08. Validation
8.1 Overview
This chapter is the catalogue of validation rules for RUML documents. Every rule has a stable id of
the form <PREFIX>-<NN>; the sections group the rules by subject, and within a section the rules
are sorted by id. Implementations, error messages and test suites SHOULD cite these ids rather than
restate the rule text, so that a violation can be identified independently of language or wording.
A parser MUST reject a document that violates a rule of this catalogue. A rule that says “a parser
MUST accept …” states what the specification permits (MAY): a parser that rejects such a document
violates the rule. Rules of the layer — bind the tool that reads, writes or evaluates a file; they
are not properties of the file.
Validation Layers
- schema: the rule can be checked by validating the document against the RUML JSON Schema alone.
This includes the
formatkeywords of the schema, such asdate-time,uriandregex; a validator that does not assert formats does not check those parts. - semantic: the rule needs logic beyond the schema, such as resolving references, comparing values or checking uniqueness across the document.
- —: the rule depends on the tool, for example on the capabilities it implements or on how it writes a file, and cannot be decided from the document alone.
A validation tool MUST check every rule of the layers schema and semantic (Rule CAP-28CAP-28 A validation tool MUST check every rule of this catalogue whose layer is schema or semantic.).
Rule Ids
A rule id is never reused. A rule that is withdrawn keeps its row, marked Retired, and its id is not given to another rule.
JSON Schema
The JSON Schema is the normative structural definition of the format. It is maintained as a file and is not restated here:
- Interactive explorer: RUML JSON Schema
- Source:
schemas/ruml.schema.json(draft 2020-12); RUML-EXT files are validated byschemas/ruml-ext.schema.json - Packages:
@law-as-code-institute/ruml-schema(npm),ruml-schema(PyPI)
The spec version the schema describes is carried in its x-ruml-spec-version annotation and
mirrored by the $specVersion value every conforming file MUST declare.
8.2 Scope
This catalogue covers the RUML document (.ruml) and the obligations of the tools that read, write
and evaluate it, including the obligations of a parser that processes RUML-EXT files (CAP-22 to
CAP-27). It does not cover:
- the result file
.ruml-result(Chapter 5); - the RUML container, its hashes and signatures (Chapter 7);
- the requirements of Chapter 4.5 on RUML-EXT files other than the parser obligations;
- the contract of a spec-test runner (Appendix A).
For ruml:draft/ and reverse-domain capabilities, the extension points of
Chapter 4.1, this catalogue covers their declaration and use in
the document – id, role and required at the root, where and how often an entry may appear, and
that a node entry has a payload – but not the content of the payload, which is defined by the
parser and evaluator that implement the capability.
8.3 File Format Rules
Rules derived from Chapter 2 and Chapter 3.2.
| Rule | § | Layer | Requirement |
|---|---|---|---|
FMT-01 | 2.1, 3.2.3 | semantic | The file MUST be encoded in UTF-8. |
FMT-02 | 2.1, 2.2, 3.8 | semantic | The file MUST be valid JSON (RFC 8259), and its root MUST be an object. |
FMT-03 | 2.2, 2.6 | schema | $format MUST be "RUML". |
FMT-04 | 2.2, 2.3 | schema | $specVersion MUST be one of the versions listed in the enum of $specVersion in the schema. |
FMT-05 | 2.2 | semantic | $format and $specVersion MUST be the first two fields of the root object. |
FMT-06 | 2.2 | schema | The root object MUST have metadata and rulemap; a parser MUST accept it with or without capabilities. |
FMT-07 | 2.2 | schema | The root object MUST NOT have fields other than $format, $specVersion, metadata, capabilities and rulemap. |
FMT-08 | 2.2, 2.5, 2.6, 3.4.1, 3.5.1 | schema | rulemap.nodes MUST contain an entry with key "1", the root node. |
FMT-09 | 2.3, 2.6 | — | A parser MUST accept a file whose $specVersion has its own MAJOR version and a MINOR version up to its own, and MUST reject a file with a higher MINOR version or a different MAJOR version. |
FMT-10 | 2.5, 2.6, 3.5.2 | schema | Every UUID MUST be a lowercase UUID version 4 (pattern in 2.5). |
FMT-11 | 2.5, 3.1, 3.5.3 | schema | Every key of rulemap.nodes and every entry of children MUST be a LexID, a string matching ^[1-9][0-9]*$. |
FMT-12 | 2.5, 2.6 | semantic | A LexID MUST NOT occur more than once as a key of rulemap.nodes. |
FMT-13 | 2.6, 3.2.3, 3.8 | schema | Every field value MUST have the data type 3.2.3 gives it: a Number a JSON number, a Boolean a JSON Boolean, a DateTime an RFC 3339 date-time, an Enum one of the values defined for the field, an Object a JSON object, an Array a JSON array. |
FMT-14 | 3.2.6 | schema | A DataPoint, text entry or context entry MUST NOT have an id field; its key is its id. |
8.4 Metadata Rules
Rules derived from Chapter 3.3.
