15046951426

Committed 15 May 2025 02:00PM UTC coverage: 23.832% (+0.02%) from 23.816%

Build # 15046951426

Build Type

push

github

Committed by

paulmthompson

Commit Message

fix includes in test files

Run Details

561 of 2354 relevant lines covered (23.83%)

2.96 hits per line

Source File
Press 'n' to go to next uncovered line, 'b' for previous

24.21

/src/WhiskerToolbox/DataManager/DataManager.cpp


#include "DataManager.hpp"
#include "AnalogTimeSeries/Analog_Time_Series.hpp"
#include "DigitalTimeSeries/Digital_Event_Series.hpp"
#include "DigitalTimeSeries/Digital_Interval_Series.hpp"
#include "Lines/Line_Data.hpp"
#include "Masks/Mask_Data.hpp"
#include "Media/Media_Data.hpp"
#include "Media/Video_Data.hpp"
#include "Points/Point_Data.hpp"
#include "Tensors/Tensor_Data.hpp"

#include "AnalogTimeSeries/Analog_Time_Series_Loader.hpp"
#include "DigitalTimeSeries/Digital_Event_Series_Loader.hpp"
#include "DigitalTimeSeries/Digital_Interval_Series_Loader.hpp"
#include "Lines/IO/CSV/Line_Data_CSV.hpp"
#include "Masks/IO/HDF5/Mask_Data_HDF5.hpp"
#include "Media/Video_Data_Loader.hpp"
#include "Points/IO/CSV/Point_Data_CSV.hpp"

#include "loaders/binary_loaders.hpp"
#include "transforms/data_transforms.hpp"
#include "transforms/Masks/mask_area.hpp"

#include "TimeFrame.hpp"

#include "nlohmann/json.hpp"
#include "utils/string_manip.hpp"

#include <filesystem>
#include <fstream>
#include <iostream>
#include <optional>
#include <regex>

using namespace nlohmann;

DataManager::DataManager() {
    _times["time"] = std::make_shared<TimeFrame>();
    _data["media"] = std::make_shared<MediaData>();

    setTimeFrame("media", "time");
    _output_path = std::filesystem::current_path();
}

bool DataManager::setTime(std::string const & key, std::shared_ptr<TimeFrame> timeframe) {

    if (!timeframe) {
        std::cerr << "Error: Cannot register a nullptr TimeFrame for key: " << key << std::endl;
        return false;
    }

    if (_times.find(key) != _times.end()) {
        std::cerr << "Error: Time key already exists in DataManager: " << key << std::endl;
        return false;
    }

    _times[key] = std::move(timeframe);
    return true;
}

std::shared_ptr<TimeFrame>  DataManager::getTime() {
    return _times["time"];
};

std::shared_ptr<TimeFrame> DataManager::getTime(std::string const & key) {
    if (_times.find(key) != _times.end()) {
        return _times[key];
    }
    return nullptr;
};

bool DataManager::removeTime(std::string const & key)
{
    if (_times.find(key) == _times.end()) {
        std::cerr << "Error: could not find time key in DataManager: " << key << std::endl;
        return false;
    }

    auto it = _times.find(key);
    _times.erase(it);
    return true;
}

bool DataManager::setTimeFrame(std::string const & data_key, std::string const & time_key) {
    if (_data.find(data_key) == _data.end()) {
        std::cerr << "Error: Data key not found in DataManager: " << data_key << std::endl;
        return false;
    }

    if (_times.find(time_key) == _times.end()) {
        std::cerr << "Error: Time key not found in DataManager: " << time_key << std::endl;
        return false;
    }

    _time_frames[data_key] = time_key;
    return true;
}

std::string DataManager::getTimeFrame(std::string const & data_key) {
    // check if data_key exists
    if (_data.find(data_key) == _data.end()) {
        std::cerr << "Error: Data key not found in DataManager: " << data_key << std::endl;
        return "";
    }

    // check if data key has time frame
    if (_time_frames.find(data_key) == _time_frames.end()) {
        std::cerr << "Error: Data key "
                  << data_key
                  << " exists, but not assigned to a TimeFrame" <<  std::endl;
        return "";
    }

    return _time_frames[data_key];
}

std::vector<std::string> DataManager::getTimeFrameKeys() {
    std::vector<std::string> keys;
    keys.reserve(_times.size());
    for (auto const & [key, value]: _times) {

        keys.push_back(key);
    }
    return keys;
}

int DataManager::addCallbackToData(std::string const & key, ObserverCallback callback) {

    int id = -1;

    if (_data.find(key) != _data.end()) {
        auto data = _data[key];

