Resolve conflitos

README.md

@@ -1,6 +1,6 @@
 ---
-title: Speech-Scope
-emoji: 🎤
+title: AIA-Scope
+emoji: 🔍
 colorFrom: gray
 colorTo: indigo
 sdk: gradio
@@ -10,9 +10,9 @@ pinned: false
 license: ecl-2.0
 ---
 
-## Speech-Scope
+## AIA-Scope
 
-Análise de métricas de áudio.
+Análise do arquivo .aia gerado pelo MIT App Inventor
 
 Confira a referência de configuração em [Hugging Face Spaces Config Reference](https://huggingface.co/docs/hub/spaces-config-reference).
 

app.py

@@ -1,83 +1,305 @@
 import gradio as gr
-import librosa
-import numpy as np
-import matplotlib.pyplot as plt
+from zipfile import ZipFile, BadZipFile
 import tempfile
-import librosa.display
+import os
+import re
+import pandas as pd
+import collections
+import json
+import glob
+from io import BytesIO
 
 
-def calculate_basic_metrics(y, sr):
-    pitches, magnitudes = librosa.piptrack(y=y, sr=sr)
-    average_pitch = np.mean(pitches[pitches > 0])
-    mfccs = librosa.feature.mfcc(y=y, sr=sr)
-    energy = np.sum(y ** 2)
-    zero_crossing_rate = np.mean(librosa.feature.zero_crossing_rate(y))
-    spectral_centroid = np.mean(librosa.feature.spectral_centroid(y=y, sr=sr))
+
+ai_patterns = [
+    "PIC*", "PersonalImageClassifier*", "Look*", "LookExtension*", "ChatBot", "ImageBot", "TMIC","TeachableMachine*",
+    "TeachableMachineImageClassifier*", "SpeechRecognizer*", "FaceExtension*","Pose*","Posenet","PosenetExtension", "Eliza*", "Alexa*"
+]
+
+
+drawing_and_animation_patterns = ["Ball", "Canvas", "ImageSprite"]
+maps_patterns = ["Map", "Marker", "Circle", "FeatureCollection", "LineString", "Navigation","Polygon", "Retangle" ]
+sensors_patterns = ["AccelerometerSensor", "BarcodeScanner", "Barometer", "Clock", "GyroscopeSensor", "Hygrometer", "LightSensor", "LocationSensor", "MagneticFieldSensor", "NearField","OrientationSensor", "ProximitySensor","Thermometer", "Pedometer"]
+social_patterns = ["ContactPicker", "EmailPicker", "PhoneCall", "PhoneNumberPicker", "Texting", "Twitter"]
+storage_patterns = ["File", "CloudDB", "DataFile", "Spreadsheet", "FusiontablesControl", "TinyDB", "TinyWebDB"]
+connectivity_patterns = ["BluetoothClient", "ActivityStarter", "Serial", "BluetoothServer", "Web"]
+
+
+def extract_components_using_regex(scm_content):
+    pattern = r'"\$Type":"(.*?)"'
+    components = re.findall(pattern, scm_content)
+    if 'roboflow' in scm_content.lower():
+        components.append("Using Roboflow")
+    return components
+
+
+def extract_category_components(components, patterns):
+    category_components = []
+    for component in components:
+        for pattern in patterns:
+            if component.startswith(pattern):
+                category_components.append(component)
+    return category_components
+
+
+def extract_extensions_from_aia(file_path: str):
+    extensions = []
+    with ZipFile(file_path, 'r') as zip_ref:
+        for file_path in zip_ref.namelist():
+            if file_path.endswith('components.json') and 'assets/external_comps/' in file_path:
+                with zip_ref.open(file_path) as file:
+                    components_json_content = file.read().decode('utf-8', errors='ignore')
+                    components_data = json.loads(components_json_content)
+                    for component in components_data:
+                        extension_type = component.get("type", "")
+                        if extension_type:
+                            extensions.append(extension_type)
+    return extensions
+
+def count_events_in_bky_file(bky_content):
+    
+    return bky_content.count('<block type="component_event"')
+
+def extract_app_name_from_scm_files(temp_dir):
+    scm_files = glob.glob(f"{temp_dir}/src/appinventor/*/*/*.scm")
+    for scm_file in scm_files:
+        with open(scm_file, 'r', encoding='utf-8', errors='ignore') as file:
+            content = file.read()
+
+            
+            regex_patterns = [
+                r'"AppName"\s*:\s*"([^"]+)"',
+                r'"AppName"\s*:\s*\'([^\']+)\''  
+            ]
+
+            for pattern in regex_patterns:
+                app_name_match = re.search(pattern, content)
+                if app_name_match:
+                    return app_name_match.group(1)
+
+    
+    print(f"Aviso: Nome do aplicativo não encontrado no diretório {temp_dir}")
+    return "N/A"
+
+def extract_project_info_from_properties(file_path):
+    
+    timestamp = "N/A"
+    app_name = "N/A"
+    app_version = "N/A"
+    authURL = "ai2.appinventor.mit.edu"
+
+    
+    with tempfile.TemporaryDirectory() as temp_dir:
+        with ZipFile(file_path, 'r') as zip_ref:
+            zip_ref.extractall(temp_dir)
+            
+            project_properties_file_path = 'youngandroidproject/project.properties'
+
+            
+            if project_properties_file_path in zip_ref.namelist():
+                with zip_ref.open(project_properties_file_path) as file:
+                    project_properties_lines = file.read().decode('utf-8').splitlines()
+
+                    
+                    timestamp = project_properties_lines[1] if len(project_properties_lines) > 1 else "N/A"
+
+                    
+                    for line in project_properties_lines:
+                        app_name_match = re.match(r'aname=(.*)', line)
+                        if app_name_match:
+                            app_name = app_name_match.group(1)
+
+                        app_version_match = re.match(r'versionname=(.*)', line)
+                        if app_version_match:
+                            app_version = app_version_match.group(1)
+
+        
+        if app_name == "N/A":
+         print("O campo App Name não foi encontrado em project.properties. Tentando encontrar em arquivos .scm...")
+        app_name = extract_app_name_from_scm_files(temp_dir)
+        print(f"Nome do App encontrado nos arquivos .scm: {app_name}")
+
+    # ...
+
 
