18634744198

Committed 19 Oct 2025 07:00PM UTC coverage: 31.618% (-29.9%) from 61.514%

Build # 18634744198

Build Type

push

github

Committed by

mpasternak

Commit Message

Merge branch 'release/v202510.1270'

Coverage Stats

657 of 9430 branches covered (6.97%)

Branch coverage included in aggregate %.

229 of 523 new or added lines in 42 files covered. (43.79%)

11303 existing lines in 316 files now uncovered.

14765 of 39346 relevant lines covered (37.53%)

0.38 hits per line

Source File
Press 'n' to go to next uncovered line, 'b' for previous

84.09

src/crossref_bpp/utils.py

from __future__ import annotations

import json
import textwrap

from django.contrib.postgres.search import TrigramSimilarity
from django.db import models


def json_format_with_wrap(item: [] | {} | str | None, width: int = 70) -> str:
    """Formatuje 'ładnie' dane JSON, w sytuacji, gdyby jakas linia
    przekraczała szerokośc ``width``, zawija.
    """
    s = json.dumps(item, indent=2).split("\n")
    news = []
    for elem in s:
        if len(elem) > width:
            elem = textwrap.fill(elem, width=width)
        news.append(elem)
    return "\n".join(news)


def perform_trigram_search(
    queryset,
    trigram_db_field,
    trigram_db_value,
    max_number_records=5,
    minimum_acceptable_similarity=0.6,
    db_trigram_target="podobienstwo",
    similarity_step_down=0.05,
    similarity_step_up=0.01,
    max_steps=20,
    initial_similarity=None,
) -> list[models.Model] | None:
    """Performs a TrigramSimilarity search, starting from similarity value and
    descending/ascending to find the optimal number of records.

    :param queryset: Django queryset, can be anything.
    :param trigram_db_field: trigram database field, can be Lower('db_column') etc,
    :param trigram_db_value: value you're looking for in trigram_db_field, will be used
    to compare similarity.
    :param initial_similarity: Starting similarity value. If None, uses minimum_acceptable_similarity + 0.2
    """

    kwargs_annotate = {
        db_trigram_target: TrigramSimilarity(trigram_db_field, trigram_db_value)
    }

    # Start from a more reasonable similarity value for better performance
    if initial_similarity is None:
        current_similarity = min(minimum_acceptable_similarity + 0.2, 0.8)
    else:
        current_similarity = initial_similarity
    current_step = 0
    while (
        current_similarity > minimum_acceptable_similarity and current_step < max_steps
    ):
        current_step += 1
        kwargs_filter = {db_trigram_target + "__gte": current_similarity}
        my_queryset = (
            queryset.annotate(**kwargs_annotate)
            .filter(**kwargs_filter)
            .order_by(f"-{db_trigram_target}")
        )

        # Use list slicing instead of count() for better performance
        results_list = list(my_queryset[: max_number_records + 1])
        cnt = len(results_list)

        if cnt == max_number_records + 1:
            # Za dużo rekordów. Zwiększ nieznacznie pożądane podobieństwo i szukaj dalej.
            current_similarity += similarity_step_up
            continue

        if cnt == 0:
            # Za mało rekordów. Zmniejsz poszukiwane prawdopodobieństwo:
            current_similarity -= similarity_step_down
            continue

        # Return queryset, not the list
        return my_queryset[:max_number_records]

1	from __future__ import annotations	1✔
2
3	import json	1✔
4	import textwrap	1✔
5
6	from django.contrib.postgres.search import TrigramSimilarity	1✔
7	from django.db import models	1✔
8
9
10	def json_format_with_wrap(item: [] \| {} \| str \| None, width: int = 70) -> str:	1✔
11	"""Formatuje 'ładnie' dane JSON, w sytuacji, gdyby jakas linia
12	przekraczała szerokośc ``width``, zawija.
13	"""
14	s = json.dumps(item, indent=2).split("\n")	1✔
15	news = []	1✔
16	for elem in s:	1✔
17	if len(elem) > width:	1✔
18	elem = textwrap.fill(elem, width=width)	1✔
19	news.append(elem)	1✔
20	return "\n".join(news)	1✔
21
22
23	def perform_trigram_search(	1✔
24	queryset,
25	trigram_db_field,
26	trigram_db_value,
27	max_number_records=5,
28	minimum_acceptable_similarity=0.6,
29	db_trigram_target="podobienstwo",
30	similarity_step_down=0.05,
31	similarity_step_up=0.01,
32	max_steps=20,
33	initial_similarity=None,
34	) -> list[models.Model] \| None:
35	"""Performs a TrigramSimilarity search, starting from similarity value and
36	descending/ascending to find the optimal number of records.
37
38	:param queryset: Django queryset, can be anything.
39	:param trigram_db_field: trigram database field, can be Lower('db_column') etc,
40	:param trigram_db_value: value you're looking for in trigram_db_field, will be used
41	to compare similarity.
42	:param initial_similarity: Starting similarity value. If None, uses minimum_acceptable_similarity + 0.2
43	"""
44
45	kwargs_annotate = {	1✔
46	db_trigram_target: TrigramSimilarity(trigram_db_field, trigram_db_value)
47	}
48
49	# Start from a more reasonable similarity value for better performance
50	if initial_similarity is None:	1!
51	current_similarity = min(minimum_acceptable_similarity + 0.2, 0.8)	1✔
52	else:
53	current_similarity = initial_similarity	×
54	current_step = 0	1✔
55	while (	1✔
56	current_similarity > minimum_acceptable_similarity and current_step < max_steps
57	):
58	current_step += 1	1✔
59	kwargs_filter = {db_trigram_target + "__gte": current_similarity}	1✔
60	my_queryset = (	1✔
61	queryset.annotate(**kwargs_annotate)
62	.filter(**kwargs_filter)
63	.order_by(f"-{db_trigram_target}")
64	)
65
66	# Use list slicing instead of count() for better performance
67	results_list = list(my_queryset[: max_number_records + 1])	1✔
68	cnt = len(results_list)	1✔
69
70	if cnt == max_number_records + 1:	1!
71	# Za dużo rekordów. Zwiększ nieznacznie pożądane podobieństwo i szukaj dalej.
72	current_similarity += similarity_step_up	×
73	continue	×
74
75	if cnt == 0:	1!
76	# Za mało rekordów. Zmniejsz poszukiwane prawdopodobieństwo:
77	current_similarity -= similarity_step_down	1✔
78	continue	1✔
79
80	# Return queryset, not the list
UNCOV 81	return my_queryset[:max_number_records]	×

iplweb / bpp / 18634744198

Source File Press 'n' to go to next uncovered line, 'b' for previous

Source File
Press 'n' to go to next uncovered line, 'b' for previous