This is the Asbru Interpreter Version {version} released on {DATE}. ====================================================================== This package contains the following. The usage of these files is described in detail in description.pdf. AsbruInterpreter/run-AsbruInterpreter.bat the batch file to start the interpreter (see below) AsbruInterpreter/run-AsbruInterpreter.sh the shell script to start the interpreter (see below) AsbruInterpreter/AsbruInterpreter.jar the executeable of the interpreter AsbruInterpreter/README.txt this file AsbruInterpreter/description.pdf a short documentation of the interpreter AsbruInterpreter/SupportedAsbruElements.txt list of all Asbru elements with status of their implementation AsbruInterpreter/roadmap.txt mapping of features to interpreter versions AsbruInterpreter/cyclical-plan_features.txt short description of implemented features of AsbruInterpreter/dtd/Asbru_7.3h.dtd the latest version of the Asbru DTD AsbruInterpreter/dtd/Asbru_7.3h_DELTA.dtd the latest version of the Asbru DTD containing additional DELTA links AsbruInterpreter/dtd/project.dtd the commented DTD of the project files AsbruInterpreter/dtd/project_dtd-doc.html automatically generated HTML documentation of project.dtd the sample files described in description.pdf can be found in AsbruInterpreter/examples. Installation ------------ Copy the content of this Archive to a directory of your choice. Prerequisites ------------- In order to start the interpreter, you need Java 1.4 (or higher) available at java.com. Starting the interpreter ------------------------ Type run-AsbruInterpreter[.bat] or run-AsbruInterpreter.sh on the command line. is the path of the project file. To generate also both HTML files, use "-transform-all -transform-plans" as additional arguments. Use "-help" as an argument to see a list of all possible arguments. shortcut for the examples: simply start 'run.bat' (or 'run.sh') in each subdirectory. This produces xxx-all.html - the detailed logfile xxx-plansonly.html - a logfile showing only plan state changes xxx-logfile.xml - the XML logfile in the same directory as the project file (where xxx is the name of the project file without extension). For details please see 'description.pdf'. Release Notes ------------- Version 1.1 has been extensively tested with more than 2000 test cases (in addition to the 3 examples). Version 1.1 supports now continuation-specification (full 'wait-for'), propagation-specification (full 'abort-if' and default specification from 'wait-for'), wait-for-optional-subplans and retry-aborted-subplans. For a complete list of supported (and unsupported) Asbru elements see 'SupportedAsbruElements.txt'. For a mapping of features to Asbru-Interpreter versions (since version 1.1.5) see 'roadmap.txt'. Since version 1.2, there is a new command-line option: -combine-plan-state-transitions=[yes|no] (default=yes) ... if set to 'no', plan-state transitions of the same plan, which occur in the same time-step, are written in single lines to the log-file (XML and HTML version), instead of only logging the last plan-state Since version 1.2, there are two changes regarding input-file format: - in the project file, "TABLE" can now be used as an alias for "ASCII" (which was not a very appropriate name) Remark: the Table format supports now also "missing" / "unknown" values - simply, concatenate two 'tab's in the input file. Note: such an "unknown" value is mapped internally to an own value (the same as for an "not-yet-defined" value) - so keep in mind that for comparisons that an unknown value never equals any "known" value. You can use inside a (together with now as time-annotation) to check if a parameter has currently a value - the interpreter now fully supports a second input file format, the so-called "key-value" format: Description: Like the Table format, the key-value format also supports qualitative values. Unlike the table format, no header line is needed. Like the table format, each line starts with date- and time-stamp. Unlike the table format, there is no fixed count of columns - instead after date/time each line can contain one or more key=value pairs, e.g. "DATE TIME parameter1=1.0 parameter4=YES". To declare a key-value file, use "KEY-VALUE" for the file-type attribute in the project file. Note: the Asbru Interpreter does not yet support multiple input files. Since version 1.2.2, the project.dtd has been slightly changed, to prepare for the support of multiple input-files. The element embeds now one or more elements. The Asbru Interpreter will try to choose the appropriate input-file for each parameter automatically. See also AsbruInterpreter/dtd/project_dtd-doc.html or look at the project files of the 3 examples. Since version 1.2.2, there is a new command-line option: -writeModuleGraph ... if specified, the Asbru Interpreter writes the internal module-graph (list of all modules and their connections) as a table to a text file named ".graph.txt". Useful to get insight how the Interpreter works internally. (first implementation; needs more feedback) Version 1.2.6 brings besides some bug-fixes: - support for multiple input files per project: see single-plan project for an example (each parameters is in an own file) - improvement of HTML output: qualtitative value names instead of integer numbers (except result of qualitative-parameter-def) and undefined instead of -1/NaN, ... New in Version 1.3.0: time-annotation in plan-activation, parameter-change, refer-to, cyclical-plan with an expression as start-time, improved HTML output New in Version 1.3.2: - important internal bug fixes - command line option -disable-trace to completely disable the XML trace-log file - improved performance (relevant for large batch processing) - new supported Asbru elements: * episode-analysis-def * comparision-def as child of parameter-def