FAQ
Recently we've had several situations where our project's Python 3.2 support blocks us from using libraries we'd like to consume (idna, characteristic, etc). Donald opened an issue (https://github.com/pyca/cryptography/issues/1809) with some evidence that we're performing significant contortions to support a version no one uses. I propose that we drop support for Python 3.2 in the next release and move forward with Python 2.6 (deprecated but no timeline for removal), Python 2.7, and 3.3+ support.


I've put in a PR (https://github.com/pyca/cryptography/pull/1846) to note this in the changelog and update our travis configuration. If there are no substantial objections in the next few days we can merge and then update jenkins to reflect this as well.


-Paul Kehrer




-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/cryptography-dev/attachments/20150414/9afc3fd7/attachment.html>

Search Discussions

  • Terry Chia at Apr 14, 2015 at 2:27 pm
    FWIW I'm fully in support of this as it doesn't seem like we'll be dropping
    support for any major platforms and 3.2 use is basically nil at this point.
    On Tue, 14 Apr 2015 at 10:23 pm Paul Kehrer wrote:

    Recently we've had several situations where our project's Python 3.2
    support blocks us from using libraries we'd like to consume (idna,
    characteristic, etc). Donald opened an issue (
    https://github.com/pyca/cryptography/issues/1809) with some evidence that
    we're performing significant contortions to support a version no one uses.
    I propose that we drop support for Python 3.2 in the next release and move
    forward with Python 2.6 (deprecated but no timeline for removal), Python
    2.7, and 3.3+ support.

    I've put in a PR (https://github.com/pyca/cryptography/pull/1846) to note
    this in the changelog and update our travis configuration. If there are no
    substantial objections in the next few days we can merge and then update
    jenkins to reflect this as well.

    -Paul Kehrer


    _______________________________________________
    Cryptography-dev mailing list
    Cryptography-dev at python.org
    https://mail.python.org/mailman/listinfo/cryptography-dev
    -------------- next part --------------
    An HTML attachment was scrubbed...
    URL: <http://mail.python.org/pipermail/cryptography-dev/attachments/20150414/d6c51bbc/attachment.html>
  • Laurens Van Houtven at Apr 14, 2015 at 2:49 pm
    I for one support any and all efforts that reduce the number of supported
    Python 3.x versions ;-)


    On Tue, Apr 14, 2015 at 10:27 AM, Terry Chia wrote:

    FWIW I'm fully in support of this as it doesn't seem like we'll be
    dropping support for any major platforms and 3.2 use is basically nil at
    this point.
    On Tue, 14 Apr 2015 at 10:23 pm Paul Kehrer wrote:

    Recently we've had several situations where our project's Python 3.2
    support blocks us from using libraries we'd like to consume (idna,
    characteristic, etc). Donald opened an issue (
    https://github.com/pyca/cryptography/issues/1809) with some evidence
    that we're performing significant contortions to support a version no one
    uses. I propose that we drop support for Python 3.2 in the next release and
    move forward with Python 2.6 (deprecated but no timeline for removal),
    Python 2.7, and 3.3+ support.

    I've put in a PR (https://github.com/pyca/cryptography/pull/1846) to
    note this in the changelog and update our travis configuration. If there
    are no substantial objections in the next few days we can merge and then
    update jenkins to reflect this as well.

    -Paul Kehrer


    _______________________________________________
    Cryptography-dev mailing list
    Cryptography-dev at python.org
    https://mail.python.org/mailman/listinfo/cryptography-dev
    _______________________________________________
    Cryptography-dev mailing list
    Cryptography-dev at python.org
    https://mail.python.org/mailman/listinfo/cryptography-dev
    -------------- next part --------------
    An HTML attachment was scrubbed...
    URL: <http://mail.python.org/pipermail/cryptography-dev/attachments/20150414/dfc5f679/attachment.html>
  • Alex Gaynor at Apr 14, 2015 at 3:22 pm
    Have we confirmed that all important downstreams (pyOpenSSL, Twisted,
    eventually Fabric/Paramiko, urllib3/requests) have dropped 3.2?


    Alex


    On Tue, Apr 14, 2015 at 10:49 AM, Laurens Van Houtven wrote:

    I for one support any and all efforts that reduce the number of supported
    Python 3.x versions ;-)
    On Tue, Apr 14, 2015 at 10:27 AM, Terry Chia wrote:

    FWIW I'm fully in support of this as it doesn't seem like we'll be
    dropping support for any major platforms and 3.2 use is basically nil at
    this point.
    On Tue, 14 Apr 2015 at 10:23 pm Paul Kehrer <paul.l.kehrer@gmail.com>
    wrote:
    Recently we've had several situations where our project's Python 3.2
    support blocks us from using libraries we'd like to consume (idna,
    characteristic, etc). Donald opened an issue (
    https://github.com/pyca/cryptography/issues/1809) with some evidence
    that we're performing significant contortions to support a version no one
    uses. I propose that we drop support for Python 3.2 in the next release and
    move forward with Python 2.6 (deprecated but no timeline for removal),
    Python 2.7, and 3.3+ support.

