def test_internet_archive():
from datetime import timedelta
from django.utils import timezone
import internetarchive
from perma.models import Link
from django.template.defaultfilters import truncatechars
start_date = timezone.now() - timedelta(days=3)
end_date = timezone.now() - timedelta(days=2)
links = Link.objects.filter(
internet_archive_upload_status="completed", creation_timestamp__range=(start_date, end_date)
)
guid_results = dict()
all_results = dict()
c = {"s3": {"access": settings.INTERNET_ARCHIVE_ACCESS_KEY, "secret": settings.INTERNET_ARCHIVE_SECRET_KEY}}
internetarchive.get_session(config=c)
for link in links:
identifier = settings.INTERNET_ARCHIVE_IDENTIFIER_PREFIX + link.guid
item = internetarchive.get_item(identifier)
warc_name = "%s.warc.gz" % link.guid
try:
fnames = [f.name for f in internetarchive.get_files(identifier, glob_pattern="*gz")]
guid_results["uploaded_file"] = warc_name in fnames
if settings.INTERNET_ARCHIVE_COLLECTION == "test_collection":
guid_results["collection"] = item.metadata["collection"] == settings.INTERNET_ARCHIVE_COLLECTION
else:
guid_results["collection"] = item.metadata["collection"][0] == settings.INTERNET_ARCHIVE_COLLECTION
guid_results["title"] = item.metadata["title"] == "%s: %s" % (
link.guid,
truncatechars(link.submitted_title, 50),
)
guid_results["mediatype"] = item.metadata["mediatype"] == "web"
guid_results["description"] = item.metadata["description"] == "Perma.cc archive of %s created on %s." % (
link.submitted_url,
link.creation_timestamp,
)
guid_results["contributor"] = item.metadata["contributor"] == "Perma.cc"
guid_results["submitted_url"] = item.metadata["submitted_url"] == link.submitted_url
guid_results["perma_url"] = item.metadata["perma_url"] == "http://%s/%s" % (settings.HOST, link.guid)
guid_results["external-identifier"] = item.metadata["external-identifier"] == "urn:X-perma:%s" % link.guid
if link.organization:
guid_results["organization"] = item.metadata["sponsor"] == "%s - %s" % (
link.organization,
link.organization.registrar,
)
except Exception as e:
guid_results["error"] = e
pass
all_results[link.guid] = guid_results
print all_results
def test_get_session_with_config_file(tmpdir):
tmpdir.chdir()
test_conf = """[s3]\naccess = key2"""
with open("ia_test.ini", "w") as fh:
fh.write(test_conf)
s = get_session(config_file="ia_test.ini")
assert s.access_key == "key2"
def test_get_files_with_get_item_kwargs(tmpdir):
tmpdir.chdir()
with responses.RequestsMock(assert_all_requests_are_fired=False) as rsps:
rsps.add(responses.GET, "{0}//archive.org/metadata/nasa".format(protocol), body=ITEM_METADATA, status=200)
s = get_session(config={"s3": {"access": "key"}})
files = get_files("nasa", files="nasa_meta.xml", archive_session=s)
files = list(files)
assert len(files) == 1
assert files[0].name == "nasa_meta.xml"
files = get_files("nasa", files="nasa_meta.xml", config={"logging": {"level": "INFO"}})
files = list(files)
assert len(files) == 1
assert files[0].name == "nasa_meta.xml"
test_conf = """[s3]\naccess = key2"""
with open("ia_test.ini", "w") as fh:
fh.write(test_conf)
files = get_files("nasa", files="nasa_meta.xml", config_file="ia_test.ini")
files = list(files)
assert len(files) == 1
assert files[0].name == "nasa_meta.xml"
files = get_files("nasa", files="nasa_meta.xml", http_adapter_kwargs={"max_retries": 3})
files = list(files)
assert len(files) == 1
assert files[0].name == "nasa_meta.xml"
files = get_files("nasa", files="nasa_meta.xml", request_kwargs={"timeout": 4})
files = list(files)
assert len(files) == 1
assert files[0].name == "nasa_meta.xml"
def get_ia_session():
try:
assert config.has_key('s3/access_key')
assert config.has_key('s3/secret_key')
assert config.has_key('cookie')
assert config.has_key('email')
except:
return None
ia_session = get_session(
config={
