Tudat update all files to new (short) license

license_update.py

import os
import sys
import re
import fnmatch
import codecs

# Define copyright header for CMakeLists
newlic1 = "".join((" #    Copyright (c) 2010-2017, Delft University of Technology\n",
                   " #    All rigths reserved\n",
                   " #\n",
                   " #    This file is part of the Tudat. Redistribution and use in source and\n",
                   " #    binary forms, with or without modification, are permitted exclusively\n",
                   " #    under the terms of the Modified BSD license. You should have received\n",
                   " #    a copy of the license with this file. If not, please or visit:\n",
                   " #    http://tudat.tudelft.nl/LICENSE."))

# Adapt for cpp and h and matlab files
newlic2 = newlic1.replace("#","*")
newlic2 = "/" + newlic2[1:]
newlic3 = newlic1.replace("#","%")

topsection = "((\/\*)|(\s{0,2}#))\s{1,4}(Copy).*(Delft).*(\n\s{0,2}(\*|#).*)*"

# Generic regex expression for matching a section

# Ends section based on white comment line (\s{0,1}[\*|#]\s{1,4}(%s)\w*\n((.)*\n)*?(\s{0,1}[\*|#](\s*|\/)(?:\n)))'
regex_sec_generic    = '(\s{0,1}(\*|#|%%)\s{1,4}(%s)\w{1,6}\s*\n)((.|\n)*?)(?=^(\s{0,1}(\*|#|%%)((\s{1,4}\w{5,10}\s*)|\/)){0,1}\s*$)'

# Specific regex for matching copyright section
regex_sec_copyright  = '((\/\*)|(\s{0,2}(#|%)))\s{1,4}(Copy)(.|\n)*?(DAMAGE\.)'

# Specific implementation of changelog, notes and references sections, resp
regex_sec_changelog  = regex_sec_generic % "[C|c][H|h][A|a][N|n]"
regex_sec_notes      = regex_sec_generic % "[N|n][O|o][T|t][E|e]"
regex_sec_references = regex_sec_generic % "[R|r][E|e][F|f][E|e]|[B|b][I|i][B|b][L|l]|[C|c][I|i][T|t][A|a]|[S|s][O|o][U|u][R|r]"

for dname, dirs, files in os.walk("/home/jacco/Gits/tudatBundle/tudat/Tudat"):
    for fname in files:

        # dname = '/home/jacco/Gits/tudatBundle/tudat/Tudat/Astrodynamics/MissionSegments/UnitTests'
        # fname = 'unitTestMathematicalShapeFunctions.cpp'

        #print(fname)

        fpath = os.path.join(dname, fname)
        if fname.endswith(('.txt', '.example', '.cmake','.in', '%m', '.cpp', '.h')):
            with codecs.open(fpath, 'r', 'utf-8') as f:
                s = f.read()
                s = s.encode('ascii', 'ignore').decode('ascii')

            # Debug
            # print("== CHANGELOG ==")
            # c = re.search(regex_sec_changelog,s, re.M)
            # print(len(c.group(0).split('\n')))
            # print(c.group(0))
            # print("== REFERENCES ==")
            # c = re.search(regex_sec_references,s, re.M)
            # print(c.group(0))
            # print(len(c.group(0).split('\n')))
            # print("== NOTES ==")
            # c = re.search(regex_sec_notes,s, re.M)
            # print(c.group(0))
            # print(len(c.group(0).split('\n')))


            # sys.exit()
            # Delete the changelog
            s = re.sub(regex_sec_changelog, "", s, flags=re.M)

            #print(s)
            #sys.exit()


            # Match references and notes
            c = re.search(regex_sec_notes, s, re.M)
            if c is not None:
                if len(c.group(0).split('\n')) <= 3:
                    s = re.sub(regex_sec_notes, "", s, flags=re.M)
            c = re.search(regex_sec_references, s, re.M)
            if c is not None:
                if len(c.group(0).split('\n')) <= 3:
                    s = re.sub(regex_sec_references, "", s, flags=re.M)

            # Replace the copyright
            c = re.search(regex_sec_copyright, s)

            if c is not None:
                if re.match("\/\*",c.group(0)):
                    s = re.sub(regex_sec_copyright, newlic2, s)
                elif re.match("\s{0,2}#",c.group(0)):
                    s = re.sub(regex_sec_copyright, newlic1, s)
                elif re.match("\s{0,2}%",c.group(0)):
                    s = re.sub(regex_sec_copyright, newlic3, s)
                else:
                    print("Couldn't determine comment type for %s" % fpath )
                with open(fpath, "w") as f:
                    f.write(s)
            else:
                print("Couldn't find match in %s" % fpath)

Features:

• recursively goes through all files and replaces the license statement in the header.
• preserves notes and references if non-empty
• deletes empty sections
• deletes the changelog section
• handles CMake, multi-line CPP and Matlab comment styles markes (#, /* */, %, respectively)

Regex build and tested with http://www.regexr.com. The section regex detects the beginning of a comment section (by the section heading) and matches everything in between till either the next section heading or end of comment block (using positive look ahead).