From f24979bc096601d16ff69bd85eb74bee79129e42 Mon Sep 17 00:00:00 2001
From: Thomas Waldmann <tw@waldmann-edv.de>
Date: Sun, 3 Apr 2022 16:49:28 +0200
Subject: [PATCH 1/2] fix scp repo url parsing for ip v6 addrs, fixes #6526

added a negative lookahead/lookbehind to make sure an ipv6 addr
(enclosed in square brackets) does not get badly matched by the
regex part intended for hostnames and ipv4 addrs only.

the other part of that regex which is actually intended to match
ipv6 addrs only matches if they are enclosed in square brackets.

also added tests for ssh and scp style repo URLs with ipv6 addrs
in brackets.

also: made regex more readable, putting these 2 cases on separate lines.
---
 src/borg/helpers/parseformat.py | 12 ++++++++----
 src/borg/testsuite/helpers.py   |  6 ++++++
 2 files changed, 14 insertions(+), 4 deletions(-)

diff --git a/src/borg/helpers/parseformat.py b/src/borg/helpers/parseformat.py
index f595c8d39..39be4a544 100644
--- a/src/borg/helpers/parseformat.py
+++ b/src/borg/helpers/parseformat.py
@@ -355,10 +355,14 @@ class Location:
     # note: scp_re is also use for local paths
     scp_re = re.compile(r"""
         (
-            """ + optional_user_re + r"""                   # user@  (optional)
-            (?P<host>([^:/]+|\[[0-9a-fA-F:.]+\])):          # host: (don't match / or [ipv6] in host to disambiguate from file:)
-        )?                                                  # user@host: part is optional
-        """ + scp_path_re + optional_archive_re, re.VERBOSE)  # path with optional archive
+            """ + optional_user_re + r"""                           # user@  (optional)
+            (?P<host>(                  # host can be ipv6 addr in brackets or something else
+                (?!\[)[^:/]+(?<!\])     # host: hostname, not containing : or /, also not ipv6: not in brackets
+                |
+                \[[0-9a-fA-F:.]+\])     # host: ipv6 addr in brackets
+            ):
+        )?                                                          # user@host: part is optional
+        """ + scp_path_re + optional_archive_re, re.VERBOSE)        # path with optional archive
 
     # get the repo from BORG_REPO env and the optional archive from param.
     # if the syntax requires giving REPOSITORY (see "borg mount"),
diff --git a/src/borg/testsuite/helpers.py b/src/borg/testsuite/helpers.py
index 1dd66d02e..8014101a5 100644
--- a/src/borg/testsuite/helpers.py
+++ b/src/borg/testsuite/helpers.py
@@ -100,6 +100,10 @@ class TestLocationWithoutEnv:
         assert repr(Location('ssh://user@[2001:db8::192.0.2.1]/some/path')) == \
             "Location(proto='ssh', user='user', host='2001:db8::192.0.2.1', port=None, path='/some/path', archive=None)"
         assert Location('ssh://user@[2001:db8::192.0.2.1]/some/path').to_key_filename() == keys_dir + '2001_db8__192_0_2_1__some_path'
+        assert repr(Location('ssh://user@[2a02:0001:0002:0003:0004:0005:0006:0007]/some/path')) == \
+            "Location(proto='ssh', user='user', host='2a02:0001:0002:0003:0004:0005:0006:0007', port=None, path='/some/path', archive=None)"
+        assert repr(Location('ssh://user@[2a02:0001:0002:0003:0004:0005:0006:0007]:1234/some/path')) == \
+            "Location(proto='ssh', user='user', host='2a02:0001:0002:0003:0004:0005:0006:0007', port=1234, path='/some/path', archive=None)"
 
     def test_file(self, monkeypatch, keys_dir):
         monkeypatch.delenv('BORG_REPO', raising=False)
@@ -132,6 +136,8 @@ class TestLocationWithoutEnv:
         assert repr(Location('user@[2001:db8::192.0.2.1]:/some/path')) == \
             "Location(proto='ssh', user='user', host='2001:db8::192.0.2.1', port=None, path='/some/path', archive=None)"
         assert Location('user@[2001:db8::192.0.2.1]:/some/path').to_key_filename() == keys_dir + '2001_db8__192_0_2_1__some_path'
+        assert repr(Location('user@[2a02:0001:0002:0003:0004:0005:0006:0007]:/some/path')) == \
+            "Location(proto='ssh', user='user', host='2a02:0001:0002:0003:0004:0005:0006:0007', port=None, path='/some/path', archive=None)"
 
     def test_smb(self, monkeypatch, keys_dir):
         monkeypatch.delenv('BORG_REPO', raising=False)

