Re: [sage-devel] Re: Question about make dependencies

[John H Palmieri]

If I respond using "verb" as a verb, do you really think I'm seriously 
criticizing you about using "vendor" as a verb? You're telling me to get a 
life? Get a sense of humor.

On Tuesday, November 14, 2023 at 3:45:07 PM UTC-8 Dima Pasechnik wrote:

> I have not invented the verb "to vendor", I merely picked it up from 
> Python docs and PEPs, e.g.
> https://www.python.org/dev/peps/pep-0518/#configparser
> (this one was written by  native English speakers, I think)
>
> You seem to prefer to pick on my supposedly broken English, rather than 
> read Python docs, as it's much less fun.
>
> Get a life.
>
>
>
> On Tue, 14 Nov 2023, 23:19 John H Palmieri,  wrote:
>
>> Not all Americans like to verb things so profligately.
>>
>> On Tuesday, November 14, 2023 at 1:15:07 PM UTC-8 Dima Pasechnik wrote:
>>
>>>
>>>
>>> On Tue, 14 Nov 2023, 20:39 Matthias Koeppe,  
>>> wrote:
>>>
 On Tuesday, November 14, 2023 at 11:39:52 AM UTC-8 Dima Pasechnik wrote:

 On Tue, Nov 14, 2023 at 7:22 PM Marc Culler  
 wrote: 
 > On Tue, Nov 14, 2023 at 11:58 AM Dima Pasechnik  
 wrote: 
 >> I imagine it might be an artifact of the design of Sage on Mac app 
 >> Marc is releasing: 
 >> vendor everything. 
 > 
 > Vendor is a noun, not a verb. 

 It's a curse, to be precise, and it slows down the project a lot. 
 We have to worry about over 400 packages!!! 
 Bloody hell, pardon my UK French, we have more interesting problems to 
 work on...


 FWIW, I consider Marc's effort to provide a self-contained macOS app of 
 Sage to be very valuable for the project.

 Dima is indeed using the word "vendor" incorrectly. I'm not concerned 
 about the grammar; I use it as a verb, too, in this specific context.
 For source code, "vendoring" (= including a copy of the source code of 
 a dependency, modified or not, in the source code) is problematic for 
 various well known reasons.
 But to accuse a binary distribution of "vendoring" is simply 
 meaningless. A binary distribution is made for the benefit of users.

>>>
>>> I am not accusing Marc of vendoring, I am saying that him asking to 
>>> supply essentially all the components of Sage in source form is asking too 
>>> much, as it saddles us with extremely heavy maintenance task, for reasons 
>>> questionable.
>>>
>>> Suppose we remove gmp from the list of packages we vendor (yes, "to 
>>> vendor", i.e. "to perform vendoring", may be a verb, as we see here. As 
>>> Americans make verbs out of everything, I find myself being gaslighted here 
>>> by one of them.)
>>>
>>> What then, will it make the task of making a macOS Sage app impossible, 
>>> or much harder, all of a sudden?
>>> I have trouble believing this.
>>>
>>>
 -- 
 You received this message because you are subscribed to the Google 
 Groups "sage-devel" group.
 To unsubscribe from this group and stop receiving emails from it, send 
 an email to sage-devel+...@googlegroups.com.

>>> To view this discussion on the web visit 
 https://groups.google.com/d/msgid/sage-devel/fa119cf9-49bc-4b16-b6af-b1b6fd6584aen%40googlegroups.com
  
 
 .

>>> -- 
>> You received this message because you are subscribed to the Google Groups 
>> "sage-devel" group.
>> To unsubscribe from this group and stop receiving emails from it, send an 
>> email to sage-devel+...@googlegroups.com.
>>
> To view this discussion on the web visit 
>> https://groups.google.com/d/msgid/sage-devel/943e37ef-0a11-4331-9e20-52141ccdddc7n%40googlegroups.com
>>  
>> 
>> .
>>
>

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/25a17216-502e-4c4d-b70d-73e6e028881cn%40googlegroups.com.

Re: [sage-devel] Symengine package fails to build in 10.2rc2

[Dima Pasechnik]

It's not on PyPI yet, as Isuru tells me there.


On Tue, 14 Nov 2023, 21:03 Marc Culler,  wrote:

>
> I see your unanswered message "ping" on issue #456 from 5 days ago.
>
> But in the Releases section I see that v0.11.0 was released 2 days ago.
>
> - Marc
>
>
> On Tuesday, November 14, 2023 at 11:50:55 AM UTC-6 Dima Pasechnik wrote:
>
>> On Tue, Nov 14, 2023 at 5:18 PM Marc Culler  wrote:
>> >
>> > The symengine package has apparently not been updated to be compatible
>> with cython 3.
>>
>> https://github.com/symengine/symengine.py/issues/456
>>
>>
>> >
>> > This is what I see in the log (full log attached):
>> >
>> > performance hint:
>> /private/var/tmp/sage-10.2-current/local/var/lib/sage/venv-python3.11.1/var/tmp/sage/build/symengine_py-0.10.0/src/symengine/lib/symengine_wrapper.pyx:4992:5:
>> Exception check on '_ctypes_callback_lambda_real' will always require the
>> GIL to be acquired.
>> > Possible solutions:
>> > 1. Declare the function as 'noexcept' if you control the definition and
>> you're sure you don't want the function to raise exceptions.
>> > 2. Use an 'int' return type on the function to allow an error code to
>> be returned.
>> >
>> > Error compiling Cython file:
>> > 
>> > ...
>> > cdef Basic X = sympify((o1).diff(c2py(symbol)._sympy_()))
>> > return X.thisptr
>> >
>> > def create_sympy_module():
>> > cdef PyModule s = PyModule.__new__(PyModule)
>> > s.thisptr = symengine.make_rcp_PyModule(_to_sympy,
>> _to_symengine, _eval,
>> > ^
>> > 
>> >
>> >
>> /private/var/tmp/sage-10.2-current/local/var/lib/sage/venv-python3.11.1/var/tmp/sage/build/symengine_py-0.10.0/src/symengine/lib/symengine_wrapper.pyx:2805:44:
>> Cannot assign type 'PyObject *(*)(rcp_const_basic) except? NULL' to
>> 'PyObject *(*)(rcp_const_basic) noexcept'. Exception values are
>> incompatible. Suggest adding 'noexcept' to type 'PyObject
>> *(rcp_const_basic) except? NULL'.
>> > make[7]: *** [symengine/lib/symengine_wrapper.cpp] Error 1
>> > make[6]: *** [symengine/lib/CMakeFiles/symengine_wrapper.dir/all] Error
>> 2
>> > make[5]: *** [all] Error 2
>> > error: error building project
>> > error: subprocess-exited-with-error
>> >
>> > --
>> > You received this message because you are subscribed to the Google
>> Groups "sage-devel" group.
>> > To unsubscribe from this group and stop receiving emails from it, send
>> an email to sage-devel+...@googlegroups.com.
>> > To view this discussion on the web visit
>> https://groups.google.com/d/msgid/sage-devel/4fd16cfa-59fc-4571-b7a8-1ac5caeda95dn%40googlegroups.com.
>>
>>
> --
> You received this message because you are subscribed to the Google Groups
> "sage-devel" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to sage-devel+unsubscr...@googlegroups.com.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/sage-devel/6db1f1d2-b942-4e1b-9cfe-fa9cc9e792d5n%40googlegroups.com
> 
> .
>

