- Allow multiple instances of Evie on 1 website. Shortcode is now parametrized.

- Allow URLs to contain other information than just HTML information. It can alose refer to e.g. PDF-files.

.gitignore

@@ -12,3 +12,32 @@ docker/tenant_files/
 **/.DS_Store
 __pycache__
 **/__pycache__
+/.idea
+*.pyc
+*.pyc
+common/.DS_Store
+common/__pycache__/__init__.cpython-312.pyc
+common/__pycache__/extensions.cpython-312.pyc
+common/models/__pycache__/__init__.cpython-312.pyc
+common/models/__pycache__/document.cpython-312.pyc
+common/models/__pycache__/interaction.cpython-312.pyc
+common/models/__pycache__/user.cpython-312.pyc
+common/utils/.DS_Store
+common/utils/__pycache__/__init__.cpython-312.pyc
+common/utils/__pycache__/celery_utils.cpython-312.pyc
+common/utils/__pycache__/nginx_utils.cpython-312.pyc
+common/utils/__pycache__/security.cpython-312.pyc
+common/utils/__pycache__/simple_encryption.cpython-312.pyc
+common/utils/__pycache__/template_filters.cpython-312.pyc
+config/.DS_Store
+config/__pycache__/__init__.cpython-312.pyc
+config/__pycache__/config.cpython-312.pyc
+config/__pycache__/logging_config.cpython-312.pyc
+eveai_app/.DS_Store
+eveai_app/__pycache__/__init__.cpython-312.pyc
+eveai_app/__pycache__/errors.cpython-312.pyc
+eveai_chat/.DS_Store
+migrations/.DS_Store
+migrations/public/.DS_Store
+scripts/.DS_Store
+scripts/__pycache__/run_eveai_app.cpython-312.pyc

common/models/user.py

@@ -35,10 +35,11 @@ class Tenant(db.Model):
     html_end_tags = db.Column(ARRAY(sa.String(10)), nullable=True, default=['p', 'li'])
     html_included_elements = db.Column(ARRAY(sa.String(50)), nullable=True)
     html_excluded_elements = db.Column(ARRAY(sa.String(50)), nullable=True)
+    html_excluded_classes = db.Column(ARRAY(sa.String(200)), nullable=True)
+
     min_chunk_size = db.Column(db.Integer, nullable=True, default=2000)
     max_chunk_size = db.Column(db.Integer, nullable=True, default=3000)
 
-
     # Embedding search variables
     es_k = db.Column(db.Integer, nullable=True, default=5)
     es_similarity_threshold = db.Column(db.Float, nullable=True, default=0.7)
@@ -80,6 +81,7 @@ class Tenant(db.Model):
             'html_end_tags': self.html_end_tags,
             'html_included_elements': self.html_included_elements,
             'html_excluded_elements': self.html_excluded_elements,
+            'html_excluded_classes': self.html_excluded_classes,
             'min_chunk_size': self.min_chunk_size,
             'max_chunk_size': self.max_chunk_size,
             'es_k': self.es_k,

common/utils/model_utils.py

@@ -86,6 +86,7 @@ def select_model_variables(tenant):
     model_variables['html_end_tags'] = tenant.html_end_tags
     model_variables['html_included_elements'] = tenant.html_included_elements
     model_variables['html_excluded_elements'] = tenant.html_excluded_elements
+    model_variables['html_excluded_classes'] = tenant.html_excluded_classes
 
     # Set Chunk Size variables
     model_variables['min_chunk_size'] = tenant.min_chunk_size
@@ -144,8 +145,12 @@ def select_model_variables(tenant):
                                                        default_headers=portkey_headers)
             tool_calling_supported = False
             match llm_model:
-                case 'gpt-4-turbo' | 'gpt-4o' | 'gpt-4o-mini':
+                case 'gpt-4o' | 'gpt-4o-mini':
                     tool_calling_supported = True
+                    PDF_chunk_size = 10000
+                    PDF_chunk_overlap = 200
+                    PDF_min_chunk_size = 8000
+                    PDF_max_chunk_size = 12000
                 case _:
                     raise Exception(f'Error setting model variables for tenant {tenant.id} '
                                     f'error: Invalid chat model')
@@ -160,10 +165,19 @@ def select_model_variables(tenant):
                                                           model=llm_model_ext,
                                                           temperature=model_variables['RAG_temperature'])
             tool_calling_supported = True
+            PDF_chunk_size = 10000
+            PDF_chunk_overlap = 200
+            PDF_min_chunk_size = 8000
+            PDF_max_chunk_size = 12000
         case _:
             raise Exception(f'Error setting model variables for tenant {tenant.id} '
                             f'error: Invalid chat provider')
 
+    model_variables['PDF_chunk_size'] = PDF_chunk_size
+    model_variables['PDF_chunk_overlap'] = PDF_chunk_overlap
+    model_variables['PDF_min_chunk_size'] = PDF_min_chunk_size
+    model_variables['PDF_max_chunk_size'] = PDF_max_chunk_size
+
     if tool_calling_supported:
         model_variables['cited_answer_cls'] = CitedAnswer
 

config/config.py

@@ -53,7 +53,7 @@ class Config(object):
     WTF_CSRF_CHECK_DEFAULT = False
 
     # file upload settings
-    MAX_CONTENT_LENGTH = 16 * 1024 * 1024
+    MAX_CONTENT_LENGTH = 50 * 1024 * 1024
     UPLOAD_EXTENSIONS = ['.txt', '.pdf', '.png', '.jpg', '.jpeg', '.gif']
 
     # supported languages

config/prompts/openai/gpt-4o.yaml

@@ -15,11 +15,12 @@ html_parse: |
 
 pdf_parse: |
   You are a top administrative aid specialized in transforming given PDF-files into markdown formatted files. The generated files will be used to generate embeddings in a RAG-system.
+  The content you get is already processed (some markdown already generated), but needs to be corrected. For large files, you may receive only portions of the full file. Consider this when processing the content.
 
