From 034d5e7e327b51c1e7b2f325230a0fde046ffee9 Mon Sep 17 00:00:00 2001 From: Daniel Rabstejnek Date: Tue, 4 Feb 2025 09:04:13 -0500 Subject: [PATCH 1/5] Initial code for deduplication workflow --- hawc/apps/lit/constants.py | 4 + ...0025_dedupesettings_duplicatecandidates.py | 84 +++++++++++++++++++ hawc/apps/lit/models.py | 66 +++++++++++++++ .../lit/templates/_duplicate_candidates.html | 1 + .../templates/lit/duplicate_candidates.html | 14 ++++ hawc/apps/lit/urls.py | 5 ++ hawc/apps/lit/views.py | 17 ++++ 7 files changed, 191 insertions(+) create mode 100644 hawc/apps/lit/migrations/0025_dedupesettings_duplicatecandidates.py create mode 100644 hawc/apps/lit/templates/_duplicate_candidates.html create mode 100644 hawc/apps/lit/templates/lit/duplicate_candidates.html diff --git a/hawc/apps/lit/constants.py b/hawc/apps/lit/constants.py index 1c51472140..039980837e 100644 --- a/hawc/apps/lit/constants.py +++ b/hawc/apps/lit/constants.py @@ -28,6 +28,10 @@ class SearchType(models.TextChoices): SEARCH = "s", "Search" IMPORT = "i", "Import" +class DuplicateResolution(models.IntegerChoices): + UNRESOLVED = 0, "Unresolved" + RESOLVED = 1, "Resolved" # TODO: change to "primary identified" + FALSE_POSITIVE = 2, "False positive" # generalized/adapted from https://www.crossref.org/blog/dois-and-matching-regular-expressions/ DOI_EXACT = re.compile(r"^10\.\d{4,9}/[^\s]+$") diff --git a/hawc/apps/lit/migrations/0025_dedupesettings_duplicatecandidates.py b/hawc/apps/lit/migrations/0025_dedupesettings_duplicatecandidates.py new file mode 100644 index 0000000000..7e28051935 --- /dev/null +++ b/hawc/apps/lit/migrations/0025_dedupesettings_duplicatecandidates.py @@ -0,0 +1,84 @@ +# Generated by Django 5.1.4 on 2025-02-03 08:23 + +import django.contrib.postgres.fields +import django.db.models.deletion +from django.conf import settings +from django.db import migrations, models + + +class Migration(migrations.Migration): + + dependencies = [ + ("assessment", "0047_alter_labeleditem_options"), + ("lit", "0024_workflows"), + migrations.swappable_dependency(settings.AUTH_USER_MODEL), + ] + + operations = [ + migrations.CreateModel( + name="DedupeSettings", + fields=[ + ( + "id", + models.AutoField( + auto_created=True, + primary_key=True, + serialize=False, + verbose_name="ID", + ), + ), + ], + ), + migrations.CreateModel( + name="DuplicateCandidates", + fields=[ + ( + "id", + models.AutoField( + auto_created=True, + primary_key=True, + serialize=False, + verbose_name="ID", + ), + ), + ( + "resolution", + models.PositiveSmallIntegerField( + choices=[ + (0, "Unresolved"), + (1, "Resolved"), + (2, "False positive"), + ], + default=0, + ), + ), + ( + "candidates", + django.contrib.postgres.fields.ArrayField( + base_field=models.IntegerField(), size=None, unique=True + ), + ), + ("primary", models.IntegerField(null=True)), + ("notes", models.TextField(blank=True)), + ("created", models.DateTimeField(auto_now_add=True)), + ("last_updated", models.DateTimeField(auto_now=True)), + ( + "assessment", + models.ForeignKey( + on_delete=django.db.models.deletion.CASCADE, + related_name="duplicates", + to="assessment.assessment", + ), + ), + ( + "resolving_user", + models.ForeignKey( + null=True, + on_delete=django.db.models.deletion.SET_NULL, + related_name="resolved_duplicates", + to=settings.AUTH_USER_MODEL, + ), + ), + ], + ), + ] diff --git a/hawc/apps/lit/models.py b/hawc/apps/lit/models.py index ed396ec210..3b82ab136e 100644 --- a/hawc/apps/lit/models.py +++ b/hawc/apps/lit/models.py @@ -1458,6 +1458,72 @@ def get_description(self) -> str: ) +# add parameters/select on literatureassessment model +class DedupeSettings(models.Model): + # deduper for use in an assessment + # for first pass maybe we just have a global deduper, or static choices, so that we don't have to build this + assessment:"Assessment" + parameters:dict # list of parameters for deduplication? ie schema of dedupe modules to use? + + def build_deduper(self): + # return deduper instance using self.parameters + return + + +# SOFT DELETES + +class SortedArrayField(ArrayField): + pass + +class DuplicateCandidates(models.Model): + assessment = models.ForeignKey( + "assessment.Assessment", on_delete=models.CASCADE, related_name="duplicates" + ) + resolution = models.PositiveSmallIntegerField( + choices=constants.DuplicateResolution, + default=constants.DuplicateResolution.UNRESOLVED + ) + resolving_user = models.ForeignKey(HAWCUser, null=True, on_delete=models.SET_NULL, related_name="resolved_duplicates") + candidates = ArrayField(models.IntegerField(),unique=True) + primary = models.IntegerField(null=True) + notes = models.TextField(blank=True) + created = models.DateTimeField(auto_now_add=True) + last_updated = models.DateTimeField(auto_now=True) + + def generate_unique_identifier(self): + return sorted(self.candidates) + + + def resolve(self,primaries:dict|list[dict]): + if primaries is None: + self.resolution = "False positive" + else: + self.resolution = "Duplicates detected" + # remove reference instances pointed to in self.data that are not primaries + # update reference instances in primaries with data? + self.result = primaries + +# where to put execute method? literatureassessment, manager for dupes model + +""" +WORKFLOW + +User defines deduper for use in assessment +User executes a session that uses a defined deduper +Session stores list of identified candidate duplicate groups +User resolves duplicates in a session; if group status != unresolved, it shows up on this page +Perhaps a seperate session page of resolved groups? ie an "in progress" list view and a "done" list view +Multiple resolutions at once? Or more like screen page in LLR where its do one, click for next (look at conflict resolution) +Should this workflow do anything proactive? ie lets say a candidate group is identified false positive, is it a big deal if it shows up again if a user executes another session w/ same settings? (yes) +Single user right? Not like conflict resolution? THIS IS CORRECT +Do we want this workflow to also happen on import? That would look slightly different + Though maybe we could just have it happen automatically AFTER import, that way it would use the same workflow + If used on import, do we add "choose a deduper" option to created search? or maybe "default" attribute to deduper, whichever one is "default" is used? + Each assessment has undeletable "default" deduper, maybe add noop setting choice for deduper for people who don't want it running on imports? +""" + + + reversion.register(LiteratureAssessment) reversion.register(Search) reversion.register(ReferenceFilterTag) diff --git a/hawc/apps/lit/templates/_duplicate_candidates.html b/hawc/apps/lit/templates/_duplicate_candidates.html new file mode 100644 index 0000000000..fb563ed668 --- /dev/null +++ b/hawc/apps/lit/templates/_duplicate_candidates.html @@ -0,0 +1 @@ +
This is a stub for a duplicate candidate group resolution. (Duplicate candidates {{object.pk}})
\ No newline at end of file diff --git a/hawc/apps/lit/templates/lit/duplicate_candidates.html b/hawc/apps/lit/templates/lit/duplicate_candidates.html new file mode 100644 index 0000000000..7c89df53d0 --- /dev/null +++ b/hawc/apps/lit/templates/lit/duplicate_candidates.html @@ -0,0 +1,14 @@ +{% extends 'assessment-rooted.html' %} + +{% load bs4 %} +{% block content %} +
+

