-
-
Notifications
You must be signed in to change notification settings - Fork 2k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
2.6.0/2.6.1 release #4196
Comments
When will it be released? |
We are indeed late on the schedule. Our current goal is the end of February, but it might be subject to another shift. |
Hi there, any update on when this will be resolved? |
I've just released '2.6.0rc1'. |
This comment was marked as outdated.
This comment was marked as outdated.
Hello. Any updates about the release? Is it possible to release, say |
This comment was marked as off-topic.
This comment was marked as off-topic.
This comment was marked as off-topic.
This comment was marked as off-topic.
Just released '2.6.0rc2'. Try it out via Expect the full 2.6.0 within 1 or 2 weeks. Edit: delayed slightly because of #4538, still on track for the end of this week. |
2.6.0 is out. |
@charles2910 Hi & sorry for the ping. Wanted to let you know that we have a new release available. Btw is there a pseudo-automated way of doing this kind of reports? Fedora has super great tools for upstream, for instance https://docs.fedoraproject.org/en-US/infra/sysadmin_guide/hotness/ automatically reports new releases to downstream package maintainers. Do you know if there's something similar for Debian? I'm realising it takes quite some time to notify all package maintainers of the new release :) To anyone reading here, do you have a preference? Some projects (like Fedora) have it automated, which is handy. We could also send a grouped email to maintainers when releases happen, if people are up for it. Feel free to tell me if interested. Thanks a lot for your help, cheers |
@dhgutteridge Hi. Really sorry for bothering. |
@bluhm Hi, also very sorry for bothering. |
Hi, @gpotter2 and thanks for pinging me! (don't be sorry, I really enjoy it since I can't keep track of everything all the time) We do have some automated tools that scans regularly the upstream's to see if there is a new release, but it didn't pickup scapy 2.6.0 yet. I'm trying to subscribe to all Github repos so I get the email for the new releases instantly but I was missing scapy (now I'm subscribed :-) Also kudos for reaching out packagers, it's a very nice gesture! Now let's see if I can prepare and upload to Debian before Sunday ends. Cheers |
Hi again :-) I'm just finishing packaging 2.6.0 and I saw the advertised support to python 3.13. At the same time, I: scapy source: uses-deprecated-python-stdlib crypt (deprecated in Python 3.11, removed in Python 3.13) [scapy/modules/krack/automaton.py:11]
I: scapy source: uses-deprecated-python-stdlib crypt (deprecated in Python 3.11, removed in Python 3.13) [scapy/modules/krack/automaton.py:12]
I: scapy source: uses-deprecated-python-stdlib crypt (deprecated in Python 3.11, removed in Python 3.13) [scapy/modules/krack/automaton.py:13]
I: scapy source: uses-deprecated-python-stdlib crypt (deprecated in Python 3.11, removed in Python 3.13) [scapy/modules/krack/crypto.py:10]
I: scapy source: uses-deprecated-python-stdlib crypt (deprecated in Python 3.11, removed in Python 3.13) [scapy/modules/krack/crypto.py:11]
I: scapy source: uses-deprecated-python-stdlib uu (deprecated in Python 3.11, removed in Python 3.13) [scapy/contrib/pnio_rpc.py:14]
I: scapy source: uses-deprecated-python-stdlib uu (deprecated in Python 3.11, removed in Python 3.13) [scapy/fields.py:24]
I: scapy source: uses-deprecated-python-stdlib uu (deprecated in Python 3.11, removed in Python 3.13) [scapy/layers/dcerpc.py:29]
I: scapy source: uses-deprecated-python-stdlib uu (deprecated in Python 3.11, removed in Python 3.13) [scapy/layers/spnego.py:20]
I: scapy source: uses-deprecated-python-stdlib uu (deprecated in Python 3.11, removed in Python 3.13) [scapy/utils.py:14] Indeed there is a removal warning on python's website 1 2. Were you guys aware of this? Should I open an issue? |
Hi, We import
That was fast ! Thanks a lot |
Oh dear, now I must apologize. I blindly trusted lintian on this one and didn't check the files... Sorry about this. I think I was a bit anxious to upload the new version and skipped some steps. Good news is I've uploaded to experimental already! |
You can ping us however you like, e.g., this is fine. (Once a maintainer knows about a new package update, we may record it centrally in a "todo" document, though that approach varies.) pkgsrc is in a freeze at the moment for stable branching, so I can't carry this update yet. Though it should be un-frozen soon. I have already looked at packaging and completed it. Though I haven't looked through test results in any detail. A couple of comments. The generated wheel (at least, what I get) doesn't install the scapy.1 man page. I see there was specific handling for it, and that was removed in refactoring relatively recently. (In my case, I've simply worked around this locally for now by copying it over to our release tree.) I was a little surprised that "Version" ends up being a date and not 2.6.0, and downstream packagers are advised to forcibly override this in the environment. But, anyway, that's obviously been handled. Thanks for all your work on this! |
Great.
