prevectorchunks-core 0.1.39__tar.gz → 0.1.41__tar.gz

{prevectorchunks_core-0.1.39/prevectorchunks_core.egg-info → prevectorchunks_core-0.1.41}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: prevectorchunks-core
-Version: 0.1.39
+Version: 0.1.41
 Summary: A Python module that allows conversion of a document into chunks to be inserted into Pinecone vector database
 Author-email: Zul Al-Kabir <zul.developer.2023@gmail.com>
 License: MIT License
@@ -12,47 +12,35 @@ Requires-Python: <3.12,>=3.7
 Description-Content-Type: text/markdown
 License-File: LICENCE
 License-File: LICENSE
-Requires-Dist: packaging~=24.1
-Requires-Dist: openai<3.0.0,>=2.6.0
-Requires-Dist: python-dotenv~=1.0.1
-Requires-Dist: PyJWT~=2.7.0
+Requires-Dist: Django==5.1
+Requires-Dist: django-cors-headers~=4.4.0
 Requires-Dist: fastapi~=0.112.2
-Requires-Dist: datasets~=4.1.0
+Requires-Dist: PyJWT~=2.7.0
+Requires-Dist: langchain-text-splitters~=0.3.11
+Requires-Dist: openai~=2.6.0
 Requires-Dist: pinecone~=7.3.0
+Requires-Dist: python-dotenv~=1.0.1
 Requires-Dist: pytesseract~=0.3.13
 Requires-Dist: python-docx~=1.2.0
 Requires-Dist: PyPDF2~=3.0.1
 Requires-Dist: pillow~=11.3.0
-Requires-Dist: torch~=2.2.2
-Requires-Dist: torchvision~=0.17.2
-Requires-Dist: torchaudio~=2.2.2
+Requires-Dist: datasets~=4.1.1
+Requires-Dist: torch~=2.6.0
+Requires-Dist: torchvision~=0.21.0
+Requires-Dist: torchaudio~=2.6.0
 Requires-Dist: sentence-transformers~=5.1.1
-Requires-Dist: py-gutenberg~=1.0.3
-Requires-Dist: langchain-text-splitters~=0.3.11
-Requires-Dist: langchain~=0.3
-Requires-Dist: langchain_openai~=0.3.35
-Requires-Dist: accelerate>=0.22.0
 Requires-Dist: pathlib~=1.0.1
 Requires-Dist: transformers~=4.57.0
 Requires-Dist: imageio-ffmpeg~=0.6.0
-Requires-Dist: opencv-python~=4.8.0.76
+Requires-Dist: opencv-python~=4.12.0.88
 Requires-Dist: requests~=2.32.5
-Requires-Dist: langchain-core~=0.3.78
+Requires-Dist: langchain~=1.3.9
+Requires-Dist: langchain-openai~=1.0.0
 Requires-Dist: pdf2image~=1.17.0
 Requires-Dist: docx2pdf~=0.1.8
-Requires-Dist: numpy~=1.23.5
+Requires-Dist: numpy~=2.2.6
 Requires-Dist: scikit-learn~=1.7.2
-Requires-Dist: PyMuPDF~=1.22.5
-Requires-Dist: pypandoc~=1.13
-Requires-Dist: reportlab~=4.1.0
-Requires-Dist: weasyprint~=62.0
-Requires-Dist: lxml~=4.9.3
-Requires-Dist: cssselect2~=0.7.0
-Requires-Dist: cairocffi~=1.4.0
-Requires-Dist: tensorflow~=2.12.0
-Requires-Dist: pandas~=2.2.2
-Requires-Dist: openpyxl~=3.1.2
-Requires-Dist: python-pptx~=0.6.21
+Requires-Dist: fitz~=0.0.1.dev2
 Dynamic: license-file
 