        id = std::visit([callback](auto & x) {
            return x.get()->addObserver(callback);
        }, data);
    }

    return id;
}

bool DataManager::removeCallbackFromData(std::string const & key, int callback_id) {
    if (_data.find(key) != _data.end()) {
        auto data = _data[key];

        std::visit([callback_id](auto & x) {
            x.get()->removeObserver(callback_id);
        }, data);

        return true;
    }

    return false;
}

void DataManager::addObserver(ObserverCallback callback) {
    _observers.push_back(std::move(callback));
}

void DataManager::_notifyObservers() {
    for (auto & observer: _observers) {
        observer();
    }
}

std::vector<std::string> DataManager::getAllKeys() {
    std::vector<std::string> keys;
    keys.reserve(_data.size());
    for (auto const & [key, value]: _data) {

        keys.push_back(key);
    }
    return keys;
}

std::optional<DataTypeVariant> DataManager::getDataVariant(std::string const & key) {
    if (_data.find(key) != _data.end()) {
        return _data[key];
    }
    return std::nullopt;
}

void DataManager::setData(std::string const & key, DataTypeVariant data) {
    _data[key] = data;
    setTimeFrame(key, "time");
    _notifyObservers();
}

std::optional<std::string> processFilePath(
        std::string const & file_path,
        std::filesystem::path const & base_path) {
    std::filesystem::path full_path = file_path;

    // Check for wildcard character
    if (file_path.find('*') != std::string::npos) {
        // Convert wildcard pattern to regex
        std::string const pattern = std::regex_replace(full_path.string(), std::regex("\\*"), ".*");
        std::regex const regex_pattern(pattern);

        // Iterate through the directory to find matching files
        for (auto const & entry: std::filesystem::directory_iterator(base_path)) {
            std::cout << "Checking " << entry.path().string() << " with full path " << full_path << std::endl;
            if (std::regex_match(entry.path().string(), regex_pattern)) {
                std::cout << "Loading file " << entry.path().string() << std::endl;
                return entry.path().string();
            }
        }
        return std::nullopt;
    } else {
        // Check if the file path is relative
        if (!std::filesystem::path(file_path).is_absolute()) {
            full_path = base_path / file_path;
        }
        // Check for the presence of the file
        if (std::filesystem::exists(full_path)) {
            std::cout << "Loading file " << full_path.string() << std::endl;
            return full_path.string();
        } else {
            return std::nullopt;
        }
    }
}

bool checkRequiredFields(json const & item, std::vector<std::string> const & requiredFields) {
    for (auto const & field: requiredFields) {
        if (!item.contains(field)) {
            std::cerr << "Error: Missing required field \"" << field << "\" in JSON item." << std::endl;
            return false;
        }
    }
    return true;
}

void checkOptionalFields(json const & item, std::vector<std::string> const & optionalFields) {
    for (auto const & field: optionalFields) {
        if (!item.contains(field)) {
            std::cout << "Warning: Optional field \"" << field << "\" is missing in JSON item." << std::endl;
        }
    }
}

DM_DataType stringToDataType(std::string const & data_type_str) {
    if (data_type_str == "video") return DM_DataType::Video;
    if (data_type_str == "points") return DM_DataType::Points;
    if (data_type_str == "mask") return DM_DataType::Mask;
    if (data_type_str == "line") return DM_DataType::Line;
    if (data_type_str == "analog") return DM_DataType::Analog;
    if (data_type_str == "digital_event") return DM_DataType::DigitalEvent;
    if (data_type_str == "digital_interval") return DM_DataType::DigitalInterval;
    if (data_type_str == "tensor") return DM_DataType::Tensor;
    if (data_type_str == "time") return DM_DataType::Time;
    return DM_DataType::Unknown;
}

std::vector<DataInfo> load_data_from_json_config(DataManager * dm, std::string const & json_filepath) {
    std::vector<DataInfo> data_info_list;
    // Open JSON file
    std::ifstream ifs(json_filepath);
    if (!ifs.is_open()) {
        std::cerr << "Failed to open JSON file: " << json_filepath << std::endl;
        return data_info_list;
    }

    // Parse JSON
    json j;
    ifs >> j;

    // get base path of filepath
    std::filesystem::path const base_path = std::filesystem::path(json_filepath).parent_path();

