",
|
185 |
+
value=contract_text2,
|
186 |
+
height=200,
|
187 |
+
key="area2")
|
188 |
|
189 |
if not (uploaded_file1 and uploaded_file2):
|
190 |
st.warning("Please upload both documents to proceed")
|
|
|
192 |
|
193 |
# ===== DOCUMENT COMPARISON SECTION =====
|
194 |
st.header("2. Document Comparison")
|
195 |
+
comparison_format = st.radio(
|
196 |
+
"Comparison Output Format",
|
197 |
+
["Side by Side", "Combined View"],
|
198 |
+
index=0
|
199 |
+
)
|
200 |
with st.expander("Show Document Differences", expanded=True):
|
201 |
if st.button("Compare Documents"):
|
202 |
with st.spinner("Analyzing documents..."):
|
|
|
205 |
return
|
206 |
|
207 |
similarity_score = calculate_similarity(contract_text1, contract_text2)
|
|
|
208 |
|
209 |
+
if comparison_format == "Side by Side":
|
210 |
+
highlighted_diff1, highlighted_diff2 = highlight_differences_side_by_side(contract_text1, contract_text2)
|
211 |
+
st.session_state.comparison_results = {
|
212 |
+
'similarity_score': similarity_score,
|
213 |
+
'highlighted_diff1': highlighted_diff1,
|
214 |
+
'highlighted_diff2': highlighted_diff2,
|
215 |
+
'format': 'side_by_side'
|
216 |
+
}
|
217 |
+
else:
|
218 |
+
highlighted_diff = highlight_differences_words(contract_text1, contract_text2)
|
219 |
+
st.session_state.comparison_results = {
|
220 |
+
'similarity_score': similarity_score,
|
221 |
+
'highlighted_diff': highlighted_diff,
|
222 |
+
'format': 'combined'
|
223 |
+
}
|
224 |
+
|
225 |
+
# Display comparison results
|
226 |
if st.session_state.comparison_results:
|
227 |
st.metric("Document Similarity Score",
|
228 |
+
f"{st.session_state.comparison_results['similarity_score']:.2f}%")
|
229 |
|
230 |
if st.session_state.comparison_results['similarity_score'] < 50:
|
231 |
st.warning("Significant differences detected")
|
232 |
|
233 |
st.markdown("**Visual Difference Highlighting:**")
|
234 |
+
if st.session_state.comparison_results['format'] == 'side_by_side':
|
235 |
+
col1, col2 = st.columns(2)
|
236 |
+
with col1:
|
237 |
+
st.markdown("### Original Document")
|
238 |
+
st.markdown(f'<div style="border:1px solid #ccc; padding:10px; white-space: pre-wrap; font-family: monospace; font-size: 0.9em;">{st.session_state.comparison_results["highlighted_diff1"]}</div>', unsafe_allow_html=True)
|
239 |
+
with col2:
|
240 |
+
st.markdown("### Modified Document")
|
241 |
+
st.markdown(f'<div style="border:1px solid #ccc; padding:10px; white-space: pre-wrap; font-family: monospace; font-size: 0.9em;">{st.session_state.comparison_results["highlighted_diff2"]}</div>', unsafe_allow_html=True)
|
242 |
+
else:
|
243 |
+
st.markdown(
|
244 |
+
f'<div style="border:1px solid #ddd; padding:10px; max-height:400px; overflow-y:auto;">{st.session_state.comparison_results["highlighted_diff"]}</div>',
|
245 |
+
unsafe_allow_html=True
|
246 |
+
)
|
247 |
|
248 |
# ===== QUESTION ANALYSIS SECTION =====
|
249 |
st.header("3. Clause Analysis")
|
|
|
291 |
except Exception as e:
|
292 |
st.session_state.analysis_results = st.session_state.analysis_results or {}
|
293 |
st.session_state.analysis_results['doc2'] = f"Analysis failed: {str(e)}"
|
294 |
+
|
295 |
+
# Display analysis results
|
296 |
if st.session_state.analysis_results:
|
297 |
col1, col2 = st.columns(2)
|
298 |
with col1:
|
|
|
304 |
st.success(st.session_state.analysis_results.get('doc2', 'No analysis performed yet'))
|
305 |
|
306 |
if __name__ == "__main__":
|
307 |
+
main()
|