Duplicate candidates

+
+ + {% include "includes/paginator.html" with plural_object_name="duplicate groups" %} +{% endblock content %} diff --git a/hawc/apps/lit/urls.py b/hawc/apps/lit/urls.py index 137e197ee4..f5d2fba513 100644 --- a/hawc/apps/lit/urls.py +++ b/hawc/apps/lit/urls.py @@ -163,4 +163,9 @@ name="workflow-htmx", ), path("api/", include((router.urls, "api"))), + path( + "assessment//duplicate-candidates/", + views.DuplicateCandidatesList.as_view(), + name="duplicate-candidates", + ), ] diff --git a/hawc/apps/lit/views.py b/hawc/apps/lit/views.py index 0eea38e364..7f3f081bc6 100644 --- a/hawc/apps/lit/views.py +++ b/hawc/apps/lit/views.py @@ -1250,3 +1250,20 @@ def venn_reference_list(self, request, *args, **kwargs): "qs": models.Reference.objects.assessment_qs(self.assessment.id).filter(id__in=ids) } return render(request, "lit/components/venn_reference_list.html", context=context) + + + +class DuplicateCandidatesList(BaseList): + parent_model = Assessment + model = models.DuplicateCandidates + template_name = "lit/duplicate_candidates.html" + breadcrumb_active_name = "Duplicate candidates" + + def get_queryset(self): + return ( + super().get_queryset().filter(assessment=self.assessment).filter(resolution=constants.DuplicateResolution.UNRESOLVED) + ) + + def get_context_data(self, **kwargs): + context = super().get_context_data(**kwargs) + From 27c0a25bbc80964c170d97895caa982e8e857621 Mon Sep 17 00:00:00 2001 From: Daniel Rabstejnek Date: Thu, 6 Feb 2025 10:03:44 -0500 Subject: [PATCH 2/5] Changes --- hawc/apps/lit/api.py | 27 +++++++++- .../lit/migrations/0026_reference_hidden.py | 18 +++++++ hawc/apps/lit/models.py | 49 +++++++++++++++---- .../lit/templates/_duplicate_candidates.html | 1 - .../templates/lit/_duplicate_candidates.html | 18 +++++++ .../templates/lit/duplicate_candidates.html | 3 ++ .../templates/lit/duplicate_candidates_2.html | 28 +++++++++++ hawc/apps/lit/urls.py | 11 +++++ hawc/apps/lit/views.py | 27 +++++++++- 9 files changed, 170 insertions(+), 12 deletions(-) create mode 100644 hawc/apps/lit/migrations/0026_reference_hidden.py delete mode 100644 hawc/apps/lit/templates/_duplicate_candidates.html create mode 100644 hawc/apps/lit/templates/lit/_duplicate_candidates.html create mode 100644 hawc/apps/lit/templates/lit/duplicate_candidates_2.html diff --git a/hawc/apps/lit/api.py b/hawc/apps/lit/api.py index c407254202..f0aa33cec0 100644 --- a/hawc/apps/lit/api.py +++ b/hawc/apps/lit/api.py @@ -18,7 +18,7 @@ from ..assessment.constants import AssessmentViewSetPermissions from ..assessment.models import Assessment from ..common.api import OncePerMinuteThrottle, PaginationWithCount -from ..common.helper import FlatExport, cacheable +from ..common.helper import FlatExport, cacheable, tryParseInt from ..common.renderers import PandasRenderers from ..common.serializers import UnusedSerializer from ..common.views import create_object_log @@ -426,3 +426,28 @@ def id_search(self, request, db_id: str, id: str): df=qs.global_df(), filename=f"global-reference-data-{id}", ) + + + + +class DuplicateViewSet( + BaseAssessmentViewSet, +): + model = models.DuplicateCandidates + http_method_names = ["post"] + + @action( + detail=True, methods=("post",), action_perms=AssessmentViewSetPermissions.CAN_EDIT_OBJECT + ) + def resolve_duplicate(self, request, pk): + instance = self.get_object() + assessment = instance.assessment + if not assessment.user_can_edit_object(self.request.user): + raise PermissionDenied() + resolution = request.POST.get("resolution") + notes = request.POST.get("notes","") + if resolution == "none": + instance.resolve(resolution=constants.DuplicateResolution.FALSE_POSITIVE,notes=notes) + if (resolution:=tryParseInt(resolution)) is not None: + instance.resolve(resolution=constants.DuplicateResolution.RESOLVED,primary=resolution,notes=notes) + return Response({"status": "ok"}) diff --git a/hawc/apps/lit/migrations/0026_reference_hidden.py b/hawc/apps/lit/migrations/0026_reference_hidden.py new file mode 100644 index 0000000000..a28c209ed9 --- /dev/null +++ b/hawc/apps/lit/migrations/0026_reference_hidden.py @@ -0,0 +1,18 @@ +# Generated by Django 5.1.4 on 2025-02-05 17:43 + +from django.db import migrations, models + + +class Migration(migrations.Migration): + + dependencies = [ + ("lit", "0025_dedupesettings_duplicatecandidates"), + ] + + operations = [ + migrations.AddField( + model_name="reference", + name="hidden", + field=models.BooleanField(default=False), + ), + ] diff --git a/hawc/apps/lit/models.py b/hawc/apps/lit/models.py index 3b82ab136e..1b1420bf0d 100644 --- a/hawc/apps/lit/models.py +++ b/hawc/apps/lit/models.py @@ -6,6 +6,7 @@ from math import ceil from typing import Self from urllib import parse +import random from celery import chain from celery.result import ResultBase @@ -849,6 +850,7 @@ class Reference(models.Model): null=True, help_text="Used internally for determining when reference was " "originally added", ) + hidden = models.BooleanField(default=False) BREADCRUMB_PARENT = "assessment" @@ -1490,20 +1492,49 @@ class DuplicateCandidates(models.Model): created = models.DateTimeField(auto_now_add=True) last_updated = models.DateTimeField(auto_now=True) + def get_assessment(self): + return self.assessment + + @classmethod + def foobar(cls,assessment): + references = assessment.references.values("pk","title") + candidate_groups = cls.random_execute(references) + cls.objects.bulk_create([cls(assessment=assessment,candidates=[ref["pk"] for ref in group]) for group in candidate_groups]) + + @classmethod + def random_execute(cls,references)->list[list[dict]]: + num_candidates = 2 + if len(references)This is a stub for a duplicate candidate group resolution. (Duplicate candidates {{object.pk}}) \ No newline at end of file diff --git a/hawc/apps/lit/templates/lit/_duplicate_candidates.html b/hawc/apps/lit/templates/lit/_duplicate_candidates.html new file mode 100644 index 0000000000..6694a25fef --- /dev/null +++ b/hawc/apps/lit/templates/lit/_duplicate_candidates.html @@ -0,0 +1,18 @@ +
+
+
+ +
+ {% for candidate in object.candidates %} +
+ +
{{candidate}}
+
+ {% endfor %} + + +
+
\ No newline at end of file diff --git a/hawc/apps/lit/templates/lit/duplicate_candidates.html b/hawc/apps/lit/templates/lit/duplicate_candidates.html index 7c89df53d0..b78dc9e70d 100644 --- a/hawc/apps/lit/templates/lit/duplicate_candidates.html +++ b/hawc/apps/lit/templates/lit/duplicate_candidates.html @@ -4,6 +4,9 @@ {% block content %}

