Add a linter for the PSL and use it on Travis CI

.gitignore

@@ -0,0 +1 @@
+linter/log

.travis.yml

@@ -1,21 +1,38 @@
 language: c
-compiler:
-  - gcc
-# Change this to your needs
+compiler: gcc
+addons:
+    apt:
+        packages:
+            - python3
+            - autoconf
+            - automake
+            - autopoint
+            - libtool
+            - gettext
+            - libidn11
+            - libidn11-dev
+            - libidn2-0
+            - libidn2-0-dev
+            - libicu48
+            - libicu-dev
+            - libunistring0
+            - libunistring-dev
+
 script:
-  - DIR=`pwd`
-  - git clone https://github.com/rockdaboot/libpsl
-  - cd libpsl
-  - echo "EXTRA_DIST =" >gtk-doc.make
-  - echo "CLEANFILES =" >>gtk-doc.make
-  - autoreconf --install --force --symlink
-  - OPTIONS="--with-psl-file=$DIR/public_suffix_list.dat --with-psl-testfile=$DIR/tests/test_psl.txt"
+    - DIR=`pwd`
+    - cd linter
+    - ./pslint_selftest.sh
+    - ./pslint.py ../public_suffix_list.dat
+    - cd $DIR
+    - git clone --depth=1 --branch newfmt https://github.com/rockdaboot/libpsl
+    - cd libpsl
+    - echo "EXTRA_DIST =" >gtk-doc.make
+    - echo "CLEANFILES =" >>gtk-doc.make
+    - autoreconf --install --force --symlink
+    - OPTIONS="--with-psl-file=$DIR/public_suffix_list.dat --with-psl-testfile=$DIR/tests/tests.txt"
 # Test PSL data with libicu (IDNA2008 UTS#46)
-  - ./configure -C --enable-runtime=libicu --enable-builtin=libicu $OPTIONS && make clean && make check -j4
+    - ./configure -C --enable-runtime=libicu --enable-builtin=libicu $OPTIONS && make clean && make check -j4
 # TEST PSL data with libidn2 (IDNA2008)
-  - ./configure -C --enable-runtime=libidn2 --enable-builtin=libidn2 $OPTIONS && make clean && make check -j4
+#    - ./configure -C --enable-runtime=libidn2 --enable-builtin=libidn2 $OPTIONS && make clean && make check -j4
 # TEST PSL data with libidn (IDNA2003)
-  - ./configure -C --enable-runtime=libidn --enable-builtin=libidn $OPTIONS && make clean && make check -j4
-before_install:
-  - sudo apt-get -qq update
-  - sudo apt-get -q install autoconf automake autopoint libtool gettext libidn11 libidn11-dev libidn2-0 libidn2-0-dev libicu48 libicu-dev libunistring0 libunistring-dev
+#    - ./configure -C --enable-runtime=libidn --enable-builtin=libidn $OPTIONS && make clean && make check -j4

linter/README.md

@@ -0,0 +1,35 @@
+This directory contains a linter for the Public Suffix List.
+
+Before you commit any changes to the PSL, please use the
+linter to check the syntax.
+
+Usage
+=====
+
+(from the repo's main directory)
+
+$ linter/pslint.py public_suffix_list.dat
+
+$? is set to 0 on success, else it is set to 1.
+
+
+Selftest
+========
+
+Every change on pslint.py should be followed by a self-test.
+
+```
+$ cd linter
+$ ./pslint_selftest.sh
+test_allowedchars: OK
+test_dots: OK
+test_duplicate: OK
+test_exception: OK
+test_punycode: OK
+test_section1: OK
+test_section2: OK
+test_section3: OK
+test_section4: OK
+test_spaces: OK
+test_wildcard: OK
+```

linter/pslint.py

@@ -0,0 +1,271 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-#
+#
+# PSL linter written in python
+#
+# Copyright 2016 Tim Rühsen (tim dot ruehsen at gmx dot de). All rights reserved.
+#
+# Permission is hereby granted, free of charge, to any person obtaining a
+# copy of this software and associated documentation files (the "Software"),
+# to deal in the Software without restriction, including without limitation
+# the rights to use, copy, modify, merge, publish, distribute, sublicense,
+# and/or sell copies of the Software, and to permit persons to whom the
+# Software is furnished to do so, subject to the following conditions:
+#
+# The above copyright notice and this permission notice shall be included in
+# all copies or substantial portions of the Software.
+#
+# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+# FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+# DEALINGS IN THE SOFTWARE.
+
+import sys
+import codecs
+
+nline = 0
+line = ""
+orig_line = ""
+warnings = 0
+errors = 0
+skip_order_check = False
+
+def warning(msg):
+	global warnings, orig_line, nline
+	print('%d: warning: %s%s' % (nline, msg, ": \'" + orig_line + "\'" if orig_line else ""))
+	warnings += 1
+
+def error(msg):
+	global errors, orig_line, nline
+	print('%d: error: %s%s' % (nline, msg, ": \'" + orig_line + "\'" if orig_line else ""))
+	errors += 1
+#	skip_order_check = True
+
+def print_psl(list):
+	for domain in list:
+		print(".".join(str(label) for label in reversed(domain)))
+
+def psl_key(s):
+	if s[0] == '*':
+		return 0
+	if s[0] == '!':
+		return 1
+	return 2
+
+def check_order(group):
+	"""Check the correct order of a domain group"""
+	global skip_order_check
+
+	try:
+		if skip_order_check or len(group) < 2:
+			skip_order_check = False
+			return
+
+		# check if the TLD is the identical within the group
+		if any(group[0][0] != labels[0] for labels in group):
+			warning('Domain group TLD is not consistent')
+
+		# sort by # of labels, label-by-label (labels are in reversed order)
+		sorted_group = sorted(group, key = lambda labels: (len(labels), psl_key(labels[-1][0]), labels))
+
+		if group != sorted_group:
+			warning('Incorrectly sorted group of domains')
+			print("  " + str(group))
+			print("  " + str(sorted_group))
+			print("Correct sorting would be:")
+			print_psl(sorted_group)
+
+	finally:
+		del group[:]
+
+
+def lint_psl(infile):
+	"""Parses PSL file and performs syntax checking"""
+	global orig_line, nline
+
+	PSL_FLAG_EXCEPTION = (1<<0)
+	PSL_FLAG_WILDCARD = (1<<1)
+	PSL_FLAG_ICANN = (1<<2) # entry of ICANN section
+	PSL_FLAG_PRIVATE = (1<<3) # entry of PRIVATE section
+	PSL_FLAG_PLAIN = (1<<4) #just used for PSL syntax checking
+
+	line2number = {}
+	line2flag = {}
+	group = []
+	section = 0
+	icann_sections = 0
+	private_sections = 0
+
+	lines = [line.strip('\n') for line in infile]
+
+	for line in lines:
+		nline += 1
+
+		# check for leadind/trailing whitespace
+		stripped = line.strip()
+		if stripped != line:
+			line = line.replace('\t','\\t')
+			line = line.replace('\r','^M')
+			orig_line = line
+			warning('Leading/Trailing whitespace')
+		orig_line = line
+		line = stripped
+
+		# empty line (end of sorted domain group)
+		if not line:
+			# check_order(group)
+			continue
+
+		# check for section begin/end
+		if line[0:2] == "//":
+			# check_order(group)
+
+			if section == 0:
+				if line == "// ===BEGIN ICANN DOMAINS===":
+					section = PSL_FLAG_ICANN
+					icann_sections += 1
+				elif line == "// ===BEGIN PRIVATE DOMAINS===":
+					section = PSL_FLAG_PRIVATE
+					private_sections += 1
+				elif line[3:11] == "===BEGIN":
+					error('Unexpected begin of unknown section')
+				elif line[3:9] == "===END":
+					error('End of section without previous begin')
+			elif section == PSL_FLAG_ICANN:
+				if line == "// ===END ICANN DOMAINS===":
+					section = 0
+				elif line[3:11] == "===BEGIN":
+					error('Unexpected begin of section: ')
+				elif line[3:9] == "===END":
+					error('Unexpected end of section')
+			elif section == PSL_FLAG_PRIVATE:
+				if line == "// ===END PRIVATE DOMAINS===":
+					section = 0
+				elif line[3:11] == "===BEGIN":
+					error('Unexpected begin of section')
+				elif line[3:9] == "===END":
+					error('Unexpected end of section')
+
+			continue # processing of comments ends here
+
+		# No rule must be outside of a section
+		if section == 0:
+			error('Rule outside of section')
+
+		group.append(list(reversed(line.split('.'))))
+
+		# decode UTF-8 input into unicode, needed only for python 2.x
+		try:
+			if sys.version_info[0] < 3:
+				line = line.decode('utf-8')
+			else:
+				line.encode('utf-8')
+		except (UnicodeDecodeError, UnicodeEncodeError):
+			orig_line = None
+			error('Invalid UTF-8 character')
+			continue
+
+		# each rule must be lowercase (or more exactly: not uppercase and not titlecase)
+		if line != line.lower():
+			error('Rule must be lowercase')
+
+		# strip leading wildcards
+		flags = section
+		# while line[0:2] == '*.':
+		if line[0:2] == '*.':
+			flags |= PSL_FLAG_WILDCARD
+			line = line[2:]
+
+		if line[0] == '!':
+			flags |= PSL_FLAG_EXCEPTION
+			line = line[1:]
+		else:
+			flags |= PSL_FLAG_PLAIN
+
+		# wildcard and exception must not combine
+		if flags & PSL_FLAG_WILDCARD and flags & PSL_FLAG_EXCEPTION:
+			error('Combination of wildcard and exception')
+			continue
+
+		labels = line.split('.')
+
+		if flags & PSL_FLAG_EXCEPTION and len(labels) > 1:
+			domain = ".".join(str(label) for label in labels[1:])
+			if not domain in line2flag:
+				error('Exception without previous wildcard')
+			elif not line2flag[domain] & PSL_FLAG_WILDCARD:
+				error('Exception without previous wildcard')
+
+		for label in labels:
+			if not label:
+				error('Leading/trailing or multiple dot')
+				continue
+
+			if label[0:4] == 'xn--':
+				error('Punycode found')
+				continue
+
+			if '--' in label:
+				error('Double minus found')
+				continue
+
+			# allowed are a-z,0-9,- and unicode >= 128 (maybe that can be finetuned a bit !?)
+			for c in label:
+				if not c.isalnum() and c != '-' and ord(c) < 128:
+					error('Illegal character')
+					break
+
+		if line in line2flag:
+			'''Found existing entry:
+			   Combination of exception and plain rule is contradictionary
+			     !foo.bar + foo.bar
+			   Doublette, since *.foo.bar implies foo.bar:
+			      foo.bar + *.foo.bar
+			   Allowed:
+			     !foo.bar + *.foo.bar
+			'''
+			error('Found doublette/ambiguity (previous line was %d)' % line2number[line])
+
+		line2number[line] = nline
+		line2flag[line] = flags
+
+	orig_line = None
+
+	if section == PSL_FLAG_ICANN:
+		error('ICANN section not closed')
+	elif section == PSL_FLAG_PRIVATE:
+		error('PRIVATE section not closed')
+
+	if icann_sections < 1:
+		warning('No ICANN section found')
+	elif icann_sections > 1:
+		warning('%d ICANN sections found' % icann_sections)
+
+	if private_sections < 1:
+		warning('No PRIVATE section found')
+	elif private_sections > 1:
+		warning('%d PRIVATE sections found' % private_sections)
+
+def usage():
+	"""Prints the usage"""
+	print('usage: %s PSLfile' % sys.argv[0])
+	print('or     %s -        # To read PSL from STDIN' % sys.argv[0])
+	exit(1)
+
+
+def main():
+	"""Check syntax of a PSL file"""
+	if len(sys.argv) < 2:
+		usage()
+
+	with sys.stdin if sys.argv[-1] == '-' else open(sys.argv[-1], 'r', encoding='utf-8', errors="surrogateescape") as infile:
+		lint_psl(infile)
+
+	return errors != 0
+
+
+if __name__ == '__main__':
+	sys.exit(main())

linter/pslint_selftest.sh

@@ -0,0 +1,31 @@
+#!/bin/sh
+
+rc=0
+rm -rf log
+mkdir -p log
+
+# add CR if missing, it won't possibly survive git
+sed -i -e 's/^e.example.com$/e.example.com\r/g' test_spaces.input
+
+for file in `ls *.input|cut -d'.' -f1`; do
+  echo -n "${file}: "
+  ./pslint.py ${file}.input >log/${file}.log 2>&1
+  diff -u ${file}.expected log/${file}.log >log/${file}.diff
+  if [ $? -eq 0 ]; then
+    echo OK
+    rm log/${file}.diff log/${file}.log
+  else
+    echo FAILED
+    cat log/${file}.diff
+    rc=1
+  fi
+done
+
+# remove CR, to not appear as changed to git
+sed -i -e 's/^e.example.com\r$/e.example.com/g' test_spaces.input
+
+if [ $rc -eq 0 ]; then
+  rmdir log
+fi
+
+exit $rc