def test_iiif_info_json():
img1 = IIIFImageClient("http://image.server/path/", "myimgid")
img2 = IIIFImageClient("http://image.server/path/", "myimgid2")
imgs = [{"image": img1}, {"image": img2}]
json_ids = corpus_extras.iiif_info_json(imgs)
# should contain the same ids but with /info.json appended
assert "http://image.server/path/myimgid/info.json" in json_ids
assert "http://image.server/path/myimgid2/info.json" in json_ids
def loris_info(image_id): # instantiate IIIFImageClient ic = IIIFImageClient(api_endpoint=localConfig.LORIS_API_ENDPOINT,image_id=image_id) # debug url info_url = ic.info() print "loris info url: %s" % info_url r = requests.get(info_url).json() return jsonify(r)
def test_iiif_image():
# copied from mep_django
myimg = IIIFImageClient("http://image.server/path/", "myimgid")
# check expected behavior
assert str(corpus_extras.iiif_image(myimg, "size:width=250")) == str(
myimg.size(width=250))
assert str(corpus_extras.iiif_image(myimg,
"size:width=250,height=300")) == str(
myimg.size(width=250, height=300))
assert str(corpus_extras.iiif_image(myimg, "format:png")) == str(
myimg.format("png"))
# check that errors don't raise exceptions
assert corpus_extras.iiif_image(myimg, "bogus") == ""
assert corpus_extras.iiif_image(myimg, "size:bogus") == ""
assert corpus_extras.iiif_image(myimg, "size:bogus=1") == ""
def iiif_client(self):
if not settings.IIIF_IMAGE_BASE_URL or not self.jp2_abs_filename:
return None
else:
return IIIFImageClient(
settings.IIIF_IMAGE_BASE_URL,
quote(os.path.join(self.issue.batch.name, "data",
self.jp2_filename),
safe=""),
)
def generate_deep_zoom(self, jekyll_site_dir):
self.log_status('Downloading deep zoom images')
imagedir = os.path.join(jekyll_site_dir, self.image_dir)
# If exporting to GitHub, add the repo name to the path.
_prefix = os.path.join(os.sep, self.github_repo, self.image_dir) if self.github_repo else os.path.join(os.sep, self.image_dir)
staticgen = IIIFStatic(dst=imagedir, prefix=_prefix)
for teipage in self.tei.page_list:
for graphic in teipage.graphics:
if graphic.rend == 'full':
imgsrc = os.path.join(jekyll_site_dir, graphic.url)
iiif_img = IIIFImageClient.init_from_url(graphic.url)
staticgen.generate(imgsrc, identifier=iiif_img.image_id)
def generate_deep_zoom(self, jekyll_site_dir):
self.log_status('Downloading deep zoom images')
imagedir = os.path.join(jekyll_site_dir, self.image_dir)
# If exporting to GitHub, add the repo name to the path.
_prefix = os.path.join(
os.sep, self.github_repo,
self.image_dir) if self.github_repo else os.path.join(
os.sep, self.image_dir)
staticgen = IIIFStatic(dst=imagedir, prefix=_prefix)
for teipage in self.tei.page_list:
for graphic in teipage.graphics:
if graphic.rend == 'full':
imgsrc = os.path.join(jekyll_site_dir, graphic.url)
iiif_img = IIIFImageClient.init_from_url(graphic.url)
staticgen.generate(imgsrc, identifier=iiif_img.image_id)
def get_image_url_from_iiif_url(iiif_url, input_size=224):
try:
image = IIIFImageClient.init_from_url(iiif_url)
except ParseError:
raise ValueError(f"{iiif_url} is not a valid iiif URL")
if "dlcs" in image.api_endpoint:
# DLCS provides a thumbnails service which only serves certain sizes of
# image. Requests for these don't touch the image server and so, as
# we're performing lots of requests, we use 400x400 thumbnails and
# resize them ourselves later on.
image.api_endpoint = image.api_endpoint.replace("iiif-img", "thumbs")
# `exact=True` equates to a `/!400,400/` image request
# https://iiif.io/api/image/1.0/#4-2-size
return str(image.size(width=400, height=400, exact=True))
return str(image.size(width=input_size, height=input_size, exact=False))
def image_tile(request, path, width, height, x1, y1, x2, y2):
width, height = map(int, (width, height))
x1, y1, x2, y2 = map(int, (x1, y1, x2, y2))
if settings.REDIRECT_IMAGES_TO_IIIF:
iiif_client = IIIFImageClient(settings.IIIF_IMAGE_BASE_URL,
quote(path, safe=""))
return redirect_to_iiif(request, iiif_client, width, height, x1, y1,
x2, y2)
else:
try:
p = os.path.join(settings.BATCH_STORAGE, path)
im = Image.open(p)
return serve_image_tile(request, im, width, height, x1, y1, x2, y2)
except EnvironmentError as e:
logging.exception("Unable to create image tile for %s", path)
return HttpResponseServerError("Unable to create image tile: %s" %
e)
def iiif_url_to_local_path(self, imgurl):
'''Convert an IIIF Image API url to a local path that can be
referenced within the jekyll site.