Duplicate candidates

+ {% actions %} + Run deduplication + {% endactions %}
    {% for object in object_list %} diff --git a/hawc/apps/lit/templates/lit/duplicate_candidates_2.html b/hawc/apps/lit/templates/lit/duplicate_candidates_2.html new file mode 100644 index 0000000000..1a3a6f6687 --- /dev/null +++ b/hawc/apps/lit/templates/lit/duplicate_candidates_2.html @@ -0,0 +1,28 @@ +{% extends 'assessment-rooted.html' %} + +{% load bs4 %} +{% block content %} +
    +

    Resolved duplicates

    +
    +
      + {% for object in object_list %} +
      + Group {{object.pk}} +
      +
      + {{object.resolution}} +
      +
      + {{object.candidates}} +
      +
      + {{object.primary}} +
      +
      + {{object.notes}} +
      + {% endfor %} +
    + {% include "includes/paginator.html" with plural_object_name="duplicate groups" %} +{% endblock content %} diff --git a/hawc/apps/lit/urls.py b/hawc/apps/lit/urls.py index f5d2fba513..0c913227be 100644 --- a/hawc/apps/lit/urls.py +++ b/hawc/apps/lit/urls.py @@ -8,6 +8,7 @@ router.register(r"reference", api.ReferenceViewSet, basename="reference") router.register(r"search", api.SearchViewSet, basename="search") router.register(r"tags", api.ReferenceFilterTagViewSet, basename="tags") +router.register(r"duplicate", api.DuplicateViewSet, basename="duplicate") app_name = "lit" urlpatterns = [ @@ -168,4 +169,14 @@ views.DuplicateCandidatesList.as_view(), name="duplicate-candidates", ), + path( + "assessment//duplicate-candidates2/", + views.DuplicateCandidatesList2.as_view(), + name="duplicate-candidates2", + ), + path( + "assessment//duplicate-task/", + views.DuplicateTask.as_view(), + name="duplicate-task", + ), ] diff --git a/hawc/apps/lit/views.py b/hawc/apps/lit/views.py index 7f3f081bc6..638581b530 100644 --- a/hawc/apps/lit/views.py +++ b/hawc/apps/lit/views.py @@ -10,7 +10,7 @@ from django.template import loader from django.urls import reverse, reverse_lazy from django.utils.decorators import method_decorator -from django.views.generic import TemplateView +from django.views.generic import TemplateView, View from ..assessment.constants import AssessmentViewPermissions from ..assessment.models import Assessment @@ -27,6 +27,7 @@ BaseUpdate, create_object_log, htmx_required, + MessageMixin ) from ..udf.cache import TagCache from . import constants, filterset, forms, models @@ -1266,4 +1267,28 @@ def get_queryset(self): def get_context_data(self, **kwargs): context = super().get_context_data(**kwargs) + return context + +class DuplicateCandidatesList2(BaseList): + parent_model = Assessment + model = models.DuplicateCandidates + template_name = "lit/duplicate_candidates_2.html" + breadcrumb_active_name = "Resolved duplicates" + + def get_queryset(self): + return ( + super().get_queryset().filter(assessment=self.assessment).exclude(resolution=constants.DuplicateResolution.UNRESOLVED) + ) + def get_context_data(self, **kwargs): + context = super().get_context_data(**kwargs) + return context + +class DuplicateTask(MessageMixin, View): + success_message = "Deduplication requested." + + def get(self, request, *args, **kwargs): + assessment = get_object_or_404(Assessment, pk=kwargs["pk"]) + models.DuplicateCandidates.foobar(assessment) + self.send_message() + return HttpResponseRedirect(request.META.get('HTTP_REFERER', '/')) \ No newline at end of file From 2ac87e6f1a1232dde9bb25369da949dbb635aeb4 Mon Sep 17 00:00:00 2001 From: Daniel Rabstejnek Date: Tue, 11 Feb 2025 09:04:08 -0500 Subject: [PATCH 3/5] Changed model fields for candidates/primary to relation based, cleanup & improvements --- hawc/apps/lit/api.py | 16 ++- hawc/apps/lit/constants.py | 4 +- ...0025_dedupesettings_duplicatecandidates.py | 1 - .../lit/migrations/0026_reference_hidden.py | 1 - ...ategroup_delete_dedupesettings_and_more.py | 80 +++++++++++ hawc/apps/lit/models.py | 125 +++++++++--------- .../templates/lit/_duplicate_candidates.html | 32 ++--- .../templates/lit/duplicate_candidates.html | 2 +- .../templates/lit/duplicate_candidates_2.html | 30 ++--- hawc/apps/lit/views.py | 24 ++-- 10 files changed, 200 insertions(+), 115 deletions(-) create mode 100644 hawc/apps/lit/migrations/0027_duplicatecandidategroup_delete_dedupesettings_and_more.py diff --git a/hawc/apps/lit/api.py b/hawc/apps/lit/api.py index f0aa33cec0..ddec6cea15 100644 --- a/hawc/apps/lit/api.py +++ b/hawc/apps/lit/api.py @@ -428,12 +428,10 @@ def id_search(self, request, db_id: str, id: str): ) - - class DuplicateViewSet( BaseAssessmentViewSet, ): - model = models.DuplicateCandidates + model = models.DuplicateCandidateGroup http_method_names = ["post"] @action( @@ -445,9 +443,13 @@ def resolve_duplicate(self, request, pk): if not assessment.user_can_edit_object(self.request.user): raise PermissionDenied() resolution = request.POST.get("resolution") - notes = request.POST.get("notes","") + notes = request.POST.get("notes", "") if resolution == "none": - instance.resolve(resolution=constants.DuplicateResolution.FALSE_POSITIVE,notes=notes) - if (resolution:=tryParseInt(resolution)) is not None: - instance.resolve(resolution=constants.DuplicateResolution.RESOLVED,primary=resolution,notes=notes) + instance.resolve(resolution=constants.DuplicateResolution.FALSE_POSITIVE, notes=notes) + if (resolution := tryParseInt(resolution)) is not None: + instance.resolve( + resolution=constants.DuplicateResolution.RESOLVED, + primary_id=resolution, + notes=notes, + ) return Response({"status": "ok"}) diff --git a/hawc/apps/lit/constants.py b/hawc/apps/lit/constants.py index 039980837e..21b2465435 100644 --- a/hawc/apps/lit/constants.py +++ b/hawc/apps/lit/constants.py @@ -28,11 +28,13 @@ class SearchType(models.TextChoices): SEARCH = "s", "Search" IMPORT = "i", "Import" + class DuplicateResolution(models.IntegerChoices): UNRESOLVED = 0, "Unresolved" - RESOLVED = 1, "Resolved" # TODO: change to "primary identified" + RESOLVED = 1, "Resolved" # TODO: change to "primary identified" FALSE_POSITIVE = 2, "False positive" + # generalized/adapted from https://www.crossref.org/blog/dois-and-matching-regular-expressions/ DOI_EXACT = re.compile(r"^10\.