Current status of the build, including test scripts:

INSTALLATION

The software is tested with GHC-8.0.1. Newer versions are likely to work too. Installation works via cabal, which comes as part of both stack and the ghc-package.

To install amperspiegel in your path, run (in the directory of this README):

cabal install

Alternatively, you can try a manual install using ghc --make amperspiegel but this will likely complain about missing packages, which you can install via cabal install. For a list of required packages, check 'build-depends:' in the amperspiegel.cabal file. The packages base, containers and text are probably already installed with ghc.

LICENSE

I put a GPL-3 license to get you started on using and modifying this software. For the license see the file LICENSE. Since I am the only contributor, you can ask me for a less restrictive license (like BSD / Apache 2.0) if you need one. I'll probably be open to giving it to you.

GETTING STARTED

Here is a script you may try after installing amperspiegel in your path

For an overview of all switches, use: amperspiegel -h

After having done all this, send me an email (or a message in github, or ..) and tell me what you think. I like knowing who is trying out my software. Don't be shy: I'll remove this notice once I'm getting too many such emails.

Implementation notes

A state assigns a population to each variable. Initially, these are the assignments:

• "parser" gets assigned a population describing the built-in parser. It contains the relations "recogniser", "continuation", "choice" and "nonTerminal" to describe the parser, plus a set of rules that is applied after parsing, using "rule", "eFst", "eSnd", "pre", "post", and some other relations.
• "asParser" gets assigned a set of rules to change a ampersand-style script that talks about "concepts", "declaration", "relation" etc into a parser.
• "switches" gets assigned the switches passed to amperspiegel.
• possibly some others, use -list to see them

All other variables get assigned the empty population by default, the following in particular:

• "population", which stands for the population we're working on currently Consequently, most switches get "population" as default argument

The switch -i uses "parser" as a parser to parse each of its arguments Next, the union over the resulting list of populations is taken To this population, "parser" is then applied as a rule-set This result is put in "population" (Resulting in a state)

The switch -asParser applies "asParser" on "population", then filters it. Consequently, "population" only contains stuff relevant for parser and rules. It then copies "population" to "parser".

Switches -count and -show displays stuff about a variable.

Future work

See the issues page.