Spaces:
Running
Running
Fix metadata field ordering and tag classification issues
Browse files- utils/ui_utils.py +13 -7
utils/ui_utils.py
CHANGED
|
@@ -21,17 +21,13 @@ def display_results(result, container, custom_prompt=""):
|
|
| 21 |
meta = {k: v for k, v in result.items()
|
| 22 |
if k not in ['pages_data', 'illustrations', 'ocr_contents', 'raw_response_data']}
|
| 23 |
|
| 24 |
-
# Create a compact metadata section
|
| 25 |
meta_html = '<div style="display: flex; flex-wrap: wrap; gap: 0.3rem; margin-bottom: 0.3rem;">'
|
| 26 |
|
| 27 |
# Document type
|
| 28 |
if 'detected_document_type' in meta:
|
| 29 |
meta_html += f'<div><strong>Type:</strong> {meta["detected_document_type"]}</div>'
|
| 30 |
|
| 31 |
-
# Processing time
|
| 32 |
-
if 'processing_time' in meta:
|
| 33 |
-
meta_html += f'<div><strong>Time:</strong> {meta["processing_time"]:.1f}s</div>'
|
| 34 |
-
|
| 35 |
# Page information
|
| 36 |
if 'limited_pages' in meta:
|
| 37 |
meta_html += f'<div><strong>Pages:</strong> {meta["limited_pages"]["processed"]}/{meta["limited_pages"]["total"]}</div>'
|
|
@@ -39,6 +35,14 @@ def display_results(result, container, custom_prompt=""):
|
|
| 39 |
meta_html += '</div>'
|
| 40 |
st.markdown(meta_html, unsafe_allow_html=True)
|
| 41 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 42 |
# Language metadata on a separate line, Subject Tags below
|
| 43 |
|
| 44 |
# First show languages if available
|
|
@@ -100,7 +104,7 @@ def display_results(result, container, custom_prompt=""):
|
|
| 100 |
# Create a separate line for Time if we have time-related tags
|
| 101 |
if 'topics' in result and result['topics']:
|
| 102 |
time_tags = [topic for topic in result['topics']
|
| 103 |
-
if any(term in topic.lower() for term in ["century", "pre-", "era"
|
| 104 |
if time_tags:
|
| 105 |
time_html = '<div style="display: flex; align-items: center; margin: 0.2rem 0; flex-wrap: wrap;">'
|
| 106 |
time_html += '<div style="margin-right: 0.3rem; font-weight: bold;">Time:</div>'
|
|
@@ -113,7 +117,7 @@ def display_results(result, container, custom_prompt=""):
|
|
| 113 |
if 'topics' in result and result['topics']:
|
| 114 |
# Filter out time-related tags which are already displayed
|
| 115 |
subject_tags = [topic for topic in result['topics']
|
| 116 |
-
if not any(term in topic.lower() for term in ["century", "pre-", "era"
|
| 117 |
|
| 118 |
if subject_tags:
|
| 119 |
# Create a separate line for Subject Tags
|
|
@@ -133,6 +137,8 @@ def display_results(result, container, custom_prompt=""):
|
|
| 133 |
tag_class += " tag-document-type" # Document types
|
| 134 |
elif any(term in topic.lower() for term in ["travel", "military", "science", "medicine", "education", "art", "literature"]):
|
| 135 |
tag_class += " tag-subject" # Subject domains
|
|
|
|
|
|
|
| 136 |
|
| 137 |
# Add each tag as an inline span
|
| 138 |
tags_html += f'<span class="{tag_class}">{topic}</span>'
|
|
|
|
| 21 |
meta = {k: v for k, v in result.items()
|
| 22 |
if k not in ['pages_data', 'illustrations', 'ocr_contents', 'raw_response_data']}
|
| 23 |
|
| 24 |
+
# Create a compact metadata section for primary metadata
|
| 25 |
meta_html = '<div style="display: flex; flex-wrap: wrap; gap: 0.3rem; margin-bottom: 0.3rem;">'
|
| 26 |
|
| 27 |
# Document type
|
| 28 |
if 'detected_document_type' in meta:
|
| 29 |
meta_html += f'<div><strong>Type:</strong> {meta["detected_document_type"]}</div>'
|
| 30 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 31 |
# Page information
|
| 32 |
if 'limited_pages' in meta:
|
| 33 |
meta_html += f'<div><strong>Pages:</strong> {meta["limited_pages"]["processed"]}/{meta["limited_pages"]["total"]}</div>'
|
|
|
|
| 35 |
meta_html += '</div>'
|
| 36 |
st.markdown(meta_html, unsafe_allow_html=True)
|
| 37 |
|
| 38 |
+
# Processing time - separate section for proper ordering of all metadata fields
|
| 39 |
+
if 'processing_time' in meta:
|
| 40 |
+
time_html = '<div style="display: flex; align-items: center; margin: 0.2rem 0; flex-wrap: wrap;">'
|
| 41 |
+
time_html += '<div style="margin-right: 0.3rem; font-weight: bold;">Time:</div>'
|
| 42 |
+
time_html += f'<div>{meta["processing_time"]:.1f}s</div>'
|
| 43 |
+
time_html += '</div>'
|
| 44 |
+
st.markdown(time_html, unsafe_allow_html=True)
|
| 45 |
+
|
| 46 |
# Language metadata on a separate line, Subject Tags below
|
| 47 |
|
| 48 |
# First show languages if available
|
|
|
|
| 104 |
# Create a separate line for Time if we have time-related tags
|
| 105 |
if 'topics' in result and result['topics']:
|
| 106 |
time_tags = [topic for topic in result['topics']
|
| 107 |
+
if any(term in topic.lower() for term in ["century", "pre-", "era"])]
|
| 108 |
if time_tags:
|
| 109 |
time_html = '<div style="display: flex; align-items: center; margin: 0.2rem 0; flex-wrap: wrap;">'
|
| 110 |
time_html += '<div style="margin-right: 0.3rem; font-weight: bold;">Time:</div>'
|
|
|
|
| 117 |
if 'topics' in result and result['topics']:
|
| 118 |
# Filter out time-related tags which are already displayed
|
| 119 |
subject_tags = [topic for topic in result['topics']
|
| 120 |
+
if not any(term in topic.lower() for term in ["century", "pre-", "era"])]
|
| 121 |
|
| 122 |
if subject_tags:
|
| 123 |
# Create a separate line for Subject Tags
|
|
|
|
| 137 |
tag_class += " tag-document-type" # Document types
|
| 138 |
elif any(term in topic.lower() for term in ["travel", "military", "science", "medicine", "education", "art", "literature"]):
|
| 139 |
tag_class += " tag-subject" # Subject domains
|
| 140 |
+
elif "historical" in topic.lower() and "document" in topic.lower():
|
| 141 |
+
tag_class += " tag-document-type" # "Historical Document Analysis" should be a document type
|
| 142 |
|
| 143 |
# Add each tag as an inline span
|
| 144 |
tags_html += f'<span class="{tag_class}">{topic}</span>'
|