'general': {
'secure': True
},
's3': {
'access': config.get('s3/access_key'),
'secret': config.get('s3/secret_key')
},
'cookies': {
'logged-in-user': config.get('email'),
'logged-in-sig': config.get('cookie')
},
},
http_adapter_kwargs={'max_retries': 10},
)
return ia_session
def getStats():
today = date.today()
back30_days = (datetime.now() - timedelta(days=30)).date()
back7_days = (datetime.now() - timedelta(days=7)).date()
collection = 'greekgovernmentgazette'
query_all = f'collection:({collection})'
res = {}
try:
s = get_session()
s.mount_http_adapter()
search_results = s.search_items(query_all,
fields=['identifier', 'addeddate'])
lst_res = list(search_results)
docs_last30days = [
i for i in lst_res
if isodate.parse_date(i['addeddate']) >= back30_days
]
docs_last7days = [
i for i in lst_res
if isodate.parse_date(i['addeddate']) >= back7_days
]
docs_today = [
i for i in lst_res
if isodate.parse_date(i['addeddate']) == today
]
res['count_all'] = len(lst_res)
res['count_last30days'] = len(docs_last30days)
res['count_last7days'] = len(docs_last7days)
res['count_today'] = len(docs_today)
finally:
return res
def __init__(self,
archive_id,
metadata=None,
config_file_path=None,
repo_base=None):
"""
:param archive_id:
:param config_file_path:
:param repo_base: In archive item, place each file in a folder mirroring its local location.
"""
self.repo_base = repo_base
self.archive_id = archive_id
self.archive_session = internetarchive.get_session(
config_file=config_file_path)
self.archive_item = internetarchive.get_item(
archive_id, config_file=config_file_path)
self.metadata = metadata
logging.info(self.archive_item.identifier)
self.original_item_files = list(
filter(
lambda x: x["source"] == "original" and not x["name"].
startswith(self.archive_item.identifier) and not x[
"name"].startswith("_"), self.archive_item.files))
self.original_item_file_names = sorted(
map(lambda x: x["name"], self.original_item_files))
def create_subcollection(collection_id, metadata=None, parent_collection=None):
"""
The expected sub-collection hierarchy is as follows top-level OSF collection -> provider
collection -> collection for nodes with multiple children -> all only child nodes
:param metadata: dict should attributes for the provider's sub-collection is being created
:param parent_collection: str the name of the sub-collection's parent
:return:
"""
if metadata is None:
metadata = {}
session = internetarchive.get_session(
config={
"s3": {"access": settings.IA_ACCESS_KEY, "secret": settings.IA_SECRET_KEY},
},
)
collection = internetarchive.Item(session, collection_id)
collection.upload(
files={"dummy.txt": BytesIO(b"dummy")},
metadata={
"mediatype": "collection",
"collection": parent_collection,
**metadata,
},
)
def get_ia_item(guid):
session = internetarchive.get_session(
config={
"s3": {"access": settings.IA_ACCESS_KEY, "secret": settings.IA_SECRET_KEY},
},
)
return session.get_item(guid)
def test_get_session_with_config_file(tmpdir):
tmpdir.chdir()
test_conf = """[s3]\naccess = key2"""
with open('ia_test.ini', 'w') as fh:
fh.write(test_conf)
s = get_session(config_file='ia_test.ini')
assert s.access_key == 'key2'
def stream_from_pbox(itemname, filename):
# TODO: handle errors etc
archive_session = get_session(config_file=settings.IATOOL_CONFIG_PATH)
item = archive_session.get_item(itemname)
files = item.get_files(filename)
file = files.__next__()
return file.download(return_responses=True)
def stream_from_pbox(itemname, filename):
# TODO: handle errors etc
archive_session = get_session(config_file=settings.IATOOL_CONFIG_PATH)
item = archive_session.get_item(itemname)
files = item.get_files(filename)
file = files.__next__()
return file.download(return_responses=True)
def new_session():
global SESSION
if SESSION is not None:
raise Exception('have another session!')
SESSION = get_session(config=CONFIG)
def test_get_session_with_config_file(tmpdir):
tmpdir.chdir()
test_conf = """[s3]\naccess = key2"""
with open('ia_test.ini', 'w') as fh:
fh.write(test_conf)
s = get_session(config_file='ia_test.ini')
assert s.access_key == 'key2'
def test_internet_archive():
from datetime import timedelta
from django.utils import timezone
import internetarchive
from perma.models import Link
from django.template.defaultfilters import truncatechars
start_date = timezone.now() - timedelta(days=3)
end_date = timezone.now() - timedelta(days=2)
links = Link.objects.filter(internet_archive_upload_status="completed", creation_timestamp__range=(start_date, end_date))
guid_results = dict()
all_results = dict()
c = {"s3":{"access":settings.INTERNET_ARCHIVE_ACCESS_KEY, "secret":settings.INTERNET_ARCHIVE_SECRET_KEY}}
internetarchive.get_session(config=c)
for link in links:
identifier = settings.INTERNET_ARCHIVE_IDENTIFIER_PREFIX + link.guid
item = internetarchive.get_item(identifier)
warc_name = "%s.warc.gz" % link.guid
try:
fnames = [f.name for f in internetarchive.get_files(identifier, glob_pattern="*gz")]
guid_results["uploaded_file"] = warc_name in fnames
if settings.INTERNET_ARCHIVE_COLLECTION == 'test_collection':
guid_results["collection"] = item.metadata["collection"] == settings.INTERNET_ARCHIVE_COLLECTION
else:
guid_results["collection"] = item.metadata["collection"][0] == settings.INTERNET_ARCHIVE_COLLECTION
guid_results["title"] = item.metadata["title"] == "%s: %s" % (link.guid, truncatechars(link.submitted_title, 50))
guid_results["mediatype"] = item.metadata["mediatype"]=="web"
guid_results["description"] = item.metadata["description"]=="Perma.cc archive of %s created on %s." % (link.submitted_url, link.creation_timestamp,)
guid_results["contributor"] = item.metadata["contributor"]=="Perma.cc"
guid_results["submitted_url"] = item.metadata["submitted_url"]==link.submitted_url
guid_results["perma_url"] = item.metadata["perma_url"]=="http://%s/%s" % (settings.HOST, link.guid)
guid_results["external-identifier"] = item.metadata["external-identifier"]=="urn:X-perma:%s" % link.guid
if link.organization:
guid_results["organization"] = item.metadata["sponsor"] == "%s - %s" % (link.organization, link.organization.registrar)
except Exception as e:
guid_results["error"] = e
pass
all_results[link.guid] = guid_results
print all_results
def createSession(self):
iaKey = decryptEnvVar('IA_ACCESS_KEY')
iaSecret = decryptEnvVar('IA_SECRET_KEY')
return get_session(
config={'s3': {
'access': iaKey,
'secret': iaSecret
}})
def start_ia_session(self):
""" starts an internet archive session """
config = dict(s3=dict(acccess=settings.INTERNET_ARCHIVE_ACCESS_KEY,
secret=settings.INTERNET_ARCHIVE_SECRET_KEY))
s = get_session(config=config, debug=True)
s.access_key = settings.INTERNET_ARCHIVE_ACCESS_KEY
s.secret_key = settings.INTERNET_ARCHIVE_SECRET_KEY
return s
def test_get_item_with_archive_session():
with responses.RequestsMock() as rsps:
rsps.add(responses.GET, '{0}//archive.org/metadata/nasa'.format(protocol),
body=ITEM_METADATA,
status=200)
s = get_session(config={'s3': {'access': 'key3'}})
item = get_item('nasa', archive_session=s)
assert item.session.access_key == 'key3'
def test_get_item_with_archive_session():
with responses.RequestsMock() as rsps:
rsps.add(responses.GET, '{0}//archive.org/metadata/nasa'.format(protocol),
body=ITEM_METADATA,
status=200)
s = get_session(config={'s3': {'access': 'key3'}})
item = get_item('nasa', archive_session=s)
assert item.session.access_key == 'key3'
def test_upload_secure_session():
with IaRequestsMock(assert_all_requests_are_fired=False) as rsps:
s = get_session(config={"general": {"secure": True}})
rsps.add_metadata_mock("nasa")
item = s.get_item("nasa")
with IaRequestsMock(assert_all_requests_are_fired=False) as rsps:
rsps.add(responses.PUT, S3_URL_RE)
