From 34d5fe41a9fa05f6f2fd306b728483c880222ae9 Mon Sep 17 00:00:00 2001
From: Bertrand Bonnefoy-Claudet <bertrand@bertrandbc.com>
Date: Wed, 9 Sep 2020 11:00:13 +0200
Subject: [PATCH] Fix parsing of unquoted values with two spaces

If a value is unquoted and has two or more adjacent spaces (like in `a=b
c`), the parser would detect an error.  This commit fixes that.

Tabs and other whitespace characters are now also considered like space
characters in this case and I added relevant test cases.
---
 CHANGELOG.md         |  5 ++++-
 src/dotenv/parser.py | 12 +++---------
 tests/test_parser.py | 36 ++++++++++++++++++++++++++++--------
 3 files changed, 35 insertions(+), 18 deletions(-)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index 34cdb324..b5305f19 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -19,8 +19,10 @@ project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
 ### Fixed
 
-- Fix potentially empty expanded value for duplicate key (#260 by [@bbc]).
+- Fix potentially empty expanded value for duplicate key (#260 by [@bbc2]).
 - Fix import error on Python 3.5.0 and 3.5.1 (#267 by [@gongqingkui]).
+- Fix parsing of unquoted values containing several adjacent space or tab characters
+  (#277 by [@bbc2], review by [@x-yuri]).
 
 ## [0.14.0] - 2020-07-03
 
@@ -226,6 +228,7 @@ project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 [@theskumar]: https://github.com/theskumar
 [@ulyssessouza]: https://github.com/ulyssessouza
 [@venthur]: https://github.com/venthur
+[@x-yuri]: https://github.com/x-yuri
 [@yannham]: https://github.com/yannham
 
 [Unreleased]: https://github.com/theskumar/python-dotenv/compare/v0.14.0...HEAD
diff --git a/src/dotenv/parser.py b/src/dotenv/parser.py
index 4eba0ac4..5cb1cdfa 100644
--- a/src/dotenv/parser.py
+++ b/src/dotenv/parser.py
@@ -24,7 +24,7 @@ def make_regex(string, extra_flags=0):
 _equal_sign = make_regex(r"(=[^\S\r\n]*)")
 _single_quoted_value = make_regex(r"'((?:\\'|[^'])*)'")
 _double_quoted_value = make_regex(r'"((?:\\"|[^"])*)"')
-_unquoted_value_part = make_regex(r"([^ \r\n]*)")
+_unquoted_value = make_regex(r"([^\r\n]*)")
 _comment = make_regex(r"(?:[^\S\r\n]*#[^\r\n]*)?")
 _end_of_line = make_regex(r"[^\S\r\n]*(?:\r\n|\n|\r|$)")
 _rest_of_line = make_regex(r"[^\r\n]*(?:\r|\n|\r\n)?")
@@ -167,14 +167,8 @@ def parse_key(reader):
 
 def parse_unquoted_value(reader):
     # type: (Reader) -> Text
-    value = u""
-    while True:
-        (part,) = reader.read_regex(_unquoted_value_part)
-        value += part
-        after = reader.peek(2)
-        if len(after) < 2 or after[0] in u"\r\n" or after[1] in u" #\r\n":
-            return value
-        value += reader.read(2)
+    (part,) = reader.read_regex(_unquoted_value)
+    return re.sub(r"\s+#.*", "", part).rstrip()
 
 
 def parse_value(reader):
diff --git a/tests/test_parser.py b/tests/test_parser.py
index f8075138..48cecdce 100644
--- a/tests/test_parser.py
+++ b/tests/test_parser.py
@@ -19,20 +19,40 @@
     (u"# a=b", [Binding(key=None, value=None, original=Original(string=u"# a=b", line=1), error=False)]),
     (u"a=b#c", [Binding(key=u"a", value=u"b#c", original=Original(string=u"a=b#c", line=1), error=False)]),
     (
-        u'a=b # comment',
-        [Binding(key=u"a", value=u"b", original=Original(string=u"a=b # comment", line=1), error=False)],
+        u'a=b #c',
+        [Binding(key=u"a", value=u"b", original=Original(string=u"a=b #c", line=1), error=False)],
     ),
     (
-        u"a=b space ",
-        [Binding(key=u"a", value=u"b space", original=Original(string=u"a=b space ", line=1), error=False)],
+        u'a=b\t#c',
+        [Binding(key=u"a", value=u"b", original=Original(string=u"a=b\t#c", line=1), error=False)],
     ),
     (
-        u"a='b space '",
-        [Binding(key=u"a", value=u"b space ", original=Original(string=u"a='b space '", line=1), error=False)],
+        u"a=b c",
+        [Binding(key=u"a", value=u"b c", original=Original(string=u"a=b c", line=1), error=False)],
     ),
     (
-        u'a="b space "',
-        [Binding(key=u"a", value=u"b space ", original=Original(string=u'a="b space "', line=1), error=False)],
+        u"a=b\tc",
+        [Binding(key=u"a", value=u"b\tc", original=Original(string=u"a=b\tc", line=1), error=False)],
+    ),
+    (
+        u"a=b  c",
+        [Binding(key=u"a", value=u"b  c", original=Original(string=u"a=b  c", line=1), error=False)],
+    ),
+    (
+        u"a=b\u00a0 c",
+        [Binding(key=u"a", value=u"b\u00a0 c", original=Original(string=u"a=b\u00a0 c", line=1), error=False)],
+    ),
+    (
+        u"a=b c ",
+        [Binding(key=u"a", value=u"b c", original=Original(string=u"a=b c ", line=1), error=False)],
+    ),
+    (
+        u"a='b c '",
+        [Binding(key=u"a", value=u"b c ", original=Original(string=u"a='b c '", line=1), error=False)],
+    ),
+    (
+        u'a="b c "',
+        [Binding(key=u"a", value=u"b c ", original=Original(string=u'a="b c "', line=1), error=False)],
     ),
     (
         u"export export_a=1",