diff --git a/src/unittest.py b/src/unittest.py
new file mode 100644
index 00000000..fb4d7d79
--- /dev/null
+++ b/src/unittest.py
@@ -0,0 +1,147 @@
+"""
+Python 2+3 script for unit testing RGL grammars.
+
+Usage: python path-to-script.py path/to/testfile.gftest ...
+The script muste be located in the RGL 'src' directory to work properly.
+
+The test file should look something like this:
+
+    LangSwe: jag sover i huset
+    LangEng: I sleep in the house
+
+    LangSwe: huset verkar stort
+    Lang: PhrUtt NoPConj (UttS (UseCl (TTAnt TPres ASimul) PPos (PredVP ...
+
+This contains two tests: Every test should be separated by empty lines.
+Every line starts with a language, followed by ":" and the sentence
+(or the abstract syntax tree if the abstract grammar is specified).
+"""
+
+from __future__ import print_function
+
+import sys
+import io
+import os.path
+from subprocess import Popen, PIPE
+from glob import glob
+
+ENCODING = 'utf-8'
+
+
+def usage():
+    print("Usage: python %s path/to/testfile.gftest ..." % (sys.argv[0],))
+    print()
+    print("(Note: to work properly this script must be located in")
+    print("the RGL 'src' directory, and gf must be in the system path)")
+    print()
+
+
+def error(linenr, *args):
+    print("[Error at line %s]" % (linenr,), *args)
+
+
+def gferror(reply):
+    return (reply.startswith('The parser failed')
+            or reply.startswith('The sentence is not complete')
+            or reply.startswith('Function') and reply.endswith('is not in scope'))
+
+
+def importfile(linenr, lang):
+    scriptdir = os.path.dirname(sys.argv[0]) or '.'
+    langfiles = glob('%s/*/%s.gf' % (scriptdir, lang))
+    if not langfiles:
+        error(linenr, "Cannot find language:", lang)
+        exit(1)
+    elif len(langfiles) > 1:
+        error(linenr, "Found multiple language files for %s:" % (lang,), *langfiles)
+        exit(1)
+    return langfiles[0]
+
+
+def stripstrings(strings):
+    return [s for s0 in strings for s in [s0.strip()] if s]
+
+
+def runtest(testlines):
+    # first we build the input to the GF process:
+    gfinput = ''
+    testing = False
+    for linenr, line in enumerate(testlines, 1):
+        if ':' in line:
+            if not testing:
+                gfinput += 'ps "### %d" \n' % (linenr,)
+                testing = True
+            lang, sent = stripstrings(line.split(':', 1))
+            gfinput += 'ps "+++ %d %s" \n' % (linenr, lang)
+            langfile = importfile(linenr, lang)
+            gfinput += 'i %s \n' % (langfile,)
+            if '/abstract/' in langfile:
+                gfinput += 'pt %s \n' % (sent,)
+            else:
+                gfinput += 'p -lang=%s "%s" \n' % (lang, sent)
+        elif not line.strip():
+            testing = False
+        else:
+            error(linenr, "Ill-formatted line in test file:", line)
+            exit(1)
+
+    # then we call GF with the script, catching stdout:
+    gf = Popen('gf -run'.split(), stdin=PIPE, stdout=PIPE)
+    stdout, _stderr = gf.communicate(gfinput.encode(ENCODING))
+    stdout = stdout.decode(ENCODING)
+
+    # then we analyse the result from the GF process:
+    totalerrors = 0
+    alltests = stripstrings(stdout.split('###'))
+    for testnr, test in enumerate(alltests, 1):
+        sents = stripstrings(test.split('+++'))
+        startline = int(sents.pop(0))
+        print("Test %d (line %d..): %d examples" % (testnr, startline, len(sents)))
+        testerrors = 0
+        oldresults = []
+        for sresults in sents:
+            alltrees = stripstrings(sresults.splitlines())
+            linenr, lang = alltrees.pop(0).split()
+            if len(alltrees) == 0 or len(alltrees) == 1 and gferror(alltrees[0]):
+                theerror = alltrees[0] if alltrees else "No parse trees found"
+                error(linenr, theerror)
+                testerrors += 1
+            else:
+                allerrors = [(sum(tree not in oldtrees for _, _, oldtrees in oldresults), tree)
+                             for tree in alltrees]
+                besterrors, besttree = min(allerrors)
+                if besterrors > 0:
+                    for oldlinenr, oldlang, oldtrees in oldresults:
+                        if besttree not in oldtrees:
+                            error(linenr, "Line %s (%s) is not a translation of line %s (%s)"
+                                    % (linenr, lang, oldlinenr, oldlang))
+                            testerrors += 1
+                oldresults.append((linenr, lang, alltrees))
+        if not testerrors:
+            print("OK!")
+        print()
+        totalerrors += testerrors
+
+    # finally we report a summary:
+    if not totalerrors:
+        print("All %d tests passed!" % (len(alltests),))
+    else:
+        print("There were %d errors in %d tests!" % (totalerrors, len(alltests)))
+    print()
+
+
+if __name__ == '__main__':
+    if len(sys.argv) <= 1:
+        usage()
+        exit(1)
+    for filename in sys.argv[1:]:
+        try:
+            print("# Testing file:", filename)
+            with io.open(filename, encoding=ENCODING) as F:
+                print()
+                runtest(F)
+        except IOError as err:
+            print(err)
+            print()
+            usage()
+            exit(1)