   # Best practices are:
-  - Respect wordings and language(s) used in the PDF.
+  - Respect wordings and language(s) used in the provided content.
   - The following items need to be considered: headings, paragraphs, listed items (numbered or not) and tables. Images can be neglected.
-  - When headings are numbered, show the numbering and define the header level. 
+  - When headings are numbered, show the numbering and define the header level. You may have to correct current header levels, as preprocessing is known to make errors.
   - A new item is started when a <return> is found before a full line is reached. In order to know the number of characters in a line, please check the document and the context within the document (e.g. an image could limit the number of characters temporarily).
   - Paragraphs are to be stripped of newlines so they become easily readable.
   - Be careful of encoding of the text. Everything needs to be human readable.

eveai_app/views/document_views.py

@@ -12,7 +12,7 @@ from werkzeug.utils import secure_filename
 from sqlalchemy.exc import SQLAlchemyError
 import requests
 from requests.exceptions import SSLError
-from urllib.parse import urlparse
+from urllib.parse import urlparse, unquote
 import io
 from minio.error import S3Error
 
@@ -89,49 +89,93 @@ def add_document():
 def add_url():
     form = AddURLForm()
 
-    # If the form is submitted
     if form.validate_on_submit():
         current_app.logger.info(f'Adding url for tenant {session["tenant"]["id"]}')
         url = form.url.data
 
-        doc_vers = DocumentVersion.query.filter_by(url=url).all()
-        if doc_vers:
-            current_app.logger.info(f'A document with url {url} already exists. No new document created.')
-            flash(f'A document with url {url} already exists. No new document created.', 'info')
+        try:
+            response = requests.head(url, allow_redirects=True)
+            content_type = response.headers.get('Content-Type', '').split(';')[0]
+
+            # Determine file extension based on Content-Type
+            extension = get_extension_from_content_type(content_type)
+
+            # Generate filename
+            parsed_url = urlparse(url)
+            path = unquote(parsed_url.path)
+            filename = os.path.basename(path)
+
+            if not filename or '.' not in filename:
+                # Use the last part of the path or a default name
+                filename = path.strip('/').split('/')[-1] or 'document'
+                filename = secure_filename(f"{filename}.{extension}")
+            else:
+                filename = secure_filename(filename)
+
+            # Check if a document with this URL already exists
+            existing_doc = DocumentVersion.query.filter_by(url=url).first()
+            if existing_doc:
+                flash(f'A document with URL {url} already exists. No new document created.', 'info')
+                return redirect(prefixed_url_for('document_bp.documents'))
+
+            # Download the content
+            response = requests.get(url)
+            response.raise_for_status()
+            file_content = response.content
+
+            # Create document and document version
+            form_dict = form_to_dict(form)
+            new_doc, new_doc_vers = create_document_stack(form_dict, file_content, filename, extension)
+
+            # Upload file to storage
+            minio_client.upload_document_file(
+                session['tenant']['id'],
+                new_doc_vers.doc_id,
+                new_doc_vers.language,
+                new_doc_vers.id,
+                filename,
+                file_content
+            )
+
+            # Start embedding task
+            task = current_celery.send_task('create_embeddings', queue='embeddings', args=[
+                session['tenant']['id'],
+                new_doc_vers.id,
+            ])
+
+            current_app.logger.info(f'Embedding creation started for tenant {session["tenant"]["id"]}, '
+                                    f'Document Version {new_doc_vers.id}. '
+                                    f'Embedding creation task: {task.id}')
+            flash(f'Processing on document {new_doc.name}, version {new_doc_vers.id} started. Task ID: {task.id}.',
+                  'success')
+
             return redirect(prefixed_url_for('document_bp.documents'))
-        # Only when no document with URL exists
-        html = fetch_html(url)
-        file = io.BytesIO(html)
 
-        parsed_url = urlparse(url)
-        path_parts = parsed_url.path.split('/')
-        filename = path_parts[-1]
-        if filename == '':
-            filename = 'index'
-        if not filename.endswith('.html'):
-            filename += '.html'
-        extension = 'html'
-        form_dict = form_to_dict(form)
-
-        new_doc, new_doc_vers = create_document_stack(form_dict, file, filename, extension)
-
-        task = current_celery.send_task('create_embeddings', queue='embeddings', args=[
-            session['tenant']['id'],
-            new_doc_vers.id,
-        ])
-        current_app.logger.info(f'Embedding creation started for tenant {session["tenant"]["id"]}, '
-                                f'Document Version {new_doc_vers.id}. '
-                                f'Embedding creation task: {task.id}')
-        flash(f'Processing on document {new_doc.name}, version {new_doc_vers.id} started. Task ID: {task.id}.',
-              'success')
-
-        return redirect(prefixed_url_for('document_bp.documents'))
-    else:
-        form_validation_failed(request, form)
+        except requests.RequestException as e:
+            current_app.logger.error(f'Error fetching URL {url}: {str(e)}')
+            flash(f'Error fetching URL: {str(e)}', 'danger')
+        except SQLAlchemyError as e:
+            current_app.logger.error(f'Database error: {str(e)}')
+            flash('An error occurred while saving the document.', 'danger')
+        except Exception as e:
+            current_app.logger.error(f'Unexpected error: {str(e)}')
+            flash('An unexpected error occurred.', 'danger')
 
     return render_template('document/add_url.html', form=form)
 
 
+def get_extension_from_content_type(content_type):
+    content_type_map = {
+        'text/html': 'html',
+        'application/pdf': 'pdf',
+        'text/plain': 'txt',
+        'application/msword': 'doc',
+        'application/vnd.openxmlformats-officedocument.wordprocessingml.document': 'docx',
+        # Add more mappings as needed
+    }
+    return content_type_map.get(content_type, 'html')  # Default to 'html' if unknown
+
+
 @document_bp.route('/add_urls', methods=['GET', 'POST'])
 @roles_accepted('Super User', 'Tenant Admin')
 def add_urls():
@@ -358,6 +402,8 @@ def handle_document_version_selection():
 
     action = request.form['action']
 
+    current_app.logger.debug(f'Triggered Document Version Action: {action}')
+
     match action:
         case 'edit_document_version':
             return redirect(prefixed_url_for('document_bp.edit_document_version', document_version_id=doc_vers_id))

eveai_app/views/user_forms.py

@@ -32,6 +32,7 @@ class TenantForm(FlaskForm):
                                 default='p, li')
     html_included_elements = StringField('HTML Included Elements', validators=[Optional()])
     html_excluded_elements = StringField('HTML Excluded Elements', validators=[Optional()])
+    html_excluded_classes = StringField('HTML Excluded Classes', validators=[Optional()])
     min_chunk_size = IntegerField('Minimum Chunk Size (2000)', validators=[NumberRange(min=0), Optional()], default=2000)
     max_chunk_size = IntegerField('Maximum Chunk Size (3000)', validators=[NumberRange(min=0), Optional()], default=3000)
     # Embedding Search variables

