profile
viewpoint

vuori/weasel-pageant 256

Deprecated: An ssh-agent compatible helper for interacting with Pageant from processes running on the Windows Subsystem for Linux.

vuori/linux-minidisc 32

Free software for accessing MiniDisc devices (fork with NetMD fixes—use upstream for HiMD devices)

vuori/xsecurelock 1

X11 screen lock utility with security in mind

vuori/aws-mfa 0

Manage AWS MFA Security Credentials

vuori/pupflare 0

A webpage proxy that request through puppeteer - can be used to bypass cloudflare anti bot / anti ddos on a non browser application (like curl)

vuori/transmission 0

Transmission fork with ipv6 fixes

vuori/xwinrestore 0

Remember X11 window positions per display configuration

issue closedneomutt/neomutt

next-unread-mailbox only works every mail_check_stats_interval seconds

<!-- Sorry you've found a bug, but thanks for reporting it here. Please try to give as much information as you can. -->

Expected Behaviour

next-unread-mailbox should go to next unread mailbox. <!-- What were you trying to do? -->

Actual Behaviour

<!-- What went wrong? -->

If called more often than mail_check_stats_interval seconds, function says that there are no unread mailboxes.

Steps to Reproduce

<!-- As much detail as possible, please -->

  1. Wait for mail to arrive in a mailbox which is not the current one.
  2. Press key bount to next-unread-mailbox, note that it will usually work once.
  3. Move to another mailbox (without reading the new mail), try it again.
  4. "No new messages" will be printed.

How often does this happen?

  • Always

When did it start to happen?

Switched to master from Ubuntu package (20191207+dfsg.1-1.1) (maybe, I'm not sure if I tried to use the function when using the packaged release).

NeoMutt Version

NeoMutt 20210205-804-7958fd-dirty
Copyright (C) 1996-2020 Michael R. Elkins and others.
NeoMutt comes with ABSOLUTELY NO WARRANTY; for details type 'neomutt -vv'.
NeoMutt is free software, and you are welcome to redistribute it
under certain conditions; type 'neomutt -vv' for details.

System: Linux 5.10.60+ (x86_64)
ncurses: ncurses 6.2.20200212 (compiled with 6.2.20200212)
libidn: 1.33 (compiled with 1.33)
GPGME: 1.13.1-unknown
OpenSSL: OpenSSL 1.1.1f  31 Mar 2020
libnotmuch: 5.4.0
PCRE2: 10.34 2019-11-21
storage: kyotocabinet
compression: lz4, zlib

Mail check settings:

set mail_check = 5
set mail_check_recent = yes
set mail_check_stats = yes
set mail_check_stats_interval = 60

Extra Info

Using an all-Maildir configuration.

The problem is that mutt_mailbox_check() sets MailboxCount to 0 even in cases when it's not going to actually check stats on mailboxes due to mail_check_stats_interval not having expired, except on the first time it sees a given mailbox.

Diff below fixes the immediate problem, though it may lead to MailboxCount being wonky after a new mailbox is added. In practice first_check_stats_done being false after the first call to the function sounds rare, but I think a canonical fix would require always either checking stats on all mailboxes or checking none of them. Basically first_check_stats_done would need to be checked in a separate pass and check_stats set to true if at least one mailbox needed a check I guess.

diff --git a/mutt_mailbox.c b/mutt_mailbox.c
index ac40682af..7e32706b0 100644
--- a/mutt_mailbox.c
+++ b/mutt_mailbox.c
@@ -191,10 +191,10 @@ int mutt_mailbox_check(struct Mailbox *m_cur, int force)
   {
     check_stats = true;
     MailboxStatsTime = t;
+    MailboxCount = 0;
   }
 
   MailboxTime = t;
-  MailboxCount = 0;
   MailboxNotify = 0;
 
   /* check device ID and serial number instead of comparing paths */

Addendum: I didn't test it but this looks like a regression caused by 0bef239d6e04316ab151bc34ea0a10750b70cfc7. Maybe the check on check_stats should be around just the mx_mbox_check_stats() call, i.e. if !check_stats then just look at m_check->has_new when deciding whether to increment MailboxCount? The different paths tatken around here by the mail_check_stats setting make things pretty tricky.

closed time in 24 days

vuori

issue commentneomutt/neomutt

next-unread-mailbox only works every mail_check_stats_interval seconds

Tested current master and the issue reported here seems to be fixed. Seeing at least some of the other wonkiness reported in https://github.com/neomutt/neomutt/issues/3072 with Maildir mailboxes, but anyway I'll close this since next-unread-mailbox now works.

vuori

comment created time in 24 days

issue commentjelmer/xandikos

Exception spew with D:allprop request on calendar collection

There's a typo in the commit, get_xmpp_url should be get_xmpp_uri to fix the exceptions.

vuori

comment created time in a month

issue commentjelmer/xandikos

Exception spew with D:allprop request on calendar collection

Corollary: <D:propname/> does not work at all:

ERROR:aiohttp.server:Error handling request
Traceback (most recent call last):
  File "/home/user/src/xandikos/.venv/lib/python3.8/site-packages/aiohttp/web_protocol.py", line 422, in _handle_request
    resp = await self._request_handler(request)
  File "/home/user/src/xandikos/.venv/lib/python3.8/site-packages/aiohttp/web_app.py", line 499, in _handle
    resp = await handler(request)
  File "/home/user/src/xandikos/.venv/lib/python3.8/site-packages/xandikos/web.py", line 1290, in xandikos_handler
    return await main_app.aiohttp_handler(request, options.route_prefix)
  File "/home/user/src/xandikos/.venv/lib/python3.8/site-packages/xandikos/webdav.py", line 2070, in aiohttp_handler
    response = await self._handle_request(request, environ)
  File "/home/user/src/xandikos/.venv/lib/python3.8/site-packages/xandikos/webdav.py", line 2028, in _handle_request
    return await do.handle(request, environ, self)
  File "/home/user/src/xandikos/.venv/lib/python3.8/site-packages/xandikos/webdav.py", line 344, in wrapper
    async for resp in req_fn(self, environ, *args, **kwargs):
  File "/home/user/src/xandikos/.venv/lib/python3.8/site-packages/xandikos/webdav.py", line 1756, in handle
    yield Status(href, "200 OK", propstat=[s async for s in propstat])
  File "/home/user/src/xandikos/.venv/lib/python3.8/site-packages/xandikos/webdav.py", line 1756, in <listcomp>
    yield Status(href, "200 OK", propstat=[s async for s in propstat])
  File "/home/user/src/xandikos/.venv/lib/python3.8/site-packages/xandikos/webdav.py", line 1101, in get_property_names
    if await prop.is_set(href, resource, environ):
  File "/home/user/src/xandikos/.venv/lib/python3.8/site-packages/xandikos/webdav.py", line 559, in is_set
    await self.get_value("/", resource, ET.Element(self.name), environ)
  File "/home/user/src/xandikos/.venv/lib/python3.8/site-packages/xandikos/xmpp.py", line 40, in get_value
    el.text = resource.get_xmpp_uri()
  File "/home/user/src/xandikos/.venv/lib/python3.8/site-packages/xandikos/caldav.py", line 175, in get_xmpp_uri
    raise NotImplementedError(self.get_xmpp_uri)
NotImplementedError: <bound method Calendar.get_xmpp_uri of CalendarCollection(TreeGitStore(<Repo at '/home/user/xandikos_data/user/calendars/calendar'>, ref=b'refs/heads/master'))>
INFO:aiohttp.access:127.0.0.1 [24/Oct/2021:19:09:14 +0000] "PROPFIND /user/calendars/calendar/ HTTP/1.1" 500 250 "-" "curl/7.68.0"
vuori

comment created time in a month

issue openedjelmer/xandikos

Exception spew with D:allprop request on calendar collection

I was testing some stuff with curl and noticed that requesting D:allprop (or no body at all to FINDPROP) on a calendar collection spews a bunch of exceptions regarding unimplemented XMPP properties in the log. This has no effect on the result, but makes the logs quite noisy. Tested with latest master.

