Files
tm_back/ted/parser.go
T
n.nakhostin 0dd877a46d Refactor TED Repository and Introduce TED Mapping Functionality
- Updated the tender repository to create a unique index on tender_id, enhancing database query performance.
- Introduced new TED mapping functions to convert parsed TED documents into tender entities, improving integration with the tender management system.
- Added error handling for contract notice processing in the TED scraper, ensuring robust logging and error management.
- Removed deprecated eform structures to streamline the codebase and focus on essential components.
2025-10-04 12:34:37 +03:30

88 lines
3.1 KiB
Go

package ted
import (
"encoding/xml"
"fmt"
"strings"
"tm/pkg/xmlparser"
)
type TEDParser struct {
ContractNotice xmlparser.Parser[ContractNotice]
ContractAwardNotice xmlparser.Parser[ContractAwardNotice]
PriorInformationNotice xmlparser.Parser[PriorInformationNotice]
// BusinessRegistrationInformationNotice xmlparser.Parser[BusinessRegistrationInformationNotice]
}
// NewTEDParser creates a new TED parser
func NewTEDParser() *TEDParser {
options := xmlparser.DefaultParserOptions()
options.MaxSize = 50 * 1024 * 1024 // 50MB limit for TED XML files
options.IgnoreUnknownElements = true // TED XML has many optional fields
contractNoticeParser := xmlparser.NewParser[ContractNotice](options)
contractAwardNoticeParser := xmlparser.NewParser[ContractAwardNotice](options)
priorInformationNoticeParser := xmlparser.NewParser[PriorInformationNotice](options)
// businessRegistrationInformationNoticeParser := xmlparser.NewParser[BusinessRegistrationInformationNotice](options)
return &TEDParser{
ContractNotice: contractNoticeParser,
ContractAwardNotice: contractAwardNoticeParser,
PriorInformationNotice: priorInformationNoticeParser,
// BusinessRegistrationInformationNotice: businessRegistrationInformationNoticeParser,
}
}
// ParseXML automatically detects and parses any supported TED XML document type
func (p *TEDParser) ParseXML(content []byte) (*ParsedDocument, string, DocumentType, error) {
// First, detect the root element type
rootElement, err := detectRootElement(content)
if err != nil {
return nil, "", DocumentType(""), fmt.Errorf("failed to detect root element: %w", err)
}
parsedDoc := &ParsedDocument{Type: DocumentType(rootElement)}
switch parsedDoc.Type {
case DocumentTypeContractNotice:
notice, err := p.ContractNotice.ParseBytes(content)
if err != nil {
return nil, "", DocumentType(""), fmt.Errorf("failed to parse contract notice: %w", err)
}
parsedDoc.ContractNotice = notice
return parsedDoc, notice.GetID(), parsedDoc.Type, nil
case DocumentTypeContractAwardNotice:
notice, err := p.ContractAwardNotice.ParseBytes(content)
if err != nil {
return nil, "", DocumentType(""), fmt.Errorf("failed to parse contract award notice: %w", err)
}
parsedDoc.ContractAwardNotice = notice
return parsedDoc, notice.GetID(), parsedDoc.Type, nil
case DocumentTypePriorInformationNotice:
notice, err := p.PriorInformationNotice.ParseBytes(content)
if err != nil {
return nil, "", DocumentType(""), fmt.Errorf("failed to parse prior information notice: %w", err)
}
parsedDoc.PriorInformationNotice = notice
return parsedDoc, notice.GetID(), parsedDoc.Type, nil
default:
return nil, "", DocumentType(""), fmt.Errorf("unsupported document type: %s", parsedDoc.Type)
}
}
// detectRootElement detects the root XML element name
func detectRootElement(content []byte) (string, error) {
decoder := xml.NewDecoder(strings.NewReader(string(content)))
for {
token, err := decoder.Token()
if err != nil {
return "", err
}
if startElement, ok := token.(xml.StartElement); ok {
return startElement.Name.Local, nil
}
}
}