BioScript

A simple type-safe Domain Specific Language (DSL) for chemistry and biology.

Usage

usage:

main.py [-h] -i INPUT [-d] [-wd WORKING_DIRECTORY] [-o OUTPUT]
       [-t {m,i,p,inkwell,l,llvm,ir,mfsim,puddle}] [-cfg] [-inline]
       [-stats] [-lu] [-sim {False,True}] [-id {0,1,2,32,4,8,16}]
       [-nf] [-smarts SMARTS] [-tcl {none,warn,error}] [-tc]
       [-tcu {complex,simple,s,c}] [-epa EPA_DEFS] [-abs ABS_INT]
       [--dbname DBNAME] [--dbuser DBUSER] [--dbpass DBPASS]
       [--dbaddr DBADDR] [--dbdriver {odbc,mysql}] [-lib LIBRARY]
       [-flow {passive,p,a,active}] [--cdb CDB] [--schema SCHEMA]
       [--validate]

Optional Arguments:

Short	Long	Flag	Purpose
-h	--help		show this help message and exit
-i	--input	path/to/input.bs	Location of input file
-d	--debug		Enable debugging.
-t	--target	{i,inkwell,p,puddle,m,mfsim,l,llvm, ir,}	What platform do you wish to target?
-o	--output	path/to/output/dir	Enable writing output. Must be set to write anything
-wd	--working-directory	path/to/directory	Directory from where you wish to work
-cfg	--write-cfg		Write the CFG to a dot file
-inline	--inline		Inline all non-recursive functions
-lu	--loopunroll		Unroll all un-rollable loops
-stats	--stats		Print the stats to std out
-cfg	--write-cfg		Write the programs control flow graph to disk

Chemistry:

Chemistry specific arguments

Short	Long	Flag	Purpose
-sim	--simulate	{True, False}	Simulate chemistry?
-id	--identify	{0,1,2,32,4,8,16}	Chemical identification level
-nf	--no-filters		Disable smart filter creation
-smarts	--smarts	0-255	Length of smarts file to use

Typing:

Typing specific arguments

Short	Long	Flag	Purpose
-tcl	--typechecklevel	{error, warn, none}	What interactions elicit notifications
-tc	--typecheck		Enable type checking of input program
-tcu	--typesused	{s, simple, c, complex}	What types to use to typecheck a program, s = {mat, nat, real}, c = {all types in chemtype}
-epa	--epa-defs	Path/to/epa_defs.json	Location of the EPA definition file (default: ./resources/epa.json)
-abs	--abs-int	path/to/abs-int.txt	Location of the abstract interaction file (default: ./resources/abstract-interaction.txt)

DB:

Database specific arguments

Flag	Argument	Purpose
--dbname	str	Name of database.
--dbuser	str	Database user.
--dbpass	str	Database password for user.
--dbaddr	{IP, Host}	Address of database.
--dbdriver	{mysql,odbc}	Database driver.

Inkwell:

Inkwell specific (-t inkwell) flags

Short	Long	Flag	Purpose
-lib	--library	path/to/json/component/lib	The path to the component library
-flow	--flow	{p, a, passive, active}	Which type of flow-based device are you targeting
-cdb		database	Name of component database (not supported)
	--schema	Path to schema.json	The ParchMint schema file
	--validate		Validate a netlist against the ParchMint schema

Setup

You'll need Python3 (apt-get install python3), and GraphViz installed (apt-get install graphviz)

If on macOS, install brew, then brew install python3 and brew install graphviz.

BioScript's grammar is attached as a submodule. Hence, if cloning this repo, make sure to include the --recursive tag:

git clone --recursive https://github.com/lilott8/BioScript

Also, if BioScriptGrammar (or any other submodule) is updated, you'll need to run git submodule foreach git pull origin master if your version of git does not do this automatically for you.

Install required python modules: pip install -r requirements.txt

Try running the example usage (below). It's possible that a C-style comment /* ... */ is present in some of the grammar files generated by Antlr. Open any offending files and use proper pythong comments # to mitigate this issue.

Some modules may have not installed properly even using the recursive pip install method above.

Continue trying the example usage (below), and pip install any missing modules.

Note: you may need to use pip3 install if python3/pip3 is not your default python/pip alias.

Example Usages:

Enable debug (-d), compile a file (-i ...), target BioScript's IR (-t ...), and place the output files in the specified location:

python main.py -d -i tests/test_cases/mix/ir_sisd.bs -t ir -o ./output -cfg

Your output folder should includ the following files: ir_sisd_main_dag.dot, ir_sisd.ir, ir_sisd_main_basic_blocks.dot.

You can either use graphviz to create a .png of the generated dag (dot -Tpng dag.dot -o dag.png), or use http://www.webgraphviz.com and paste the contents of the .dot or .dag file for visualization.