    // Iterate through JSON array
    for (auto const & item: j) {

        if (!checkRequiredFields(item, {"data_type", "name", "filepath"})) {
            continue;// Exit if any required field is missing
        }

        std::string const data_type_str = item["data_type"];
        auto const data_type = stringToDataType(data_type_str);
        if (data_type == DM_DataType::Unknown) {
            std::cout << "Unknown data type: " << data_type_str << std::endl;
            continue;
        }

        std::string const name = item["name"];

        auto file_exists = processFilePath(item["filepath"], base_path);
        if (!file_exists) {
            std::cerr << "File does not exist: " << item["filepath"] << std::endl;
            continue;
        }

        std::string const file_path = file_exists.value();

        switch (data_type) {
            case DM_DataType::Video: {

                auto video_data = load_video_into_VideoData(file_path);
                dm->setData<VideoData>("media", video_data);

                data_info_list.push_back({name, "VideoData", ""});
                break;
            }
            case DM_DataType::Points: {

                auto point_data = load_into_PointData(file_path, item);

                dm->setData<PointData>(name, point_data);

                std::string const color = item.value("color", "#0000FF");
                data_info_list.push_back({name, "PointData", color});
                break;
            }
            case DM_DataType::Mask: {

                auto mask_data = load_into_MaskData(file_path, item);

                std::string const color = item.value("color", "0000FF");
                dm->setData<MaskData>(name, mask_data);

                data_info_list.push_back({name, "MaskData", color});

                if (item.contains("operations")) {

                    for (auto const & operation: item["operations"]) {

                        std::string const operation_type = operation["type"];

                        if (operation_type == "area") {
                            std::cout << "Calculating area for mask: " << name << std::endl;
                            auto area_data = area(dm->getData<MaskData>(name).get());
                            std::string const output_name = name + "_area";
                            dm->setData<AnalogTimeSeries>(output_name, area_data);
                        }
                    }
                }
                break;
            }
            case DM_DataType::Line: {

                auto line_map = load_line_csv(file_path);

                //Get the whisker name from the filename using filesystem
                auto whisker_filename = std::filesystem::path(file_path).filename().string();

                //Remove .csv suffix from filename
                auto whisker_name = remove_extension(whisker_filename);

                dm->setData<LineData>(whisker_name, std::make_shared<LineData>(line_map));

                std::string const color = item.value("color", "0000FF");

                data_info_list.push_back({name, "LineData", color});

                break;
            }
            case DM_DataType::Analog: {

                auto analog_time_series = load_into_AnalogTimeSeries(file_path, item);

                for (int channel = 0; channel < analog_time_series.size(); channel++) {
                    std::string const channel_name = name + "_" + std::to_string(channel);

                    dm->setData<AnalogTimeSeries>(channel_name, analog_time_series[channel]);

                    if (item.contains("clock")) {
                        std::string const clock = item["clock"];
                        dm->setTimeFrame(channel_name, clock);
                    }
                }
                break;
            }
            case DM_DataType::DigitalEvent: {

                auto digital_event_series = load_into_DigitalEventSeries(file_path, item);

                for (int channel = 0; channel < digital_event_series.size(); channel++) {
                    std::string const channel_name = name + "_" + std::to_string(channel);

                    dm->setData<DigitalEventSeries>(channel_name, digital_event_series[channel]);

                    if (item.contains("clock")) {
                        std::string const clock = item["clock"];
                        dm->setTimeFrame(channel_name, clock);
                    }
                }
                break;
            }
            case DM_DataType::DigitalInterval: {

                auto digital_interval_series = load_into_DigitalIntervalSeries(file_path, item);
                dm->setData<DigitalIntervalSeries>(name, digital_interval_series);

                break;
            }
            case DM_DataType::Tensor: {

                if (item["format"] == "numpy") {

                    TensorData tensor_data;
                    loadNpyToTensorData(file_path, tensor_data);

                    dm->setData<TensorData>(name, std::make_shared<TensorData>(tensor_data));

                } else {
                    std::cout << "Format " << item["format"] << " not found for " << name << std::endl;
                }
                break;
            }
            case DM_DataType::Time: {

                if (item["format"] == "uint16") {

                    int const channel = item["channel"];
                    std::string const transition = item["transition"];

                    int const header_size = item.value("header_size", 0);

                    auto opts = Loader::BinaryAnalogOptions{.file_path = file_path,
                                                            .header_size_bytes = static_cast<size_t>(header_size)};
                    auto data = readBinaryFile<uint16_t>(opts);

                    auto digital_data = Loader::extractDigitalData(data, channel);
                    auto events = Loader::extractEvents(digital_data, transition);

