Ladies and gentlemen,



Some of you have been asking NV Access and the global NVDA community
regarding porting NVDA source code and dependencies to Python 3. I and some
contributors have been brainstorming best ways to accomplish this, and NV
Access, contributors, some add-on writers, users, and NVDA observers agree
that it is important that we look into eventually porting NVDA to Python 3.



Until recently, the possibility of moving NVDA to Python 3 seemed to be only
a dream. However, a major hurdle was overcome earlier this year when
wxPython 4 alpha was released, promising support for Python versions 2.7 and
3.x. Apart from a major problem regarding certain internal routines from
wxPython 4 that may not be compatible with NVDA source code, the stage is
almost set to begin our transition to Python 3, dubbed "Project Pylennium".



Project Pylennium is a new project I'm planning and working on that should
make it easier for NV Access, code contributors, add-on writers, users and
many others to experience smooth transition to Python 3. The overall goal of
this project is to prepare NVDA's source code so it can run on both Python 2
and 3 in most cases until developers are ready to move on. This project also
seeks to uncover possible compatibility issues for NVDA community when the
transition happens, and to document these findings so it can serve as
guidelines to current and future developers.



The current iteration of Project Pylennium (based on NVDA GitHub issue 7105
as a roadmap) can be found on my copy of NVDA source code at:

https://github.com/josephsl/nvda

The branch to use is "py3000".



Project pylennium involves the following:



1. Gather dependencies required for Python 3 transition, such as newer
versions of Python, ConfigObj and many others. A crucial dependency is the
Six module, which adds compatibility layer functions for both Python 2 and
3.
2. Document porting notes, which will eventually be published on our
NVDA Community wiki.
3. Create source-level compatibility layer in order to mitigate changes
in module names, standard library folder structure, function and class names
and so on.
4. Migrate most code to Python 3 syntax.
5. Frequent and regular testing to make sure regressions are not
introduced.



Progress so far:



1. Imports: a compatibility layer is in place to import Python 2 or 3
version of winreg, pickle, io.StringIO and what not.
2. Compiles fine via SCons 2.x.
3. Possible to see the first initialization phase of NVDA launcher
(nvda.pyw) via command line. The biggest showstopper at this time is the
fact that logHandler.StreamHandler constructor uses file function when
Python 3 says it isn't available, which suggests we need to modernize log
handler module.



Benefits:



1. Allows NV Access to look at porting data and make informed
decisions.
2. Allows contributors and Python experts to provide expertise and
prepare the source code for Python 3 transition.
3. Allows add-on authors to look at changes and prepare their add-ons
to run on Python 3.
4. Allows interested users to be involved in a major project and
provide test data.



Disadvantages:



1. We need to part ways with Windows XP and Server 2003.
2. Stress throughout the project, as it is nothing like what we've seen
recently.
3. Some add-ons will need community intervention to get them to run on
Python 3, as some are abandonware (no longer maintained by authors or the
add-ons community don't know who wrote some of them and lost contact with
some authors).
4. Possible issues in code and other ones, especially when looking at
code and attitude conflicts.



Next steps:



1. I'm looking for volunteers and subproject leads: if you are
passionate about Python, NVDA, and want to help NVDA achieve the next
milestone, please help and find folks who can assist in this project. In
regards to subprojects, mostly have to do with dependency checks and tests,
testing, documentation, talking to Python Software Foundation (PSF),
recruiting Python experts and what not.
2. Cooperation: as Project Pylennium is a grand-scale project, we
cannot move forward and make it a reality unless cooperation between
developers, add-on authors, users, observers and many others is there.
3. Promote: one way to keep the momentum going is promoting this
project far and wide.
4. Continued work: I'm looking into how to port urllib imports, as this
module has seen huge folder restructuring in Python 3 and other work.



What you'll need for this project:



1. Python 3.6.0 32-bit or later (3.6.2 is recommended).
2. wxPython 4 alpha 3 or later with the wheel for Python 3.5 or 3.6.
3. Six module (from PYPI).
4. ConfigObj 5.0.6.



Guidelines:



1. When sending pull requests, please do NOT send them to NV Access
yet, as folks over there are working on high priority work. Instead, send a
pull request against josephsl/nvda branch py3000.
2. Please provide a clear commit message when sending in pull requests
so folks can track what has changed.
3. Before you commit, please make sure to test NVDA in order to catch
regressions and write workarounds.
4. At least once a month, I will merge NV Access master branch to keep
the code up to date and incorporate changes into this project, sometimes
requiring transforming syntax to fit Python 3 requirements.
5. If you want to port add-ons, please do so at your earliest
convenience or when you are ready. As part of this work, I myself will port
my own add-ons to Python 3 (one of them is done; throughout this year, I'll
port rest of my add-ons). Please do not port your add-ons to Python 3
completely - I recommend making it so it'll run happily on both Python
versions.



Thank you.

Sincerely,

Joseph

thanks for good news joseph and brian.
do you continue supporting of xp?
i became worried when i read the message for the first time.
God bless you all!

On 7/22/17, Brian's Mail list account via Groups.Io
<bglists=blueyonder.co.uk@groups.io> wrote:

Sounds sensible. At least this will preserve xp etc with the old Python
for

as long as possible. One thing you do not say, which may be obvious to
you,

but perhaps not to the average user is..
What are the reasons for change.
is it that its going to be faster, 64 bit or has better ways of doing
stuff

nvda needs to be able to do in the future?
Obviously not in your remit is that at some point an agreed last working
version of the old NVDA that can work on xp etc is going to have to be
made

available and then when add ons are updated, those that will and won't
run
on the old version will need to be noted and if possible trapped by the
old

version so it cannot be scrambled by an add on.


Brian

bglists@blueyonder.co.uk
Sent via blueyonder.
Please address personal email to:-
briang1@blueyonder.co.uk, putting 'Brian Gaff'
in the display name field.
----- Original Message -----
From: "Joseph Lee" <joseph.lee22590@gmail.com>
To: <nvda@nvda.groups.io>
Sent: Saturday, July 22, 2017 7:55 AM
Subject: [nvda] Announcing Project Pylennium: porting NVDA code base,
dependencies, and developer attitudes to Python 3.6.x

Ladies and gentlemen,



Some of you have been asking NV Access and the global NVDA community
regarding porting NVDA source code and dependencies to Python 3. I and
some
contributors have been brainstorming best ways to accomplish this, and
NV
Access, contributors, some add-on writers, users, and NVDA observers
agree
that it is important that we look into eventually porting NVDA to Python
3.



Until recently, the possibility of moving NVDA to Python 3 seemed to be
only
a dream. However, a major hurdle was overcome earlier this year when
wxPython 4 alpha was released, promising support for Python versions 2.7
and
3.x. Apart from a major problem regarding certain internal routines from
wxPython 4 that may not be compatible with NVDA source code, the stage
is
almost set to begin our transition to Python 3, dubbed "Project
Pylennium".



Project Pylennium is a new project I'm planning and working on that
should
make it easier for NV Access, code contributors, add-on writers, users
and
many others to experience smooth transition to Python 3. The overall
goal

of
this project is to prepare NVDA's source code so it can run on both
Python

2
and 3 in most cases until developers are ready to move on. This project
also
seeks to uncover possible compatibility issues for NVDA community when
the
transition happens, and to document these findings so it can serve as
guidelines to current and future developers.



The current iteration of Project Pylennium (based on NVDA GitHub issue
7105
as a roadmap) can be found on my copy of NVDA source code at:

https://github.com/josephsl/nvda

The branch to use is "py3000".



Project pylennium involves the following:



1. Gather dependencies required for Python 3 transition, such as newer
versions of Python, ConfigObj and many others. A crucial dependency is
the
Six module, which adds compatibility layer functions for both Python 2
and
3.
2. Document porting notes, which will eventually be published on our
NVDA Community wiki.
3. Create source-level compatibility layer in order to mitigate changes
in module names, standard library folder structure, function and class
names
and so on.
4. Migrate most code to Python 3 syntax.
5. Frequent and regular testing to make sure regressions are not
introduced.



Progress so far:



1. Imports: a compatibility layer is in place to import Python 2 or 3
version of winreg, pickle, io.StringIO and what not.
2. Compiles fine via SCons 2.x.
3. Possible to see the first initialization phase of NVDA launcher
(nvda.pyw) via command line. The biggest showstopper at this time is the
fact that logHandler.StreamHandler constructor uses file function when
Python 3 says it isn't available, which suggests we need to modernize
log
handler module.



Benefits:



1. Allows NV Access to look at porting data and make informed
decisions.
2. Allows contributors and Python experts to provide expertise and
prepare the source code for Python 3 transition.
3. Allows add-on authors to look at changes and prepare their add-ons
to run on Python 3.
4. Allows interested users to be involved in a major project and
provide test data.



Disadvantages:



1. We need to part ways with Windows XP and Server 2003.
2. Stress throughout the project, as it is nothing like what we've seen
recently.
3. Some add-ons will need community intervention to get them to run on
Python 3, as some are abandonware (no longer maintained by authors or
the
add-ons community don't know who wrote some of them and lost contact
with
some authors).
4. Possible issues in code and other ones, especially when looking at
code and attitude conflicts.



Next steps:



1. I'm looking for volunteers and subproject leads: if you are
passionate about Python, NVDA, and want to help NVDA achieve the next
milestone, please help and find folks who can assist in this project. In
regards to subprojects, mostly have to do with dependency checks and
tests,
testing, documentation, talking to Python Software Foundation (PSF),
recruiting Python experts and what not.
2. Cooperation: as Project Pylennium is a grand-scale project, we
cannot move forward and make it a reality unless cooperation between
developers, add-on authors, users, observers and many others is there.
3. Promote: one way to keep the momentum going is promoting this
project far and wide.
4. Continued work: I'm looking into how to port urllib imports, as this
module has seen huge folder restructuring in Python 3 and other work.



What you'll need for this project:



1. Python 3.6.0 32-bit or later (3.6.2 is recommended).
2. wxPython 4 alpha 3 or later with the wheel for Python 3.5 or 3.6.
3. Six module (from PYPI).
4. ConfigObj 5.0.6.



Guidelines:



1. When sending pull requests, please do NOT send them to NV Access
yet, as folks over there are working on high priority work. Instead,
send

a
pull request against josephsl/nvda branch py3000.
2. Please provide a clear commit message when sending in pull requests
so folks can track what has changed.
3. Before you commit, please make sure to test NVDA in order to catch
regressions and write workarounds.
4. At least once a month, I will merge NV Access master branch to keep
the code up to date and incorporate changes into this project, sometimes
requiring transforming syntax to fit Python 3 requirements.
5. If you want to port add-ons, please do so at your earliest
convenience or when you are ready. As part of this work, I myself will
port
my own add-ons to Python 3 (one of them is done; throughout this year,
I'll
port rest of my add-ons). Please do not port your add-ons to Python 3
completely - I recommend making it so it'll run happily on both Python
versions.



Thank you.

