Ian Ward

@JVickery-TBS is our date range hack on the publishing queue search something that could be shared as an example?

> `update.py` is perhaps not the best home as it can delegate to any type of call, but I was hesitant to make a whole new module. This kind of...

Is it possible volume containing your solr data is not being kept and it being reverted to a previous state when you restart the service?

@amercader would we normally consider this kind of change for a 2.10 patch release?

@petercpg absolutely agree, would you like to submit a pull request changing this to something like: ```jinja2 {% set pkg_link = (h.url_for('dataset.read', id=change.pkg_id))%} {% if change.method == "change" %} {%...

Thanks @Engerrs Instead of adding a new `last_id` parameter to the `datastore_search` API let's implement a special case of the [upcoming generic search filters](https://github.com/ckan/ckanext-search/blob/main/doc/filters.md) feature. e.g. If `datastore_search` parameters contain:...

@Engerrs @EricSoroos in [5893dc5](https://github.com/ckan/ckan/pull/9028/commits/5893dc5627c086b94d69b76a9e7365c3f8a943e6) we're now getting the next id value as part of the same query that retrieves the data when possible, and the `next_page` return value can now...

I was hoping to fix the issue of duplicated/missing rows that can happen when a rows are added/removed while a dump is running. Getting the next id as part of...

CSV dump time for a real 13M record dataset: before | after | change --- | --- | --- 30:09 | 2:08 | 15× faster

`term_translation_*` and the multilingual extension take the approach of mixing translations for user-provided metadata with translations from ckan and its extensions and passing all strings through a gettext call that...