Initial space app

.gitattributes

@@ -1,3 +1,4 @@
+index.tsv filter=lfs diff=lfs merge=lfs -text
 *.7z filter=lfs diff=lfs merge=lfs -text
 *.arrow filter=lfs diff=lfs merge=lfs -text
 *.bin filter=lfs diff=lfs merge=lfs -text
@@ -33,3 +34,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+index.tsv filter=lfs diff=lfs merge=lfs -text

README.md

@@ -1,10 +1,10 @@
 ---
 title: Waqfeya
-emoji: 🌍
-colorFrom: yellow
-colorTo: yellow
+emoji: 📚
+colorFrom: pink
+colorTo: pink
 sdk: gradio
-sdk_version: 5.26.0
+sdk_version: 5.9.1
 app_file: app.py
 pinned: false
 license: mit

app.py

@@ -0,0 +1,165 @@
+import csv
+import json
+import urllib.parse
+
+from pathlib import Path
+
+import gradio as gr
+
+from fuzzywuzzy import fuzz
+from pyarabic.araby import strip_tashkeel
+
+
+def main():
+  with gr.Blocks(
+    theme=gr.themes.Default(font=[gr.themes.GoogleFont('Noto Sans Arabic'), 'Arial', 'sans-serif']),
+    css='\n'.join([
+      'html, body, .gradio-container { direction: rtl !important; }',
+      'h1 { text-align: center; display: block; }',
+      'th, td { text-align: right !important; }',
+      'th span { white-space: nowrap !important; }',
+      '.icon-wrap { right: unset !important; left: var(--size-3) !important; }',
+    ])
+  ) as demo:
+    index_state = gr.State()
+    results_data = gr.State()
+
+    gr.Markdown('# ابحث في كتب المكتبة الوقفية 📚', rtl=True)
+
+    title = gr.Textbox(label='عنوان الكتاب', placeholder='اكتب عنوان الكتاب', lines=1, rtl=True)
+
+    with gr.Row():
+      category = gr.Dropdown(choices=['جارٍ التحميل...'], label='التصنيف (اختياري)', interactive=False)
+      author = gr.Dropdown(choices=['جارٍ التحميل...'], label='المؤلف (اختياري)', interactive=False)
+
+    search_button = gr.Button('ابحث')
+
+    gr.Markdown('## النتائج 🎯', rtl=True)
+
+    results = gr.Dataframe(headers=['#', 'العنوان', 'المؤلف', 'التصنيف', 'درجة التطابق'], interactive=False)
+
+    download_label = gr.Markdown('### تحميل ملفات الكتاب 📥', visible=False, rtl=True)
+    with gr.Tabs(visible=False) as details_box:
+      with gr.Tab('PDF'):
+        pdf_tab = gr.Markdown(rtl=True)
+      with gr.Tab('TXT'):
+        txt_tab = gr.Markdown(rtl=True)
+      with gr.Tab('DOCX'):
+        docx_tab = gr.Markdown(rtl=True)
+
+    def load_data():
+      _index = load_index()
+      _categories = get_categories(_index)
+      _authors = get_authors(_index)
+
+      return (
+        _index,
+        gr.update(choices=_categories, value=_categories[0], interactive=True),
+        gr.update(choices=_authors, value=_authors[0], interactive=True),
+      )
+
+    def show_details(evt: gr.SelectData, index_state, results_data):
+      book_details = index_state[results_data[evt.index[0]][0] - 1]
+
+      pdf_paths = generate_download_url(book_details[-4])
+      txt_paths = generate_download_url(book_details[-3])
+      docx_paths = generate_download_url(book_details[-2])
+
+      return [
+        gr.update(visible=True, value=pdf_paths),
+        gr.update(visible=True, value=txt_paths),
+        gr.update(visible=True, value=docx_paths),
+        gr.update(visible=True),
+        gr.update(visible=True),
+      ]
+
+    search_button.click(
+      fn=lambda t, c, a, idx: handle_search(idx, t, c, a),
+      inputs=[title, category, author, index_state],
+      outputs=[results, results_data, details_box],
+    )
+
+    results.select(
+      fn=show_details,
+      inputs=[index_state, results_data],
+      outputs=[pdf_tab, txt_tab, docx_tab, details_box, download_label],
+    )
+
+    demo.load(load_data, outputs=[index_state, category, author])
+
+  demo.launch()
+
+
+def load_index():
+  with open('index.tsv', 'r', encoding='utf-8') as file:
+    data = list(csv.reader(file, delimiter='\t'))[1:]
+
+  for i in range(len(data)):
+    data[i] = [i + 1] + data[i] + [normalize_text(data[i][2])]
+
+  return data
+
+
+def get_categories(index):
+  return [''] + sorted(set([row[1] for row in index if row[1]]))
+
+
+def get_authors(index):
+  return [''] + sorted(set([row[2] for row in index if row[2]]))
+
+
+def handle_search(index, title: str, category: str, author: str):
+  title = normalize_text(title)
+
+  if not title.strip():
+    return [[['', 'يرجى إدخال عنوان للبحث.', '', '', '']], [['', 'يرجى إدخال عنوان للبحث.', '', '', '']], gr.update(visible=False)]
+
+  filtered = index
+
+  if category and category != '':
+    filtered = [row for row in filtered if row[1] == category]
+
+  if author and author != '':
+    filtered = [row for row in filtered if row[2] == author]
+
+  scored_results = []
+  for row in filtered:
+    score = fuzz.partial_ratio(title, row[-1])
+
+    if score > 50:
+      scored_results.append((score, row))
+
+  if not scored_results:
+    return [[['', 'لم يتم العثور على نتائج مطابقة.', '', '', '']], [['', 'لم يتم العثور على نتائج مطابقة.', '', '', '']], gr.update(visible=False)]
+
+  scored_results.sort(reverse=True)
+
+  result_table = [[row[0], row[3], row[2], row[1], score] for score, row in scored_results[:100]]
+
+  return [result_table, result_table, gr.update(visible=False)]
+
+
+def generate_download_url(paths):
+  formatted_paths = []
+
+  for path in json.loads(paths.replace("'", '"')):
+    encoded_path = f'https://huggingface.co/datasets/ieasybooks-org/waqfeya-library/resolve/main/{urllib.parse.quote(path[2:])}'
+    formatted_path = f'{encoded_path}?download=true'
+    formatted_paths.append(f'- [{Path(path).name}]({formatted_path})')
+
+  return '\n'.join(formatted_paths)
+
+
+def normalize_text(text):
+  text = strip_tashkeel(text)
+  text = text.replace('أ', 'ا')
+  text = text.replace('إ', 'ا')
+  text = text.replace('آ', 'ا')
+  text = text.replace('ي', 'ى')
+  text = text.replace('ة', 'ه')
+
+  return text
+
+
+if __name__ == '__main__':
+  main()

index.tsv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4c8c53cf6fbe2340fb6ebc3194c09c7b485d06adc627cc9d28a0c39e2c7a0df0
+size 16784007

requirements.txt

@@ -0,0 +1,4 @@
+PyArabic==0.6.15
+fuzzywuzzy==0.18.0
+gradio==5.9.1
+python-Levenshtein==0.27.1