     return {
-        'Average Pitch': average_pitch,
-        'Number of MFCCs': mfccs.shape[1],
-        'Energy': energy,
-        'Zero Crossing Rate': zero_crossing_rate,
-        'Spectral Centroid': spectral_centroid
+        'timestamp': timestamp,
+        'app_name': app_name,
+        'app_version': app_version,
+        'authURL': authURL
     }
 
 
-def calculate_advanced_metrics(y, sr):
-    metrics = {}
 
-    f0, _, _ = librosa.pyin(y, fmin=50, fmax=4000)
-    if f0 is not None:
-        metrics['Average F0 (YIN)'] = np.nanmean(f0)
 
-    chroma = librosa.feature.chroma_stft(y=y, sr=sr)
-    metrics['Average Chroma'] = np.mean(chroma)
 
-    spectral_contrast = librosa.feature.spectral_contrast(y=y, sr=sr)
-    metrics['Average Spectral Contrast'] = np.mean(spectral_contrast)
+def extract_ai_components(components):
+    ai_components = []
+    for component in components:
+        for pattern in ai_patterns:
+            if '*' in pattern and component.startswith(pattern[:-1]):
+                ai_components.append(component)
+            elif component == pattern:
+                ai_components.append(component)
+    if "roboflow" in ' '.join(components).lower():
+        ai_components.append("Using Roboflow")
+    return ai_components
+
+def extract_media_files(file_path: str):
+    media_files = []
+    with ZipFile(file_path, 'r') as zip_ref:
+        for file_path in zip_ref.namelist():
+            if 'assets/' in file_path and not file_path.endswith('/'):
+                media_files.append(os.path.basename(file_path))
+    return media_files
+
+def list_components_in_aia_file(file_path):
+    
+    results_df = pd.DataFrame(columns=[
+        'aia_file', 'project_info', 'components', 'IA components', 'screens', 'operators',
+        'variables', 'events', 'extensions', 'Media',
+        'Drawing and Animation', 'Maps', 'Sensors', 'Social', 'Storage', 'Connectivity'])
 
-    return metrics
+ 
+    pd.set_option('display.max_colwidth', None)
+    file_name = os.path.basename(file_path)
+    
 
+    components_list = []
+    number_of_screens = 0
+    operators_count = 0
+    variables_count = 0
+    events_count = 0 
+    
+    media_files = extract_media_files(file_path)
+    media_summary = ', '.join(media_files)
+    
+    project_info = extract_project_info_from_properties(file_path)
+    project_info_str = f"Timestamp: {project_info['timestamp']}, App Name: {project_info['app_name']}, Version: {project_info['app_version']}, AuthURL: {project_info['authURL']}"
 
-def generate_spectrogram(y, sr):
-    plt.figure(figsize=(10, 4))
-    librosa.display.specshow(librosa.amplitude_to_db(np.abs(librosa.stft(y)), ref=np.max), sr=sr, x_axis='time', y_axis='log')
-    plt.colorbar(format='%+2.0f dB')
-    plt.title('Spectrogram')
-    plt.tight_layout()
 
+    with tempfile.TemporaryDirectory() as temp_dir:
+        with ZipFile(file_path, 'r') as zip_ref:
+            zip_ref.extractall(temp_dir)
+        scm_files = glob.glob(temp_dir + '/src/appinventor/*/*/*.scm')
+        bky_files = glob.glob(temp_dir + '/src/appinventor/*/*/*.bky') 
 
