Script 'mail_helper' called by obssrc
Hello community,
here is the log from the commit of package python-elasticsearch-dsl for
openSUSE:Factory checked in at 2024-01-21 23:08:41
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Comparing /work/SRC/openSUSE:Factory/python-elasticsearch-dsl (Old)
and /work/SRC/openSUSE:Factory/.python-elasticsearch-dsl.new.16006 (New)
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Package is "python-elasticsearch-dsl"
Sun Jan 21 23:08:41 2024 rev:9 rq:1140106 version:8.12.0
Changes:
--------
---
/work/SRC/openSUSE:Factory/python-elasticsearch-dsl/python-elasticsearch-dsl.changes
2023-11-27 22:45:24.942350224 +0100
+++
/work/SRC/openSUSE:Factory/.python-elasticsearch-dsl.new.16006/python-elasticsearch-dsl.changes
2024-01-21 23:09:01.870981527 +0100
@@ -1,0 +2,9 @@
+Sat Jan 20 12:39:22 UTC 2024 - Dirk Müller <dmuel...@suse.com>
+
+- update to 8.12.0:
+ * Added Search.knn() method
+ * Added Search.rank() method (undocumented as it still is in
+ technical preview)
+ * Fixed importing collapse from dictionary
+
+-------------------------------------------------------------------
Old:
----
elasticsearch-dsl-8.11.0.tar.gz
New:
----
elasticsearch-dsl-8.12.0.tar.gz
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Other differences:
------------------
++++++ python-elasticsearch-dsl.spec ++++++
--- /var/tmp/diff_new_pack.LD6Tza/_old 2024-01-21 23:09:03.307033874 +0100
+++ /var/tmp/diff_new_pack.LD6Tza/_new 2024-01-21 23:09:03.323034457 +0100
@@ -1,7 +1,7 @@
#
# spec file for package python-elasticsearch-dsl
#
-# Copyright (c) 2023 SUSE LLC
+# Copyright (c) 2024 SUSE LLC
#
# All modifications and additions to the file contributed by third parties
# remain the property of their copyright owners, unless otherwise agreed
@@ -16,15 +16,18 @@
#
+%{?sle15_python_module_pythons}
Name: python-elasticsearch-dsl
-Version: 8.11.0
+Version: 8.12.0
Release: 0
Summary: Python client for Elasticsearch
License: Apache-2.0
Group: Development/Languages/Python
URL: https://github.com/elasticsearch/elasticsearch-dsl-py
Source:
https://github.com/elastic/elasticsearch-dsl-py/archive/refs/tags/v%{version}.tar.gz#/elasticsearch-dsl-%{version}.tar.gz
+BuildRequires: %{python_module pip}
BuildRequires: %{python_module setuptools}
+BuildRequires: %{python_module wheel}
BuildRequires: fdupes
BuildRequires: python-rpm-macros
# gh#elastic/elasticsearch-dsl-py#1569
@@ -47,10 +50,10 @@
%autosetup -p1 -n elasticsearch-dsl-py-%{version}
%build
-%python_build
+%pyproject_wheel
%install
-%python_install
+%pyproject_install
%python_expand %fdupes %{buildroot}%{$python_sitelib}
%check
@@ -59,6 +62,6 @@
%files %{python_files}
%doc AUTHORS Changelog.rst README
%license LICENSE
-%{python_sitelib}/elasticsearch_dsl/
-%{python_sitelib}/elasticsearch_dsl-%{version}*-info/
+%{python_sitelib}/elasticsearch_dsl
+%{python_sitelib}/elasticsearch_dsl-%{version}.dist-info
++++++ elasticsearch-dsl-8.11.0.tar.gz -> elasticsearch-dsl-8.12.0.tar.gz ++++++
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/elasticsearch-dsl-py-8.11.0/.gitignore
new/elasticsearch-dsl-py-8.12.0/.gitignore
--- old/elasticsearch-dsl-py-8.11.0/.gitignore 2023-11-13 12:54:53.000000000
+0100
+++ new/elasticsearch-dsl-py-8.12.0/.gitignore 2024-01-19 12:11:44.000000000
+0100
@@ -14,3 +14,6 @@
venv
.idea
.pytest_cache
+
+# sample code for GitHub issues
+issues
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/elasticsearch-dsl-py-8.11.0/Changelog.rst
new/elasticsearch-dsl-py-8.12.0/Changelog.rst