                    // convert to int with std::transform
                    std::vector<int> events_int;
                    events_int.reserve(events.size());
                    for (auto e: events) {
                        events_int.push_back(static_cast<int>(e));
                    }
                    std::cout << "Loaded " << events_int.size() << " events for " << name << std::endl;

                    auto timeframe = std::make_shared<TimeFrame>(events_int);
                    dm->setTime(name, timeframe);
                }

                if (item["format"] == "uint16_length") {

                    int const header_size = item.value("header_size", 0);

                    auto opts = Loader::BinaryAnalogOptions{.file_path = file_path,
                                                            .header_size_bytes = static_cast<size_t>(header_size)};
                    auto data = readBinaryFile<uint16_t>(opts);

                    std::vector<int> t(data.size());
                    std::iota(std::begin(t), std::end(t), 0);

                    std::cout << "Total of " << t.size() << " timestamps for " << name << std::endl;

                    auto timeframe = std::make_shared<TimeFrame>(t);
                    dm->setTime(name, timeframe);
                }
                break;
            }
            default:
                std::cout << "Unsupported data type: " << data_type_str << std::endl;
                continue;
        }
        if (item.contains("clock")) {
            std::string const clock = item["clock"];
            std::cout << "Setting time for " << name << " to " << clock << std::endl;
            dm->setTimeFrame(name, clock);
        }
    }

    return data_info_list;
}

DM_DataType DataManager::getType(std::string const & key) const {
    auto it = _data.find(key);
    if (it != _data.end()) {
        if (std::holds_alternative<std::shared_ptr<MediaData>>(it->second)) {
            return DM_DataType::Video;
        } else if (std::holds_alternative<std::shared_ptr<PointData>>(it->second)) {
            return DM_DataType::Points;
        } else if (std::holds_alternative<std::shared_ptr<LineData>>(it->second)) {
            return DM_DataType::Line;
        } else if (std::holds_alternative<std::shared_ptr<MaskData>>(it->second)) {
            return DM_DataType::Mask;
        } else if (std::holds_alternative<std::shared_ptr<AnalogTimeSeries>>(it->second)) {
            return DM_DataType::Analog;
        } else if (std::holds_alternative<std::shared_ptr<DigitalEventSeries>>(it->second)) {
            return DM_DataType::DigitalEvent;
        } else if (std::holds_alternative<std::shared_ptr<DigitalIntervalSeries>>(it->second)) {
            return DM_DataType::DigitalInterval;
        } else if (std::holds_alternative<std::shared_ptr<TensorData>>(it->second)) {
            return DM_DataType::Tensor;
        }
        return DM_DataType::Unknown;
    }
    return DM_DataType::Unknown;
}

std::string convert_data_type_to_string(DM_DataType type) {
    switch (type) {
        case DM_DataType::Video:
            return "video";
        case DM_DataType::Points:
            return "points";
        case DM_DataType::Mask:
            return "mask";
        case DM_DataType::Line:
            return "line";
        case DM_DataType::Analog:
            return "analog";
        case DM_DataType::DigitalEvent:
            return "digital_event";
        case DM_DataType::DigitalInterval:
            return "digital_interval";
        case DM_DataType::Tensor:
            return "tensor";
        case DM_DataType::Time:
            return "time";
        default:
            return "unknown";
    }
}