Sincerely,

Joseph

--
we have not sent you but as a mercy to the creation.
holy quran, chapter 21, verse 107.
in the very authentic narration is:
imam hosein is the beacon of light and the ark of salvation.
best website for studying islamic book in different languages
al-islam.org

Sounds sensible. At least this will preserve xp etc with the old
Python for

as long as possible. One thing you do not say, which may be obvious to
you,

but perhaps not to the average user is..
What are the reasons for change.
is it that its going to be faster, 64 bit or has better ways of doing
stuff

nvda needs to be able to do in the future?
Obviously not in your remit is that at some point an agreed last
working version of the old NVDA that can work on xp etc is going to
have to be made

available and then when add ons are updated, those that will and won't
run on the old version will need to be noted and if possible trapped
by the old

version so it cannot be scrambled by an add on.


Brian

bglists@blueyonder.co.uk
Sent via blueyonder.
Please address personal email to:-
briang1@blueyonder.co.uk, putting 'Brian Gaff'
in the display name field.
----- Original Message -----
From: "Joseph Lee" <joseph.lee22590@gmail.com>
To: <nvda@nvda.groups.io>
Sent: Saturday, July 22, 2017 7:55 AM
Subject: [nvda] Announcing Project Pylennium: porting NVDA code base,
dependencies, and developer attitudes to Python 3.6.x

Ladies and gentlemen,



Some of you have been asking NV Access and the global NVDA community
regarding porting NVDA source code and dependencies to Python 3. I
and some contributors have been brainstorming best ways to accomplish
this, and NV Access, contributors, some add-on writers, users, and
NVDA observers agree that it is important that we look into
eventually porting NVDA to Python 3.



Until recently, the possibility of moving NVDA to Python 3 seemed to
be only a dream. However, a major hurdle was overcome earlier this
year when wxPython 4 alpha was released, promising support for Python
versions 2.7 and 3.x. Apart from a major problem regarding certain
internal routines from wxPython 4 that may not be compatible with
NVDA source code, the stage is almost set to begin our transition to
Python 3, dubbed "Project Pylennium".



Project Pylennium is a new project I'm planning and working on that
should make it easier for NV Access, code contributors, add-on
writers, users and many others to experience smooth transition to
Python 3. The overall goal

of
this project is to prepare NVDA's source code so it can run on both
Python

2
and 3 in most cases until developers are ready to move on. This
project also seeks to uncover possible compatibility issues for NVDA
community when the transition happens, and to document these findings
so it can serve as guidelines to current and future developers.



The current iteration of Project Pylennium (based on NVDA GitHub issue
7105
as a roadmap) can be found on my copy of NVDA source code at:

https://github.com/josephsl/nvda

The branch to use is "py3000".



Project pylennium involves the following:



1. Gather dependencies required for Python 3 transition, such as newer
versions of Python, ConfigObj and many others. A crucial dependency is
the
Six module, which adds compatibility layer functions for both Python 2
and
3.
2. Document porting notes, which will eventually be published on our
NVDA Community wiki.
3. Create source-level compatibility layer in order to mitigate changes
in module names, standard library folder structure, function and class
names
and so on.
4. Migrate most code to Python 3 syntax.
5. Frequent and regular testing to make sure regressions are not
introduced.



Progress so far:



1. Imports: a compatibility layer is in place to import Python 2 or 3
version of winreg, pickle, io.StringIO and what not.
2. Compiles fine via SCons 2.x.
3. Possible to see the first initialization phase of NVDA launcher
(nvda.pyw) via command line. The biggest showstopper at this time is the
fact that logHandler.StreamHandler constructor uses file function when
Python 3 says it isn't available, which suggests we need to modernize log
handler module.



Benefits:



1. Allows NV Access to look at porting data and make informed
decisions.
2. Allows contributors and Python experts to provide expertise and
prepare the source code for Python 3 transition.
3. Allows add-on authors to look at changes and prepare their add-ons
to run on Python 3.
4. Allows interested users to be involved in a major project and
provide test data.



Disadvantages:



1. We need to part ways with Windows XP and Server 2003.
2. Stress throughout the project, as it is nothing like what we've seen
recently.
3. Some add-ons will need community intervention to get them to run on
Python 3, as some are abandonware (no longer maintained by authors or the
add-ons community don't know who wrote some of them and lost contact with
some authors).
4. Possible issues in code and other ones, especially when looking at
code and attitude conflicts.



Next steps:



1. I'm looking for volunteers and subproject leads: if you are
passionate about Python, NVDA, and want to help NVDA achieve the next
milestone, please help and find folks who can assist in this project. In
regards to subprojects, mostly have to do with dependency checks and
tests,
testing, documentation, talking to Python Software Foundation (PSF),
recruiting Python experts and what not.
2. Cooperation: as Project Pylennium is a grand-scale project, we
cannot move forward and make it a reality unless cooperation between
developers, add-on authors, users, observers and many others is there.
3. Promote: one way to keep the momentum going is promoting this
project far and wide.
4. Continued work: I'm looking into how to port urllib imports, as this
module has seen huge folder restructuring in Python 3 and other work.



What you'll need for this project:



1. Python 3.6.0 32-bit or later (3.6.2 is recommended).
2. wxPython 4 alpha 3 or later with the wheel for Python 3.5 or 3.6.
3. Six module (from PYPI).
4. ConfigObj 5.0.6.



Guidelines:



1. When sending pull requests, please do NOT send them to NV Access
yet, as folks over there are working on high priority work. Instead, send

a
pull request against josephsl/nvda branch py3000.
2. Please provide a clear commit message when sending in pull requests
so folks can track what has changed.
3. Before you commit, please make sure to test NVDA in order to catch
regressions and write workarounds.
4. At least once a month, I will merge NV Access master branch to keep
the code up to date and incorporate changes into this project, sometimes
requiring transforming syntax to fit Python 3 requirements.
5. If you want to port add-ons, please do so at your earliest
convenience or when you are ready. As part of this work, I myself will
port
my own add-ons to Python 3 (one of them is done; throughout this year,
I'll
port rest of my add-ons). Please do not port your add-ons to Python 3
completely - I recommend making it so it'll run happily on both Python
versions.



Thank you.

Sincerely,

Joseph

Sounds sensible. At least this will preserve xp etc with the old
Python for

as long as possible. One thing you do not say, which may be obvious to
you,

but perhaps not to the average user is..
What are the reasons for change.
is it that its going to be faster, 64 bit or has better ways of doing
stuff

nvda needs to be able to do in the future?
Obviously not in your remit is that at some point an agreed last
working version of the old NVDA that can work on xp etc is going to
have to be made

available and then when add ons are updated, those that will and won't
run on the old version will need to be noted and if possible trapped
by the old

version so it cannot be scrambled by an add on.


Brian

bglists@blueyonder.co.uk
Sent via blueyonder.
Please address personal email to:-
briang1@blueyonder.co.uk, putting 'Brian Gaff'
in the display name field.
----- Original Message -----
From: "Joseph Lee" <joseph.lee22590@gmail.com>
To: <nvda@nvda.groups.io>
Sent: Saturday, July 22, 2017 7:55 AM
Subject: [nvda] Announcing Project Pylennium: porting NVDA code base,
dependencies, and developer attitudes to Python 3.6.x

Ladies and gentlemen,



Some of you have been asking NV Access and the global NVDA community
regarding porting NVDA source code and dependencies to Python 3. I
and some contributors have been brainstorming best ways to accomplish
this, and NV Access, contributors, some add-on writers, users, and
NVDA observers agree that it is important that we look into
eventually porting NVDA to Python 3.



Until recently, the possibility of moving NVDA to Python 3 seemed to
be only a dream. However, a major hurdle was overcome earlier this
year when wxPython 4 alpha was released, promising support for Python
versions 2.7 and 3.x. Apart from a major problem regarding certain
internal routines from wxPython 4 that may not be compatible with
NVDA source code, the stage is almost set to begin our transition to
Python 3, dubbed "Project Pylennium".



Project Pylennium is a new project I'm planning and working on that
should make it easier for NV Access, code contributors, add-on
writers, users and many others to experience smooth transition to
Python 3. The overall goal

of
this project is to prepare NVDA's source code so it can run on both
Python

2
and 3 in most cases until developers are ready to move on. This
project also seeks to uncover possible compatibility issues for NVDA
community when the transition happens, and to document these findings
so it can serve as guidelines to current and future developers.



The current iteration of Project Pylennium (based on NVDA GitHub issue
7105
as a roadmap) can be found on my copy of NVDA source code at:

https://github.com/josephsl/nvda

The branch to use is "py3000".



Project pylennium involves the following:



1. Gather dependencies required for Python 3 transition, such as newer
versions of Python, ConfigObj and many others. A crucial dependency is
the
Six module, which adds compatibility layer functions for both Python 2
and
3.
2. Document porting notes, which will eventually be published on our
NVDA Community wiki.
3. Create source-level compatibility layer in order to mitigate changes
in module names, standard library folder structure, function and class
names
and so on.
4. Migrate most code to Python 3 syntax.
5. Frequent and regular testing to make sure regressions are not
introduced.



Progress so far:



1. Imports: a compatibility layer is in place to import Python 2 or 3
version of winreg, pickle, io.StringIO and what not.
2. Compiles fine via SCons 2.x.
3. Possible to see the first initialization phase of NVDA launcher
(nvda.pyw) via command line. The biggest showstopper at this time is the
fact that logHandler.StreamHandler constructor uses file function when
Python 3 says it isn't available, which suggests we need to modernize log
handler module.



Benefits:



1. Allows NV Access to look at porting data and make informed
decisions.
2. Allows contributors and Python experts to provide expertise and
prepare the source code for Python 3 transition.
3. Allows add-on authors to look at changes and prepare their add-ons
to run on Python 3.
4. Allows interested users to be involved in a major project and
provide test data.



Disadvantages:



1. We need to part ways with Windows XP and Server 2003.
2. Stress throughout the project, as it is nothing like what we've seen
recently.
3. Some add-ons will need community intervention to get them to run on
Python 3, as some are abandonware (no longer maintained by authors or the
add-ons community don't know who wrote some of them and lost contact with
some authors).
4. Possible issues in code and other ones, especially when looking at
code and attitude conflicts.



Next steps:



1. I'm looking for volunteers and subproject leads: if you are
passionate about Python, NVDA, and want to help NVDA achieve the next
milestone, please help and find folks who can assist in this project. In
regards to subprojects, mostly have to do with dependency checks and
tests,
testing, documentation, talking to Python Software Foundation (PSF),
recruiting Python experts and what not.
2. Cooperation: as Project Pylennium is a grand-scale project, we
cannot move forward and make it a reality unless cooperation between
developers, add-on authors, users, observers and many others is there.
3. Promote: one way to keep the momentum going is promoting this
project far and wide.
4. Continued work: I'm looking into how to port urllib imports, as this
module has seen huge folder restructuring in Python 3 and other work.



What you'll need for this project:



1. Python 3.6.0 32-bit or later (3.6.2 is recommended).
2. wxPython 4 alpha 3 or later with the wheel for Python 3.5 or 3.6.
3. Six module (from PYPI).
4. ConfigObj 5.0.6.



