Skip to Content
Last modified by Helena on 2025/09/10 11:19
From version 2.1
edited by Helena
on 2025/05/15 15:48
Change comment: Created by office importer.
To version 3.2
edited by Helena
on 2025/05/16 10:57
Change comment: There is no comment for this version

Summary

Details

Page properties
Title
... ... @@ -1,0 +1,1 @@
1 +14 ANNEX Semantic Versioning
Content
... ... @@ -1,4 +1,6 @@
1 -= 14 ANNEX Semantic Versioning =
1 +{{box title="**Contents**"}}
2 +{{toc/}}
3 +{{/box}}
2 2  
3 3  == 14.1 Introduction to Semantic Versioning ==
4 4  
... ... @@ -6,9 +6,9 @@
6 6  
7 7  In systems with many dependencies, releasing new artefact versions can quickly become a nightmare. If the dependency specifications are too tight, you are in danger of version lock (the inability to upgrade an artefact without having to release new versions of every dependent artefact). If dependencies are specified too loosely, you will inevitably be bitten by version promiscuity (assuming compatibility with more future versions than is reasonable). Dependency hell is where you are when version lock and/or version promiscuity prevent you from easily and safely moving your data modelling forward.
8 8  
9 -As a very successful solution to the similar problem in software development, "Semantic Versioning" [[__semver.org__>>url:http://semver.org/]][[ >>url:http://semver.org/]]proposes a simple set of rules and requirements that dictate how version numbers are assigned and incremented. These rules make also perfect sense in the world of versioned data modelling and help to solve the "dependency hell" encountered with previous versions of SDMX. SDMX 3.0 applies thus the Semantic Versioning rules on all versioned SDMX artefacts. Once you release a versioned SDMX artefact, you communicate changes to it with specific increments to your version number.
11 +As a very successful solution to the similar problem in software development, "Semantic Versioning" [[__semver.org__>>https://semver.org/]] proposes a simple set of rules and requirements that dictate how version numbers are assigned and incremented. These rules make also perfect sense in the world of versioned data modelling and help to solve the "dependency hell" encountered with previous versions of SDMX. SDMX 3.0 applies thus the Semantic Versioning rules on all versioned SDMX artefacts. Once you release a versioned SDMX artefact, you communicate changes to it with specific increments to your version number.
10 10  
11 -**This SDMX 3.0(.0) specification inherits the original **[[__**semver.org**__>>url:https://semver.org/]]**[[ >>url:https://semver.org/]]2.0.0 wording (license: __[[Creative Commons >>url:http://creativecommons.org/licenses/by/3.0/]][[- >>url:http://creativecommons.org/licenses/by/3.0/]][[CC BY 3.0>>url:http://creativecommons.org/licenses/by/3.0/]]__[[)>>url:http://creativecommons.org/licenses/by/3.0/]] and applies it to versioned SDMX structural artefacts.** Under this scheme, version numbers and the way they change convey meaning about the underlying data structures and what has been modified from one version to the next.
13 +**This SDMX 3.0(.0) specification inherits the original **[[__**semver.org**__>>https://xwiki:semver.org]]** 2.0.0 wording (license: [[__Creative Commons - CC BY 3.0__>>https://creativecommons.org/licenses/by/3.0/]]) and applies it to versioned SDMX structural artefacts.** Under this scheme, version numbers and the way they change convey meaning about the underlying data structures and what has been modified from one version to the next.
