Skip to content

fix: Download from http with stream enabled #407

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 2 commits into from
Aug 6, 2024
Merged

fix: Download from http with stream enabled #407

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
69c3108
fix: download with stream
wphyojpl Aug 6, 2024
c75b339
chore: checking other test cases
wphyojpl Aug 6, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion cumulus_lambda_functions/stage_in_out/download_granules_http.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -24,7 +24,7 @@ def _set_props_from_env(self):
return self

def _download_one_item(self, downloading_url):
downloading_response = requests.get(downloading_url)
downloading_response = requests.get(downloading_url, stream=True)
downloading_response.raise_for_status()
downloading_response.raw.decode_content = True
local_file_path = os.path.join(self._download_dir, os.path.basename(downloading_url))
Expand Down
154 changes: 151 additions & 3 deletions tests/integration_tests/test_docker_stage_in.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,3 +1,6 @@
import os
os.environ['OBJC_DISABLE_INITIALIZE_FORK_SAFETY'] = 'YES'
os.environ['no_proxy'] = '*'
import logging

import requests
Expand All @@ -7,7 +10,6 @@
import math
from unittest.mock import patch, MagicMock
import json
import os
import tempfile
from glob import glob
from sys import argv
Expand Down Expand Up @@ -1226,10 +1228,155 @@ def test_02_download__from_file(self):
download_result = download_result['features']
self.assertTrue('assets' in download_result[0], f'no assets in download_result: {download_result}')
downloaded_file_hrefs = set([k['assets']['data']['href'] for k in download_result])
print(downloaded_file_hrefs)
for each_granule in zip(granule_json['features'], download_result):
remote_filename = os.path.basename(each_granule[0]['assets']['data']['href'])
self.assertTrue(os.path.join('.', remote_filename) in downloaded_file_hrefs,
f'mismatched: {remote_filename}')
self.assertTrue(FileUtils.get_size(os.path.join(downloading_dir, remote_filename)) > 0, f'empty file: {remote_filename}')
print()
self.assertTrue(FileUtils.file_exist(os.environ['OUTPUT_FILE']), f'missing output file')
return

