Skip to content

Improved indexing progress feedback #2516

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

Merged
henrikek merged 1 commit into from
Oct 13, 2026
Merged

Improved indexing progress feedback #2516

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
10 changes: 0 additions & 10 deletions ESSArch_Core/cli/commands/search.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -85,16 +85,6 @@ def rebuild(indexes, batch_size, remove_stale, do_not_delete_old, index_file_con
For the document index, file content extraction is disabled by default.
Use --index-file-content to enable content extraction from the original
files during the rebuild.

:param indexes: Names of the indices to rebuild. Multiple indices can
be specified.
:param batch_size: Number of documents to process in each batch.
:param remove_stale: Whether to remove indexed documents that no longer
exist in the database.
:param do_not_delete_old: Whether to keep the existing index before
rebuilding.
:param index_file_content: Whether to extract and index content from
original files in the document index.
"""
indexes = get_indexes(indexes)

Expand Down
52 changes: 48 additions & 4 deletions ESSArch_Core/ip/models.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2804,6 +2804,7 @@ def write_to_search_index(self, task):
:param task: The task associated with the indexing operation
"""
logger = logging.getLogger('essarch.ip')
task.set_progress(1, total=100)
srcdir = self.object_path
ct_profile = self.get_profile('content_type')
indexed_files = []
Expand All @@ -2812,6 +2813,7 @@ def write_to_search_index(self, task):

profile_type = self.get_package_type_display().lower()
index_cits = self.get_profile_data(profile_type).get('index_cits', True)
total_cits_imported = 0
if ct_profile is not None and (index_cits is True or index_cits == 'True'):
cts = self.get_content_type_file()
if cts is not None:
Expand All @@ -2822,8 +2824,11 @@ def write_to_search_index(self, task):
except KeyError:
logger.exception('No content type importer specified in profile')
raise
ct_importer = get_importer(ct_importer_name)(task)
ct_importer = get_importer(ct_importer_name)(task.get_processtask())
indexed_files = ct_importer.import_content(cts, ip=self)
if indexed_files is not None:
total_cits_imported = len(indexed_files)
task.create_success_event("Content type specification imported")
else:
err = "Content type specification file not found"
logger.error('{err}: {path}'.format(err=err, path=cts))
Expand All @@ -2835,15 +2840,23 @@ def write_to_search_index(self, task):
except KeyError:
logger.exception('No content type importer specified in profile')
raise
ct_importer = get_importer(ct_importer_name)(task)
ct_importer = get_importer(ct_importer_name)(task.get_processtask())
indexed_files = ct_importer.import_content(cts, ip=self)
if indexed_files is not None:
total_cits_imported = len(indexed_files)
task.create_success_event("Content type specification imported")
task.set_progress(2, total=100)

# Profile settings control file indexing and content extraction.
index_files = self.get_profile_data(profile_type).get('index_files', True)
index_files_content = self.get_profile_data(profile_type).get('index_files_content', True)

# Reserved for optionally reusing content from existing Elasticsearch documents.
index_files_content_reuse_if_exists = False

total_documents_to_index = 0
total_dirs = 0

if index_files is True or index_files == 'True':
group = None
try:
Expand All @@ -2857,6 +2870,7 @@ def write_to_search_index(self, task):
for d in dirs:
src = os.path.join(root, d)
index_path(self, src, group=group, index_file_content=index_files_content)
total_dirs += 1

for f in files:
src = os.path.join(root, f)
Expand All @@ -2878,16 +2892,22 @@ def write_to_search_index(self, task):
file_ids.append(str(tag_version.pk))
file_paths.append((src, tag_version))

task.set_progress(3, total=100)
existing_content = {}
if index_files_content_reuse_if_exists:
task.set_progress(4, total=100)
existing_content = get_existing_file_content(file_ids)
task.set_progress(5, total=100)

documents = []
document_paths = {}
total_files_to_process = len(file_paths)

for src, tag_version in file_paths:
for processed_files, (src, tag_version) in enumerate(file_paths, start=1):
content = None
if index_files_content_reuse_if_exists:
content = existing_content.get(str(tag_version.pk))

doc, _ = index_document(
tag_version,
src,
Expand All @@ -2902,13 +2922,37 @@ def write_to_search_index(self, task):
group.add_object(tag_version)

documents.append(doc)
document_paths[str(doc.meta.id)] = src

if total_files_to_process:
# Reserve 10–70% for file processing and content extraction.
progress = 10 + int(processed_files / total_files_to_process * 60)
task.set_progress(progress, total=100)