To reproduce: curl -u user:pass -X PROPFIND -H "Depth: 0" https://example.com/user/calendars/calendar/

Log:

ERROR:root:Not implemented while getting {http://calendarserver.org/ns/}xmpp-uri for CalendarCollection(TreeGitStore(<Repo at '/home/user/xandikos_data/user/calendars/calendar'>, ref=b'refs/heads/master'))
Traceback (most recent call last):
  File "/home/user/src/xandikos/.venv/lib/python3.8/site-packages/xandikos/webdav.py", line 1043, in get_property_from_element
    get_value_ext = prop.get_value_ext  # type: ignore
AttributeError: 'XmppUriProperty' object has no attribute 'get_value_ext'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/home/user/src/xandikos/.venv/lib/python3.8/site-packages/xandikos/webdav.py", line 1045, in get_property_from_element
    await prop.get_value(href, resource, ret, environ)
  File "/home/user/src/xandikos/.venv/lib/python3.8/site-packages/xandikos/xmpp.py", line 40, in get_value
    el.text = resource.get_xmpp_uri()
  File "/home/user/src/xandikos/.venv/lib/python3.8/site-packages/xandikos/caldav.py", line 175, in get_xmpp_uri
    raise NotImplementedError(self.get_xmpp_uri)
NotImplementedError: <bound method Calendar.get_xmpp_uri of CalendarCollection(TreeGitStore(<Repo at '/home/user/xandikos_data/user/calendars/calendar'>, ref=b'refs/heads/master'))>
ERROR:root:Not implemented while getting {http://calendarserver.org/ns/}xmpp-server for CalendarCollection(TreeGitStore(<Repo at '/home/user/xandikos_data/user/calendars/calendar'>, ref=b'refs/heads/master'))
Traceback (most recent call last):
  File "/home/user/src/xandikos/.venv/lib/python3.8/site-packages/xandikos/webdav.py", line 1043, in get_property_from_element
    get_value_ext = prop.get_value_ext  # type: ignore
AttributeError: 'XmppServerProperty' object has no attribute 'get_value_ext'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/home/user/src/xandikos/.venv/lib/python3.8/site-packages/xandikos/webdav.py", line 1045, in get_property_from_element
    await prop.get_value(href, resource, ret, environ)
  File "/home/user/src/xandikos/.venv/lib/python3.8/site-packages/xandikos/xmpp.py", line 76, in get_value
    server = resource.get_xmpp_server()
  File "/home/user/src/xandikos/.venv/lib/python3.8/site-packages/xandikos/caldav.py", line 169, in get_xmpp_server
    raise NotImplementedError(self.get_xmpp_server)
NotImplementedError: <bound method Calendar.get_xmpp_server of CalendarCollection(TreeGitStore(<Repo at '/home/user/xandikos_data/user/calendars/calendar'>, ref=b'refs/heads/master'))>
ERROR:root:Not implemented while getting {http://calendarserver.org/ns/}xmpp-heartbeat for CalendarCollection(TreeGitStore(<Repo at '/home/user/xandikos_data/user/calendars/calendar'>, ref=b'refs/heads/master'))
Traceback (most recent call last):
  File "/home/user/src/xandikos/.venv/lib/python3.8/site-packages/xandikos/webdav.py", line 1043, in get_property_from_element
    get_value_ext = prop.get_value_ext  # type: ignore
AttributeError: 'XmppHeartbeatProperty' object has no attribute 'get_value_ext'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/home/user/src/xandikos/.venv/lib/python3.8/site-packages/xandikos/webdav.py", line 1045, in get_property_from_element
    await prop.get_value(href, resource, ret, environ)
  File "/home/user/src/xandikos/.venv/lib/python3.8/site-packages/xandikos/xmpp.py", line 55, in get_value
    (uri, minutes) = resource.get_xmpp_heartbeat()
  File "/home/user/src/xandikos/.venv/lib/python3.8/site-packages/xandikos/caldav.py", line 172, in get_xmpp_heartbeat
    raise NotImplementedError(self.get_xmpp_heartbeat)
NotImplementedError: <bound method Calendar.get_xmpp_heartbeat of CalendarCollection(TreeGitStore(<Repo at '/home/user/xandikos_data/user/calendars/calendar'>, ref=b'refs/heads/master'))>
INFO:aiohttp.access:127.0.0.1 [24/Oct/2021:18:52:35 +0000] "PROPFIND /user/calendars/calendar/ HTTP/1.1" 207 2349 "-" "curl/7.68.0"

created time in a month

issue commentpimutils/khal

Feature: VJOURNAL support

Some functions in new/edit should be disabled. VALARM is not allowed, but attendees can be added (supposedly for indicating who was present at a meeting that the journal entry applies to). They should not get invitations though. I don't see it in the current UI, but free/busy and opaque/transparent selection are also not applicable.

Recurrences are technically allowed, though it's hard to imagine why you'd want a recurring journal entry. Maybe to use as a template for recording some kind of weekly meeting? Only IETF knows I guess. They would probably be better off disabled.