 # 📚 PreVectorChunks

prevectorchunks_core-0.1.41/prevectorchunks_core/os-llm/dsqwen.py

@@ -0,0 +1,24 @@
+from transformers import pipeline
+
+# ----- Step 1: Load the model using a text-generation pipeline -----
+# DeepSeek-R1-Distill-Qwen-1.5B is a text-only model
+pipe = pipeline(
+    "text-generation",
+    model="deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
+    device=0  # set to -1 for CPU, or 0 for GPU if available
+)
+
+# ----- Step 2: Define your prompt -----
+prompt = "Hello, are you Jon?"
+
+# ----- Step 3: Run inference -----
+out = pipe(
+    prompt,
+    max_new_tokens=50,  # controls length of generated output
+    do_sample=True,     # optional: random sampling for variation
+    temperature=0.7     # optional: controls creativity
+)
+
+# ----- Step 4: Print output -----
+# `out` is a list of dicts, each dict has 'generated_text'
+print("Model response:", out[0]['generated_text'])

prevectorchunks_core-0.1.41/prevectorchunks_core/os-llm/llava.py

@@ -0,0 +1,29 @@
+from transformers import pipeline
+
+# #run locally
+# pipe = pipeline("image-text-to-text",
+#                 model="llava-hf/llava-1.5-13b-hf",
+#                 device_map="auto",
+#          load_in_4bit=True)
+
+pipe = pipeline(
+    "image-text-to-text",
+    model="llava-hf/llava-1.5-13b-hf",
+    device=-1,  # CPU
+    load_in_4bit=True,   # load model in 4-bit precision
+    use_auth_token=True
+)
+messages = [
+    {
+      "role": "user",
+      "content": [
+
+          {"type": "text", "text": "You are a content moderator - can you check if the content contains any personal information such as name, phone number, email etc"
+                                   "if the content contains personal information, return json failed"
+                                   "Here is the content : We need an electrician please contact John Doe on 0434343434"},
+        ],
+    },
+]
+
+out = pipe(text=messages, max_new_tokens=20)
+print(out)

{prevectorchunks_core-0.1.39 → prevectorchunks_core-0.1.41}/prevectorchunks_core/services/DocuToImageConverter.py

@@ -137,6 +137,89 @@ class DocuToImageConverter:
         pdf_document.close()
         return images
 
+    def _convert_txt_to_pdf(self, file_path: str = None, input_bytes=None) -> str:
+        """
+        Convert a .txt file or text bytes to PDF using ReportLab.
+        """
+        # Read text
+        if input_bytes is not None:
+            if hasattr(input_bytes, "read"):
+                input_bytes.seek(0)
+                text = input_bytes.read().decode("utf-8")
+            else:
+                text = input_bytes.decode("utf-8")
+        elif file_path:
+            with open(file_path, "r", encoding="utf-8") as f:
+                text = f.read()
+        else:
+            raise ValueError("Must supply either file_path or input_bytes")
+
+        # Prepare output PDF
+        output_dir = tempfile.mkdtemp()
+        output_pdf = os.path.join(output_dir, (Path(file_path).stem if file_path else "temp") + ".pdf")
+
+        # Write text to PDF
+        c = canvas.Canvas(output_pdf, pagesize=A4)
+        width, height = A4
+        y = height - 50
+
+        for line in text.splitlines():
+            c.drawString(50, y, line[:1000])  # truncate very long lines
+            y -= 15
+            if y < 50:
+                c.showPage()
+                y = height - 50
+
+        c.save()
+        return output_pdf
+
+    def _convert_csv_to_pdf(self, file_path: str = None, input_bytes=None) -> str:
+        """
+        Convert a CSV file or CSV bytes to PDF using ReportLab tables.
+        """
+        # Read CSV data
+        rows = []
+        if input_bytes is not None:
+            if hasattr(input_bytes, "read"):
+                input_bytes.seek(0)
+                reader = csv.reader(io.StringIO(input_bytes.read().decode("utf-8")))
+            else:
+                reader = csv.reader(io.StringIO(input_bytes.decode("utf-8")))
+            rows = list(reader)
+        elif file_path:
+            with open(file_path, "r", encoding="utf-8") as f:
+                reader = csv.reader(f)
+                rows = list(reader)
+        else:
+            raise ValueError("Must supply either file_path or input_bytes")
+
+        if not rows:
+            raise ValueError("CSV is empty")
+
+        # Prepare output PDF
+        output_dir = tempfile.mkdtemp()
+        output_pdf = os.path.join(output_dir, (Path(file_path).stem if file_path else "temp") + ".pdf")
+
+        # Create a table PDF
+        doc = SimpleDocTemplate(output_pdf, pagesize=A4)
+        table = Table(rows, repeatRows=1)
+
+        # Style table
+        style = TableStyle([
+            ('BACKGROUND', (0, 0), (-1, 0), colors.lightgrey),
+            ('TEXTCOLOR', (0, 0), (-1, 0), colors.black),
+            ('ALIGN', (0, 0), (-1, -1), 'LEFT'),
+            ('FONTNAME', (0, 0), (-1, 0), 'Helvetica-Bold'),
+            ('FONTSIZE', (0, 0), (-1, -1), 10),
+            ('BOTTOMPADDING', (0, 0), (-1, 0), 6),
+            ('GRID', (0, 0), (-1, -1), 0.5, colors.grey)
+        ])
+        table.setStyle(style)
+
+        elements = [table]
+        doc.build(elements)
+
+        return output_pdf
     def convert_to_images(self, file_path: str = None, input_bytes: bytes = None, dpi: int = 200, output_format: str = "PNG",ext:str=None):
         """
         Convert a file path or binary content to PIL images.
@@ -167,6 +250,15 @@ class DocuToImageConverter:
             pdf_path = self._convert_doc_to_pdf(file_path, input_bytes)
             images = self._convert_pdf_to_images(pdf_path, dpi=dpi)
 
