GRS syntax

The syntax descried here appeared in version 0.44.0 (2017/09/05). For the previous syntax, view old_grs page.

Global structure

A GRS is composed by a set of declarations that may be provided in several files. These files are expected to used the .grs or the .dom file extension.

Five kinds of declarations can be used:

The first two kinds (Feature domain and label domain) can only be used at the top level and they cannot be nested (see below the multi-file handling).

Feature domains

In graphs and in rules, nodes contain feature structures. To control these feature structures, a feature domain may be given first. In the feature domain declaration, feature names are identifiers and are defined as:

In closed features definition, feature values can be any strings; double quotes are required for string that are not lexical identifier (like values for pers).

❓ Explain merging of two feature domain declarations.

features {
  cat: n, np, v, adj;
  mood: inf, ind, subj, pastp, presp;
  lemma: STRING;
  phon: STRING;
  pers: "1","2","3";
  position: NUMERIC;

REM: values of pers feature are numerals but the only way to restrict to the finite domain {1, 2, 3} is to declare it as a closed feature and possible values as strings.

Label domains

An explicit set of valid labels for edges may be given after the labels keyword. It is possible to give several label domain declarations; the union of the different sets is then considered (❓ what about duplicates?, see #4).

By default, edges are drawn with a black solid line and above the figure in DEP representation.

To modify the color or the position of the edges, the user can add attributes to a label with suffixes:

Several suffixes can be used simultaneously.

labels { OBJ, SUJ, DE_OBJ, ANT, ANT_REL@red, ANT_REP@blue@bottom@dash }

⚠️ the color, style and position management will change soon (see #5)


Rule declaration is introduced by the keyword rule. For the syntax, see rule page.


Strategies are used specify the way rules are applied during transformation. The syntax of strategies definition is:

strat strat_id {
  <<< strategy_description >>>

Strategy descriptions are defined by the following syntax:

  S ::= rule_id               % Apply the gives rule
      | package_id            % Apply any rule defined in the package (not in sub-packages)
      | strat_id              % Called a strategy defined elsewhere
      | Pick (S)              % Select arbitrary one of the graph produced by the strategy S
      | Alt (S_1, …, S_n)     % Collect graphs produced by each sub-strategies (union)
      | Seq (S_1, …, S_n)     % Apply sequentially S_1, then S_2 on S_1 output …
      | Iter (S)              % Iterate the application of the strategy S until normal forms
      | If (S, S_1, S_2)      % If S is productive then it is equivalent to S_1 else it is equivalent to S_2

Other constructor are provided for some strategies

  Empty ≜ Seq()               % The Empty strategy always returns the input graph
  Try (S) ≜ If (S, S, Empty)  % Equivalent to S if S is productive else it returns the input graph

Computing one normal form

To compute only one normal form with a strategy S, one can used the strategy: Pick (Iter (S)): the strategy Iter (S) computes the full set of normal forms and the Pick operator choses one of them. But this may be not efficient if the number of normal forms is high.

For this case, another implementation of the rewriting is available with the operator Onf (the name stands for ‘One normal form’). With this operator, only one normal form is built, and so :

  Onf (S) = Pick (Iter (S))

⚠️ But Onf can be safely used only if the strategy is terminating. More info about this on the rewriting page.


Packages are used to organize the set of declarations and to define scopes of definitions. Syntax of packages definition:

package package_id {

where declarations_list is a list of declarations of rules, packages and strategies.

The syntax for accessing to some element e defined in package P is P.e. In case of nested packages, an identifier may look like P1.P2.P3.e. When a reference is made to an element P1.P2.e, the system tries to find inside the current package a sub-package P1 which contains a sub-package P2 which contains an element e. If no such element is found, the same thing is searched recursively, first in the mother package of the current one, up to the root package.

Note that it is not allowed to have a domain declaration inside a package.

Multi-file management

When a GRS become large and contains an high number of rules, it is sensible to define it in through a set of files. Two mechanisms are available for this purpose: external file import and external file inclusion.

External file import

At any place in a list of declaration in a GRS file, one can use the syntax:

import "filename.grs"

This creates a new package with the same name as the file (without the .grs extension). Hence, the meaning is the same as the following code:

package filename {
  <<< content of the file "filename.grs" >>>

As a consequence, it is not allowed to import a file which contains domain declarations because it would be equivalent to a domain declaration inside a package and this is forbidden. To use a external domain declaration, one should use the file inclusion.

External file inclusion

With file inclusion, the content of the external file is interpreted as if it was placed directly in the file at the same place. In other words the code:

include "filename.grs"

has the same meaning as

<<< content of the file "filename.grs" >>>

A complete example

We consider the same GRS defined through the multi-file mechanism and with a single file.

Multi-file declaration

Consider a folder with the five files:

Single file declaration

The five files above define a GRS, equivalent to the one below:

Apply the GRS to a graph

Consider the graph defined in