    I've put in a PR (https://github.com/pyca/cryptography/pull/1846) to
    note this in the changelog and update our travis configuration. If there
    are no substantial objections in the next few days we can merge and then
    update jenkins to reflect this as well.

    -Paul Kehrer


    _______________________________________________
    Cryptography-dev mailing list
    Cryptography-dev at python.org
    https://mail.python.org/mailman/listinfo/cryptography-dev
    _______________________________________________
    Cryptography-dev mailing list
    Cryptography-dev at python.org
    https://mail.python.org/mailman/listinfo/cryptography-dev
    _______________________________________________
    Cryptography-dev mailing list
    Cryptography-dev at python.org
    https://mail.python.org/mailman/listinfo/cryptography-dev



    --
    "I disapprove of what you say, but I will defend to the death your right to
    say it." -- Evelyn Beatrice Hall (summarizing Voltaire)
    "The people's good is the highest law." -- Cicero
    GPG Key fingerprint: 125F 5C67 DFE9 4084
    -------------- next part --------------
    An HTML attachment was scrubbed...
    URL: <http://mail.python.org/pipermail/cryptography-dev/attachments/20150414/186f4375/attachment-0001.html>
  • Glyph at Apr 15, 2015 at 4:07 am
    Is there a way to answer this question as a query against PyPI metadata? It seems like the information ought to be there, in some form...


    -g

    On Apr 14, 2015, at 11:22, Alex Gaynor wrote:

    Have we confirmed that all important downstreams (pyOpenSSL, Twisted, eventually Fabric/Paramiko, urllib3/requests) have dropped 3.2?

    Alex

    On Tue, Apr 14, 2015 at 10:49 AM, Laurens Van Houtven <_ at lvh.io wrote:
    I for one support any and all efforts that reduce the number of supported Python 3.x versions ;-)

    On Tue, Apr 14, 2015 at 10:27 AM, Terry Chia <terrycwk1994 at gmail.com wrote:
    FWIW I'm fully in support of this as it doesn't seem like we'll be dropping support for any major platforms and 3.2 use is basically nil at this point.
    On Tue, 14 Apr 2015 at 10:23 pm Paul Kehrer <paul.l.kehrer at gmail.com wrote:
    Recently we've had several situations where our project's Python 3.2 support blocks us from using libraries we'd like to consume (idna, characteristic, etc). Donald opened an issue (https://github.com/pyca/cryptography/issues/1809 <https://github.com/pyca/cryptography/issues/1809>) with some evidence that we're performing significant contortions to support a version no one uses. I propose that we drop support for Python 3.2 in the next release and move forward with Python 2.6 (deprecated but no timeline for removal), Python 2.7, and 3.3+ support.

    I've put in a PR (https://github.com/pyca/cryptography/pull/1846 <https://github.com/pyca/cryptography/pull/1846>) to note this in the changelog and update our travis configuration. If there are no substantial objections in the next few days we can merge and then update jenkins to reflect this as well.

    -Paul Kehrer


    _______________________________________________
    Cryptography-dev mailing list
    Cryptography-dev at python.org <mailto:cryptography-dev@python.org>
    https://mail.python.org/mailman/listinfo/cryptography-dev <https://mail.python.org/mailman/listinfo/cryptography-dev>

    _______________________________________________
    Cryptography-dev mailing list
    Cryptography-dev at python.org <mailto:cryptography-dev@python.org>
    https://mail.python.org/mailman/listinfo/cryptography-dev <https://mail.python.org/mailman/listinfo/cryptography-dev>



    _______________________________________________
    Cryptography-dev mailing list
    Cryptography-dev at python.org <mailto:cryptography-dev@python.org>
    https://mail.python.org/mailman/listinfo/cryptography-dev <https://mail.python.org/mailman/listinfo/cryptography-dev>




    --
    "I disapprove of what you say, but I will defend to the death your right to say it." -- Evelyn Beatrice Hall (summarizing Voltaire)
    "The people's good is the highest law." -- Cicero
    GPG Key fingerprint: 125F 5C67 DFE9 4084
    _______________________________________________
    Cryptography-dev mailing list
    Cryptography-dev at python.org
    https://mail.python.org/mailman/listinfo/cryptography-dev

    -------------- next part --------------
    An HTML attachment was scrubbed...
    URL: <http://mail.python.org/pipermail/cryptography-dev/attachments/20150415/bd3e2794/attachment.html>
  • Cory Benfield at Apr 15, 2015 at 7:37 am

    On 15 April 2015 at 05:07, Glyph wrote:
    Is there a way to answer this question as a query against PyPI metadata? It
    seems like the information ought to be there, in some form...