\d{4,9}/[^\s]+$") DOI_EXTRACT = re.compile(r"10\.\d{4,9}/[^\s]+") diff --git a/hawc/apps/lit/migrations/0025_dedupesettings_duplicatecandidates.py b/hawc/apps/lit/migrations/0025_dedupesettings_duplicatecandidates.py index 7e28051935..29322ed29a 100644 --- a/hawc/apps/lit/migrations/0025_dedupesettings_duplicatecandidates.py +++ b/hawc/apps/lit/migrations/0025_dedupesettings_duplicatecandidates.py @@ -7,7 +7,6 @@ class Migration(migrations.Migration): - dependencies = [ ("assessment", "0047_alter_labeleditem_options"), ("lit", "0024_workflows"), diff --git a/hawc/apps/lit/migrations/0026_reference_hidden.py b/hawc/apps/lit/migrations/0026_reference_hidden.py index a28c209ed9..91dddf36a1 100644 --- a/hawc/apps/lit/migrations/0026_reference_hidden.py +++ b/hawc/apps/lit/migrations/0026_reference_hidden.py @@ -4,7 +4,6 @@ class Migration(migrations.Migration): - dependencies = [ ("lit", "0025_dedupesettings_duplicatecandidates"), ] diff --git a/hawc/apps/lit/migrations/0027_duplicatecandidategroup_delete_dedupesettings_and_more.py b/hawc/apps/lit/migrations/0027_duplicatecandidategroup_delete_dedupesettings_and_more.py new file mode 100644 index 0000000000..6b0c400c7d --- /dev/null +++ b/hawc/apps/lit/migrations/0027_duplicatecandidategroup_delete_dedupesettings_and_more.py @@ -0,0 +1,80 @@ +# Generated by Django 5.1.4 on 2025-02-10 12:18 + +import django.db.models.deletion +from django.conf import settings +from django.db import migrations, models + + +class Migration(migrations.Migration): + dependencies = [ + ("assessment", "0047_alter_labeleditem_options"), + ("lit", "0026_reference_hidden"), + migrations.swappable_dependency(settings.AUTH_USER_MODEL), + ] + + operations = [ + migrations.CreateModel( + name="DuplicateCandidateGroup", + fields=[ + ( + "id", + models.AutoField( + auto_created=True, + primary_key=True, + serialize=False, + verbose_name="ID", + ), + ), + ( + "resolution", + models.PositiveSmallIntegerField( + choices=[ + (0, "Unresolved"), + (1, "Resolved"), + (2, "False positive"), + ], + default=0, + ), + ), + ("notes", models.TextField(blank=True)), + ("created", models.DateTimeField(auto_now_add=True)), + ("last_updated", models.DateTimeField(auto_now=True)), + ( + "assessment", + models.ForeignKey( + on_delete=django.db.models.deletion.CASCADE, + related_name="duplicates", + to="assessment.assessment", + ), + ), + ( + "candidates", + models.ManyToManyField(related_name="duplicate_candidates", to="lit.reference"), + ), + ( + "primary", + models.ForeignKey( + null=True, + on_delete=django.db.models.deletion.SET_NULL, + related_name="duplicate_primaries", + to="lit.reference", + ), + ), + ( + "resolving_user", + models.ForeignKey( + null=True, + on_delete=django.db.models.deletion.SET_NULL, + related_name="resolved_duplicates", + to=settings.AUTH_USER_MODEL, + ), + ), + ], + ), + migrations.DeleteModel( + name="DedupeSettings", + ), + migrations.DeleteModel( + name="DuplicateCandidates", + ), + ] diff --git a/hawc/apps/lit/models.py b/hawc/apps/lit/models.py index 1b1420bf0d..5cd5534e9e 100644 --- a/hawc/apps/lit/models.py +++ b/hawc/apps/lit/models.py @@ -1,12 +1,12 @@ import html import json import logging +import random import re from copy import copy from math import ceil from typing import Self from urllib import parse -import random from celery import chain from celery.result import ResultBase @@ -1460,100 +1460,93 @@ def get_description(self) -> str: ) -# add parameters/select on literatureassessment model -class DedupeSettings(models.Model): - # deduper for use in an assessment - # for first pass maybe we just have a global deduper, or static choices, so that we don't have to build this - assessment:"Assessment" - parameters:dict # list of parameters for deduplication? ie schema of dedupe modules to use? - - def build_deduper(self): - # return deduper instance using self.parameters - return - - -# SOFT DELETES - -class SortedArrayField(ArrayField): - pass - -class DuplicateCandidates(models.Model): +class DuplicateCandidateGroup(models.Model): assessment = models.ForeignKey( "assessment.Assessment", on_delete=models.CASCADE, related_name="duplicates" ) resolution = models.PositiveSmallIntegerField( - choices=constants.DuplicateResolution, - default=constants.DuplicateResolution.UNRESOLVED + choices=constants.DuplicateResolution, default=constants.DuplicateResolution.UNRESOLVED + ) + resolving_user = models.ForeignKey( + HAWCUser, null=True, on_delete=models.SET_NULL, related_name="resolved_duplicates" + ) + candidates = models.ManyToManyField(Reference, related_name="duplicate_candidates") + primary = models.ForeignKey( + Reference, null=True, on_delete=models.SET_NULL, related_name="duplicate_primaries" ) - resolving_user = models.ForeignKey(HAWCUser, null=True, on_delete=models.SET_NULL, related_name="resolved_duplicates") - candidates = ArrayField(models.IntegerField(),unique=True) - primary = models.IntegerField(null=True) notes = models.TextField(blank=True) created = models.DateTimeField(auto_now_add=True) last_updated = models.DateTimeField(auto_now=True) + @property + def secondaries(self): + return self.candidates.exclude(pk=self.primary_id) + def get_assessment(self): return self.assessment @classmethod - def foobar(cls,assessment): - references = assessment.references.values("pk","title") - candidate_groups = cls.random_execute(references) - cls.objects.bulk_create([cls(assessment=assessment,candidates=[ref["pk"] for ref in group]) for group in candidate_groups]) + def validate_candidates(cls, candidates: list[int]): + qs = cls.objects.annotate(candidates_count=models.Count("candidates")).filter( + candidates_count=len(candidates) + ) + for candidate in candidates: + qs = qs.filter(candidates=candidate) + return not qs.exists() @classmethod - def random_execute(cls,references)->list[list[dict]]: + def find_duplicate_candidate_groups(cls, references) -> list[list[dict]]: num_candidates = 2 - if len(references) list["DuplicateCandidateGroup"]: + references = assessment.references.values("pk", "title") + candidate_groups = cls.find_duplicate_candidate_groups(references) + candidate_groups = [ + group + for group in candidate_groups + if cls.validate_candidates([ref["pk"] for ref in group]) + ] + objs = cls.objects.bulk_create([cls(assessment=assessment) for group in candidate_groups]) + m2m_objs = cls.candidates.through.objects.bulk_create( + [ + cls.candidates.through(duplicatecandidategroup_id=obj.pk, reference_id=ref["pk"]) + for obj, group in zip(objs, candidate_groups, strict=False) + for ref in group + ] + ) - def generate_unique_identifier(self): - return sorted(self.candidates) - def _update_references(self): - # TODO also make primary not hidden? may be unnecessary - duplicate_ids = set(self.candidates)-{self.primary} + duplicate_ids = self.secondaries.values_list("pk", flat=True) self.assessment.references.filter(pk__in=duplicate_ids).update(hidden=True) - - def resolve(self,resolution:constants.DuplicateResolution,primary:int=None,notes:str=""): + # if a "hidden" reference was selected as primary, unhide it + if self.primary.hidden: + self.primary.hidden = False + self.primary.save() + + def resolve( + self, + resolution: constants.DuplicateResolution, + primary_id: int | None = None, + notes: str = "", + ): if resolution == constants.DuplicateResolution.UNRESOLVED: raise ValueError("Resolution must not be unresolved.") if resolution == constants.DuplicateResolution.RESOLVED: - if primary is None: + if primary_id is None: raise ValueError("Primary must not be None if duplicate identified.") - if primary not in self.candidates: + if primary_id not in self.candidates.values_list("pk", flat=True): raise ValueError("Primary must be a candidate.") - self.primary = primary - #self._update_references() + self.primary_id = primary_id + self._update_references() self.resolution = resolution self.notes = notes self.save() -# where to put execute method? literatureassessment, manager for dupes model - - -# DuplicateCandidateGroup - -""" -WORKFLOW - -User defines deduper for use in assessment -User executes a session that uses a defined deduper -Session stores list of identified candidate duplicate groups -User resolves duplicates in a session; if group status != unresolved, it shows up on this page -Perhaps a seperate session page of resolved groups? ie an "in progress" list view and a "done" list view -Multiple resolutions at once? Or more like screen page in LLR where its do one, click for next (look at conflict resolution) -Should this workflow do anything proactive? ie lets say a candidate group is identified false positive, is it a big deal if it shows up again if a user executes another session w/ same settings? (yes) -Single user right? Not like conflict resolution? THIS IS CORRECT -Do we want this workflow to also happen on import? That would look slightly different - Though maybe we could just have it happen automatically AFTER import, that way it would use the same workflow - If used on import, do we add "choose a deduper" option to created search? or maybe "default" attribute to deduper, whichever one is "default" is used? - Each assessment has undeletable "default" deduper, maybe add noop setting choice for deduper for people who don't want it running on imports? -""" - - reversion.register(LiteratureAssessment) reversion.register(Search) diff --git a/hawc/apps/lit/templates/lit/_duplicate_candidates.html b/hawc/apps/lit/templates/lit/_duplicate_candidates.html index 6694a25fef..c0ac4e33a1 100644 --- a/hawc/apps/lit/templates/lit/_duplicate_candidates.html +++ b/hawc/apps/lit/templates/lit/_duplicate_candidates.html @@ -1,18 +1,20 @@
    -
    -
    - + +
    + +
    + {% for candidate in object.candidates.all %} +
    + +
    + {% include 'lit/_reference_with_tags.html' with ref=candidate %}
    - {% for candidate in object.candidates %} -
    - -
    {{candidate}}
    -
    - {% endfor %} - - - +
    + {% endfor %} + + +
    \ No newline at end of file diff --git a/hawc/apps/lit/templates/lit/duplicate_candidates.html b/hawc/apps/lit/templates/lit/duplicate_candidates.html index b78dc9e70d..2c67b17fdb 100644 --- a/hawc/apps/lit/templates/lit/duplicate_candidates.html +++ b/hawc/apps/lit/templates/lit/duplicate_candidates.html @@ -10,7 +10,7 @@