'''
# check if we have an info url, so that it can be duplicated
is_info = imgurl.endswith('info.json')
iiif_img = IIIFImageClient.init_from_url(imgurl)
# convert api endpoint to local path
iiif_img.api_endpoint = self.image_dir
# simplify image id: strip out iiif id prefix and pidspace
# prefix, if possible
# (leaving any id suffix, since those are likely needed to
# guarantee uniqueness)
iiif_img.image_id = iiif_img.image_id \
.replace(settings.IIIF_ID_PREFIX, '') \
.replace('%s:' % settings.FEDORA_PIDSPACE, '')
# serialize updated iiif image url for use as local image path
if is_info:
return unicode(iiif_img.info())
return unicode(iiif_img)
def iiif_url_to_local_path(self, imgurl):
'''Convert an IIIF Image API url to a local path that can be
referenced within the jekyll site.
'''
# check if we have an info url, so that it can be duplicated
is_info = imgurl.endswith('info.json')
iiif_img = IIIFImageClient.init_from_url(imgurl)
# convert api endpoint to local path
iiif_img.api_endpoint = self.image_dir
# simplify image id: strip out iiif id prefix and pidspace
# prefix, if possible
# (leaving any id suffix, since those are likely needed to
# guarantee uniqueness)
iiif_img.image_id = iiif_img.image_id \
.replace(settings.IIIF_ID_PREFIX, '') \
.replace('%s:' % settings.FEDORA_PIDSPACE, '')
# serialize updated iiif image url for use as local image path
if is_info:
return unicode(iiif_img.info())
return unicode(iiif_img)
def index_traversing(size, index_name):
# run for 1 time and get a last_result
es = create_es_client()
#####document this after first run
# # create a new index
# es.indices.create(
# index='lab_clustering_hex',
# )
#####
response = es.search(
index=os.environ['INDEX_NAME'],
body={
"query": {
"match_all": {}
},
"size": size,
"from": 0,
"sort": {
"_id": "desc", # document ID.
},
})
# post the first 20 dictionaries
for ID_url_dict in Extracting_url_ID(response):
image_url = str(IIIFImageClient().init_from_url(
ID_url_dict['url']).size(width=50))
cluster_centers_list, cluster_number, cluster_size, cluster_proportion, lab_centers_list = clustering_api_request(
image_url) # Clustering in lab space.
cluster_centers_list_hex = rgblist2hexlist(cluster_centers_list)
searchstr = labcenters2searchterm(cluster_number, lab_centers_list)
# update the document
es.index(index=index_name,
id=ID_url_dict['id'],
body={
"iiif_url": ID_url_dict['url'],
"image_url": image_url,
"cluster_info": {
"cluster_centers_list": cluster_centers_list,
"cluster_centers_list_hex": cluster_centers_list_hex,
"cluster_number": cluster_number,
"cluster_size": cluster_size,
"cluster_proportion": cluster_proportion,
"lab_centers_list": lab_centers_list,
"searchstr": searchstr,
}
})
print(ID_url_dict['id'])
last_result_id = response['hits']['hits'][size - 1]['_id']
counter = size # just to make the waiting time less painful
while len(response['hits']['hits']) == size: # get in the loop!
response = es.search(
index=os.environ['INDEX_NAME'],
body={
"query": {
"match_all": {}
},
"size": size,
"from": 0,
"sort": {
"_id": "desc", # document ID.
},
"search_after": [last_result_id],
})
for ID_url_dict in Extracting_url_ID(
response): # post the dictionaries
image_url = str(IIIFImageClient().init_from_url(
ID_url_dict['url']).size(width=50))
cluster_centers_list, cluster_number, cluster_size, cluster_proportion, lab_centers_list = clustering_api_request(
image_url)
cluster_centers_list_hex = rgblist2hexlist(cluster_centers_list)
searchstr = labcenters2searchterm(cluster_number, lab_centers_list)
es.index(index=index_name,
id=ID_url_dict['id'],
body={
"iiif_url": ID_url_dict['url'],
"image_url": image_url,
"cluster_info": {
"cluster_centers_list": cluster_centers_list,
"cluster_centers_list_hex":
cluster_centers_list_hex,
"cluster_number": cluster_number,
"cluster_size": cluster_size,
"cluster_proportion": cluster_proportion,
"lab_centers_list": lab_centers_list,
"searchstr": searchstr,
}
})
print(ID_url_dict['id']
) # just to make the waiting time less painful
last_result_id = ID_url_dict['id'] # update the last ID
counter += size
print(counter)