12 12  
13 13  == 14.2 Semantic Versioning Specification for SDMX 3.0(.0) ==
14 14  
... ... @@ -43,78 +43,7 @@
43 43  
44 44  == 14.3 Backus–Naur Form Grammar for Valid SDMX 3.0(.0) Semantic Versions ==
45 45  
46 -|(((
47 -**<valid semver> ::= <version core>**
48 48  
49 -**| <version core> "-" <extension>**
50 -
51 -**<version core> ::= <major> "." <minor> "." <patch>**
52 -
53 -**<major> ::= <numeric identifier>**
54 -
55 -**<minor> ::= <numeric identifier>**
56 -
57 -**<patch> ::= <numeric identifier>**
58 -
59 -**<extension> ::= <dot-separated extension identifiers>**
60 -
61 -**<dot-separated extension identifiers> ::= <extension identifier>**
62 -
63 -**| <extension identifier> "." <dotseparated extension identifiers>**
64 -
65 -**<extension identifier> ::= <alphanumeric identifier>**
66 -
67 -**| <numeric identifier>**
68 -
69 -**<alphanumeric identifier> ::= <non-digit>**
70 -
71 -**| <non-digit> <identifier characters>**
72 -
73 -**| <identifier characters> <non-digit>**
74 -
75 -**| <identifier characters> <non-digit> <identifier characters>**
76 -
77 -**<numeric identifier> ::= "0"**
78 -
79 -**| <positive digit>**
80 -
81 -**| <positive digit> <digits>**
82 -
83 -**<identifier characters> ::= <identifier character>**
84 -
85 -**| <identifier character> <identifier characters>**
86 -)))
87 -
88 -**<identifier character> ::= <digit>**
89 -
90 -**| <non-digit>**
91 -
92 -**<non-digit> ::= <letter>**
93 -
94 -**| "-"**
95 -
96 -**<digits> ::= <digit>**
97 -
98 -**| <digit> <digits>**
99 -
100 -**<digit> ::= "0"**
101 -
102 -**| <positive digit>**
103 -
104 -**<positive digit> ::= "1" | "2" | "3" | "4" | "5" | "6" | "7" | "8" | "9"**
105 -
106 -**<letter> ::= "A" | "B" | "C" | "D" | "E" | "F" | "G" | "H" | "I" | "J"**
107 -
108 -**| "K" | "L" | "M" | "N" | "O" | "P" | "Q" | "R" | "S" | "T"**
109 -
110 -**| "U" | "V" | "W" | "X" | "Y" | "Z" | "a" | "b" | "c" | "d"**
111 -
112 -**| "e" | "f" | "g" | "h" | "i" | "j" | "k" | "l" | "m" | "n"**
113 -
114 -**| "o" | "p" | "q" | "r" | "s" | "t" | "u" | "v" | "w" | "x"**
115 -
116 -**| "y" | "z"**
117 -
118 118  == 14.4 Dependency Management in SDMX 3.0(.0): ==
119 119  
120 120  MAJOR, MINOR or PATCH version parts in SDMX 3.0 artefact references CAN be wildcarded using "+" as extension:
... ... @@ -226,24 +226,16 @@
226 226  
227 227  One with named groups for those systems that support them (PCRE [Perl Compatible Regular Expressions, i.e. Perl, PHP and R], Python and Go).
228 228  
229 -Reduced version (without original SemVer "build metadata") from: [[__https:~~/~~/regex101.com/r/Ly7O1x/3/__>>url:https://regex101.com/r/Ly7O1x/3/]][[url:https://regex101.com/r/Ly7O1x/3/]]
160 +Reduced version (without original SemVer "build metadata") from: [[__https:~~/~~/regex101.com/r/Ly7O1x/3/__https:~~/~~/regex101.com/r/Ly7O1x/3/>>https://https:regex101.comrLy7O1x3https:regex101.comrLy7O1x3]]
230 230  
231 -^(?P<major>0|[1-9]\d*)\.(?P<minor>0|[1-9]\d*)\.(?P<patch>0|[1-
162 +^(?P<major>0|[1-9]\d*)\.(?P<minor>0|[1-9]\d*)\.(?P<patch>0|[1-9]\d*)(?:-(?P<extension>(?:0|[1-9]\d*|\d*[a-zA-Z-][0-9a-zA-Z-]*)(?:\.(?:0|[1-9]\d*|\d*[a-zA-Z-][0-9a-zA-Z-]*))*))?$
232 232  
233 -9]\d*)(?:-(?P<extension>(?:0|[1-9]\d*|\d*[a-zA-Z-][0-9a-zA-Z-
234 -
235 -]*)(?:\.(?:0|[1-9]\d*|\d*[a-zA-Z-][0-9a-zA-Z-]*))*))?$
236 -
237 237  And one with numbered capture groups instead (so cg1 = major, cg2 = minor, cg3 = patch and cg4 = extension) that is compatible with ECMA Script (JavaScript), PCRE (Perl Compatible Regular Expressions, i.e. Perl, PHP and R), Python and Go.
