Video tutorial | Example 1 video | Example 2 video

1 Select how to provide the input

File must be in one of the following format: rdf, rj, jsonld, nt, nq, trix, trig, tql, ttl, n3, brf.
File may be compressed using: gz, bz2, xz, 7z.


2 Combine processors

Insert here the commands you want to use for the RDFpro processing. The list of the available commands is documented in the right side window. Load an example.

Additional files

These additional files can be included in the commands using the labels #file1, #file2, #file3, #file4.
Some other pre-loaded files are available, too (see list).


3 Select how to get the output


Extended documentation Usage examples

AVAILABLE PROCESSORS:

@groovy         Transform the stream using a user-supplied Groovy script
  [-p]          use multiple script instances in parallel (script pooling)
  SCRIPT        either the script expression or the name of the script file
  [ARG...]      optional arguments to be passed to the script

@mapreduce      Performs a map/reduce like computation
  [-b PRED]     directly emits input quads matching bypass predicate PRED
  [-r REDUCER]  apply REDUCER to quad partitions (default: identity reducer)
  [-e PRED]     keeps only partitions with a quad matching predicate PRED
  [-a PRED]     keeps only partitions whose quads all match predicate PRED
  [-u]          deduplicate partition quads before reducing
  MAPPER...     map input quads using MAPPERs: either a script, 'e' to map by
                entity (subject or object resource) or string matching regex
                [spoc]+ to map by hash of selected s, p, o, c components

@prefix|@p      Adds missing prefix-to-namespace bindings
  [-f FILE]     use prefixes from FILE instead of prefix.cc

@rdfs           Emits the RDFS closure of input quads
  [-e RULES]    exclude RULES in comma-separated list (default: no exclusions)
  [-d]          decompose OWL axioms to RDFS (e.g. equivalentClass -> subClass)
  [-t]          drop uninformative <x rdf:type _:b> statements (default: keep)
  [-C | -c URI] emits closed TBox to default graph [-C] or graph URI [-c]
  [-b URI][-w]  use base URI [-b] and optional BNode rewriting [-w] to load TBox
  [FILE...]     load TBox from FILE...

@rules          Emit the closure of input quads using a set of rules
  [-r RULESETS] use comma separated list of RULESETs (rdfs, owl2rl, custom file)
  [-B BINDINGS] use comma separated list of var=value BINDING to customize rules
  [-p] MODE     set partitioning MODE: none (default), entity, graph, rules
  [-g] MODE     set graph inference MODE: none (default), global, separate, star
  [-G] URI      set global graph URI for inference modes global and star
  [-t]          drop uninformative <x rdf:type _:b> statements (default: keep)
  [-u]          emit unique statements (may be faster than separate @unique)
  [-C | -c URI] emit TBox data closure unchanged [-C] or to graph URI [-c]
  [-b URI][-w]  use base URI [-b] and BNode rewriting [-w] to load TBox data
  [FILE...]     load TBox data (e.g., TBox) from FILE...

@smush          Performs smushing, using a single URI for each sameAs cluster
  URI...        use ranked namespace URIs to select canonical URIs

@stats          Emits VOID structural statistics for its input
  [-n URI]      use namespace URI to mint URIs for VOID dataset instances
  [-p URI]      create a dataset for graphs linked to a source via property URI
  [-c URI]      look for graph-to-source quads in graph URI
  [-t NUM]      emits only VOID partitions with at least NUM entities or triples
  [-o]          enable computation of void:classes and void:properties (costly)

@tbox           Emits only quads belonging to RDFS or OWL TBox axioms.

@transform|@t   Discards/replaces quads based on matching and replace exp.
  [EXP]         sequence of rules on quad components X (values: s, p, o, c):
                  +X value list => quad dropped if X not belongs to list
                  -X value list => quad dropped if X belongs to list
                  =X value => component X replaced with value (after filters)
                values are in Turtle and include following wildcard values: <*>
                => any URI; _:* => any BNode; * => any plain literal; *@* => any
                lang literal; *^^* => any typed literal; *@xyz => literals lang
                xyz; *^^<uri> => literals type <uri>; *^^ns:iri => literals type
                ns:iri; *^^ns:* => literals any type with prefix ns; ns:* => any
                uri with prefix ns; <ns*> => any uri in namespace ns

@unique|@u      Discards duplicates in the input stream
  [-m]          merges quads with same <s,p,o> and different graphs in a unique
                quad, put in a graph described with quads of all source graphs


QUICK SYNTAX:

CMD ::=
  @p args1                          builtin processor (see below)
  @p1 args1 ... @pN argsN           sequence composition
  { @p1 args1 , ... , @pN argsN }S  parallel composition, set operator S

S ::=
  a   multiset sum of @p1 .. @pn outputs, keep duplicates (fast, default)
  u   set union of @p1 .. @pn outputs
  U   multiset union of @p1 .. @pn outputs (with duplicates)
  i   set intersection of @p1 .. @pn outputs
  I   multiset intersection of @p1 .. @pn outputs (with duplicates)
  d   set difference @p1 \ (union of @p2 .. @pN)
  D   multiset difference @p1 \ (union of @p2 .. @pN) (with duplicates)
  s   symmetric set difference: quads in at least a @pi but not all of them
  S   multiset symmetric set difference
  n+  emits quads with at least n (number) occurrences (no duplicates out) 
  n-  emits quads with at most n (number) occurrences (no duplicates out)

RDFpro is public domain software | Official website