eveai_app/views/user_views.py

@@ -68,6 +68,8 @@ def tenant():
             if form.html_included_elements.data else []
         new_tenant.html_excluded_elements = [tag.strip() for tag in form.html_excluded_elements.data.split(',')] \
             if form.html_excluded_elements.data else []
+        new_tenant.html_excluded_classes = [cls.strip() for cls in form.html_excluded_classes.data.split(',')] \
+            if form.html_excluded_classes.data else []
 
         current_app.logger.debug(f'html_tags: {new_tenant.html_tags},'
                                  f'html_end_tags: {new_tenant.html_end_tags},'
@@ -123,6 +125,8 @@ def edit_tenant(tenant_id):
             form.html_included_elements.data = ', '.join(tenant.html_included_elements)
         if tenant.html_excluded_elements:
             form.html_excluded_elements.data = ', '.join(tenant.html_excluded_elements)
+        if tenant.html_excluded_classes:
+            form.html_excluded_classes.data = ', '.join(tenant.html_excluded_classes)
 
     if form.validate_on_submit():
         # Populate the tenant with form data
@@ -134,6 +138,8 @@ def edit_tenant(tenant_id):
                                          elem.strip()]
         tenant.html_excluded_elements = [elem.strip() for elem in form.html_excluded_elements.data.split(',') if
                                          elem.strip()]
+        tenant.html_excluded_classes = [elem.strip() for elem in form.html_excluded_classes.data.split(',') if
+                                        elem.strip()]
 
         db.session.commit()
         flash('Tenant updated successfully.', 'success')

eveai_workers/Processors/PDF_Processor.py

@@ -0,0 +1,271 @@
+import io
+import pdfplumber
+from flask import current_app
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+from langchain_core.output_parsers import StrOutputParser
+from langchain_core.prompts import ChatPromptTemplate
+import re
+
+from langchain_core.runnables import RunnablePassthrough
+
+from common.extensions import minio_client
+from common.utils.model_utils import create_language_template
+
+
+class PDFProcessor:
+    def __init__(self, tenant, model_variables, document_version):
+        self.tenant = tenant
+        self.model_variables = model_variables
+        self.document_version = document_version
+
+        # Configuration parameters from model_variables
+        self.chunk_size = model_variables['PDF_chunk_size']
+        self.chunk_overlap = model_variables['PDF_chunk_overlap']
+        self.min_chunk_size = model_variables['PDF_min_chunk_size']
+        self.max_chunk_size = model_variables['PDF_max_chunk_size']
+
+        # Set tuning variable for easy use
+        self.embed_tuning = model_variables['embed_tuning']
+
+    def process_pdf(self):
+        self._log("Starting PDF processing")
+        try:
+            file_data = minio_client.download_document_file(
+                self.tenant.id,
+                self.document_version.doc_id,
+                self.document_version.language,
+                self.document_version.id,
+                self.document_version.file_name
+            )
+
+            extracted_content = self._extract_content(file_data)
+            structured_content, title = self._structure_content(extracted_content)
+
+            llm_chunks = self._split_content_for_llm(structured_content)
+            markdown = self._process_chunks_with_llm(llm_chunks)
+
+            self._save_markdown(markdown)
+            self._log("Finished processing PDF")
+            return markdown, title
+        except Exception as e:
+            self._log(f"Error processing PDF: {str(e)}", level='error')
+            raise
+
+    def _log(self, message, level='debug'):
+        logger = current_app.logger
+        log_method = getattr(logger, level)
+        log_method(f"PDFProcessor - Tenant {self.tenant.id}, Document {self.document_version.id}: {message}")
+
+    def _extract_content(self, file_data):
+        extracted_content = []
+        with pdfplumber.open(io.BytesIO(file_data)) as pdf:
+            figure_counter = 1
+            for page_num, page in enumerate(pdf.pages):
+                self._log(f"Extracting content from page {page_num + 1}")
+                page_content = {
+                    'text': page.extract_text(),
+                    'figures': self._extract_figures(page, page_num, figure_counter),
+                    'tables': self._extract_tables(page)
+                }
+                if self.embed_tuning:
+                    self._log(f'Extracted PDF Content for page {page_num + 1}')
+                    self._log(f"{page_content }")
+                figure_counter += len(page_content['figures'])
+                extracted_content.append(page_content)
+
+            # if self.embed_tuning:
+            #     current_app.embed_tuning_logger.debug(f'Extracted PDF Content')
+            #     current_app.embed_tuning_logger.debug(f'---------------------')
+            #     current_app.embed_tuning_logger.debug(f'Page: {page_content}')
+            #     current_app.embed_tuning_logger.debug(f'End of Extracted PDF Content')
+            #     current_app.embed_tuning_logger.debug(f'----------------------------')
+
+        return extracted_content
+
+    def _extract_figures(self, page, page_num, figure_counter):
+        figures = []
+        # Omit figure processing for now!
+        # for img in page.images:
+        #     try:
+        #         # Try to get the bbox, use full page dimensions if not available
+        #         bbox = img.get('bbox', (0, 0, page.width, page.height))
+        #
+        #         figure = {
+        #             'figure_number': figure_counter,
+        #             'filename': f"figure_{page_num + 1}_{figure_counter}.png",
+        #             'caption': self._find_figure_caption(page, bbox)
+        #         }
+        #
+        #         # Extract the figure as an image
+        #         figure_image = page.within_bbox(bbox).to_image()
+        #
+        #         # Save the figure using MinIO
+        #         with io.BytesIO() as output:
+        #             figure_image.save(output, format='PNG')
+        #             output.seek(0)
+        #             minio_client.upload_document_file(
+        #                 self.tenant.id,
+        #                 self.document_version.doc_id,
+        #                 self.document_version.language,
+        #                 self.document_version.id,
+        #                 figure['filename'],
