Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Improve code search #501

Merged
merged 3 commits into from
Feb 11, 2025
Merged

Improve code search #501

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
babeba2
added explicit join conditions (on clauses)
bigabig Feb 11, 2025
f523396
code search now includes SentenceAnnotations
bigabig Feb 11, 2025
3c414d2
improved search statistic code query
bigabig Feb 11, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 2 additions & 4 deletions backend/src/app/core/analysis/search_statistics/search_statistics.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,7 +6,6 @@
from app.core.data.crud.project_metadata import crud_project_meta
from app.core.data.dto.search_stats import KeywordStat, SpanEntityStat, TagStat
from app.core.data.orm.annotation_document import AnnotationDocumentORM
from app.core.data.orm.code import CodeORM
from app.core.data.orm.document_tag import (
DocumentTagORM,
SourceDocumentDocumentTagLinkTable,
Expand Down Expand Up @@ -141,10 +140,9 @@ def compute_code_statistics(
)
.join(SpanTextORM.span_annotations)
.join(SpanAnnotationORM.annotation_document)
.join(SpanAnnotationORM.code)
.group_by(SpanTextORM.id)
.filter(
CodeORM.id == code_id,
SpanAnnotationORM.code_id == code_id,
AnnotationDocumentORM.source_document_id.in_(list(sdoc_ids)),
)
.order_by(count.desc())
Expand All @@ -169,7 +167,7 @@ def compute_code_statistics(
.join(SpanAnnotationORM.code)
.group_by(SpanTextORM.id)
.filter(
CodeORM.id == code_id,
SpanAnnotationORM.code_id == code_id,
SpanTextORM.id.in_(span_text_ids),
)
.order_by(func.array_position(span_text_ids, SpanTextORM.id))
Expand Down
56 changes: 35 additions & 21 deletions backend/src/app/core/analysis/timeline_analysis/timeline_analysis_columns.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,13 +2,12 @@
from sqlalchemy.dialects.postgresql import ARRAY, array, array_agg

from app.core.data.orm.annotation_document import AnnotationDocumentORM
from app.core.data.orm.code import CodeORM
from app.core.data.orm.document_tag import DocumentTagORM
from app.core.data.orm.sentence_annotation import SentenceAnnotationORM
from app.core.data.orm.source_document import SourceDocumentORM
from app.core.data.orm.span_annotation import SpanAnnotationORM
from app.core.data.orm.span_text import SpanTextORM
from app.core.data.orm.user import UserORM
from app.core.db.sql_utils import aggregate_ids
from app.core.db.sql_utils import aggregate_ids, aggregate_two_ids
from app.core.search.column_info import AbstractColumns
from app.core.search.filtering_operators import FilterOperator, FilterValueType
from app.core.search.search_builder import SearchBuilder
Expand Down Expand Up @@ -100,57 +99,72 @@ def add_subquery_filter_statements(self, query_builder: SearchBuilder):
)
case TimelineAnalysisColumns.CODE_ID_LIST:
query_builder._add_subquery_column(
aggregate_ids(
CodeORM.id, label=TimelineAnalysisColumns.CODE_ID_LIST.value
aggregate_two_ids(
SpanAnnotationORM.code_id,
SentenceAnnotationORM.code_id,
label=TimelineAnalysisColumns.CODE_ID_LIST.value,
)
)
query_builder._join_subquery(
SourceDocumentORM.annotation_documents,
AnnotationDocumentORM,
AnnotationDocumentORM.source_document_id == SourceDocumentORM.id,
isouter=True,
)
query_builder._join_subquery(
SpanAnnotationORM,
SpanAnnotationORM.annotation_document_id
== AnnotationDocumentORM.id,
isouter=True,
)
query_builder._join_subquery(
SpanAnnotationORM.code,
SentenceAnnotationORM,
SentenceAnnotationORM.annotation_document_id
== AnnotationDocumentORM.id,
isouter=True,
)

case TimelineAnalysisColumns.USER_ID_LIST:
query_builder._add_subquery_column(
aggregate_ids(
UserORM.id, TimelineAnalysisColumns.USER_ID_LIST.value
AnnotationDocumentORM.user_id,
TimelineAnalysisColumns.USER_ID_LIST.value,
)
)
query_builder._join_subquery(
SourceDocumentORM.annotation_documents,
isouter=True,
)
query_builder._join_subquery(
AnnotationDocumentORM.user,
AnnotationDocumentORM,
AnnotationDocumentORM.source_document_id == SourceDocumentORM.id,
isouter=True,
)
case TimelineAnalysisColumns.SPAN_ANNOTATIONS:
query_builder._add_subquery_column(
cast(
array_agg(
func.distinct(
array([cast(CodeORM.id, String), SpanTextORM.text])
array(
[
cast(SpanAnnotationORM.code_id, String),
SpanTextORM.text,
]
)
),
),
ARRAY(String, dimensions=2),
).label(TimelineAnalysisColumns.SPAN_ANNOTATIONS.value)
)
query_builder._join_subquery(
SourceDocumentORM.annotation_documents,
isouter=True,
)
query_builder._join_subquery(
AnnotationDocumentORM.span_annotations,
AnnotationDocumentORM,
AnnotationDocumentORM.source_document_id == SourceDocumentORM.id,
isouter=True,
)
query_builder._join_subquery(
SpanAnnotationORM.span_text,
SpanAnnotationORM,
SpanAnnotationORM.annotation_document_id
== AnnotationDocumentORM.id,
isouter=True,
)
query_builder._join_subquery(
SpanAnnotationORM.code,
SpanTextORM,
SpanTextORM.id == SpanAnnotationORM.span_text_id,
isouter=True,
)