def test_02_download__from_file_large_github_data_file(self):
granule_json = '''
{
"numberMatched": 20,
"numberReturned": 20,
"stac_version": "1.0.0",
"type": "FeatureCollection",
"links": [
{
"rel": "self",
"href": "https://58nbcawrvb.execute-api.us-west-2.amazonaws.com/test/am-uds-dapa/collections/SNDR_SNPP_ATMS_L1A___1/items?datetime=2016-01-14T08:00:00Z/2016-01-14T11:59:59Z&limit=100&offset=0"
},
{
"rel": "root",
"href": "https://58nbcawrvb.execute-api.us-west-2.amazonaws.com"
},
{
"rel": "next",
"href": "https://58nbcawrvb.execute-api.us-west-2.amazonaws.com/test/am-uds-dapa/collections/SNDR_SNPP_ATMS_L1A___1/items?datetime=2016-01-14T08:00:00Z/2016-01-14T11:59:59Z&limit=100&offset=100"
},
{
"rel": "prev",
"href": "https://58nbcawrvb.execute-api.us-west-2.amazonaws.com/test/am-uds-dapa/collections/SNDR_SNPP_ATMS_L1A___1/items?datetime=2016-01-14T08:00:00Z/2016-01-14T11:59:59Z&limit=100&offset=0"
}
],
"features": [
{
"type": "Feature",
"stac_version": "1.0.0",
"id": "SNDR.SNPP.ATMS.L1A.nominal2.01",
"properties": {
"start_datetime": "2016-01-14T09:54:00Z",
"end_datetime": "2016-01-14T10:00:00Z",
"created": "2020-12-14T13:50:00Z",
"updated": "2022-08-15T06:26:39.830000Z",
"datetime": "2022-08-15T06:26:37.029000Z"
},
"geometry": {
"type": "Point",
"coordinates": [
0.0,
0.0
]
},
"links": [
{
"rel": "collection",
"href": "."
}
],
"assets": {
"data": {
"href": "https://raw.githubusercontent.com/unity-sds/unity-tutorial-application/main/test/stage_in/SNDR.SS1330.CHIRP.20160822T0005.m06.g001.L1_AQ.std.v02_48.G.200425095850.nc",
"title": "SNDR.SS1330.CHIRP.20160822T0005.m06.g001.L1_AQ.std.v02_48.G.200425095850.nc",
"description": "SNDR.SS1330.CHIRP.20160822T0005.m06.g001.L1_AQ.std.v02_48.G.200425095850.nc"
}
},
"bbox": [
0.0,
0.0,
0.0,
0.0
],
"stac_extensions": [],
"collection": "SNDR_SNPP_ATMS_L1A___1"
},
{
"type": "Feature",
"stac_version": "1.0.0",
"id": "SNDR.SNPP.ATMS.L1A.nominal2.08",
"properties": {
"start_datetime": "2016-01-14T10:36:00Z",
"end_datetime": "2016-01-14T10:42:00Z",
"created": "2020-12-14T13:50:00Z",
"updated": "2022-08-15T06:26:26.078000Z",
"datetime": "2022-08-15T06:26:19.333000Z"
},
"geometry": {
"type": "Point",
"coordinates": [
0.0,
0.0
]
},
"links": [
{
"rel": "collection",
"href": "."
}
],
"assets": {
"data": {
"href": "https://raw.githubusercontent.com/unity-sds/unity-tutorial-application/main/test/stage_in/SNDR.SS1330.CHIRP.20160822T0011.m06.g002.L1_AQ.std.v02_48.G.200425095901.nc",
"title": "SNDR.SS1330.CHIRP.20160822T0011.m06.g002.L1_AQ.std.v02_48.G.200425095901.nc",
"description": "SNDR.SS1330.CHIRP.20160822T0011.m06.g002.L1_AQ.std.v02_48.G.200425095901.nc"
}
},
"bbox": [
0.0,
0.0,
0.0,
0.0
],
"stac_extensions": [],
"collection": "SNDR_SNPP_ATMS_L1A___1"
}
]
}
'''
granule_json = json.loads(granule_json)
if len(argv) > 1:
argv.pop(-1)
argv.append('DOWNLOAD')
os.environ['GRANULES_DOWNLOAD_TYPE'] = 'HTTP'
with tempfile.TemporaryDirectory() as tmp_dir_name:
# tmp_dir_name = '/tmp/unity-ds'
os.environ['OUTPUT_FILE'] = os.path.join(tmp_dir_name, 'some_output', 'output.json')
granule_json_file = os.path.join(tmp_dir_name, 'input_file.json')
downloading_dir = os.path.join(tmp_dir_name, 'downloading_dir')
FileUtils.mk_dir_p(downloading_dir)
FileUtils.write_json(granule_json_file, granule_json)
os.environ['STAC_JSON'] = granule_json_file
os.environ['DOWNLOAD_DIR'] = downloading_dir
download_result_str = choose_process()
download_result = json.loads(download_result_str)
self.assertTrue('features' in download_result, f'missing features in download_result')
self.assertEqual(len(download_result['features']) + 1, len(glob(os.path.join(downloading_dir, '*'))),
f'downloaded file does not match: {download_result["features"]}')
error_file = os.path.join(downloading_dir, 'error.log')
if FileUtils.file_exist(error_file):
self.assertTrue(False, f'some downloads failed. error.log exists. {FileUtils.read_json(error_file)}')
download_result = download_result['features']
self.assertTrue('assets' in download_result[0], f'no assets in download_result: {download_result}')
downloaded_file_hrefs = set([k['assets']['data']['href'] for k in download_result])
for each_granule in zip(granule_json['features'], download_result):
remote_filename = os.path.basename(each_granule[0]['assets']['data']['href'])
self.assertTrue(os.path.join('.', remote_filename) in downloaded_file_hrefs,
f'mismatched: {remote_filename}')
self.assertTrue(FileUtils.get_size(os.path.join(downloading_dir, remote_filename)) > 40 * 2**20, f'empty file: {remote_filename}')
self.assertTrue(FileUtils.file_exist(os.environ['OUTPUT_FILE']), f'missing output file')
return

Expand Down Expand Up @@ -1415,6 +1562,7 @@ def test_02_download__from_http(self):
remote_filename = os.path.basename(each_granule[0]['assets']['data']['href'])
self.assertTrue(os.path.join('.', remote_filename) in downloaded_file_hrefs,
f'mismatched: {remote_filename}')
self.assertTrue(FileUtils.get_size(os.path.join(downloading_dir, remote_filename)) > 0, f'empty file: {remote_filename}')
self.assertTrue(FileUtils.file_exist(os.environ['OUTPUT_FILE']), f'missing output file')
return

Expand Down Expand Up @@ -1769,7 +1917,7 @@ def test_02_download__from_http_with_role(self):
for each_granule in zip(granule_json['features'], download_result):
remote_filename = [k['href'] for k in each_granule[0]['assets'].values() if 'data' in k['roles']]
remote_filename = os.path.basename(remote_filename[0])
self.assertTrue(os.path.join('.', remote_filename) in downloaded_file_hrefs,
f'mismatched: {remote_filename}')
self.assertTrue(os.path.join('.', remote_filename) in downloaded_file_hrefs, f'mismatched: {remote_filename}')
self.assertTrue(FileUtils.get_size(os.path.join(downloading_dir, remote_filename)) > 0, f'empty file: {remote_filename}')
self.assertTrue(FileUtils.file_exist(os.environ['OUTPUT_FILE']), f'missing output file')
return
Loading