ari/minerva
1
0
Fork 0
forked from bellwether/minerva
Code Pull requests Activity

add tempfile deletion to query; ls to s3

Browse source
This commit is contained in:
Ari Brown 2023-08-07 18:10:10 -04:00
parent 1d509f544c
commit 972e6dd210
2 changed files with 23 additions and 9 deletions
Download patch file Download diff file Expand all files Collapse all files

19
minerva/athena.py
View file

@ -39,6 +39,7 @@ class Execute:
self.sql = sql self.sql = sql
self.params = [str(p) for p in params] self.params = [str(p) for p in params]
self.info_cache = None self.info_cache = None
self.files = []
def query(self): def query(self):
return self.sql return self.sql
@ -96,13 +97,11 @@ class Query(Execute):
# Because we're using `UNLOAD`, we get a manifest of the files # Because we're using `UNLOAD`, we get a manifest of the files
# that make up our data. # that make up our data.
manif = self.info_cache['Statistics']['DataManifestLocation'] manif = self.info_cache['Statistics']['DataManifestLocation']
tmp = self.handler.s3.download(manif) files = self.handler.s3.read(manif).split("\n")
with open(tmp, "r") as f:
txt = f.read()
files = txt.strip().split("\n")
files = [f.strip() for f in files if f.strip()] # filter empty files = [f.strip() for f in files if f.strip()] # filter empty
self.files = files
return files return files
else: else:
print("Error") print("Error")
@ -116,3 +115,13 @@ class Query(Execute):
self.ds = pa.dataset.dataset(local) self.ds = pa.dataset.dataset(local)
return self.ds return self.ds
def __enter__(self):
return self
def __exit__(self):
self.close()
def close(self):
for file in self.files:
os.remove(file)

9
minerva/s3.py
View file

@ -6,10 +6,10 @@ class S3:
def __init__(self, handler): def __init__(self, handler):
self.handler = handler self.handler = handler
self.s3 = handler.session.client("s3") self.s3 = handler.session.client("s3")
self.resource = handler.session.resource("s3")
def parse(self, uri): def parse(self, uri):
bucket = uri.split("/")[2] _, _, bucket, file = uri.split("/", 3)
file = os.path.join(*uri.split("/")[3:])
return bucket, file return bucket, file
def download(self, uri, loc=None): def download(self, uri, loc=None):
@ -31,3 +31,8 @@ class S3:
bucket, key = self.parse(remote) bucket, key = self.parse(remote)
return self.s3.upload_file(local, bucket, key) return self.s3.upload_file(local, bucket, key)
def ls(self, uri, **kwargs):
bucket, key = self.parse(uri)
#self.s3.list_objects_v2(Bucket=bucket, Prefix=key)
return self.resource.Bucket(bucket).objects.filter(Prefix=key, **kwargs)