Files

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

151 lines
5.2 KiB
Python
Raw Permalink Normal View History

#!/usr/bin/env python
# Dan Blankenberg
# Very simple example of using the API to run Data Managers
# Script makes the naive assumption that dbkey==sequence id, which in many cases is not true nor desired
# *** This script is not recommended for use as-is on a production server ***
import optparse
import time
2020-10-06 18:49:19 +01:00
from urllib.parse import urljoin
2025-03-15 16:43:25 +00:00
from common import (
2017-12-01 11:09:57 +00:00
get,
post,
2022-02-02 15:26:21 -05:00
)
DEFAULT_SLEEP_TIME = 3
FETCH_GENOME_TOOL_ID = "testtoolshed.g2.bx.psu.edu/repos/blankenberg/data_manager_fetch_genome_all_fasta/data_manager_fetch_genome_all_fasta/0.0.1"
2017-08-17 00:33:02 +01:00
BUILD_INDEX_TOOLS_ID = [
"testtoolshed.g2.bx.psu.edu/repos/blankenberg/data_manager_bwa_index_builder/bwa_index_builder_data_manager/0.0.1",
"testtoolshed.g2.bx.psu.edu/repos/blankenberg/data_manager_bwa_index_builder/bwa_color_space_index_builder_data_manager/0.0.1",
]
2017-08-17 00:33:02 +01:00
def run_tool(tool_id, history_id, params, api_key, galaxy_url, wait=True, sleep_time=None, **kwargs):
sleep_time = sleep_time or DEFAULT_SLEEP_TIME
2017-08-17 00:33:02 +01:00
tools_url = urljoin(galaxy_url, "api/tools")
payload = {
2016-05-17 23:52:13 +01:00
"tool_id": tool_id,
}
if history_id:
payload["history_id"] = history_id
2017-08-17 00:33:02 +01:00
payload["inputs"] = params
rval = post(api_key, tools_url, payload)
if wait:
2017-08-17 00:33:02 +01:00
outputs = list(rval["outputs"])
while outputs:
finished_datasets = []
2017-08-17 00:33:02 +01:00
for i, dataset_dict in enumerate(outputs):
if dataset_is_terminal(dataset_dict["id"], api_key=api_key, galaxy_url=galaxy_url):
finished_datasets.append(i)
2020-10-06 18:49:19 +01:00
for _ in reversed(finished_datasets):
2017-08-17 00:33:02 +01:00
outputs.pop(0)
if wait and outputs:
2017-08-17 00:33:02 +01:00
time.sleep(sleep_time)
return rval
2017-08-17 00:33:02 +01:00
def get_dataset_state(hda_id, api_key, galaxy_url):
2024-04-27 16:03:22 +01:00
datasets_url = urljoin(galaxy_url, f"api/datasets/{hda_id}")
2017-08-17 00:33:02 +01:00
dataset_info = get(api_key, datasets_url)
return dataset_info["state"]
2017-08-17 00:33:02 +01:00
def dataset_is_terminal(hda_id, api_key, galaxy_url):
dataset_state = get_dataset_state(hda_id, api_key, galaxy_url)
return dataset_state in ["ok", "error"]
2016-11-14 21:12:14 +00:00
if __name__ == "__main__":
parser = optparse.OptionParser()
2017-08-17 00:33:02 +01:00
parser.add_option("-k", "--key", dest="api_key", action="store", type="string", default=None, help="API Key.")
parser.add_option(
"-u",
"--url",
dest="base_url",
action="store",
type="string",
default="http://localhost:8080",
help="Base URL of Galaxy Server",
)
parser.add_option(
"-d",
"--dbkey",
dest="dbkeys",
action="append",
type="string",
default=[],
help="List of dbkeys to download and Index",
)
parser.add_option(
"-s",
"--sleep_time",
dest="sleep_time",
action="store",
type="int",
default=DEFAULT_SLEEP_TIME,
help="How long to sleep between check loops",
)
2026-01-19 22:33:34 +00:00
options, args = parser.parse_args()
# check options
2017-08-17 00:33:02 +01:00
assert options.api_key is not None, ValueError("You must specify an API key.")
assert options.dbkeys, ValueError("You must specify at least one dbkey to use.")
# check user is admin
2017-08-17 00:33:02 +01:00
configuration_options = get(options.api_key, urljoin(options.base_url, "api/configuration"))
if "library_import_dir" not in configuration_options: # hack to check if is admin user
print(
"Warning: Data Managers are only available to admin users. The API Key provided does not appear to belong to an admin user. Will attempt to run anyway."
)
# Fetch Genomes
dbkeys = {}
for dbkey in options.dbkeys:
if dbkey not in dbkeys:
2017-08-17 00:33:02 +01:00
dbkeys[dbkey] = run_tool(
FETCH_GENOME_TOOL_ID,
None,
{
"dbkey": dbkey,
"reference_source|reference_source_selector": "ucsc",
"reference_source|requested_dbkey": dbkey,
},
options.api_key,
options.base_url,
wait=False,
)
else:
2024-04-27 16:03:22 +01:00
f"dbkey ({dbkey}) was specified more than once, skipping additional specification."
print("Genomes Queued for downloading.")
# Start indexers
indexing_tools = []
while dbkeys:
for dbkey, value in dbkeys.items():
2017-08-17 00:33:02 +01:00
if dataset_is_terminal(value["outputs"][0]["id"], options.api_key, options.base_url):
del dbkeys[dbkey]
for tool_id in BUILD_INDEX_TOOLS_ID:
2017-08-17 00:33:02 +01:00
indexing_tools.append(
run_tool(
tool_id, None, {"all_fasta_source": dbkey}, options.api_key, options.base_url, wait=False
)
2022-02-02 15:26:21 -05:00
)
if dbkeys:
2017-08-17 00:33:02 +01:00
time.sleep(options.sleep_time)
print("All genomes downloaded and indexers now queued.")
# Wait for indexers to finish
while indexing_tools:
2017-08-17 00:33:02 +01:00
for i, indexing_tool_value in enumerate(indexing_tools):
if dataset_is_terminal(indexing_tool_value["outputs"][0]["id"], options.api_key, options.base_url):
print("Finished:", indexing_tool_value)
del indexing_tools[i]
break
if indexing_tools:
2017-08-17 00:33:02 +01:00
time.sleep(options.sleep_time)
print("All indexers have been run, please check results.")