From b2eaa2fba46cc12a5bcbebe0a23f3701c893de5b Mon Sep 17 00:00:00 2001
From: Thomas Waldmann <tw@waldmann-edv.de>
Date: Sun, 3 Apr 2022 19:01:11 +0200
Subject: [PATCH 2/2] use same host regex for ssh and scp style, refactor/clean
 up

although bug #6526 did not show with ssh style URLs, we should
not have different regexes for the host part for ssh and scp style.

thus i extracted the host_re from both and also cleaned up a bit.
---
 src/borg/helpers/parseformat.py | 35 +++++++++++++++++++--------------
 1 file changed, 20 insertions(+), 15 deletions(-)

diff --git a/src/borg/helpers/parseformat.py b/src/borg/helpers/parseformat.py
index 39be4a544..08ec8df8e 100644
--- a/src/borg/helpers/parseformat.py
+++ b/src/borg/helpers/parseformat.py
@@ -341,28 +341,33 @@ class Location:
             (?P<archive>[^/]+)                              # archive name must not contain "/"
         )?$"""                                              # must match until the end
 
+    # host NAME, or host IP ADDRESS (v4 or v6, v6 must be in square brackets)
+    host_re = r"""
+        (?P<host>(
+            (?!\[)[^:/]+(?<!\])     # hostname or v4 addr, not containing : or / (does not match v6 addr: no brackets!)
+            |
+            \[[0-9a-fA-F:.]+\])     # ipv6 address in brackets
+        )
+    """
+
     # regexes for misc. kinds of supported location specifiers:
     ssh_re = re.compile(r"""
-        (?P<proto>ssh)://                                   # ssh://
-        """ + optional_user_re + r"""                       # user@  (optional)
-        (?P<host>([^:/]+|\[[0-9a-fA-F:.]+\]))(?::(?P<port>\d+))?  # host or host:port or [ipv6] or [ipv6]:port
-        """ + abs_path_re + optional_archive_re, re.VERBOSE)  # path or path::archive
+        (?P<proto>ssh)://                                       # ssh://
+        """ + optional_user_re + host_re + r"""                 # user@  (optional), host name or address
+        (?::(?P<port>\d+))?                                     # :port (optional)
+        """ + abs_path_re + optional_archive_re, re.VERBOSE)    # path or path::archive
 
     file_re = re.compile(r"""
-        (?P<proto>file)://                                  # file://
-        """ + file_path_re + optional_archive_re, re.VERBOSE)  # servername/path, path or path::archive
+        (?P<proto>file)://                                      # file://
+        """ + file_path_re + optional_archive_re, re.VERBOSE)   # servername/path, path or path::archive
 
-    # note: scp_re is also use for local paths
+    # note: scp_re is also used for local paths
     scp_re = re.compile(r"""
         (
-            """ + optional_user_re + r"""                           # user@  (optional)
-            (?P<host>(                  # host can be ipv6 addr in brackets or something else
-                (?!\[)[^:/]+(?<!\])     # host: hostname, not containing : or /, also not ipv6: not in brackets
-                |
-                \[[0-9a-fA-F:.]+\])     # host: ipv6 addr in brackets
-            ):
-        )?                                                          # user@host: part is optional
-        """ + scp_path_re + optional_archive_re, re.VERBOSE)        # path with optional archive
+            """ + optional_user_re + host_re + r"""             # user@  (optional), host name or address
+            :                                                   # : (required!)
+        )?                                                      # user@host: part is optional
+        """ + scp_path_re + optional_archive_re, re.VERBOSE)    # path with optional archive
 
     # get the repo from BORG_REPO env and the optional archive from param.
     # if the syntax requires giving REPOSITORY (see "borg mount"),