    Duplicate candidates

      {% for object in object_list %} - {% include 'lit/_duplicate_candidates.html' %} + {% include 'lit/_duplicate_candidates.html' %} {% endfor %}
    {% include "includes/paginator.html" with plural_object_name="duplicate groups" %} diff --git a/hawc/apps/lit/templates/lit/duplicate_candidates_2.html b/hawc/apps/lit/templates/lit/duplicate_candidates_2.html index 1a3a6f6687..1732684076 100644 --- a/hawc/apps/lit/templates/lit/duplicate_candidates_2.html +++ b/hawc/apps/lit/templates/lit/duplicate_candidates_2.html @@ -7,21 +7,21 @@

    Resolved duplicates

      {% for object in object_list %} -
      - Group {{object.pk}} -
      -
      - {{object.resolution}} -
      -
      - {{object.candidates}} -
      -
      - {{object.primary}} -
      -
      - {{object.notes}} -
      +
      + Group {{object.pk}} +
      +
      + {{object.resolution}} +
      +
      + {{object.candidates}} +
      +
      + {{object.primary}} +
      +
      + {{object.notes}} +
      {% endfor %}
    {% include "includes/paginator.html" with plural_object_name="duplicate groups" %} diff --git a/hawc/apps/lit/views.py b/hawc/apps/lit/views.py index 638581b530..97986d8861 100644 --- a/hawc/apps/lit/views.py +++ b/hawc/apps/lit/views.py @@ -25,9 +25,9 @@ BaseFilterList, BaseList, BaseUpdate, + MessageMixin, create_object_log, htmx_required, - MessageMixin ) from ..udf.cache import TagCache from . import constants, filterset, forms, models @@ -1253,42 +1253,50 @@ def venn_reference_list(self, request, *args, **kwargs): return render(request, "lit/components/venn_reference_list.html", context=context) - class DuplicateCandidatesList(BaseList): parent_model = Assessment - model = models.DuplicateCandidates + model = models.DuplicateCandidateGroup template_name = "lit/duplicate_candidates.html" breadcrumb_active_name = "Duplicate candidates" def get_queryset(self): return ( - super().get_queryset().filter(assessment=self.assessment).filter(resolution=constants.DuplicateResolution.UNRESOLVED) + super() + .get_queryset() + .filter(assessment=self.assessment) + .filter(resolution=constants.DuplicateResolution.UNRESOLVED) + .prefetch_related("candidates", "candidates__identifiers", "candidates__tags") ) def get_context_data(self, **kwargs): context = super().get_context_data(**kwargs) return context + class DuplicateCandidatesList2(BaseList): parent_model = Assessment - model = models.DuplicateCandidates + model = models.DuplicateCandidateGroup template_name = "lit/duplicate_candidates_2.html" breadcrumb_active_name = "Resolved duplicates" def get_queryset(self): return ( - super().get_queryset().filter(assessment=self.assessment).exclude(resolution=constants.DuplicateResolution.UNRESOLVED) + super() + .get_queryset() + .filter(assessment=self.assessment) + .exclude(resolution=constants.DuplicateResolution.UNRESOLVED) ) def get_context_data(self, **kwargs): context = super().get_context_data(**kwargs) return context + class DuplicateTask(MessageMixin, View): success_message = "Deduplication requested." def get(self, request, *args, **kwargs): assessment = get_object_or_404(Assessment, pk=kwargs["pk"]) - models.DuplicateCandidates.foobar(assessment) + models.DuplicateCandidateGroup.create_duplicate_candidate_groups(assessment) self.send_message() - return HttpResponseRedirect(request.META.get('HTTP_REFERER', '/')) \ No newline at end of file + return HttpResponseRedirect(request.META.get("HTTP_REFERER", "/")) From 55817a7179606d373a7a809a046c748d497d0dbc Mon Sep 17 00:00:00 2001 From: Daniel Rabstejnek Date: Tue, 11 Feb 2025 14:33:12 -0500 Subject: [PATCH 4/5] Refactor names, create task, cleanup duplicate resolution ui --- hawc/apps/lit/models.py | 20 ++-------- hawc/apps/lit/tasks.py | 26 ++++++++++++ .../templates/lit/_duplicate_candidates.html | 40 +++++++++++-------- .../templates/lit/duplicate_candidates.html | 17 -------- .../templates/lit/duplicate_resolution.html | 21 ++++++++++ hawc/apps/lit/templates/lit/overview.html | 1 + ...idates_2.html => resolved_duplicates.html} | 3 ++ hawc/apps/lit/urls.py | 18 ++++----- hawc/apps/lit/views.py | 29 +++++++++----- 9 files changed, 106 insertions(+), 69 deletions(-) delete mode 100644 hawc/apps/lit/templates/lit/duplicate_candidates.html create mode 100644 hawc/apps/lit/templates/lit/duplicate_resolution.html rename hawc/apps/lit/templates/lit/{duplicate_candidates_2.html => resolved_duplicates.html} (80%) diff --git a/hawc/apps/lit/models.py b/hawc/apps/lit/models.py index 5cd5534e9e..f8f030e362 100644 --- a/hawc/apps/lit/models.py +++ b/hawc/apps/lit/models.py @@ -1500,25 +1500,11 @@ def find_duplicate_candidate_groups(cls, references) -> list[list[dict]]: if len(references) < num_candidates: return [] num_groups = min(3, len(references) / num_candidates) - return [random.choices(references, k=num_candidates) for i in range(num_groups)] + return [random.choices(references, k=num_candidates) for i in range(num_groups)] # noqa: S311 @classmethod - def create_duplicate_candidate_groups(cls, assessment) -> list["DuplicateCandidateGroup"]: - references = assessment.references.values("pk", "title") - candidate_groups = cls.find_duplicate_candidate_groups(references) - candidate_groups = [ - group - for group in candidate_groups - if cls.validate_candidates([ref["pk"] for ref in group]) - ] - objs = cls.objects.bulk_create([cls(assessment=assessment) for group in candidate_groups]) - m2m_objs = cls.candidates.through.objects.bulk_create( - [ - cls.candidates.through(duplicatecandidategroup_id=obj.pk, reference_id=ref["pk"]) - for obj, group in zip(objs, candidate_groups, strict=False) - for ref in group - ] - ) + def create_duplicate_candidate_groups(cls, assessment_id: int): + tasks.create_duplicate_candidate_groups.delay(assessment_id) def _update_references(self): duplicate_ids = self.secondaries.values_list("pk", flat=True) diff --git a/hawc/apps/lit/tasks.py b/hawc/apps/lit/tasks.py index 3304a67cac..37535e0d7b 100644 --- a/hawc/apps/lit/tasks.py +++ b/hawc/apps/lit/tasks.py @@ -142,3 +142,29 @@ def fix_pubmed_without_content(): logger.info(f"Attempting to update pubmed content for {num_ids} identifiers") if num_ids > 0: Identifiers.update_pubmed_content(ids) + + +@shared_task +def create_duplicate_candidate_groups(assessment_id: int): + DuplicateCandidateGroup = apps.get_model("lit", "DuplicateCandidateGroup") + assessment = apps.get_model("assessment", "Assessment").objects.get(pk=assessment_id) + references = assessment.references.values("pk", "title") + candidate_groups = DuplicateCandidateGroup.find_duplicate_candidate_groups(references) + candidate_groups = [ + group + for group in candidate_groups + if DuplicateCandidateGroup.validate_candidates([ref["pk"] for ref in group]) + ] + with transaction.atomic(): + objs = DuplicateCandidateGroup.objects.bulk_create( + [DuplicateCandidateGroup(assessment=assessment) for group in candidate_groups] + ) + DuplicateCandidateGroup.candidates.through.objects.bulk_create( + [ + DuplicateCandidateGroup.candidates.through( + duplicatecandidategroup_id=obj.pk, reference_id=ref["pk"] + ) + for obj, group in zip(objs, candidate_groups, strict=False) + for ref in group + ] + ) diff --git a/hawc/apps/lit/templates/lit/_duplicate_candidates.html b/hawc/apps/lit/templates/lit/_duplicate_candidates.html index c0ac4e33a1..f85177de08 100644 --- a/hawc/apps/lit/templates/lit/_duplicate_candidates.html +++ b/hawc/apps/lit/templates/lit/_duplicate_candidates.html @@ -1,20 +1,26 @@
    -
    -
    - -
    - {% for candidate in object.candidates.all %} -
    - -
    - {% include 'lit/_reference_with_tags.html' with ref=candidate %} +
    +
    + +
    + + +
    + {% for candidate in object.candidates.all %} +
    + +
    + {% include 'lit/_reference_with_tags.html' with ref=candidate %} +
    +
    + {% endfor %} +
    +
    -
    - {% endfor %} - - - + + +
    +
    \ No newline at end of file diff --git a/hawc/apps/lit/templates/lit/duplicate_candidates.html b/hawc/apps/lit/templates/lit/duplicate_candidates.html deleted file mode 100644 index 2c67b17fdb..0000000000 --- a/hawc/apps/lit/templates/lit/duplicate_candidates.html +++ /dev/null @@ -1,17 +0,0 @@ -{% extends 'assessment-rooted.html' %} - -{% load bs4 %} -{% block content %} -
    -

