Paperless-ngx 文档管理系统

   git clone https://github.com/paperless-ngx/paperless-ngx.git   cd paperless-ngx

   cp docker-compose.env.yml.example docker-compose.env.yml

   docker-compose up -d

   http://localhost:8000

import requests
# 获取文档列表response = requests.get("http://localhost:8000/api/documents/",                        auth=("username", "password"))documents = response.json()
# 上传新文档files = {"document": open("example.pdf", "rb")}response = requests.post("http://localhost:8000/api/documents/post_document/",                         files=files,                         auth=("username", "password"))

class BarcodePlugin(ConsumeTaskPlugin):    """    条形码处理插件，用于识别文档中的条形码    """    NAME: str = "BarcodePlugin"
    @property    def able_to_run(self) -> bool:        return (            settings.CONSUMER_ENABLE_BARCODES            and self.input_doc.mime_type == "application/pdf"        )
    def run(self) -> str | None:        # 识别条形码并处理文档        barcodes = self._get_barcodes(self.input_doc.original_file)        if not barcodes:            return None                # 处理找到的条形码        for barcode in barcodes:            if barcode.is_separator:                self._handle_separator_barcode(barcode)            elif barcode.is_asn:                self._handle_asn_barcode(barcode)                return f"Found {len(barcodes)} barcodes"

class DocumentClassifier:    """    文档自动分类器，用于预测文档的对应关系、类型和标签    """    def predict_correspondent(self, content: str) -> int | None:        """        预测文档的对应关系        """        if not self._clf_correspondent:            return None                X = self._vectorizer_correspondent.transform([content])        pred = self._clf_correspondent.predict(X)        if len(pred) > 0 and pred[0] != -1:            return pred[0]        return None
    def predict_document_type(self, content: str) -> int | None:        """        预测文档类型        """        if not self._clf_document_type:            return None                X = self._vectorizer_document_type.transform([content])        pred = self._clf_document_type.predict(X)        if len(pred) > 0 and pred[0] != -1:            return pred[0]        return None

def get_schema() -> Schema:    """    定义文档索引的搜索模式    """    return Schema(        id=NUMERIC(stored=True, unique=True),        title=TEXT(sortable=True),        content=TEXT(),        asn=NUMERIC(sortable=True, signed=False),        correspondent=TEXT(sortable=True),        correspondent_id=NUMERIC(),        document_type=TEXT(sortable=True),        document_type_id=NUMERIC(),        tags=KEYWORD(stored=True, commas=True, scorable=True),        tags_ids=KEYWORD(stored=True, commas=True),        created=DATETIME(sortable=True),        modified=DATETIME(sortable=True),        added=DATETIME(sortable=True),        path_id=KEYWORD(stored=True),        path_name=TEXT(),        owner_id=NUMERIC(),        owner_username=TEXT(),    )