Guidelines:



1. When sending pull requests, please do NOT send them to NV Access
yet, as folks over there are working on high priority work. Instead, send

a
pull request against josephsl/nvda branch py3000.
2. Please provide a clear commit message when sending in pull requests
so folks can track what has changed.
3. Before you commit, please make sure to test NVDA in order to catch
regressions and write workarounds.
4. At least once a month, I will merge NV Access master branch to keep
the code up to date and incorporate changes into this project, sometimes
requiring transforming syntax to fit Python 3 requirements.
5. If you want to port add-ons, please do so at your earliest
convenience or when you are ready. As part of this work, I myself will
port
my own add-ons to Python 3 (one of them is done; throughout this year,
I'll
port rest of my add-ons). Please do not port your add-ons to Python 3
completely - I recommend making it so it'll run happily on both Python
versions.



Thank you.

Sincerely,

Joseph

I think at present at least, XP is not really very good for web browsing any more due to the tendency for even the working third party browsers to be quite slow on single core machines of that vintage.

It is however still a joy to use for those of us who hate all the issues over too much security on single user computers that 7 8 and now 10 impose on us all.
Brian

bglists@blueyonder.co.uk
Sent via blueyonder.
Please address personal email to:-
briang1@blueyonder.co.uk, putting 'Brian Gaff'
in the display name field.
----- Original Message ----- From: "Joseph Lee" <joseph.lee22590@gmail.com>
To: <nvda@nvda.groups.io>
Sent: Saturday, July 22, 2017 7:07 PM
Subject: Re: [nvda] Announcing Project Pylennium: porting NVDA code base, dependencies, and developer attitudes to Python 3.6.x


Hi,
NV Access said they'll support XP as long as possible, but it won't be forever. I'm using Python 3.4 and 3.6 internally for this reason.
Cheers,
Joseph

-----Original Message-----
From: nvda@nvda.groups.io [mailto:nvda@nvda.groups.io] On Behalf Of nasrin khaksar
Sent: Saturday, July 22, 2017 4:46 AM
To: nvda@nvda.groups.io
Subject: Re: [nvda] Announcing Project Pylennium: porting NVDA code base, dependencies, and developer attitudes to Python 3.6.x

thanks for good news joseph and brian.
do you continue supporting of xp?
i became worried when i read the message for the first time.
God bless you all!

On 7/22/17, Brian's Mail list account via Groups.Io <bglists=blueyonder.co.uk@groups.io> wrote:

Sounds sensible. At least this will preserve xp etc with the old
Python for

as long as possible. One thing you do not say, which may be obvious to
you,

but perhaps not to the average user is..
What are the reasons for change.
is it that its going to be faster, 64 bit or has better ways of doing
stuff

nvda needs to be able to do in the future?
Obviously not in your remit is that at some point an agreed last
working version of the old NVDA that can work on xp etc is going to
have to be made

available and then when add ons are updated, those that will and won't
run on the old version will need to be noted and if possible trapped
by the old

version so it cannot be scrambled by an add on.


Brian

bglists@blueyonder.co.uk
Sent via blueyonder.
Please address personal email to:-
briang1@blueyonder.co.uk, putting 'Brian Gaff'
in the display name field.
----- Original Message -----
From: "Joseph Lee" <joseph.lee22590@gmail.com>
To: <nvda@nvda.groups.io>
Sent: Saturday, July 22, 2017 7:55 AM
Subject: [nvda] Announcing Project Pylennium: porting NVDA code base,
dependencies, and developer attitudes to Python 3.6.x

Ladies and gentlemen,



Some of you have been asking NV Access and the global NVDA community
regarding porting NVDA source code and dependencies to Python 3. I
and some contributors have been brainstorming best ways to accomplish
this, and NV Access, contributors, some add-on writers, users, and
NVDA observers agree that it is important that we look into
eventually porting NVDA to Python 3.



Until recently, the possibility of moving NVDA to Python 3 seemed to
be only a dream. However, a major hurdle was overcome earlier this
year when wxPython 4 alpha was released, promising support for Python
versions 2.7 and 3.x. Apart from a major problem regarding certain
internal routines from wxPython 4 that may not be compatible with
NVDA source code, the stage is almost set to begin our transition to
Python 3, dubbed "Project Pylennium".



Project Pylennium is a new project I'm planning and working on that
should make it easier for NV Access, code contributors, add-on
writers, users and many others to experience smooth transition to
Python 3. The overall goal

of
this project is to prepare NVDA's source code so it can run on both
Python

2
and 3 in most cases until developers are ready to move on. This
project also seeks to uncover possible compatibility issues for NVDA
community when the transition happens, and to document these findings
so it can serve as guidelines to current and future developers.



The current iteration of Project Pylennium (based on NVDA GitHub issue
7105
as a roadmap) can be found on my copy of NVDA source code at:

https://github.com/josephsl/nvda

The branch to use is "py3000".



Project pylennium involves the following:



1. Gather dependencies required for Python 3 transition, such as newer
versions of Python, ConfigObj and many others. A crucial dependency is
the
Six module, which adds compatibility layer functions for both Python 2
and
3.
2. Document porting notes, which will eventually be published on our
NVDA Community wiki.
3. Create source-level compatibility layer in order to mitigate changes
in module names, standard library folder structure, function and class
names
and so on.
4. Migrate most code to Python 3 syntax.
5. Frequent and regular testing to make sure regressions are not
introduced.



Progress so far:



1. Imports: a compatibility layer is in place to import Python 2 or 3
version of winreg, pickle, io.StringIO and what not.
2. Compiles fine via SCons 2.x.
3. Possible to see the first initialization phase of NVDA launcher
(nvda.pyw) via command line. The biggest showstopper at this time is the
fact that logHandler.StreamHandler constructor uses file function when
Python 3 says it isn't available, which suggests we need to modernize log
handler module.



Benefits:



1. Allows NV Access to look at porting data and make informed
decisions.
2. Allows contributors and Python experts to provide expertise and
prepare the source code for Python 3 transition.
3. Allows add-on authors to look at changes and prepare their add-ons
to run on Python 3.
4. Allows interested users to be involved in a major project and
provide test data.



Disadvantages:



1. We need to part ways with Windows XP and Server 2003.
2. Stress throughout the project, as it is nothing like what we've seen
recently.
3. Some add-ons will need community intervention to get them to run on
Python 3, as some are abandonware (no longer maintained by authors or the
add-ons community don't know who wrote some of them and lost contact with
some authors).
4. Possible issues in code and other ones, especially when looking at
code and attitude conflicts.



Next steps:



1. I'm looking for volunteers and subproject leads: if you are
passionate about Python, NVDA, and want to help NVDA achieve the next
milestone, please help and find folks who can assist in this project. In
regards to subprojects, mostly have to do with dependency checks and
tests,
testing, documentation, talking to Python Software Foundation (PSF),
recruiting Python experts and what not.
2. Cooperation: as Project Pylennium is a grand-scale project, we
cannot move forward and make it a reality unless cooperation between
developers, add-on authors, users, observers and many others is there.
3. Promote: one way to keep the momentum going is promoting this
project far and wide.
4. Continued work: I'm looking into how to port urllib imports, as this
module has seen huge folder restructuring in Python 3 and other work.



What you'll need for this project:



1. Python 3.6.0 32-bit or later (3.6.2 is recommended).
2. wxPython 4 alpha 3 or later with the wheel for Python 3.5 or 3.6.
3. Six module (from PYPI).
4. ConfigObj 5.0.6.



Guidelines:



1. When sending pull requests, please do NOT send them to NV Access
yet, as folks over there are working on high priority work. Instead, send

a
pull request against josephsl/nvda branch py3000.
2. Please provide a clear commit message when sending in pull requests
so folks can track what has changed.
3. Before you commit, please make sure to test NVDA in order to catch
regressions and write workarounds.
4. At least once a month, I will merge NV Access master branch to keep
the code up to date and incorporate changes into this project, sometimes
requiring transforming syntax to fit Python 3 requirements.
5. If you want to port add-ons, please do so at your earliest
convenience or when you are ready. As part of this work, I myself will
port
my own add-ons to Python 3 (one of them is done; throughout this year,
I'll
port rest of my add-ons). Please do not port your add-ons to Python 3
completely - I recommend making it so it'll run happily on both Python
versions.



Thank you.

Sincerely,

Joseph

Well I wouldn't use xp for the net anymore its far to insecure for that.

However its good for all those 32 bit programs you need to use, ie old
games especially those that use directx 8 or whatever that is.

They are also usefull well xp is for old nokia 32 bit apps and other
older games.




On 24/07/2017 7:52 p.m., Brian's Mail list account via Groups.Io wrote:

I think at present at least, XP is not really very good for web
browsing any more due to the tendency for even the working third party
browsers to be quite slow on single core machines of that vintage.

It is however still a joy to use for those of us who hate all the
issues over too much security on single user computers that 7 8 and
now 10 impose on us all.
Brian

bglists@blueyonder.co.uk
Sent via blueyonder.
Please address personal email to:-
briang1@blueyonder.co.uk, putting 'Brian Gaff'
in the display name field.
----- Original Message ----- From: "Joseph Lee"
<joseph.lee22590@gmail.com>
To: <nvda@nvda.groups.io>
Sent: Saturday, July 22, 2017 7:07 PM
Subject: Re: [nvda] Announcing Project Pylennium: porting NVDA code
base, dependencies, and developer attitudes to Python 3.6.x


Hi,
NV Access said they'll support XP as long as possible, but it won't be
forever. I'm using Python 3.4 and 3.6 internally for this reason.
Cheers,
Joseph

-----Original Message-----
From: nvda@nvda.groups.io [mailto:nvda@nvda.groups.io] On Behalf Of
nasrin khaksar
Sent: Saturday, July 22, 2017 4:46 AM
To: nvda@nvda.groups.io
Subject: Re: [nvda] Announcing Project Pylennium: porting NVDA code
base, dependencies, and developer attitudes to Python 3.6.x

thanks for good news joseph and brian.
do you continue supporting of xp?
i became worried when i read the message for the first time.
God bless you all!

On 7/22/17, Brian's Mail list account via Groups.Io
<bglists=blueyonder.co.uk@groups.io> wrote:

Sounds sensible. At least this will preserve xp etc with the old
Python for

as long as possible. One thing you do not say, which may be obvious to
you,

but perhaps not to the average user is..
What are the reasons for change.
is it that its going to be faster, 64 bit or has better ways of doing
stuff

nvda needs to be able to do in the future?
Obviously not in your remit is that at some point an agreed last
working version of the old NVDA that can work on xp etc is going to
have to be made

available and then when add ons are updated, those that will and won't
run on the old version will need to be noted and if possible trapped
by the old

version so it cannot be scrambled by an add on.


Brian

bglists@blueyonder.co.uk
Sent via blueyonder.
Please address personal email to:-
briang1@blueyonder.co.uk, putting 'Brian Gaff'
in the display name field.
----- Original Message -----
From: "Joseph Lee" <joseph.lee22590@gmail.com>
To: <nvda@nvda.groups.io>
Sent: Saturday, July 22, 2017 7:55 AM
Subject: [nvda] Announcing Project Pylennium: porting NVDA code base,
dependencies, and developer attitudes to Python 3.6.x

Ladies and gentlemen,



Some of you have been asking NV Access and the global NVDA community
regarding porting NVDA source code and dependencies to Python 3. I
and some contributors have been brainstorming best ways to accomplish
this, and NV Access, contributors, some add-on writers, users, and
NVDA observers agree that it is important that we look into
eventually porting NVDA to Python 3.



Until recently, the possibility of moving NVDA to Python 3 seemed to
be only a dream. However, a major hurdle was overcome earlier this
year when wxPython 4 alpha was released, promising support for Python
versions 2.7 and 3.x. Apart from a major problem regarding certain
internal routines from wxPython 4 that may not be compatible with
NVDA source code, the stage is almost set to begin our transition to
Python 3, dubbed "Project Pylennium".



Project Pylennium is a new project I'm planning and working on that
should make it easier for NV Access, code contributors, add-on
writers, users and many others to experience smooth transition to
Python 3. The overall goal

of
this project is to prepare NVDA's source code so it can run on both
Python

2
and 3 in most cases until developers are ready to move on. This
project also seeks to uncover possible compatibility issues for NVDA
community when the transition happens, and to document these findings
so it can serve as guidelines to current and future developers.



The current iteration of Project Pylennium (based on NVDA GitHub issue
7105
as a roadmap) can be found on my copy of NVDA source code at:

https://github.com/josephsl/nvda

The branch to use is "py3000".



Project pylennium involves the following:



1. Gather dependencies required for Python 3 transition, such as newer
versions of Python, ConfigObj and many others. A crucial dependency is
the
Six module, which adds compatibility layer functions for both Python 2
and
3.
2. Document porting notes, which will eventually be published on our
NVDA Community wiki.
3. Create source-level compatibility layer in order to mitigate changes
in module names, standard library folder structure, function and class
names
and so on.
4. Migrate most code to Python 3 syntax.
5. Frequent and regular testing to make sure regressions are not
introduced.



Progress so far:



1. Imports: a compatibility layer is in place to import Python 2 or 3
version of winreg, pickle, io.StringIO and what not.
2. Compiles fine via SCons 2.x.
3. Possible to see the first initialization phase of NVDA launcher
(nvda.pyw) via command line. The biggest showstopper at this time is
the
fact that logHandler.StreamHandler constructor uses file function when
Python 3 says it isn't available, which suggests we need to
modernize log
handler module.



Benefits:



1. Allows NV Access to look at porting data and make informed
decisions.
2. Allows contributors and Python experts to provide expertise and
prepare the source code for Python 3 transition.
3. Allows add-on authors to look at changes and prepare their add-ons
to run on Python 3.
4. Allows interested users to be involved in a major project and
provide test data.



Disadvantages:



1. We need to part ways with Windows XP and Server 2003.
2. Stress throughout the project, as it is nothing like what we've seen
recently.
3. Some add-ons will need community intervention to get them to run on
Python 3, as some are abandonware (no longer maintained by authors
or the
add-ons community don't know who wrote some of them and lost contact
with
some authors).
4. Possible issues in code and other ones, especially when looking at
code and attitude conflicts.



Next steps:



1. I'm looking for volunteers and subproject leads: if you are
passionate about Python, NVDA, and want to help NVDA achieve the next
milestone, please help and find folks who can assist in this
project. In
regards to subprojects, mostly have to do with dependency checks and
tests,
testing, documentation, talking to Python Software Foundation (PSF),
recruiting Python experts and what not.
2. Cooperation: as Project Pylennium is a grand-scale project, we
cannot move forward and make it a reality unless cooperation between
developers, add-on authors, users, observers and many others is there.
3. Promote: one way to keep the momentum going is promoting this
project far and wide.
4. Continued work: I'm looking into how to port urllib imports, as this
module has seen huge folder restructuring in Python 3 and other work.



What you'll need for this project:



1. Python 3.6.0 32-bit or later (3.6.2 is recommended).
2. wxPython 4 alpha 3 or later with the wheel for Python 3.5 or 3.6.
3. Six module (from PYPI).
4. ConfigObj 5.0.6.



Guidelines:



1. When sending pull requests, please do NOT send them to NV Access
yet, as folks over there are working on high priority work. Instead,
send

a
pull request against josephsl/nvda branch py3000.
2. Please provide a clear commit message when sending in pull requests
so folks can track what has changed.
3. Before you commit, please make sure to test NVDA in order to catch
regressions and write workarounds.
4. At least once a month, I will merge NV Access master branch to keep
the code up to date and incorporate changes into this project,
sometimes
requiring transforming syntax to fit Python 3 requirements.
5. If you want to port add-ons, please do so at your earliest
convenience or when you are ready. As part of this work, I myself will
port
my own add-ons to Python 3 (one of them is done; throughout this year,
I'll
port rest of my add-ons). Please do not port your add-ons to Python 3
completely - I recommend making it so it'll run happily on both Python
versions.



Thank you.

Sincerely,

Joseph

It would be nice if Microsoft had made XP so bulletproof that you could
use it forever, but that is not the world we find ourselves in.

Well I wouldn't use xp for the net anymore its far to insecure for that.

However its good for all those 32 bit programs you need to use, ie old
games especially those that use directx 8 or whatever that is.

They are also usefull well xp is for old nokia 32 bit apps and other
older games.




On 24/07/2017 7:52 p.m., Brian's Mail list account via Groups.Io wrote:

I think at present at least, XP is not really very good for web
browsing any more due to the tendency for even the working third
party browsers to be quite slow on single core machines of that vintage.

It is however still a joy to use for those of us who hate all the
issues over too much security on single user computers that 7 8 and
now 10 impose on us all.
Brian

bglists@blueyonder.co.uk
Sent via blueyonder.
Please address personal email to:-
briang1@blueyonder.co.uk, putting 'Brian Gaff'
in the display name field.
----- Original Message ----- From: "Joseph Lee"
<joseph.lee22590@gmail.com>
To: <nvda@nvda.groups.io>
Sent: Saturday, July 22, 2017 7:07 PM
Subject: Re: [nvda] Announcing Project Pylennium: porting NVDA code
base, dependencies, and developer attitudes to Python 3.6.x


Hi,
NV Access said they'll support XP as long as possible, but it won't
be forever. I'm using Python 3.4 and 3.6 internally for this reason.
Cheers,
Joseph

-----Original Message-----
From: nvda@nvda.groups.io [mailto:nvda@nvda.groups.io] On Behalf Of
nasrin khaksar
Sent: Saturday, July 22, 2017 4:46 AM
To: nvda@nvda.groups.io
Subject: Re: [nvda] Announcing Project Pylennium: porting NVDA code
base, dependencies, and developer attitudes to Python 3.6.x

thanks for good news joseph and brian.
do you continue supporting of xp?
i became worried when i read the message for the first time.
God bless you all!

On 7/22/17, Brian's Mail list account via Groups.Io
<bglists=blueyonder.co.uk@groups.io> wrote:

Sounds sensible. At least this will preserve xp etc with the old
Python for

as long as possible. One thing you do not say, which may be obvious
to you,

but perhaps not to the average user is..
What are the reasons for change.
is it that its going to be faster, 64 bit or has better ways of
doing stuff

nvda needs to be able to do in the future?
Obviously not in your remit is that at some point an agreed last
working version of the old NVDA that can work on xp etc is going to
have to be made

available and then when add ons are updated, those that will and
won't run on the old version will need to be noted and if possible
trapped by the old

version so it cannot be scrambled by an add on.


Brian

bglists@blueyonder.co.uk
Sent via blueyonder.
Please address personal email to:-
briang1@blueyonder.co.uk, putting 'Brian Gaff'
in the display name field.
----- Original Message -----
From: "Joseph Lee" <joseph.lee22590@gmail.com>
To: <nvda@nvda.groups.io>
Sent: Saturday, July 22, 2017 7:55 AM
Subject: [nvda] Announcing Project Pylennium: porting NVDA code
base, dependencies, and developer attitudes to Python 3.6.x

Ladies and gentlemen,



Some of you have been asking NV Access and the global NVDA
community regarding porting NVDA source code and dependencies to
Python 3. I and some contributors have been brainstorming best ways
to accomplish this, and NV Access, contributors, some add-on
writers, users, and NVDA observers agree that it is important that
we look into eventually porting NVDA to Python 3.



Until recently, the possibility of moving NVDA to Python 3 seemed
to be only a dream. However, a major hurdle was overcome earlier
this year when wxPython 4 alpha was released, promising support for
Python versions 2.7 and 3.x. Apart from a major problem regarding
certain internal routines from wxPython 4 that may not be
compatible with NVDA source code, the stage is almost set to begin
our transition to Python 3, dubbed "Project Pylennium".



Project Pylennium is a new project I'm planning and working on that
should make it easier for NV Access, code contributors, add-on
writers, users and many others to experience smooth transition to
Python 3. The overall goal

of
this project is to prepare NVDA's source code so it can run on both
Python

2
and 3 in most cases until developers are ready to move on. This
project also seeks to uncover possible compatibility issues for
NVDA community when the transition happens, and to document these
findings so it can serve as guidelines to current and future developers.



The current iteration of Project Pylennium (based on NVDA GitHub
issue
7105
as a roadmap) can be found on my copy of NVDA source code at:

https://github.com/josephsl/nvda

The branch to use is "py3000".



Project pylennium involves the following:



1. Gather dependencies required for Python 3 transition, such as
newer versions of Python, ConfigObj and many others. A crucial
dependency is the Six module, which adds compatibility layer
functions for both Python 2 and 3.
2. Document porting notes, which will eventually be published on
our NVDA Community wiki.
3. Create source-level compatibility layer in order to mitigate
changes in module names, standard library folder structure,
function and class names and so on.
4. Migrate most code to Python 3 syntax.
5. Frequent and regular testing to make sure regressions are not
introduced.



Progress so far:



1. Imports: a compatibility layer is in place to import Python 2 or
3 version of winreg, pickle, io.StringIO and what not.
2. Compiles fine via SCons 2.x.
3. Possible to see the first initialization phase of NVDA launcher
(nvda.pyw) via command line. The biggest showstopper at this time
is the fact that logHandler.StreamHandler constructor uses file
function when Python 3 says it isn't available, which suggests we
need to modernize log handler module.



Benefits:



1. Allows NV Access to look at porting data and make informed
decisions.
2. Allows contributors and Python experts to provide expertise and
prepare the source code for Python 3 transition.
3. Allows add-on authors to look at changes and prepare their
add-ons to run on Python 3.
4. Allows interested users to be involved in a major project and
provide test data.



Disadvantages:



1. We need to part ways with Windows XP and Server 2003.
2. Stress throughout the project, as it is nothing like what we've
seen recently.
3. Some add-ons will need community intervention to get them to run
on Python 3, as some are abandonware (no longer maintained by
authors or the add-ons community don't know who wrote some of them
and lost contact with some authors).
4. Possible issues in code and other ones, especially when looking
at code and attitude conflicts.



Next steps:



1. I'm looking for volunteers and subproject leads: if you are
passionate about Python, NVDA, and want to help NVDA achieve the next
milestone, please help and find folks who can assist in this
project. In
regards to subprojects, mostly have to do with dependency checks and
tests,
testing, documentation, talking to Python Software Foundation (PSF),
recruiting Python experts and what not.
2. Cooperation: as Project Pylennium is a grand-scale project, we
cannot move forward and make it a reality unless cooperation between
developers, add-on authors, users, observers and many others is there.
3. Promote: one way to keep the momentum going is promoting this
project far and wide.
4. Continued work: I'm looking into how to port urllib imports, as this
module has seen huge folder restructuring in Python 3 and other work.



What you'll need for this project:



1. Python 3.6.0 32-bit or later (3.6.2 is recommended).
2. wxPython 4 alpha 3 or later with the wheel for Python 3.5 or 3.6.
3. Six module (from PYPI).
4. ConfigObj 5.0.6.



Guidelines:



1. When sending pull requests, please do NOT send them to NV Access
yet, as folks over there are working on high priority work. Instead,
send

a
pull request against josephsl/nvda branch py3000.
2. Please provide a clear commit message when sending in pull requests
so folks can track what has changed.
3. Before you commit, please make sure to test NVDA in order to catch
regressions and write workarounds.
4. At least once a month, I will merge NV Access master branch to keep
the code up to date and incorporate changes into this project,
sometimes
requiring transforming syntax to fit Python 3 requirements.
5. If you want to port add-ons, please do so at your earliest
convenience or when you are ready. As part of this work, I myself will
port
my own add-ons to Python 3 (one of them is done; throughout this year,
I'll
port rest of my add-ons). Please do not port your add-ons to Python 3
completely - I recommend making it so it'll run happily on both Python
versions.



Thank you.

Sincerely,

Joseph

I'm sorry but microsoft stopped supporting windows XP in 2014. Why can't you upgrade to a better system instead of staying with something that isn't supported anymore. I hope you don't get malware or something that could destroy your computer so you can't use it anymore.

-----Original Message-----
From: nvda@nvda.groups.io [mailto:nvda@nvda.groups.io] On Behalf Of nasrin khaksar
Sent: Monday, July 24, 2017 6:11 AM
To: nvda@nvda.groups.io
Subject: Re: [nvda] Announcing Project Pylennium: porting NVDA code base, dependencies, and developer attitudes to Python 3.6.x

i use xp, love it and dont believe negative statement against it.
its realy great operating system forever.

On 7/24/17, Shaun Everiss <sm.everiss@gmail.com> wrote:

Well I wouldn't use xp for the net anymore its far to insecure for that.

However its good for all those 32 bit programs you need to use, ie old
games especially those that use directx 8 or whatever that is.

They are also usefull well xp is for old nokia 32 bit apps and other
older games.




On 24/07/2017 7:52 p.m., Brian's Mail list account via Groups.Io wrote:

I think at present at least, XP is not really very good for web
browsing any more due to the tendency for even the working third
party browsers to be quite slow on single core machines of that vintage.

It is however still a joy to use for those of us who hate all the
issues over too much security on single user computers that 7 8 and
now 10 impose on us all.
Brian

bglists@blueyonder.co.uk
Sent via blueyonder.
Please address personal email to:-
briang1@blueyonder.co.uk, putting 'Brian Gaff'
in the display name field.
----- Original Message ----- From: "Joseph Lee"
<joseph.lee22590@gmail.com>
To: <nvda@nvda.groups.io>
Sent: Saturday, July 22, 2017 7:07 PM
Subject: Re: [nvda] Announcing Project Pylennium: porting NVDA code
base, dependencies, and developer attitudes to Python 3.6.x


Hi,
NV Access said they'll support XP as long as possible, but it won't
be forever. I'm using Python 3.4 and 3.6 internally for this reason.
Cheers,
Joseph

-----Original Message-----
From: nvda@nvda.groups.io [mailto:nvda@nvda.groups.io] On Behalf Of
nasrin khaksar
Sent: Saturday, July 22, 2017 4:46 AM
To: nvda@nvda.groups.io
Subject: Re: [nvda] Announcing Project Pylennium: porting NVDA code
base, dependencies, and developer attitudes to Python 3.6.x

thanks for good news joseph and brian.
do you continue supporting of xp?
i became worried when i read the message for the first time.
God bless you all!

On 7/22/17, Brian's Mail list account via Groups.Io
<bglists=blueyonder.co.uk@groups.io> wrote:

Sounds sensible. At least this will preserve xp etc with the old
Python for

as long as possible. One thing you do not say, which may be obvious
to you,

but perhaps not to the average user is..
What are the reasons for change.
is it that its going to be faster, 64 bit or has better ways of
doing stuff

nvda needs to be able to do in the future?
Obviously not in your remit is that at some point an agreed last
working version of the old NVDA that can work on xp etc is going to
have to be made

available and then when add ons are updated, those that will and
won't run on the old version will need to be noted and if possible
trapped by the old

version so it cannot be scrambled by an add on.


Brian

bglists@blueyonder.co.uk
Sent via blueyonder.
Please address personal email to:-
briang1@blueyonder.co.uk, putting 'Brian Gaff'
in the display name field.
----- Original Message -----
From: "Joseph Lee" <joseph.lee22590@gmail.com>
To: <nvda@nvda.groups.io>
Sent: Saturday, July 22, 2017 7:55 AM
Subject: [nvda] Announcing Project Pylennium: porting NVDA code
base, dependencies, and developer attitudes to Python 3.6.x

Ladies and gentlemen,



Some of you have been asking NV Access and the global NVDA
community regarding porting NVDA source code and dependencies to
Python 3. I and some contributors have been brainstorming best ways
to accomplish this, and NV Access, contributors, some add-on
writers, users, and NVDA observers agree that it is important that
we look into eventually porting NVDA to Python 3.



Until recently, the possibility of moving NVDA to Python 3 seemed
to be only a dream. However, a major hurdle was overcome earlier
this year when wxPython 4 alpha was released, promising support for
Python versions 2.7 and 3.x. Apart from a major problem regarding
certain internal routines from wxPython 4 that may not be
compatible with NVDA source code, the stage is almost set to begin
our transition to Python 3, dubbed "Project Pylennium".



Project Pylennium is a new project I'm planning and working on that
should make it easier for NV Access, code contributors, add-on
writers, users and many others to experience smooth transition to
Python 3. The overall goal

of
this project is to prepare NVDA's source code so it can run on both
Python

2
and 3 in most cases until developers are ready to move on. This
project also seeks to uncover possible compatibility issues for
NVDA community when the transition happens, and to document these
findings so it can serve as guidelines to current and future developers.



The current iteration of Project Pylennium (based on NVDA GitHub
issue
7105
as a roadmap) can be found on my copy of NVDA source code at:

https://github.com/josephsl/nvda

The branch to use is "py3000".



Project pylennium involves the following:



1. Gather dependencies required for Python 3 transition, such as
newer versions of Python, ConfigObj and many others. A crucial
dependency is the Six module, which adds compatibility layer
functions for both Python 2 and 3.
2. Document porting notes, which will eventually be published on
our NVDA Community wiki.
3. Create source-level compatibility layer in order to mitigate
changes in module names, standard library folder structure,
function and class names and so on.
4. Migrate most code to Python 3 syntax.
5. Frequent and regular testing to make sure regressions are not
introduced.



Progress so far:



1. Imports: a compatibility layer is in place to import Python 2 or
3 version of winreg, pickle, io.StringIO and what not.
2. Compiles fine via SCons 2.x.
3. Possible to see the first initialization phase of NVDA launcher
(nvda.pyw) via command line. The biggest showstopper at this time
is the fact that logHandler.StreamHandler constructor uses file
function when Python 3 says it isn't available, which suggests we
need to modernize log handler module.



Benefits:



1. Allows NV Access to look at porting data and make informed
decisions.
2. Allows contributors and Python experts to provide expertise and
prepare the source code for Python 3 transition.
3. Allows add-on authors to look at changes and prepare their
add-ons to run on Python 3.
4. Allows interested users to be involved in a major project and
provide test data.



Disadvantages:



1. We need to part ways with Windows XP and Server 2003.
2. Stress throughout the project, as it is nothing like what we've
seen recently.
3. Some add-ons will need community intervention to get them to run
on Python 3, as some are abandonware (no longer maintained by
authors or the add-ons community don't know who wrote some of them
and lost contact with some authors).
4. Possible issues in code and other ones, especially when looking
at code and attitude conflicts.



Next steps:



1. I'm looking for volunteers and subproject leads: if you are
passionate about Python, NVDA, and want to help NVDA achieve the next
milestone, please help and find folks who can assist in this
project. In
regards to subprojects, mostly have to do with dependency checks and
tests,
testing, documentation, talking to Python Software Foundation (PSF),
recruiting Python experts and what not.
2. Cooperation: as Project Pylennium is a grand-scale project, we
cannot move forward and make it a reality unless cooperation between
developers, add-on authors, users, observers and many others is there.
3. Promote: one way to keep the momentum going is promoting this
project far and wide.
4. Continued work: I'm looking into how to port urllib imports, as this
module has seen huge folder restructuring in Python 3 and other work.



What you'll need for this project:



1. Python 3.6.0 32-bit or later (3.6.2 is recommended).
2. wxPython 4 alpha 3 or later with the wheel for Python 3.5 or 3.6.
3. Six module (from PYPI).
4. ConfigObj 5.0.6.



Guidelines:



1. When sending pull requests, please do NOT send them to NV Access
yet, as folks over there are working on high priority work. Instead,
send

a
pull request against josephsl/nvda branch py3000.
2. Please provide a clear commit message when sending in pull requests
so folks can track what has changed.
3. Before you commit, please make sure to test NVDA in order to catch
regressions and write workarounds.
4. At least once a month, I will merge NV Access master branch to keep
the code up to date and incorporate changes into this project,
sometimes
requiring transforming syntax to fit Python 3 requirements.
5. If you want to port add-ons, please do so at your earliest
convenience or when you are ready. As part of this work, I myself will
port
my own add-ons to Python 3 (one of them is done; throughout this year,
I'll
port rest of my add-ons). Please do not port your add-ons to Python 3
completely - I recommend making it so it'll run happily on both Python
versions.



Thank you.

Sincerely,

Joseph

Hmm, I don't think anyone disputes the facts, its just that some either have little choice or do not want to change.
I do think you overstate the risks for the average user though, as it seems to me that security holes are in any operating system, the danger is that we do not know where they are yet in 10, but we probably do in Xp. If there are very few XP users around soon the target will move squarely onto the newer systems and then we are back in effect where we started. sigh. Brian

bglists@blueyonder.co.uk
Sent via blueyonder.
Please address personal email to:-
briang1@blueyonder.co.uk, putting 'Brian Gaff'
in the display name field.
----- Original Message ----- From: "Brian Vogel" <britechguy@gmail.com>
To: <nvda@nvda.groups.io>
Sent: Monday, July 24, 2017 2:44 PM
Subject: Re: [nvda] Announcing Project Pylennium: porting NVDA code base, dependencies, and developer attitudes to Python 3.6.x


On Mon, Jul 24, 2017 at 06:20 am, Marshall handheld Flax wrote:

It would be nice if Microsoft had made XP so bulletproof that you could
use it forever, but that is not the world we find ourselves in.

And being a computer technician by trade I will say that, at this point in time, it is foolish, yes foolish, to use a Windows XP machine if you are interacting with cyberspace with it. There are massive security holes in the OS itself that will never be patched and the version of IE that will run under it is even worse.

Using an unsupported operating system, regardless of which operating system, is just asking for trouble. It will come, it's just a matter of when.

--
Brian - Windows 10 Home, 64-Bit, Version 1703, Build 15063 (dot level on request - it changes too often to keep in signature)

* * *The opposite of a correct statement is a false statement.* * But the opposite of a profound truth may well be another* * profound truth.*

* ~ * *Niels Bohr*



.

I'm sorry, but IBM stopped supporting os/2 in 2006, why do folks still
use it?

Mostly because it suits their purposes, and it does what they want, so
they use it.

People use what works for them, and if that just happens to be something
that isn't officially supported anymore, then so what. Everyone who uses
an unsupported os probably already knows that it isn't supported, and
there's reasons they choose to use it anyway. As long as there are
operating systems, some people are going to use them regardless of their
supported status. Dos is still used very heavily in industrial
settings, and I don't see anyone yelling at those industries to update
to something that is still supported by microsoft, ibm, novell, paragon,
or any number of other companies. Why? Because the dos being used suits
the purpose for what it's being used for, and the industrial users
already know there's no support, and that makes no difference to them
whatsoever.

Unless you're willing to shoulder the financial burden of upgrading each
and every XP user in the world, perhaps you should attempt to offer
suggestions on how they could make their systems more secure instead of
just telling them to scrap a perfectly working system and move to
something they neither want, or can afford.

I'm frankly a bit tired of the whole I'm better than you because I use a
modern os attitude that is prevalent on this list (and other places too,
don't worry, it's not just here), let's attempt to provide support
(which is the point of this list in the first place) instead of putting
folks down because they don't want or can't afford to run the latest and
greatest hardware/os, ok?



On 7/24/2017 1:53 PM, Rosemarie Chavarria wrote:

I'm sorry but microsoft stopped supporting windows XP in 2014. Why can't
you upgrade to a better system instead of staying with something that
isn't supported anymore. I hope you don't get malware or something that
could destroy your computer so you can't use it anymore.

-----Original Message-----
From: nvda@nvda.groups.io [mailto:nvda@nvda.groups.io] On Behalf Of nasrin
khaksar
Sent: Monday, July 24, 2017 6:11 AM
To: nvda@nvda.groups.io
Subject: Re: [nvda] Announcing Project Pylennium: porting NVDA code base,
dependencies, and developer attitudes to Python 3.6.x

i use xp, love it and dont believe negative statement against it.
its realy great operating system forever.

On 7/24/17, Shaun Everiss <sm.everiss@gmail.com> wrote:

Well I wouldn't use xp for the net anymore its far to insecure for that.

However its good for all those 32 bit programs you need to use, ie old
games especially those that use directx 8 or whatever that is.

They are also usefull well xp is for old nokia 32 bit apps and other
older games.




On 24/07/2017 7:52 p.m., Brian's Mail list account via Groups.Io wrote:

I think at present at least, XP is not really very good for web
browsing any more due to the tendency for even the working third
party browsers to be quite slow on single core machines of that
vintage.

It is however still a joy to use for those of us who hate all the
issues over too much security on single user computers that 7 8 and
now 10 impose on us all.
Brian

bglists@blueyonder.co.uk
Sent via blueyonder.
Please address personal email to:-
briang1@blueyonder.co.uk, putting 'Brian Gaff'
in the display name field.
----- Original Message ----- From: "Joseph Lee"
<joseph.lee22590@gmail.com>
To: <nvda@nvda.groups.io>
Sent: Saturday, July 22, 2017 7:07 PM
Subject: Re: [nvda] Announcing Project Pylennium: porting NVDA code
base, dependencies, and developer attitudes to Python 3.6.x


Hi,
NV Access said they'll support XP as long as possible, but it won't
be forever. I'm using Python 3.4 and 3.6 internally for this reason.
Cheers,
Joseph

-----Original Message-----
From: nvda@nvda.groups.io [mailto:nvda@nvda.groups.io] On Behalf Of
nasrin khaksar
Sent: Saturday, July 22, 2017 4:46 AM
To: nvda@nvda.groups.io
Subject: Re: [nvda] Announcing Project Pylennium: porting NVDA code
base, dependencies, and developer attitudes to Python 3.6.x

thanks for good news joseph and brian.
do you continue supporting of xp?
i became worried when i read the message for the first time.
God bless you all!

On 7/22/17, Brian's Mail list account via Groups.Io
<bglists=blueyonder.co.uk@groups.io> wrote:

Sounds sensible. At least this will preserve xp etc with the old
Python for

as long as possible. One thing you do not say, which may be obvious
to you,

but perhaps not to the average user is..
What are the reasons for change.
is it that its going to be faster, 64 bit or has better ways of
doing stuff

nvda needs to be able to do in the future?
Obviously not in your remit is that at some point an agreed last
working version of the old NVDA that can work on xp etc is going to
have to be made

available and then when add ons are updated, those that will and
won't run on the old version will need to be noted and if possible
trapped by the old

version so it cannot be scrambled by an add on.


Brian

bglists@blueyonder.co.uk
Sent via blueyonder.
Please address personal email to:-
briang1@blueyonder.co.uk, putting 'Brian Gaff'
in the display name field.
----- Original Message -----
From: "Joseph Lee" <joseph.lee22590@gmail.com>
To: <nvda@nvda.groups.io>
Sent: Saturday, July 22, 2017 7:55 AM
Subject: [nvda] Announcing Project Pylennium: porting NVDA code
base, dependencies, and developer attitudes to Python 3.6.x

Ladies and gentlemen,



Some of you have been asking NV Access and the global NVDA
community regarding porting NVDA source code and dependencies to
Python 3. I and some contributors have been brainstorming best ways
to accomplish this, and NV Access, contributors, some add-on
writers, users, and NVDA observers agree that it is important that
we look into eventually porting NVDA to Python 3.



Until recently, the possibility of moving NVDA to Python 3 seemed
to be only a dream. However, a major hurdle was overcome earlier
this year when wxPython 4 alpha was released, promising support for
Python versions 2.7 and 3.x. Apart from a major problem regarding
certain internal routines from wxPython 4 that may not be
compatible with NVDA source code, the stage is almost set to begin
our transition to Python 3, dubbed "Project Pylennium".



Project Pylennium is a new project I'm planning and working on that
should make it easier for NV Access, code contributors, add-on
writers, users and many others to experience smooth transition to
Python 3. The overall goal

of
this project is to prepare NVDA's source code so it can run on both
Python

2
and 3 in most cases until developers are ready to move on. This
project also seeks to uncover possible compatibility issues for
NVDA community when the transition happens, and to document these
findings so it can serve as guidelines to current and future
developers.



The current iteration of Project Pylennium (based on NVDA GitHub
issue
7105
as a roadmap) can be found on my copy of NVDA source code at:

https://github.com/josephsl/nvda

The branch to use is "py3000".



Project pylennium involves the following:



1. Gather dependencies required for Python 3 transition, such as
newer versions of Python, ConfigObj and many others. A crucial
dependency is the Six module, which adds compatibility layer
functions for both Python 2 and 3.
2. Document porting notes, which will eventually be published on
our NVDA Community wiki.
3. Create source-level compatibility layer in order to mitigate
changes in module names, standard library folder structure,
function and class names and so on.
4. Migrate most code to Python 3 syntax.
5. Frequent and regular testing to make sure regressions are not
introduced.



Progress so far:



1. Imports: a compatibility layer is in place to import Python 2 or
3 version of winreg, pickle, io.StringIO and what not.
2. Compiles fine via SCons 2.x.
3. Possible to see the first initialization phase of NVDA launcher
(nvda.pyw) via command line. The biggest showstopper at this time
is the fact that logHandler.StreamHandler constructor uses file
function when Python 3 says it isn't available, which suggests we
need to modernize log handler module.



Benefits:



1. Allows NV Access to look at porting data and make informed
decisions.
2. Allows contributors and Python experts to provide expertise and
prepare the source code for Python 3 transition.
3. Allows add-on authors to look at changes and prepare their
add-ons to run on Python 3.
4. Allows interested users to be involved in a major project and
provide test data.



Disadvantages:



1. We need to part ways with Windows XP and Server 2003.
2. Stress throughout the project, as it is nothing like what we've
seen recently.
3. Some add-ons will need community intervention to get them to run
on Python 3, as some are abandonware (no longer maintained by
authors or the add-ons community don't know who wrote some of them
and lost contact with some authors).
4. Possible issues in code and other ones, especially when looking
at code and attitude conflicts.



Next steps:



1. I'm looking for volunteers and subproject leads: if you are
passionate about Python, NVDA, and want to help NVDA achieve the next
milestone, please help and find folks who can assist in this
project. In
regards to subprojects, mostly have to do with dependency checks and
tests,
testing, documentation, talking to Python Software Foundation (PSF),
recruiting Python experts and what not.
2. Cooperation: as Project Pylennium is a grand-scale project, we
cannot move forward and make it a reality unless cooperation between
developers, add-on authors, users, observers and many others is
there.
3. Promote: one way to keep the momentum going is promoting this
project far and wide.
4. Continued work: I'm looking into how to port urllib imports, as
this
module has seen huge folder restructuring in Python 3 and other work.



What you'll need for this project:



1. Python 3.6.0 32-bit or later (3.6.2 is recommended).
2. wxPython 4 alpha 3 or later with the wheel for Python 3.5 or 3.6.
3. Six module (from PYPI).
4. ConfigObj 5.0.6.



Guidelines:



1. When sending pull requests, please do NOT send them to NV Access
yet, as folks over there are working on high priority work. Instead,
send

a
pull request against josephsl/nvda branch py3000.
2. Please provide a clear commit message when sending in pull
requests
so folks can track what has changed.
3. Before you commit, please make sure to test NVDA in order to catch
regressions and write workarounds.
4. At least once a month, I will merge NV Access master branch to
keep
the code up to date and incorporate changes into this project,
sometimes
requiring transforming syntax to fit Python 3 requirements.
5. If you want to port add-ons, please do so at your earliest
convenience or when you are ready. As part of this work, I myself
will
port
my own add-ons to Python 3 (one of them is done; throughout this
year,
I'll
port rest of my add-ons). Please do not port your add-ons to Python 3
completely - I recommend making it so it'll run happily on both
Python
versions.



Thank you.

Sincerely,

Joseph

I'm sorry but microsoft stopped supporting windows XP in 2014. Why can't you upgrade to a better system instead of staying with something that isn't supported anymore. I hope you don't get malware or something that could destroy your computer so you can't use it anymore.

-----Original Message-----
From: nvda@nvda.groups.io [mailto:nvda@nvda.groups.io] On Behalf Of nasrin khaksar
Sent: Monday, July 24, 2017 6:11 AM
To: nvda@nvda.groups.io
Subject: Re: [nvda] Announcing Project Pylennium: porting NVDA code base, dependencies, and developer attitudes to Python 3.6.x

i use xp, love it and dont believe negative statement against it.
its realy great operating system forever.

On 7/24/17, Shaun Everiss <sm.everiss@gmail.com> wrote:

Well I wouldn't use xp for the net anymore its far to insecure for that.

However its good for all those 32 bit programs you need to use, ie old
games especially those that use directx 8 or whatever that is.

They are also usefull well xp is for old nokia 32 bit apps and other
older games.




On 24/07/2017 7:52 p.m., Brian's Mail list account via Groups.Io wrote:

I think at present at least, XP is not really very good for web
browsing any more due to the tendency for even the working third
party browsers to be quite slow on single core machines of that vintage.

It is however still a joy to use for those of us who hate all the
issues over too much security on single user computers that 7 8 and
now 10 impose on us all.
Brian

bglists@blueyonder.co.uk
Sent via blueyonder.
Please address personal email to:-
briang1@blueyonder.co.uk, putting 'Brian Gaff'
in the display name field.
----- Original Message ----- From: "Joseph Lee"
<joseph.lee22590@gmail.com>
To: <nvda@nvda.groups.io>
Sent: Saturday, July 22, 2017 7:07 PM
Subject: Re: [nvda] Announcing Project Pylennium: porting NVDA code
base, dependencies, and developer attitudes to Python 3.6.x


Hi,
NV Access said they'll support XP as long as possible, but it won't
be forever. I'm using Python 3.4 and 3.6 internally for this reason.
Cheers,
Joseph

-----Original Message-----
From: nvda@nvda.groups.io [mailto:nvda@nvda.groups.io] On Behalf Of
nasrin khaksar
Sent: Saturday, July 22, 2017 4:46 AM
To: nvda@nvda.groups.io
Subject: Re: [nvda] Announcing Project Pylennium: porting NVDA code
base, dependencies, and developer attitudes to Python 3.6.x

thanks for good news joseph and brian.
do you continue supporting of xp?
i became worried when i read the message for the first time.
God bless you all!

On 7/22/17, Brian's Mail list account via Groups.Io
<bglists=blueyonder.co.uk@groups.io> wrote:

Sounds sensible. At least this will preserve xp etc with the old
Python for

as long as possible. One thing you do not say, which may be obvious
to you,

but perhaps not to the average user is..
What are the reasons for change.
is it that its going to be faster, 64 bit or has better ways of
doing stuff

nvda needs to be able to do in the future?
Obviously not in your remit is that at some point an agreed last
working version of the old NVDA that can work on xp etc is going to
have to be made

available and then when add ons are updated, those that will and
won't run on the old version will need to be noted and if possible
trapped by the old

version so it cannot be scrambled by an add on.


Brian

bglists@blueyonder.co.uk
Sent via blueyonder.
Please address personal email to:-
briang1@blueyonder.co.uk, putting 'Brian Gaff'
in the display name field.
----- Original Message -----
From: "Joseph Lee" <joseph.lee22590@gmail.com>
To: <nvda@nvda.groups.io>
Sent: Saturday, July 22, 2017 7:55 AM
Subject: [nvda] Announcing Project Pylennium: porting NVDA code
base, dependencies, and developer attitudes to Python 3.6.x

Ladies and gentlemen,



Some of you have been asking NV Access and the global NVDA
community regarding porting NVDA source code and dependencies to
Python 3. I and some contributors have been brainstorming best ways
to accomplish this, and NV Access, contributors, some add-on
writers, users, and NVDA observers agree that it is important that
we look into eventually porting NVDA to Python 3.



Until recently, the possibility of moving NVDA to Python 3 seemed
to be only a dream. However, a major hurdle was overcome earlier
this year when wxPython 4 alpha was released, promising support for
Python versions 2.7 and 3.x. Apart from a major problem regarding
certain internal routines from wxPython 4 that may not be
compatible with NVDA source code, the stage is almost set to begin
our transition to Python 3, dubbed "Project Pylennium".



Project Pylennium is a new project I'm planning and working on that
should make it easier for NV Access, code contributors, add-on
writers, users and many others to experience smooth transition to
Python 3. The overall goal

of
this project is to prepare NVDA's source code so it can run on both
Python

2
and 3 in most cases until developers are ready to move on. This
project also seeks to uncover possible compatibility issues for
NVDA community when the transition happens, and to document these
findings so it can serve as guidelines to current and future developers.



The current iteration of Project Pylennium (based on NVDA GitHub
issue
7105
as a roadmap) can be found on my copy of NVDA source code at:

https://github.com/josephsl/nvda

The branch to use is "py3000".



Project pylennium involves the following:



1. Gather dependencies required for Python 3 transition, such as
newer versions of Python, ConfigObj and many others. A crucial
dependency is the Six module, which adds compatibility layer
functions for both Python 2 and 3.
2. Document porting notes, which will eventually be published on
our NVDA Community wiki.
3. Create source-level compatibility layer in order to mitigate
changes in module names, standard library folder structure,
function and class names and so on.
4. Migrate most code to Python 3 syntax.
5. Frequent and regular testing to make sure regressions are not
introduced.



Progress so far:



1. Imports: a compatibility layer is in place to import Python 2 or
3 version of winreg, pickle, io.StringIO and what not.
2. Compiles fine via SCons 2.x.
3. Possible to see the first initialization phase of NVDA launcher
(nvda.pyw) via command line. The biggest showstopper at this time
is the fact that logHandler.StreamHandler constructor uses file
function when Python 3 says it isn't available, which suggests we
need to modernize log handler module.



Benefits:



1. Allows NV Access to look at porting data and make informed
decisions.
2. Allows contributors and Python experts to provide expertise and
prepare the source code for Python 3 transition.
3. Allows add-on authors to look at changes and prepare their
add-ons to run on Python 3.
4. Allows interested users to be involved in a major project and
provide test data.



Disadvantages:



1. We need to part ways with Windows XP and Server 2003.
2. Stress throughout the project, as it is nothing like what we've
seen recently.
3. Some add-ons will need community intervention to get them to run
on Python 3, as some are abandonware (no longer maintained by
authors or the add-ons community don't know who wrote some of them
and lost contact with some authors).
4. Possible issues in code and other ones, especially when looking
at code and attitude conflicts.



Next steps:



1. I'm looking for volunteers and subproject leads: if you are
passionate about Python, NVDA, and want to help NVDA achieve the next
milestone, please help and find folks who can assist in this
project. In
regards to subprojects, mostly have to do with dependency checks and
tests,
testing, documentation, talking to Python Software Foundation (PSF),
recruiting Python experts and what not.
2. Cooperation: as Project Pylennium is a grand-scale project, we
cannot move forward and make it a reality unless cooperation between
developers, add-on authors, users, observers and many others is there.
3. Promote: one way to keep the momentum going is promoting this
project far and wide.
4. Continued work: I'm looking into how to port urllib imports, as this
module has seen huge folder restructuring in Python 3 and other work.



What you'll need for this project:



1. Python 3.6.0 32-bit or later (3.6.2 is recommended).
2. wxPython 4 alpha 3 or later with the wheel for Python 3.5 or 3.6.
3. Six module (from PYPI).
4. ConfigObj 5.0.6.



Guidelines:



1. When sending pull requests, please do NOT send them to NV Access
yet, as folks over there are working on high priority work. Instead,
send

a
pull request against josephsl/nvda branch py3000.
2. Please provide a clear commit message when sending in pull requests
so folks can track what has changed.
3. Before you commit, please make sure to test NVDA in order to catch
regressions and write workarounds.
4. At least once a month, I will merge NV Access master branch to keep
the code up to date and incorporate changes into this project,
sometimes
requiring transforming syntax to fit Python 3 requirements.
5. If you want to port add-ons, please do so at your earliest
convenience or when you are ready. As part of this work, I myself will
port
my own add-ons to Python 3 (one of them is done; throughout this year,
I'll
port rest of my add-ons). Please do not port your add-ons to Python 3
completely - I recommend making it so it'll run happily on both Python
versions.



Thank you.

Sincerely,

Joseph

I'm sorry, but IBM stopped supporting os/2 in 2006, why do folks still use it?

Mostly because it suits their purposes, and it does what they want, so they use it.

People use what works for them, and if that just happens to be something that isn't officially supported anymore, then so what. Everyone who uses an unsupported os probably already knows that it isn't supported, and there's reasons they choose to use it anyway. As long as there are operating systems, some people are going to use them regardless of their supported status. Dos is still used very heavily in industrial settings, and I don't see anyone yelling at those industries to update to something that is still supported by microsoft, ibm, novell, paragon, or any number of other companies. Why? Because the dos being used suits the purpose for what it's being used for, and the industrial users already know there's no support, and that makes no difference to them whatsoever.

Unless you're willing to shoulder the financial burden of upgrading each and every XP user in the world, perhaps you should attempt to offer suggestions on how they could make their systems more secure instead of just telling them to scrap a perfectly working system and move to something they neither want, or can afford.

I'm frankly a bit tired of the whole I'm better than you because I use a modern os attitude that is prevalent on this list (and other places too, don't worry, it's not just here), let's attempt to provide support (which is the point of this list in the first place) instead of putting folks down because they don't want or can't afford to run the latest and greatest hardware/os, ok?



On 7/24/2017 1:53 PM, Rosemarie Chavarria wrote:

I'm sorry but microsoft stopped supporting windows XP in 2014. Why can't you upgrade to a better system instead of staying with something that isn't supported anymore. I hope you don't get malware or something that could destroy your computer so you can't use it anymore.

-----Original Message-----
From: nvda@nvda.groups.io [mailto:nvda@nvda.groups.io] On Behalf Of nasrin khaksar
Sent: Monday, July 24, 2017 6:11 AM
To: nvda@nvda.groups.io
Subject: Re: [nvda] Announcing Project Pylennium: porting NVDA code base, dependencies, and developer attitudes to Python 3.6.x

i use xp, love it and dont believe negative statement against it.
its realy great operating system forever.

On 7/24/17, Shaun Everiss <sm.everiss@gmail.com> wrote:

Well I wouldn't use xp for the net anymore its far to insecure for that.

However its good for all those 32 bit programs you need to use, ie old
games especially those that use directx 8 or whatever that is.

They are also usefull well xp is for old nokia 32 bit apps and other
older games.




On 24/07/2017 7:52 p.m., Brian's Mail list account via Groups.Io wrote:

I think at present at least, XP is not really very good for web
browsing any more due to the tendency for even the working third
party browsers to be quite slow on single core machines of that vintage.

It is however still a joy to use for those of us who hate all the
issues over too much security on single user computers that 7 8 and
now 10 impose on us all.
Brian

bglists@blueyonder.co.uk
Sent via blueyonder.
Please address personal email to:-
briang1@blueyonder.co.uk, putting 'Brian Gaff'
in the display name field.
----- Original Message ----- From: "Joseph Lee"
<joseph.lee22590@gmail.com>
To: <nvda@nvda.groups.io>
Sent: Saturday, July 22, 2017 7:07 PM
Subject: Re: [nvda] Announcing Project Pylennium: porting NVDA code
base, dependencies, and developer attitudes to Python 3.6.x


Hi,
NV Access said they'll support XP as long as possible, but it won't
be forever. I'm using Python 3.4 and 3.6 internally for this reason.
Cheers,
Joseph

-----Original Message-----
From: nvda@nvda.groups.io [mailto:nvda@nvda.groups.io] On Behalf Of
nasrin khaksar
Sent: Saturday, July 22, 2017 4:46 AM
To: nvda@nvda.groups.io
Subject: Re: [nvda] Announcing Project Pylennium: porting NVDA code
base, dependencies, and developer attitudes to Python 3.6.x

thanks for good news joseph and brian.
do you continue supporting of xp?
i became worried when i read the message for the first time.
God bless you all!