r = item.upload(NASA_METADATA_PATH)
assert r[0].url == "https://s3.us.archive.org/nasa/nasa.json"
def start_ia_session(self):
""" starts an internet archive session """
config = dict(s3=dict(acccess=settings.INTERNET_ARCHIVE_ACCESS_KEY,
secret=settings.INTERNET_ARCHIVE_SECRET_KEY))
s = get_session(config=config,
debug=True)
s.access_key = settings.INTERNET_ARCHIVE_ACCESS_KEY
s.secret_key = settings.INTERNET_ARCHIVE_SECRET_KEY
return s
def upload_to_ia(force=set()):
s = get_session()
item = s.get_item("NotoFonts")
hashdict = {f["name"]: f["md5"] for f in item.files}
fonts_modified = False
for path in tqdm(sorted(pathset)):
filename = path.name
file = open(path, "rb").read()
hash = md5(file).hexdigest()
if "fonts" not in force:
try:
if hashdict[filename] == hash:
print("SKIPPING: " + filename)
continue
except KeyError:
pass
fonts_modified = True
print("WORKING: " + filename)
upload_paths = []
ttf = TTFont(path)
print(" CONVERTING TO woff2...")
ttf.flavor = "woff2"
woff2_path = "upload/" + path.with_suffix(".woff2").name
try:
ttf.save(open(woff2_path, "wb"))
upload_paths.append(woff2_path)
except TTLibError:
print("could not convert to woff2")
print(" CONVERTING TO woff...")
ttf.flavor = "woff"
woff_path = "upload/" + path.with_suffix(".woff").name
ttf.save(open(woff_path, "wb"))
upload_paths.append(woff_path)
print(" UPLOADING...")
r = item.upload(files=[*upload_paths, str(path)], retries=100)
for upath in [woff2_path, woff_path]:
remove(upath)
if "css" in force or fonts_modified:
from generate_css import build_all_css
print(" GENERATING CSS...")
build_all_css()
css_files = glob("*.css")
for path in [Path(p) for p in sorted(css_files)]:
filename = path.name
file = open(path, "rb").read()
hash = md5(file).hexdigest()
# if "css" not in force:
try:
if hashdict[filename] == hash:
print("SKIPPING: " + filename)
continue
except KeyError:
pass
print(" UPLOADING " + filename)
r = item.upload(files=css_files, retries=100)
def test_upload_secure_session():
with IaRequestsMock(assert_all_requests_are_fired=False) as rsps:
s = get_session(config={'general': {'secure': True}})
rsps.add_metadata_mock('nasa')
item = s.get_item('nasa')
with IaRequestsMock(assert_all_requests_are_fired=False) as rsps:
rsps.add(responses.PUT, S3_URL_RE)
r = item.upload(NASA_METADATA_PATH)
assert r[0].url == 'https://s3.us.archive.org/nasa/nasa.json'
def test_get_session_with_config():
s = get_session(config={
's3': {
'access': 'key'
},
'gengeral': {
'secure': False
}
})
assert s.access_key == 'key'
def __init__(
self,
issue_db: IssueDB,
sandcrawler_db_client: SandcrawlerPostgrestClient,
sandcrawler_s3_client: SandcrawlerMinioClient,
):
self.issue_db: IssueDB = issue_db
self.ia_client = internetarchive.get_session()
self.sandcrawler_db_client = sandcrawler_db_client
self.sandcrawler_s3_client = sandcrawler_s3_client
def test_upload_secure_session():
with IaRequestsMock(assert_all_requests_are_fired=False) as rsps:
c = {'s3': {'access': 'foo', 'secret': 'bar'}, 'general': {'secure': True}}
s = get_session(config=c)
rsps.add_metadata_mock('nasa')
item = s.get_item('nasa')
with IaRequestsMock(assert_all_requests_are_fired=False) as rsps:
rsps.add(responses.PUT, S3_URL_RE)
r = item.upload(NASA_METADATA_PATH)
assert r[0].url == 'https://s3.us.archive.org/nasa/nasa.json'
def get_stats(days=30):
back_days = (datetime.now() - timedelta(days=days)).date()
query_all = 'collection:(greekgovernmentgazette) AND date:[{1} TO {0}]' \
.format(datetime.now().strftime('%Y-%m-%d'), back_days.strftime('%Y-%m-%d'))
s = get_session()
s.mount_http_adapter()
search_results = s.search_items(query_all,
fields=['identifier', 'addeddate'])
return '{}\t{}'.format(query_all, len(search_results))
def test_upload_secure_session(testitem_metadata, json_filename):
with responses.RequestsMock(assert_all_requests_are_fired=False) as rsps:
s = get_session(config={'general': {'secure': True}})