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/CAAWYfq0CgEXVxnyhj1e3_dh4ONormduAjnXw%2B4JhuO%3DO7Np1kA%40mail.gmail.com.

Re: [sage-devel] Re: Question about make dependencies

[Dima Pasechnik]

I'm only worried that I know more English words than native speakers...

On Tue, 14 Nov 2023, 23:56 Michael Orlitzky,  wrote:

> On 2023-11-14 23:44:50, Dima Pasechnik wrote:
> > I have not invented the verb "to vendor"
>
> Don't worry, you are in good company:
>
> https://www.gocomics.com/calvinandhobbes/1993/01/25
>
>
>
>
>
> --
> You received this message because you are subscribed to the Google Groups
> "sage-devel" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to sage-devel+unsubscr...@googlegroups.com.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/sage-devel/ZVQJLHJ08qXQ2I2r%40stitch.
>

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/CAAWYfq0puMGv4gw9vFxOv-Kx0TGOa9N3nRL2aeiD%3DOsa4cuyGA%40mail.gmail.com.

Re: [sage-devel] Re: Question about make dependencies

[Michael Orlitzky]

On 2023-11-14 23:44:50, Dima Pasechnik wrote:
> I have not invented the verb "to vendor"

Don't worry, you are in good company:

https://www.gocomics.com/calvinandhobbes/1993/01/25

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/ZVQJLHJ08qXQ2I2r%40stitch.

Re: [sage-devel] Re: Question about make dependencies

[Dima Pasechnik]

I have not invented the verb "to vendor", I merely picked it up from Python
docs and PEPs, e.g.
https://www.python.org/dev/peps/pep-0518/#configparser
(this one was written by  native English speakers, I think)

You seem to prefer to pick on my supposedly broken English, rather than
read Python docs, as it's much less fun.

Get a life.



On Tue, 14 Nov 2023, 23:19 John H Palmieri,  wrote:

> Not all Americans like to verb things so profligately.
>
> On Tuesday, November 14, 2023 at 1:15:07 PM UTC-8 Dima Pasechnik wrote:
>
>>
>>
>> On Tue, 14 Nov 2023, 20:39 Matthias Koeppe,  wrote:
>>
>>> On Tuesday, November 14, 2023 at 11:39:52 AM UTC-8 Dima Pasechnik wrote:
>>>
>>> On Tue, Nov 14, 2023 at 7:22 PM Marc Culler  wrote:
>>> > On Tue, Nov 14, 2023 at 11:58 AM Dima Pasechnik 
>>> wrote:
>>> >> I imagine it might be an artifact of the design of Sage on Mac app
>>> >> Marc is releasing:
>>> >> vendor everything.
>>> >
>>> > Vendor is a noun, not a verb.
>>>
>>> It's a curse, to be precise, and it slows down the project a lot.
>>> We have to worry about over 400 packages!!!
>>> Bloody hell, pardon my UK French, we have more interesting problems to
>>> work on...
>>>
>>>
>>> FWIW, I consider Marc's effort to provide a self-contained macOS app of
>>> Sage to be very valuable for the project.
>>>
>>> Dima is indeed using the word "vendor" incorrectly. I'm not concerned
>>> about the grammar; I use it as a verb, too, in this specific context.
>>> For source code, "vendoring" (= including a copy of the source code of a
>>> dependency, modified or not, in the source code) is problematic for various
>>> well known reasons.
>>> But to accuse a binary distribution of "vendoring" is simply
>>> meaningless. A binary distribution is made for the benefit of users.
>>>
>>
>> I am not accusing Marc of vendoring, I am saying that him asking to
>> supply essentially all the components of Sage in source form is asking too
>> much, as it saddles us with extremely heavy maintenance task, for reasons
>> questionable.
>>
>> Suppose we remove gmp from the list of packages we vendor (yes, "to
>> vendor", i.e. "to perform vendoring", may be a verb, as we see here. As
>> Americans make verbs out of everything, I find myself being gaslighted here
>> by one of them.)
>>
>> What then, will it make the task of making a macOS Sage app impossible,
>> or much harder, all of a sudden?
>> I have trouble believing this.
>>
>>
>>> --
>>> You received this message because you are subscribed to the Google
>>> Groups "sage-devel" group.
>>> To unsubscribe from this group and stop receiving emails from it, send
>>> an email to sage-devel+...@googlegroups.com.
>>>
>> To view this discussion on the web visit
>>> https://groups.google.com/d/msgid/sage-devel/fa119cf9-49bc-4b16-b6af-b1b6fd6584aen%40googlegroups.com
>>> 
>>> .
>>>
>> --
> You received this message because you are subscribed to the Google Groups
> "sage-devel" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to sage-devel+unsubscr...@googlegroups.com.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/sage-devel/943e37ef-0a11-4331-9e20-52141ccdddc7n%40googlegroups.com
> 
> .
>

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/CAAWYfq0PMxGug%3Dwv1FRMZjbeGW1UdU5jPB57Z-k7CatETAuemg%40mail.gmail.com.

Re: [sage-devel] Re: Question about make dependencies

[John H Palmieri]

Not all Americans like to verb things so profligately.