238 238  
239 -Reduced version (without original SemVer "build metadata") from: [[__https:~~/~~/regex101.com/r/vkijKf/1/__>>url:https://regex101.com/r/vkijKf/1/]][[url:https://regex101.com/r/vkijKf/1/]]
166 +Reduced version (without original SemVer "build metadata") from: [[__https:~~/~~/regex101.com/r/vkijKf/1/__https:~~/~~/regex101.com/r/vkijKf/1/>>https://https:regex101.comrvkijKf1https:regex101.comrvkijKf1]]
240 240  
241 -^(0|[1-9]\d*)\.(0|[1-9]\d*)\.(0|[1-9]\d*)(?:-((?:0|[1-
168 +^(0|[1-9]\d*)\.(0|[1-9]\d*)\.(0|[1-9]\d*)(?:-((?:0|[1-9]\d*|\d*[a-zA-Z-][0-9a-zA-Z-]*)(?:\.(?:0|[1-9]\d*|\d*[a-zA-Z-][0-9a-zA-Z-]*))*))?$
242 242  
243 -9]\d*|\d*[a-zA-Z-][0-9a-zA-Z-]*)(?:\.(?:0|[1-9]\d*|\d*[a-zA-Z-
244 -
245 -][0-9a-zA-Z-]*))*))?$
246 -
247 247  **Must I adopt semantic versioning rules when switching to SDMX 3.0?**
248 248  
249 249  No. If backwards compatibility with pre-existing tools and processes is required, then it is possible to continue using the previous versioning scheme (with up to two version parts MAJOR.MINOR). Semantic versioning is indicated only for those use cases where a proper artefact versioning is required. If versioning does not apply to some or all of your artefacts, then rather migrate to non-versioned SDMX 3.0 artefacts.
... ... @@ -270,92 +270,4 @@
270 270  
271 271  In practice, the migration approach will often mirror the way in which organisations have migrated between earlier SDMX versions. Rarely, the new data models used mixed SDMX standard versions in their dependencies, and if they did then standard conversions were put in place. A typical method is to first migrate the re-used artefacts from the previous SDMX version to SDMX 3.0 and while doing so to apply the appropriate new semantic version string. From that point onwards, you can enjoy the advantages of the new SDMX versioning features for all those artefacts that require appropriate versioning.
272 272  
273 -[[1>>path:#sdfootnote1anc||name="sdfootnote1sym"]] Regular expressions, as specified in [[__W3C XML Schema Definition Language (XSD)__>>url:https://www.w3.org/TR/xmlschema11-2/]][[ >>url:https://www.w3.org/TR/xmlschema11-2/]][[__1.1 Part 2: Datatypes__>>url:https://www.w3.org/TR/xmlschema11-2/]][[.>>url:https://www.w3.org/TR/xmlschema11-2/]]
274 -
275 -[[2>>path:#sdfootnote2anc||name="sdfootnote2sym"]] The seconds can be reported fractionally
276 -
277 -[[3>>path:#sdfootnote3anc||name="sdfootnote3sym"]] ISO 8601 defines alternative definitions for the first week, all of which produce equivalent results. Any of these definitions could be substituted so long as they are in relation to the reporting year start day.
278 -
279 -[[4>>path:#sdfootnote4anc||name="sdfootnote4sym"]] The rules for adding durations to a date time are described in the W3C XML Schema specification. See __[[http:~~/~~/www.w3.org/TR/xmlschema>>url:http://www.w3.org/TR/xmlschema-2/#adding-durations-to-dateTimes]][[->>url:http://www.w3.org/TR/xmlschema-2/#adding-durations-to-dateTimes]][[2/#adding>>url:http://www.w3.org/TR/xmlschema-2/#adding-durations-to-dateTimes]][[->>url:http://www.w3.org/TR/xmlschema-2/#adding-durations-to-dateTimes]][[durations>>url:http://www.w3.org/TR/xmlschema-2/#adding-durations-to-dateTimes]][[->>url:http://www.w3.org/TR/xmlschema-2/#adding-durations-to-dateTimes]][[to>>url:http://www.w3.org/TR/xmlschema-2/#adding-durations-to-dateTimes]][[dateTimes>>url:http://www.w3.org/TR/xmlschema-2/#adding-durations-to-dateTimes]]__[[ >>url:http://www.w3.org/TR/xmlschema-2/#adding-durations-to-dateTimes]]for further details.
