User visible changes -------------------- Add version1 output module: writes out the version number of the used requirement set to a file. Output module latex2 is configurable in the way, that it is possible to give a list of attributes which are printed in the small table beneath the requirement itself. Output module graph is configurable: it is possible to specify which requirement attributes are transposed into node properties. latex2 output enhancement: comma seperates two requirements - not ends them. There is no additional comma at the end of the list anymore. oopricing1 enhancements: added supplier column, added result sheet, added version information to file properties. Removed xml formatting problem. graph enhancements: version information was added to the output graph. Utility added which reads in the csv output of the oopricing1 and converts it into a dependency graph. Nodes include information about requirement pricing and dependent pricing. Also the nodes are colored after the compliance: green=fully, orange=partial and red=none compliant. Fix 3052358: a specific error is logged if a topic does not exists at all. (This is another case when the requirement is not used because the topic is not included - in the later case a debug message is logged, because this is seen as a normal use case.) Add man page for version1 output. Add man page for rmtoo-pricing-graph. Fixed bug which made it possible that a requirement references itself: self-dependency is excluded now. Internal Changes ---------------- Blackbox tests comparison enhancement: on different systems / different times the XML files were stored in different orders. Therefore all XML files are now parsed and the content is compared - instead of one seriallized representation. Changed xml comparison from simple do-it-yourself to xmldiff: the later one really works in all cases. The first one had problems when the nodes are sorted in a different order. This does not solve the problem completely: differences are still found where no differences should be. What DOES fix this (i.e. getting always the same XML document) is sorting the styles (which are read from a dictionary) before outputting them.