1
2	#include "DataManager.hpp"
3	#include "AnalogTimeSeries/Analog_Time_Series.hpp"
4	#include "DigitalTimeSeries/Digital_Event_Series.hpp"
5	#include "DigitalTimeSeries/Digital_Interval_Series.hpp"
6	#include "Lines/Line_Data.hpp"
7	#include "Masks/Mask_Data.hpp"
8	#include "Media/Media_Data.hpp"
9	#include "Media/Video_Data.hpp"
10	#include "Points/Point_Data.hpp"
11	#include "Tensors/Tensor_Data.hpp"
12
13	#include "AnalogTimeSeries/Analog_Time_Series_Loader.hpp"
14	#include "DigitalTimeSeries/Digital_Event_Series_Loader.hpp"
15	#include "DigitalTimeSeries/Digital_Interval_Series_Loader.hpp"
16	#include "Lines/IO/CSV/Line_Data_CSV.hpp"
17	#include "Masks/IO/HDF5/Mask_Data_HDF5.hpp"
18	#include "Media/Video_Data_Loader.hpp"
19	#include "Points/IO/CSV/Point_Data_CSV.hpp"
20
21	#include "loaders/binary_loaders.hpp"
22	#include "transforms/data_transforms.hpp"
23	#include "transforms/Masks/mask_area.hpp"
24
25	#include "TimeFrame.hpp"
26
27	#include "nlohmann/json.hpp"
28	#include "utils/string_manip.hpp"
29
30	#include <filesystem>
31	#include <fstream>
32	#include <iostream>
33	#include <optional>
34	#include <regex>
35
36	using namespace nlohmann;
37
38	DataManager::DataManager() {	39✔
39	_times["time"] = std::make_shared<TimeFrame>();	117✔
40	_data["media"] = std::make_shared<MediaData>();	117✔
41
42	setTimeFrame("media", "time");	195✔
43	_output_path = std::filesystem::current_path();	39✔
44	}	39✔
45
46	bool DataManager::setTime(std::string const & key, std::shared_ptr<TimeFrame> timeframe) {	16✔
47
48	if (!timeframe) {	16✔
49	std::cerr << "Error: Cannot register a nullptr TimeFrame for key: " << key << std::endl;	1✔
50	return false;	1✔
51	}
52
53	if (_times.find(key) != _times.end()) {	15✔
54	std::cerr << "Error: Time key already exists in DataManager: " << key << std::endl;	2✔
55	return false;	2✔
56	}
57
58	_times[key] = std::move(timeframe);	13✔
59	return true;	13✔
60	}
61
62	std::shared_ptr<TimeFrame> DataManager::getTime() {	1✔
63	return _times["time"];	3✔
64	};
65
66	std::shared_ptr<TimeFrame> DataManager::getTime(std::string const & key) {	7✔
67	if (_times.find(key) != _times.end()) {	7✔
68	return _times[key];	5✔
69	}
70	return nullptr;	2✔
71	};
72
73	bool DataManager::removeTime(std::string const & key)	×
74	{
75	if (_times.find(key) == _times.end()) {	×
76	std::cerr << "Error: could not find time key in DataManager: " << key << std::endl;	×
77	return false;	×
78	}
79
80	auto it = _times.find(key);	×
81	_times.erase(it);	×
82	return true;	×
83	}
84
85	bool DataManager::setTimeFrame(std::string const & data_key, std::string const & time_key) {	78✔
86	if (_data.find(data_key) == _data.end()) {	78✔
87	std::cerr << "Error: Data key not found in DataManager: " << data_key << std::endl;	1✔
88	return false;	1✔
89	}
90
91	if (_times.find(time_key) == _times.end()) {	77✔
92	std::cerr << "Error: Time key not found in DataManager: " << time_key << std::endl;	1✔
93	return false;	1✔
94	}
95
96	_time_frames[data_key] = time_key;	76✔
97	return true;	76✔
98	}
99
100	std::string DataManager::getTimeFrame(std::string const & data_key) {	6✔
101	// check if data_key exists
102	if (_data.find(data_key) == _data.end()) {	6✔
103	std::cerr << "Error: Data key not found in DataManager: " << data_key << std::endl;	1✔
104	return "";	3✔
105	}
106
107	// check if data key has time frame
108	if (_time_frames.find(data_key) == _time_frames.end()) {	5✔
109	std::cerr << "Error: Data key "
110	<< data_key
111	<< " exists, but not assigned to a TimeFrame" << std::endl;	×
112	return "";	×
113	}
114
115	return _time_frames[data_key];	5✔
116	}
117
118	std::vector<std::string> DataManager::getTimeFrameKeys() {	8✔
119	std::vector<std::string> keys;	8✔
120	keys.reserve(_times.size());	8✔
121	for (auto const & [key, value]: _times) {	24✔
122
123	keys.push_back(key);	16✔
124	}
125	return keys;	8✔
126	}	×
127
128	int DataManager::addCallbackToData(std::string const & key, ObserverCallback callback) {	7✔
129
130	int id = -1;	7✔
131
132	if (_data.find(key) != _data.end()) {	7✔
133	auto data = _data[key];	6✔
134
135	id = std::visit([callback](auto & x) {	12✔
136	return x.get()->addObserver(callback);	6✔
137	}, data);
138	}	6✔
139
140	return id;	7✔
141	}
142
143	bool DataManager::removeCallbackFromData(std::string const & key, int callback_id) {	4✔