Expand Down
58 changes: 37 additions & 21 deletions backend/src/app/core/analysis/word_frequency_analysis/word_frequency_columns.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,14 +2,13 @@
from sqlalchemy.dialects.postgresql import ARRAY, array, array_agg

from app.core.data.orm.annotation_document import AnnotationDocumentORM
from app.core.data.orm.code import CodeORM
from app.core.data.orm.document_tag import DocumentTagORM
from app.core.data.orm.sentence_annotation import SentenceAnnotationORM
from app.core.data.orm.source_document import SourceDocumentORM
from app.core.data.orm.span_annotation import SpanAnnotationORM
from app.core.data.orm.span_text import SpanTextORM
from app.core.data.orm.user import UserORM
from app.core.data.orm.word_frequency import WordFrequencyORM
from app.core.db.sql_utils import aggregate_ids
from app.core.db.sql_utils import aggregate_ids, aggregate_two_ids
from app.core.search.column_info import AbstractColumns
from app.core.search.filtering_operators import FilterOperator, FilterValueType
from app.core.search.search_builder import SearchBuilder
Expand Down Expand Up @@ -156,55 +155,72 @@ def add_subquery_filter_statements(self, query_builder: SearchBuilder):
)
case WordFrequencyColumns.CODE_ID_LIST:
query_builder._add_subquery_column(
aggregate_ids(
CodeORM.id, label=WordFrequencyColumns.CODE_ID_LIST.value
aggregate_two_ids(
SpanAnnotationORM.code_id,
SentenceAnnotationORM.code_id,
label=WordFrequencyColumns.CODE_ID_LIST.value,
)
)
query_builder._join_subquery(
SourceDocumentORM.annotation_documents,
AnnotationDocumentORM,
AnnotationDocumentORM.source_document_id == SourceDocumentORM.id,
isouter=True,
)
query_builder._join_subquery(
SpanAnnotationORM.code,
SpanAnnotationORM,
SpanAnnotationORM.annotation_document_id
== AnnotationDocumentORM.id,
isouter=True,
)
case WordFrequencyColumns.USER_ID_LIST:
query_builder._add_subquery_column(
aggregate_ids(UserORM.id, WordFrequencyColumns.USER_ID_LIST.value)
)
query_builder._join_subquery(
SourceDocumentORM.annotation_documents,
SentenceAnnotationORM,
SentenceAnnotationORM.annotation_document_id
== AnnotationDocumentORM.id,
isouter=True,
)

case WordFrequencyColumns.USER_ID_LIST:
query_builder._add_subquery_column(
aggregate_ids(
AnnotationDocumentORM.user_id,
WordFrequencyColumns.USER_ID_LIST.value,
)
)
query_builder._join_subquery(
AnnotationDocumentORM.user,
AnnotationDocumentORM,
AnnotationDocumentORM.source_document_id == SourceDocumentORM.id,
isouter=True,
)
case WordFrequencyColumns.SPAN_ANNOTATIONS:
query_builder._add_subquery_column(
cast(
array_agg(
func.distinct(
array([cast(CodeORM.id, String), SpanTextORM.text])
array(
[
cast(SpanAnnotationORM.code_id, String),
SpanTextORM.text,
]
)
),
),
ARRAY(String, dimensions=2),
).label(WordFrequencyColumns.SPAN_ANNOTATIONS.value)
)
query_builder._join_subquery(
SourceDocumentORM.annotation_documents,
isouter=True,
)
query_builder._join_subquery(
AnnotationDocumentORM.span_annotations,
AnnotationDocumentORM,
AnnotationDocumentORM.source_document_id == SourceDocumentORM.id,
isouter=True,
)
query_builder._join_subquery(
SpanAnnotationORM.span_text,
SpanAnnotationORM,
SpanAnnotationORM.annotation_document_id
== AnnotationDocumentORM.id,
isouter=True,
)
query_builder._join_subquery(
SpanAnnotationORM.code,
SpanTextORM,
SpanTextORM.id == SpanAnnotationORM.span_text_id,
isouter=True,
)

