From 4ce143736860e71fd59c6506875717197c800510 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Henrik=20tom=20W=C3=B6rden?= <h.tomwoerden@indiscale.com>
Date: Tue, 9 Jul 2024 16:44:11 +0200
Subject: [PATCH] MAITN: move code from old crawler
---
CHANGELOG.md | 1 +
src/caoscrawler/crawl.py | 29 ++++++++++++++++++--
src/doc/getting_started/optionalfeatures.rst | 7 +++++
3 files changed, 34 insertions(+), 3 deletions(-)
diff --git a/CHANGELOG.md b/CHANGELOG.md
index 2c9160a5..995ad6ee 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -28,6 +28,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
* Better internal handling of identifiable/reference resolving and merging of entities. This also
includes more understandable output for users.
* Better handling of missing imports, with nice messages for users.
+* No longer use configuration of advancedtools to set to and from email addresses
### Deprecated ###
diff --git a/src/caoscrawler/crawl.py b/src/caoscrawler/crawl.py
index 627e1dc4..dab9fbba 100644
--- a/src/caoscrawler/crawl.py
+++ b/src/caoscrawler/crawl.py
@@ -749,9 +749,31 @@ one with the entities that need to be updated and the other with entities to be
def inform_about_pending_changes(pending_changes, run_id, path, inserts=False):
# Sending an Email with a link to a form to authorize updates is
if get_config_setting("send_crawler_notifications"):
- filename = OldCrawler.save_form(
- [el[3] for el in pending_changes], path, run_id)
- OldCrawler.send_mail([el[3] for el in pending_changes], filename)
+ filename = OldCrawler.save_form([el[3] for el in pending_changes], path, run_id)
+ text = """Dear Curator,
+ there where changes that need your authorization. Please check the following
+ carefully and if the changes are ok, click on the following link:
+
+ {url}/Shared/{filename}
+
+ {changes}
+ """.format(url=db.configuration.get_config()["Connection"]["url"],
+ filename=filename,
+ changes="\n".join([el[3] for el in pending_changes]))
+ try:
+ fro=get_config_setting("sendmail_from_address")
+ to=get_config_setting("sendmail_to_address")
+ except KeyError:
+ logger.error("Server Configuration is missing a setting for "
+ "sending mails. The administrator should check "
+ "'from_mail' and 'to_mail'.")
+ return
+
+ send_mail(
+ from_addr=fro,
+ to=to,
+ subject="Crawler Update",
+ body=text)
for i, el in enumerate(pending_changes):
@@ -862,6 +884,7 @@ def _notify_about_inserts_and_updates(n_inserts, n_updates, logfile, run_id):
The email contains some basic information and a link to the log and the CrawlerRun Record.
"""
if not get_config_setting("send_crawler_notifications"):
+ logger.debug("Crawler email notifications are disabled.")
return
if n_inserts == 0 and n_updates == 0:
return
diff --git a/src/doc/getting_started/optionalfeatures.rst b/src/doc/getting_started/optionalfeatures.rst
index d326d7fc..7b776465 100644
--- a/src/doc/getting_started/optionalfeatures.rst
+++ b/src/doc/getting_started/optionalfeatures.rst
@@ -30,6 +30,13 @@ to decide what tool is used for sending mails (use the upper one if you
want to actually send mails. See ``sendmail`` configuration in the
LinkAhead docs.
+You can even supply the name of a custom CSS file that shall be used:
+
+.. code:: ini
+
+ [advancedtools]
+ crawler.customcssfile = theme-research.css
+
Crawler Status Records
----------------------
--
GitLab