280 -
281 -[[5>>path:#sdfootnote5anc||name="sdfootnote5sym"]] The Validation and Transformation Language is a standard language designed and published under the SDMX initiative. VTL is described in the VTL User and Reference Guides available on the SDMX website [[__https:~~/~~/sdmx.org__>>url:https://sdmx.org/]][[.>>url:https://sdmx.org/]]
282 -
283 -[[6>>path:#sdfootnote6anc||name="sdfootnote6sym"]] In this chapter, in order to distinguish VTL and SDMX model artefacts, the VTL ones are written in the Arial font while the SDMX ones in Courier New
284 -
285 -[[7>>path:#sdfootnote7anc||name="sdfootnote7sym"]] See also the section "VTL-DL Rulesets" in the VTL Reference Manual.
286 -
287 -[[8>>path:#sdfootnote8anc||name="sdfootnote8sym"]] The VTLMappings are used also for User Defined Operators (UDO). Although UDOs are envisaged to be defined on generic operands, so that the specific artefacts to be manipulated are passed as parameters at their invocation, it is also possible that an UDO invokes directly some specific SDMX artefacts. These SDMX artefacts have to be mapped to the corresponding aliases used in the definition of the UDO through the VtlMappingScheme and VtlMapping classes as well.
288 -
289 -[[9>>path:#sdfootnote9anc||name="sdfootnote9sym"]] For a complete description of the structure of the URN see the SDMX 2.1 Standards - Section 5 - Registry Specifications, paragraph 6.2.2 ("Universal Resource Name (URN)").
290 -
291 -[[10>>path:#sdfootnote10anc||name="sdfootnote10sym"]] The container-object-id can repeat and may not be present.
292 -
293 -[[11>>path:#sdfootnote11anc||name="sdfootnote11sym"]] i.e., the artefact belongs to a maintainable class
294 -
295 -[[12>>path:#sdfootnote12anc||name="sdfootnote12sym"]] Since these references to SDMX objects include non-permitted characters as per the VTL ID notation, they need to be included between single quotes, according to the VTL rules for irregular names.
296 -
297 -[[13>>path:#sdfootnote13anc||name="sdfootnote13sym"]] For the syntax of the VTL operators see the VTL Reference Manual
298 -
299 -[[14>>path:#sdfootnote14anc||name="sdfootnote14sym"]] In case the invoked artefact is a VTL component, which can be invoked only within the invocation of a VTL data set (SDMX Dataflow), the specific SDMX class-name (e.g. Dimension, TimeDimension, Measure or DataAttribute) can be deduced from the data structure of the SDMX Dataflow, which the component belongs to.
300 -
301 -[[15>>path:#sdfootnote15anc||name="sdfootnote15sym"]] If the Agency is composite (for example AgencyA.Dept1.Unit2), the agency is considered different even if only part of the composite name is different (for example AgencyA.Dept1.Unit3 is a different Agency than the previous one). Moreover the agency-id cannot be omitted in part (i.e., if a TransformationScheme owned by AgencyA.Dept1.Unit2 references an artefact coming from AgencyA.Dept1.Unit3, the specification of the agency-id becomes mandatory and must be complete, without omitting the possibly equal parts like AgencyA.Dept1)
302 -
303 -[[16>>path:#sdfootnote16anc||name="sdfootnote16sym"]] Single quotes are needed because this reference is not a VTL regular name. ^^19^^ Single quotes are not needed in this case because CL_FREQ is a VTL regular name.
304 -
305 -[[17>>path:#sdfootnote17anc||name="sdfootnote17sym"]] The result DFR(1.0.0) is be equal to DF1(1.0.0) save that the component SECTOR is called SEC
306 -
307 -[[18>>path:#sdfootnote18anc||name="sdfootnote18sym"]] Rulesets of this kind cannot be reused when the referenced Concept has a different representation.
308 -
309 -[[19>>path:#sdfootnote19anc||name="sdfootnote19sym"]] See also the section "VTL-DL Rulesets" in the VTL Reference Manual.