You are right, this was lost in the migration to
The date is a fallback and isn't supposed to happen. If you cloned the repo while fetching the git tags, they're supposed to be used. If you downloaded a zip file it's supposed to be automatically baked in by git. Could be a bug though, what method did you use? Thanks |
I've opened this as issue #4549 for tracking. Thanks.
We download from GitHub as https://github.com/secdev/scapy/archive/v2.6.0.tar.gz. I had inferred this was expected behaviour because of the section in Having analyzed, I see that |
Hey guys, I saw there were some bugfixes after the release. Are you guys considering doing a patch release (2.6.1)? If not, could you consider it? For what I have seen #4558, #4560, #4567, #4569 and 28287eb are the most important fixes and the ones I'm considering to add as patches if there is no patch release. Should I add something else? |
Hi, thanks for reaching out. The ones you listed are enough, #4560 being to me the most critical (as it can't really be worked around). I want to take a second look at the code handling XDG and to make sure I didn't miss some other case that would result in a crash, then we'll be good to go. |
FTR |
This issue tracks the associated 2.6.0 release. As usual, feel free to comment down below to have some features/bugs included before the final release.
Note to package maintainers: it is important to point out that special care should be taken when porting/testing this release. The plateform-specific code aimed at reading the network configuration (interfaces, routes, etc.) has been entirely rewritten on both Linux and *BSD flavors. Plateforms that were tested include: Windows, Linux, OpenBSD, NetBSD, FreeBSD, Darwin. Other plateforms have not been tested, therefore we encourage maintainers to perform additional testing.
Changelog
General
[removal] DROP SUPPORT OF PYTHON 2.7
pyproject.toml
) and version handling. Scapy will now include wheels on pypi.Main changes
sr(IP(dst="224.0.0.1%eth0")/..., multi=True)
iface=
argument is deprecated on level3 functions (send, sr, sr1), as its behavior was undefined. It remains in use for level2 functions (sendp, srp, srp1). RFC6874-like scope identifiers (see just above) should be used.DCERPC_Client
andDCERPC_Server
with support forNCACN_IP_TCP
andNCACN_NP
smbserver()
) (2.0.2 to 3.1.1)smbclient()
) (2.0.2 to 3.1.1)LDAP_client
(support for various binding mechanisms, encryption, etc.)cryptography
(42/43.0) versions~/.config/scapy/startup.py
. This follows XDG variables. (Older~/.scapy_startup.py
is now non functional)bpython
,ptpython
andptipython
load_extcap()
)StreamSocket
changes, support for TCP reassembly, etc. TCPSession(app=True) must no longer be used withStreamSocket
. Custom sessions are marked as unstable.L3RawSocket(6)
automatically on the loopback interface on linuxL3pcapSocket
(the default L3 on Windows or when libpcap is used) now follows the same behavior as other L3 sockets when routingsr*
class of functions now properly supports sending on multiple interfaces (Windows & Linux)sr*
class of functions have also been fixedmanufdb
(from wireshark) is now bundled and cached in~/.cache/scapy
, as it is no longer shipped as a standalone file in Wireshark.dnsd
,llmnrd
,nbnsd
,dhcpd
...). Addmdnsd
for mDNS support*ListFields
conf.nameservers
contains the DNS servers. Also addsdns_resolve()
Session
objectssendpfast
loop argument to be consistent withsendp
graph()
to include implicit linksHTTP_Client
andHTTP_Server
which support the same SSPs as Windowshttp_client
get_if_hwaddr
for performancearping
without IPAutomotive changes
TODO
The text was updated successfully, but these errors were encountered: