Spaces:
Running
Running
Update test5.html
Browse files- test5.html +110 -166
test5.html
CHANGED
@@ -3,211 +3,156 @@
|
|
3 |
<head>
|
4 |
<meta charset="UTF-8">
|
5 |
<meta name="viewport" content="width=device-width, initial-scale=1.0">
|
6 |
-
<title>
|
|
|
|
|
7 |
<style>
|
8 |
-
|
9 |
-
|
10 |
-
|
11 |
-
margin-top: 10px;
|
12 |
-
}
|
13 |
</style>
|
14 |
</head>
|
15 |
<body>
|
16 |
-
<h1>
|
17 |
-
|
18 |
-
<!-- Formulario para cargar archivos (permitir carga m煤ltiple) -->
|
19 |
<input type="file" id="fileInput" accept=".pdf,.txt,.docx" multiple>
|
20 |
-
<button onclick="processFiles()">Cargar y Analizar Archivos</button>
|
21 |
-
|
22 |
-
<!-- 脕rea de texto para la pregunta -->
|
23 |
<br><br>
|
|
|
|
|
|
|
|
|
|
|
|
|
24 |
<label for="question">Pregunta:</label>
|
25 |
-
<input type="text" id="question" placeholder="Escribe tu pregunta
|
26 |
-
|
27 |
-
|
28 |
-
<button onclick="askQuestion()">
|
29 |
-
|
30 |
-
<!-- 脕rea para mostrar el texto extra铆do de los archivos -->
|
31 |
-
<h3>Texto Extra铆do:</h3>
|
32 |
-
<textarea id="extractedText" readonly></textarea>
|
33 |
-
|
34 |
-
<!-- 脕rea para mostrar la respuesta -->
|
35 |
-
<h3>Respuesta:</h3>
|
36 |
<div id="response"></div>
|
37 |
|
38 |
-
<!-- Cargar la librer铆a PDF.js -->
|
39 |
-
<script src="https://cdnjs.cloudflare.com/ajax/libs/pdf.js/2.10.377/pdf.min.js"></script>
|
40 |
-
<script src="https://cdn.jsdelivr.net/npm/[email protected]/dist/jszip.min.js"></script>
|
41 |
-
|
42 |
<script>
|
43 |
-
//
|
44 |
-
|
45 |
-
|
46 |
-
|
47 |
-
let extractedText = '';
|
48 |
-
let modelReady = false;
|
49 |
|
50 |
-
//
|
51 |
function processFiles() {
|
52 |
const fileInput = document.getElementById('fileInput');
|
53 |
-
const
|
54 |
-
|
55 |
-
alert('Por favor, selecciona al menos un archivo.');
|
56 |
-
return;
|
57 |
-
}
|
58 |
-
|
59 |
-
extractedText = ''; // Reset the extracted text
|
60 |
-
let filePromises = [];
|
61 |
|
62 |
-
|
63 |
-
for (let file of files) {
|
64 |
-
const fileExtension = file.name.split('.').pop().toLowerCase();
|
65 |
const reader = new FileReader();
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
66 |
|
67 |
-
|
68 |
-
reader.onload = function(event) {
|
69 |
-
const fileContent = event.target.result;
|
70 |
-
|
71 |
-
// Extract text based on the file type
|
72 |
-
if (fileExtension === 'pdf') {
|
73 |
-
extractTextFromPDF(fileContent, resolve);
|
74 |
-
} else if (fileExtension === 'txt') {
|
75 |
-
extractTextFromTXT(fileContent, resolve);
|
76 |
-
} else if (fileExtension === 'docx') {
|
77 |
-
extractTextFromDOCX(fileContent, resolve);
|
78 |
-
} else {
|
79 |
-
reject(`Formato de archivo no soportado: ${file.name}`);
|
80 |
-
}
|
81 |
-
};
|
82 |
-
|
83 |
reader.readAsArrayBuffer(file);
|
84 |
-
}
|
85 |
-
|
86 |
-
filePromises.push(promise);
|
87 |
-
}
|
88 |
-
|
89 |
-
// Wait for all file promises to finish
|
90 |
-
Promise.all(filePromises)
|
91 |
-
.then(() => {
|
92 |
-
// Display extracted text in the textarea
|
93 |
-
document.getElementById('extractedText').value = extractedText;
|
94 |
-
alert('Texto extra铆do de los archivos.');
|
95 |
-
})
|
96 |
-
.catch((error) => {
|
97 |
-
console.error('Error al procesar los archivos:', error);
|
98 |
-
alert('Hubo un error al procesar los archivos.');
|
99 |
-
});
|
100 |
-
}
|
101 |
-
|
102 |
-
// Function to extract text from a PDF file
|
103 |
-
function extractTextFromPDF(fileContent, resolve) {