    -g

    IME a lot of projects don't keep their PyPI metadata up-to-date.


    I already responded to Paul on Twitter, but requests abandoned 3.2 a
    long time ago. urllib3 nominally supports it, but I don't think we'd
    need much convincing to drop 3.2 support really.
  • Donald Stufft at Apr 15, 2015 at 8:06 am

    On Apr 15, 2015, at 3:37 AM, Cory Benfield wrote:
    On 15 April 2015 at 05:07, Glyph wrote:
    Is there a way to answer this question as a query against PyPI metadata? It
    seems like the information ought to be there, in some form...

    -g
    IME a lot of projects don't keep their PyPI metadata up-to-date.

    I already responded to Paul on Twitter, but requests abandoned 3.2 a
    long time ago. urllib3 nominally supports it, but I don't think we'd
    need much convincing to drop 3.2 support really.

    lol whoops, pip supports 3.2, requests still works on 3.2 though!


    ---
    Donald Stufft
    PGP: 7C6B 7C5D 5E2B 6356 A926 F04F 6E3C BCE9 3372 DCFA


    -------------- next part --------------
    A non-text attachment was scrubbed...
    Name: signature.asc
    Type: application/pgp-signature
    Size: 801 bytes
    Desc: Message signed with OpenPGP using GPGMail
    URL: <http://mail.python.org/pipermail/cryptography-dev/attachments/20150415/40451e69/attachment.sig>
  • Cory Benfield at Apr 15, 2015 at 8:58 am

    On 15 April 2015 at 09:06, Donald Stufft wrote:
    lol whoops, pip supports 3.2, requests still works on 3.2 though!

    Heh, yeah: we don't support 3.2 but we do support pip, so as long as
    it works well enough for you then that's fine by me.


    That said, 3.2 is bad and people using it should feel bad.
  • Alex Gaynor at Apr 18, 2015 at 7:52 pm
    For those not following the issue, Donald posted some graphs:
    https://github.com/pyca/cryptography/pull/1846#issuecomment-94195717


    Basically, no one uses this, it's old and burdensome, I'm now also +1 on
    dropping.


    Paramiko still supports 3.2 though, so I sent bitprophet a message to
    confirm if it's a problem for him. That said, I'm planning on merging this
    as soon as Paul updates it with the last issue (;-)), and we can revert if
    it turns out it'll be a giant issue for paramiko.


    Cheers,
    Alex


    On Wed, Apr 15, 2015 at 4:58 AM, Cory Benfield wrote:

    On 15 April 2015 at 09:06, Donald Stufft wrote:
    lol whoops, pip supports 3.2, requests still works on 3.2 though!
    Heh, yeah: we don't support 3.2 but we do support pip, so as long as
    it works well enough for you then that's fine by me.

    That said, 3.2 is bad and people using it should feel bad.
    _______________________________________________
    Cryptography-dev mailing list
    Cryptography-dev at python.org
    https://mail.python.org/mailman/listinfo/cryptography-dev





    --
    "I disapprove of what you say, but I will defend to the death your right to
    say it." -- Evelyn Beatrice Hall (summarizing Voltaire)
    "The people's good is the highest law." -- Cicero
    GPG Key fingerprint: 125F 5C67 DFE9 4084
    -------------- next part --------------
    An HTML attachment was scrubbed...
    URL: <http://mail.python.org/pipermail/cryptography-dev/attachments/20150418/1b6c26f1/attachment.html>
  • Donald Stufft at Apr 18, 2015 at 7:57 pm
    If Jeff wants it I can generate graphs for paramiko as well.



    On Apr 18, 2015, at 3:52 PM, Alex Gaynor wrote:

    For those not following the issue, Donald posted some graphs:
    https://github.com/pyca/cryptography/pull/1846#issuecomment-94195717 <https://github.com/pyca/cryptography/pull/1846#issuecomment-94195717>

    Basically, no one uses this, it's old and burdensome, I'm now also +1 on dropping.

    Paramiko still supports 3.2 though, so I sent bitprophet a message to confirm if it's a problem for him. That said, I'm planning on merging this as soon as Paul updates it with the last issue (;-)), and we can revert if it turns out it'll be a giant issue for paramiko.

    Cheers,
    Alex

    On Wed, Apr 15, 2015 at 4:58 AM, Cory Benfield <cory at lukasa.co.uk wrote:
    On 15 April 2015 at 09:06, Donald Stufft <donald at stufft.io wrote:
    lol whoops, pip supports 3.2, requests still works on 3.2 though!
    Heh, yeah: we don't support 3.2 but we do support pip, so as long as
    it works well enough for you then that's fine by me.

    That said, 3.2 is bad and people using it should feel bad.
    _______________________________________________
    Cryptography-dev mailing list
    Cryptography-dev at python.org <mailto:cryptography-dev@python.org>
    https://mail.python.org/mailman/listinfo/cryptography-dev <https://mail.python.org/mailman/listinfo/cryptography-dev>



    --
    "I disapprove of what you say, but I will defend to the death your right to say it." -- Evelyn Beatrice Hall (summarizing Voltaire)
    "The people's good is the highest law." -- Cicero
    GPG Key fingerprint: 125F 5C67 DFE9 4084
    _______________________________________________
    Cryptography-dev mailing list
    Cryptography-dev at python.org
    https://mail.python.org/mailman/listinfo/cryptography-dev

    ---
    Donald Stufft
    PGP: 7C6B 7C5D 5E2B 6356 A926 F04F 6E3C BCE9 3372 DCFA


    -------------- next part --------------
    An HTML attachment was scrubbed...
    URL: <http://mail.python.org/pipermail/cryptography-dev/attachments/20150418/79898c97/attachment.html>
    -------------- next part --------------
    A non-text attachment was scrubbed...
    Name: signature.asc
    Type: application/pgp-signature
    Size: 801 bytes
    Desc: Message signed with OpenPGP using GPGMail
    URL: <http://mail.python.org/pipermail/cryptography-dev/attachments/20150418/79898c97/attachment.sig>
  • Alex Gaynor at Apr 18, 2015 at 8:00 pm
    I'll wait to see what he says first.


    Alex


    On Sat, Apr 18, 2015 at 3:57 PM, Donald Stufft wrote:

    If Jeff wants it I can generate graphs for paramiko as well.


    On Apr 18, 2015, at 3:52 PM, Alex Gaynor wrote:

    For those not following the issue, Donald posted some graphs:
    https://github.com/pyca/cryptography/pull/1846#issuecomment-94195717

    Basically, no one uses this, it's old and burdensome, I'm now also +1 on
    dropping.

    Paramiko still supports 3.2 though, so I sent bitprophet a message to
    confirm if it's a problem for him. That said, I'm planning on merging this
    as soon as Paul updates it with the last issue (;-)), and we can revert if
    it turns out it'll be a giant issue for paramiko.

    Cheers,
    Alex
    On Wed, Apr 15, 2015 at 4:58 AM, Cory Benfield wrote:
    On 15 April 2015 at 09:06, Donald Stufft wrote:
    lol whoops, pip supports 3.2, requests still works on 3.2 though!
    Heh, yeah: we don't support 3.2 but we do support pip, so as long as
    it works well enough for you then that's fine by me.

    That said, 3.2 is bad and people using it should feel bad.
    _______________________________________________
    Cryptography-dev mailing list
    Cryptography-dev at python.org
    https://mail.python.org/mailman/listinfo/cryptography-dev


    --
    "I disapprove of what you say, but I will defend to the death your right
    to say it." -- Evelyn Beatrice Hall (summarizing Voltaire)
    "The people's good is the highest law." -- Cicero
    GPG Key fingerprint: 125F 5C67 DFE9 4084
    _______________________________________________
    Cryptography-dev mailing list
    Cryptography-dev at python.org
    https://mail.python.org/mailman/listinfo/cryptography-dev


    ---
    Donald Stufft
    PGP: 7C6B 7C5D 5E2B 6356 A926 F04F 6E3C BCE9 3372 DCFA


    _______________________________________________
    Cryptography-dev mailing list
    Cryptography-dev at python.org
    https://mail.python.org/mailman/listinfo/cryptography-dev



    --
    "I disapprove of what you say, but I will defend to the death your right to
    say it." -- Evelyn Beatrice Hall (summarizing Voltaire)
    "The people's good is the highest law." -- Cicero
    GPG Key fingerprint: 125F 5C67 DFE9 4084
    -------------- next part --------------
    An HTML attachment was scrubbed...
    URL: <http://mail.python.org/pipermail/cryptography-dev/attachments/20150418/baeee6fc/attachment-0001.html>

Related Discussions

Discussion Navigation
viewthread | post
Discussion Overview
groupcryptography-dev @
categoriespython
postedApr 14, '15 at 2:23p
activeApr 18, '15 at 8:00p
posts11
users7
websitepython.org

People

Translate

site design / logo © 2017 Grokbase