Remove unused tables #976
Assignees
Labels
Database
Issues with Frank Lab database, not Spyglass code
Comments
|
All these would be good to clean up as far as I know. |
CBroz1
added a commit
to CBroz1/spyglass
that referenced
this issue
Jun 10, 2024
edeno
pushed a commit
that referenced
this issue
Jun 18, 2024
* Give UUID to artifact interval * Add ability to set smoothing sigma in get_firing_rate (#994) * add option to set spike smoothing sigma * update changelog * Add docstrings to SortedSpikesGroup and Decoding methods (#996) * Add docstrings * update changelog * fix spelling --------- Co-authored-by: Samuel Bray <[email protected]> * Add Common Errors doc (#997) * Add Common Errors * Update changelog * Mua notebook (#998) * documented some of mua notebook * mua notebook documented * documented some of mua notebook * synced py script * Dandi export and read (#956) * compile exported files, download dandiset, and organize * add function to translate files into dandi-compatible names * add table to store dandi name translation and steps to populate * add dandiset validation * add function to fetch nwb from dandi * add function to change obj_id of nwb_file * add dandi upload call and fix circular import * debug dandi file streaming * fix circular import * resolve dandi-streamed files with fetch_nwb * implement review comments * add admin tools to fix common dandi discrepencies * implement tool to cleanup common dandi errors * add dandi export to tutorial * fix linting * update changelog * fix spelling * style changes from review * reorganize function locations * fix circular import * make dandi dependency optional in imports * store dandi instance of data in DandiPath * resolve case of pre-existing dandi entries for export * cleanup bugs from refactor * update notebook * Apply suggestions from code review Co-authored-by: Chris Broz <[email protected]> * add requested changes from review * make method check_admin_privilege in LabMember --------- Co-authored-by: Chris Broz <[email protected]> * Minor fixes (#999) * give analysis nwb new uuid when created * fix function argument * update changelog * Fix bug in change in analysis_file object_id (#1004) * fix bug in change in analysis_file_object_id * update changelog * Remove classes for usused tables (#1003) * #976 * Remove notebook reference * Non-daemon parallel populate (#1001) * initial non daemon parallel commit * resolve namespace and pickling errors * fix linting * update changelog * implement review comments * add parallel_make flag to spikesorting recording tables * fix multiprocessing spawn error on mac * move propert --------- Co-authored-by: Samuel Bray <[email protected]> * Update pipeline column for IntervalList --------- Co-authored-by: Samuel Bray <[email protected]> Co-authored-by: Samuel Bray <[email protected]> Co-authored-by: Chris Broz <[email protected]> Co-authored-by: Denisse Morales-Rodriguez <[email protected]> Co-authored-by: Samuel Bray <[email protected]>
|
At @samuelbray32 's suggestion, I wrote a script to look at tables on shared prefixes, without a Spyglass class, with contents. They are listed at the end of the script below. Scriptimport importlib
import inspect
import pkgutil
import datajoint as dj
from datajoint.user_tables import TableMeta
from tqdm import tqdm
from spyglass.utils.database_settings import SHARED_MODULES
class SearchPkg:
def __init__(self, package_name="spyglass"):
self.package = importlib.import_module(package_name)
self.class_cache = set()
self.load_cache()
def find_subclasses(self, module, parent_class):
for name, obj in inspect.getmembers(parent_class, inspect.isclass):
if isinstance(obj, TableMeta) and obj.__module__ == module.__name__:
name = obj.full_table_name.replace("`", "")
if name in self.class_cache:
continue
print(f"Found class: {name}")
self.class_cache.add(obj.full_table_name.replace("`", ""))
self.find_subclasses(module, obj)
def load_cache(self, package_name="spyglass"):
package = self.package
for loader, module_name, is_pkg in tqdm(
pkgutil.walk_packages(package.__path__, package.__name__ + "."),
desc="Loading cache",
total=64,
):
module = importlib.import_module(module_name)
if not getattr(module, "schema", None):
continue
print(f"Checking module: {module_name}")
self.find_subclasses(module, module)
my_query = """
select table_schema as database_name, table_name from information_schema.tables
where table_type = 'BASE TABLE'
and table_schema not in('information_schema', 'sys', 'performance_schema', 'mysql')
order by table_schema, table_name;
"""
mysql_tbls = [
".".join(t)
for t in dj.conn().query(my_query).fetchall()
if t[0].split("_")[0] in SHARED_MODULES and t[1][0] != "~"
]
class_cache = SearchPkg().class_cache
no_class = set(mysql_tbls).difference(class_cache)
len_cache = dict()
for t in no_class:
len_cache[t] = len(dj.FreeTable(dj.conn(), t))
non_empty = {k: v for k, v in len_cache.items() if v > 0}
non_empty = {
"common_analytic_signal.__analytic_signal": 1,