Usually a journal will have a long description, so a key binding to invoke an external editor to edit either single fields or just a summary+description (separated by newline) combination seems desirable.

Anyway I haven't even looked at khal's code yet so I guess I'll get to that soonish.

vuori

comment created time in a month

issue commentpimutils/khal

Feature: VJOURNAL support

As precedent, KOrganizer (which has pretty advanced support) shows journal entries in calendar/agenda with a notebook icon. Orage only shows them in a separate listing of journal entries, though that's a sub-view of the selected date's details. There seems to be little support elsewhere.

One alternative I was thinking of was a three-way flag -J/--journal (with a configurable default) for at/calendar/list/search and ikhal: "yes" to show both VEVENT and VJOURNAL (default), "no" for VEVENT only, "only" for VJOURNAL only.

The other way would be to have --journal be a yes/no boolean for existing subcommands plus a separate "journal" subcommand to show a list of journal entries. While this is a bit non-orthogonal, it would allow having a separate default time range: seeing journals for the last year or month seems like the most useful default view (you rarely care about past VEVENTs while you rarely have future VJOURNALs, so existing calendar/list default ranges are not too useful).

From a technical perspective, I was going to integrate this with khal because VJOURNAL is almost exactly a VEVENT with just a few different defaults. It's also almost the same conceptually: a summary and description attached to a date (unlike a VTODO which is conceptually quite different, since todos involve all sorts of priority and completion status bits and may not have dates at all). Implementing journals as a whole separate project would involve quite a bit of duplicate work in both UI and internals.

vuori

comment created time in a month

issue commentpimutils/khal

Feature: VJOURNAL support

"What I did today" journal/diary type entries in a more verbose and unstructured format than what VTODOs allow for. Basically what RFC 5545 § 3.6.3 suggests: "descriptive text notes associated with a particular calendar date."

I have been using a text file for this, but having journal entries show up in my calendar would be nice.

vuori

comment created time in a month

issue openedpimutils/khal

Feature: VJOURNAL support

I've been thinking about adding support for VJOURNAL entries to khal. AFAICT these are almost the same as VEVENT, the main difference being that they are implicitly FREE and TRANSPARENT. So it seems that this would mostly need a few UI tweaks to show/hide journal entries in agenda views and maybe allow editing long DESCRIPTIONs in an external editor.

Is there interest in having this in mainline?

(Since many CalDAV servers don't allow pushing VJOURNALs by default a nice extra would be to have a setting to create "compatible" journal entries which are VEVENTs with an X-flag to indicate that these are actually journal entries. KOrganizer has a pretty nice UI for VJOURNALs so there's some benefit to having real ones though.)

created time in a month

pull request commenttransmission/transmission

Wrong IPv6 address used in announces

I can't easily test the original problem scenario any more, but with light testing this version seems to be working otherwise.

vuori

comment created time in 2 months

push eventvuori/transmission

geertjan

commit sha 23e469a6b2e253dc54c479a98447d8bdbc799be8

Fix setting the ETA timestamps Field etaDLSpeedCalculatedAt was set too early, causing the condition following it to always be false. The same for etaULSpeedCalculatedAt.

view details

Mike Gelfand

commit sha cefd8a314962bd919c162a5eb96e88eec9994f15

Bump version to 2.93+, fix NEWS date

view details

Mike Gelfand

commit sha 4d5fe6ede92b8c22e8a022b91648dbf2c84ce1cb

Add support for mbedtls (formely polarssl)

view details

Mike Gilbert

commit sha 357cccb9cf87e9c1808df4a7c48f600d7fc9f204

cmake: link against libmbedcrypto if available In recent versions of mbed TLS, several symbols are moved to libmbedcrypto. Fixes: https://github.com/transmission/transmission/issues/115

view details

Mike Gelfand

commit sha b21a6140e963491f146293c9b0a2863a9ae99ad3

Fix FTCBFS due to AC_RUN_IFELSE (patch by Helmut Grohne) Fixes: #475

view details

dubhater

commit sha e19ab0b7a9139da65e502520551deabee8452118

Qt: Fix bad downloaded percentage in DetailsDialog It was always 0.0% as long as the torrent was not finished.

