content_type_by_signature.h

/*********************************************************************************************************************************************/
/*  DocWire SDK: Award-winning modern data processing in C++20. SourceForge Community Choice & Microsoft support. AI-driven processing.      */
/*  Supports nearly 100 data formats, including email boxes and OCR. Boost efficiency in text extraction, web data extraction, data mining,  */
/*  document analysis. Offline processing possible for security and confidentiality                                                          */
/*                                                                                                                                           */
/*  Copyright (c) SILVERCODERS Ltd, http://silvercoders.com                                                                                  */
/*  Project homepage: https://github.com/docwire/docwire                                                                                     */
/*                                                                                                                                           */
/*  SPDX-License-Identifier: AGPL-3.0-only OR LicenseRef-DocWire-Commercial                                                                  */
/*********************************************************************************************************************************************/
 
#ifndef DOCWIRE_CONTENT_TYPE_BY_SIGNATURE_H
#define DOCWIRE_CONTENT_TYPE_BY_SIGNATURE_H
 
#include "chain_element.h"
#include "content_type_export.h"
#include "data_source.h"
#include "document_elements.h"
#include "make_error.h"
#include "nested_exception.h"
#include "ref_or_owned.h"
 
namespace docwire::content_type::by_signature
{
 
struct allow_multiple
{
    bool v;
};
 
class DOCWIRE_CONTENT_TYPE_EXPORT database : public with_pimpl<database>
{
public:
    explicit database();
    friend DOCWIRE_CONTENT_TYPE_EXPORT void detect(data_source& data, const database& database_to_use, allow_multiple allow_multiple);
};
 
DOCWIRE_CONTENT_TYPE_EXPORT void detect(data_source& data, const database& database_to_use = database{}, allow_multiple allow_multiple = {false});
 
class detector : public chain_element
{
public:
 
    explicit detector(ref_or_owned<database> database_to_use = database{}, allow_multiple allow_multiple = {false})
        : m_database_to_use(database_to_use), m_allow_multiple{allow_multiple} {}
 
    continuation operator()(message_ptr msg, const message_callbacks& emit_message) override
    {
        try
        {
            if (msg->is<data_source>())
            {
                data_source& data = msg->get<data_source>();
                detect(data, m_database_to_use.get(), m_allow_multiple);
            }
            else if (msg->is<document::image>())
            {
                data_source& data = msg->get<document::image>().source;
                detect(data, m_database_to_use.get(), m_allow_multiple);
            }
        }
        catch (const std::exception& e)
        {
            emit_message(make_nested_ptr(std::current_exception(), DOCWIRE_MAKE_ERROR("Content type detection by signature failed")));
        }
        return emit_message(std::move(msg));
    }
 
    bool is_leaf() const override
    {
        return false;
    }
 
private:
    ref_or_owned<database> m_database_to_use;
    allow_multiple m_allow_multiple;
};
 
} // namespace docwire::content_type::by_signature
 
#endif // DOCWIRE_CONTENT_TYPE_BY_SIGNATURE_H