| Rule | § | Layer | Requirement |
|---|---|---|---|
META-01 | 3.3.1, 3.8 | schema | metadata MUST have id, version, created, modified and status. |
META-02 | 3.3.1, 3.3.3 | schema | status MUST be one of the seven values of 3.3.3. |
META-03 | 3.3.1 | semantic | modified MUST NOT be earlier than created. |
META-04 | 3.3.2, 3.3.5 | schema | A parser MUST accept each optional field of 3.3.2 and derivedFrom.originalHash, present or absent, including replaces and replacedBy together. |
META-05 | 3.3.2, 3.3.6 | schema | documentType MUST be one of the values of 3.3.6. |
META-06 | 3.3.2 | schema | authors MUST have at least one entry, and every entry MUST have name. |
META-07 | 3.3.2 | semantic | jurisdiction MUST begin with an ISO 3166-1 alpha-2 country code; a parser MUST accept hyphen-separated subdivisions after it. |
META-08 | 3.3.2 | semantic | language MUST be an ISO 639-1 code. |
META-09 | 3.3.2 | semantic | When both are set, validUntil MUST be later than validFrom. |
META-10 | 3.3.3 | schema | Status inForce MUST have validFrom. |
META-11 | 3.3.3 | semantic | Status expired MUST have validUntil, and it MUST lie in the past. |
META-12 | 3.3.3 | schema | Status replaced MUST have replacedBy. |
META-13 | 3.3.5 | schema | derivedFrom MUST have originalId, originalStatus, derivationType and derivedAt. |
META-14 | 3.3.5 | semantic | A working copy (derivedFrom.derivationType: "workingCopy") MUST NOT have status inForce. |
8.5 Rulemap and Node Rules
Rules derived from Chapters 3.1, 3.4 and 3.5.
| Rule | § | Layer | Requirement |
|---|---|---|---|
NODE-01 | 2.2, 3.2.6, 3.5.2 | schema | A node MUST have id and label. |
NODE-02 | 3.1, 3.4.1 | schema | rulemap MUST have nodes, an object whose values are node objects. |
NODE-03 | 3.1, 3.2.6, 3.5.1, 3.5.2 | schema | A node MUST NOT have fields other than those of 3.5; in particular it MUST NOT store its LexID (lexId), its parent or its type. |
NODE-04 | 3.1 | — | A tool that writes a file MUST keep the order of every children array. |
NODE-05 | 3.5.1 | semantic | The root node "1" MUST NOT appear in any children array. |
NODE-06 | 3.5.1 | schema | children MUST NOT be empty. |
NODE-07 | 3.5.2a, 3.5.3, 3.6, 3.6.4, 3.6.5 | schema | A parser MUST accept the optional node fields of 3.5.3, present or absent, except operator on a leaf node: a leaf node MUST NOT have operator unless metadata.status is draft; in a draft, a parser MUST accept any of the eight operators of 3.6 on a leaf node. |
NODE-08 | 3.5.2, 3.8 | semantic | The id of a node MUST NOT equal the id of another node or metadata.id. |
NODE-09 | 3.5.2 | schema | label MUST NOT be empty. |
NODE-10 | 3.5.3, 3.8 | semantic | Every entry of children MUST be a key of rulemap.nodes. |
NODE-11 | 3.5.3 | schema | description MUST NOT exceed 10,000 characters. |
NODE-12 | 3.5.2a | schema | A node with children MUST have operator. |
8.6 Operator Rules
Rules derived from Chapter 3.6.
