Implementation plan: Catalog Data Cleaning (#3848)

Co-authored-by: Olga Bulat <obulat@gmail.com>
Co-authored-by: Madison Swain-Bowden <bowdenm@spu.edu>

documentation/projects/proposals/data_normalization/20240227-implementation_plan_catalog_data_cleaning.md

@@ -0,0 +1,220 @@
+# 2024-02-27 Implementation Plan: Catalog Data Cleaning
+
+**Author**: @krysal
+
+## Reviewers
+
+- [x] @obulat
+- [x] @AetherUnbound
+
+## Project links
+
+<!-- Enumerate any references to other documents/pages, including milestones and other plans -->
+
+- [Project Thread](https://github.com/WordPress/openverse/issues/430)
+
+This project does not have a project proposal because the scope and rationale of
+the project are clear, as defined in the project thread. In doubt, check the
+[Expected Outcomes](#expected-outcomes) section below.
+
+## Overview
+
+This document describes a mechanism for rectifying incorrect data in the catalog
+database (DB) that currently has to be cleaned up every time a data refresh is
+run. This one-time fix is an effort to avoid wasting resources and data refresh
+runtime.
+
+## Background
+
+One of the steps of the [data refresh process for images][img-data-refresh] is
+cleaning the data that is not fit for production. This process is triggered
+weekly by an Airflow DAG, which then runs in the Ingestion Server, taking
+approximately just over **20 hours** to complete, according to a inspection of
+recent executions as of the time of drafting this document. The cleaned data is
+only saved to the API database, which is replaced each time during the same data
+refresh, meaning this process has to be repeated each time to make the _same_
+corrections.
+
+This cleaning process was designed this way to optimize writes to the API
+database, since the most important factor was to provide the correct data to
+users via the API. Most of the rows affected were added prior to the creation of
+the `MediaStore` class in the Catalog (possibly by the discontinued CommonCrawl
+ingestion) which is nowadays responsible for validating the provider data prior
+to upserting the records into the upstream database. However, the current
+approach entails a problem of wasting resources both in time, which continues to
+increase, and in the machines (CPU) it uses, which could easily be avoided
+making the changes permanent by saving them in the upstream database.
+
+[img-data-refresh]: ./../../../catalog/reference/DAGs.md#image_data_refresh
+
+## Expected Outcomes
+
+<!-- List any succinct expected products from this implementation plan. -->
+
+- The catalog database (upstream) contains the cleaned data outputs of the
+  current Ingestion Server's cleaning steps
+- The image Data Refresh process is simplified by reducing significantly
+  cleaning times.
+
+<!-- removing the cleaning steps from the Ingestion Server. -->
+
+## Step-by-step plan
+
+The cleaning functions that the Ingestion Server applies (see the
+[cleanup][ing_server_cleanup] file) are already implemented in the Catalog in
+the `MediaStore` class: see its [`_tag_blacklisted` method][tag_blacklisted]
+(which probably should be renamed) and the [url utilities][url_utils] file. The
+only part that it's not there and can't be ported is the filtering of
+low-confidence tags, since provider scripts don't save an "accuracy" by tag.
+
+With this the plan then starts in the Ingestion Server with the following steps:
+
+1. [Save TSV files of cleaned data to AWS S3](#save-tsv-files-of-cleaned-data-to-aws-s3)
+1. [Make and run a batched update DAG for one-time cleanup](#make-and-run-a-batched-update-dag-for-one-time-cleanups)
+1. [Run an image Data Refresh to confirm cleaning time is reduced](#run-an-image-data-refresh-to-confirm-cleaning-time-is-reduced)
+
+[ing_server_cleanup]:
+  https://github.com/WordPress/openverse/blob/f8971fdbea36fe0eaf5b7d022b56e4edfc03bebd/ingestion_server/ingestion_server/cleanup.py#L79-L168
+[tag_blacklisted]:
+  https://github.com/WordPress/openverse/blob/f8971fdbea36fe0eaf5b7d022b56e4edfc03bebd/catalog/dags/common/storage/media.py#L245-L259
+[url_utils]:
+  https://github.com/WordPress/openverse/blob/a930ee0f1f116bac77cf56d1fb0923989613df6d/catalog/dags/common/urls.py
+
+## Step details
+
+### Save TSV files of cleaned data to AWS S3
+
+In a previous exploration, the Ingestion Server was set to [store TSV files of
+the cleaned data][pr-saving-tsv] in the form of `<identifier> <cleaned_field>`,
+which can be used later to perform the updates efficiently in the catalog DB,
+which only had indexes for the `identifier` field. These files are saved to the
+disk of the Ingestion Server EC2 instances, and worked fine for files with URL
+corrections since this type of fields is relatively short, but became a problem
+when trying to save tags, as the file turned too large and filled up the disk,
+causing issues to the data refresh execution.
+
+[aws_mpu]:
+  https://docs.aws.amazon.com/AmazonS3/latest/userguide/mpuoverview.html
+
+To have some numbers of the problem we are dealing with, the following table
+shows the number of records cleaned by field for last runs at the moment of
+writing this IP, except for tags, which we don't have accurate registries since
+file saving was disabled.
+
+| timestamp (UTC)     | 'url' | 'creator_url' | 'foreign_landing_url' | 'tags' |
+| ------------------- | :---: | :-----------: | :-------------------: | :----: |
+| 2024-02-27 04:05:26 | 22156 |    9035458    |        8809213        |   0    |
+| 2024-02-20 04:06:56 | 22157 |    9035456    |        8809209        |   0    |
+| 2024-02-13 04:41:22 | 22155 |    9035451    |        8809204        |   0    |
+
+The alternative is to upload TSV files to the Amazon Simple Storage Service
+(S3), creating a new bucket or using a subfolder within `openverse-catalog`. The
+benefit of using S3 buckets is that they have streaming capabilities and will
+allow us to read the files in chunks later if necessary for performance. The
+downside is that objects in S3 don't allow appending natviely, so it may require
+to upload files with different part numbers or evaluate if the [multipart upload
+process][aws_mpu] or more easily, the [`smart_open`][smart_open] package could
+serve us here.
+
+[smart_open]: https://github.com/piskvorky/smart_open
+
+### Make and run a batched update DAG for one-time cleanups
+
+A batched catalog cleaner DAG (or potentially a `batched_update_from_file`)
+should take the files of the previous step to perform a batched update on the
+catalog's image table, while handling deadlocking and timeout concerns, similar
+to the [batched_update][batched_update]. This table is constantly in use by
+other DAGs, such as those from providers ingestion or the data refresh process,
+and ideally can't be singly blocked by any DAG.
+
+[batched_update]: ./../../../catalog/reference/DAGs.md#batched_update
+
+A [proof of concept PR](https://github.com/WordPress/openverse/pull/3601)
+consisted of uploading each file to temporary `UNLOGGED` DB tables (which
+provides huge gains in writing performance while their disadventages are not
+relevant to us, they won't be permanent), and include a `row_id` serial number
+used later to query it in batches. The following must be included:
+
+- Add an index for the `identifier` column in the temporary table after filling
+  it up, to improve the query performance
+- An adaptation to handle the column type of tags (`jsonb`) and modify the
+  `metadata`
+- Include a DAG task for reporting the number of rows affected by column to
+  Slack
+
+### Run an image data refresh to confirm cleaning time is reduced
+
+Finally, after the previous steps are done, running a data refresh will confirm
+there are no more updates applied at ingestion. If time isn't significantly
+reduced then it will be necessary to check what was missing in the previous
+steps. Looking at files generated in the
+[step 1](#save-tsv-files-of-cleaned-data-to-aws-s3) may yield clues.
+
+If confirmed the time is reduced to zero, optionally the cleaning steps can be
+removed, or leave them in case we want to perform a similar cleaning effort
+later, e.g. see the [Other projects or work](#other-projects-or-work) section.
+
+## Dependencies
+
+### Infrastructure
+
+No changes needed. The Ingestion Server already has the credentials required to
+[connect with AWS](https://github.com/WordPress/openverse/blob/a930ee0f1f116bac77cf56d1fb0923989613df6d/ingestion_server/ingestion_server/indexer_worker.py#L23-L28).
+
+### Tools & packages
+
+<!-- Describe any tools or packages which this work might be dependent on. If multiple options are available, try to list as many as are reasonable with your own recommendation. -->
+
+Requires installing and familiarizing with the [smart_open][smart_open] utility.
+
+### Other projects or work
+
+Once the steps have been completed and proved the method works we could make
+additional similar corrections following the same procedure. Some potentially
+related issues are:
+
+- [Some images have duplicate incorrectly decoded unicode tags #1303](https://github.com/WordPress/openverse/issues/1303)
+- [Provider scripts may include html tags in record titles #1441](https://github.com/WordPress/openverse/issues/1441)
+- [Fix Wikimedia image titles #1728](https://github.com/WordPress/openverse/issues/1728)
+- [Add filetype to all images in the catalog DB #1560](https://github.com/WordPress/openverse/issues/1560),
+  for when the file type can be derived from the URL.
+
+This will also open up space for more structural changes to the Openverse DB
+schemas in a [second phase](https://github.com/WordPress/openverse/issues/244)
+of the Data Normalization endeavor.
+
+## Alternatives
+
+A previous proposal was to use the `ImageStore` to re-evaluate every image in
+the catalog DB. While this could theoretically be performed in a batched way
+too, and presented the advantage of future validations to be easily incorpored
+in a single place, it also came with significant shortcomings and complexities.
+The class would have to adapt to validations for images ingested by the
+CommonCrawl process, for which it was not planned and could open a can of extra
+problems. It would also have to go through the entire database to update the bad
+rows, unless a mix of both proposal is implemented, but ultimately the process
+of this IP is considered more direct and simple for the goal.
+
+## Rollback
+
+<!-- How do we roll back this solution in the event of failure? Are there any steps that can not easily be rolled back? -->
+
+In the rare case we need the old data back, we can resort to DB backups, which
+are performed [weekly][db_snapshots].
+
+[db_snapshots]: ./../../../catalog/reference/DAGs.md#rotate_db_snapshots
+
+<!--
+## Risks
+
+What risks are we taking with this solution? Are there risks that once taken can’t be undone?-->
+
+## Prior art
+
+- Previous attempt from cc-archive: [Clean preexisting data using ImageStore
+  #517][mathemancer_pr]
+- @obulat's PR to [add logging and save cleaned up data in the Ingestion
+  Server][pr-saving-tsv]
+
+[pr-saving-tsv]: https://github.com/WordPress/openverse/pull/904
+[mathemancer_pr]: https://github.com/cc-archive/cccatalog/pull/517

documentation/projects/proposals/data_normalization/index.md

@@ -0,0 +1,8 @@
+# Data Normalization
+
+```{toctree}
+:titlesonly:
+:glob:
+
+*
+```