--- old/elasticsearch-dsl-py-8.11.0/Changelog.rst 2023-11-13
12:54:53.000000000 +0100
+++ new/elasticsearch-dsl-py-8.12.0/Changelog.rst 2024-01-19
12:11:44.000000000 +0100
@@ -3,11 +3,22 @@
Changelog
=========
+8.12.0 (2024-01-18)
+-------------------
+
+* Added ``Search.knn()`` method (`#1691`_)
+* Added ``Search.rank()`` method (undocumented as it still is in technical
preview) (`#1692`_)
+* Fixed importing collapse from dictionary (`#1689`_)
+
+.. _#1689: https://github.com/elastic/elasticsearch-dsl-py/pull/1689
+.. _#1691: https://github.com/elastic/elasticsearch-dsl-py/pull/1691
+.. _#1692: https://github.com/elastic/elasticsearch-dsl-py/pull/1692
+
8.11.0 (2023-11-13)
-------------------
* Added support for Python 3.12 (`#1680`_)
-* Added support for Search.collase() (`#1649`_, contributed by `@qcoumes`_)
+* Added ``Search.collapse()`` (`#1649`_, contributed by `@qcoumes`_)
.. _@qcoumes: https://github.com/qcoumes
.. _#1680: https://github.com/elastic/elasticsearch-dsl-py/pull/1680
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/elasticsearch-dsl-py-8.11.0/docs/search_dsl.rst
new/elasticsearch-dsl-py-8.12.0/docs/search_dsl.rst
--- old/elasticsearch-dsl-py-8.11.0/docs/search_dsl.rst 2023-11-13
12:54:53.000000000 +0100
+++ new/elasticsearch-dsl-py-8.12.0/docs/search_dsl.rst 2024-01-19
12:11:44.000000000 +0100
@@ -14,6 +14,8 @@
* aggregations
+ * k-nearest neighbor searches
+
* sort
* pagination
@@ -352,6 +354,31 @@
done in-place (does not return a copy).
+K-Nearest Neighbor Searches
+~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+To issue a kNN search, use the ``.knn()`` method:
+
+.. code:: python
+
+ s = Search()
+ vector = get_embedding("search text")
+
+ s = s.knn(
+ field="embedding",
+ k=5,
+ num_candidates=10,
+ query_vector=vector
+ )
+
+The ``field``, ``k`` and ``num_candidates`` arguments can be given as
+positional or keyword arguments and are required. In addition to these,
+``query_vector`` or ``query_vector_builder`` must be given as well.
+
+The ``.knn()`` method can be invoked multiple times to include multiple kNN
+searches in the request.
+
+
Sorting
~~~~~~~
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore'
old/elasticsearch-dsl-py-8.11.0/elasticsearch_dsl/__init__.py
new/elasticsearch-dsl-py-8.12.0/elasticsearch_dsl/__init__.py
--- old/elasticsearch-dsl-py-8.11.0/elasticsearch_dsl/__init__.py
2023-11-13 12:54:53.000000000 +0100
+++ new/elasticsearch-dsl-py-8.12.0/elasticsearch_dsl/__init__.py
2024-01-19 12:11:44.000000000 +0100
@@ -84,7 +84,7 @@
from .utils import AttrDict, AttrList, DslBase
from .wrappers import Range
-VERSION = (8, 11, 0)
+VERSION = (8, 12, 0)
__version__ = VERSION
__versionstr__ = ".".join(map(str, VERSION))
__all__ = [
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore'
old/elasticsearch-dsl-py-8.11.0/elasticsearch_dsl/search.py
new/elasticsearch-dsl-py-8.12.0/elasticsearch_dsl/search.py
--- old/elasticsearch-dsl-py-8.11.0/elasticsearch_dsl/search.py 2023-11-13
12:54:53.000000000 +0100
+++ new/elasticsearch-dsl-py-8.12.0/elasticsearch_dsl/search.py 2024-01-19
12:11:44.000000000 +0100
@@ -24,7 +24,7 @@
from .aggs import A, AggBase
from .connections import get_connection
from .exceptions import IllegalOperation
-from .query import Bool, Q
+from .query import Bool, Q, Query
from .response import Hit, Response
from .utils import AttrDict, DslBase, recursive_to_dict
@@ -120,7 +120,6 @@
self._doc_type = []
self._doc_type_map = {}
- self._collapse = {}
if isinstance(doc_type, (tuple, list)):
self._doc_type.extend(doc_type)
