UnisKB/apps/common/handle/impl/text/doc_split_handle.py

# coding=utf-8
"""
    @project: maxkb
    @Author：虎
    @file： text_split_handle.py
    @date：2024/3/27 18:19
    @desc:
"""
import io
import os
import re
import traceback
from functools import reduce
from typing import List

import uuid_utils.compat as uuid
from docx import Document, ImagePart
from docx.oxml import ns
from docx.table import Table
from docx.text.paragraph import Paragraph

from common.handle.base_split_handle import BaseSplitHandle
from common.utils.logger import maxkb_logger
from common.utils.split_model import SplitModel
from knowledge.models import File

default_pattern_list = [re.compile('(?<=^)# .*|(?<=\\n)# .*'),
                        re.compile('(?<=\\n)(?<!#)## (?!#).*|(?<=^)(?<!#)## (?!#).*'),
                        re.compile("(?<=\\n)(?<!#)### (?!#).*|(?<=^)(?<!#)### (?!#).*"),
                        re.compile("(?<=\\n)(?<!#)#### (?!#).*|(?<=^)(?<!#)#### (?!#).*"),
                        re.compile("(?<=\\n)(?<!#)##### (?!#).*|(?<=^)(?<!#)##### (?!#).*"),
                        re.compile("(?<=\\n)(?<!#)###### (?!#).*|(?<=^)(?<!#)###### (?!#).*")]

old_docx_nsmap = {'v': 'urn:schemas-microsoft-com:vml'}
combine_nsmap = {**ns.nsmap, **old_docx_nsmap}


def image_to_mode(image, doc: Document, images_list, get_image_id):
    image_ids = image['get_image_id_handle'](image.get('image'))
    for img_id in image_ids:  # 获取图片id
        part = doc.part.related_parts[img_id]  # 根据图片id获取对应的图片
        if isinstance(part, ImagePart):
            image_uuid = get_image_id(img_id)
            if len([i for i in images_list if i.id == image_uuid]) == 0:
                image = File(id=image_uuid, file_name=part.filename, meta={'debug': False, 'content': part.blob})
                images_list.append(image)
            return f'![{part.filename.replace("[", "").replace("]", "")}](./oss/file/{image_uuid})'
        return None
    return None


def get_paragraph_element_images(paragraph_element, doc: Document, images_list, get_image_id):
    images_xpath_list = [(".//pic:pic", lambda img: img.xpath('.//a:blip/@r:embed')),
                         (".//w:pict", lambda img: img.xpath('.//v:imagedata/@r:id', namespaces=combine_nsmap))]
    images = []
    for images_xpath, get_image_id_handle in images_xpath_list:
        try:
            _images = paragraph_element.xpath(images_xpath)
            if _images is not None and len(_images) > 0:
                for image in _images:
                    images.append({'image': image, 'get_image_id_handle': get_image_id_handle})
        except Exception as e:
            pass
    return images


def images_to_string(images, doc: Document, images_list, get_image_id):
    return "".join(
        [item for item in [image_to_mode(image, doc, images_list, get_image_id) for image in images] if
         item is not None])


def get_paragraph_element_txt(paragraph_element, doc: Document, images_list, get_image_id):
    try:
        images = get_paragraph_element_images(paragraph_element, doc, images_list, get_image_id)
        if len(images) > 0:
            return images_to_string(images, doc, images_list, get_image_id)
        elif paragraph_element.text is not None:
            return paragraph_element.text
        return ""
    except Exception as e:
        maxkb_logger.error(f'Error getting paragraph element text: {e}')
    return ""


def get_paragraph_txt(paragraph: Paragraph, doc: Document, images_list, get_image_id):
    try:
        return "".join([get_paragraph_element_txt(e, doc, images_list, get_image_id) for e in paragraph._element])
    except Exception as e:
        return ""


def get_cell_text(cell, doc: Document, images_list, get_image_id):
    try:
        return "".join(
            [get_paragraph_txt(paragraph, doc, images_list, get_image_id) for paragraph in cell.paragraphs]).replace(
            "\n", '</br>')
    except Exception as e:
        return ""


def get_image_id_func():
    image_map = {}

    def get_image_id(image_id):
        _v = image_map.get(image_id)
        if _v is None:
            image_map[image_id] = uuid.uuid7()
            return image_map.get(image_id)
        return _v

    return get_image_id


title_font_list = [
    [36, 100],
    [30, 36]
]