# Index all collected file documents in a single bulk operation
# instead of sending an individual Elasticsearch request per file.
bulk_index_documents(documents)
total_documents_to_index = len(documents)

def report_progress(processed, succeeded, failed):
if total_documents_to_index:
progress = 70 + int(processed / total_documents_to_index * 30)
task.set_progress(progress, total=100)

succeeded, failed = bulk_index_documents(
documents,
progress_callback=report_progress,
document_paths=document_paths,
)

task.create_success_event(
f"Indexed {succeeded} files successfully; "
f"{failed} files failed"
)

InformationPackageDocument.from_obj(self).save()

return total_documents_to_index, total_dirs, total_cits_imported

def get_cached_storage_object(self):
cache_method = self.policy.cache_storage
return self.storage.get(
Expand Down
6 changes: 4 additions & 2 deletions ESSArch_Core/ip/tasks.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -582,10 +582,12 @@ def PreserveInformationPackage(self, storage_method_pk, temp_path=None):
pretty_mb_per_sec(mb_per_sec))


@app.task(bind=True)
@app.task(bind=True, event_type=50770)
def WriteInformationPackageToSearchIndex(self):
ip = self.get_information_package()
ip.write_to_search_index(self.get_processtask())
files_count, dirs_count, cits_count = ip.write_to_search_index(self)

return f"Indexed {files_count} files, {dirs_count} dirs and {cits_count} cits"


@app.task(bind=True)
Expand Down
71 changes: 52 additions & 19 deletions ESSArch_Core/search/ingest.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,7 +4,7 @@

from django.conf import settings
from elasticsearch import ConnectionError, RequestError, TransportError
from elasticsearch.helpers import streaming_bulk
from elasticsearch.helpers import BulkIndexError, streaming_bulk
from elasticsearch_dsl.connections import get_connection

from ESSArch_Core.fixity.format import FormatIdentifier
Expand All @@ -24,10 +24,54 @@
format_identifier = FormatIdentifier()


def bulk_index_documents(documents, batch_size=100):
def bulk_index_documents(documents, batch_size=100, progress_callback=None, document_paths=None):
client = get_connection('default')
logger = logging.getLogger('essarch.search.ingest')

document_paths = document_paths or {}
actions = []
succeeded = 0
failed = 0
processed = 0

def index_batch(batch):
nonlocal succeeded, failed, processed

try:
for success, info in streaming_bulk(
client,
batch,
chunk_size=batch_size,
raise_on_error=True,
):
if success:
succeeded += 1
else:
failed += 1
logger.error(
'Failed to bulk index document: %s', info
)

processed += 1

if progress_callback:
progress_callback(processed, succeeded, failed)

except BulkIndexError as exc:
for error in exc.errors:
operation = next(iter(error))
details = error[operation]
document_id = details.get('_id')
path = document_paths.get(document_id, '<unknown path>')

logger.error(
'Failed to index file "%s" (document ID: %s): %s',
path,
document_id,
details.get('error'),
)

raise

for doc in documents:
actions.append({
Expand All @@ -37,26 +81,13 @@ def bulk_index_documents(documents, batch_size=100):
})

if len(actions) >= batch_size:
for success, info in streaming_bulk(
client,
actions,
chunk_size=batch_size,
):
if not success:
logger = logging.getLogger('essarch.search.ingest')
logger.error('Failed to bulk index document: %s', info)

index_batch(actions)
actions = []

if actions:
for success, info in streaming_bulk(
client,
actions,
chunk_size=batch_size,
):
if not success:
logger = logging.getLogger('essarch.search.ingest')
logger.error('Failed to bulk index document: %s', info)
index_batch(actions)

return succeeded, failed


def index_document(tag_version, filepath, index_file_content=True, existing_content=None, save=True):
Expand Down Expand Up @@ -256,6 +287,8 @@ def index_path(ip, path, parent=None, group=None, index_file_content=True,

logger.debug('indexing {}'.format(path))

doc = None

if index:
if isfile:
doc, tag_version = index_document(tag_version, path, index_file_content,
Expand Down
Loading