On Tuesday, November 14, 2023 at 1:15:07 PM UTC-8 Dima Pasechnik wrote:

>
>
> On Tue, 14 Nov 2023, 20:39 Matthias Koeppe,  wrote:
>
>> On Tuesday, November 14, 2023 at 11:39:52 AM UTC-8 Dima Pasechnik wrote:
>>
>> On Tue, Nov 14, 2023 at 7:22 PM Marc Culler  wrote: 
>> > On Tue, Nov 14, 2023 at 11:58 AM Dima Pasechnik  
>> wrote: 
>> >> I imagine it might be an artifact of the design of Sage on Mac app 
>> >> Marc is releasing: 
>> >> vendor everything. 
>> > 
>> > Vendor is a noun, not a verb. 
>>
>> It's a curse, to be precise, and it slows down the project a lot. 
>> We have to worry about over 400 packages!!! 
>> Bloody hell, pardon my UK French, we have more interesting problems to 
>> work on...
>>
>>
>> FWIW, I consider Marc's effort to provide a self-contained macOS app of 
>> Sage to be very valuable for the project.
>>
>> Dima is indeed using the word "vendor" incorrectly. I'm not concerned 
>> about the grammar; I use it as a verb, too, in this specific context.
>> For source code, "vendoring" (= including a copy of the source code of a 
>> dependency, modified or not, in the source code) is problematic for various 
>> well known reasons.
>> But to accuse a binary distribution of "vendoring" is simply meaningless. 
>> A binary distribution is made for the benefit of users.
>>
>
> I am not accusing Marc of vendoring, I am saying that him asking to supply 
> essentially all the components of Sage in source form is asking too much, 
> as it saddles us with extremely heavy maintenance task, for reasons 
> questionable.
>
> Suppose we remove gmp from the list of packages we vendor (yes, "to 
> vendor", i.e. "to perform vendoring", may be a verb, as we see here. As 
> Americans make verbs out of everything, I find myself being gaslighted here 
> by one of them.)
>
> What then, will it make the task of making a macOS Sage app impossible, or 
> much harder, all of a sudden?
> I have trouble believing this.
>
>
>> -- 
>> You received this message because you are subscribed to the Google Groups 
>> "sage-devel" group.
>> To unsubscribe from this group and stop receiving emails from it, send an 
>> email to sage-devel+...@googlegroups.com.
>>
> To view this discussion on the web visit 
>> https://groups.google.com/d/msgid/sage-devel/fa119cf9-49bc-4b16-b6af-b1b6fd6584aen%40googlegroups.com
>>  
>> 
>> .
>>
>

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/943e37ef-0a11-4331-9e20-52141ccdddc7n%40googlegroups.com.

Re: [sage-devel] Re: Question about make dependencies

[Dima Pasechnik]

On Tue, 14 Nov 2023, 20:39 Matthias Koeppe, 
wrote:

> On Tuesday, November 14, 2023 at 11:39:52 AM UTC-8 Dima Pasechnik wrote:
>
> On Tue, Nov 14, 2023 at 7:22 PM Marc Culler  wrote:
> > On Tue, Nov 14, 2023 at 11:58 AM Dima Pasechnik 
> wrote:
> >> I imagine it might be an artifact of the design of Sage on Mac app
> >> Marc is releasing:
> >> vendor everything.
> >
> > Vendor is a noun, not a verb.
>
> It's a curse, to be precise, and it slows down the project a lot.
> We have to worry about over 400 packages!!!
> Bloody hell, pardon my UK French, we have more interesting problems to
> work on...
>
>
> FWIW, I consider Marc's effort to provide a self-contained macOS app of
> Sage to be very valuable for the project.
>
> Dima is indeed using the word "vendor" incorrectly. I'm not concerned
> about the grammar; I use it as a verb, too, in this specific context.
> For source code, "vendoring" (= including a copy of the source code of a
> dependency, modified or not, in the source code) is problematic for various
> well known reasons.
> But to accuse a binary distribution of "vendoring" is simply meaningless.
> A binary distribution is made for the benefit of users.
>

I am not accusing Marc of vendoring, I am saying that him asking to supply
essentially all the components of Sage in source form is asking too much,
as it saddles us with extremely heavy maintenance task, for reasons
questionable.

Suppose we remove gmp from the list of packages we vendor (yes, "to
vendor", i.e. "to perform vendoring", may be a verb, as we see here. As
Americans make verbs out of everything, I find myself being gaslighted here
by one of them.)

What then, will it make the task of making a macOS Sage app impossible, or
much harder, all of a sudden?
I have trouble believing this.


> --
> You received this message because you are subscribed to the Google Groups
> "sage-devel" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to sage-devel+unsubscr...@googlegroups.com.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/sage-devel/fa119cf9-49bc-4b16-b6af-b1b6fd6584aen%40googlegroups.com
> 
> .
>

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/CAAWYfq3zx9DeBB5%3DvsO1%2BDC7UHugKyykaVUOC4Z1-tD3xfKDtQ%40mail.gmail.com.

Re: [sage-devel] Symengine package fails to build in 10.2rc2

[Marc Culler]

I see your unanswered message "ping" on issue #456 from 5 days ago.

But in the Releases section I see that v0.11.0 was released 2 days ago.

- Marc   


On Tuesday, November 14, 2023 at 11:50:55 AM UTC-6 Dima Pasechnik wrote:

> On Tue, Nov 14, 2023 at 5:18 PM Marc Culler  wrote:
> >
> > The symengine package has apparently not been updated to be compatible 
> with cython 3.
>
> https://github.com/symengine/symengine.py/issues/456
>
>
> >
> > This is what I see in the log (full log attached):
> >
> > performance hint: 
> /private/var/tmp/sage-10.2-current/local/var/lib/sage/venv-python3.11.1/var/tmp/sage/build/symengine_py-0.10.0/src/symengine/lib/symengine_wrapper.pyx:4992:5:
>  
> Exception check on '_ctypes_callback_lambda_real' will always require the 
> GIL to be acquired.
> > Possible solutions:
> > 1. Declare the function as 'noexcept' if you control the definition and 
> you're sure you don't want the function to raise exceptions.
> > 2. Use an 'int' return type on the function to allow an error code to be 
> returned.
> >
> > Error compiling Cython file:
> > 
> > ...
> > cdef Basic X = sympify((o1).diff(c2py(symbol)._sympy_()))
> > return X.thisptr
> >
> > def create_sympy_module():
> > cdef PyModule s = PyModule.__new__(PyModule)
> > s.thisptr = symengine.make_rcp_PyModule(_to_sympy, 
> _to_symengine, _eval,
> > ^
> > 
> >
> > 
> /private/var/tmp/sage-10.2-current/local/var/lib/sage/venv-python3.11.1/var/tmp/sage/build/symengine_py-0.10.0/src/symengine/lib/symengine_wrapper.pyx:2805:44:
>  
> Cannot assign type 'PyObject *(*)(rcp_const_basic) except? NULL' to 
> 'PyObject *(*)(rcp_const_basic) noexcept'. Exception values are 
> incompatible. Suggest adding 'noexcept' to type 'PyObject 
> *(rcp_const_basic) except? NULL'.
> > make[7]: *** [symengine/lib/symengine_wrapper.cpp] Error 1
> > make[6]: *** [symengine/lib/CMakeFiles/symengine_wrapper.dir/all] Error 2
> > make[5]: *** [all] Error 2
> > error: error building project
> > error: subprocess-exited-with-error
> >
> > --
> > You received this message because you are subscribed to the Google 
> Groups "sage-devel" group.
> > To unsubscribe from this group and stop receiving emails from it, send 
> an email to sage-devel+...@googlegroups.com.
> > To view this discussion on the web visit 
> https://groups.google.com/d/msgid/sage-devel/4fd16cfa-59fc-4571-b7a8-1ac5caeda95dn%40googlegroups.com
> .
>

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/6db1f1d2-b942-4e1b-9cfe-fa9cc9e792d5n%40googlegroups.com.

Re: [sage-devel] Re: Question about make dependencies

[Matthias Koeppe]

On Tuesday, November 14, 2023 at 11:39:52 AM UTC-8 Dima Pasechnik wrote:

On Tue, Nov 14, 2023 at 7:22 PM Marc Culler  wrote: 
> On Tue, Nov 14, 2023 at 11:58 AM Dima Pasechnik  wrote: 
>> I imagine it might be an artifact of the design of Sage on Mac app 
>> Marc is releasing: 
>> vendor everything. 
> 
> Vendor is a noun, not a verb. 

It's a curse, to be precise, and it slows down the project a lot. 
We have to worry about over 400 packages!!! 
Bloody hell, pardon my UK French, we have more interesting problems to 
work on...


FWIW, I consider Marc's effort to provide a self-contained macOS app of 
Sage to be very valuable for the project.

Dima is indeed using the word "vendor" incorrectly. I'm not concerned about 
the grammar; I use it as a verb, too, in this specific context.
For source code, "vendoring" (= including a copy of the source code of a 
dependency, modified or not, in the source code) is problematic for various 
well known reasons.
But to accuse a binary distribution of "vendoring" is simply meaningless. A 
binary distribution is made for the benefit of users.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/fa119cf9-49bc-4b16-b6af-b1b6fd6584aen%40googlegroups.com.

Re: [sage-devel] Re: Question about make dependencies

[Dima Pasechnik]

On Tue, Nov 14, 2023 at 7:22 PM Marc Culler  wrote:
>
> I was hesitant to send that message because I didn't want to have to go 
> through this again.
>
> On Tue, Nov 14, 2023 at 11:58 AM Dima Pasechnik  wrote:
>>
>> > Does ./configure say why the system copy is unsuitable?
>
>
> There is no system copy.
>
>> I imagine it might be an artifact of the design of Sage on Mac app
>> Marc is releasing:
>
>
> Very likely.
>
>> vendor everything.
>
>
> Vendor is a noun, not a verb.

It's a curse, to be precise, and it slows down the project a lot.
We have to worry  about over 400 packages!!!
Bloody hell, pardon my UK French, we have more interesting problems to
work on...


>
>> Is Apple's software distribution system so dumb that this is the only way?
>
>
> If you think that Apple's requirement that apps be notarized is dumb, then I 
> encourage you to take it up with Apple.  Maybe Don Quixote will join you in 
> the effort.

Apple sucks, but still I don't see how this is relevant here.
You need to package a number of external libraries in your bundle,
why do you need them to be built all at the same swoop?
If a commercial vendor gives you a library, without full source, then
you cannot package it in your app?
Sorry, this sounds rather improbable.
Can't you base your app on a Python.org Python bundle, instead, to begin with?

>
> I am serving people who use macOS; more than 8.800 of them in the case of 
> Sage 10.1.  I am fairly confident in claiming that the vast majority of Sage 
> 10.1 users are using it on macOS from the Sage_macOS app.  Even homebrew 
> users are using that app, since it is what gets installed when you run "brew 
> install sage" , which works fine by the way.

They might well be using Conda.

And Homebrew users might very well build Sage from source, as it's
relatively quick thanks to a large
number of libraries they can use.

>
> When I started this project you were griping about how many sage-support 
> problems were related to macOS installation.  I don't think you can claim 
> that to be a problem any more.

It's still the most cursed platform we need to worry about.

>
> - Marc
>
> --
> You received this message because you are subscribed to the Google Groups 
> "sage-devel" group.
> To unsubscribe from this group and stop receiving emails from it, send an 
> email to sage-devel+unsubscr...@googlegroups.com.
> To view this discussion on the web visit 
> https://groups.google.com/d/msgid/sage-devel/CALcZXREOLVXQemMvTbVZDgf%3DWT3zOi-tmokJV_3%3DX%3D0zDO-cXw%40mail.gmail.com.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/CAAWYfq3Y1r2UxGjj_KjufdxyCRd09r6tpOFMC7uP1zyVMpi9zQ%40mail.gmail.com.

Re: [sage-devel] Re: Question about make dependencies

[Marc Culler]

Thanks, Matthias.  I will try to remember to check the timestamps next time.