def get_title_level(paragraph: Paragraph):
    try:
        if paragraph.style is not None:
            psn = paragraph.style.name
            if psn.startswith('Heading') or psn.startswith('TOC 标题') or psn.startswith('标题'):
                return int(psn.replace("Heading ", '').replace('TOC 标题', '').replace('标题',
                                                                                       ''))
        if len(paragraph.runs) == 1:
            font_size = paragraph.runs[0].font.size
            pt = font_size.pt
            if pt >= 30:
                for _value, index in zip(title_font_list, range(len(title_font_list))):
                    if pt >= _value[0] and pt < _value[1]:
                        return index + 1
    except Exception as e:
        pass
    return None


class DocSplitHandle(BaseSplitHandle):
    @staticmethod
    def paragraph_to_md(paragraph: Paragraph, doc: Document, images_list, get_image_id):
        try:
            title_level = get_title_level(paragraph)
            if title_level is not None:
                title = "".join(["#" for i in range(title_level)]) + " " + paragraph.text
                images = reduce(lambda x, y: [*x, *y],
                                [get_paragraph_element_images(e, doc, images_list, get_image_id) for e in
                                 paragraph._element],
                                [])
                if len(images) > 0:
                    return title + '\n' + images_to_string(images, doc, images_list, get_image_id) if len(
                        paragraph.text) > 0 else images_to_string(images, doc, images_list, get_image_id)
                return title

        except Exception as e:
            maxkb_logger.error(f"Error processing DOC file: {e}, {traceback.format_exc()}")
            return paragraph.text
        return get_paragraph_txt(paragraph, doc, images_list, get_image_id)

    @staticmethod
    def table_to_md(table, doc: Document, images_list, get_image_id):
        rows = table.rows

        # 创建 Markdown 格式的表格
        md_table = '| ' + ' | '.join(
            [get_cell_text(cell, doc, images_list, get_image_id) for cell in rows[0].cells]) + ' |\n'
        md_table += '| ' + ' | '.join(['---' for i in range(len(rows[0].cells))]) + ' |\n'
        for row in rows[1:]:
            md_table += '| ' + ' | '.join(
                [get_cell_text(cell, doc, images_list, get_image_id) for cell in row.cells]) + ' |\n'
        return md_table

    def to_md(self, doc, images_list, get_image_id):
        elements = []
        for element in doc.element.body:
            tag = str(element.tag)
            if tag.endswith('tbl'):
                # 处理表格
                table = Table(element, doc)
                elements.append(table)
            elif tag.endswith('p'):
                # 处理段落
                paragraph = Paragraph(element, doc)
                elements.append(paragraph)
        return "\n".join(
            [self.paragraph_to_md(element, doc, images_list, get_image_id) if isinstance(element,
                                                                                         Paragraph) else self.table_to_md(
                element,
                doc,
                images_list, get_image_id)
             for element
             in elements])

    def handle(self, file, pattern_list: List, with_filter: bool, limit: int, get_buffer, save_image):
        file_name = os.path.basename(file.name)
        try:
            if type(limit) is str:
                limit = int(limit)
            if type(with_filter) is str:
                with_filter = with_filter.lower() == 'true'
            image_list = []
            buffer = get_buffer(file)
            doc = Document(io.BytesIO(buffer))
            content = self.to_md(doc, image_list, get_image_id_func())
            if len(image_list) > 0:
                save_image(image_list)
            if pattern_list is not None and len(pattern_list) > 0:
                split_model = SplitModel(pattern_list, with_filter, limit)
            else:
                split_model = SplitModel(default_pattern_list, with_filter=with_filter, limit=limit)
        except BaseException as e:
            maxkb_logger.error(f"Error processing XLSX file {file.name}: {e}, {traceback.format_exc()}")
            return {
                'name': file_name,
                'content': []
            }
        return {
            'name': file_name,
            'content': split_model.parse(content)
        }

    def support(self, file, get_buffer):
        file_name: str = file.name.lower()
        if file_name.endswith(".docx") or file_name.endswith(".doc") or file_name.endswith(
                ".DOC") or file_name.endswith(".DOCX"):
            return True
        return False

    def get_content(self, file, save_image):
        try:
            image_list = []
            buffer = file.read()
            doc = Document(io.BytesIO(buffer))
            content = self.to_md(doc, image_list, get_image_id_func())
            if len(image_list) > 0:
                save_image(image_list)
            return content
        except BaseException as e:
            traceback.print_exception(e)
            return f'{e}'