rsps.add(responses.GET, 'https://archive.org/metadata/nasa',
body=testitem_metadata,
status=200)
item = s.get_item('nasa')
with responses.RequestsMock(
assert_all_requests_are_fired=False) as rsps:
rsps.add(responses.PUT, S3_URL_RE, status=200)
r = item.upload(json_filename)
assert r[0].url == 'https://s3.us.archive.org/nasa/nasa_meta.json'
def test_upload_secure_session(testitem_metadata, json_filename):
with responses.RequestsMock(assert_all_requests_are_fired=False) as rsps:
s = get_session(config={'general': {'secure': True}})
rsps.add(responses.GET,
'https://archive.org/metadata/nasa',
body=testitem_metadata,
status=200)
item = s.get_item('nasa')
with responses.RequestsMock(
assert_all_requests_are_fired=False) as rsps:
rsps.add(responses.PUT, S3_URL_RE, status=200)
r = item.upload(json_filename)
assert r[0].url == 'https://s3.us.archive.org/nasa/nasa_meta.json'
def __init__(self, file_storage, access_key, secret_key, loglevel, logfile):
self.file_storage = file_storage
self.access_key = access_key
self.secret_key = secret_key
session_data = {'access': access_key, 'secret': secret_key}
if logfile:
logconfig = {'logging': {'level': loglevel, 'file': logfile}}
else:
logconfig = {'logging': {'level': loglevel}}
self.session = get_session({'s3': session_data, 'logging': logconfig})
self.logger = logging.getLogger('iasync')
def __init__(self, top_dir, access_key, secret_key, loglevel, logfile):
self.top_dir = top_dir
self.access_key = access_key
self.secret_key = secret_key
session_data = {'access': access_key, 'secret': secret_key}
if logfile:
logconfig = {'logging': {'level': loglevel, 'file': logfile}}
else:
logconfig = {'logging': {'level': loglevel}}
self.session = get_session({'s3': session_data, 'logging': logconfig})
self.logger = logging.getLogger('gvision.ia')
def ia_get_session() -> ArchiveSession:
"""
Get an IA session based on the configuration supplied via environment
variables.
Because get_session() starts with values found in ~/.ia for the current
user then merges in the "additional" supplied values, the configuration
file reference is explicitly eliminated. For desktop testing, this
guarantees that the application has the same dependence on environment
variables as it would when deployed.
"""
return internetarchive.get_session(IA_CONFIG, config_file='/dev/null')
def main(collection: str, name: str, concurrency: int, dummy: bool=False,
dummy_text: str=None):
if dummy:
t = DummyTrainer(dummy_text)
else:
urls = from_cdx_url(cdx_url(collection), session=internetarchive.get_session())
t = Trainer(urls, concurrency=concurrency)
identifier = str(int(time.time()))
filename_base = collection + '_dictionary_' + identifier
upload_urls = t.upload(filename_base + '.zstdict.zst', filename_base)
print(upload_urls)
add_entry(identifier, name, t.sha256, upload_urls['public_url'],
upload_urls['backup_url'])
def __init__(self,
ckm_repo: CkanMetaRepo,
ia_access: str,
ia_secret: str,
ia_collection: str,
token: str = None) -> None:
self.ckm_repo = ckm_repo
self.ia_collection = ia_collection
self.ia_access = ia_access
self.ia_session = internetarchive.get_session(
config={'s3': {
'access': ia_access,
'secret': ia_secret,
}})
self._gh = github.Github(token) if token else github.Github()
def get_dictionary(filename: str) -> zstandard.ZstdCompressionDict:
s = internetarchive.get_session()
r = s.get('https://archive.org/download/' + filename,
headers={'Range': 'bytes=0-7'})
if r.content[:4] != b'\x5D\x2A\x4D\x18':
return None
data_size = struct.unpack('<L', r.content[4:])[0]
r = s.get('https://archive.org/download/' + filename,
headers={'Range': 'bytes=8-{}'.format(8 + data_size - 1)})
dictionary = r.content
if r.content[:4] == b'\x28\xB5\x2F\xFD':
dictionary = zstandard.ZstdDecompressor().decompress(dictionary)
if dictionary[:4] != b'\x37\xA4\x30\xEC':
raise ValueError('Not a dictionary.')