310 -
311 -[[20>>path:#sdfootnote20anc||name="sdfootnote20sym"]] If a calculated artefact is persistent, it needs a persistent definition, i.e. a SDMX definition in a SDMX environment. In addition, possible calculated artefact that are not persistent may require a SDMX definition, for example when the result of a nonpersistent calculation is disseminated through SDMX tools (like an inquiry tool).
312 -
313 -[[21>>path:#sdfootnote21anc||name="sdfootnote21sym"]] See the VTL 2.0 User Manual
314 -
315 -[[22>>path:#sdfootnote22anc||name="sdfootnote22sym"]] See the SDMX Standards Section 2 – Information Model
316 -
317 -[[23>>path:#sdfootnote23anc||name="sdfootnote23sym"]] Besides the mapping between one SDMX Dataflow and one VTL Data Set, it is also possible to map distinct parts of a SDMX Dataflow to different VTL Data Set, as explained in a following paragraph.
318 -
319 -[[24>>path:#sdfootnote24anc||name="sdfootnote24sym"]] E.g., if in the data structure there exist 3 Dimensions C,D,E having the role of MeasureDimension, they should be considered as a joint MeasureDimension Z=(C,D,E); therefore when the description says “each possible value Cj of the MeasureDimension …” it means “each possible combination of values (Cj, Dk, Ew) of the joint MeasureDimension Z=(C,D,E)”.
320 -
321 -[[25>>path:#sdfootnote25anc||name="sdfootnote25sym"]] A typical example of this kind is the validation, and more in general the manipulation, of individual time series belonging to the same Dataflow, identifiable through the DimensionComponents of the Dataflow except the TimeDimension. The coding of these kind of operations might be simplified by mapping distinct time series (i.e. different parts of a SDMX Dataflow) to distinct VTL Data Sets.
322 -
323 -[[26>>path:#sdfootnote26anc||name="sdfootnote26sym"]] Please note that this kind of mapping is only an option at disposal of the definer of VTL Transformations; in fact it remains always possible to manipulate the needed parts of SDMX Dataflows by means of VTL operators (e.g. “sub”, “filter”, “calc”, “union” …), maintaining a mapping one-to-one between SDMX Dataflows and VTL Data Sets.
324 -
325 -[[27>>path:#sdfootnote27anc||name="sdfootnote27sym"]] This definition is made through the ToVtlSubspace and ToVtlSpaceKey classes and/or the FromVtlSuperspace and FromVtlSpaceKey classes, depending on the direction of the mapping (“key” means “dimension”). The mapping of Dataflow subsets can be applied independently in the two directions, also according to different Dimensions. When no Dimension is declared for a given direction, it is assumed that the option of mapping different parts of a SDMX Dataflow to different VTL Data Sets is not used.
326 -
327 -[[28>>path:#sdfootnote28anc||name="sdfootnote28sym"]] As a consequence of this formalism, a slash in the name of the VTL Data Set assumes the specific meaning of separator between the name of the Dataflow and the values of some of its Dimensions.
328 -
329 -[[29>>path:#sdfootnote29anc||name="sdfootnote29sym"]] This is the order in which the dimensions are defined in the ToVtlSpaceKey class or in the FromVtlSpaceKey class, depending on the direction of the mapping.
330 -
331 -[[30>>path:#sdfootnote30anc||name="sdfootnote30sym"]] It should be remembered that, according to the VTL consistency rules, a given VTL dataset cannot be the result of more than one VTL Transformation.
332 -
333 -[[31>>path:#sdfootnote31anc||name="sdfootnote31sym"]] If these DimensionComponents would not be dropped, the various VTL Data Sets resulting from this kind of mapping would have non-matching values for the Identifiers corresponding to the mapping Dimensions (e.g. POPULATION and COUNTRY). As a consequence, taking into account that the typical binary VTL operations at dataset level (+, -, *, / and so on) are executed on the observations having matching values for the identifiers, it would not be possible to compose the resulting VTL datasets one another (e.g. it would not be possible to calculate the population ratio between USA and CANADA).
