Formatting a CSV for Constraint Import

To bulk import constraints, the CSV file should be formatted as described in this article.

General Column/Data Requirements

The following column headers must exist and contain data in each row:

• Constraint_Name or Constraint_GUID
• Constraint_Type

The following columns are optional (for example, the Package columns are only there to specify destination if desired):

• Constraint_Description
• Package_Name
• Package_GUID
• LB
• UB
• LB_Inclusive
• UB_Inclusive
• Regex

The order of the columns is not enforced, it does not have to be in the same order as the suggested format above.

Any additional user-made columns will be ignored and will not interfere with import.

Bulk import can currently import Package and Constraint elements.

Column-specific Requirements

Constraint Name, Constraint_GUID

• Either a Constraint Name or Constraint GUID is required. If a Constraint GUID is given, it must be associated with an existing constraint in the model.

Constraint Type

• Must be one of the following values: Integer, String, Real.

Package Name, Package GUID

• Leaving this column blank will default Package to the "Phenom" package.
• Setting this column to an existing Package will default to the first matching Conceptual Package within the model.
• If there is no matching Package, a new Package will be created within the "Phenom" Package.
• If a Package GUID is given, it must reference an existing UoP package in the model.

LB, UB

• Denote the lower and upper bounds for a constraint.
• Must be a number.

LB Inclusive, UB Inclusive

• Must be a boolean: true, false.

Regex

• Does not need to be valid regex.

Additional Requirements

• Data within Package_Name and Constraint_Name must all begin with a letter, and only contain numbers, letters, or underscores. The data must also not match any of the following reserved words:
  • IDL Reserved words:

    "and", "begin", "break", "case", "common", "compile_opt", "continue", "do", "else", "end", "endcase", "endelse", "endfor", "endforeach", "endif", "endrep", "endswitch", "endwhile", "eq", "for", "foreach", "forward_function", "function", "ge", "goto", "gt", "if", "inherits", "le", "lt", "mod", "ne", "not", "of", "on_ioerror", "or", "pro", "repeat", "switch", "then", "until", "while", "xor"

  • OCL Reserved words:

    "abstract", "any", "attribute", "boolean", "case", "char", "component", "const", "consumes", "context", "custom", "default", "double", "emits", "enum", "eventtype", "exception", "factory", "false", "finder", "fixed", "float", "getraises", "home", "import", "in", "inout", "interface", "local", "long", "manages", "module", "multiple", "native", "object", "octet", "oneway", "out", "primarykey", "private", "provides", "public", "publishes", "raises", "readonly", "sequence", "setraises", "short", "string", "struct", "supports", "switch", "true", "truncatable", "typedef", "typeid", "typeprefix", "union", "unsigned", "uses", "valuebase", "valuetype", "void", "wchar", "wstring"

• If there are any errors during validation of the CSV file, the model import will not be initiated and a list of errors will be returned. They will be labeled by the row they occur on for convenience.
• There is a 2MB limit in place - if a larger file needs to be imported, it needs to be split up into smaller files.
• In the raw CSV (i.e., as observed from notepad), any field with commas must have surrounding quotes (Excel normally handles this).