return zstandard.ZstdCompressionDict(dictionary)
def main():
session = internetarchive.get_session()
if len(sys.argv) == 3:
item_name = sys.argv[1]
release_id = sys.argv[2]
item_to_fileset(item_name, release_id=release_id, session=session)
else:
for line in sys.stdin:
line = line.strip()
if not line:
continue
fields = line.split("\t")
assert len(fields) == 2
item_name = fields[0]
release_id = fields[1]
item_to_fileset(item_name, release_id=release_id, session=session)
def test_get_files_with_get_item_kwargs(tmpdir):
tmpdir.chdir()
with responses.RequestsMock(assert_all_requests_are_fired=False) as rsps:
rsps.add(responses.GET,
'{0}//archive.org/metadata/nasa'.format(protocol),
body=ITEM_METADATA,
status=200)
s = get_session(config={'s3': {'access': 'key'}})
files = get_files('nasa', files='nasa_meta.xml', archive_session=s)
files = list(files)
assert len(files) == 1
assert files[0].name == 'nasa_meta.xml'
files = get_files('nasa',
files='nasa_meta.xml',
config={'logging': {
'level': 'INFO'
}})
files = list(files)
assert len(files) == 1
assert files[0].name == 'nasa_meta.xml'
test_conf = """[s3]\naccess = key2"""
with open('ia_test.ini', 'w') as fh:
fh.write(test_conf)
files = get_files('nasa',
files='nasa_meta.xml',
config_file='ia_test.ini')
files = list(files)
assert len(files) == 1
assert files[0].name == 'nasa_meta.xml'
files = get_files('nasa',
files='nasa_meta.xml',
http_adapter_kwargs={'max_retries': 3})
files = list(files)
assert len(files) == 1
assert files[0].name == 'nasa_meta.xml'
files = get_files('nasa',
files='nasa_meta.xml',
request_kwargs={'timeout': 4})
files = list(files)
assert len(files) == 1
assert files[0].name == 'nasa_meta.xml'
def test_get_files_with_get_item_kwargs(tmpdir):
tmpdir.chdir()
with IaRequestsMock(assert_all_requests_are_fired=False) as rsps:
rsps.add_metadata_mock('nasa')
s = get_session(config={'s3': {'access': 'key'}})
files = get_files('nasa', files='nasa_meta.xml', archive_session=s)
files = list(files)
assert len(files) == 1
assert files[0].name == 'nasa_meta.xml'
files = get_files('nasa',
files='nasa_meta.xml',
config={'logging': {
'level': 'INFO'
}})
files = list(files)
assert len(files) == 1
assert files[0].name == 'nasa_meta.xml'
test_conf = """[s3]\naccess = key2"""
with open('ia_test.ini', 'w') as fh:
fh.write(test_conf)
files = get_files('nasa',
files='nasa_meta.xml',
config_file='ia_test.ini')
files = list(files)
assert len(files) == 1
assert files[0].name == 'nasa_meta.xml'
files = get_files('nasa',
files='nasa_meta.xml',
http_adapter_kwargs={'max_retries': 3})
files = list(files)
assert len(files) == 1
assert files[0].name == 'nasa_meta.xml'
files = get_files('nasa',
files='nasa_meta.xml',
request_kwargs={'timeout': 4})
files = list(files)
assert len(files) == 1
assert files[0].name == 'nasa_meta.xml'
def cdx_url(collection: str) -> str:
session = internetarchive.get_session()
response = session.get(
'https://archive.org/advancedsearch.php',
params={
'q': (
'collection:archiveteam* '
'AND format:(Item CDX Index) '
'AND identifier:{}*'.format(collection)
),
'fl[]': 'identifier',
'sort[]': 'addeddate desc',
'rows': '1',
'output': 'json',
'scope': 'all'
}
).json()
print(response)
identifier = response['response']['docs'][0]['identifier']
return 'https://archive.org/download/{0}/{0}.cdx.gz'.format(identifier)
def test_get_files_with_get_item_kwargs(tmpdir):
tmpdir.chdir()
with responses.RequestsMock(
assert_all_requests_are_fired=False) as rsps:
rsps.add(responses.GET, '{0}//archive.org/metadata/nasa'.format(protocol),