    Duplicate candidates

    - {% actions %} - Run deduplication - {% endactions %} -
    -
      - {% for object in object_list %} - {% include 'lit/_duplicate_candidates.html' %} - {% endfor %} -
    - {% include "includes/paginator.html" with plural_object_name="duplicate groups" %} -{% endblock content %} diff --git a/hawc/apps/lit/templates/lit/duplicate_resolution.html b/hawc/apps/lit/templates/lit/duplicate_resolution.html new file mode 100644 index 0000000000..0b49227202 --- /dev/null +++ b/hawc/apps/lit/templates/lit/duplicate_resolution.html @@ -0,0 +1,21 @@ +{% extends 'assessment-rooted.html' %} + +{% load bs4 %} +{% block content %} +
    +

    Duplicate candidates

    + {% actions %} + Resolved duplicates + Identify duplicates + {% endactions %} +
    +
      + {% for object in object_list %} +
    • + {% include 'lit/_duplicate_candidates.html' %} +
      • + {% endfor %} +
    + {% include "includes/paginator.html" with plural_object_name="duplicate groups" %} +{% endblock content %} diff --git a/hawc/apps/lit/templates/lit/overview.html b/hawc/apps/lit/templates/lit/overview.html index 57a7bcbb31..37ba401033 100644 --- a/hawc/apps/lit/templates/lit/overview.html +++ b/hawc/apps/lit/templates/lit/overview.html @@ -27,6 +27,7 @@

    Literature Review

    Tag manually added references Tag untagged references Upload full text URLs + Duplicate resolution
    Exports Download all references diff --git a/hawc/apps/lit/templates/lit/duplicate_candidates_2.html b/hawc/apps/lit/templates/lit/resolved_duplicates.html similarity index 80% rename from hawc/apps/lit/templates/lit/duplicate_candidates_2.html rename to hawc/apps/lit/templates/lit/resolved_duplicates.html index 1732684076..2a833bae9e 100644 --- a/hawc/apps/lit/templates/lit/duplicate_candidates_2.html +++ b/hawc/apps/lit/templates/lit/resolved_duplicates.html @@ -4,6 +4,9 @@ {% block content %}