| Rule | § | Layer | Requirement |
|---|---|---|---|
OP-01 | 3.5.2a | schema | operator MUST be one of the eight values of 3.6. |
OP-02 | 3.6.4, 4.1, 4.3, 4.14 | semantic | Outside a draft, a node with operator: "DELEGATED" MUST have children. A node with operator: "DELEGATED" and children MUST have exactly one entry of a capability with the combiner role; for a ruml:draft/ or reverse-domain capability, the role is the one declared at the root. In a draft, a parser MUST accept that operator on a leaf node without a combiner (NODE-07). |
OP-03 | 3.6.4 | — | An evaluator MUST NOT short-circuit a node with operator: "DELEGATED"; it evaluates every child before the combiner runs. |
OP-04 | 3.6.4, 4.14 | semantic | The node-level payload of a combiner MUST have a non-empty prose description. |
OP-05 | 3.6.4, 4.14 | semantic | A parser MUST accept a combiner that uses non-deterministic input, including a ruml:expression that calls $random, $now, $millis or $eval. |
OP-06 | 3.6.4 | semantic | A parser MUST accept any capability declared at the root with the combiner role, including a reverse-domain one, as the combiner of a DELEGATED node. |
OP-07 | 3.6.5 | schema | operator: "undefined" MUST NOT occur unless metadata.status is draft; in a draft, a parser MUST accept it. |
8.7 Capability Rules
Rules derived from Chapter 4, including the parser obligations for
RUML-EXT files (4.5). For those, the schema is
schemas/ruml-ext.schema.json.
| Rule | § | Layer | Requirement |
|---|---|---|---|
CAP-01 | 1.4, 4.1, 4.2 | semantic | A parser MUST reject a file that declares a capability with required: true that the parser does not implement, and MUST NOT process such a file partially. |
CAP-02 | 4.1, 4.2 | semantic | A parser MUST accept a file that declares a capability it does not implement with required: false or without required, and MUST ignore the payload and the node-level entries of that capability. |
CAP-03 | 4.1, 4.2, 4.4 | schema | The root entry of a ruml:draft/ or reverse-domain capability MUST declare role. |
CAP-04 | 4.1, 4.4 | semantic | A parser that implements a ruml:draft/ or reverse-domain capability MUST reject the file if the declared role differs from the role it implements. |
CAP-05 | 4.1 | schema | The root entry of an official ruml: capability MUST NOT declare role. |
CAP-06 | 4.1, 4.3 | semantic | On one node, a capability with the data role MUST NOT have more than one entry; a parser MUST accept several entries of a capability with the decision or reaction role. |
CAP-07 | 4.1, 4.3, 4.8, 4.16 | semantic | An entry of a capability with the decision role MUST NOT appear on a node with children. |
CAP-08 | 3.6.4, 4.1, 4.14 | semantic | An entry of a capability with the combiner role MUST NOT appear on a node whose operator is not DELEGATED. |
CAP-09 | 4.2 | schema | Every entry of the root capabilities MUST have id; a parser MUST accept required and payload in it. |
CAP-10 | 4.2 | semantic | An id MUST NOT occur more than once in the root capabilities. |
CAP-11 | 4.2, 4.14 | schema | The root entry of ruml:expression MUST have required: true. |
CAP-12 | 4.2 | semantic | The data of a root entry’s payload MUST be an object keyed by id; a parser MUST accept a payload with data, config or both where the capability provides for them. |
CAP-13 | 4.2 | — | A parser that implements a capability MUST validate its payload.data and payload.config against that capability’s specification. |
CAP-14 | 4.2 | schema | When the root capabilities is absent or empty, a node MUST NOT have capabilities. |
CAP-15 | 3.5.3, 4.2, 4.3, 4.8, 4.9, 4.15 | semantic | The id of every entry of a node’s capabilities MUST be declared in the root capabilities. |
CAP-16 | 4.3, 4.14 | schema | A parser MUST accept capabilities on any node, and entries of capabilities with the data or reaction role on any node, in any order, also beside the combiner of a DELEGATED node. |
CAP-17 | 3.5.3, 4.3 | schema | Every entry of a node’s capabilities MUST have id and payload, and payload MUST be an object. |
CAP-18 | 4.3, 4.13 | schema | ruml:protocol MUST NOT appear in a node’s capabilities, neither inline nor in a RUML-EXT file. |
CAP-19 | 4.3 | schema | A parser MUST accept decisionMode on a node with children and MUST ignore it there. |
CAP-20 | 4.3 | — | A tool that writes a file MUST NOT reorder the entries of a node’s capabilities. |
CAP-21 | 4.4 | schema | A capability id MUST be a ruml: id defined by this specification version, a ruml:draft/<name> id or a <reverse-domain>/<name> id; it MUST consist of lowercase letters, digits, _, - and the delimiters, and MUST NOT exceed 128 characters. |
CAP-22 | 4.5 | schema | The $format of a RUML-EXT file MUST be "RUML-EXT"; a parser MUST check it before it processes the file further. |
CAP-23 | 4.5 | schema | The root of a RUML-EXT file MUST NOT contain metadata or rulemap. |
CAP-24 | 2.6, 4.5 | semantic | Before a merge, the $base.id of a RUML-EXT file MUST equal the metadata.id of the base file, and its $base.hash the hash of the base file’s canonical form; otherwise the parser MUST abort the merge. Each RUML-EXT file is checked on its own. |
CAP-25 | 4.5 | — | A parser MUST NOT skip the $base comparison, not even in a development environment or on an explicit user override. |
CAP-26 | 2.6, 4.5 | semantic | Every node UUID in the nodes of a RUML-EXT file MUST be the id of a node of the base file. |
CAP-27 | 4.5 | semantic | The same capability MUST NOT be defined for the same node both inline and in a RUML-EXT file, nor in two RUML-EXT files of the same base file; a parser MUST accept several RUML-EXT files for one base file. |
CAP-28 | 4.6 | — | A validation tool MUST check every rule of this catalogue whose layer is schema or semantic. |
CAP-29 | 4.8 | semantic | A node’s ruml:query payload MUST have expression, a valid JSONata expression; a parser MUST accept description in it. |
CAP-30 | 4.9 | schema | A node’s ruml:ai-query payload MUST have model; a parser MUST accept prompt, description, inputs and config in it. |
CAP-31 | 4.9 | schema | A parser MUST accept context, dataPoints and texts in the inputs of a ruml:ai-query payload; an object entry of context MUST have id, and a parser MUST accept role (reference or subject) in it. |
CAP-32 | 4.9 | schema | A parser MUST accept deterministic (a Boolean), effort (low, medium or high) and temperature (a number from 0 to 1) in the config of a node’s ruml:ai-query payload, and systemPrompt in the config of its root payload. |
CAP-33 | 4.9 | schema | A parser MUST accept a leaf node that has both ruml:query and ruml:ai-query. |
CAP-34 | 4.10 | schema | A parser MUST accept ruml:actions on any node, also as its only capability. |
CAP-35 | 4.10 | schema | A node’s ruml:actions payload MUST have at least one of onTrue, onFalse and onUndefined; an action MUST have operation and ref, and a parser MUST accept description in the payload, config in an action and any string as operation. |
CAP-36 | 4.13 | schema | A parser MUST accept config.level with the value standard or verbose in the root payload of ruml:protocol. |
CAP-37 | 4.14 | schema | ruml:expression MUST NOT appear in a RUML-EXT file. |
CAP-38 | 4.14 | semantic | A node’s ruml:expression payload MUST have a non-empty expression, a valid JSONata expression. |
8.8 DataPoint Rules
Rules derived from Chapter 4.7.
| Rule | § | Layer | Requirement |
|---|---|---|---|
DP-01 | 4.7 | schema | A DataPoint MUST have name, type and binding, and type MUST be one of the data types of 4.7.1. |
DP-02 | 1.2, 4.7, 4.7.4 | schema | The key of a DataPoint, the id of a set and the keys of a set’s fields MUST match the pattern [a-z_][a-z0-9_]*. |
DP-03 | 4.7 | schema | A parser MUST accept description, validation and sensitive on any DataPoint, unit for number and percentage, format for location, identifier and address, and cardinality for object. |
DP-04 | 4.7, 4.7.3 | schema | A DataPoint MUST NOT have unit, format, options, set or cardinality, nor a validation rule, that 4.7 and 4.7.3 do not provide for its type. |
DP-05 | 4.7, 4.7.1 | schema | A DataPoint with type: "currency" MUST have unit. |
DP-06 | 4.7, 4.7.1 | schema | A DataPoint with type: "enum" MUST have a non-empty options, and every option MUST have value. |
DP-07 | 4.7 | schema | A DataPoint with type: "object" MUST have set. |
DP-08 | 4.7, 4.7.3 | schema | cardinality.min MUST be an integer of at least 0 and cardinality.max an integer of at least 1; minLength and maxLength MUST be integers from 0 to 253 − 1. |
DP-09 | 4.7.2 | schema | A parser MUST accept hint, placeholder and defaultValue in an input binding; description, sourceType and fallback in an external binding; and description in a computed binding. |
DP-10 | 4.7.2 | schema | An external binding MUST have ref. |
DP-11 | 4.7.2 | semantic | A computed binding MUST have expression, a valid JSONata expression; a parser MUST accept one that reads other DataPoints, including computed ones. |
DP-12 | 4.7.2 | semantic | The references between computed DataPoints MUST NOT form a cycle, directly or transitively. |
DP-13 | 4.7.3 | schema | validation.pattern MUST be a valid ECMA-262 regular expression. |
DP-14 | 4.7.3 | schema | validation.min and validation.max MUST have the type 4.7.3 gives for the data type. |
DP-15 | 4.7.4 | schema | An entry of sets MUST have name and fields; a parser MUST accept description in it and a DataPoint of type: "object" among its fields. |
DP-16 | 4.7.4 | semantic | The set of a DataPoint MUST be the id of an entry of sets. |
DP-17 | 4.7.4 | semantic | A set MUST NOT reference itself, directly or transitively. |
DP-18 | 4.8, 4.9, 4.14 | semantic | Every DataPoint id that the expression of a ruml:query or of a ruml:expression reads, and every id in inputs.dataPoints of a ruml:ai-query, MUST be a key of the data of ruml:datapoints. |
8.9 Text Reference Rules
Rules derived from Chapter 4.12.
