bump.

Noticed while doing some testing that this was never reviewed/applied.

On 8/6/21 3:22 PM, Dylan Whyte wrote:
This prevents cases in which a string containing a percent character is
inadvertently utf-8 decoded before being displayed in notes.

Signed-off-by: Dylan Whyte <[email protected]>
---
  src/PVE/Tools.pm | 4 ++--
  1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/src/PVE/Tools.pm b/src/PVE/Tools.pm
index 807bc03..b8d6dc9 100644
--- a/src/PVE/Tools.pm
+++ b/src/PVE/Tools.pm
@@ -1200,8 +1200,8 @@ sub upid_normalize_status_type {
  sub encode_text {
      my ($text) = @_;
- # all control and hi-bit characters, and ':'
-    my $unsafe = "^\x20-\x39\x3b-\x7e";
+    # all control and hi-bit characters, ':', and '%'
+    my $unsafe = "\x00-\x1f\x25\x3a\x7f-\xff";
      return uri_escape(Encode::encode("utf8", $text), $unsafe);
  }


_______________________________________________
pve-devel mailing list
[email protected]
https://lists.proxmox.com/cgi-bin/mailman/listinfo/pve-devel

Reply via email to