- Marc

On Tue, Nov 14, 2023 at 12:56 PM Matthias Koeppe 
wrote:

> I don't think this happens in normal development. When this happens in
> your setup, check the mtime of the timestamps in
> local/var/lib/sage/installed.
>
> On Tuesday, November 14, 2023 at 9:37:09 AM UTC-8 Marc Culler wrote:
>
>> For me, one of the most frustrating parts of building sage is that,
>> whenever a build fails, the make system decides that gmp is out of date and
>> hence I have to wait for it to recompile gmp and all of its dependencies.
>>
>> Is this intentional?
>>
>> Does anyone know why it happens?
>>
>> Is it avoidable?
>>
>> Or am I the only one who sees this?
>>
>> - Marc
>>
>> --
> You received this message because you are subscribed to a topic in the
> Google Groups "sage-devel" group.
> To unsubscribe from this topic, visit
> https://groups.google.com/d/topic/sage-devel/rBvXdVKC4nw/unsubscribe.
> To unsubscribe from this group and all its topics, send an email to
> sage-devel+unsubscr...@googlegroups.com.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/sage-devel/96d12845-208c-46cb-966b-8a991ef4a68cn%40googlegroups.com
> 
> .
>

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/CALcZXRE3M%2BuV-Hd69%2BJdt3aAEZJcW8JhzFi0%3DpcdB%2BgFGndBrA%40mail.gmail.com.

Re: [sage-devel] Re: Question about make dependencies

[Marc Culler]

I was hesitant to send that message because I didn't want to have to go
through this again.

On Tue, Nov 14, 2023 at 11:58 AM Dima Pasechnik  wrote:

> > Does ./configure say why the system copy is unsuitable?
>

There is no system copy.

I imagine it might be an artifact of the design of Sage on Mac app
> Marc is releasing:
>

Very likely.

vendor everything.
>

Vendor is a noun, not a verb.

Is Apple's software distribution system so dumb that this is the only way?
>

If you think that Apple's requirement that apps be notarized is dumb, then
I encourage you to take it up with Apple.  Maybe Don Quixote will join you
in the effort.

I am serving people who use macOS; more than 8.800 of them in the case of
Sage 10.1.  I am fairly confident in claiming that the vast majority of
Sage 10.1 users are using it on macOS from the Sage_macOS app.  Even
homebrew users are using that app, since it is what gets installed when you
run "brew install sage" , which works
fine by the way.

When I started this project you were griping about how many sage-support
problems were related to macOS installation.  I don't think you can claim
that to be a problem any more.

- Marc

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/CALcZXREOLVXQemMvTbVZDgf%3DWT3zOi-tmokJV_3%3DX%3D0zDO-cXw%40mail.gmail.com.

[sage-devel] Re: Question about make dependencies

[Matthias Koeppe]

I don't think this happens in normal development. When this happens in your 
setup, check the mtime of the timestamps in local/var/lib/sage/installed.

On Tuesday, November 14, 2023 at 9:37:09 AM UTC-8 Marc Culler wrote:

> For me, one of the most frustrating parts of building sage is that, 
> whenever a build fails, the make system decides that gmp is out of date and 
> hence I have to wait for it to recompile gmp and all of its dependencies.
>
> Is this intentional?
>
> Does anyone know why it happens?
>
> Is it avoidable?
>
> Or am I the only one who sees this?
>
> - Marc
>
>

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/96d12845-208c-46cb-966b-8a991ef4a68cn%40googlegroups.com.

Re: [sage-devel] Re: Question about make dependencies

[Dima Pasechnik]

On Tue, Nov 14, 2023 at 5:50 PM Michael Orlitzky  wrote:
>
> On Tue, 2023-11-14 at 09:42 -0800, Marc Culler wrote:
> > Of course I meant that I have to wait for everything that *depends on* gmp
> > to be recompiled.  Also, this happens when there is nothing wrong with the
> > gmp build.  The make system decides that it is out of date even though the
> > build was successful and the package was installed.
>
> I think most of us have the system copy of gmp detected by ./configure,
> so there could very well be a bug that we haven't noticed. But ideally
> you should also be using the system copy of gmp.
>
> Does ./configure say why the system copy is unsuitable?

I imagine it might be an artifact of the design of Sage on Mac app
Marc is releasing:
vendor everything.
Is Apple's software distribution system so dumb that this is the only way?
I hope it is not, and we can get rid of "sage toolchain" kludge which
is most probably behind
this weirdness...


>
> --
> You received this message because you are subscribed to the Google Groups 
> "sage-devel" group.
> To unsubscribe from this group and stop receiving emails from it, send an 
> email to sage-devel+unsubscr...@googlegroups.com.
> To view this discussion on the web visit 
> https://groups.google.com/d/msgid/sage-devel/25eeaa1d9b467afe24e61994e324a4f120c49b75.camel%40orlitzky.com.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/CAAWYfq18s8LeOkVAbGmT-e3fUFYRTCeupVJvN_OW9_TRWw8xZw%40mail.gmail.com.

Re: [sage-devel] Symengine package fails to build in 10.2rc2

[Dima Pasechnik]

On Tue, Nov 14, 2023 at 5:18 PM Marc Culler  wrote:
>
> The symengine package has apparently not been updated to be compatible with 
> cython 3.

https://github.com/symengine/symengine.py/issues/456


>
> This is what I see in the log (full log attached):
>
>   performance hint: 
> /private/var/tmp/sage-10.2-current/local/var/lib/sage/venv-python3.11.1/var/tmp/sage/build/symengine_py-0.10.0/src/symengine/lib/symengine_wrapper.pyx:4992:5:
>  Exception check on '_ctypes_callback_lambda_real' will always require the 
> GIL to be acquired.
>   Possible solutions:
> 1. Declare the function as 'noexcept' if you control the definition 
> and you're sure you don't want the function to raise exceptions.
> 2. Use an 'int' return type on the function to allow an error code to 
> be returned.
>
>   Error compiling Cython file:
>   
>   ...
>   cdef Basic X = sympify((o1).diff(c2py(symbol)._sympy_()))
>   return X.thisptr
>
>   def create_sympy_module():
>   cdef PyModule s = PyModule.__new__(PyModule)
>   s.thisptr = symengine.make_rcp_PyModule(_to_sympy, 
> _to_symengine, _eval,
>   ^
>   
>
>   
> /private/var/tmp/sage-10.2-current/local/var/lib/sage/venv-python3.11.1/var/tmp/sage/build/symengine_py-0.10.0/src/symengine/lib/symengine_wrapper.pyx:2805:44:
>  Cannot assign type 'PyObject *(*)(rcp_const_basic) except? NULL' to 
> 'PyObject *(*)(rcp_const_basic) noexcept'. Exception values are incompatible. 
> Suggest adding 'noexcept' to type 'PyObject *(rcp_const_basic) except? NULL'.
>   make[7]: *** [symengine/lib/symengine_wrapper.cpp] Error 1
>   make[6]: *** [symengine/lib/CMakeFiles/symengine_wrapper.dir/all] Error 2
>   make[5]: *** [all] Error 2
>   error: error building project
>   error: subprocess-exited-with-error
>
> --
> You received this message because you are subscribed to the Google Groups 
> "sage-devel" group.
> To unsubscribe from this group and stop receiving emails from it, send an 
> email to sage-devel+unsubscr...@googlegroups.com.
> To view this discussion on the web visit 
> https://groups.google.com/d/msgid/sage-devel/4fd16cfa-59fc-4571-b7a8-1ac5caeda95dn%40googlegroups.com.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/CAAWYfq24%3DM9_YfwUjp3-vsc7E4Qus6%3DuqtxFfXeWxZXVS5Oarw%40mail.gmail.com.

Re: [sage-devel] Re: Question about make dependencies

[Michael Orlitzky]

On Tue, 2023-11-14 at 09:42 -0800, Marc Culler wrote:
> Of course I meant that I have to wait for everything that *depends on* gmp 
> to be recompiled.  Also, this happens when there is nothing wrong with the 
> gmp build.  The make system decides that it is out of date even though the 
> build was successful and the package was installed.

I think most of us have the system copy of gmp detected by ./configure,
so there could very well be a bug that we haven't noticed. But ideally
you should also be using the system copy of gmp.

Does ./configure say why the system copy is unsuitable?

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/25eeaa1d9b467afe24e61994e324a4f120c49b75.camel%40orlitzky.com.

Re: [sage-devel] Question about make dependencies

[Dima Pasechnik]

On Tue, Nov 14, 2023 at 5:37 PM Marc Culler  wrote:
>
> For me, one of the most frustrating parts of building sage is that, whenever 
> a build fails, the make system decides that gmp is out of date and hence I 
> have to wait for it to recompile gmp and all of its dependencies.
>
> Is this intentional?
>
> Does anyone know why it happens?
>
> Is it avoidable?
>
> Or am I the only one who sees this?

you are probably the only one who does not use a prebuilt gmp :-)

>
> - Marc
>
> --
> You received this message because you are subscribed to the Google Groups 
> "sage-devel" group.
> To unsubscribe from this group and stop receiving emails from it, send an 
> email to sage-devel+unsubscr...@googlegroups.com.
> To view this discussion on the web visit 
> https://groups.google.com/d/msgid/sage-devel/233fc79f-0564-4a90-aade-3c7882cec5b2n%40googlegroups.com.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/CAAWYfq0vOxKfa2%3DynvbxsxEz_RHpZ0jx6oEpKG8-N%2Bdh8sS82w%40mail.gmail.com.

[sage-devel] Re: Question about make dependencies

[Marc Culler]

Of course I meant that I have to wait for everything that *depends on* gmp 
to be recompiled.  Also, this happens when there is nothing wrong with the 
gmp build.  The make system decides that it is out of date even though the 
build was successful and the package was installed.

- Marc

On Tuesday, November 14, 2023 at 11:37:09 AM UTC-6 Marc Culler wrote:

> For me, one of the most frustrating parts of building sage is that, 
> whenever a build fails, the make system decides that gmp is out of date and 
> hence I have to wait for it to recompile gmp and all of its dependencies.
>
> Is this intentional?
>
> Does anyone know why it happens?
>
> Is it avoidable?
>
> Or am I the only one who sees this?
>
> - Marc
>
>

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/ab50a937-2ac2-4610-86c3-b954451dd84an%40googlegroups.com.

[sage-devel] Question about make dependencies

[Marc Culler]

For me, one of the most frustrating parts of building sage is that, 
whenever a build fails, the make system decides that gmp is out of date and 
hence I have to wait for it to recompile gmp and all of its dependencies.

Is this intentional?

Does anyone know why it happens?

Is it avoidable?

Or am I the only one who sees this?

- Marc

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/233fc79f-0564-4a90-aade-3c7882cec5b2n%40googlegroups.com.

[sage-devel] Symengine package fails to build in 10.2rc2

[Marc Culler]

The symengine package has apparently not been updated to be compatible with 
cython 3.

This is what I see in the log (full log attached):

  performance hint: 
/private/var/tmp/sage-10.2-current/local/var/lib/sage/venv-python3.11.1/var/tmp/sage/build/symengine_py-0.10.0/src/symengine/lib/symengine_wrapper.pyx:4992:5:
 
Exception check on '_ctypes_callback_lambda_real' will always require the 
GIL to be acquired.
  Possible solutions:
1. Declare the function as 'noexcept' if you control the definition 
and you're sure you don't want the function to raise exceptions.
2. Use an 'int' return type on the function to allow an error code 
to be returned.

  Error compiling Cython file:
  
  ...
  cdef Basic X = sympify((o1).diff(c2py(symbol)._sympy_()))
  return X.thisptr

  def create_sympy_module():
  cdef PyModule s = PyModule.__new__(PyModule)
  s.thisptr = symengine.make_rcp_PyModule(_to_sympy, 
_to_symengine, _eval,
  ^
  

  
/private/var/tmp/sage-10.2-current/local/var/lib/sage/venv-python3.11.1/var/tmp/sage/build/symengine_py-0.10.0/src/symengine/lib/symengine_wrapper.pyx:2805:44:
 
Cannot assign type 'PyObject *(*)(rcp_const_basic) except? NULL' to 
'PyObject *(*)(rcp_const_basic) noexcept'. Exception values are 
incompatible. Suggest adding 'noexcept' to type 'PyObject 
*(rcp_const_basic) except? NULL'.
  make[7]: *** [symengine/lib/symengine_wrapper.cpp] Error 1
  make[6]: *** [symengine/lib/CMakeFiles/symengine_wrapper.dir/all] Error 2
  make[5]: *** [all] Error 2
  error: error building project
  error: subprocess-exited-with-error

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/4fd16cfa-59fc-4571-b7a8-1ac5caeda95dn%40googlegroups.com.
Using cached file 
/private/var/tmp/sage-10.2-current/upstream/symengine.py-0.10.0.tar.gz
symengine_py-0.10.0

Setting up build directory for symengine_py-0.10.0
Finished extraction
No patch files found in ../patches

Host system:
Darwin abner.local 22.5.0 Darwin Kernel Version 22.5.0: Mon Apr 24 20:51:50 PDT 
2023; root:xnu-8796.121.2~5/RELEASE_X86_64 x86_64

C compiler: gcc
C compiler version:
Apple clang version 13.1.6 (clang-1316.0.21.2.5)
Target: x86_64-apple-darwin22.5.0
Thread model: posix
InstalledDir: 
/Applications/Xcode_13.4.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/bin

Package 'symengine_py' is currently not installed
No legacy uninstaller found for 'symengine_py'; nothing to do
Installing symengine_py-0.10.0
Looking in links: 
/private/var/tmp/sage-10.2-current/local/var/lib/sage/venv-python3.11.1/var/lib/sage/wheels
Processing 
/private/var/tmp/sage-10.2-current/local/var/lib/sage/venv-python3.11.1/var/tmp/sage/build/symengine_py-0.10.0/src
  Preparing metadata (setup.py): started
  Running command python setup.py egg_info
  
/private/var/tmp/sage-10.2-current/local/var/lib/sage/venv-python3.11.1/lib/python3.11/site-packages/setuptools/__init__.py:80:
 _DeprecatedInstaller: setuptools.installer and fetch_build_eggs are deprecated.
  !!

  

  Requirements should be satisfied by a PEP 517 installer.
  If you are using pip, you can try `pip install --use-pep517`.
  


  !!
dist.fetch_build_eggs(dist.setup_requires)
  running egg_info
  creating 
/private/var/folders/sl/wck1hwwx6pv2mcl04pv2092wgn/T/pip-pip-egg-info-2xeqdjam/symengine.egg-info
  writing 
/private/var/folders/sl/wck1hwwx6pv2mcl04pv2092wgn/T/pip-pip-egg-info-2xeqdjam/symengine.egg-info/PKG-INFO
  writing dependency_links to 
/private/var/folders/sl/wck1hwwx6pv2mcl04pv2092wgn/T/pip-pip-egg-info-2xeqdjam/symengine.egg-info/dependency_links.txt
  writing top-level names to 
/private/var/folders/sl/wck1hwwx6pv2mcl04pv2092wgn/T/pip-pip-egg-info-2xeqdjam/symengine.egg-info/top_level.txt
  writing manifest file 
'/private/var/folders/sl/wck1hwwx6pv2mcl04pv2092wgn/T/pip-pip-egg-info-2xeqdjam/symengine.egg-info/SOURCES.txt'
  reading manifest file 
'/private/var/folders/sl/wck1hwwx6pv2mcl04pv2092wgn/T/pip-pip-egg-info-2xeqdjam/symengine.egg-info/SOURCES.txt'
  reading manifest template 'MANIFEST.in'
  warning: no files found matching '*.hpp' under directory 'symengine'
  warning: no files found matching 

Re: [sage-devel] Sage-relevant environment variables

[John Cremona]

Thajd Dima, very helpful as always!  I can't believe that after years of
using git I did not know about git grep (but I'm sure that you can believe
it).

Incidentally, setting SAGE_ATLAS_LIB was one of the early ways to help
developers by saying to use a system build of atlas which otherwise took a
very long time.  We have come a long way since then.

John

On Tue, 14 Nov 2023 at 14:59, Dima Pasechnik  wrote:

> Of these listed, only SAGE_INSTALL_CCACHE and SAGE_MATPLOTLIB_GUI
> are mentioned in Sage's source, as you can see by running
>
> git grep SAGE_MATPLOTLIB_GUI
> git grep foobar
>
> etc
>
> On Tue, Nov 14, 2023 at 2:33 PM John Cremona 
> wrote:
> >
> > I have machines I have been building Sage on (and using for development)
> for 10 years or more, and they have several environment variables of the
> form SAGE* which I set in my .bashrc script.  I suspect that all or most of
> these are redundant; some may even be harmful.  Is there a list somewhere
> of which envoronment variables *are* currently used?
> >
> > Here's an example:
> >
> > jec@lehner:~/sage (branch: develop)$ env | grep SAGE
> > SAGE_MATPLOTLIB_GUI=yes
> > SAGE_PARALLEL_SPKG_BUILD=yes
> > SAGE_ATLAS_LIB=/usr/lib
> >
> > Another also has
> > jec@atkin:/usr/local/include$ env | grep SAGE
> > SAGE_UPGRADING=yes
> > SAGE_INSTALL_CCACHE=yes
> >
> > John
> >
> >
> > --
> > You received this message because you are subscribed to the Google
> Groups "sage-devel" group.
> > To unsubscribe from this group and stop receiving emails from it, send
> an email to sage-devel+unsubscr...@googlegroups.com.
> > To view this discussion on the web visit
> https://groups.google.com/d/msgid/sage-devel/CAD0p0K4is68YT6zb%2B8mrjOiB-86FASFSWj%2B4%2Br%3Din90P8sn18w%40mail.gmail.com
> .
>
> --
> You received this message because you are subscribed to the Google Groups
> "sage-devel" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to sage-devel+unsubscr...@googlegroups.com.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/sage-devel/CAAWYfq2VifUqRABxL_M1_EPNuA0LTVMbVUzDPP6VmLAf7eNU%3Dg%40mail.gmail.com
> .
>

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/CAD0p0K5NAg1Ri%2BVaJRyJuGzZyQpKhgJaf7%3DPyyNfbW%3D1NUjHOg%40mail.gmail.com.

[sage-devel] Relative Simplicial Complex

[Andrés Moreno]

Hi everyone,

I am collaborating with a colleague to somehow extend the class 
SimplicialComplex to include functionalities of relative simplicial 
complexes. We've started coding but seek your input before moving forward, 
particularly on the initial design of the class.

Initially, we considered designing a new class with with a constructor 
having as an input two instances of the class SimplicialComplex, namely a 
complex and its subcomplex. However, we're now leaning towards inheriting 
directly from SimplicialComplex and adding a property for the relative 
part. This approach seems more logical, allowing us access to methods and 
behaviors from Sage's base classes.

We've also noted that some existing methods in Sage already include 
parameters for relative parts. An alternative approach could be to 
integrate this functionality into all existing methods.
We value your thoughts and feedback on these approaches, as we aim to make 
a meaningful contribution to the software.

Best regards,

Andres Moreno

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/3cea93fa-9784-488a-9696-e76cbdacbccdn%40googlegroups.com.