|
104 |
-
const loadingTask = pdfjsLib.getDocument({ data: fileContent });
|
105 |
-
loadingTask.promise.then(function(pdf) {
|
106 |
-
let text = '';
|
107 |
-
const numPages = pdf.numPages;
|
108 |
-
let pagePromises = [];
|
109 |
-
|
110 |
-
for (let i = 1; i <= numPages; i++) {
|
111 |
-
pagePromises.push(pdf.getPage(i).then(function(page) {
|
112 |
-
return page.getTextContent().then(function(textContent) {
|
113 |
-
// Concatenate text extracted from each page
|
114 |
-
text += textContent.items.map(item => item.str).join(' ') + '\n';
|
115 |
-
});
|
116 |
-
}));
|
117 |
}
|
118 |
-
|
119 |
-
Promise.all(pagePromises).then(function() {
|
120 |
-
extractedText += text + '\n'; // Add the extracted text
|
121 |
-
resolve();
|
122 |
-
});
|
123 |
-
}).catch(function(error) {
|
124 |
-
console.error('Error al extraer texto del PDF:', error);
|
125 |
-
resolve();
|
126 |
});
|
127 |
}
|
128 |
|
129 |
-
//
|
130 |
-
function
|
131 |
-
|
132 |
-
|
133 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
134 |
|
135 |
-
|
136 |
-
|
137 |
-
const zip = new JSZip();
|
138 |
-
zip.loadAsync(fileContent).then(function(zip) {
|
139 |
-
zip.file('word/document.xml').async('string').then(function(content) {
|
140 |
-
const parser = new DOMParser();
|
141 |
-
const xmlDoc = parser.parseFromString(content, 'text/xml');
|
142 |
-
const texts = xmlDoc.getElementsByTagName('w:t');
|
143 |
-
extractedText += Array.from(texts).map(t => t.textContent).join(' ') + '\n';
|
144 |
-
resolve();
|
145 |
-
}).catch(function(error) {
|
146 |
-
console.error('Error al extraer texto del DOCX:', error);
|
147 |
-
resolve();
|
148 |
-
});
|
149 |
});
|
150 |
}
|
151 |
|
152 |
-
//
|
153 |
-
|
154 |
-
|
155 |
-
|
156 |
-
|
157 |
-
|
158 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
159 |
}
|
160 |
});
|
161 |
-
|
162 |
-
const result = await response.json();
|
163 |
-
|
164 |
-
if (result.error && result.error.includes('currently loading')) {
|
165 |
-
const estimatedTime = result.estimated_time || 20;
|
166 |
-
alert(`El modelo est谩 cargando. Estimaci贸n de tiempo restante: ${Math.round(estimatedTime)} segundos.`);
|
167 |
-
return false;
|
168 |
-
}
|
169 |
-
|
170 |
-
modelReady = true;
|
171 |
-
return true;
|
172 |
-
} catch (error) {
|
173 |
-
console.error('Error al verificar si el modelo est谩 listo:', error);
|
174 |
-
return false;
|
175 |
-
}
|
176 |
}
|
177 |
|
178 |
-
//
|
179 |
async function askQuestion() {
|
180 |
const question = document.getElementById('question').value;
|
181 |
-
const context =
|
182 |
|
183 |
-
|
184 |
-
|
185 |
-
alert('Por favor, ingresa una pregunta y aseg煤rate de que el contexto no est茅 vac铆o.');
|
186 |
return;
|
187 |
}
|
188 |
|
189 |
-
// Check if the model is ready
|
190 |
-
const modelReady = await checkModelReady();
|
191 |
-
if (!modelReady) {
|
192 |
-
return; // Don't continue if the model isn't ready
|
193 |
-
}
|
194 |
-
|
195 |
-
// Prepare the request data
|
196 |
const data = {
|
197 |
inputs: {
|
198 |
-
question: question,
|
199 |
-
context: context
|
200 |
}
|
201 |
};
|
202 |
-
|
|
|
|
|
|
|
203 |
console.log("key : " + cucu);
|
|
|
|
|
|
|
|
|
|
|
204 |
try {
|
205 |
-
const response = await fetch(
|
206 |
method: 'POST',
|
207 |
-
headers:
|
208 |
-
'Authorization': 'Bearer ' + cucu, // Replace with your Hugging Face API key
|
209 |
-
'Content-Type': 'application/json'
|
210 |
-
},
|
211 |
body: JSON.stringify(data)
|
212 |
});
|
213 |
|
@@ -215,12 +160,11 @@
|
|
215 |
if (response.ok) {
|
216 |
document.getElementById('response').innerText = result.answer;
|
217 |
} else {
|
218 |
-
|
219 |
-
alert(`Hubo un error al procesar la solicitud: ${JSON.stringify(result)}`);
|
220 |
}
|
221 |
} catch (error) {
|
222 |
console.error('Error al hacer la consulta:', error);
|
223 |
-
|
224 |
}
|
225 |
}
|
226 |
</script>
|
|
|
3 |
<head>
|
4 |
<meta charset="UTF-8">
|
5 |
<meta name="viewport" content="width=device-width, initial-scale=1.0">
|
6 |
+
<title>Extracci贸n de Texto y Preguntas con Modelo Multiling眉e</title>
|
7 |
+
<script src="https://cdnjs.cloudflare.com/ajax/libs/pdf.js/2.16.105/pdf.min.js"></script>
|
8 |
+
<script src="https://cdnjs.cloudflare.com/ajax/libs/jszip/3.10.1/jszip.min.js"></script>
|
9 |
<style>
|
10 |
+
body { font-family: Arial, sans-serif; margin: 20px; }
|
11 |
+
#output { margin-top: 20px; }
|
12 |
+
textarea { width: 100%; height: 300px; }
|
|
|
|
|
13 |
</style>
|
14 |
</head>
|
15 |
<body>
|
16 |
+
<h1>Sube tu archivo (PDF, TXT, DOCX) y haz preguntas</h1>
|
|
|
|
|
17 |
<input type="file" id="fileInput" accept=".pdf,.txt,.docx" multiple>
|
|
|
|
|
|
|
18 |
<br><br>
|
19 |
+
<button onclick="processFiles()">Cargar y procesar archivos</button>
|
20 |
+
<br><br>
|
21 |
+
|
22 |
+
<textarea id="output" placeholder="Texto extra铆do del archivo..."></textarea>
|
23 |
+
<br><br>
|
24 |
+
|
25 |
<label for="question">Pregunta:</label>
|
26 |
+
<input type="text" id="question" placeholder="Escribe tu pregunta aqu铆...">
|
27 |
+
<br><br>
|
28 |
+
|
29 |
+
<button onclick="askQuestion()">Preguntar</button>
|
30 |
+
|
|
|
|
|
|
|
|
|
|
|
|
|
31 |
<div id="response"></div>
|
32 |
|
|
|
|
|
|
|
|
|
33 |
<script>
|
34 |
+
// Funci贸n para limpiar el texto extra铆do
|
35 |
+
function cleanText(text) {
|
36 |
+
return text.replace(/\s+/g, ' ').trim(); // Elimina espacios y saltos de l铆nea innecesarios
|
37 |
+
}
|
|
|
|
|
38 |
|
39 |
+
// Funci贸n para procesar los archivos cargados
|
40 |
function processFiles() {
|
41 |
const fileInput = document.getElementById('fileInput');
|
42 |
+
const output = document.getElementById('output');
|
43 |
+
let extractedText = '';
|
|
|
|
|
|
|
|
|
|
|
|
|
44 |
|
45 |
+
Array.from(fileInput.files).forEach(file => {
|
|
|
|
|
46 |
const reader = new FileReader();
|
47 |
+
reader.onload = function(e) {
|
48 |
+
const fileType = file.name.split('.').pop().toLowerCase();
|
49 |
+
|
50 |
+
if (fileType === 'txt') {
|
51 |
+
extractedText += cleanText(e.target.result) + '\n';
|
52 |
+
} else if (fileType === 'pdf') {
|
53 |
+
extractTextFromPDF(e.target.result).then(text => {
|
54 |
+
extractedText += cleanText(text) + '\n';
|
55 |
+
output.value = extractedText;
|
56 |
+
});
|
57 |
+
} else if (fileType === 'docx') {
|
58 |
+
extractTextFromDOCX(e.target.result).then(text => {
|
59 |
+
extractedText += cleanText(text) + '\n';
|
60 |
+
output.value = extractedText;
|
61 |
+
});
|
62 |
+
}
|
63 |
+
};
|
64 |
|
65 |
+
if (file.type === 'application/pdf') {
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
66 |
reader.readAsArrayBuffer(file);
|
67 |
+
} else {
|
68 |
+
reader.readAsText(file);
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
69 |
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
70 |
});
|
71 |
}
|
72 |
|
73 |
+
