Mercurial > repos > diodupima > data_manager_coast_taxonomic_filters

changeset 8:3c7247991227 draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
"planemo upload commit 5a255cfbba89d13ea62aeaaabd2801ac618cbb6b-dirty"
author diodupima
date Wed, 07 Jul 2021 23:00:55 +0000
parents 98f4b03a05dc
children 8a3febefd3e5
files ho2s.py
diffstat 1 files changed, 15 insertions(+), 2 deletions(-) [+]
line wrap: on
line diff
--- a/ho2s.py	Wed Jul 07 21:40:56 2021 +0000
+++ b/ho2s.py	Wed Jul 07 23:00:55 2021 +0000
@@ -1,7 +1,9 @@
 import argparse
 import os
+from urllib import request
+from shutil import unpack_archive
+import datetime
 import subprocess
-import datetime
 from galaxy.util.json import from_json_string, to_json_string
 
 
@@ -25,7 +27,18 @@
     tool_data_table_name = args.tool_data_table_name
     data_id = "_".join([taxid, datetime.datetime.now().strftime("%Y_%m_%d_%H_%M_%S")])
 
-    subprocess.call(f". high_order2species.sh {taxid} {data_id}", shell=True, cwd=target_directory)
+    url = 'ftp://ftp.ncbi.nih.gov/pub/taxonomy/taxdump.tar.gz'
+    request.urlretrieve(url, os.path.join(target_directory, "taxdump.tar.gz"))
+    unpack_archive(os.path.join(target_directory, "taxdump.tar.gz"), extract_dir=target_directory)
+
+    subprocess.call(
+        f"taxonkit list --ids {taxid} --data-dir {target_directory}" 
+        f"| taxonkit filter --equal-to Species --lower-than Species --save-predictable-norank --data-dir {target_directory}" 
+        f"> {data_id}.txids",
+        shell=True,
+        cwd=target_directory
+    )
+
     data_table_entry = {
         "value": data_id,
         "name": name,