Naming Scheme

Description

A Naming Scheme describes how free-form names should be turned into consistent identifiers. Typical uses include cleaning field names that arrive with spaces (for example Order ID → order_id), normalizing file or folder names, and applying a shared project convention (snake_case, camelCase, PascalCase, and so on).

Naming Schemes are managed as project metadata. Apply one without adding buttons to every dialog:

  • N indicator — widgets that accept a naming scheme show a small N on the right (same diamond plate as $ for variables and # for expanded integers). Click N to open the matching scheme in a dialog. If several schemes of that type exist, pick one first. If none exist, Hop creates a new scheme of that type and opens it.

  • Keyboard shortcut — CTRL-SHIFT-N on a name field (transform/action name, field cell, table name, file name, …) lists schemes of the matching type and replaces the current identifier. When the field does not accept variables, CTRL-SPACE does the same thing.

  • TableView toolbar — Apply a Naming Scheme to a column still rewrites a whole column at once (with undo).

Naming Scheme N indicator

If only one scheme of the required type exists, apply and open both skip the picker.

A General scheme is offered on every name field that does not have a scheme of its own type. Create one General scheme (for example lower-case + underscore) and it covers transform names, field names, tables, and files. Add a type-specific scheme only when that kind of name needs different rules; when such a scheme exists, General is no longer listed for that type.

Get Fields

CSV File Input and Text File Input have an optional Naming scheme field. When you click Get Fields, discovered names are rewritten with that scheme.

If the field is empty, Get Fields still auto-applies when exactly one matching scheme exists (a type-specific hop-field scheme, or a General scheme). The same automatic apply runs after Get Fields in other dialogs that fill annotated name columns (Excel, Parquet, Select Values, and so on).

Transform, action, pipeline, workflow and metadata name fields use the shortcut with variable insertion disabled — they stay identifiers, not '${…​}' expressions. They still show N, not $.

Annotated filename and folder widgets (run configurations, execution information locations, and so on) show N automatically. Filename widgets still accept variables (CTRL-SPACE); CTRL-SHIFT-N rewrites the last path segment only.

Variable names (project described variables, pipeline and workflow parameters, Set Variables) use the Hop variable names type. Values that already contain '${…​}' are left unchanged.

Options

Naming Scheme editor

The screenshot above shows a scheme named lowercase-dashes: lower case words joined with dashes, so Order ID previews as order-id.

Option Description

Name

Unique metadata name of this scheme.

Description

Optional free-text description.

Type

Target kind. General applies to every name field unless a scheme of the widget’s specific type exists (Hop field, transform, action, pipeline, workflow, metadata, variable, database table, database column, file, or folder). File schemes rewrite only the last path segment and keep the parent folder and the file extension (/data/Order ID.csv → /data/order_id.csv). Folder schemes rewrite the last segment and keep the parent path (and a trailing slash).

Case style

As is, Lower case, Upper case, camelCase, or PascalCase.

Word separator

How words are joined when not using camel/Pascal case: none, underscore (_), dash (-), or space.

Capitalize first word

When enabled, uppercase the first character of the first word. Typical for transform and action names with a space separator (Table input).

Extra delimiters

Additional characters treated as word boundaries (beyond spaces, _, -, and camelCase edges).

Remove special characters

When enabled, strip non letter/digit characters from each word.

Collapse repeated separators

When enabled, collapse repeated separators such as __ to a single separator.

Trim leading/trailing separators

When enabled, remove separators at the start or end of the result.

Prefix / Suffix

Optional strings prepended or appended after the transform.

Preview input / output

Live sample showing how the current options transform a name.

Examples

Input Case style Word separator Result

Order ID

Lower case

Underscore (_)

order_id

Order ID

As is

Underscore (_)

Order_ID

Order ID

camelCase

_

orderId

Order ID

PascalCase

_

OrderId

order-id

Upper case

Underscore (_)

ORDER_ID

Order ID

Lower case

Dash (-)

order-id

read_customers

Lower case

Space

read customers

read_customers (capitalize first word)

Lower case

Space

Read customers

/data/Order ID.csv (file type)

Lower case

Underscore (_)

/data/order_id.csv

/data/My Folder/ (folder type)

Lower case

Underscore (_)

/data/my_folder/

Applying from a name field

The screenshot below shows an action name field with the N indicator. Hover it for the shortcut reminder; click it to open the scheme in a dialog.

Action name field with Naming Scheme indicator and tooltip

  1. Create at least one Naming Scheme in the Metadata perspective (pick the type that matches the field).

  2. Focus a name widget — transform name, action name, a field-name cell, a table name, a file or folder path, a variable or parameter name, …

  3. Press CTRL-SHIFT-N (or CTRL-SPACE on name fields that do not accept variables).

  4. If several schemes of that type exist, pick one. The list shows a live preview (Order ID → order_id).

  5. The current identifier is replaced. Values that already contain ${…​} are left unchanged.

To review or change the rules instead of applying them, click the N indicator. That opens the scheme in a dialog (or a picker when more than one scheme matches the field type). It does not open another tab in the metadata perspective. If no scheme of that type (and no General scheme) exists yet, Hop creates one of the matching type and opens it for you to save.

The same shortcut works on F2 inline rename in the Metadata and File Explorer trees (metadata names, file names, and folder names). There is no N indicator on those tiny tree editors — press CTRL-SHIFT-N after F2.

Applying from TableView

  1. Open a transform or dialog that shows a field list (or any other editable table).

  2. Ensure table toolbars are visible (Hop Gui options).

  3. Click the Apply a Naming Scheme to a column toolbar icon (label / tag icon on the TableView toolbar).

  4. Select the column to rename if the table has more than one name column.

  5. Select a Naming Scheme of the matching type.

  6. Confirm — values in non-empty rows are updated. Use the table Undo action if needed.

CTRL-SHIFT-N on a focused cell applies the scheme to that cell only. The toolbar is the bulk action.

Example: CSV File Input fields

The following example uses the lowercase-dashes scheme on the Name column of a CSV File Input field list.

Before — mixed styles (ID, FirstName, birth_date, stateCode, …):

CSV File Input fields before applying a Naming Scheme

After — consistent lower-case names with dashes (id, first-name, birth-date, state-code, …):

CSV File Input fields after applying the lowercase-dashes Naming Scheme

Types as plugins

The Type combo lists every registered naming-scheme kind. Built-in kinds include General, Hop field/transform/action/pipeline/workflow/metadata/variable, database table/column, file, and folder.

Other plugins can add kinds without changing Hop. For example hop-data-vault can register dv-hub, dv-satellite, dv-link, dm-dimension, dm-fact, and so on. Create one Naming Scheme metadata object per kind (or one General scheme). Use one scheme per type in a project so CI has a single expected spelling.

Checking names in CI

A name is valid when applying the matching scheme leaves the string unchanged. Empty values, <null>, and values that contain ${…​} are skipped.

The same check runs in Hop Gui Verify this pipeline / Verify this workflow (errors only) and from the command line:

hop naming-check --project my-project
hop naming-check --path /path/to/project --format json --require-scheme
hop naming-check --type hop-field,hop-transform

  • Exit 0 when there are no errors.

  • Exit 1 when at least one name would be rewritten (or when --require-scheme is set and a type has no scheme).

  • Exit 2 on unexpected failure (file load issues are warnings).

  • --format json prints one object per finding for GitHub Actions or other reporters.

Example GitHub Actions step after Hop is on PATH and the project is checked out:

- name: Check naming schemes
  run: hop naming-check --project my-project --require-scheme

Create one General scheme (or one scheme per type) in the project’s metadata. There is no --fix flag: rewriting committed pipelines is a separate, explicit action.

Related

  • GitHub issue #2683 — field name cleanup / naming toolbar