+        #                 output.getvalue()
+        #             )
+        #
+        #         figures.append(figure)
+        #         figure_counter += 1
+        #     except Exception as e:
+        #         self._log(f"Error processing figure on page {page_num + 1}: {str(e)}", level='error')
+
+        return figures
+
+    def _find_figure_caption(self, page, bbox):
+        try:
+            # Look for text below the figure
+            caption_bbox = (bbox[0], bbox[3], bbox[2], min(bbox[3] + 50, page.height))
+            caption_text = page.crop(caption_bbox).extract_text()
+            if caption_text and caption_text.lower().startswith('figure'):
+                return caption_text
+        except Exception as e:
+            self._log(f"Error finding figure caption: {str(e)}", level='error')
+        return None
+
+    def _extract_tables(self, page):
+        tables = []
+        try:
+            for table in page.extract_tables():
+                if table:
+                    markdown_table = self._table_to_markdown(table)
+                    if markdown_table:  # Only add non-empty tables
+                        tables.append(markdown_table)
+        except Exception as e:
+            self._log(f"Error extracting tables from page: {str(e)}", level='error')
+        return tables
+
+    def _table_to_markdown(self, table):
+        if not table or not table[0]:  # Check if table is empty or first row is empty
+            return ""  # Return empty string for empty tables
+
+        def clean_cell(cell):
+            if cell is None:
+                return ""  # Convert None to empty string
+            return str(cell).replace("|", "\\|")  # Escape pipe characters and convert to string
+
+        header = [clean_cell(cell) for cell in table[0]]
+        markdown = "| " + " | ".join(header) + " |\n"
+        markdown += "| " + " | ".join(["---"] * len(header)) + " |\n"
+
+        for row in table[1:]:
+            cleaned_row = [clean_cell(cell) for cell in row]
+            markdown += "| " + " | ".join(cleaned_row) + " |\n"
+
+        return markdown
+
+    def _structure_content(self, extracted_content):
+        structured_content = ""
+        title = "Untitled Document"
+        current_heading_level = 0
+        heading_pattern = re.compile(r'^(\d+(\.\d+)*\.?\s*)?(.+)$')
+
+        def identify_heading(text):
+            match = heading_pattern.match(text.strip())
+            if match:
+                numbering, _, content = match.groups()
+                if numbering:
+                    level = numbering.count('.') + 1
+                    return level, f"{numbering}{content}"
+                else:
+                    return 1, content  # Assume it's a top-level heading if no numbering
+            return 0, text  # Not a heading
+
+        for page in extracted_content:
+            # Assume the title is on the first page
+            if page == extracted_content[0]:
+                lines = page.get('text', '').split('\n')
+                if lines:
+                    title = lines[0].strip()  # Use the first non-empty line as the title
+
+            # Process text
+            paragraphs = page['text'].split('\n\n')
+
+            for para in paragraphs:
+                lines = para.strip().split('\n')
+                if len(lines) == 1:  # Potential heading
+                    level, text = identify_heading(lines[0])
+                    if level > 0:
+                        heading_marks = '#' * level
+                        structured_content += f"\n\n{heading_marks} {text}\n\n"
+                        if level == 1 and not title:
+                            title = text  # Use the first top-level heading as the title if not set
+                    else:
+                        structured_content += f"{para}\n\n"  # Treat as normal paragraph
+                else:
+                    structured_content += f"{para}\n\n"  # Multi-line paragraph
+
+            # Process figures
+            for figure in page.get('figures', []):
+                structured_content += f"\n\n![Figure {figure['figure_number']}]({figure['filename']})\n\n"
+                if figure['caption']:
+                    structured_content += f"*Figure {figure['figure_number']}: {figure['caption']}*\n\n"
+
+            # Add tables
+            if 'tables' in page:
+                for table in page['tables']:
+                    structured_content += f"\n{table}\n"
+
+        if self.embed_tuning:
+            self._save_intermediate(structured_content, "structured_content.md")
+
+        return structured_content, title
+
+    def _split_content_for_llm(self, content):
+        text_splitter = RecursiveCharacterTextSplitter(
+            chunk_size=self.chunk_size,
+            chunk_overlap=self.chunk_overlap,
+            length_function=len,
+            separators=["\n\n", "\n", " ", ""]
+        )
+        return text_splitter.split_text(content)
+
+    def _process_chunks_with_llm(self, chunks):
+        llm = self.model_variables['llm']
+        template = self.model_variables['pdf_parse_template']
+        pdf_prompt = ChatPromptTemplate.from_template(template)
+        setup = RunnablePassthrough()
+        output_parser = StrOutputParser()
+        chain = setup | pdf_prompt | llm | output_parser
+
+        markdown_chunks = []
+        for chunk in chunks:
+            input = {"pdf_content": chunk}
+            result = chain.invoke(input)
+            # Remove Markdown code block delimiters if present
+            result = result.strip()
+            if result.startswith("```markdown"):
+                result = result[len("```markdown"):].strip()
+            if result.endswith("```"):
+                result = result[:-3].strip()
+            markdown_chunks.append(result)
+
+        return "\n\n".join(markdown_chunks)
+
+    def _save_markdown(self, markdown):
+        markdown_filename = f"{self.document_version.id}.md"
+        minio_client.upload_document_file(
+            self.tenant.id,
+            self.document_version.doc_id,
+            self.document_version.language,
+            self.document_version.id,
+            markdown_filename,
+            markdown.encode('utf-8')
+        )
+
+    def _save_intermediate(self, content, filename):
+        minio_client.upload_document_file(
+            self.tenant.id,
+            self.document_version.doc_id,
+            self.document_version.language,
+            self.document_version.id,
+            filename,
+            content.encode('utf-8')
+        )