144	if (_data.find(key) != _data.end()) {	4✔
145	auto data = _data[key];	3✔
146
147	std::visit([callback_id](auto & x) {	6✔
148	x.get()->removeObserver(callback_id);	3✔
149	}, data);	3✔
150
151	return true;	3✔
152	}	3✔
153
154	return false;	1✔
155	}
156
157	void DataManager::addObserver(ObserverCallback callback) {	5✔
158	_observers.push_back(std::move(callback));	5✔
159	}	5✔
160
161	void DataManager::_notifyObservers() {	33✔
162	for (auto & observer: _observers) {	41✔
163	observer();	8✔
164	}
165	}	33✔
166
167	std::vector<std::string> DataManager::getAllKeys() {	5✔
168	std::vector<std::string> keys;	5✔
169	keys.reserve(_data.size());	5✔
170	for (auto const & [key, value]: _data) {	17✔
171
172	keys.push_back(key);	12✔
173	}
174	return keys;	5✔
175	}	×
176
177	std::optional<DataTypeVariant> DataManager::getDataVariant(std::string const & key) {	×
178	if (_data.find(key) != _data.end()) {	×
179	return _data[key];	×
180	}
181	return std::nullopt;	×
182	}
183
184	void DataManager::setData(std::string const & key, DataTypeVariant data) {	1✔
185	_data[key] = data;	1✔
186	setTimeFrame(key, "time");	3✔
187	_notifyObservers();	1✔
188	}	1✔
189
190	std::optional<std::string> processFilePath(	×
191	std::string const & file_path,
192	std::filesystem::path const & base_path) {
193	std::filesystem::path full_path = file_path;	×
194
195	// Check for wildcard character
196	if (file_path.find('*') != std::string::npos) {	×
197	// Convert wildcard pattern to regex
198	std::string const pattern = std::regex_replace(full_path.string(), std::regex("\\"), ".");	×
199	std::regex const regex_pattern(pattern);	×
200
201	// Iterate through the directory to find matching files
202	for (auto const & entry: std::filesystem::directory_iterator(base_path)) {	×
203	std::cout << "Checking " << entry.path().string() << " with full path " << full_path << std::endl;	×
204	if (std::regex_match(entry.path().string(), regex_pattern)) {	×
205	std::cout << "Loading file " << entry.path().string() << std::endl;	×
206	return entry.path().string();	×
207	}
208	}	×
209	return std::nullopt;	×
210	} else {	×
211	// Check if the file path is relative
212	if (!std::filesystem::path(file_path).is_absolute()) {	×
213	full_path = base_path / file_path;	×
214	}
215	// Check for the presence of the file
216	if (std::filesystem::exists(full_path)) {	×
217	std::cout << "Loading file " << full_path.string() << std::endl;	×
218	return full_path.string();	×
219	} else {
220	return std::nullopt;	×
221	}
222	}
223	}	×
224
225	bool checkRequiredFields(json const & item, std::vector<std::string> const & requiredFields) {	×
226	for (auto const & field: requiredFields) {	×
227	if (!item.contains(field)) {	×
228	std::cerr << "Error: Missing required field \"" << field << "\" in JSON item." << std::endl;	×
229	return false;	×
230	}
231	}
232	return true;	×
233	}
234
235	void checkOptionalFields(json const & item, std::vector<std::string> const & optionalFields) {	×
236	for (auto const & field: optionalFields) {	×
237	if (!item.contains(field)) {	×
238	std::cout << "Warning: Optional field \"" << field << "\" is missing in JSON item." << std::endl;	×
239	}
240	}
241	}	×
242
243	DM_DataType stringToDataType(std::string const & data_type_str) {	×
244	if (data_type_str == "video") return DM_DataType::Video;	×
245	if (data_type_str == "points") return DM_DataType::Points;	×
246	if (data_type_str == "mask") return DM_DataType::Mask;	×
247	if (data_type_str == "line") return DM_DataType::Line;	×
248	if (data_type_str == "analog") return DM_DataType::Analog;	×
249	if (data_type_str == "digital_event") return DM_DataType::DigitalEvent;	×
250	if (data_type_str == "digital_interval") return DM_DataType::DigitalInterval;	×
251	if (data_type_str == "tensor") return DM_DataType::Tensor;	×
252	if (data_type_str == "time") return DM_DataType::Time;	×
253	return DM_DataType::Unknown;	×
254	}
255
256	std::vector<DataInfo> load_data_from_json_config(DataManager * dm, std::string const & json_filepath) {	×
257	std::vector<DataInfo> data_info_list;	×
258	// Open JSON file
259	std::ifstream ifs(json_filepath);	×
260	if (!ifs.is_open()) {	×
261	std::cerr << "Failed to open JSON file: " << json_filepath << std::endl;	×
262	return data_info_list;	×
263	}
264
265	// Parse JSON
266	json j;	×
267	ifs >> j;	×
268
269	// get base path of filepath
270	std::filesystem::path const base_path = std::filesystem::path(json_filepath).parent_path();	×
271
272	// Iterate through JSON array
273	for (auto const & item: j) {	×
274
275	if (!checkRequiredFields(item, {"data_type", "name", "filepath"})) {	×
276	continue;// Exit if any required field is missing	×
277	}
278
279	std::string const data_type_str = item["data_type"];	×
280	auto const data_type = stringToDataType(data_type_str);	×
281	if (data_type == DM_DataType::Unknown) {	×
282	std::cout << "Unknown data type: " << data_type_str << std::endl;	×
283	continue;	×
284	}
285
286	std::string const name = item["name"];	×
287
288	auto file_exists = processFilePath(item["filepath"], base_path);	×
289	if (!file_exists) {	×
290	std::cerr << "File does not exist: " << item["filepath"] << std::endl;	×
291	continue;	×
292	}
293
294	std::string const file_path = file_exists.value();	×
295
296	switch (data_type) {	×
297	case DM_DataType::Video: {	×
298
299	auto video_data = load_video_into_VideoData(file_path);	×
300	dm->setData<VideoData>("media", video_data);	×
301
302	data_info_list.push_back({name, "VideoData", ""});	×
303	break;	×
304	}	×
305	case DM_DataType::Points: {	×
306
307	auto point_data = load_into_PointData(file_path, item);	×
308
309	dm->setData<PointData>(name, point_data);	×
310
311	std::string const color = item.value("color", "#0000FF");	×
312	data_info_list.push_back({name, "PointData", color});	×
313	break;	×
314	}	×
315	case DM_DataType::Mask: {	×
316
317	auto mask_data = load_into_MaskData(file_path, item);	×
318
319	std::string const color = item.value("color", "0000FF");	×
320	dm->setData<MaskData>(name, mask_data);	×
321
322	data_info_list.push_back({name, "MaskData", color});	×
323
324	if (item.contains("operations")) {	×
325
326	for (auto const & operation: item["operations"]) {	×
327
328	std::string const operation_type = operation["type"];	×
329
330	if (operation_type == "area") {	×
331	std::cout << "Calculating area for mask: " << name << std::endl;	×
332	auto area_data = area(dm->getData<MaskData>(name).get());	×
333	std::string const output_name = name + "_area";	×
334	dm->setData<AnalogTimeSeries>(output_name, area_data);	×
335	}	×
336	}	×
337	}
338	break;	×
339	}	×
340	case DM_DataType::Line: {	×
341
342	auto line_map = load_line_csv(file_path);	×
343
344	//Get the whisker name from the filename using filesystem
345	auto whisker_filename = std::filesystem::path(file_path).filename().string();	×
346
347	//Remove .csv suffix from filename
348	auto whisker_name = remove_extension(whisker_filename);	×
349
350	dm->setData<LineData>(whisker_name, std::make_shared<LineData>(line_map));	×
351
352	std::string const color = item.value("color", "0000FF");	×
353
354	data_info_list.push_back({name, "LineData", color});	×
355
356	break;	×
357	}	×
358	case DM_DataType::Analog: {	×
359
360	auto analog_time_series = load_into_AnalogTimeSeries(file_path, item);	×
361
362	for (int channel = 0; channel < analog_time_series.size(); channel++) {	×
363	std::string const channel_name = name + "_" + std::to_string(channel);	×
364
365	dm->setData<AnalogTimeSeries>(channel_name, analog_time_series[channel]);	×
366
367	if (item.contains("clock")) {	×
368	std::string const clock = item["clock"];	×
369	dm->setTimeFrame(channel_name, clock);	×
370	}	×
371	}	×
372	break;	×
373	}	×
374	case DM_DataType::DigitalEvent: {	×
375
376	auto digital_event_series = load_into_DigitalEventSeries(file_path, item);	×
377
378	for (int channel = 0; channel < digital_event_series.size(); channel++) {	×
379	std::string const channel_name = name + "_" + std::to_string(channel);	×
380
381	dm->setData<DigitalEventSeries>(channel_name, digital_event_series[channel]);	×
382
383	if (item.contains("clock")) {	×
384	std::string const clock = item["clock"];	×
385	dm->setTimeFrame(channel_name, clock);	×
386	}	×
387	}	×
388	break;	×
389	}	×
390	case DM_DataType::DigitalInterval: {	×
391
392	auto digital_interval_series = load_into_DigitalIntervalSeries(file_path, item);	×
393	dm->setData<DigitalIntervalSeries>(name, digital_interval_series);	×
394
395	break;	×
396	}	×
397	case DM_DataType::Tensor: {	×
398
399	if (item["format"] == "numpy") {	×
400
