klbr/khoj
1
0
Fork 0
mirror of https://github.com/khoaliber/khoj.git synced 2026-03-06 05:39:12 +00:00
Code Issues Packages Projects Releases Wiki Activity

Rename Files, Classes from X_To_JSONL to more appropriate X_To_Entries

Browse Source 
These content processors are converting content into entries in DB
instead of entries in JSONL file
This commit is contained in:
Debanjum Singh Solanky
2023-11-01 14:51:33 -07:00
parent 2ad2055bcb
commit d92a2d03a7
16 changed files with 127 additions and 125 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
tests/test_plaintext_to_jsonl.py
View File

@@ -6,7 +6,7 @@ from pathlib import Path
# Internal Packages
from khoj.utils.fs_syncer import get_plaintext_files
from khoj.utils.rawconfig import TextContentConfig
from khoj.processor.plaintext.plaintext_to_jsonl import PlaintextToJsonl
from khoj.processor.plaintext.plaintext_to_entries import PlaintextToEntries
from database.models import LocalPlaintextConfig, KhojUser
@@ -27,14 +27,14 @@ def test_plaintext_file(tmp_path):
f"{plaintextfile}": entry,
}
maps = PlaintextToJsonl.convert_plaintext_entries_to_maps(entry_to_file_map=data)
maps = PlaintextToEntries.convert_plaintext_entries_to_maps(entry_to_file_map=data)
# Convert each entry.file to absolute path to make them JSON serializable
for map in maps:
map.file = str(Path(map.file).absolute())
# Process Each Entry from All Notes Files
jsonl_string = PlaintextToJsonl.convert_entries_to_jsonl(maps)
jsonl_string = PlaintextToEntries.convert_entries_to_jsonl(maps)
jsonl_data = [json.loads(json_string) for json_string in jsonl_string.splitlines()]
# Assert
@@ -100,7 +100,7 @@ def test_parse_html_plaintext_file(content_config, default_user: KhojUser):
extracted_plaintext_files = get_plaintext_files(config=config)
# Act
maps = PlaintextToJsonl.convert_plaintext_entries_to_maps(extracted_plaintext_files)
maps = PlaintextToEntries.convert_plaintext_entries_to_maps(extracted_plaintext_files)
# Assert
assert len(maps) == 1