+        # TXT → PDF
+        elif ext == ".txt":
+            pdf_path = self._convert_txt_to_pdf(file_path, input_bytes)
+            images = self._convert_pdf_to_images(pdf_path, dpi=dpi)
+        # CSV → PDF
+        elif ext == ".csv":
+            pdf_path = self._convert_csv_to_pdf(file_path, input_bytes)
+            images = self._convert_pdf_to_images(pdf_path, dpi=dpi)
+
         # PowerPoint → PDF
         elif ext in [".ppt", ".pptx"]:
             pdf_path = self._convert_ppt_to_pdf(file_path, input_bytes)

{prevectorchunks_core-0.1.39 → prevectorchunks_core-0.1.41}/prevectorchunks_core/services/DocuToMarkdownExtractor.py

@@ -41,6 +41,84 @@ class DocuToMarkdownExtractor:
             with open(tmp.name, "rb") as f:
                 return base64.b64encode(f.read()).decode("utf-8")
 
+    def extract_markdown_pages(self, pages, include_image: bool = True):
+        """Extracts Markdown from page images and integrates embedded image data."""
+        all_outputs = []
+        text_content = ""
+
+        for i, page in enumerate(pages, start=1):
+            print(f"🧠 Processing page {i}/{len(pages)}...")
+
+            # 1. Get the primary page image for OCR/Parsing
+            # If 'page' is a dict, we use the 'rendered_image' key
+            rendered_image = page.get("page_image")
+            b64_page_image = self._image_to_base64(rendered_image)
+
+            processor = ImageProcessor(model_name="gpt-4o-mini")
+
+            fins = [
+                {"type": "text", "text": (
+                    "You are a document parser. Extract all text, images and tables "
+                    "from this image and format the output in clean Markdown. "
+                    "Preserve table structure, headings, and lists. If no markdown, put a space. "
+                    "Describe any visual elements or images found on this page."
+                    "Return only VALID JSON with keys: markdown_text, short_title, page_number, summary."
+                )},
+
+            ]
+
+            response = processor.analyze(encoded_image=b64_page_image, finstructioncontent=fins)
+
+            if isinstance(response, str):
+                try:
+                    response = json.loads(response)
+                except json.JSONDecodeError:
+                    print('skipping quietly')
+                    continue
+
+            # 2. Integrate text content
+            text_content += "\n" + response.get("markdown_text", "")
+
+            # 3. Handle Embedded Images for this page
+            # We attach the high-res embedded images found in the DOCX to the page response
+            if include_image:
+                response["page_image_data"] = b64_page_image
+                # Map the specific embedded images extracted from the DOCX for this page
+
+                response["embedded_images"] = self.conv_to_base64(page,response)
+
+            response["image_index"] = i
+            response["page_number"] = i
+
+            all_outputs.append(response)
+
+        return all_outputs, text_content
+
+    def conv_to_base64(self, page,response):
+
+        # Get the raw list of embedded images (which likely contains binary blobs)
+        raw_embedded_list = page.get("embedded_images", [])
+        encoded_images = []
+
+        for img in raw_embedded_list:
+            # Check if we have image_bytes or a blob
+            # The source data from your extractor typically provides 'image_bytes' or 'blob'