-    with tempfile.NamedTemporaryFile(delete=False, suffix='.png', mode='w+b') as f:
-        plt.savefig(f.name, format='png')
-        plt.close()
-        return f.name
+        number_of_screens = len(scm_files)
+        for scm_file in scm_files:
+            with open(scm_file, 'r', encoding='utf-8', errors='ignore') as file:
+                content = file.read()
+                components = extract_components_using_regex(content)
+                components_list.extend(components)
+                operators_count += len(re.findall(r'[+\-*/<>!=&|]', content))
+                variables_count += len(re.findall(r'"\$Name":"(.*?)"', content))
 
+        
+        drawing_and_animation_summary = ', '.join(extract_category_components(components_list, drawing_and_animation_patterns))
+        maps_summary = ', '.join(extract_category_components(components_list, maps_patterns))
+        sensors_summary = ', '.join(extract_category_components(components_list, sensors_patterns))
+        social_summary = ', '.join(extract_category_components(components_list, social_patterns))
+        storage_summary = ', '.join(extract_category_components(components_list, storage_patterns))
+        connectivity_summary = ', '.join(extract_category_components(components_list, connectivity_patterns))
 
-def process_audio(file):
-    if file is None:
-        return {}, "placeholder.png"  
 
-    sr, y = file
+        
+        
+        extensions_list = []
+        extensions_list = extract_extensions_from_aia(file_path)
 
-    if y.dtype != np.float32:
-        y = y.astype(np.float32) / np.iinfo(y.dtype).max
+        for bky_file in bky_files:
+            with open(bky_file, 'r', encoding='utf-8', errors='ignore') as file:
+                bky_content = file.read()
+                events_count += count_events_in_bky_file(bky_content)
+                
 
-    basic_metrics = calculate_basic_metrics(y, sr)
-    advanced_metrics = calculate_advanced_metrics(y, sr)
 
-    metrics = {**basic_metrics, **advanced_metrics}
+        
+        extensions_summary = ', '.join(list(set(extensions_list)))
 
-    image_path = generate_spectrogram(y, sr)
+    components_count = collections.Counter(components_list)
+    components_summary = [f'{comp} ({count} x)' if count > 1 else comp for comp, count in components_count.items()]
+    ai_components_summary = extract_ai_components(components_list)
+    new_row = pd.DataFrame([{
+    'aia_file': file_name,
+    'project_info': project_info_str,
+    'components': ', '.join(components_summary),
+    'IA components': ', '.join(ai_components_summary),
+    'screens': number_of_screens,
+    'operators': operators_count,
+    'variables': variables_count,
+    'events': events_count,
+    'extensions': extensions_summary,
+    'Media': media_summary,
+    'Drawing and Animation': drawing_and_animation_summary,
+    'Maps': maps_summary,
+    'Sensors': sensors_summary,
+    'Social': social_summary,
+    'Storage': storage_summary,
+    'Connectivity': connectivity_summary
+    }])
 
-    return metrics, image_path
+    
+    results_df = pd.concat([results_df, new_row], ignore_index=True)
+    return results_df
+#
+
+
+output_style = """
+<style>
+    .output-container {
+        max-height: 500px; /* Ajuste a altura máxima conforme necessário */
+        overflow: auto; /* Isso permite a rolagem vertical e horizontal se necessário */
+        display: block; /* Isso garante que o container seja renderizado abaixo do botão submit */
+    }
+    .output-container table {
+        width: 100%; /* Isso faz com que a tabela utilize toda a largura do container */
+        border-collapse: collapse;
+    }
+    .output-container th, .output-container td {
+        border: 1px solid #ddd; /* Isso adiciona bordas às células para melhor visualização */
+        text-align: left;
+        padding: 8px;
+    }
+</style>
+"""
 
 
+def analyze_aia(uploaded_file):
+    try:
+        
+        file_path = uploaded_file.name if hasattr(uploaded_file, 'name') else None
+
+        if file_path and os.path.exists(file_path):
+            with ZipFile(file_path, 'r') as zip_ref:
+                with tempfile.TemporaryDirectory() as temp_dir:
+                    zip_ref.extractall(temp_dir)
+                    results_df = list_components_in_aia_file(file_path)
+                    
+                    html_result = results_df.to_html(escape=False, classes="output-html")
+                    return output_style + f'<div class="output-container">{html_result}</div>'
+
+
+        else:
+            return output_style + "Não foi possível localizar o arquivo .aia."
+
+    except BadZipFile:
+        return output_style + "Falha ao abrir o arquivo .aia como um arquivo zip. Ele pode estar corrompido ou não é um arquivo .aia válido."
+
+    except Exception as e:
+        return output_style + f"Erro ao processar o arquivo: {str(e)}"
+
 iface = gr.Interface(
-    fn=process_audio,
-    inputs=gr.Audio(label="Upload Audio"),
-    outputs=["json", "image"],
-    title="Speech-Scope",
-    description="Speech and audio Metrics Analysis"
+    fn=analyze_aia,
+    inputs=gr.File(),
+    outputs=gr.HTML(),
+    title="AIA-Scope",
+    description="Upload an .aia file to analyze its components.",
+    live=False  
 )
 
-iface.launch(debug=True)
+if __name__ == "__main__":
+    iface.launch(debug=True)
+
+
+