    Resolved duplicates

    + {% actions %} + Duplicate resolution + {% endactions %}
      {% for object in object_list %} diff --git a/hawc/apps/lit/urls.py b/hawc/apps/lit/urls.py index 0c913227be..560e926a7c 100644 --- a/hawc/apps/lit/urls.py +++ b/hawc/apps/lit/urls.py @@ -165,18 +165,18 @@ ), path("api/", include((router.urls, "api"))), path( - "assessment//duplicate-candidates/", - views.DuplicateCandidatesList.as_view(), - name="duplicate-candidates", + "assessment//duplicate-resolution/", + views.DuplicateResolution.as_view(), + name="duplicate-resolution", ), path( - "assessment//duplicate-candidates2/", - views.DuplicateCandidatesList2.as_view(), - name="duplicate-candidates2", + "assessment//resolved-duplicates/", + views.ResolvedDuplicates.as_view(), + name="resolved-duplicates", ), path( - "assessment//duplicate-task/", - views.DuplicateTask.as_view(), - name="duplicate-task", + "assessment//identify-duplicates/", + views.IdentifyDuplicates.as_view(), + name="identify-duplicates", ), ] diff --git a/hawc/apps/lit/views.py b/hawc/apps/lit/views.py index 97986d8861..d18e9a4edc 100644 --- a/hawc/apps/lit/views.py +++ b/hawc/apps/lit/views.py @@ -1253,11 +1253,13 @@ def venn_reference_list(self, request, *args, **kwargs): return render(request, "lit/components/venn_reference_list.html", context=context) -class DuplicateCandidatesList(BaseList): +class DuplicateResolution(BaseList): parent_model = Assessment model = models.DuplicateCandidateGroup - template_name = "lit/duplicate_candidates.html" - breadcrumb_active_name = "Duplicate candidates" + template_name = "lit/duplicate_resolution.html" + breadcrumb_active_name = "Duplicate resolution" + + paginate_by = 5 def get_queryset(self): return ( @@ -1270,13 +1272,16 @@ def get_queryset(self): def get_context_data(self, **kwargs): context = super().get_context_data(**kwargs) + context["breadcrumbs"] = lit_overview_crumbs( + self.request.user, self.assessment, "Duplicate resolution" + ) return context -class DuplicateCandidatesList2(BaseList): +class ResolvedDuplicates(BaseList): parent_model = Assessment model = models.DuplicateCandidateGroup - template_name = "lit/duplicate_candidates_2.html" + template_name = "lit/resolved_duplicates.html" breadcrumb_active_name = "Resolved duplicates" def get_queryset(self): @@ -1289,14 +1294,20 @@ def get_queryset(self): def get_context_data(self, **kwargs): context = super().get_context_data(**kwargs) + context["breadcrumbs"] = lit_overview_crumbs( + self.request.user, self.assessment, "Resolved duplicates" + ) return context -class DuplicateTask(MessageMixin, View): - success_message = "Deduplication requested." +class IdentifyDuplicates(MessageMixin, View): + success_message = "Duplicate identification requested." def get(self, request, *args, **kwargs): assessment = get_object_or_404(Assessment, pk=kwargs["pk"]) - models.DuplicateCandidateGroup.create_duplicate_candidate_groups(assessment) + if not assessment.user_is_team_member_or_higher(request.user): + raise PermissionDenied() + url = reverse("lit:duplicate-resolution", args=(assessment.pk,)) + models.DuplicateCandidateGroup.create_duplicate_candidate_groups(assessment.pk) self.send_message() - return HttpResponseRedirect(request.META.get("HTTP_REFERER", "/")) + return HttpResponseRedirect(url) From fc55194ac1328bc1cd9adcf0883a071f58eae9d4 Mon Sep 17 00:00:00 2001 From: Daniel Rabstejnek Date: Tue, 11 Feb 2025 14:58:05 -0500 Subject: [PATCH 5/5] Updated permissions --- hawc/apps/lit/api.py | 9 +++++-- hawc/apps/lit/models.py | 2 ++ .../templates/lit/_duplicate_candidates.html | 26 ------------------- .../templates/lit/duplicate_resolution.html | 25 +++++++++++++++++- hawc/apps/lit/templates/lit/overview.html | 6 ++++- .../templates/lit/resolved_duplicates.html | 8 +++--- hawc/apps/lit/views.py | 4 ++- 7 files changed, 46 insertions(+), 34 deletions(-) delete mode 100644 hawc/apps/lit/templates/lit/_duplicate_candidates.html diff --git a/hawc/apps/lit/api.py b/hawc/apps/lit/api.py index ddec6cea15..18ef4ed3ed 100644 --- a/hawc/apps/lit/api.py +++ b/hawc/apps/lit/api.py @@ -440,15 +440,20 @@ class DuplicateViewSet( def resolve_duplicate(self, request, pk): instance = self.get_object() assessment = instance.assessment - if not assessment.user_can_edit_object(self.request.user): + if not assessment.user_can_edit_object(request.user): raise PermissionDenied() resolution = request.POST.get("resolution") notes = request.POST.get("notes", "") if resolution == "none": - instance.resolve(resolution=constants.DuplicateResolution.FALSE_POSITIVE, notes=notes) + instance.resolve( + resolution=constants.DuplicateResolution.FALSE_POSITIVE, + resolving_user=request.user, + notes=notes, + ) if (resolution := tryParseInt(resolution)) is not None: instance.resolve( resolution=constants.DuplicateResolution.RESOLVED, + resolving_user=request.user, primary_id=resolution, notes=notes, ) diff --git a/hawc/apps/lit/models.py b/hawc/apps/lit/models.py index f8f030e362..d21c36f67f 100644 --- a/hawc/apps/lit/models.py +++ b/hawc/apps/lit/models.py @@ -1517,6 +1517,7 @@ def _update_references(self): def resolve( self, resolution: constants.DuplicateResolution, + resolving_user: HAWCUser, primary_id: int | None = None, notes: str = "", ): @@ -1530,6 +1531,7 @@ def resolve( self.primary_id = primary_id self._update_references() self.resolution = resolution + self.resolving_user = resolving_user self.notes = notes self.save() diff --git a/hawc/apps/lit/templates/lit/_duplicate_candidates.html b/hawc/apps/lit/templates/lit/_duplicate_candidates.html deleted file mode 100644 index f85177de08..0000000000 --- a/hawc/apps/lit/templates/lit/_duplicate_candidates.html +++ /dev/null @@ -1,26 +0,0 @@ -
      -
      -
      -
      -
      - - -
      - {% for candidate in object.candidates.all %} -
      - -
      - {% include 'lit/_reference_with_tags.html' with ref=candidate %} -
      -
      - {% endfor %} -
      - -
      - -
      -
      -
      -
      \ No newline at end of file diff --git a/hawc/apps/lit/templates/lit/duplicate_resolution.html b/hawc/apps/lit/templates/lit/duplicate_resolution.html index 0b49227202..e2b6e0c7d7 100644 --- a/hawc/apps/lit/templates/lit/duplicate_resolution.html +++ b/hawc/apps/lit/templates/lit/duplicate_resolution.html @@ -13,7 +13,30 @@

      Duplicate candidates

      {% for object in object_list %}
    • - {% include 'lit/_duplicate_candidates.html' %} +
      +
      +
      +
      + + +
      + {% for candidate in object.candidates.all %} +
      + +
      + {% include 'lit/_reference_with_tags.html' with ref=candidate %} +
      +
      + {% endfor %} +
      + +
      + +
      +
      +
      • {% endfor %}
    diff --git a/hawc/apps/lit/templates/lit/overview.html b/hawc/apps/lit/templates/lit/overview.html index 37ba401033..d83f3c885f 100644 --- a/hawc/apps/lit/templates/lit/overview.html +++ b/hawc/apps/lit/templates/lit/overview.html @@ -27,7 +27,11 @@

    Literature Review

    Tag manually added references Tag untagged references Upload full text URLs - Duplicate resolution + {% if obj_perms.edit %} + Duplicate resolution + {% else %} + Resolved duplicates + {% endif %}
    Exports Download all references diff --git a/hawc/apps/lit/templates/lit/resolved_duplicates.html b/hawc/apps/lit/templates/lit/resolved_duplicates.html index 2a833bae9e..7bda5535f1 100644 --- a/hawc/apps/lit/templates/lit/resolved_duplicates.html +++ b/hawc/apps/lit/templates/lit/resolved_duplicates.html @@ -4,9 +4,11 @@ {% block content %}

    Resolved duplicates

    - {% actions %} - Duplicate resolution - {% endactions %} + {% if obj_perms.edit %} + {% actions %} + Duplicate resolution + {% endactions %} + {% endif %}
      {% for object in object_list %} diff --git a/hawc/apps/lit/views.py b/hawc/apps/lit/views.py index d18e9a4edc..e607ed933b 100644 --- a/hawc/apps/lit/views.py +++ b/hawc/apps/lit/views.py @@ -1258,6 +1258,7 @@ class DuplicateResolution(BaseList): model = models.DuplicateCandidateGroup template_name = "lit/duplicate_resolution.html" breadcrumb_active_name = "Duplicate resolution" + assessment_permission = AssessmentViewPermissions.TEAM_MEMBER_EDITABLE paginate_by = 5 @@ -1283,6 +1284,7 @@ class ResolvedDuplicates(BaseList): model = models.DuplicateCandidateGroup template_name = "lit/resolved_duplicates.html" breadcrumb_active_name = "Resolved duplicates" + assessment_permission = AssessmentViewPermissions.TEAM_MEMBER def get_queryset(self): return ( @@ -1305,7 +1307,7 @@ class IdentifyDuplicates(MessageMixin, View): def get(self, request, *args, **kwargs): assessment = get_object_or_404(Assessment, pk=kwargs["pk"]) - if not assessment.user_is_team_member_or_higher(request.user): + if not assessment.user_can_edit_object(request.user): raise PermissionDenied() url = reverse("lit:duplicate-resolution", args=(assessment.pk,)) models.DuplicateCandidateGroup.create_duplicate_candidate_groups(assessment.pk)