body=ITEM_METADATA,
status=200)
s = get_session(config={'s3': {'access': 'key'}})
files = get_files('nasa', files='nasa_meta.xml', archive_session=s)
files = list(files)
assert len(files) == 1
assert files[0].name == 'nasa_meta.xml'
files = get_files('nasa',
files='nasa_meta.xml',
config={'logging': {'level': 'INFO'}})
files = list(files)
assert len(files) == 1
assert files[0].name == 'nasa_meta.xml'
test_conf = """[s3]\naccess = key2"""
with open('ia_test.ini', 'w') as fh:
fh.write(test_conf)
files = get_files('nasa', files='nasa_meta.xml',
config_file='ia_test.ini')
files = list(files)
assert len(files) == 1
assert files[0].name == 'nasa_meta.xml'
files = get_files('nasa',
files='nasa_meta.xml',
http_adapter_kwargs={'max_retries': 3})
files = list(files)
assert len(files) == 1
assert files[0].name == 'nasa_meta.xml'
files = get_files('nasa', files='nasa_meta.xml',
request_kwargs={'timeout': 4})
files = list(files)
assert len(files) == 1
assert files[0].name == 'nasa_meta.xml'
def __init__(self, **kwargs):
super().__init__()
self.ingest_strategy = IngestStrategy.ArchiveorgFileset
# TODO: enable cleanup when confident (eg, safe path parsing)
self.skip_cleanup_local_files = kwargs.get("skip_cleanup_local_files",
True)
self.working_dir = os.environ.get("SANDCRAWLER_WORKING_DIR",
"/tmp/sandcrawler/")
try:
os.mkdir(self.working_dir)
except FileExistsError:
pass
self.http_session = requests_retry_session()
self.ia_session = internetarchive.get_session(
config={
"s3": {
"access": os.environ.get("IA_ACCESS_KEY"),
"secret": os.environ.get("IA_SECRET_KEY"),
},
})
def test_get_files_with_get_item_kwargs(tmpdir):
tmpdir.chdir()
with IaRequestsMock(assert_all_requests_are_fired=False) as rsps:
rsps.add_metadata_mock('nasa')
s = get_session(config={'s3': {'access': 'key'}})
files = get_files('nasa', files='nasa_meta.xml', archive_session=s)
files = list(files)
assert len(files) == 1
assert files[0].name == 'nasa_meta.xml'
files = get_files('nasa',
files='nasa_meta.xml',
config={'logging': {'level': 'INFO'}})
files = list(files)
assert len(files) == 1
assert files[0].name == 'nasa_meta.xml'
test_conf = """[s3]\naccess = key2"""
with open('ia_test.ini', 'w') as fh:
fh.write(test_conf)
files = get_files('nasa', files='nasa_meta.xml',
config_file='ia_test.ini')
files = list(files)
assert len(files) == 1
assert files[0].name == 'nasa_meta.xml'
files = get_files('nasa',
files='nasa_meta.xml',
http_adapter_kwargs={'max_retries': 3})
files = list(files)
assert len(files) == 1
assert files[0].name == 'nasa_meta.xml'
files = get_files('nasa', files='nasa_meta.xml',
request_kwargs={'timeout': 4})
files = list(files)
assert len(files) == 1
assert files[0].name == 'nasa_meta.xml'
def archive_url(data: typing.Dict[str, bytes],
url_data: typing.Tuple[str, int, int, str, bool]):
url, offset, length, filename, redownload = url_data
if redownload:
data[url] = requests.get(url, allow_redirects=False).content
else:
if filename.endswith('.zst'):
with dictionary_lock:
dictionary = get_dictionary(filename)
r = internetarchive.get_session().get(
'https://archive.org/download/' + filename,
headers={
'Range': 'bytes={}-{}'.format(offset, offset + length - 1)
})
if filename.endswith('.zst'):
data[url] = zstandard.ZstdDecompressor(dict_data=dictionary) \
.decompressobj().decompress(r.content)
elif filename.endswith('.gz'):
data[url] = gzip.decompress(r.content)
elif filename.endswith('.warc'):
data[url] = r.content
else:
raise ValueError('WARC type not supported.')
print(len(data[url]), url)
logger.warning("Timeout or unknown RequestException. Unable to upload "
"to IA. Trying again if retries not exceeded: %s" % rd)
if self.request.retries == self.max_retries:
# Give up for now. It'll get done next time cron is run.
return
raise self.retry(exc=exc)
if all(r.ok for r in responses):
rd.filepath_ia = "https://archive.org/download/%s/%s" % (
bucket_name, file_name)
rd.save(do_extraction=False, index=False)
access_key = settings.IA_ACCESS_KEY
secret_key = settings.IA_SECRET_KEY
session = ia.get_session({'s3': {
'access': access_key,
'secret': secret_key,
}})
def upload_to_ia(identifier, files, metadata=None):
"""Upload an item and its files to the Internet Archive
On the Internet Archive there are Items and files. Items have a global
identifier, and files go inside the item:
https://internetarchive.readthedocs.io/en/latest/items.html
This function mirrors the IA library's similar upload function, but builds
in retries and various assumptions that make sense. Note that according to
emails with IA staff, it is best to maximize the number of files uploaded to
an Item at a time, rather than uploading each file in a separate go.
def test_get_session_with_config():
s = get_session(config={'s3': {'access': 'key'}, 'gengeral': {'secure': False}})
assert s.access_key == 'key'
def test_get_session_with_config():
s = get_session(config={'s3': {'access': 'key'}})
assert s.access_key == 'key'
def test_get_session_with_config():
s = get_session(config={"s3": {"access": "key"}})
assert s.access_key == "key"
inc_path = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
sys.path.insert(0, inc_path)
from copy import deepcopy
import responses
from internetarchive.cli import ia_list
from internetarchive import get_session
protocol = 'https:'
ROOT_DIR = os.getcwd()
TEST_JSON_FILE = os.path.join(ROOT_DIR, 'tests/data/nasa_meta.json')
SESSION = get_session()
with open(TEST_JSON_FILE, 'r') as fh:
ITEM_METADATA = fh.read().strip()
NASA_FILES = set([
'NASAarchiveLogo.jpg',
'globe_west_540.jpg',
'nasa_reviews.xml',
'nasa_meta.xml',
'nasa_archive.torrent',
'nasa_files.xml'
])
def test_ia_list(capsys):
with responses.RequestsMock() as rsps:
def nasa_item():
session = get_session()
with IaRequestsMock() as mocker:
mocker.add_metadata_mock('nasa')
yield session.get_item('nasa')
def session():
return get_session(config=dict(s3=dict(access='access', secret='secret')))
def session_with_logging():
return get_session(config={'logging': {'level': 'INFO'}})
def session():
return get_session()
def test_get_item_with_archive_session():
with responses.RequestsMock() as rsps:
rsps.add(responses.GET, "{0}//archive.org/metadata/nasa".format(protocol), body=ITEM_METADATA, status=200)
s = get_session(config={"s3": {"access": "key3"}})
item = get_item("nasa", archive_session=s)
assert item.session.access_key == "key3"
def test_get_item_with_archive_session(nasa_mocker):
s = get_session(config={'s3': {'access': 'key3'}})
item = get_item('nasa', archive_session=s)
assert item.session.access_key == 'key3'