elif isinstance(doc_type, collections.abc.Mapping):
@@ -294,7 +293,6 @@
s = self.__class__(
using=self._using, index=self._index, doc_type=self._doc_type
)
- s._collapse = self._collapse.copy()
s._doc_type_map = self._doc_type_map.copy()
s._extra = self._extra.copy()
s._params = self._params.copy()
@@ -320,6 +318,8 @@
self.aggs = AggsProxy(self)
self._sort = []
+ self._knn = []
+ self._rank = {}
self._collapse = {}
self._source = None
self._highlight = {}
@@ -408,6 +408,9 @@
s = super()._clone()
s._response_class = self._response_class
+ s._knn = [knn.copy() for knn in self._knn]
+ s._rank = self._rank.copy()
+ s._collapse = self._collapse.copy()
s._sort = self._sort[:]
s._source = copy.copy(self._source) if self._source is not None else
None
s._highlight = self._highlight.copy()
@@ -446,6 +449,14 @@
self.aggs._params = {
"aggs": {name: A(value) for (name, value) in aggs.items()}
}
+ if "knn" in d:
+ self._knn = d.pop("knn")
+ if isinstance(self._knn, dict):
+ self._knn = [self._knn]
+ if "rank" in d:
+ self._rank = d.pop("rank")
+ if "collapse" in d:
+ self._collapse = d.pop("collapse")
if "sort" in d:
self._sort = d.pop("sort")
if "_source" in d:
@@ -493,6 +504,85 @@
s._script_fields.update(kwargs)
return s
+ def knn(
+ self,
+ field,
+ k,
+ num_candidates,
+ query_vector=None,
+ query_vector_builder=None,
+ boost=None,
+ filter=None,
+ similarity=None,
+ ):
+ """
+ Add a k-nearest neighbor (kNN) search.
+
+ :arg field: the name of the vector field to search against
+ :arg k: number of nearest neighbors to return as top hits
+ :arg num_candidates: number of nearest neighbor candidates to consider
per shard
+ :arg query_vector: the vector to search for
+ :arg query_vector_builder: A dictionary indicating how to build a
query vector
+ :arg boost: A floating-point boost factor for kNN scores
+ :arg filter: query to filter the documents that can match
+ :arg similarity: the minimum similarity required for a document to be
considered a match, as a float value
+
+ Example::
+
+ s = Search()
+ s = s.knn(field='embedding', k=5, num_candidates=10,
query_vector=vector,
+ filter=Q('term', category='blog')))
+ """
+ s = self._clone()
+ s._knn.append(
+ {
+ "field": field,
+ "k": k,
+ "num_candidates": num_candidates,
+ }
+ )
+ if query_vector is None and query_vector_builder is None:
+ raise ValueError("one of query_vector and query_vector_builder is
required")
+ if query_vector is not None and query_vector_builder is not None:
+ raise ValueError(
+ "only one of query_vector and query_vector_builder must be
given"
+ )
+ if query_vector is not None:
+ s._knn[-1]["query_vector"] = query_vector
+ if query_vector_builder is not None:
+ s._knn[-1]["query_vector_builder"] = query_vector_builder
+ if boost is not None:
+ s._knn[-1]["boost"] = boost
+ if filter is not None:
+ if isinstance(filter, Query):
+ s._knn[-1]["filter"] = filter.to_dict()
+ else:
+ s._knn[-1]["filter"] = filter
+ if similarity is not None:
+ s._knn[-1]["similarity"] = similarity
+ return s
+
+ def rank(self, rrf=None):
+ """
+ Defines a method for combining and ranking results sets from a
combination
+ of searches. Requires a minimum of 2 results sets.
+
+ :arg rrf: Set to ``True`` or an options dictionary to set the rank
method to reciprocal rank fusion (RRF).
+
+ Example::
+ s = Search()
+ s = s.query('match', content='search text')
+ s = s.knn(field='embedding', k=5, num_candidates=10,
query_vector=vector)
+ s = s.rank(rrf=True)
+
+ Note: This option is in technical preview and may change in the
future. The syntax will likely change before GA.