Expand Down
13 changes: 13 additions & 0 deletions backend/src/app/core/db/sql_utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,3 +9,16 @@ def aggregate_ids(column: InstrumentedAttribute, label: str):
None,
type_=ARRAY(Integer),
).label(label)


def aggregate_two_ids(
column1: InstrumentedAttribute, column2: InstrumentedAttribute, label: str
):
return func.array_remove(
func.array_cat(
array_agg(func.distinct(column1), type_=ARRAY(Integer)),
array_agg(func.distinct(column2), type_=ARRAY(Integer)),
),
None,
type_=ARRAY(Integer),
).label(label)
11 changes: 9 additions & 2 deletions backend/src/app/core/search/bbox_anno_search/bbox_anno_search_columns.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -81,8 +81,15 @@ def add_subquery_filter_statements(self, query_builder: SearchBuilder):
label=BBoxColumns.DOCUMENT_TAG_ID_LIST.value,
)
)
query_builder._join_subquery(BBoxAnnotationORM.annotation_document)
query_builder._join_subquery(AnnotationDocumentORM.source_document)
query_builder._join_subquery(
AnnotationDocumentORM,
AnnotationDocumentORM.id
== BBoxAnnotationORM.annotation_document_id,
)
query_builder._join_subquery(
SourceDocumentORM,
SourceDocumentORM.id == AnnotationDocumentORM.source_document_id,
)
query_builder._join_subquery(
SourceDocumentORM.document_tags, isouter=True
)
Expand Down
60 changes: 40 additions & 20 deletions backend/src/app/core/search/sdoc_search/sdoc_search_columns.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,13 +2,15 @@
from sqlalchemy.dialects.postgresql import ARRAY, array, array_agg

from app.core.data.orm.annotation_document import AnnotationDocumentORM
from app.core.data.orm.code import CodeORM
from app.core.data.orm.document_tag import DocumentTagORM
from app.core.data.orm.sentence_annotation import SentenceAnnotationORM
from app.core.data.orm.source_document import SourceDocumentORM
from app.core.data.orm.span_annotation import SpanAnnotationORM
from app.core.data.orm.span_text import SpanTextORM
from app.core.data.orm.user import UserORM
from app.core.db.sql_utils import aggregate_ids
from app.core.db.sql_utils import (
aggregate_ids,
aggregate_two_ids,
)
from app.core.search.column_info import AbstractColumns
from app.core.search.filtering_operators import FilterOperator, FilterValueType
from app.core.search.search_builder import SearchBuilder
Expand Down Expand Up @@ -111,53 +113,71 @@ def add_subquery_filter_statements(self, query_builder: SearchBuilder):
)
case SdocColumns.CODE_ID_LIST:
query_builder._add_subquery_column(
aggregate_ids(CodeORM.id, label=SdocColumns.CODE_ID_LIST.value)
aggregate_two_ids(
SpanAnnotationORM.code_id,
SentenceAnnotationORM.code_id,
label=SdocColumns.CODE_ID_LIST.value,
)
)
query_builder._join_subquery(
SourceDocumentORM.annotation_documents,
AnnotationDocumentORM,
AnnotationDocumentORM.source_document_id == SourceDocumentORM.id,
isouter=True,
)
query_builder._join_subquery(
SpanAnnotationORM.code,
SpanAnnotationORM,
SpanAnnotationORM.annotation_document_id
== AnnotationDocumentORM.id,
isouter=True,
)
case SdocColumns.USER_ID_LIST:
query_builder._add_subquery_column(
aggregate_ids(UserORM.id, SdocColumns.USER_ID_LIST.value)
)
query_builder._join_subquery(
SourceDocumentORM.annotation_documents,
SentenceAnnotationORM,
SentenceAnnotationORM.annotation_document_id
== AnnotationDocumentORM.id,
isouter=True,
)

case SdocColumns.USER_ID_LIST:
query_builder._add_subquery_column(
aggregate_ids(
AnnotationDocumentORM.user_id, SdocColumns.USER_ID_LIST.value
)
)
query_builder._join_subquery(
AnnotationDocumentORM.user,
AnnotationDocumentORM,
AnnotationDocumentORM.source_document_id == SourceDocumentORM.id,
isouter=True,
)
case SdocColumns.SPAN_ANNOTATIONS:
query_builder._add_subquery_column(
cast(
array_agg(
func.distinct(
array([cast(CodeORM.id, String), SpanTextORM.text])
array(
[
cast(SpanAnnotationORM.code_id, String),
SpanTextORM.text,
]
)
),
),
ARRAY(String, dimensions=2),
).label(SdocColumns.SPAN_ANNOTATIONS.value)
)
query_builder._join_subquery(
SourceDocumentORM.annotation_documents,
isouter=True,
)
query_builder._join_subquery(
AnnotationDocumentORM.span_annotations,
AnnotationDocumentORM,
AnnotationDocumentORM.source_document_id == SourceDocumentORM.id,
isouter=True,
)
query_builder._join_subquery(
SpanAnnotationORM.span_text,
SpanAnnotationORM,
SpanAnnotationORM.annotation_document_id
== AnnotationDocumentORM.id,
isouter=True,
)
query_builder._join_subquery(
SpanAnnotationORM.code,
SpanTextORM,
SpanTextORM.id == SpanAnnotationORM.span_text_id,
isouter=True,
)

Expand Down
Loading
Loading