Re: [sage-devel] Sage-relevant environment variables

[Dima Pasechnik]

Of these listed, only SAGE_INSTALL_CCACHE and SAGE_MATPLOTLIB_GUI
are mentioned in Sage's source, as you can see by running

git grep SAGE_MATPLOTLIB_GUI
git grep foobar

etc

On Tue, Nov 14, 2023 at 2:33 PM John Cremona  wrote:
>
> I have machines I have been building Sage on (and using for development) for 
> 10 years or more, and they have several environment variables of the form 
> SAGE* which I set in my .bashrc script.  I suspect that all or most of these 
> are redundant; some may even be harmful.  Is there a list somewhere of which 
> envoronment variables *are* currently used?
>
> Here's an example:
>
> jec@lehner:~/sage (branch: develop)$ env | grep SAGE
> SAGE_MATPLOTLIB_GUI=yes
> SAGE_PARALLEL_SPKG_BUILD=yes
> SAGE_ATLAS_LIB=/usr/lib
>
> Another also has
> jec@atkin:/usr/local/include$ env | grep SAGE
> SAGE_UPGRADING=yes
> SAGE_INSTALL_CCACHE=yes
>
> John
>
>
> --
> You received this message because you are subscribed to the Google Groups 
> "sage-devel" group.
> To unsubscribe from this group and stop receiving emails from it, send an 
> email to sage-devel+unsubscr...@googlegroups.com.
> To view this discussion on the web visit 
> https://groups.google.com/d/msgid/sage-devel/CAD0p0K4is68YT6zb%2B8mrjOiB-86FASFSWj%2B4%2Br%3Din90P8sn18w%40mail.gmail.com.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/CAAWYfq2VifUqRABxL_M1_EPNuA0LTVMbVUzDPP6VmLAf7eNU%3Dg%40mail.gmail.com.

[sage-devel] Sage-relevant environment variables

[John Cremona]

I have machines I have been building Sage on (and using for development)
for 10 years or more, and they have several environment variables of the
form SAGE* which I set in my .bashrc script.  I suspect that all or most of
these are redundant; some may even be harmful.  Is there a list somewhere
of which envoronment variables *are* currently used?

Here's an example:

jec@lehner:~/sage (branch: develop)$ env | grep SAGE
SAGE_MATPLOTLIB_GUI=yes
SAGE_PARALLEL_SPKG_BUILD=yes
SAGE_ATLAS_LIB=/usr/lib

Another also has
jec@atkin:/usr/local/include$ env | grep SAGE
SAGE_UPGRADING=yes
SAGE_INSTALL_CCACHE=yes

John

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/CAD0p0K4is68YT6zb%2B8mrjOiB-86FASFSWj%2B4%2Br%3Din90P8sn18w%40mail.gmail.com.

Re: [sage-devel] Accidential push to develop

[Dima Pasechnik]

Hi,

I took the liberty to fix the develop branch by force-pushing 10.2.rc2
branch.
All should be in order now, and PRs can proceed.

A copy of the accidentally pushed to develop is here:
https://github.com/dimpase/sage/tree/develop_accidentally_pushed_to_



On Tue, Nov 14, 2023 at 9:43 AM Dima Pasechnik  wrote:
>
> On Tue, Nov 14, 2023 at 9:29 AM tobia...@gmx.de  wrote:
> >
> > Hi everybody,
> >
> > I just accidentally pushed to the develop branch (instead of to a new
branch in my fork). I'm very sorry! I leave it to the release manager to
revert/fix it to not introduce more issues.
> >
> > What confuses me, however, is how this was possible in first place?! I
thought we had branch protection rules in place that prevented such things.
Did someone recently changed something in these rules?
>
> Maybe github had introduced a bug; anyway, " Restrict who can push to
> matching branches " is not
> on for develop [1], and  this seems to mean that non-force pushed may
> be done by anyone in an appropriate team.
>
> I've switched it on now, for the time being.
>
> (force pushed are limited to Volker)
>
> [1]
https://github.com/sagemath/sage/settings/branch_protection_rules/33965567
>
>
> >
> > --
> > You received this message because you are subscribed to the Google
Groups "sage-devel" group.
> > To unsubscribe from this group and stop receiving emails from it, send
an email to sage-devel+unsubscr...@googlegroups.com.
> > To view this discussion on the web visit
https://groups.google.com/d/msgid/sage-devel/842f9089-e9cc-4c5d-b706-1eefc055e455n%40googlegroups.com
.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/CAAWYfq278D-1sp8vjb27Q-aw4-PQgARDp%3DZu4jBEQFKpV9VbOg%40mail.gmail.com.

Re: [sage-devel] Accidential push to develop

[Dima Pasechnik]

On Tue, Nov 14, 2023 at 9:29 AM tobia...@gmx.de  wrote:
>
> Hi everybody,
>
> I just accidentally pushed to the develop branch (instead of to a new branch 
> in my fork). I'm very sorry! I leave it to the release manager to revert/fix 
> it to not introduce more issues.
>
> What confuses me, however, is how this was possible in first place?! I 
> thought we had branch protection rules in place that prevented such things. 
> Did someone recently changed something in these rules?

Maybe github had introduced a bug; anyway, " Restrict who can push to
matching branches " is not
on for develop [1], and  this seems to mean that non-force pushed may
be done by anyone in an appropriate team.

I've switched it on now, for the time being.

(force pushed are limited to Volker)

[1] https://github.com/sagemath/sage/settings/branch_protection_rules/33965567


>
> --
> You received this message because you are subscribed to the Google Groups 
> "sage-devel" group.
> To unsubscribe from this group and stop receiving emails from it, send an 
> email to sage-devel+unsubscr...@googlegroups.com.
> To view this discussion on the web visit 
> https://groups.google.com/d/msgid/sage-devel/842f9089-e9cc-4c5d-b706-1eefc055e455n%40googlegroups.com.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/CAAWYfq3pAAf0M%3D-YJ2U1hF8KtSKqxKfpi%3DWnYjNFUNu4gvpEXg%40mail.gmail.com.

[sage-devel] Accidential push to develop

[tobia...@gmx.de]

Hi everybody,

I just accidentally pushed to the develop branch (instead of to a new 
branch in my fork). I'm very sorry! I leave it to the release manager to 
revert/fix it to not introduce more issues.

What confuses me, however, is how this was possible in first place?! I 
thought we had branch protection rules in place that prevented such things. 
Did someone recently changed something in these rules?

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/842f9089-e9cc-4c5d-b706-1eefc055e455n%40googlegroups.com.