update load_db script (#252)

.gitignore

@@ -140,4 +140,5 @@ postgres/
 .idea
 environment/local.env
 qdrant_storage/
-postgres-data/
+postgres-data/
+local_cache/

scripts/README.md

@@ -1,10 +1,13 @@
 # pephub scripts
+
 These are useful scripts for pephub. They provide utilities to manage the interface between PEPs, a PEP database, and PEPhub. Read below about each script.
 
 ### `load_db.py`
-This will help you load a database of PEPs from a directory of PEPs.
+
+This will help you load a database of PEPs from a directory of PEPs. Note, that a namespace is **required** for each PEP. This is the name of the folder that the PEP is in. For example, if you have a PEP in `peps/ChangLab/PEP_1`, then the namespace is `ChangLab`. If you have a PEP in `peps/demo/basic`, then the namespace is `demo`. The namespace is used to identify the PEP in the database.
 
 **Usage:**
+
 ```console
 python load_db.py \
 --username $POSTGRES_USERNAME \
@@ -64,4 +67,4 @@ peps
 │   ├── GSE101516
 │   │   ├── GSE101516_samples.csv
 │   │   └── GSE101516_samples.yaml
-```
+```

scripts/load_db.py

@@ -1,7 +1,7 @@
 import argparse
 import os
 from tqdm import tqdm
-from pepdbagent import Connection
+from pepdbagent import PEPDatabaseAgent as Connection
 import peppy
 
 from utils import is_valid_namespace, is_valid_project, extract_project_file_name
@@ -76,6 +76,7 @@ def build_connection_string(args: argparse.Namespace) -> str:
 # get file path
 FILE_PATH = args.files
 
+
 # traverse directory
 for name in tqdm(os.listdir(FILE_PATH), desc="Uploading repository", leave=True):
     # build a path to the namespace
@@ -95,4 +96,6 @@ def build_connection_string(args: argparse.Namespace) -> str:
                 p = peppy.Project(
                     f"{path_to_proj}/{extract_project_file_name(path_to_proj)}"
                 )
-                pagent.upload_project(p, name)
+                pagent.project.create(
+                    p, name, p.name, description=f"Uploaded from, {path_to_proj}"
+                )