view details

Bernard Spil

commit sha bdffe25b42d710c9a2dff54ec895772100f6e1d5

Fix build with LibreSSL 2.7 LibreSSL 2.7 adds OpenSSL 1.1 API See also: https://bugs.freebsd.org/226953 Signed-off-by: Bernard Spil <brnrd@FreeBSD.org>

view details

Mike Gelfand

commit sha 11f1d46e1b24aeb10125158771ca313b31665715

Merge pull request #570 from Sp1l/master Fix build with LibreSSL 2.7

view details

Mike Gelfand

commit sha 9d96f87ec3b8a531b176644fc93ad6fec1d6c9eb

Fix coding style and building with !TR_LIGHTWEIGHT

view details

Niklas Haas

commit sha 11d29ff62d528319f9b9596fd746dd7795aff140

LibreSSL compatibility This seems to be the idiomatic way to fix libressl compatibility issues, judging by what most other open source projects seem to be doing. I've confirmed that transmission builds with libressl for me after this patch is applied.

view details

Bernard Spil

commit sha 06a1644e298afc948945e44b463348944988be03

Fix build with LibreSSL 2.7 LibreSSL 2.7 adds OpenSSL 1.1 API See also: https://bugs.freebsd.org/226953 Signed-off-by: Bernard Spil <brnrd@FreeBSD.org>

view details

Mike Gelfand

commit sha ab4d7073053d694b91d5c59ddac596fe5aa233e6

Merge pull request #528 from neheb/patches Fix compile errors for 2.9x

view details

Mike Gelfand

commit sha 0b5ec3a00da2013cba3dd5ddcf9811e17eae9cf8

Merge pull request #547 from dubhater/patch-6 Qt: Fix bad downloaded percentage in DetailsDialog

view details

dubhater

commit sha 85c0399d8c93b85bcb75dfc86b2b69428d0f2ad6

Qt: Fix bad downloaded percentage in DetailsDialog It was always 0.0% as long as the torrent was not finished.

view details

Mike Gelfand

commit sha 444f15edf82ca85d9a1691a9b3a615af10848328

Merge pull request #522 from gjalsem/master Fix setting the ETA timestamps

view details

geertjan

commit sha cc0b93143191d4b1648d8c990b2079886d745d13

Fix setting the ETA timestamps Field etaDLSpeedCalculatedAt was set too early, causing the condition following it to always be false. The same for etaULSpeedCalculatedAt.

view details

Mike Gelfand

commit sha 69d9afd75a64fb7ebaec0d45e348abe5f9cc1ec4

Adjust CFBundleVersion format to fix Sparkle update check Set the first version component to be the same as the last Subversion-based release build version and add two more components (major and minor version numbers). To allow for nightly build updates this should probably include another component (e.g. build timestamp), but we're not there yet.

view details

Mike Gelfand

commit sha 523d8c1317725143556d771d960cf957d72ffbb8

Tracker error XSS in inspector (web client) Tracker error messages are inadequately output encoded when rendered by the tracker information page inside the WebUI, allowing a malicious tracker to inject an XSS payload into the page. Esploiting this issue allows an attacker to supply arbitrary client-side code that will ultimately be rendered and executed within the end user's web browser. Found by Rory McNamara (Gotham Digital Science). CVE pending.

view details

Mike Gelfand

commit sha 913d9d3f37a8ec9702abaf50410ca1ac126e094a

HTML-escape torrent name displayed in trackers inspector tab This will prevent injection of arbitrary HTML when multiple torrents are selected. Follow-up to the previous commit.

view details

Zachary J. Slater

commit sha f388d17c2db8f7f06cfb6a879898588a9c776de3

Updated copyright string to 2018 Small update to copyright string

view details

push time in 2 months

push eventvuori/transmission

geertjan

commit sha 23e469a6b2e253dc54c479a98447d8bdbc799be8

Fix setting the ETA timestamps Field etaDLSpeedCalculatedAt was set too early, causing the condition following it to always be false. The same for etaULSpeedCalculatedAt.

view details

Mike Gelfand

commit sha cefd8a314962bd919c162a5eb96e88eec9994f15

Bump version to 2.93+, fix NEWS date

view details

Mike Gelfand

commit sha 4d5fe6ede92b8c22e8a022b91648dbf2c84ce1cb

Add support for mbedtls (formely polarssl)

view details

Mike Gilbert

commit sha 357cccb9cf87e9c1808df4a7c48f600d7fc9f204

cmake: link against libmbedcrypto if available In recent versions of mbed TLS, several symbols are moved to libmbedcrypto. Fixes: https://github.com/transmission/transmission/issues/115

view details

Mike Gelfand

commit sha b21a6140e963491f146293c9b0a2863a9ae99ad3

Fix FTCBFS due to AC_RUN_IFELSE (patch by Helmut Grohne) Fixes: #475

view details

dubhater

commit sha e19ab0b7a9139da65e502520551deabee8452118

Qt: Fix bad downloaded percentage in DetailsDialog It was always 0.0% as long as the torrent was not finished.

