%reload_ext autoreload
%autoreload 2
%matplotlib inline

#!pip install EXCAT-Sync
#!pip install requests_toolbelt

import numpy as np
from pathlib import Path
import openslide
from tqdm import tqdm
import cv2
import pandas as pd
import math
import matplotlib.pyplot as plt

from exact_sync.v1.api.annotations_api import AnnotationsApi
from exact_sync.v1.api.images_api import ImagesApi
from exact_sync.v1.api.image_sets_api import ImageSetsApi
from exact_sync.v1.api.annotation_types_api import AnnotationTypesApi
from exact_sync.v1.api.products_api import ProductsApi
from exact_sync.v1.api.teams_api import TeamsApi

from exact_sync.v1.models import ImageSet, Team, Product, AnnotationType, Annotation, AnnotationMediaFile
from exact_sync.v1.rest import ApiException
from exact_sync.v1.configuration import Configuration
from exact_sync.v1.api_client import ApiClient

configuration = Configuration()
configuration.username = 'exact'
configuration.password = 'exact'
configuration.host = "http://127.0.0.1:8000"

client = ApiClient(configuration)

image_sets_api = ImageSetsApi(client)
annotations_api = AnnotationsApi(client)
annotation_types_api = AnnotationTypesApi(client)
images_api = ImagesApi(client)
product_api = ProductsApi(client)

images = {}
annotation_types = {}

total_annotations = {}
for image_set in [169]:
    total_annotations[image_set] = []
    for image in images_api.list_images(pagination=False, image_set=image_set).results:
        images[image.name] = image
    for product in [138]:
        for anno_type in annotation_types_api.list_annotation_types(product=product).results:
            annotation_types[anno_type.id] = anno_type

context_offset = 0.15
patch_width = 100
patch_height = 100
result_image_names = {}

result_annotations = []
for label in annotations["label"].unique():
    
    labelDF = annotations[annotations["label"] == label]
        
    annotation_count = len(labelDF)
    
    x_images = math.ceil(math.sqrt(annotation_count))
    y_images = math.ceil(math.sqrt(annotation_count))
    
    x_total_size = int(x_images * patch_width)
    y_total_size = int(y_images * patch_height)
    
    result_image = np.zeros(shape=(y_total_size, x_total_size, 3), dtype=np.uint8)
    
    ids = list(labelDF.index)
    name = '{0}_map.tiff'.format(label)
    result_image_names[name] = annotation_count
    
    x_total_max = 0
    y_total_max = 0
    
    for x_row in range(x_images):
        x_start = x_total_max
        y_max = 0
    
        for y_row in range(y_images):
            if len(ids) == 0:
                break
                
            id = ids.pop()
            anno = annotations.iloc[id]
            
            source_image = cv2.imread(anno.file)
            
            x_ori = int(anno.vector["x1"])
            y_ori = int(anno.vector["y1"])
            w_ori = int(anno.vector["x2"] - x_ori)
            h_ori = int(anno.vector["y2"] - y_ori)
            
            patch = source_image[y_ori:y_ori+h_ori,x_ori:x_ori+w_ori]
            if patch.shape[0] <= 1 or patch.shape[1] <= 1:
                continue
            
            patch = cv2.resize(patch, (patch_width, patch_height))
            
            x_min = x_start
            x_max = x_min + patch.shape[1]
            x_total_max = x_max if x_max > x_total_max else x_total_max

            y_min = y_max
            y_max = y_min + patch.shape[0]
            y_total_max = y_max if y_max > y_total_max else y_total_max

            result_image[y_min:y_max, x_min:x_max] = patch
            
            vector = {'x1': x_min, "y1": y_min, "x2": x_max, "y2": y_max}
            result_annotations.append([name, label, vector])
             
    cv2.imwrite(name, result_image)

result_annotations.to_csv("labels.txt", sep="|", header=False, index=False)