From cd8c3bfe876beb20534fabff98bb74c3241d8c13 Mon Sep 17 00:00:00 2001
From: Alexander Schlemmer <alexander@mail-schlemmer.de>
Date: Thu, 23 Mar 2023 18:53:14 +0100
Subject: [PATCH] REVIEW: added deprecated function start_crawling
---
CHANGELOG.md | 2 ++
src/caoscrawler/crawl.py | 19 ++++++++++++++++++-
2 files changed, 20 insertions(+), 1 deletion(-)
diff --git a/CHANGELOG.md b/CHANGELOG.md
index 0572dda7..08b47840 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -36,6 +36,8 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
- The functions ``load_definition``, ``initialize_converters`` and ``load_converters`` are deprecated. Please use
the functions ``load_definition``, ``initialize_converters`` and ``create_converter_registry`` from the scanner module
instead.
+- The function ``start_crawling`` is deprecated. The function ``scan_structure_elements`` in the
+ scanner module mostly covers its functionality.
### Removed ###
- The command line argument ``--prefix``. Use the new argument ``--remove-prefix`` instead.
diff --git a/src/caoscrawler/crawl.py b/src/caoscrawler/crawl.py
index 3dd10011..7de19275 100644
--- a/src/caoscrawler/crawl.py
+++ b/src/caoscrawler/crawl.py
@@ -70,7 +70,8 @@ from .version import check_cfood_version
from .scanner import (scan_directory,
load_definition,
create_converter_registry,
- initialize_converters)
+ initialize_converters,
+ scan_structure_elements)
from .debug_tree import DebugTree
logger = logging.getLogger(__name__)
@@ -244,6 +245,22 @@ class Crawler(object):
def generate_run_id(self):
self.run_id = uuid.uuid1()
+
+ def start_crawling(self, items: Union[list[StructureElement], StructureElement],
+ crawler_definition: dict,
+ converter_registry: dict,
+ restricted_path: Optional[list[str]] = None):
+
+ warnings.warn(DeprecationWarning(
+ "The function start_crawling in the crawl module is deprecated. "
+ "Please use scan_structure_elements from the scanner module."))
+
+ self.generate_run_id()
+
+ return scan_structure_elements(
+ items, crawler_definition, converter_registry, restrict_path)
+
+
def crawl_directory(self,
crawled_directory: str,
crawler_definition_path: str,
--
GitLab