401	TensorData tensor_data;	×
402	loadNpyToTensorData(file_path, tensor_data);	×
403
404	dm->setData<TensorData>(name, std::make_shared<TensorData>(tensor_data));	×
405
406	} else {	×
407	std::cout << "Format " << item["format"] << " not found for " << name << std::endl;	×
408	}
409	break;	×
410	}
411	case DM_DataType::Time: {	×
412
413	if (item["format"] == "uint16") {	×
414
415	int const channel = item["channel"];	×
416	std::string const transition = item["transition"];	×
417
418	int const header_size = item.value("header_size", 0);	×
419
420	auto opts = Loader::BinaryAnalogOptions{.file_path = file_path,	×
421	.header_size_bytes = static_cast<size_t>(header_size)};	×
422	auto data = readBinaryFile<uint16_t>(opts);	×
423
424	auto digital_data = Loader::extractDigitalData(data, channel);	×
425	auto events = Loader::extractEvents(digital_data, transition);	×
426
427	// convert to int with std::transform
428	std::vector<int> events_int;	×
429	events_int.reserve(events.size());	×
430	for (auto e: events) {	×
431	events_int.push_back(static_cast<int>(e));	×
432	}
433	std::cout << "Loaded " << events_int.size() << " events for " << name << std::endl;	×
434
435	auto timeframe = std::make_shared<TimeFrame>(events_int);	×
436	dm->setTime(name, timeframe);	×
437	}	×
438
439	if (item["format"] == "uint16_length") {	×
440
441	int const header_size = item.value("header_size", 0);	×
442
443	auto opts = Loader::BinaryAnalogOptions{.file_path = file_path,	×
444	.header_size_bytes = static_cast<size_t>(header_size)};	×
445	auto data = readBinaryFile<uint16_t>(opts);	×
446
447	std::vector<int> t(data.size());	×
448	std::iota(std::begin(t), std::end(t), 0);	×
449
450	std::cout << "Total of " << t.size() << " timestamps for " << name << std::endl;	×
451
452	auto timeframe = std::make_shared<TimeFrame>(t);	×
453	dm->setTime(name, timeframe);	×
454	}	×
455	break;	×
456	}
457	default:	×
458	std::cout << "Unsupported data type: " << data_type_str << std::endl;	×
459	continue;	×
460	}	×
461	if (item.contains("clock")) {	×
462	std::string const clock = item["clock"];	×
463	std::cout << "Setting time for " << name << " to " << clock << std::endl;	×
464	dm->setTimeFrame(name, clock);	×
465	}	×
466	}	×
467
468	return data_info_list;
469	}	×
470
471	DM_DataType DataManager::getType(std::string const & key) const {	×
472	auto it = _data.find(key);	×
473	if (it != _data.end()) {	×
474	if (std::holds_alternative<std::shared_ptr<MediaData>>(it->second)) {	×
475	return DM_DataType::Video;	×
476	} else if (std::holds_alternative<std::shared_ptr<PointData>>(it->second)) {	×
477	return DM_DataType::Points;	×
478	} else if (std::holds_alternative<std::shared_ptr<LineData>>(it->second)) {	×
479	return DM_DataType::Line;	×
480	} else if (std::holds_alternative<std::shared_ptr<MaskData>>(it->second)) {	×
481	return DM_DataType::Mask;	×
482	} else if (std::holds_alternative<std::shared_ptr<AnalogTimeSeries>>(it->second)) {	×
483	return DM_DataType::Analog;	×
484	} else if (std::holds_alternative<std::shared_ptr<DigitalEventSeries>>(it->second)) {	×
485	return DM_DataType::DigitalEvent;	×
486	} else if (std::holds_alternative<std::shared_ptr<DigitalIntervalSeries>>(it->second)) {	×
487	return DM_DataType::DigitalInterval;	×
488	} else if (std::holds_alternative<std::shared_ptr<TensorData>>(it->second)) {	×
489	return DM_DataType::Tensor;	×
490	}
491	return DM_DataType::Unknown;	×
492	}
493	return DM_DataType::Unknown;	×
494	}
495
496	std::string convert_data_type_to_string(DM_DataType type) {	×
497	switch (type) {	×
498	case DM_DataType::Video:	×
499	return "video";	×
500	case DM_DataType::Points:	×
501	return "points";	×
502	case DM_DataType::Mask:	×
503	return "mask";	×
504	case DM_DataType::Line:	×
505	return "line";	×
506	case DM_DataType::Analog:	×
507	return "analog";	×
508	case DM_DataType::DigitalEvent:	×
509	return "digital_event";	×
510	case DM_DataType::DigitalInterval:	×
511	return "digital_interval";	×
512	case DM_DataType::Tensor:	×
513	return "tensor";	×
514	case DM_DataType::Time:	×
515	return "time";	×
516	default:	×
517	return "unknown";	×
518	}
519	}

paulmthompson / WhiskerToolbox / 15046951426

Source File Press 'n' to go to next uncovered line, 'b' for previous

Source File
Press 'n' to go to next uncovered line, 'b' for previous