Update app.py

app.py

@@ -3,45 +3,7 @@ import requests
 import urllib.parse
 import re
 import xmltodict
-from itertools import product, combinations
-import nltk
-from nltk.corpus import stopwords
-from nltk.stem import SnowballStemmer
-from collections import Counter
-import concurrent.futures
-import ssl
-import os
-
-# Configurar SSL para la descarga de NLTK
-try:
-    _create_unverified_https_context = ssl._create_unverified_context
-except AttributeError:
-    pass
-else:
-    ssl._create_default_https_context = _create_unverified_https_context
-
-# Función para descargar recursos de NLTK
-def download_nltk_resources():
-    resources = ['punkt', 'stopwords', 'words']
-    for resource in resources:
-        try:
-            nltk.data.find(f'tokenizers/{resource}')
-        except LookupError:
-            print(f"Descargando {resource}...")
-            nltk.download(resource, quiet=True)
-
-# Descargar recursos de NLTK
-download_nltk_resources()
-
-# Configurar el directorio de datos de NLTK
-nltk_data_dir = '/tmp/nltk_data'
-os.makedirs(nltk_data_dir, exist_ok=True)
-nltk.data.path.append(nltk_data_dir)
-
-# Inicializar el stemmer y la lista de stopwords
-stemmer = SnowballStemmer("spanish")
-stop_words = set(stopwords.words('spanish'))
-english_words = set(nltk.corpus.words.words())
+import inflect
 
 # Función para obtener sugerencias de DuckDuckGo
 def fetch_duckduckgo_suggestions(query, lang_code="es"):
@@ -95,7 +57,7 @@ def fetch_youtube_suggestions(query, lang_code="es"):
         return []
 
 # Función para obtener sugerencias de Bing
-def fetch_bing_suggestions(query, market="es-ES"):
+def fetch_bing_suggestions(query, market="en-US"):
     url = "https://api.bing.com/qsml.aspx"
     params = {
         "Market": market,
@@ -125,7 +87,7 @@ def fetch_bing_suggestions(query, market="es-ES"):
         return []
 
 # Función para obtener sugerencias de Amazon
-def fetch_amazon_suggestions(query, market_id="A1F83G8C2ARO7P", alias="aps"):
+def fetch_amazon_suggestions(query, market_id="ATVPDKIKX0DER", alias="aps"):
     url = "https://completion.amazon.com/api/2017/suggestions"
     params = {
         "mid": market_id,
@@ -144,115 +106,160 @@ def fetch_amazon_suggestions(query, market_id="A1F83G8C2ARO7P", alias="aps"):
     else:
         return []
 
-# Función para expandir la palabra clave
-def expand_keyword(keyword):
-    # Tokenizar la palabra clave
-    tokens = nltk.word_tokenize(keyword.lower())
-    
-    # Eliminar stopwords y aplicar stemming
-    tokens = [stemmer.stem(token) for token in tokens if token not in stop_words]
-    
-    # Generar variaciones
-    variations = []
-    for i in range(1, len(tokens) + 1):
-        variations.extend(list(combinations(tokens, i)))
-    
-    # Convertir tuplas a strings
-    expanded_keywords = [" ".join(variation) for variation in variations]
-    
-    # Añadir variaciones con prefijos y sufijos
-    prefixes = ['como', 'que', 'donde', 'cuando', 'por que', 'cual', 'mejor', 'peor', 'top']
-    suffixes = ['gratis', 'online', 'pdf', 'precios', 'opiniones', 'review', 'tutorial', 'curso']
-    
-    for prefix in prefixes:
-        expanded_keywords.append(f"{prefix} {keyword}")
-    
-    for suffix in suffixes:
-        expanded_keywords.append(f"{keyword} {suffix}")
-    
-    # Añadir plurales y singulares (simplificado)
-    expanded_keywords.extend([f"{kw}s" for kw in expanded_keywords])
-    expanded_keywords.extend([kw[:-1] if kw.endswith('s') else kw for kw in expanded_keywords])
-    
-    # Añadir variaciones con números
-    expanded_keywords.extend([f"{kw} {i}" for kw in expanded_keywords for i in range(1, 11)])
-    
-    # Eliminar duplicados y la palabra clave original vacía
-    expanded_keywords = list(set(expanded_keywords) - {''})
-    
+# Función para generar plurales y variaciones
+def generate_plural_variations(keyword):
+    p = inflect.engine()
+    singular = keyword
+    plural = p.plural(keyword)
+    return [singular, plural]
+
+# Función para expandir la palabra clave con stopwords y conjunciones
+STOP_WORDS = ["de", "en", "por", "para", "con", "sin", "y", "o"]
+
+def expand_keyword_with_stopwords(keyword):
+    expanded_keywords = [keyword]
+    for word in STOP_WORDS:
+        expanded_keywords.append(f"{keyword} {word}")
+        expanded_keywords.append(f"{word} {keyword}")
     return expanded_keywords
 
