hjxdl 0.1.1__py3-none-any.whl → 0.1.3__py3-none-any.whl

hdl/_version.py

@@ -12,5 +12,5 @@ __version__: str
 __version_tuple__: VERSION_TUPLE
 version_tuple: VERSION_TUPLE
 
-__version__ = version = '0.1.1'
-__version_tuple__ = version_tuple = (0, 1, 1)
+__version__ = version = '0.1.3'
+__version_tuple__ = version_tuple = (0, 1, 3)

hdl/utils/llm/extract.py

@@ -0,0 +1,119 @@
+import pdfplumber
+import pytesseract
+from PIL import Image
+import pandas as pd
+import io
+from spire.doc import Document
+from spire.doc.common import *
+
+
+class DocExtractor():
+    def __init__(
+        self,
+        lang: str = "chi_sim"
+    ) -> None:
+        self.lang = lang
+
+    @classmethod
+    def text_from_doc(
+        doc_path
+    ):
+        document = Document()
+        # Load a Word document
+        document.LoadFromFile(doc_path)
+        document_text = document.GetText()
+        return document_text
+    
+    @staticmethod
+    def text_from_plain(
+        txt_path
+    ):
+        with open(txt_path, "r") as f:
+            text = f.read()
+        return text
+    
+    @staticmethod
+    def extract_text_from_image(
+        image: Image.Image,
+    ) -> str:
+        return pytesseract.image_to_string(image, lang=self.lang)
+
+    @staticmethod
+    def is_within_bbox(
+        bbox1, bbox2
+    ):
+        """Check if bbox1 is within bbox2."""
+        return bbox1[0] >= bbox2[0] and bbox1[1] >= bbox2[1] and bbox1[2] <= bbox2[2] and bbox1[3] <= bbox2[3]
+
+    def text_tables_from_pdf(
+        self,
+        pdf_path,
+        table_from_pic: bool = False
+    ):
+        all_tables = []
+        all_texts = []
+        with pdfplumber.open(pdf_path) as pdf:
+            for page_number, page in enumerate(pdf.pages):
+                tables = page.find_tables()
+                page_text = page.extract_text(x_tolerance=0.1, y_tolerance=0.1) or ''
+                page_text_lines = page_text.split('\n')
+
+                # Extract tables
+                if tables:
+                    for table in tables:
+                        if table and len(table.extract()) > 1:
+                            table_data = table.extract()
+                            df = pd.DataFrame(table_data[1:], columns=table_data[0])
+                            df['Page'] = page_number + 1  # 添加页码信息
+                            all_tables.append(df)
+                
+                # Get bounding boxes for tables
+                table_bboxes = [table.bbox for table in tables]
+
+                # Filter out text within table bounding boxes
+                non_table_text = []
+                for char in page.chars:
+                    char_bbox = (char['x0'], char['top'], char['x1'], char['bottom'])
+                    if not any(self.is_within_bbox(char_bbox, table_bbox) for table_bbox in table_bboxes):
+                        non_table_text.append(char['text'])
+                remaining_text = ''.join(non_table_text).strip()
+                if remaining_text:
+                    all_texts.append(remaining_text)
+
+                # Extract tables from images if specified
+                if table_from_pic:
+                    for img in page.images:
+                        try:
+                            x0, top, x1, bottom = img["x0"], img["top"], img["x1"], img["bottom"]
+                            if x0 < 0 or top < 0 or x1 > page.width or bottom > page.height:
+                                print(f"Skipping image with invalid bounds on page {page_number + 1}")
+                                continue
+                            
+                            cropped_image = page.within_bbox((x0, top, x1, bottom)).to_image()
+                            img_bytes = io.BytesIO()
+                            cropped_image.save(img_bytes, format='PNG')
+                            img_bytes.seek(0)
+                            pil_image = Image.open(img_bytes)
+                            
+                            ocr_text = self.extract_text_from_image(pil_image, lang=self.lang)
+                            
+                            table = [line.split() for line in ocr_text.split('\n') if line.strip()]
+                            
+                            if table:
+                                num_columns = max(len(row) for row in table)
+                                for row in table:
+                                    if len(row) != num_columns:
+                                        row.extend([''] * (num_columns - len(row)))
+                                
+                                df = pd.DataFrame(table[1:], columns=table[0])
+                                df['Page'] = page_number + 1
+                                all_tables.append(df)
+                        except Exception as e:
+                            print(f"Error processing image on page {page_number + 1}: {e}")
+
+        if all_tables:
+            return all_texts, all_tables
+        else:
+            return all_texts, [pd.DataFrame()]
+
+    

{hjxdl-0.1.1.dist-info → hjxdl-0.1.3.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: hjxdl
-Version: 0.1.1
+Version: 0.1.3
 Summary: A collection of functions for Jupyter notebooks
 Home-page: https://github.com/huluxiaohuowa/hdl
 Author: Jianxing Hu

{hjxdl-0.1.1.dist-info → hjxdl-0.1.3.dist-info}/RECORD

@@ -1,5 +1,5 @@
 hdl/__init__.py,sha256=5sZZNySv08wwfzJcSDssGTqUn9wlmDsR6R4XB8J8mFM,70
-hdl/_version.py,sha256=PKIMyjdUACH4-ONvtunQCnYE2UhlMfp9su83e3HXl5E,411
+hdl/_version.py,sha256=L5DCMp1QAlSqy-8bW7d51bLubTxNjZGYc5fMQkb752U,411
 hdl/args/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 hdl/args/loss_args.py,sha256=s7YzSdd7IjD24rZvvOrxLLFqMZQb9YylxKeyelSdrTk,70
 hdl/controllers/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
@@ -86,9 +86,10 @@ hdl/utils/general/glob.py,sha256=8-RCnt6L297wMIfn34ZAMCsGCZUjHG3MGglGZI1cX0g,491
 hdl/utils/llm/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 hdl/utils/llm/chat.py,sha256=H2c8assJlSdZQKIfPkYrVZHqv66TsdsxtaLXv0kNe1w,11565
 hdl/utils/llm/embs.py,sha256=sC8tga7HgDwPI2m7TDWKp9kkxEIMxEyMtgmEhfRi4vI,6362
+hdl/utils/llm/extract.py,sha256=eF-oHu5sMtes8I6ZfNXnEykPfzqbn-2WvnIKiUMz6BA,4573
 hdl/utils/schedulers/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 hdl/utils/schedulers/norm_lr.py,sha256=bDwCmdEK-WkgxQMFBiMuchv8Mm7C0-GZJ6usm-PQk14,4461
-hjxdl-0.1.1.dist-info/METADATA,sha256=ORYQSX57x_WZQ1pG6KIjvesKNPshg1vCKv_8kw0hyas,542
-hjxdl-0.1.1.dist-info/WHEEL,sha256=pd56usn78UTvq1xeX_ZwFhoK6jE5u5wzu4TTBIG5cQ0,91
-hjxdl-0.1.1.dist-info/top_level.txt,sha256=-kxwTM5JPhylp06z3zAVO3w6_h7wtBfBo2zgM6YZoTk,4
-hjxdl-0.1.1.dist-info/RECORD,,
+hjxdl-0.1.3.dist-info/METADATA,sha256=e-L25DrhaIVW_yfpSpmoqH4k-J6yaTooOGviFgDXFwo,542
+hjxdl-0.1.3.dist-info/WHEEL,sha256=-oYQCr74JF3a37z2nRlQays_SX2MqOANoqVjBBAP2yE,91
+hjxdl-0.1.3.dist-info/top_level.txt,sha256=-kxwTM5JPhylp06z3zAVO3w6_h7wtBfBo2zgM6YZoTk,4
+hjxdl-0.1.3.dist-info/RECORD,,

{hjxdl-0.1.1.dist-info → hjxdl-0.1.3.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: setuptools (71.0.0)
+Generator: setuptools (71.0.3)
 Root-Is-Purelib: true
 Tag: py3-none-any