view details

Bernard Spil

commit sha bdffe25b42d710c9a2dff54ec895772100f6e1d5

Fix build with LibreSSL 2.7 LibreSSL 2.7 adds OpenSSL 1.1 API See also: https://bugs.freebsd.org/226953 Signed-off-by: Bernard Spil <brnrd@FreeBSD.org>

view details

Mike Gelfand

commit sha 11f1d46e1b24aeb10125158771ca313b31665715

Merge pull request #570 from Sp1l/master Fix build with LibreSSL 2.7

view details

Mike Gelfand

commit sha 9d96f87ec3b8a531b176644fc93ad6fec1d6c9eb

Fix coding style and building with !TR_LIGHTWEIGHT

view details

Niklas Haas

commit sha 11d29ff62d528319f9b9596fd746dd7795aff140

LibreSSL compatibility This seems to be the idiomatic way to fix libressl compatibility issues, judging by what most other open source projects seem to be doing. I've confirmed that transmission builds with libressl for me after this patch is applied.

view details

Bernard Spil

commit sha 06a1644e298afc948945e44b463348944988be03

Fix build with LibreSSL 2.7 LibreSSL 2.7 adds OpenSSL 1.1 API See also: https://bugs.freebsd.org/226953 Signed-off-by: Bernard Spil <brnrd@FreeBSD.org>

view details

Mike Gelfand

commit sha ab4d7073053d694b91d5c59ddac596fe5aa233e6

Merge pull request #528 from neheb/patches Fix compile errors for 2.9x

view details

Mike Gelfand

commit sha 0b5ec3a00da2013cba3dd5ddcf9811e17eae9cf8

Merge pull request #547 from dubhater/patch-6 Qt: Fix bad downloaded percentage in DetailsDialog

view details

dubhater

commit sha 85c0399d8c93b85bcb75dfc86b2b69428d0f2ad6

Qt: Fix bad downloaded percentage in DetailsDialog It was always 0.0% as long as the torrent was not finished.

view details

Mike Gelfand

commit sha 444f15edf82ca85d9a1691a9b3a615af10848328

Merge pull request #522 from gjalsem/master Fix setting the ETA timestamps

view details

geertjan

commit sha cc0b93143191d4b1648d8c990b2079886d745d13

Fix setting the ETA timestamps Field etaDLSpeedCalculatedAt was set too early, causing the condition following it to always be false. The same for etaULSpeedCalculatedAt.

view details

Mike Gelfand

commit sha 69d9afd75a64fb7ebaec0d45e348abe5f9cc1ec4

Adjust CFBundleVersion format to fix Sparkle update check Set the first version component to be the same as the last Subversion-based release build version and add two more components (major and minor version numbers). To allow for nightly build updates this should probably include another component (e.g. build timestamp), but we're not there yet.

view details

Mike Gelfand

commit sha 523d8c1317725143556d771d960cf957d72ffbb8

Tracker error XSS in inspector (web client) Tracker error messages are inadequately output encoded when rendered by the tracker information page inside the WebUI, allowing a malicious tracker to inject an XSS payload into the page. Esploiting this issue allows an attacker to supply arbitrary client-side code that will ultimately be rendered and executed within the end user's web browser. Found by Rory McNamara (Gotham Digital Science). CVE pending.

view details

Mike Gelfand

commit sha 913d9d3f37a8ec9702abaf50410ca1ac126e094a

HTML-escape torrent name displayed in trackers inspector tab This will prevent injection of arbitrary HTML when multiple torrents are selected. Follow-up to the previous commit.

view details

Zachary J. Slater

commit sha f388d17c2db8f7f06cfb6a879898588a9c776de3

Updated copyright string to 2018 Small update to copyright string

view details

push time in 2 months

pull request commenttransmission/transmission

Wrong IPv6 address used in announces

Sure, I'll try to rebase and validate this week.

vuori

comment created time in 2 months

issue commentneomutt/neomutt

next-unread-mailbox only works every mail_check_stats_interval seconds

A quick delve into the history... It seems that function was created ~14 years ago and barely touched since. The original <next-folder> function: "open next folder with new mail" It got renamed to unread, but the docs never changed.