| Rule | § | Layer | Requirement |
|---|---|---|---|
TXT-01 | 4.12 | schema | The payload of a node’s ruml:text-reference entry MUST have refs with at least one anchor. An anchor MUST have id and type (represents or refers-to) and MUST NOT have fields other than these and element (a lowercase UUID), which a parser MUST accept; a node MUST NOT have more than one represents anchor, and the same anchor MUST NOT appear twice. |
TXT-02 | 4.12 | schema | A text key MUST match the pattern [a-z_][a-z0-9_]*. |
TXT-03 | 4.12 | — | A text key MUST NOT change once anchors may point at it, and the GUID of an element MUST NOT change; a tool that imports a new version of a text MUST carry the GUID of every element it matches forward. |
TXT-04 | 4.12 | schema | A text entry MUST have title and timestamp, and status unless it has content. |
TXT-05 | 4.12 | semantic | In a text entry, timestamp and validFrom MUST be RFC 3339 date-times, url a URI and language an ISO 639-1 code. |
TXT-06 | 4.12 | schema | A parser MUST accept a text entry with any of the optional fields type, text, content, version, validFrom, validUntil, language and isReferenced. |
TXT-07 | 4.12 | schema | A text entry MUST NOT have both text and content; an embedded entry with status inForce MUST have a non-empty text or a content. |
TXT-08 | 4.12 | schema | A text entry with content MUST have digest, sha3-256: followed by 64 lowercase hexadecimal digits. |
TXT-09 | 4.12 | semantic | digest MUST equal the SHA3-256 hash of the UTF-8 bytes of the RFC 8785 form of content. |
TXT-10 | 4.12 | schema | A text entry with type url, container or external MUST have url, containerPath or ref respectively. |
TXT-11 | 4.12 | schema | The containerPath of a text entry MUST start with .texts/. |
TXT-12 | 4.12 | semantic | When a text entry has both, validUntil MUST be later than validFrom. |
TXT-13 | 4.12 | semantic | In a text entry with content, status, validFrom, validUntil, language and version MUST agree with the metadata of the text itself. |
TXT-14 | 4.12 | semantic | In content, a content array MUST NOT contain two adjacent strings. |
TXT-15 | 4.12 | schema | The root of content MUST have type, attrs and content, and MUST declare the namespace of its dialect as its default namespace (attrs.xmlns). |
TXT-16 | 4.12 | semantic | In content, every element with an eId MUST have a GUID, a lowercase UUID, unique within the entry; an element other than the root MUST NOT declare xmlns; every namespace prefix MUST be declared by an xmlns:<prefix> attribute on the element or an ancestor; and an element MUST NOT have xsi:schemaLocation. |
TXT-17 | 4.12 | — | A tool that writes RUML Text MUST keep the elements and attributes of every namespace, including those it does not understand, and MUST NOT add sentence markers that the source does not contain. |
TXT-18 | 4.12 | semantic | Converted to XML, content MUST be valid against the schemas of the dialect its root namespace names; a fragment MUST be valid where it stands in a complete document. |
TXT-19 | 4.9, 4.12 | semantic | Every id of an anchor in the refs of a node’s ruml:text-reference entry, and every id in inputs.texts of a ruml:ai-query, MUST be a key of the data of ruml:text-reference. |
TXT-20 | 4.12 | semantic | The element of an anchor MUST be the GUID of an element in the content of the entry its id names; an anchor into an entry without content MUST NOT have element. |
8.10 Context Rules
Rules derived from Chapter 4.11.
| Rule | § | Layer | Requirement |
|---|---|---|---|
CTX-01 | 4.11 | schema | A context entry MUST have title, category and timestamp. |
CTX-02 | 4.11 | schema | category MUST be one of the categories of 4.11.1. |
CTX-03 | 4.11, 4.11.2 | schema | In a context entry, timestamp MUST be an RFC 3339 date-time, accessDate an RFC 3339 date and url a URI. |
CTX-04 | 4.11 | schema | A context key MUST match the pattern [a-z_][a-z0-9_]*. |
CTX-05 | 4.11 | schema | The same context id MUST NOT appear more than once in the refs of a node’s ruml:context entry. |
CTX-06 | 4.11 | schema | A parser MUST accept a context entry with any of the optional fields type, subject, metadata, content, source, hash and accessDate, the last two with any type. |
CTX-07 | 4.11, 4.11.2 | schema | A context entry with type url, container or external MUST have url, containerPath or ref respectively. |
CTX-08 | 4.11 | schema | hash MUST be sha3-256, sha256 or blake3, then :, then 64 lowercase hexadecimal digits. |
CTX-09 | 4.11.2 | schema | The containerPath of a context entry MUST start with .context/. |
CTX-10 | 4.9, 4.11 | semantic | Every id in the refs of a node’s ruml:context entry, and every context id in inputs.context of a ruml:ai-query, MUST be a key of the data of ruml:context. |
8.11 Modular Node Rules
Rules derived from Chapter 6.
| Rule | § | Layer | Requirement |
|---|---|---|---|
MOD-01 | 3.5.3, 6.2, 6.4 | semantic | A node whose LexID appears more than once in the children arrays of the rulemap (a self-reference counts) MUST have modular: true. |
MOD-02 | 6.4 | semantic | A parser MUST accept a modular node that lists its own LexID in its children (recursion). |
MOD-03 | 6.4 | — | A processor MUST warn about a recursive node whose operator is AND, NOT_AND, OR or NOT_OR; the warning alone does not make the file invalid. |
MOD-04 | 6.4 | semantic | A self-reference MUST NOT be the first entry of children. |
MOD-05 | 6.4 | semantic | A recursive node MUST have at least one child that is not a self-reference. |
MOD-06 | 6.5 | semantic | A parser MUST accept a modular node among the children of another modular node, also when it is referenced elsewhere. |
8.12 Tree Reference Rules
Rules derived from Chapters 4.15 and 4.16.
| Rule | § | Layer | Requirement |
|---|---|---|---|
TREE-01 | 4.15, 4.16 | schema | The root entries of ruml:tree-reference and ruml:tree-include MUST NOT have a payload. A node’s ruml:tree-reference payload MUST contain represents, refersTo (with at least one entry) or both, and represents and every refersTo entry MUST have exactly uri; a node’s ruml:tree-include payload MUST contain uri. Both payloads MUST NOT contain other fields. |
TREE-02 | 4.15, 4.16 | schema | Every uri of a tree reference or inclusion MUST be a node anchor <ruml-version-uri>#node/<id>; a same-document anchor (#node/…) is not one. |
TREE-03 | 4.15 | schema | The same uri MUST NOT appear twice in refersTo. |
TREE-04 | 4.15, 4.16 | semantic | The target node of a node anchor MUST exist in the referenced version. |
TREE-05 | 4.15 | semantic | A parser MUST accept ruml:tree-reference entries that form cycles across files. |
TREE-06 | 4.15, 4.16 | semantic | A node anchor MUST name an immutable RUML version; a file that is exported or exchanged MUST NOT contain an anchor without a version. All node anchors of a file into the same other RUML file MUST use the same version of it. |
TREE-07 | 4.15 | schema | A parser MUST accept a node that has ruml:text-reference anchors besides a ruml:tree-reference entry. |
TREE-08 | 4.16 | schema | A node with ruml:tree-include MUST NOT have represents in its ruml:tree-reference. |
TREE-09 | 4.16 | semantic | Inclusions MUST NOT form a cycle across files, directly or through further files. |