+ """
+ s = self._clone()
+ s._rank = {}
+ if rrf is not None and rrf is not False:
+ s._rank["rrf"] = {} if rrf is True else rrf
+ return s
+
def source(self, fields=None, **kwargs):
"""
Selectively control how the _source field is returned.
@@ -676,6 +766,15 @@
if self.query:
d["query"] = self.query.to_dict()
+ if self._knn:
+ if len(self._knn) == 1:
+ d["knn"] = self._knn[0]
+ else:
+ d["knn"] = self._knn
+
+ if self._rank:
+ d["rank"] = self._rank
+
# count request doesn't care for sorting and other things
if not count:
if self.post_filter:
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/elasticsearch-dsl-py-8.11.0/setup.py
new/elasticsearch-dsl-py-8.12.0/setup.py
--- old/elasticsearch-dsl-py-8.11.0/setup.py 2023-11-13 12:54:53.000000000
+0100
+++ new/elasticsearch-dsl-py-8.12.0/setup.py 2024-01-19 12:11:44.000000000
+0100
@@ -19,7 +19,7 @@
from setuptools import find_packages, setup
-VERSION = (8, 11, 0)
+VERSION = (8, 12, 0)
__version__ = VERSION
__versionstr__ = ".".join(map(str, VERSION))
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/elasticsearch-dsl-py-8.11.0/tests/test_search.py
new/elasticsearch-dsl-py-8.12.0/tests/test_search.py
--- old/elasticsearch-dsl-py-8.11.0/tests/test_search.py 2023-11-13
12:54:53.000000000 +0100
+++ new/elasticsearch-dsl-py-8.12.0/tests/test_search.py 2024-01-19
12:11:44.000000000 +0100
@@ -234,6 +234,72 @@
assert s._doc_type_map == {}
+def test_knn():
+ s = search.Search()
+
+ with raises(TypeError):
+ s.knn()
+ with raises(TypeError):
+ s.knn("field")
+ with raises(TypeError):
+ s.knn("field", 5)
+ with raises(ValueError):
+ s.knn("field", 5, 100)
+ with raises(ValueError):
+ s.knn("field", 5, 100, query_vector=[1, 2, 3], query_vector_builder={})
+
+ s = s.knn("field", 5, 100, query_vector=[1, 2, 3])
+ assert {
+ "knn": {
+ "field": "field",
+ "k": 5,
+ "num_candidates": 100,
+ "query_vector": [1, 2, 3],
+ }
+ } == s.to_dict()
+
+ s = s.knn(
+ k=4,
+ num_candidates=40,
+ boost=0.8,
+ field="name",
+ query_vector_builder={
+ "text_embedding": {"model_id": "foo", "model_text": "search text"}
+ },
+ )
+ assert {
+ "knn": [
+ {
+ "field": "field",
+ "k": 5,
+ "num_candidates": 100,
+ "query_vector": [1, 2, 3],
+ },
+ {
+ "field": "name",
+ "k": 4,
+ "num_candidates": 40,
+ "query_vector_builder": {
+ "text_embedding": {"model_id": "foo", "model_text":
"search text"}
+ },
+ "boost": 0.8,
+ },
+ ]
+ } == s.to_dict()
+
+
+def test_rank():
+ s = search.Search()
+ s.rank(rrf=False)
+ assert {} == s.to_dict()
+
+ s = s.rank(rrf=True)
+ assert {"rank": {"rrf": {}}} == s.to_dict()
+
+ s = s.rank(rrf={"window_size": 50, "rank_constant": 20})
+ assert {"rank": {"rrf": {"window_size": 50, "rank_constant": 20}}} ==
s.to_dict()
+
+
def test_sort():
s = search.Search()
s = s.sort("fielda", "-fieldb")
@@ -544,10 +610,12 @@
s = search.Search()
s.update_from_dict({"indices_boost": [{"important-documents": 2}]})
s.update_from_dict({"_source": ["id", "name"]})
+ s.update_from_dict({"collapse": {"field": "user_id"}})
assert {
"indices_boost": [{"important-documents": 2}],
"_source": ["id", "name"],
+ "collapse": {"field": "user_id"},
} == s.to_dict()