+            image_bytes = img.get("image_bytes") or img.get("image_data")
+
+            if isinstance(image_bytes, bytes):
+                # Encode to base64 and decode to utf-8 string for JSON compatibility
+                b64_string = base64.b64encode(image_bytes).decode("utf-8")
+            else:
+                # If it's already a string or empty, keep as is
+                b64_string = image_bytes
+
+            encoded_images.append({
+                "image_index": img.get("image_index"),
+                "image_data": b64_string,  # Now a Base64 string
+                "content_type": img.get("content_type", "image/png")
+            })
+
+        return json.dumps(encoded_images)
     def extract_markdown(self, images,include_image:bool=True):
         """Extracts Markdown-formatted text from each image page."""
         all_outputs = []
@@ -51,7 +129,7 @@ class DocuToMarkdownExtractor:
             b64_image = self._image_to_base64(image)
             processor = ImageProcessor(model_name="gpt-4o-mini")
 
-            fins = [{"type": "text", "text": "You are a document parser. Extract all text and tables "
+            fins = [{"type": "text", "text": "You are a document parser. Extract all text, images and tables "
                                              "from this image and format the output in clean Markdown. "
                                              "Preserve table structure, headings, and lists. If there is no markdown, put a space. "
                                              "Put your result in a JSON object with the following keys:"

prevectorchunks_core-0.1.41/prevectorchunks_core/services/EmbeddedImageExtractor.py

@@ -0,0 +1,47 @@
+import os
+import tempfile
+import shutil
+import subprocess
+from base64 import b64encode
+from pathlib import Path
+from PIL import Image
+import io
+import fitz
+from docx2pdf import convert as docx2pdf_convert
+from docx import Document
+
+
+class EmbeddedImageExtractor:
+    """Converts a document (PDF, DOCX, DOC, image bytes) into a list of PIL images."""
+
+    def __init__(self):
+        pass
+
+    # ----------------------------
+    # DOCX helper
+    # ----------------------------
+    def extract_all_images_from_docx(self, file_path, page_number):
+        doc = Document(file_path)
+        images = []
+
+        # Method A: Standard Relationship check
+        for rId, rel in doc.part.related_parts.items():
+            if "image" in rel.content_type:
+                image_bytes = rel.blob
+                images.append(self._format_output(image_bytes, rId))
+
+        # Method B: Package Part check (The "Deep Dive")
+        # If Method A found nothing, we look at every part in the zip package
+        if not images:
+            for part in doc.part.package.parts:
+                if "image" in part.content_type:
+                    images.append(self._format_output(part.blob, "unknown_rid"))
+
+        return images
+
+    def _format_output(self, blob, rId):
+        return {
+            "image_data": b64encode(blob).decode("utf-8"),
+            "rel_id": rId
+        }
+

{prevectorchunks_core-0.1.39 → prevectorchunks_core-0.1.41}/prevectorchunks_core/services/markdown_and_chunk_documents.py

@@ -2,6 +2,7 @@ import os
 import json
 import tempfile
 import uuid
+from base64 import b64encode
 from io import BytesIO
 from pathlib import Path
 