"common_analytic_signal.analytic_signal_parameters": 1,
"common_analytic_signal.analytic_signal_selection": 1,
"common_artifact.__artifact_detection": 1,
"common_artifact.artifact_detection_parameters": 3,
"common_artifact.artifact_detection_selection": 1,
"common_artifact.artifact_removed_interval_list": 1,
"common_backup.curated_spike_sorting_back_up": 413,
"common_backup.spike_sorting_back_up": 2810,
"common_filter.fir_filter": 3,
"common_interval.sort_interval": 52,
"common_lab.analysis_nwbfile": 5,
"common_lab.nwbfile": 4,
"common_metrics.metric_parameters": 1,
"common_nwbfile.kachery_channel": 1,
"common_ripple.ripple_artifact_detection_parameters": 2,
"common_ripple.ripple_artifact_removed_interval_list": 2,
"common_sortingview.__sortingview_workspace": 1,
"common_spikesorting.__spike_sorting_recording": 1,
"common_spikesorting.automatic_curation_parameters": 3,
"common_spikesorting.sort_group": 1665,
"common_spikesorting.sort_group__sort_group_electrode": 7427,
"common_spikesorting.spike_sorter": 16,
"common_spikesorting.spike_sorter_parameters": 20,
"common_spikesorting.spike_sorting_artifact_detection_parameters": 1,
"common_spikesorting.spike_sorting_metric_parameters": 6,
"common_spikesorting.spike_sorting_metrics": 3,
"common_spikesorting.spike_sorting_preprocessing_parameters": 2,
"common_spikesorting.spike_sorting_recording_selection": 1,
"common_spikesorting.spike_sorting_selection": 1,
"common_spikesorting.spike_sorting_waveform_parameters": 1,
"common_spikesorting.unit_inclusion_parameters": 2,
"common_temp.temp": 2,
"common_waveforms.waveform_parameters": 1,
"decoding_clusterless.classifier_parameters": 3,
"decoding_clusterless.multiunit_high_synchrony_events_parameters": 1,
"decoding_clusterless.sorted_spikes_classifier_parameters": 4,
"decoding_sortedspikes.#my_sorted_spikes_indicator_selection": 13,
"lfp_v1.__l_f_p": 81,
"lfp_v1.__l_f_p_band": 4,
"lfp_v1.l_f_p_band_selection": 5,
"lfp_v1.l_f_p_band_selection__l_f_p_band_electrode": 76,
"lfp_v1.l_f_p_electrode_group": 65,
"lfp_v1.l_f_p_electrode_group__l_f_p_electrode": 3248,
"lfp_v1.l_f_p_output": 80,
"lfp_v1.l_f_p_output__l_f_p": 80,
"position_dlc_centroid.__d_l_c_centroid": 52,
"position_dlc_centroid.d_l_c_centroid_params": 6,
"position_dlc_centroid.d_l_c_centroid_selection": 60,
"position_dlc_cohort.__d_l_c_smooth_interp_cohort": 52,
"position_dlc_cohort.__d_l_c_smooth_interp_cohort__body_part": 206,
"position_dlc_cohort.d_l_c_smooth_interp_cohort_selection": 52,
"position_dlc_model.__d_l_c_model": 3,
"position_dlc_model.__d_l_c_model__body_part": 15,
"position_dlc_model.d_l_c_model_input": 3,
"position_dlc_model.d_l_c_model_params": 1,
"position_dlc_model.d_l_c_model_selection": 4,
"position_dlc_model.d_l_c_model_source": 5,
"position_dlc_model.d_l_c_model_source__from_import": 3,
"position_dlc_model.d_l_c_model_source__from_upstream": 2,
"position_dlc_orient.__d_l_c_orientation": 52,
"position_dlc_orient.d_l_c_orientation_params": 2,
"position_dlc_orient.d_l_c_orientation_selection": 60,
"position_dlc_pose_estimation.__d_l_c_pose_estimation": 64,
"position_dlc_pose_estimation.__d_l_c_pose_estimation__body_part": 320,
"position_dlc_pose_estimation.d_l_c_pose_estimation_selection": 307,
"position_dlc_position.__d_l_c_smooth_interp": 206,
"position_dlc_position.d_l_c_smooth_interp_params": 5,
"position_dlc_position.d_l_c_smooth_interp_selection": 242,
"position_dlc_project.body_part": 7,
"position_dlc_project.d_l_c_project": 8,
"position_dlc_project.d_l_c_project__body_part": 37,
"position_dlc_project.d_l_c_project__file": 55,
"position_dlc_selection.__d_l_c_pos_v1": 36,
"position_dlc_selection.d_l_c_pos_selection": 52,
"position_dlc_selection.d_l_c_pos_video_params": 10,
"position_dlc_selection.d_l_c_pos_video_selection": 26,
"position_dlc_training.#d_l_c_model_training_params": 2,
"position_dlc_training.__d_l_c_model_training": 2,
"position_dlc_training.d_l_c_model_training_selection": 3,
"position_merge.pose_output": 1,
"position_merge.pose_output__d_l_c_pose_estimation": 1,
"position_position.final_position": 35,
"position_position.final_position__d_l_c_pos_v1": 26,
"waveform_features.#waveform_features_params": 4,
"waveform_features.unit_waveform_features_selection": 327,
}These seem to primarily be first drafts of tables that were later renamed. Unless there are objections, I plan to drop these tables as well. |
|
Would these table's exist on other labs' databases? Just checking if the drop calls should be in the release notes |
|
Only those we removed in #1003. The rest are remnants of bad drafting practices. Ideally, tables with no Spy class would not be declared on shared schemas, which suggests that we should revoke |
7 tasks
|
This task is 95% there. Here are the remaining empty tables that are part of the package empty_tables = [
# PARTS - cannot drop
"lfp_imported._imported_l_f_p",
"lfp_merge.l_f_p_output__imported_l_f_p",
"position_linearization_merge.linearized_position_output__linearized_position_v0",