-# Función para obtener sugerencias de todas las plataformas
-def fetch_all_suggestions(keyword):
-    platforms = {
-        'Google': fetch_google_suggestions,
-        'DuckDuckGo': fetch_duckduckgo_suggestions,
-        'YouTube': fetch_youtube_suggestions,
-        'Bing': fetch_bing_suggestions,
-        'Amazon': fetch_amazon_suggestions
-    }
-    
-    all_suggestions = []
-    
-    with concurrent.futures.ThreadPoolExecutor(max_workers=5) as executor:
-        future_to_platform = {executor.submit(func, keyword): platform for platform, func in platforms.items()}
-        for future in concurrent.futures.as_completed(future_to_platform):
-            platform = future_to_platform[future]
-            try:
-                suggestions = future.result()
-                all_suggestions.extend([(suggestion, platform) for suggestion in suggestions])
-            except Exception as exc:
-                print(f'{platform} generated an exception: {exc}')
-    
-    return all_suggestions
+# Función para expandir las palabras clave
+def expand_keyword(keyword):
+    expanded_keywords = expand_keyword_with_stopwords(keyword)
+    expanded_keywords += generate_plural_variations(keyword)
+    for letter in 'abcdefghijklmnopqrstuvwxyz*_':
+        expanded_keywords.append(keyword + " " + letter)
+        expanded_keywords.append(letter + " " + keyword)
+    return expanded_keywords
 
 # Función principal
 def main(keyword):
     expanded_keywords = expand_keyword(keyword)
-    all_suggestions = []
-    
-    with concurrent.futures.ThreadPoolExecutor(max_workers=10) as executor:
-        future_to_keyword = {executor.submit(fetch_all_suggestions, kw): kw for kw in expanded_keywords}
-        for future in concurrent.futures.as_completed(future_to_keyword):
-            kw = future_to_keyword[future]
-            try:
-                suggestions = future.result()
-                all_suggestions.extend(suggestions)
-            except Exception as exc:
-                print(f'{kw} generated an exception: {exc}')
-    
-    # Contar las sugerencias y sus plataformas
-    suggestion_counter = Counter(sugg for sugg, _ in all_suggestions)
-    platform_counter = Counter(platform for _, platform in all_suggestions)
-    
-    # Ordenar las sugerencias por frecuencia
-    sorted_suggestions = sorted(suggestion_counter.items(), key=lambda x: x[1], reverse=True)
-    
-    # Crear el HTML de salida
-    html_output = "<h3>Todas las sugerencias ordenadas por relevancia:</h3><ul>"
-    for suggestion, count in sorted_suggestions:
-        platforms = [platform for sugg, platform in all_suggestions if sugg == suggestion]
-        platforms_str = ", ".join(set(platforms))
-        relevance_score = count * len(set(platforms))
-        html_output += f"<li><strong>{suggestion}</strong> (Frecuencia: {count}, Plataformas: {platforms_str}, Puntuación de relevancia: {relevance_score})</li>"
-    html_output += "</ul>"
-    
-    # Añadir estadísticas por plataforma
-    html_output += "<h3>Estadísticas por plataforma:</h3><ul>"
-    for platform, count in platform_counter.most_common():
-        html_output += f"<li>{platform}: {count} sugerencias</li>"
-    html_output += "</ul>"
-    
-    # Añadir botón para copiar todas las sugerencias
-    all_suggestions_text = ", ".join(sugg for sugg, _ in sorted_suggestions)
-    html_output += f"""
+    all_suggestions = {}
+    google_suggestions_all = []
+    duckduckgo_suggestions_all = []
+    youtube_suggestions_all = []
+    bing_suggestions_all = []
+    amazon_suggestions_all = []
+
+    # Obtener sugerencias de cada fuente
+    for exp_keyword in expanded_keywords:
+        suggestions = fetch_duckduckgo_suggestions(exp_keyword)
+        duckduckgo_suggestions_all.extend(suggestions)
+        for suggestion in suggestions:
+            if suggestion in all_suggestions:
+                all_suggestions[suggestion] += 1
+            else:
+                all_suggestions[suggestion] = 1
+
+        suggestions = fetch_google_suggestions(exp_keyword)
+        google_suggestions_all.extend(suggestions)
+        for suggestion in suggestions:
+            if suggestion in all_suggestions:
+                all_suggestions[suggestion] += 1
+            else:
+                all_suggestions[suggestion] = 1
+
+        suggestions = fetch_youtube_suggestions(exp_keyword)
+        youtube_suggestions_all.extend(suggestions)
+        for suggestion in suggestions:
+            if suggestion in all_suggestions:
+                all_suggestions[suggestion] += 1
+            else:
+                all_suggestions[suggestion] = 1
+
+        suggestions = fetch_bing_suggestions(exp_keyword)
+        bing_suggestions_all.extend(suggestions)
+        for suggestion in suggestions:
+            if suggestion in all_suggestions:
+                all_suggestions[suggestion] += 1
+            else:
+                all_suggestions[suggestion] = 1
+
+        suggestions = fetch_amazon_suggestions(exp_keyword)
+        amazon_suggestions_all.extend(suggestions)
+        for suggestion in suggestions:
+            if suggestion in all_suggestions:
+                all_suggestions[suggestion] += 1
+            else:
+                all_suggestions[suggestion] = 1
+
+    # Filtrar las top 10 sugerencias de cada plataforma
+    google_top_10 = list(set(google_suggestions_all))[:10]
+    duckduckgo_top_10 = list(set(duckduckgo_suggestions_all))[:10]
+    youtube_top_10 = list(set(youtube_suggestions_all))[:10]
+    bing_top_10 = list(set(bing_suggestions_all))[:10]
+    amazon_top_10 = list(set(amazon_suggestions_all))[:10]
+
+    # Ordenar y filtrar las sugerencias combinadas
+    sorted_suggestions = sorted(all_suggestions.items(), key=lambda item: item[1], reverse=True)
+    combined_top_10_suggestions = [sug for sug, freq in sorted_suggestions if freq >= 2][:10]
+    suggestions_str = ", ".join(combined_top_10_suggestions)
+
+    # Crear la lista de todas las palabras clave con su número de repeticiones
+    all_suggestions_str = "<ul>"
+    for suggestion, freq in sorted_suggestions:
+        all_suggestions_str += f"<li>{suggestion} - {freq} repeticiones</li>"
+    all_suggestions_str += "</ul>"
+
+    # Crear el HTML de salida con un botón de copia
+    html_output = f"""
     <div>
-        <h3>Copiar todas las sugerencias:</h3>
+        <b>Sugerencias combinadas de Google, DuckDuckGo, YouTube, Bing y Amazon (Top 10 combinadas):</b> <span id='suggestions_text'>{suggestions_str}</span>
         <button class="lg secondary svelte-cmf5ev" style="font-size: small; padding: 2px; color: #808080ba; border: none; margin-left: 5px;" 
-        onclick='navigator.clipboard.writeText(`{all_suggestions_text}`).then(() => alert("Todas las sugerencias copiadas al portapapeles"))'>&nbsp;Copiar todas las sugerencias&nbsp;</button>
+        onclick='navigator.clipboard.writeText(document.getElementById("suggestions_text").innerText).then(() => alert("Texto copiado al portapapeles"))'>&nbsp;✂&nbsp;</button>
     </div>
+
+    <h4>Top 10 Sugerencias de Google:</h4>
+    <ul>
     """
-    
+    for suggestion in google_top_10:
+        freq = all_suggestions[suggestion]
+        html_output += f"<li>{suggestion} ({freq})</li>"
+    html_output += "</ul>"
+
+    html_output += """
+    <h4>Top 10 Sugerencias de DuckDuckGo:</h4>
+    <ul>
+    """
+    for suggestion in duckduckgo_top_10:
+        freq = all_suggestions[suggestion]
+        html_output += f"<li>{suggestion} ({freq})</li>"
+    html_output += "</ul>"
+
+    html_output += """
+    <h4>Top 10 Sugerencias de YouTube:</h4>
+    <ul>
+    """
+    for suggestion in youtube_top_10:
+        freq = all_suggestions[suggestion]
+        html_output += f"<li>{suggestion} ({freq})</li>"
+    html_output += "</ul>"
+
+    html_output += """
+    <h4>Top 10 Sugerencias de Bing:</h4>
+    <ul>
+    """
+    for suggestion in bing_top_10:
+        freq = all_suggestions[suggestion]
+        html_output += f"<li>{suggestion} ({freq})</li>"
+    html_output += "</ul>"
+
+    html_output += """
+    <h4>Top 10 Sugerencias de Amazon:</h4>
+    <ul>
+    """
+    for suggestion in amazon_top_10:
+        freq = all_suggestions[suggestion]
+        html_output += f"<li>{suggestion} ({freq})</li>"
+    html_output += "</ul>"
+
+    # Agregar la lista completa de todas las palabras clave
+    html_output += """
+    <h4>Lista completa de palabras clave con su número de repeticiones:</h4>
+    """
+    html_output += all_suggestions_str
+
     return html_output
 