I think that if we're going to change the behaviour, we should add a function for the new case.

  • <next-unread-mailbox>
  • <next-new-mailbox>

Note that as it stands, next-unread-mailbox goes to neither new nor unread mailboxes with master for me due to cached MailboxCount getting stuck at 0.

I much prefer unread meaning unread, but adding next-new-mailbox for those who prefer unread=new would be nice I guess. The standard for consistency is not very high though: sidebar-next/prev-new will select non-new unread folders despite the name (they also are not affected by mail_check_recent unlike next-unread-mailbox, at least when mark_old=no).

vuori

comment created time in 2 months

issue commentneomutt/neomutt

next-unread-mailbox only works every mail_check_stats_interval seconds

So after digging a bit more, it seems to me that https://github.com/neomutt/neomutt/pull/3023/commits/46dcb33f21ce87529512841ecab352efff5bad5a should be reverted and the problem mentioned in the PR https://github.com/neomutt/neomutt/pull/3023 regarding Notmuch ignoring the check_stats flag should be fixed in the Notmuch backend instead?

While second-guessing the backend in mailbox_check() can be made to work, it seems like a recipe for trouble in the future, especially when there's already a mechanism for telling the backend not to calculate stats.

vuori

comment created time in 2 months

issue openedneomutt/neomutt

next-unread-mailbox only works every mail_check_stats_interval seconds

<!-- Sorry you've found a bug, but thanks for reporting it here. Please try to give as much information as you can. -->

Expected Behaviour

next-unread-mailbox should go to next unread mailbox. <!-- What were you trying to do? -->

Actual Behaviour

<!-- What went wrong? -->

If called more often than mail_check_stats_interval seconds, function says that there are no unread mailboxes.

Steps to Reproduce

<!-- As much detail as possible, please -->

  1. Wait for mail to arrive in a mailbox which is not the current one.
  2. Press key bount to next-unread-mailbox, note that it will usually work once.
  3. Move to another mailbox (without reading the new mail), try it again.
  4. "No new messages" will be printed.

How often does this happen?

  • Always

When did it start to happen?

Switched to master from Ubuntu package (20191207+dfsg.1-1.1) (maybe, I'm not sure if I tried to use the function when using the packaged release).

NeoMutt Version

NeoMutt 20210205-804-7958fd-dirty
Copyright (C) 1996-2020 Michael R. Elkins and others.
NeoMutt comes with ABSOLUTELY NO WARRANTY; for details type 'neomutt -vv'.
NeoMutt is free software, and you are welcome to redistribute it
under certain conditions; type 'neomutt -vv' for details.

System: Linux 5.10.60+ (x86_64)
ncurses: ncurses 6.2.20200212 (compiled with 6.2.20200212)
libidn: 1.33 (compiled with 1.33)
GPGME: 1.13.1-unknown
OpenSSL: OpenSSL 1.1.1f  31 Mar 2020
libnotmuch: 5.4.0
PCRE2: 10.34 2019-11-21
storage: kyotocabinet
compression: lz4, zlib

Extra Info

The problem is that mutt_mailbox_check() sets MailboxCount to 0 even in cases when it's not going to actually check stats on mailboxes due to mail_check_stats_interval not having expired, except on the first time it sees a given mailbox.

Diff below fixes the immediate problem, though it may lead to MailboxCount being wonky after a new mailbox is added. In practice first_check_stats_done being false after the first call to the function sounds rare, but I think a canonical fix would require always either checking stats on all mailboxes or checking none of them. Basically first_check_stats_done would need to be checked in a separate pass and check_stats set to true if at least one mailbox needed a check I guess.

diff --git a/mutt_mailbox.c b/mutt_mailbox.c
index ac40682af..7e32706b0 100644
--- a/mutt_mailbox.c
+++ b/mutt_mailbox.c
@@ -191,10 +191,10 @@ int mutt_mailbox_check(struct Mailbox *m_cur, int force)
   {
     check_stats = true;
     MailboxStatsTime = t;
+    MailboxCount = 0;
   }
 
   MailboxTime = t;
-  MailboxCount = 0;
   MailboxNotify = 0;
 
   /* check device ID and serial number instead of comparing paths */

created time in 2 months

more