"spikesorting_group_v1.sorted_spikes_group__sort_group",
# Empty is informative
"common_usage.activity_log",
# See 1025
"common_session.session_group_session",
# Can drop?
"common_position.__position_video",
"position_v1_dlc_model.__d_l_c_model_evaluation",
]Can we deprecate the last 4, @edeno ? This includes ...
|
|
These tables have been dropped in the short term, but a subset that have been removed from the package will be redeclared by anyone with a version that retains the deprecated table Script for deleting empty tablesimport os
import datajoint as dj
import pandas as pd
from datajoint.errors import DataJointError
from datajoint.schemas import VirtualModule
from datajoint.utils import to_camel_case
from networkx import NetworkXError
from pymysql import IntegrityError, OperationalError
from spyglass.utils.database_settings import SHARED_MODULES
class TableData:
def __init__(self, tables=None):
self.processed = False
self.tables = (
list(tables.keys()) if isinstance(tables, dict) else tables
) or self.fetch_empty_tables()
self._by_schemas = None
self._vmods = None
def fetch_empty_tables(self):
query = (
"SELECT table_schema AS database_name, table_name"
+ "FROM information_schema.tables WHERE table_rows < 1;"
)
empty = dj.conn().query(query).fetchall()
in_spy = []
for db, tbl in empty:
if db.split("_")[0] not in SHARED_MODULES or tbl[0] == "~":
continue
full_name = f"{db}.{tbl}"
ft = dj.FreeTable(dj.conn(), full_name)
rows = len(ft)
if len(ft) > 0:
print(f"{rows:03}: {full_name}")
continue
in_spy.append(f"{db}.{tbl}")
return in_spy
@property
def by_schemas(self):
if self._by_schemas is None:
self._by_schemas = {}
for table in self.tables:
schema, table_name = table.split(".")
if schema not in self._by_schemas:
self._by_schemas[schema] = []
self._by_schemas[schema].append(to_camel_case(table_name))
return self._by_schemas
@property
def vmods(self):
if self._vmods is None:
self._vmods = {}
for schema in self.by_schemas:
self._vmods[schema] = VirtualModule("vmod", schema)
return self._vmods
def drop(self, table):
schema, table_name = table.split(".")
class_obj = getattr(self.vmods[schema], to_camel_case(table_name))
getattr(self.vmods[schema], to_camel_case(table_name)).drop()
self.tables.remove(table)
def drop_all(self):
for table in self.tables:
try:
self.drop(table)
except KeyboardInterrupt:
print("Interrupted")
except (AttributeError, DataJointError, NetworkXError) as e:
print(f"Error: {e}")
if __name__ == "__main__":
data = TableData(tbls)
data.drop_all()This could only really be addressed by revoking table declaration privileges for average users, which would result in error messages that could be fixed by deleting the table from their outdated packages. To be discussed as a group |
|
This has been resolved on our production server |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Process
Using a sql command, I fetched tables with no rows, as recorded in the info schema. Some weren't actually empty, so I narrowed down the list by fetching with dj, and then looked for which items weren't represented in the package
Script
Resulting Lists
First pass with an older version of the script above, will update soon
Result
There are ....
common_analytic_signal: last used by Xulu 05/23common_artifact: last used by Alison 11/22common_backup: last used by Kyu 02/22common_curation: last used by Eric 02/22common_metrics: last used by Jen 03/22common_sortingview: last used by Alison 11/22common_waveform: last used by Jen 03/22decoding: last used by Eric 05/23position_position: last used by Daniel 04/23common_behav._head_dir). See 'neither' in list above.spikesorting_curation.unit_inclusion_parameters: declared 07/22common_session.session_group_session: declared 03/22common_ephys.electrode_brain_region: declared 05/22position_merge.position_video_selection: 04/23common_usage.insert_error: declared 02/24 - worth preserving for outside-lab insert errors?Proposed
0.6.0. Inserts into these tables can be overwritten to temporarily launch a deprecation warning and then log incommon_usageto let us know they're being used.Steps
The text was updated successfully, but these errors were encountered: