Add test for locations of regexp terms.

Author: Max Schaefer

python/ql/test/library-tests/regexparser/Locations.expected

@@ -0,0 +1,20 @@
+| locations.py | 14 | 5 |
+| locations.py | 19 | 5 |
+| locations.py | 24 | 5 |
+| locations.py | 29 | 5 |
+| locations.py | 34 | 5 |
+| locations.py | 39 | 5 |
+| locations.py | 44 | 5 |
+| locations.py | 49 | 5 |
+| locations.py | 54 | 5 |
+| locations.py | 54 | 26 |
+| locations.py | 59 | 5 |
+| locations.py | 59 | 26 |
+| locations.py | 65 | 5 |
+| locations.py | 65 | 26 |
+| locations.py | 72 | 6 |
+| locations.py | 72 | 27 |
+| locations.py | 80 | 6 |
+| locations.py | 85 | 7 |
+| locations.py | 90 | 5 |
+| locations.py | 90 | 26 |

python/ql/test/library-tests/regexparser/Locations.ql

@@ -0,0 +1,7 @@
+import semmle.python.regexp.RegexTreeView::RegexTreeView
+
+from RegExpTerm t, string file, int line, int column
+where
+  t.toString() = "[this]" and
+  t.hasLocationInfo(file, line, column, _, _)
+select file, line, column

python/ql/test/library-tests/regexparser/locations.py

@@ -0,0 +1,92 @@
+import re
+
+# This file contains tests for the regexp parser's location tracking.
+
+# To keep the expected results manageable, we only test the locations of the
+# regexp term `[this]`, appearing in various kinds of regexps.
+#
+# To make the location information easier to understand, we generally put each
+# regexp on its own line, even though this is not the way one would normally
+# write regexps in Python.
+
+# plain string
+re.compile(
+'[this] is a test'
+)
+
+# raw string
+re.compile(
+r'[this] is a test'
+)
+
+# byte string
+re.compile(
+b'[this] is a test'
+)
+
+# byte raw string
+re.compile(
+br'[this] is a test'
+)
+
+# multiline string
+re.compile(
+'''[this] is a test'''
+)
+
+# multiline raw string
+re.compile(
+r'''[this] is a test'''
+)
+
+# multiline byte string
+re.compile(
+b'''[this] is a test'''
+)
+
+# multiline byte raw string
+re.compile(
+br'''[this] is a test'''
+)
+
+# plain string with multiple parts
+re.compile(
+'[this] is a test' ' and [this] is another test'
+)
+
+# plain string with multiple parts across lines
+re.compile(
+'[this] is a test'
+' and [this] is another test'
+)
+
+# plain string with multiple parts across lines and comments
+re.compile(
+'[this] is a test'
+# comment
+' and [this] is another test'
+)
+
+# actual multiline string
+re.compile(
+r'''
+[this] is a test
+and [this] is another test
+'''
+)
+
+# plain string with escape sequences
+re.compile(
+'\t[this] is a test'
+)
+
+# raw string with escape sequences
+re.compile(
+r'\A[this] is a test'
+)
+
+# plain string with escaped newline
+re.compile(
+'[this] is a test\
+ and [this] is another test'
+)