Checking configuration and input data
Overview
You should always check your configuration and input prior to creating, updating, or deleting content. You can do this by running Workbench with the --check
option, e.g.:
./workbench --config config.yml --check
Note
If you're on Windows, you will likely need to run Workbench by explicitly invoking Python, e.g. python workbench --config config.yml --check
instead of using ./workbench
as illustrated above. Similarly, if you are on a Mac that has the Homebrew version of Python, you may need to run Workbench by providing the full path to the Homebrew Python interpreter, e.g., /opt/homebrew/bin/python3 workbench --config config.yml --check
.
If you do this, Workbench will check the following conditions and report any errors that require your attention before proceeding:
- Configuration file
- Whether your configuration file is valid YAML (i.e., no YAML syntax errors).
- Whether your configuration file contains all required values.
- Connection to Drupal
- Whether your Drupal has the required Workbench Integration module enabled, and that the module is up to date.
- Whether the
host
you provided will accept theusername
andpassword
you provided.
- Input directory
- Whether the directory named in the
input_dir
configuration setting exists.
- Whether the directory named in the
- Rollback files
- Whether the rollback config file and CSV file can be written.
- CSV file
- Whether the CSV file is encoded in either ASCII or UTF-8.
- Whether each row contains the same number of columns as there are column headers.
- Whether there are any duplicate column headers.
- Whether your CSV file contains required columns headers, including the field defined as the unique ID for each record (defaults to "id" if the
id_field
key is not in your config file) - Whether your CSV column headers correspond to existing Drupal field labels or machine names.
- Whether all Drupal fields that are configured to be required are present in the CSV file.
- Whether required fields in your CSV contain values (i.e., they are not blank).
- Whether the columns required to create paged content are present (see "Creating paged content" below).
- If creating compound/paged content using the "With page/child-level metadata" method,
--check
will tell you if any child item rows in your CSV precede their parent rows. - If your config file includes
csv_headers: labels
,--check
will tell you if it detects any duplicate field labels.
- Media files
- Whether the files named in the CSV file are present, or in the case of remote files, are accessible (but this check is skipped if
allow_missing_files: true
is present in your config file for "create" tasks). Ifnodes_only
is true, this check is skipped. - Whether files in the
file
CSV column have extensions that are registered with the media's file field in Drupal. Note that validation of file extensions does not yet apply to files named using theadditional_files
configuration or for remote files (see this issue for more info). - Whether the media types configured for specific file extensions are configured on the target Drupal. Islandora Workbench will default to the 'file' media type if it can't find another more specific media type for a file, so the most likely cause for this check to fail is that the assigned media type does not exist on the target Drupal.
- If creating media track files,
--check
will tell you if your media_use_tid value (either in the media_use_tid configuration setting or in row-level values in your CSV) does not include the "Service File" taxonomy term.
- Whether the files named in the CSV file are present, or in the case of remote files, are accessible (but this check is skipped if
- Field values
- Base fields
- If the
langcode
field is present in your CSV, whether values in it are valid Drupal language codes. - Whether your CSV file contains a
title
field (create
task only) - Whether values in the
title
field exceed Drupal's maximum length for titles of 255 characters, or whatever the value of themax_node_title_length
configuration setting is. - If the
created
field is present in your CSV file, whether the values in it are formatted correctly (like "2020-11-15T23:49:22+00:00") and whether the date is in the past (both of which are Drupal requirements). - If the
uid
field is present in your CSV file, whether the user IDs in that field exist in the target Drupal. Note that this check does not inspect permissions or roles, only that the user ID exists.
- If the
- Whether aliases in the
url_alias
field in your CSV already exist, and whether they start with a leading slash (/
). - Taxonomy
- Whether term ID and term URIs used in CSV fields correspond to existing terms.
- Whether the length of new terms exceeds 255 characters, which is the maximum length for a term name.
- Whether the term ID (or term URI) provided for
media_use_tid
is a member of the "Islandora Media Use" vocabulary. - Whether term names in your CSV require a vocabulary namespace.
- Typed Relation fields
- Whether values used in typed relation fields are in the required format
- Whether values need to be namespaced
- Whether the term IDs/term names/term URIs used in the values exist in the vocabularies configured for the field.
- "List" text fields
- Whether values in CSV fields of this Drupal field type are in the field's configured "Allowed values list".
- If using the pages from directories configuration (
paged_content_from_directories: true
):- Whether page filenames contain an occurrence of the sequence separator.
- Whether any page directories are empty.
- Whether the content type identified in the
content_type
configuration option exists. - Whether multivalued fields exceed their allowed number of values.
- Whether values in text-type fields exceed their configured maximum length.
- Whether the nodes referenced in
field_member_of
(if that field is present in the CSV) exist. - Whether values used in geolocation fields are valid lat,long coordinates.
- Whether values used in EDTF fields are valid EDTF date/time values (subset of date/time values only; see documentation for more detail). Also validates whether dates are valid Gregorian calendar dates.
- Base fields
- Hook scripts
- Whether registered bootstrap, preprocessor, and post-action scripts exist and are executable.
If Workbench detects a configuration or input data violation, it will either stop and tell you why it stopped, or (if the violation will not cause Workbench's interaction with Drupal to fail), tell you that it found an anomaly and to check the log file for more detail.
Note
Adding perform_soft_checks: true
to you configuration file will tell --check
to not stop when it encounters an error with 1) parent/child order in your input CSV, 2) file extensions in your CSV that are not registered with the Drupal configuration of their media type's file fields, or 3) invalid EDTF dates. Workbench will continue to log all of these errors, but will exit after it has checked every row in your CSV file.
A successful outcome of running --check
confirms that all of the conditions listed above are in place, but it does not guarantee a successful job. There are a lot of factors in play during ingest/update/delete interactions with Drupal that can't be checked in advance, most notably network stability, load on the Drupal server, or failure of an Islandora microservice. But in general --check
will tell you if there's a problem that you can investigate and resolve before proceeding with your task.
Typical (and recommended) Islandora Workbench usage
You will probably need to run Workbench using --check
a few times before you will be ready to run it without --check
and commit your data to Islandora. For example, you may need to correct errors in taxonomy term IDs or names, fix errors in media filenames, or wrap values in your CSV files in quotation marks.
It's also a good idea to check the Workbench log file after running --check
. All warnings and errors are printed to the console, but the log file may contain additional information or detail that will help you resolve issues.
Once you have used --check
to detect all of the problems with your CSV data, committing it to Islandora will work very reliably.
Also, it is good practice to check your log after each time you run Islandora Workbench, since it may contain information that is not printed to the console.`
Prompting the user to run --check
As decribed elsewhere, you can configure Workbench to prompt the user to remind them to run --check
. To do so, include remind_user_to_run_check: true
in your config file. If this setting is present, the user will be prompted "Have you run --check? (y/n)". Responding "y" resumes normal operation, "n" (or any other response) causes Workbench to exit. Note that this setting does not force the user to run --check
, it merely ask them if they have run it.
Requiring --check
You can require a successful --check
to have to have been run by including check_lock_file_path
with the name (or path) of a file as its value, for example check_lock_file_path: checklock.txt
. If this setting is present in your config file, and it has a file name or path as its value, when Workbench is run without --check
using the same configuration file, it will look for the "lock" file. It compares the data in this lock file with an expected value, and if they are the same, Workbench executes normally. If they differ, Workbench logs this difference and exits.
If --check
has detected any errors, Workbench will not execute.
This can be useful if you want to force the user to run check, or if you are running Workbench in a scheduled or scripted environment and you want to only execute Workbench if --check
has been successful.