 # Interfaz de Gradio
@@ -261,8 +268,8 @@ iface = gr.Interface(
     inputs="text",
     outputs="html",
     title="<div style='margin:0 auto;text-align:center'><div style='margin:0 auto;text-align:center'><img style='width:100px;display: inline-table;margin-bottom:-10px' src='https://artxeweb.com/media/files/search.jpg'><p>Sugerencias Combinadas de Google, DuckDuckGo, YouTube, Bing y Amazon</p></div>",
-    description="<p style='margin-bottom:10px;text-align:center;background: #ffffff; padding: 8px; border-radius: 8px; border-width: 1px; border: solid 1px #e5e7eb;'>Ingrese una palabra clave para obtener sugerencias de búsqueda relacionadas de Google, DuckDuckGo, YouTube, Bing y Amazon. Se mostrarán todas las sugerencias ordenadas por relevancia.</p>",
+    description="<p style='margin-bottom:10px;text-align:center;background: #ffffff; padding: 8px; border-radius: 8px; border-width: 1px; border: solid 1px #e5e7eb;'>Ingrese una palabra clave para obtener sugerencias de búsqueda relacionadas de Google, DuckDuckGo, YouTube, Bing y Amazon. Se mostrarán las 10 primeras sugerencias combinadas y también las 10 principales de cada plataforma por separado.</p>",
     article="<div style='margin-top:10px'><p style='text-align: center !important; background: #ffffff; padding: 5px 30px; border-radius: 8px; border-width: 1px; border: solid 1px #e5e7eb; width: fit-content; margin: auto;'>Desarrollada por <a style='text-decoration: none !important; color: #e12a31 !important;' href='https://artxeweb.com'>© Artxe Web</a></p></div>"
 )
 
-iface.launch()
+iface.launch()