From 3c8ec9af68d593e7e5eef52aca3d001e22c8e0f7 Mon Sep 17 00:00:00 2001 From: Frederick Muriuki Muriithi Date: Mon, 9 Sep 2024 16:39:50 -0500 Subject: Enable samples uploads. --- uploader/samples/models.py | 85 ++++++++++++++++++++++ uploader/samples/views.py | 173 ++++++++++++++++----------------------------- 2 files changed, 145 insertions(+), 113 deletions(-) (limited to 'uploader/samples') diff --git a/uploader/samples/models.py b/uploader/samples/models.py index 15e509e..d7d5384 100644 --- a/uploader/samples/models.py +++ b/uploader/samples/models.py @@ -1,7 +1,12 @@ """Functions for handling samples.""" +import csv +from typing import Iterator + import MySQLdb as mdb from MySQLdb.cursors import DictCursor +from functional_tools import take + def samples_by_species_and_population( conn: mdb.Connection, species_id: int, @@ -17,3 +22,83 @@ def samples_by_species_and_population( "AND iset.InbredSetId=%(population_id)s", {"species_id": species_id, "population_id": population_id}) return tuple(cursor.fetchall()) + + +def read_samples_file(filepath, separator: str, firstlineheading: bool, **kwargs) -> Iterator[dict]: + """Read the samples file.""" + with open(filepath, "r", encoding="utf-8") as inputfile: + reader = csv.DictReader( + inputfile, + fieldnames=( + None if firstlineheading + else ("Name", "Name2", "Symbol", "Alias")), + delimiter=separator, + quotechar=kwargs.get("quotechar", '"')) + for row in reader: + yield row + + +def save_samples_data(conn: mdb.Connection, + speciesid: int, + file_data: Iterator[dict]): + """Save the samples to DB.""" + data = ({**row, "SpeciesId": speciesid} for row in file_data) + total = 0 + with conn.cursor() as cursor: + while True: + batch = take(data, 5000) + if len(batch) == 0: + break + cursor.executemany( + "INSERT INTO Strain(Name, Name2, SpeciesId, Symbol, Alias) " + "VALUES(" + " %(Name)s, %(Name2)s, %(SpeciesId)s, %(Symbol)s, %(Alias)s" + ") ON DUPLICATE KEY UPDATE Name=Name", + batch) + total += len(batch) + print(f"\tSaved {total} samples total so far.") + + +def cross_reference_samples(conn: mdb.Connection, + species_id: int, + population_id: int, + strain_names: Iterator[str]): + """Link samples to their population.""" + with conn.cursor(cursorclass=DictCursor) as cursor: + cursor.execute( + "SELECT MAX(OrderId) AS loid FROM StrainXRef WHERE InbredSetId=%s", + (population_id,)) + last_order_id = (cursor.fetchone()["loid"] or 10) + total = 0 + while True: + batch = take(strain_names, 5000) + if len(batch) == 0: + break + params_str = ", ".join(["%s"] * len(batch)) + ## This query is slow -- investigate. + cursor.execute( + "SELECT s.Id FROM Strain AS s LEFT JOIN StrainXRef AS sx " + "ON s.Id = sx.StrainId WHERE s.SpeciesId=%s AND s.Name IN " + f"({params_str}) AND sx.StrainId IS NULL", + (species_id,) + tuple(batch)) + strain_ids = (sid["Id"] for sid in cursor.fetchall()) + params = tuple({ + "pop_id": population_id, + "strain_id": strain_id, + "order_id": last_order_id + (order_id * 10), + "mapping": "N", + "pedigree": None + } for order_id, strain_id in enumerate(strain_ids, start=1)) + cursor.executemany( + "INSERT INTO StrainXRef( " + " InbredSetId, StrainId, OrderId, Used_for_mapping, PedigreeStatus" + ")" + "VALUES (" + " %(pop_id)s, %(strain_id)s, %(order_id)s, %(mapping)s, " + " %(pedigree)s" + ")", + params) + last_order_id += (len(params) * 10) + total += len(batch) + print(f"\t{total} total samples cross-referenced to the population " + "so far.") diff --git a/uploader/samples/views.py b/uploader/samples/views.py index 6af90f4..6e3dc4b 100644 --- a/uploader/samples/views.py +++ b/uploader/samples/views.py @@ -1,7 +1,6 @@ """Code regarding samples""" import os import sys -import csv import uuid from pathlib import Path from typing import Iterator @@ -18,8 +17,6 @@ from flask import ( render_template, current_app as app) -from functional_tools import take - from uploader import jobs from uploader.files import save_file from uploader.datautils import order_by_family @@ -34,7 +31,8 @@ from uploader.species.models import (all_species, order_species_by_family) from uploader.population.models import(save_population, population_by_id, - populations_by_species) + populations_by_species, + population_by_species_and_id) from .models import samples_by_species_and_population @@ -123,86 +121,6 @@ def list_samples(species_id: int, population_id: int): activelink="list-samples") -def read_samples_file(filepath, separator: str, firstlineheading: bool, **kwargs) -> Iterator[dict]: - """Read the samples file.""" - with open(filepath, "r", encoding="utf-8") as inputfile: - reader = csv.DictReader( - inputfile, - fieldnames=( - None if firstlineheading - else ("Name", "Name2", "Symbol", "Alias")), - delimiter=separator, - quotechar=kwargs.get("quotechar", '"')) - for row in reader: - yield row - - -def save_samples_data(conn: mdb.Connection, - speciesid: int, - file_data: Iterator[dict]): - """Save the samples to DB.""" - data = ({**row, "SpeciesId": speciesid} for row in file_data) - total = 0 - with conn.cursor() as cursor: - while True: - batch = take(data, 5000) - if len(batch) == 0: - break - cursor.executemany( - "INSERT INTO Strain(Name, Name2, SpeciesId, Symbol, Alias) " - "VALUES(" - " %(Name)s, %(Name2)s, %(SpeciesId)s, %(Symbol)s, %(Alias)s" - ") ON DUPLICATE KEY UPDATE Name=Name", - batch) - total += len(batch) - print(f"\tSaved {total} samples total so far.") - - -def cross_reference_samples(conn: mdb.Connection, - species_id: int, - population_id: int, - strain_names: Iterator[str]): - """Link samples to their population.""" - with conn.cursor(cursorclass=DictCursor) as cursor: - cursor.execute( - "SELECT MAX(OrderId) AS loid FROM StrainXRef WHERE InbredSetId=%s", - (population_id,)) - last_order_id = (cursor.fetchone()["loid"] or 10) - total = 0 - while True: - batch = take(strain_names, 5000) - if len(batch) == 0: - break - params_str = ", ".join(["%s"] * len(batch)) - ## This query is slow -- investigate. - cursor.execute( - "SELECT s.Id FROM Strain AS s LEFT JOIN StrainXRef AS sx " - "ON s.Id = sx.StrainId WHERE s.SpeciesId=%s AND s.Name IN " - f"({params_str}) AND sx.StrainId IS NULL", - (species_id,) + tuple(batch)) - strain_ids = (sid["Id"] for sid in cursor.fetchall()) - params = tuple({ - "pop_id": population_id, - "strain_id": strain_id, - "order_id": last_order_id + (order_id * 10), - "mapping": "N", - "pedigree": None - } for order_id, strain_id in enumerate(strain_ids, start=1)) - cursor.executemany( - "INSERT INTO StrainXRef( " - " InbredSetId, StrainId, OrderId, Used_for_mapping, PedigreeStatus" - ")" - "VALUES (" - " %(pop_id)s, %(strain_id)s, %(order_id)s, %(mapping)s, " - " %(pedigree)s" - ")", - params) - last_order_id += (len(params) * 10) - total += len(batch) - print(f"\t{total} total samples cross-referenced to the population " - "so far.") - - def build_sample_upload_job(# pylint: disable=[too-many-arguments] speciesid: int, populationid: int, @@ -219,14 +137,15 @@ def build_sample_upload_job(# pylint: disable=[too-many-arguments] ] + (["--firstlineheading"] if firstlineheading else []) -@samplesbp.route("/upload/species//populations//samples", +@samplesbp.route("/populations//upload-samples", methods=["GET", "POST"]) @require_login def upload_samples(species_id: int, population_id: int):#pylint: disable=[too-many-return-statements] """Upload the samples.""" - samples_uploads_page = redirect(url_for("samples.upload_samples", - species_id=species_id, - population_id=population_id)) + samples_uploads_page = redirect(url_for( + "species.populations.samples.upload_samples", + species_id=species_id, + population_id=population_id)) if not is_integer_input(species_id): flash("You did not provide a valid species. Please select one to " "continue.", @@ -241,14 +160,14 @@ def upload_samples(species_id: int, population_id: int):#pylint: disable=[too-ma flash("You did not provide a valid population. Please select one " "to continue.", "alert-danger") - return redirect(url_for("samples.select_population", + return redirect(url_for("species.populations.samples.select_population", species_id=species_id), code=307) population = with_db_connection( lambda conn: population_by_id(conn, int(population_id))) if not bool(population): flash("Invalid grouping/population!", "alert-error") - return redirect(url_for("samples.select_population", + return redirect(url_for("species.populations.samples.select_population", species_id=species_id), code=307) @@ -296,34 +215,62 @@ def upload_samples(species_id: int, population_id: int):#pylint: disable=[too-ma redisuri, f"{app.config['UPLOAD_FOLDER']}/job_errors") return redirect(url_for( - "samples.upload_status", job_id=the_job["jobid"])) + "species.populations.samples.upload_status", + species_id=species_id, + population_id=population_id, + job_id=the_job["jobid"])) + -@samplesbp.route("/upload/status/", methods=["GET"]) -def upload_status(job_id: uuid.UUID): +@samplesbp.route("/populations//" + "upload-samples/status/", + methods=["GET"]) +def upload_status(species_id: int, population_id: int, job_id: uuid.UUID): """Check on the status of a samples upload job.""" - job = with_redis_connection(lambda rconn: jobs.job( - rconn, jobs.jobsnamespace(), job_id)) - if job: - status = job["status"] - if status == "success": - return render_template("samples/upload-success.html", job=job) - - if status == "error": - return redirect(url_for("samples.upload_failure", job_id=job_id)) - - error_filename = Path(jobs.error_filename( - job_id, f"{app.config['UPLOAD_FOLDER']}/job_errors")) - if error_filename.exists(): - stat = os.stat(error_filename) - if stat.st_size > 0: + with database_connection(app.config["SQL_URI"]) as conn: + species = species_by_id(conn, species_id) + if not bool(species): + flash("You must provide a valid species.", "alert-danger") + return redirect(url_for("species.populations.samples.index")) + + population = population_by_species_and_id( + conn, species_id, population_id) + if not bool(population): + flash("You must provide a valid population.", "alert-danger") + return redirect(url_for( + "species.populations.samples.select_population", + species_id=species_id)) + + job = with_redis_connection(lambda rconn: jobs.job( + rconn, jobs.jobsnamespace(), job_id)) + if job: + status = job["status"] + if status == "success": + return render_template("samples/upload-success.html", + job=job, + species=species, + population=population,) + + if status == "error": return redirect(url_for( - "samples.upload_failure", job_id=job_id)) + "species.populations.samples.upload_failure", job_id=job_id)) - return render_template( - "samples/upload-progress.html", - job=job) # maybe also handle this? + error_filename = Path(jobs.error_filename( + job_id, f"{app.config['UPLOAD_FOLDER']}/job_errors")) + if error_filename.exists(): + stat = os.stat(error_filename) + if stat.st_size > 0: + return redirect(url_for( + "samples.upload_failure", job_id=job_id)) - return render_template("no_such_job.html", job_id=job_id), 400 + return render_template("samples/upload-progress.html", + species=species, + population=population, + job=job) # maybe also handle this? + + return render_template("no_such_job.html", + job_id=job_id, + species=species, + population=population), 400 @samplesbp.route("/upload/failure/", methods=["GET"]) def upload_failure(job_id: uuid.UUID): -- cgit v1.2.3