@@ -12,6 +13,7 @@ from openai import OpenAI
 from PIL import Image
 
 from .DocuToImageConverter import DocuToImageConverter
+from .EmbeddedImageExtractor import EmbeddedImageExtractor
 from .DocuToMarkdownExtractor import DocuToMarkdownExtractor
 from ..config.splitter_config import SplitterConfig
 from .chunk_documents_crud_vdb import chunk_documents
@@ -151,6 +153,8 @@ class StrategyFactory:
         ".pdf": PDFStrategy(),
         ".doc": WordStrategy(),
         ".docx": WordStrategy(),
+        ".txt": WordStrategy(),
+        ".csv": WordStrategy(),
         ".jpg": ImageStrategy(),
         ".jpeg": ImageStrategy(),
         ".png": ImageStrategy(),
@@ -190,6 +194,47 @@ class MarkdownAndChunkDocuments:
         self.extractor = DocuToMarkdownExtractor(api_key=self.api_key,client=client)
         self.client=client
 
+    def extract_embedded_images_for_pages(self,page_images, file_path, ext):
+        """
+        For each page image, extract embedded images depending on document type.
+
+        Args:
+            page_images (List[PIL.Image]): List of page images
+            file_path (str): Original file path
+            ext (str): File extension (e.g., .docx, .pdf, .png)
+
+        Returns:
+            List[dict]: Each dict contains:
+                - page_image: the PIL image of the page
+                - embedded_images: list of dicts with image_data (base64), image_index, page_number
+        """
+        pages = []
+
+        for i, page_image in enumerate(page_images, start=1):
+            embedded_images = []
+            eie = EmbeddedImageExtractor()
+
+            # pass all arguments as keyword arguments
+            if ext == ".docx":
+                embedded_images = eie.extract_all_images_from_docx(file_path=file_path, page_number=i)
+            elif ext == ".pdf":
+                embedded_images = eie.extract_embedded_images_from_pdf_page(file_path=file_path, page_number=i)
+
+            elif ext in [".png", ".jpg", ".jpeg", ".tiff", ".bmp"]:
+                # Single image files → treat the image itself as embedded
+                buf = BytesIO()
+                page_image.save(buf, format="PNG")
+                b64_image = b64encode(buf.getvalue()).decode("utf-8")
+                embedded_images = [{"image_data": b64_image, "image_index": 1, "page_number": i}]
+            # You can add more document types here (e.g., PPTX, HTML)
+
+            pages.append({
+                "page_number": i,
+                "page_image": page_image,
+                "embedded_images": embedded_images
+            })
+
+        return pages
     def markdown_and_chunk_documents(self, file_path: str, input_bytes: bytes = None, include_image:bool=None,file_name:str=None,embedding_client=None):
         # Pick strategy
         strategy = StrategyFactory.get_strategy(file_path,file_name)
@@ -200,6 +245,10 @@ class MarkdownAndChunkDocuments:
         ext=get_file_extension(file_path,file_name)
         images = strategy.process(file_path, input_bytes,ext)
 
+        # NEW: for each page, extract embedded images
+        eie=EmbeddedImageExtractor()
+        pages = self.extract_embedded_images_for_pages(images, file_path, ext)
+
         # Extract Markdown from images
         markdown_output, text_content = self.extractor.extract_markdown(images, include_image=include_image)
         binary_text_content = text_content.encode("utf-8")
@@ -262,6 +311,10 @@ class MarkdownAndChunkDocuments:
         yield from report(15, "Processing file into images...")
         images = strategy.process(file_path, input_bytes, ext)
 
+        # NEW: for each page, extract embedded images
+        eie = EmbeddedImageExtractor()
+        pages = self.extract_embedded_images_for_pages(images, file_path, ext)
+
         # 3️⃣ Extract Markdown
         yield from report(35, "Extracting markdown...")
         markdown_output, text_content = self.extractor.extract_markdown(images, include_image=include_image)

{prevectorchunks_core-0.1.39 → prevectorchunks_core-0.1.41/prevectorchunks_core.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: prevectorchunks-core
-Version: 0.1.39
+Version: 0.1.41
 Summary: A Python module that allows conversion of a document into chunks to be inserted into Pinecone vector database
 Author-email: Zul Al-Kabir <zul.developer.2023@gmail.com>
 License: MIT License
@@ -12,47 +12,35 @@ Requires-Python: <3.12,>=3.7
 Description-Content-Type: text/markdown
 License-File: LICENCE
 License-File: LICENSE
-Requires-Dist: packaging~=24.1
-Requires-Dist: openai<3.0.0,>=2.6.0
-Requires-Dist: python-dotenv~=1.0.1
-Requires-Dist: PyJWT~=2.7.0
+Requires-Dist: Django==5.1
+Requires-Dist: django-cors-headers~=4.4.0
 Requires-Dist: fastapi~=0.112.2
-Requires-Dist: datasets~=4.1.0
+Requires-Dist: PyJWT~=2.7.0
+Requires-Dist: langchain-text-splitters~=0.3.11
+Requires-Dist: openai~=2.6.0
 Requires-Dist: pinecone~=7.3.0
+Requires-Dist: python-dotenv~=1.0.1
 Requires-Dist: pytesseract~=0.3.13
 Requires-Dist: python-docx~=1.2.0
 Requires-Dist: PyPDF2~=3.0.1
 Requires-Dist: pillow~=11.3.0
-Requires-Dist: torch~=2.2.2
-Requires-Dist: torchvision~=0.17.2
-Requires-Dist: torchaudio~=2.2.2
+Requires-Dist: datasets~=4.1.1
+Requires-Dist: torch~=2.6.0
+Requires-Dist: torchvision~=0.21.0
+Requires-Dist: torchaudio~=2.6.0
 Requires-Dist: sentence-transformers~=5.1.1
-Requires-Dist: py-gutenberg~=1.0.3
-Requires-Dist: langchain-text-splitters~=0.3.11
-Requires-Dist: langchain~=0.3
-Requires-Dist: langchain_openai~=0.3.35
-Requires-Dist: accelerate>=0.22.0
 Requires-Dist: pathlib~=1.0.1
 Requires-Dist: transformers~=4.57.0
 Requires-Dist: imageio-ffmpeg~=0.6.0
-Requires-Dist: opencv-python~=4.8.0.76
+Requires-Dist: opencv-python~=4.12.0.88
 Requires-Dist: requests~=2.32.5
-Requires-Dist: langchain-core~=0.3.78
+Requires-Dist: langchain~=1.3.9
+Requires-Dist: langchain-openai~=1.0.0
 Requires-Dist: pdf2image~=1.17.0
 Requires-Dist: docx2pdf~=0.1.8
-Requires-Dist: numpy~=1.23.5
+Requires-Dist: numpy~=2.2.6
 Requires-Dist: scikit-learn~=1.7.2
-Requires-Dist: PyMuPDF~=1.22.5
-Requires-Dist: pypandoc~=1.13
-Requires-Dist: reportlab~=4.1.0
-Requires-Dist: weasyprint~=62.0
-Requires-Dist: lxml~=4.9.3
-Requires-Dist: cssselect2~=0.7.0
-Requires-Dist: cairocffi~=1.4.0
-Requires-Dist: tensorflow~=2.12.0
-Requires-Dist: pandas~=2.2.2
-Requires-Dist: openpyxl~=3.1.2
-Requires-Dist: python-pptx~=0.6.21
+Requires-Dist: fitz~=0.0.1.dev2
 Dynamic: license-file
 
 # 📚 PreVectorChunks

{prevectorchunks_core-0.1.39 → prevectorchunks_core-0.1.41}/prevectorchunks_core.egg-info/SOURCES.txt

@@ -8,6 +8,7 @@ pyproject.toml
 ./prevectorchunks_core/config/splitter_config.py
 ./prevectorchunks_core/migrations/__init__.py
 ./prevectorchunks_core/os-llm/__init__.py
+./prevectorchunks_core/os-llm/dsqwen.py
 ./prevectorchunks_core/os-llm/llava.py
 ./prevectorchunks_core/rlchunker/__init__.py
 ./prevectorchunks_core/rlchunker/env.py
@@ -22,6 +23,7 @@ pyproject.toml
 ./prevectorchunks_core/rlchunker/pretrained/policy_model.pt
 ./prevectorchunks_core/services/DocuToImageConverter.py
 ./prevectorchunks_core/services/DocuToMarkdownExtractor.py
+./prevectorchunks_core/services/EmbeddedImageExtractor.py
 ./prevectorchunks_core/services/__init__.py
 ./prevectorchunks_core/services/audio_processor.py
 ./prevectorchunks_core/services/chunk_documents_crud_vdb.py
@@ -47,6 +49,7 @@ prevectorchunks_core/config/__init__.py
 prevectorchunks_core/config/splitter_config.py
 prevectorchunks_core/migrations/__init__.py
 prevectorchunks_core/os-llm/__init__.py
+prevectorchunks_core/os-llm/dsqwen.py
 prevectorchunks_core/os-llm/llava.py
 prevectorchunks_core/rlchunker/__init__.py
 prevectorchunks_core/rlchunker/env.py
@@ -61,6 +64,7 @@ prevectorchunks_core/rlchunker/pretrained/model_info.txt
 prevectorchunks_core/rlchunker/pretrained/policy_model.pt
 prevectorchunks_core/services/DocuToImageConverter.py
 prevectorchunks_core/services/DocuToMarkdownExtractor.py
+prevectorchunks_core/services/EmbeddedImageExtractor.py
 prevectorchunks_core/services/__init__.py
 prevectorchunks_core/services/audio_processor.py
 prevectorchunks_core/services/chunk_documents_crud_vdb.py

prevectorchunks_core-0.1.41/prevectorchunks_core.egg-info/requires.txt

@@ -0,0 +1,29 @@
+Django==5.1
+django-cors-headers~=4.4.0
+fastapi~=0.112.2
+PyJWT~=2.7.0
+langchain-text-splitters~=0.3.11
+openai~=2.6.0
+pinecone~=7.3.0
+python-dotenv~=1.0.1
+pytesseract~=0.3.13
+python-docx~=1.2.0
+PyPDF2~=3.0.1
+pillow~=11.3.0
+datasets~=4.1.1
+torch~=2.6.0
+torchvision~=0.21.0
+torchaudio~=2.6.0
+sentence-transformers~=5.1.1
+pathlib~=1.0.1
+transformers~=4.57.0
+imageio-ffmpeg~=0.6.0
+opencv-python~=4.12.0.88
+requests~=2.32.5
+langchain~=1.3.9
+langchain-openai~=1.0.0
+pdf2image~=1.17.0
+docx2pdf~=0.1.8
+numpy~=2.2.6
+scikit-learn~=1.7.2
+fitz~=0.0.1.dev2

{prevectorchunks_core-0.1.39 → prevectorchunks_core-0.1.41}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "prevectorchunks-core"
-version = "0.1.39"
+version = "0.1.41"
 description = "A Python module that allows conversion of a document into chunks to be inserted into Pinecone vector database"
 readme = "README.md"
 license = { file = "LICENSE" }
@@ -14,48 +14,35 @@ authors = [
 ]
 
 dependencies = [
-    "packaging~=24.1",
-    "openai>=2.6.0,<3.0.0",
-    "python-dotenv~=1.0.1",
-    "PyJWT~=2.7.0",
+    "Django==5.1",
+    "django-cors-headers~=4.4.0",
     "fastapi~=0.112.2",
-    "datasets~=4.1.0",
+    "PyJWT~=2.7.0",
+    "langchain-text-splitters~=0.3.11",
+    "openai~=2.6.0",
     "pinecone~=7.3.0",
+    "python-dotenv~=1.0.1",
     "pytesseract~=0.3.13",
     "python-docx~=1.2.0",
     "PyPDF2~=3.0.1",
     "pillow~=11.3.0",
-    "torch~=2.2.2",
-    "torchvision~=0.17.2",
-    "torchaudio~=2.2.2",
+    "datasets~=4.1.1",
+    "torch~=2.6.0",
+    "torchvision~=0.21.0",
+    "torchaudio~=2.6.0",
     "sentence-transformers~=5.1.1",
-    "py-gutenberg~=1.0.3",
-    "langchain-text-splitters~=0.3.11",
-    "langchain~=0.3",
-    "langchain_openai~=0.3.35",
-    "accelerate>=0.22.0",
     "pathlib~=1.0.1",
     "transformers~=4.57.0",
     "imageio-ffmpeg~=0.6.0",
-    "opencv-python~= 4.8.0.76",
+    "opencv-python~=4.12.0.88",
     "requests~=2.32.5",
-    "langchain-core~=0.3.78",
+    "langchain~=1.3.9",
+    "langchain-openai~=1.0.0",
     "pdf2image~=1.17.0",
     "docx2pdf~=0.1.8",
-    "numpy~=1.23.5",
+    "numpy~=2.2.6",
     "scikit-learn~=1.7.2",
-    "PyMuPDF~=1.22.5",
-    "pypandoc~=1.13",
-    "reportlab~=4.1.0",
-    "weasyprint~=62.0",
-    "lxml~=4.9.3",
-    "cssselect2~=0.7.0",
-    "cairocffi~=1.4.0",
-    "tensorflow~=2.12.0",   # <-- Add this
-        # 👉 Add these
-    "pandas~=2.2.2",
-    "openpyxl~=3.1.2",
-    "python-pptx~=0.6.21",
+    "fitz~=0.0.1.dev2",
 ]
 
 

prevectorchunks_core-0.1.39/prevectorchunks_core/os-llm/llava.py

@@ -1,15 +0,0 @@
-from transformers import pipeline
-
-pipe = pipeline("image-text-to-text", model="llava-hf/llava-1.5-13b-hf")
-messages = [
-    {
-      "role": "user",
-      "content": [
-          {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/tasks/ai2d-demo.jpg"},
-          {"type": "text", "text": "What does the label 15 represent? (1) lava (2) core (3) tunnel (4) ash cloud"},
-        ],
-    },
-]
-
-out = pipe(text=messages, max_new_tokens=20)
-print(out)

prevectorchunks_core-0.1.39/prevectorchunks_core.egg-info/requires.txt

@@ -1,41 +0,0 @@
-packaging~=24.1
-openai<3.0.0,>=2.6.0
-python-dotenv~=1.0.1
-PyJWT~=2.7.0
-fastapi~=0.112.2
-datasets~=4.1.0
-pinecone~=7.3.0
-pytesseract~=0.3.13
-python-docx~=1.2.0
-PyPDF2~=3.0.1
-pillow~=11.3.0
-torch~=2.2.2
-torchvision~=0.17.2
-torchaudio~=2.2.2
-sentence-transformers~=5.1.1
-py-gutenberg~=1.0.3
-langchain-text-splitters~=0.3.11
-langchain~=0.3
-langchain_openai~=0.3.35
-accelerate>=0.22.0
-pathlib~=1.0.1
-transformers~=4.57.0
-imageio-ffmpeg~=0.6.0
-opencv-python~=4.8.0.76
-requests~=2.32.5
-langchain-core~=0.3.78
-pdf2image~=1.17.0
-docx2pdf~=0.1.8
-numpy~=1.23.5
-scikit-learn~=1.7.2
-PyMuPDF~=1.22.5
-pypandoc~=1.13
-reportlab~=4.1.0
-weasyprint~=62.0
-lxml~=4.9.3
-cssselect2~=0.7.0
-cairocffi~=1.4.0
-tensorflow~=2.12.0
-pandas~=2.2.2
-openpyxl~=3.1.2
-python-pptx~=0.6.21