// Funci贸n para extraer texto de un archivo PDF
|
74 |
+
function extractTextFromPDF(pdfData) {
|
75 |
+
return new Promise((resolve, reject) => {
|
76 |
+
const loadingTask = pdfjsLib.getDocument({data: pdfData});
|
77 |
+
loadingTask.promise.then(pdf => {
|
78 |
+
let text = '';
|
79 |
+
let pageNumber = 1;
|
80 |
+
|
81 |
+
function extractPageText(pageNum) {
|
82 |
+
pdf.getPage(pageNum).then(page => {
|
83 |
+
page.getTextContent().then(content => {
|
84 |
+
content.items.forEach(item => {
|
85 |
+
text += item.str + ' ';
|
86 |
+
});
|
87 |
+
|
88 |
+
if (pageNum < pdf.numPages) {
|
89 |
+
extractPageText(pageNum + 1);
|
90 |
+
} else {
|
91 |
+
resolve(text);
|
92 |
+
}
|
93 |
+
});
|
94 |
+
});
|
95 |
+
}
|
96 |
|
97 |
+
extractPageText(pageNumber);
|
98 |
+
}, reject);
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
99 |
});
|
100 |
}
|
101 |
|
102 |
+
// Funci贸n para extraer texto de un archivo DOCX
|
103 |
+
function extractTextFromDOCX(docxData) {
|
104 |
+
return new Promise((resolve, reject) => {
|
105 |
+
JSZip.loadAsync(docxData).then(function(zip) {
|
106 |
+
const xmlFile = zip.file("word/document.xml");
|
107 |
+
if (!xmlFile) {
|
108 |
+
reject("Archivo XML no encontrado.");
|
109 |
+
} else {
|
110 |
+
xmlFile.async("string").then(function(xmlText) {
|
111 |
+
const parser = new DOMParser();
|
112 |
+
const xmlDoc = parser.parseFromString(xmlText, "text/xml");
|
113 |
+
let text = '';
|
114 |
+
const paragraphs = xmlDoc.getElementsByTagName('w:t');
|
115 |
+
|
116 |
+
for (let i = 0; i < paragraphs.length; i++) {
|
117 |
+
text += paragraphs[i].textContent + ' ';
|
118 |
+
}
|
119 |
+
resolve(text);
|
120 |
+
});
|
121 |
}
|
122 |
});
|
123 |
+
});
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
124 |
}
|
125 |
|
126 |
+
// Funci贸n para hacer preguntas al modelo de Hugging Face
|
127 |
async function askQuestion() {
|
128 |
const question = document.getElementById('question').value;
|
129 |
+
const context = document.getElementById('output').value;
|
130 |
|
131 |
+
if (!question || !context) {
|
132 |
+
alert("Por favor, aseg煤rate de que hay texto y una pregunta.");
|
|
|
133 |
return;
|
134 |
}
|
135 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
136 |
const data = {
|
137 |
inputs: {
|
138 |
+
question: question,
|
139 |
+
context: context
|
140 |
}
|
141 |
};
|
142 |
+
|
143 |
+
const modelUrl = "https://api-inference.huggingface.co/models/distilbert-base-multilingual-cased";
|
144 |
+
const cucu = window.huggingface.variables["API_KEY_2"];
|
145 |
+
const token = window.huggingface.variables["API_KEY_2"];
|
146 |
console.log("key : " + cucu);
|
147 |
+
const headers = {
|
148 |
+
"Authorization": `Bearer ${token}`,
|
149 |
+
"Content-Type": "application/json"
|
150 |
+
};
|
151 |
+
|
152 |
try {
|
153 |
+
const response = await fetch(modelUrl, {
|
154 |
method: 'POST',
|
155 |
+
headers: headers,
|
|
|
|
|
|
|
156 |
body: JSON.stringify(data)
|
157 |
});
|
158 |
|
|
|
160 |
if (response.ok) {
|
161 |
document.getElementById('response').innerText = result.answer;
|
162 |
} else {
|
163 |
+
document.getElementById('response').innerText = `Error: ${result.error}`;
|
|
|
164 |
}
|
165 |
} catch (error) {
|
166 |
console.error('Error al hacer la consulta:', error);
|
167 |
+
document.getElementById('response').innerText = `Error al procesar la solicitud: ${error.message}`;
|
168 |
}
|
169 |
}
|
170 |
</script>
|