@uipath/skills 1.199.0 → 1.200.0

package/skills/uipath-platform/references/data-fabric/bulk-import.md

@@ -15,21 +15,25 @@ Response: `{ Code: "RecordsImported", Data: { InsertedRecords, TotalRecords, Err
 - `InsertedRecords` — number of rows successfully imported
 - `TotalRecords` — total rows in the CSV (including failures)
 - `ErrorFileLink` — download URL for a CSV of failed rows (only present if there were failures)
-- `--folder-key` — required when the parent entity is folder-scoped (CLI ≥ `1.197.0`)
+- `--folder-key` — required when the parent entity is folder-scoped
 
 **Pre-flight check (required by Rule 20):**
 
 ```bash
-# Inspect schema before importing — tell the user which columns will be dropped
+# Inspect schema before importing — tell the user which columns will be ignored
 uip df entities get <entity-id> --output json
 # Flag every field whose Fields[].fieldDataType.Name ∈ {CHOICE_SET_SINGLE, CHOICE_SET_MULTIPLE, RELATIONSHIP, FILE, AUTO_NUMBER}
 ```
 
 ## CSV Format Requirements
 
-- **Header row is required** and must exactly match entity field names (case-sensitive)
-- Use `uip df entities get <entity-id> --output json` to discover exact field names before importing
-- System fields (`Id`, `CreatedBy`, `CreateTime`, `UpdatedBy`, `UpdateTime`) must NOT appear in the CSV
+- **Header row is required** and must exactly match each field's `DisplayName`
+  (case-sensitive), not its internal `Name`. For example, a field returned as
+  `{ "Name": "SKU", "DisplayName": "Stock-Keeping Unit" }` requires the CSV
+  header `Stock-Keeping Unit`; `SKU` is rejected as an invalid header.
+- Use `uip df entities get <entity-id> --output json` to discover exact
+  `Fields[].DisplayName` values before importing
+- System fields (`Id`, `CreatedBy`, `CreateTime`, `UpdatedBy`, `UpdateTime`, `RecordOwner`) must NOT appear in the CSV
 
 ### Example CSV
 
@@ -50,32 +54,11 @@ For an entity with fields: `Name` (STRING), `Score` (DECIMAL, `decimalPrecision:
 
 For entities with any complex field, use `records insert --file <json>` instead — the insert endpoint handles every type except `FILE` (which is exclusively written through `files upload`, data-fabric.md Rule 6). See [`records-query.md`](records-query.md#writing-choice-set-and-relationship-values) for the value form.
 
-## Full Workflow
-
-```bash
-# 1. Discover entity and field names
-uip df entities list --output json
-uip df entities get <entity-id> --output json   # note exact field names
-
-# 2. Create CSV with matching headers
-cat > /tmp/data.csv <<EOF
-Name,Score,Active
-Alice,95,true
-Bob,82,true
-EOF
-
-# 3. Import
-uip df records import <entity-id> --file /tmp/data.csv --output json
-
-# 4. Verify
-uip df records list <entity-id> --output json
-```
-
 ## Error Handling
 
 | Error | Cause | Fix |
 |-------|-------|-----|
-| `Import errors in CSV` | Header names don't match field names | Run `entities get` and check exact field names (case-sensitive) |
+| `Import errors in CSV` / `Invalid column header` | Header names don't match field display names | Run `entities get` and use exact `Fields[].DisplayName` values (case-sensitive), not internal `Name` values |
 | `Entity not found` | Wrong entity ID | Run `entities list` to get correct ID |
 | Row-level errors | Invalid data types (e.g. text in number field) | Check data values match field types |
 

package/skills/uipath-platform/references/data-fabric/choice-sets.md

@@ -6,7 +6,7 @@ Reusable picklists that back `CHOICE_SET_SINGLE` and `CHOICE_SET_MULTIPLE` entit
 
 ## Commands
 
-Every command below accepts `--folder-key <GUID>` for folder-scoped choice sets — required on writes against a folder-scoped set, recommended on reads. `list` additionally accepts `--include-folders` (mutually exclusive with `--folder-key`). See [Folder scope](#folder-scope) below for the picker rules. CLI floor: `@uipath/data-fabric-tool@1.197.0+`.
+Every command below accepts `--folder-key <GUID>` for folder-scoped choice sets — required on writes against a folder-scoped set, recommended on reads. `list` additionally accepts `--include-folders` (mutually exclusive with `--folder-key`). See [Folder scope](#folder-scope) below for the picker rules.
 
 | Command | Use |
 |---------|-----|
@@ -41,14 +41,14 @@ Bind a folder-scoped choice set to an entity in a different folder by passing on
 
 ## Value `Name` validation
 
-A choice-set value's `Name` must be alphanumeric and start with a letter. The server rejects C# / VB reserved keywords with *"Choiceset member name must … not be C# keyword"* — but this is a **separate code path from the entity/field-name validator (data-fabric.md Rule 4)**, and the two behave differently:
+A choice-set value's `Name` must start with a letter and then contain letters and digits only, up to 250 characters. Underscores are rejected. Its `DisplayName` is limited to 500 characters. The server rejects C# / VB reserved keywords with *"Choiceset member name must … not be C# keyword"* — but this is a **separate code path from the entity/field-name validator (data-fabric.md Rule 4)**, and the two behave differently:
 
 | Aspect | Entity / field name (Rule 4) | Choice-set value `Name` (here) |
 |---|---|---|
 | Case match | **case-insensitive** (`Class`, `class`, `CLASS` all rejected) | **case-sensitive** (`class` rejected, `Class` may pass — empirically verified: `New` accepted while `new` would be rejected) |
 | Keyword list | full C#/VB reserved list — incl. `Select`, `Return`, `New`, `Internal`, … | partial list — some keywords missing (empirically `select` is NOT rejected as a choice-set value, but `Select` IS rejected as a field name) |
 
-Do not assume a name is legal in one place because it's legal in the other. The safe, portable convention for choice-set value `Name`s: **all-lowercase, snake_case, namespaced to dodge the C# keyword list outright** — `internal_audit`, `new_lead`, `class_a`. Move the human label to `DisplayName`: `Name: "internal_audit"` with `DisplayName: "Internal"`. The dropdown shows "Internal"; the validator sees `internal_audit`. Lowercase tokens that the choice-value validator does reject: `internal`, `public`, `private`, `class`, `case`, `new`, `default`, `static`, `void`, `event`, `lock`, `object`, `string`, `int`.
+Do not assume a name is legal in one place because it's legal in the other. Use a descriptive alphanumeric token that is not itself a language keyword, such as `internalAudit`, `newLead`, or `classOption`. Move the human label to `DisplayName`: `Name: "internalAudit"` with `DisplayName: "Internal"`. Lowercase tokens that the choice-value validator rejects include `internal`, `public`, `private`, `class`, `case`, `new`, `default`, `static`, `void`, `event`, `lock`, `object`, `string`, and `int`.
 
 ## Sourcing `NumberId` after batch value creates
 
@@ -61,67 +61,10 @@ Two rules for any script that batch-creates values:
 
 ## Add a choice-set field to an entity
 
-### Step 1 — Get or create the choice set
-
-**Contract:**
-
-```
-uip df choice-sets create <name> [--display-name "<label>"] [--description "<…>"] --output json
-```
-
-| Arg | Required | Notes |
-|---|---|---|
-| `<name>` | yes | System name. Alphanumeric, starts with a letter, not a C#/VB/SQL reserved keyword. |
-| `--display-name "<label>"` | no | User-facing label in dropdowns. Defaults to `<name>` when omitted. |
-| `--description "<…>"` | no | Free text. |
-
-**Example:**
-
-```bash
-uip df choice-sets list --output json                                                          # check for an existing match first
-uip df choice-sets create ExpenseTypes --display-name "Expense Types" --output json            # create when none matches
-```
-
-### Step 2 — Add each value to the set
-
-**Contract:**
-
-```
-uip df choice-set-values create <choice-set-id> <name> [--display-name "<label>"] --output json
-```
-
-| Arg | Required | Notes |
-|---|---|---|
-| `<choice-set-id>` | yes | UUID from `choice-sets list` / `create`. |
-| `<name>` | yes | System name. Same alphanumeric + no-reserved-keyword rule as `<name>` above (see [Value `Name` validation](#value-name-validation)). |
-| `--display-name "<label>"` | no | User-facing label. Defaults to `<name>` when omitted. |
-
-`NumberId` is assigned 0-based by creation order — order matters. See [Sourcing `NumberId` after batch value creates](#sourcing-numberid-after-batch-value-creates) for the per-value error handling rule.
-
-**Example — `travel` and `meals` on the ExpenseTypes set:**
-
-```bash
-uip df choice-set-values create <choice-set-id> travel --display-name "Travel" --output json
-uip df choice-set-values create <choice-set-id> meals  --display-name "Meals"  --output json
-```
-
-### Step 3 — Bind the choice set to an entity field
-
-```bash
-# New entity
-uip df entities create "Expense" --body '{
-  "fields":[
-    {"fieldName":"amount",   "type":"DECIMAL", "isRequired": true},
-    {"fieldName":"category", "type":"CHOICE_SET_SINGLE",   "choiceSetId":"<choice-set-id>"},
-    {"fieldName":"tags",     "type":"CHOICE_SET_MULTIPLE", "choiceSetId":"<choice-set-id>"}
-  ]
-}' --output json
-
-# Existing entity
-uip df entities update <entity-id> --body '{
-  "addFields":[{"fieldName":"category","type":"CHOICE_SET_SINGLE","choiceSetId":"<choice-set-id>"}]
-}' --output json
-```
+1. List choice sets and let the user reuse one or approve a new set.
+2. For a new set, create it, then create each value separately in the approved order. `choice-set-values create` takes the set ID and value name as positional arguments; it does not accept a batch `--body`.
+3. Re-list values to obtain their assigned `NumberId`s.
+4. Bind the set with `{"fieldName":"<field>","type":"CHOICE_SET_SINGLE","choiceSetId":"<id>"}` (or `CHOICE_SET_MULTIPLE`) in `entities create` or `entities update`.
 
 ## Write / read / filter record values