From 12ce6eefb05286f4af64fd8e4afdc370619d1426 Mon Sep 17 00:00:00 2001
From: onb1259 <onb1259@onb.ac.at>
Date: Wed, 15 Nov 2023 13:15:48 +0100
Subject: [PATCH] updated iiif_utils to newest version
---
iiif_utils.py | 59 ++++++++++++++++++++++++++++++++++++++++-----------
1 file changed, 47 insertions(+), 12 deletions(-)
diff --git a/iiif_utils.py b/iiif_utils.py
index b739cb4..70e7860 100644
--- a/iiif_utils.py
+++ b/iiif_utils.py
@@ -2,6 +2,7 @@ import requests
import json
import threading
import os
+from time import sleep
def get_json_by_url(url):
@@ -23,32 +24,66 @@ def get_imgurls_from_manifesturl(url):
def get_imgurls_from_collectionurl(url):
res = []
- i = 0
for manifest_url in get_manifesturls_from_collectionurl(url):
tmp = get_imgurls_from_manifesturl(manifest_url)
- print(i)
- i += 1
for el in tmp:
res.append(el)
return res
+def downlaod_image(img_url, img_path):
+ img_bytes = requests.get(img_url).content
+ with open(img_path, '+wb') as img_file:
+ img_file.write(img_bytes)
+
def download_images(img_urls, img_paths):
for u, n in zip(img_urls, img_paths):
- img_bytes = requests.get(u).content
- with open(n, '+wb') as img_file:
- img_file.write(img_bytes)
+ downlaod_image(u, n)
-def download_images_multithreded(img_urls, img_paths, nb_threads=5):
- chunk_idcs = list(range(0, len(img_urls), int(len(img_urls)/nb_threads))) + [len(img_urls)]
+def download_images_multithreded(img_urls, img_paths, nb_threads=10):
+ chunk_idcs = list(range(0, len(img_urls), int(len(img_urls)/nb_threads) if int(len(img_urls)/nb_threads)!=0 else 1)) + [len(img_urls)]
chunks_urls = [img_urls[i:j] for i,j in zip(chunk_idcs[:-1], chunk_idcs[1:])]
chunks_paths = [img_paths[i:j] for i,j in zip(chunk_idcs[:-1], chunk_idcs[1:])]
+ i = 0
for urls, paths in zip(chunks_urls, chunks_paths):
- threading.Thread(target=download_images, args=[urls, paths]).start()
+ threading.Thread(target=download_images, args=[urls, paths], name=f'DownloadThread-{i}').start()
+ i += 1
+ # wait for all download threads to finish
+ while any(['DownloadThread' in e.name for e in threading.enumerate()]):
+ sleep(1)
def create_paths_from_iiifurls(img_urls, base_path=''):
- res = [el.split('images/')[1] for el in img_urls]
- res = ['_'.join(el.split('/')[:-4]) for el in res]
- res = [el+'.'+u.split('.')[-1] for el, u in zip(res, img_urls)]
+ res = [e.replace(':', '-').replace('/', '_') for e in img_urls]
res = [os.path.join(base_path, el) for el in res]
return res
+def add_annotations(list_of_boundingboxes, manifest):
+ for i, _ in enumerate(manifest['sequences'][0]['canvases']):
+ id = manifest['sequences'][0]['canvases'][i]['images'][0]['@id']
+ bounding_box = list_of_boundingboxes[i]
+ # if bounding box is None, skip
+ if not bounding_box:
+ continue
+ annotation = {
+ "@context": "http://iiif.io/api/presentation/2/context.json",
+ "@type": "sc:AnnotationList",
+ "@id": f"{id}/annotations",
+ "resources": [
+ {
+ "@type": "oa:Annotation",
+ "motivation": "sc:painting",
+ "resource": {
+ "@type": "oa:SpecificResource",
+ "selector": {
+ "@type": "oa:FragmentSelector",
+ "value": f"xywh={bounding_box[0]},{bounding_box[1]},{bounding_box[0]+bounding_box[2]},{bounding_box[2]+bounding_box[3]}"
+ },
+ "style": "rect",
+ "label": "Detected figure"
+ },
+ "on": f"{id}"
+ }
+ ]
+ }
+ manifest['sequences'][0]['canvases'][i]['otherContent'].append(annotation)
+ return manifest
+
--
GitLab