eveai_workers/tasks.py

@@ -3,7 +3,6 @@ import os
 from datetime import datetime as dt, timezone as tz
 import subprocess
 
-
 import gevent
 from bs4 import BeautifulSoup
 import html
@@ -12,6 +11,7 @@ from flask import current_app
 # OpenAI imports
 from langchain.chains.summarize import load_summarize_chain
 from langchain.text_splitter import CharacterTextSplitter, MarkdownHeaderTextSplitter
+from langchain_core.documents import Document
 from langchain_core.exceptions import LangChainException
 from langchain_core.output_parsers import StrOutputParser
 from langchain_core.prompts import ChatPromptTemplate
@@ -29,6 +29,7 @@ from common.utils.celery_utils import current_celery
 from common.utils.database import Database
 from common.utils.model_utils import select_model_variables, create_language_template
 from common.utils.os_utils import safe_remove, sync_folder
+from eveai_workers.Processors.PDF_Processor import PDFProcessor
 
 
 @current_celery.task(name='create_embeddings', queue='embeddings')
@@ -103,33 +104,67 @@ def create_embeddings(tenant_id, document_version_id):
         raise
 
 
+# def process_pdf(tenant, model_variables, document_version):
+#     file_data = minio_client.download_document_file(tenant.id, document_version.doc_id, document_version.language,
+#                                                     document_version.id, document_version.file_name)
+#
+#     pdf_text = ''
+#     pdf_reader = PyPDF2.PdfReader(io.BytesIO(file_data))
+#     for page in pdf_reader.pages:
+#         pdf_text += page.extract_text()
+#
+#     markdown = generate_markdown_from_pdf(tenant, model_variables, document_version, pdf_text)
+#     markdown_file_name = f'{document_version.id}.md'
+#     minio_client.upload_document_file(tenant.id, document_version.doc_id, document_version.language,
+#                                       document_version.id,
+#                                       markdown_file_name, markdown.encode())
+#
+#     potential_chunks = create_potential_chunks_for_markdown(tenant.id, document_version, markdown_file_name)
+#     chunks = combine_chunks_for_markdown(potential_chunks, model_variables['min_chunk_size'],
+#                                          model_variables['max_chunk_size'])
+#
+#     if len(chunks) > 1:
+#         summary = summarize_chunk(tenant, model_variables, document_version, chunks[0])
+#         document_version.system_context = f'Summary: {summary}\n'
+#     else:
+#         document_version.system_context = ''
+#
+#     enriched_chunks = enrich_chunks(tenant, document_version, chunks)
+#     embeddings = embed_chunks(tenant, model_variables, document_version, enriched_chunks)
+#
+#     try:
+#         db.session.add(document_version)
+#         document_version.processing_finished_at = dt.now(tz.utc)
+#         document_version.processing = False
+#         db.session.add_all(embeddings)
+#         db.session.commit()
+#     except SQLAlchemyError as e:
+#         current_app.logger.error(f'Error saving embedding information for tenant {tenant.id} '
+#                                  f'on HTML, document version {document_version.id}'
+#                                  f'error: {e}')
+#         raise
+#
+#     current_app.logger.info(f'Embeddings created successfully for tenant {tenant.id} '
+#                             f'on document version {document_version.id} :-)')
+
 def process_pdf(tenant, model_variables, document_version):
-    file_data = minio_client.download_document_file(tenant.id, document_version.doc_id, document_version.language,
-                                       document_version.id, document_version.file_name)
+    processor = PDFProcessor(tenant, model_variables, document_version)
+    markdown, title = processor.process_pdf()
 
-    pdf_text = ''
-    pdf_reader = PyPDF2.PdfReader(io.BytesIO(file_data))
-    for page in pdf_reader.pages:
-        pdf_text += page.extract_text()
+    # Create potential chunks for embedding
+    potential_chunks = create_potential_chunks_for_markdown(tenant.id, document_version, f"{document_version.id}.md")
 
-    markdown = generate_markdown_from_pdf(tenant, model_variables, document_version, pdf_text)
-    markdown_file_name = f'{document_version.id}.md'
-    minio_client.upload_document_file(tenant.id, document_version.doc_id, document_version.language, document_version.id,
-                         markdown_file_name, markdown.encode())
-
-    potential_chunks = create_potential_chunks_for_markdown(tenant.id, document_version, markdown_file_name)
+    # Combine chunks for embedding
     chunks = combine_chunks_for_markdown(potential_chunks, model_variables['min_chunk_size'],
                                          model_variables['max_chunk_size'])
 
-    if len(chunks) > 1:
-        summary = summarize_chunk(tenant, model_variables, document_version, chunks[0])
-        document_version.system_context = f'Summary: {summary}\n'
-    else:
-        document_version.system_context = ''
+    # Enrich chunks
+    enriched_chunks = enrich_chunks(tenant, document_version, title, chunks)
 
-    enriched_chunks = enrich_chunks(tenant, document_version, chunks)
+    # Create embeddings
     embeddings = embed_chunks(tenant, model_variables, document_version, enriched_chunks)
 
+    # Update document version and save embeddings
     try:
         db.session.add(document_version)
         document_version.processing_finished_at = dt.now(tz.utc)
@@ -138,7 +173,7 @@ def process_pdf(tenant, model_variables, document_version):
         db.session.commit()
     except SQLAlchemyError as e:
         current_app.logger.error(f'Error saving embedding information for tenant {tenant.id} '
-                                 f'on HTML, document version {document_version.id}'
+                                 f'on PDF, document version {document_version.id}'
                                  f'error: {e}')
         raise
 
@@ -160,7 +195,7 @@ def delete_embeddings_for_document_version(document_version):
 
 def process_html(tenant, model_variables, document_version):
     file_data = minio_client.download_document_file(tenant.id, document_version.doc_id, document_version.language,
-                                       document_version.id, document_version.file_name)
+                                                    document_version.id, document_version.file_name)
     html_content = file_data.decode('utf-8')
 
     # The tags to be considered can be dependent on the tenant
@@ -173,13 +208,15 @@ def process_html(tenant, model_variables, document_version):
                                        excluded_elements=html_excluded_elements)
 
     extracted_file_name = f'{document_version.id}-extracted.html'
-    minio_client.upload_document_file(tenant.id, document_version.doc_id, document_version.language, document_version.id,
-                         extracted_file_name, extracted_html.encode())
+    minio_client.upload_document_file(tenant.id, document_version.doc_id, document_version.language,
+                                      document_version.id,
+                                      extracted_file_name, extracted_html.encode())
 
     markdown = generate_markdown_from_html(tenant, model_variables, document_version, extracted_html)
     markdown_file_name = f'{document_version.id}.md'
-    minio_client.upload_document_file(tenant.id, document_version.doc_id, document_version.language, document_version.id,
-                         markdown_file_name, markdown.encode())
+    minio_client.upload_document_file(tenant.id, document_version.doc_id, document_version.language,
+                                      document_version.id,
+                                      markdown_file_name, markdown.encode())
 
     potential_chunks = create_potential_chunks_for_markdown(tenant.id, document_version, markdown_file_name)
     chunks = combine_chunks_for_markdown(potential_chunks, model_variables['min_chunk_size'],
@@ -236,18 +273,73 @@ def enrich_chunks(tenant, document_version, title, chunks):
 
 
 def generate_markdown_from_html(tenant, model_variables, document_version, html_content):
-    current_app.logger.debug(f'Generating Markdown from HTML for tenant {tenant.id} '
+    current_app.logger.debug(f'Generating markdown from HTML for tenant {tenant.id} '
                              f'on document version {document_version.id}')
+
     llm = model_variables['llm']
     template = model_variables['html_parse_template']
     parse_prompt = ChatPromptTemplate.from_template(template)
     setup = RunnablePassthrough()
     output_parser = StrOutputParser()
-
     chain = setup | parse_prompt | llm | output_parser
-    input_html = {"html": html_content}
 
-    markdown = chain.invoke(input_html)
+    soup = BeautifulSoup(html_content, 'lxml')
+
+    def split_content(soup, max_size=20000):
+        chunks = []
+        current_chunk = []
+        current_size = 0
+
+        for element in soup.find_all(['h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'p', 'div', 'span', 'table']):
+            element_html = str(element)
+            element_size = len(element_html)
+
+            if current_size + element_size > max_size and current_chunk:
+                chunks.append(''.join(map(str, current_chunk)))
+                current_chunk = []
+                current_size = 0
+
+            current_chunk.append(element)
+            current_size += element_size
+
+            if element.name in ['h1', 'h2', 'h3'] and current_size > max_size:
+                chunks.append(''.join(map(str, current_chunk)))
+                current_chunk = []
+                current_size = 0
+
+        if current_chunk:
+            chunks.append(''.join(map(str, current_chunk)))
+
+        return chunks
+
+    chunks = split_content(soup)
+
+    markdown_chunks = []
+
+    for chunk in chunks:
+        current_app.logger.debug(f'Processing chunk to generate markdown from HTML for tenant {tenant.id} '
+                                 f'on document version {document_version.id}')
+        if tenant.embed_tuning:
+            current_app.embed_tuning_logger.debug(f'Processing chunk: \n '
+                                                  f'------------------\n'
+                                                  f'{chunk}\n'
+                                                  f'------------------\n')
+        input_html = {"html": chunk}
+        markdown_chunk = chain.invoke(input_html)
+        markdown_chunks.append(markdown_chunk)
+        if tenant.embed_tuning:
+            current_app.embed_tuning_logger.debug(f'Processed markdown chunk: \n '
+                                                  f'-------------------------\n'
+                                                  f'{markdown_chunk}\n'
+                                                  f'-------------------------\n')
+        current_app.logger.debug(f'Finished processing chunk to generate markdown from HTML for tenant {tenant.id} '
+                                 f'on document version {document_version.id}')
+
+    # Combine all markdown chunks
+    markdown = "\n\n".join(markdown_chunks)
+
+    current_app.logger.debug(f'Finished generating markdown from HTML for tenant {tenant.id} '
+                             f'on document version {document_version.id}')
 
     return markdown
 
@@ -324,36 +416,73 @@ def embed_chunks(tenant, model_variables, document_version, chunks):
 
 
 def parse_html(tenant, html_content, tags, included_elements=None, excluded_elements=None):
+    current_app.logger.debug(f'Parsing HTML for tenant {tenant.id}')
     soup = BeautifulSoup(html_content, 'html.parser')
     extracted_html = ''
+    excluded_classes = parse_excluded_classes(tenant.html_excluded_classes)
 
     if included_elements:
         elements_to_parse = soup.find_all(included_elements)
     else:
-        elements_to_parse = [soup]  # parse the entire document if no included_elements specified
+        elements_to_parse = [soup]
 
+    log_parsing_info(tenant, tags, included_elements, excluded_elements, excluded_classes, elements_to_parse)
+
+    for element in elements_to_parse:
+        for sub_element in element.find_all(tags):
+            if should_exclude_element(sub_element, excluded_elements, excluded_classes):
+                continue
+            extracted_html += extract_element_content(sub_element)
+
+    title = soup.find('title').get_text(strip=True) if soup.find('title') else ''
+
+    current_app.logger.debug(f'Finished parsing HTML for tenant {tenant.id}')
+
+    return extracted_html, title
+
+
+def parse_excluded_classes(excluded_classes):
+    parsed = {}
+    for rule in excluded_classes:
+        element, cls = rule.split('.', 1)
+        parsed.setdefault(element, set()).add(cls)
+    return parsed
+
+
+def should_exclude_element(element, excluded_elements, excluded_classes):
+    if excluded_elements and element.find_parent(excluded_elements):
+        return True
+    return is_element_excluded_by_class(element, excluded_classes)
+
+
+def is_element_excluded_by_class(element, excluded_classes):
+    for parent in element.parents:
+        if element_matches_exclusion(parent, excluded_classes):
+            return True
+    return element_matches_exclusion(element, excluded_classes)
+
+
+def element_matches_exclusion(element, excluded_classes):
+    if '*' in excluded_classes and any(cls in excluded_classes['*'] for cls in element.get('class', [])):
+        return True
+    return element.name in excluded_classes and \
+        any(cls in excluded_classes[element.name] for cls in element.get('class', []))
+
+
+def extract_element_content(element):
+    content = ' '.join(child.strip() for child in element.stripped_strings)
+    return f'<{element.name}>{content}</{element.name}>\n'
+
+
+def log_parsing_info(tenant, tags, included_elements, excluded_elements, excluded_classes, elements_to_parse):
     if tenant.embed_tuning:
         current_app.embed_tuning_logger.debug(f'Tags to parse: {tags}')
         current_app.embed_tuning_logger.debug(f'Included Elements: {included_elements}')
-        current_app.embed_tuning_logger.debug(f'Included Elements: {len(included_elements)}')
         current_app.embed_tuning_logger.debug(f'Excluded Elements: {excluded_elements}')
+        current_app.embed_tuning_logger.debug(f'Excluded Classes: {excluded_classes}')
         current_app.embed_tuning_logger.debug(f'Found {len(elements_to_parse)} elements to parse')
         current_app.embed_tuning_logger.debug(f'First element to parse: {elements_to_parse[0]}')
 
-    # Iterate through the found included elements
-    for element in elements_to_parse:
-        # Find all specified tags within each included element
-        for sub_element in element.find_all(tags):
-            if tenant.embed_tuning:
-                current_app.embed_tuning_logger.debug(f'Found element: {sub_element.name}')
-            if excluded_elements and sub_element.find_parent(excluded_elements):
-                continue  # Skip this sub_element if it's within any of the excluded_elements
-            extracted_html += f'<{sub_element.name}>{sub_element.get_text(strip=True)}</{sub_element.name}>\n'
-
-    title = soup.find('title').get_text(strip=True)
-
-    return extracted_html, title
-
 
 def process_youtube(tenant, model_variables, document_version):
     base_path = os.path.join(current_app.config['UPLOAD_FOLDER'],
@@ -414,8 +543,9 @@ def download_youtube(url, tenant_id, document_version, file_name):
             with open(temp_file.name, 'rb') as f:
                 file_data = f.read()
 
-        minio_client.upload_document_file(tenant_id, document_version.doc_id, document_version.language, document_version.id,
-                             file_name, file_data)
+        minio_client.upload_document_file(tenant_id, document_version.doc_id, document_version.language,
+                                          document_version.id,
+                                          file_name, file_data)
 
         current_app.logger.info(f'Downloaded YouTube video: {url} for tenant: {tenant_id}')
         return file_name, yt.title, yt.description, yt.author
@@ -429,7 +559,7 @@ def compress_audio(tenant_id, document_version, input_file, output_file):
         current_app.logger.info(f'Compressing audio for tenant: {tenant_id}')
 
         input_data = minio_client.download_document_file(tenant_id, document_version.doc_id, document_version.language,
-                                            document_version.id, input_file)
+                                                         document_version.id, input_file)
 
         with tempfile.NamedTemporaryFile(delete=False, suffix='.mp4') as temp_input:
             temp_input.write(input_data)
@@ -448,8 +578,9 @@ def compress_audio(tenant_id, document_version, input_file, output_file):
                 with open(temp_output.name, 'rb') as f:
                     compressed_data = f.read()
 
-        minio_client.upload_document_file(tenant_id, document_version.doc_id, document_version.language, document_version.id,
-                             output_file, compressed_data)
+        minio_client.upload_document_file(tenant_id, document_version.doc_id, document_version.language,
+                                          document_version.id,
+                                          output_file, compressed_data)
 
         current_app.logger.info(f'Compressed audio for tenant: {tenant_id}')
     except Exception as e:
@@ -465,7 +596,7 @@ def transcribe_audio(tenant_id, document_version, input_file, output_file, model
 
         # Download the audio file from MinIO
         audio_data = minio_client.download_document_file(tenant_id, document_version.doc_id, document_version.language,
-                                            document_version.id, input_file)
+                                                         document_version.id, input_file)
 
         # Load the audio data into pydub
         audio = AudioSegment.from_mp3(io.BytesIO(audio_data))
@@ -648,7 +779,4 @@ def combine_chunks_for_markdown(potential_chunks, min_chars, max_chars):
         actual_chunks.append(current_chunk)
 
     return actual_chunks
-    pass
-
-
 

integrations/Wordpress/eveai-chat-widget/eveai-chat_plugin.php

@@ -3,7 +3,7 @@
 Plugin Name: EveAI Chat Widget
 Plugin URI: https://askeveai.com/
 Description: Integrates the EveAI chat interface into your WordPress site.
-Version: 1.3.21
+Version: 1.4.1
 Author: Josako, Pieter Laroy
 Author URI: https://askeveai.com/about/
 */
@@ -18,15 +18,28 @@ function eveai_chat_enqueue_scripts() {
     wp_enqueue_style('eveai-chat-style', plugin_dir_url(__FILE__) . 'css/eveai-chat-style.css');
 }
 add_action('wp_enqueue_scripts', 'eveai_chat_enqueue_scripts');
+add_action('admin_enqueue_scripts', 'eveai_chat_enqueue_scripts');
 
 // Shortcode function
 function eveai_chat_shortcode($atts) {
-    $options = get_option('eveai_chat_options');
-    $tenant_id = esc_js($options['tenant_id']);
-    $api_key = esc_js($options['api_key']);
-    $domain = esc_js($options['domain']);
-    $language = esc_js($options['language']);
-    $supported_languages = esc_js($options['supported_languages']);
+    // Default values
+    $defaults = array(
+        'tenant_id' => '',
+        'api_key' => '',
+        'domain' => '',
+        'language' => 'en',
+        'supported_languages' => 'en,fr,de,es'
+    );
+
+    // Merge provided attributes with defaults
+    $atts = shortcode_atts($defaults, $atts, 'eveai_chat');
+
+    // Sanitize inputs
+    $tenant_id = sanitize_text_field($atts['tenant_id']);
+    $api_key = sanitize_text_field($atts['api_key']);
+    $domain = esc_url_raw($atts['domain']);
+    $language = sanitize_text_field($atts['language']);
+    $supported_languages = sanitize_text_field($atts['supported_languages']);
 
     // Generate a unique ID for this instance of the chat widget
     $chat_id = 'chat-container-' . uniqid();
@@ -49,80 +62,3 @@ function eveai_chat_shortcode($atts) {
 }
 add_shortcode('eveai_chat', 'eveai_chat_shortcode');
 
-// Add admin menu
-function eveai_chat_admin_menu() {
-    add_options_page('EveAI Chat Settings', 'EveAI Chat', 'manage_options', 'eveai-chat-settings', 'eveai_chat_settings_page');
-}
-add_action('admin_menu', 'eveai_chat_admin_menu');
-
-// Settings page
-function eveai_chat_settings_page() {
-    ?>
-    <div class="wrap">
-        <h1>EveAI Chat Settings</h1>
-        <form method="post" action="options.php">
-            <?php
-            settings_fields('eveai_chat_options');
-            do_settings_sections('eveai-chat-settings');
-            submit_button();
-            ?>
-        </form>
-    </div>
-    <?php
-}
-
-// Register settings
-function eveai_chat_register_settings() {
-    register_setting('eveai_chat_options', 'eveai_chat_options', 'eveai_chat_options_validate');
-
-    add_settings_section('eveai_chat_main', 'Main Settings', 'eveai_chat_section_text', 'eveai-chat-settings');
-
-    add_settings_field('eveai_chat_tenant_id', 'Tenant ID', 'eveai_chat_tenant_id_input', 'eveai-chat-settings', 'eveai_chat_main');
-    add_settings_field('eveai_chat_api_key', 'API Key', 'eveai_chat_api_key_input', 'eveai-chat-settings', 'eveai_chat_main');
-    add_settings_field('eveai_chat_domain', 'Domain', 'eveai_chat_domain_input', 'eveai-chat-settings', 'eveai_chat_main');
-    add_settings_field('eveai_chat_language', 'Default Language', 'eveai_chat_language_input', 'eveai-chat-settings', 'eveai_chat_main');
-    add_settings_field('eveai_chat_supported_languages', 'Supported Languages', 'eveai_chat_supported_languages_input', 'eveai-chat-settings', 'eveai_chat_main');
-}
-add_action('admin_init', 'eveai_chat_register_settings');
-
-function eveai_chat_section_text() {
-    echo '<p>Enter your EveAI Chat configuration details below:</p>';
-}
-
-function eveai_chat_tenant_id_input() {
-    $options = get_option('eveai_chat_options');
-    echo "<input id='eveai_chat_tenant_id' name='eveai_chat_options[tenant_id]' type='text' value='" . esc_attr($options['tenant_id']) . "' />";
-}
-
-function eveai_chat_api_key_input() {
-    $options = get_option('eveai_chat_options');
-    echo "<input id='eveai_chat_api_key' name='eveai_chat_options[api_key]' type='password' value='" . esc_attr($options['api_key']) . "' />";
-}
-
-function eveai_chat_domain_input() {
-    $options = get_option('eveai_chat_options');
-    echo "<input id='eveai_chat_domain' name='eveai_chat_options[domain]' type='text' value='" . esc_attr($options['domain']) . "' />";
-}
-
-function eveai_chat_language_input() {
-    $options = get_option('eveai_chat_options');
-    echo "<input id='eveai_chat_language' name='eveai_chat_options[language]' type='text' value='" . esc_attr($options['language']) . "' />";
-}
-
-function eveai_chat_supported_languages_input() {
-    $options = get_option('eveai_chat_options');
-    $supported_languages = isset($options['supported_languages']) ? $options['supported_languages'] : 'en,fr,de,es';
-    echo "<input id='eveai_chat_supported_languages' name='eveai_chat_options[supported_languages]' type='text' value='" . esc_attr($supported_languages) . "' />";
-    echo "<p class='description'>Enter comma-separated language codes (e.g., en,fr,de,es)</p>";
-}
-
-
-function eveai_chat_options_validate($input) {
-    $new_input = array();
-    $new_input['tenant_id'] = sanitize_text_field($input['tenant_id']);
-    $new_input['api_key'] = sanitize_text_field($input['api_key']);
-    $new_input['domain'] = esc_url_raw($input['domain']);
-    $new_input['language'] = sanitize_text_field($input['language']);
-    $new_input['supported_languages'] = sanitize_text_field($input['supported_languages']);
-    return $new_input;
-}