334 -
335 -[[32>>path:#sdfootnote32anc||name="sdfootnote32sym"]] In case the ordered concatenation notation is used, the VTL Transformation described above, e.g. ‘DF1(1.0)/POPULATION.USA’ := DF1(1.0) [ sub INDICATOR=“POPULATION”, COUNTRY=“USA”], is implicitly executed. In order to test the overall compliance of the VTL program to the VTL consistency rules, it has to be considered as part of the VTL program even if it is not explicitly coded.
336 -
337 -[[33>>path:#sdfootnote33anc||name="sdfootnote33sym"]] If the whole DF2(1.0) is calculated by means of just one VTL Transformation, then the mapping between the SDMX Dataflow and the corresponding VTL dataset is one-to-one and this kind of mapping (one SDMX Dataflow to many VTL datasets) does not apply.
338 -
339 -[[34>>path:#sdfootnote34anc||name="sdfootnote34sym"]] This is possible as each VTL dataset corresponds to one particular combination of values of INDICATOR and COUNTRY.
340 -
341 -[[35>>path:#sdfootnote35anc||name="sdfootnote35sym"]] The mapping dimensions are defined as FromVtlSpaceKeys of the FromVtlSuperSpace of the VtlDataflowMapping relevant to DF2(1.0).
342 -
343 -[[36>>path:#sdfootnote36anc||name="sdfootnote36sym"]] the symbol of the VTL persistent assignment is used (<-)
344 -
345 -[[37>>path:#sdfootnote37anc||name="sdfootnote37sym"]] The result is persistent in this example but it can be also non persistent if needed.
346 -
347 -[[38>>path:#sdfootnote38anc||name="sdfootnote38sym"]] In case the ordered concatenation notation from VTL to SDMX is used, the set of Transformations described above is implicitly performed; therefore, in order to test the overall compliance of the VTL program to the VTL consistency rules, these implicit Transformations have to be considered as part of the VTL program even if they are not explicitly coded.
348 -
349 -[[39>>path:#sdfootnote39anc||name="sdfootnote39sym"]] Through SDMX Constraints, it is possible to specify the values that a Component of a Dataflow can assume.
350 -
351 -[[40>>path:#sdfootnote40anc||name="sdfootnote40sym"]] By using represented variables, VTL can assume that data structures having the same variables as identifiers can be composed one another because the correspondent values can match.
352 -
353 -[[41>>path:#sdfootnote41anc||name="sdfootnote41sym"]] A Concept becomes a Component in a DataStructureDefinition, and Components can have different LocalRepresentations in different DataStructureDefinitions, also overriding the (possible) base representation of the Concept.
354 -
355 -[[42>>path:#sdfootnote42anc||name="sdfootnote42sym"]] The representation given in the DSD should obviously be compatible with the VTL data type.
356 -
357 -[[43>>path:#sdfootnote43anc||name="sdfootnote43sym"]] Unidimensional datasets are those with a single 'indicator' or 'series code' dimension.
358 -
359 -[[44>>path:#sdfootnote44anc||name="sdfootnote44sym"]] A list of commonly used locales can be found in the Java supported locales: __[[https:~~/~~/www.oracle.com/java/technologies/javase/jdk8>>url:https://www.oracle.com/java/technologies/javase/jdk8-jre8-suported-locales.html]][[->>url:https://www.oracle.com/java/technologies/javase/jdk8-jre8-suported-locales.html]][[jre8>>url:https://www.oracle.com/java/technologies/javase/jdk8-jre8-suported-locales.html]][[->>url:https://www.oracle.com/java/technologies/javase/jdk8-jre8-suported-locales.html]][[suported>>url:https://www.oracle.com/java/technologies/javase/jdk8-jre8-suported-locales.html]][[->>url:https://www.oracle.com/java/technologies/javase/jdk8-jre8-suported-locales.html]][[locales.html>>url:https://www.oracle.com/java/technologies/javase/jdk8-jre8-suported-locales.html]]__[[url:https://www.oracle.com/java/technologies/javase/jdk8-jre8-suported-locales.html]]
360 -
361 -[[45>>path:#sdfootnote45anc||name="sdfootnote45sym"]] yyyy represents the calendar year while YYYY represents the year of the week, which is only relevant for 53 week years
196 +