integrations/Wordpress/eveai-chat-widget/js/eveai-chat-widget.js

@@ -161,24 +161,32 @@ class EveAIChatWidget extends HTMLElement {
 
     this.socket.on('connect', (data) => {
       console.log('Socket connected OK');
+      console.log('Connect event data:', data);
+      console.log('Connect event this:', this);
       this.setStatusMessage('Connected to EveAI.');
       this.updateConnectionStatus(true);
       this.startHeartbeat();
-      if (data.room) {
+      if (data && data.room) {
         this.room = data.room;
         console.log(`Joined room: ${this.room}`);
+      } else {
+        console.log('Room information not received on connect');
       }
     });
 
     this.socket.on('authenticated', (data) => {
-      console.log('Authenticated event received: ', data);
+      console.log('Authenticated event received');
+      console.log('Authentication event data:', data);
+      console.log('Authentication event this:', this);
       this.setStatusMessage('Authenticated.');
-      if (data.token) {
-        this.jwtToken = data.token; // Store the JWT token received from the server
+      if (data && data.token) {
+        this.jwtToken = data.token;
       }
-      if (data.room) {
+      if (data && data.room) {
         this.room = data.room;
         console.log(`Confirmed room: ${this.room}`);
+      } else {
+        console.log('Room information not received on authentication');
       }
     });
 

integrations/Wordpress/eveai-chat-widget/readme.txt

@@ -3,7 +3,7 @@ Contributors: Josako
 Tags: chat, ai
 Requires at least: 5.0
 Tested up to: 5.9
-Stable tag: 1.3.0
+Stable tag: 1.4.1
 License: GPLv2 or later
 License URI: http://www.gnu.org/licenses/gpl-2.0.html
 
@@ -17,7 +17,18 @@ This plugin allows you to easily add the EveAI chat widget to your WordPress sit
 
 1. Upload the `eveai-chat-widget` folder to the `/wp-content/plugins/` directory
 2. Activate the plugin through the 'Plugins' menu in WordPress
-3. Go to Settings > EveAI Chat to configure your chat widget parameters
+3. Add EveAI Chat Widget to your page or post using the instructions below.
+
+== Usage ==
+
+To add an EveAI Chat Widget to your page or post, use the following shortcode:
+
+[eveai_chat tenant_id="YOUR_TENANT_ID" api_key="YOUR_API_KEY" domain="YOUR_DOMAIN" language="LANGUAGE_CODE" supported_languages="COMMA_SEPARATED_LANGUAGE_CODES"]
+
+Example:
+[eveai_chat tenant_id="123456" api_key="your_api_key_here" domain="https://your-domain.com" language="en" supported_languages="en,fr,de,es"]
+
+You can add multiple chat widgets with different configurations by using the shortcode multiple times with different parameters.
 
 == Frequently Asked Questions ==
 
@@ -27,6 +38,13 @@ Contact your EveAI service provider to obtain your Tenant ID, API Key, and Domai
 
 == Changelog ==
 
+= 1.4.1 - 1.4...=
+* Bug fixes
+
+= 1.4.0 =
+* Allow for multiple instances of Evie on the same website
+* Parametrization of the shortcode
+
 = 1.3.3 -  =
 * ensure all attributes (also height and supportedLanguages) are set before initializing the socket
 * Bugfixing

nginx/nginx.conf

@@ -32,7 +32,7 @@ http {
     #keepalive_timeout  0;
     keepalive_timeout  65;
 
-    client_max_body_size 16M;
+    client_max_body_size 50M;
 
     #gzip  on;
 

nginx/static/assets/css/eveai.css

@@ -386,6 +386,7 @@ input[type="radio"] {
 .btn-danger:hover {
     background-color: darken(var(--bs-danger), 10%) !important; /* Darken the background on hover */
     border-color: darken(var(--bs-danger), 10%) !important; /* Darken the border on hover */
+    color: var(--bs-white) !important; /* Ensure the text remains white and readable */
 }
 
 /* Success Alert Styling */

requirements.txt

@@ -43,7 +43,6 @@ pgvector~=0.2.5
 pycryptodome~=3.20.0
 pydantic~=2.7.4
 PyJWT~=2.8.0
-pypdf~=4.2.0
 PySocks~=1.7.1
 python-dateutil~=2.9.0.post0
 python-engineio~=4.9.1
@@ -61,16 +60,17 @@ urllib3~=2.2.2
 WTForms~=3.1.2
 wtforms-html5~=0.6.1
 zxcvbn~=4.4.28
-
 pytube~=15.0.0
-PyPDF2~=3.0.1
 groq~=0.9.0
 pydub~=0.25.1
 argparse~=1.4.0
-portkey_ai~=1.7.0
-
+portkey_ai~=1.8.2
 minio~=7.2.7
 Werkzeug~=3.0.3
 itsdangerous~=2.2.0
 cryptography~=43.0.0
-graypy~=2.1.0
+graypy~=2.1.0
+lxml~=5.3.0
+pillow~=10.4.0
+pdfplumber~=0.11.4
+PyPDF2~=3.0.1