On 7/22/17, Brian's Mail list account via Groups.Io
<bglists=blueyonder.co.uk@groups.io> wrote:

Sounds sensible. At least this will preserve xp etc with the old
Python for

as long as possible. One thing you do not say, which may be obvious
to you,

but perhaps not to the average user is..
What are the reasons for change.
is it that its going to be faster, 64 bit or has better ways of
doing stuff

nvda needs to be able to do in the future?
Obviously not in your remit is that at some point an agreed last
working version of the old NVDA that can work on xp etc is going to
have to be made

available and then when add ons are updated, those that will and
won't run on the old version will need to be noted and if possible
trapped by the old

version so it cannot be scrambled by an add on.


Brian

bglists@blueyonder.co.uk
Sent via blueyonder.
Please address personal email to:-
briang1@blueyonder.co.uk, putting 'Brian Gaff'
in the display name field.
----- Original Message -----
From: "Joseph Lee" <joseph.lee22590@gmail.com>
To: <nvda@nvda.groups.io>
Sent: Saturday, July 22, 2017 7:55 AM
Subject: [nvda] Announcing Project Pylennium: porting NVDA code
base, dependencies, and developer attitudes to Python 3.6.x

Ladies and gentlemen,



Some of you have been asking NV Access and the global NVDA
community regarding porting NVDA source code and dependencies to
Python 3. I and some contributors have been brainstorming best ways
to accomplish this, and NV Access, contributors, some add-on
writers, users, and NVDA observers agree that it is important that
we look into eventually porting NVDA to Python 3.



Until recently, the possibility of moving NVDA to Python 3 seemed
to be only a dream. However, a major hurdle was overcome earlier
this year when wxPython 4 alpha was released, promising support for
Python versions 2.7 and 3.x. Apart from a major problem regarding
certain internal routines from wxPython 4 that may not be
compatible with NVDA source code, the stage is almost set to begin
our transition to Python 3, dubbed "Project Pylennium".



Project Pylennium is a new project I'm planning and working on that
should make it easier for NV Access, code contributors, add-on
writers, users and many others to experience smooth transition to
Python 3. The overall goal

of
this project is to prepare NVDA's source code so it can run on both
Python

2
and 3 in most cases until developers are ready to move on. This
project also seeks to uncover possible compatibility issues for
NVDA community when the transition happens, and to document these
findings so it can serve as guidelines to current and future developers.



The current iteration of Project Pylennium (based on NVDA GitHub
issue
7105
as a roadmap) can be found on my copy of NVDA source code at:

https://github.com/josephsl/nvda

The branch to use is "py3000".



Project pylennium involves the following:



1. Gather dependencies required for Python 3 transition, such as
newer versions of Python, ConfigObj and many others. A crucial
dependency is the Six module, which adds compatibility layer
functions for both Python 2 and 3.
2. Document porting notes, which will eventually be published on
our NVDA Community wiki.
3. Create source-level compatibility layer in order to mitigate
changes in module names, standard library folder structure,
function and class names and so on.
4. Migrate most code to Python 3 syntax.
5. Frequent and regular testing to make sure regressions are not
introduced.



Progress so far:



1. Imports: a compatibility layer is in place to import Python 2 or
3 version of winreg, pickle, io.StringIO and what not.
2. Compiles fine via SCons 2.x.
3. Possible to see the first initialization phase of NVDA launcher
(nvda.pyw) via command line. The biggest showstopper at this time
is the fact that logHandler.StreamHandler constructor uses file
function when Python 3 says it isn't available, which suggests we
need to modernize log handler module.



Benefits:



1. Allows NV Access to look at porting data and make informed
decisions.
2. Allows contributors and Python experts to provide expertise and
prepare the source code for Python 3 transition.
3. Allows add-on authors to look at changes and prepare their
add-ons to run on Python 3.
4. Allows interested users to be involved in a major project and
provide test data.



Disadvantages:



1. We need to part ways with Windows XP and Server 2003.
2. Stress throughout the project, as it is nothing like what we've
seen recently.
3. Some add-ons will need community intervention to get them to run
on Python 3, as some are abandonware (no longer maintained by
authors or the add-ons community don't know who wrote some of them
and lost contact with some authors).
4. Possible issues in code and other ones, especially when looking
at code and attitude conflicts.



Next steps:



1. I'm looking for volunteers and subproject leads: if you are
passionate about Python, NVDA, and want to help NVDA achieve the next
milestone, please help and find folks who can assist in this
project. In
regards to subprojects, mostly have to do with dependency checks and
tests,
testing, documentation, talking to Python Software Foundation (PSF),
recruiting Python experts and what not.
2. Cooperation: as Project Pylennium is a grand-scale project, we
cannot move forward and make it a reality unless cooperation between
developers, add-on authors, users, observers and many others is there.
3. Promote: one way to keep the momentum going is promoting this
project far and wide.
4. Continued work: I'm looking into how to port urllib imports, as this
module has seen huge folder restructuring in Python 3 and other work.



What you'll need for this project:



1. Python 3.6.0 32-bit or later (3.6.2 is recommended).
2. wxPython 4 alpha 3 or later with the wheel for Python 3.5 or 3.6.
3. Six module (from PYPI).
4. ConfigObj 5.0.6.



Guidelines:



1. When sending pull requests, please do NOT send them to NV Access
yet, as folks over there are working on high priority work. Instead,
send

a
pull request against josephsl/nvda branch py3000.
2. Please provide a clear commit message when sending in pull requests
so folks can track what has changed.
3. Before you commit, please make sure to test NVDA in order to catch
regressions and write workarounds.
4. At least once a month, I will merge NV Access master branch to keep
the code up to date and incorporate changes into this project,
sometimes
requiring transforming syntax to fit Python 3 requirements.
5. If you want to port add-ons, please do so at your earliest
convenience or when you are ready. As part of this work, I myself will
port
my own add-ons to Python 3 (one of them is done; throughout this year,
I'll
port rest of my add-ons). Please do not port your add-ons to Python 3
completely - I recommend making it so it'll run happily on both Python
versions.



Thank you.

Sincerely,

Joseph

Well I wouldn't use xp for the net anymore its far to insecure for that.

However its good for all those 32 bit programs you need to use, ie old
games especially those that use directx 8 or whatever that is.

They are also usefull well xp is for old nokia 32 bit apps and other
older games.




On 24/07/2017 7:52 p.m., Brian's Mail list account via Groups.Io wrote:

I think at present at least, XP is not really very good for web
browsing any more due to the tendency for even the working third
party browsers to be quite slow on single core machines of that vintage.

It is however still a joy to use for those of us who hate all the
issues over too much security on single user computers that 7 8 and
now 10 impose on us all.
Brian

bglists@blueyonder.co.uk
Sent via blueyonder.
Please address personal email to:-
briang1@blueyonder.co.uk, putting 'Brian Gaff'
in the display name field.
----- Original Message ----- From: "Joseph Lee"
<joseph.lee22590@gmail.com>
To: <nvda@nvda.groups.io>
Sent: Saturday, July 22, 2017 7:07 PM
Subject: Re: [nvda] Announcing Project Pylennium: porting NVDA code
base, dependencies, and developer attitudes to Python 3.6.x


Hi,
NV Access said they'll support XP as long as possible, but it won't
be forever. I'm using Python 3.4 and 3.6 internally for this reason.
Cheers,
Joseph

-----Original Message-----
From: nvda@nvda.groups.io [mailto:nvda@nvda.groups.io] On Behalf Of
nasrin khaksar
Sent: Saturday, July 22, 2017 4:46 AM
To: nvda@nvda.groups.io
Subject: Re: [nvda] Announcing Project Pylennium: porting NVDA code
base, dependencies, and developer attitudes to Python 3.6.x

thanks for good news joseph and brian.
do you continue supporting of xp?
i became worried when i read the message for the first time.
God bless you all!

On 7/22/17, Brian's Mail list account via Groups.Io
<bglists=blueyonder.co.uk@groups.io> wrote:

Sounds sensible. At least this will preserve xp etc with the old
Python for

as long as possible. One thing you do not say, which may be obvious
to you,

but perhaps not to the average user is..
What are the reasons for change.
is it that its going to be faster, 64 bit or has better ways of
doing stuff

nvda needs to be able to do in the future?
Obviously not in your remit is that at some point an agreed last
working version of the old NVDA that can work on xp etc is going to
have to be made

available and then when add ons are updated, those that will and
won't run on the old version will need to be noted and if possible
trapped by the old

version so it cannot be scrambled by an add on.


Brian

bglists@blueyonder.co.uk
Sent via blueyonder.
Please address personal email to:-
briang1@blueyonder.co.uk, putting 'Brian Gaff'
in the display name field.
----- Original Message -----
From: "Joseph Lee" <joseph.lee22590@gmail.com>
To: <nvda@nvda.groups.io>
Sent: Saturday, July 22, 2017 7:55 AM
Subject: [nvda] Announcing Project Pylennium: porting NVDA code
base, dependencies, and developer attitudes to Python 3.6.x

Ladies and gentlemen,



Some of you have been asking NV Access and the global NVDA
community regarding porting NVDA source code and dependencies to
Python 3. I and some contributors have been brainstorming best ways
to accomplish this, and NV Access, contributors, some add-on
writers, users, and NVDA observers agree that it is important that
we look into eventually porting NVDA to Python 3.



Until recently, the possibility of moving NVDA to Python 3 seemed
to be only a dream. However, a major hurdle was overcome earlier
this year when wxPython 4 alpha was released, promising support for
Python versions 2.7 and 3.x. Apart from a major problem regarding
certain internal routines from wxPython 4 that may not be
compatible with NVDA source code, the stage is almost set to begin
our transition to Python 3, dubbed "Project Pylennium".



Project Pylennium is a new project I'm planning and working on that
should make it easier for NV Access, code contributors, add-on
writers, users and many others to experience smooth transition to
Python 3. The overall goal

of
this project is to prepare NVDA's source code so it can run on both
Python

2
and 3 in most cases until developers are ready to move on. This
project also seeks to uncover possible compatibility issues for
NVDA community when the transition happens, and to document these
findings so it can serve as guidelines to current and future developers.



The current iteration of Project Pylennium (based on NVDA GitHub
issue
7105
as a roadmap) can be found on my copy of NVDA source code at:

https://github.com/josephsl/nvda

The branch to use is "py3000".



Project pylennium involves the following:



1. Gather dependencies required for Python 3 transition, such as
newer versions of Python, ConfigObj and many others. A crucial
dependency is the Six module, which adds compatibility layer
functions for both Python 2 and 3.
2. Document porting notes, which will eventually be published on
our NVDA Community wiki.
3. Create source-level compatibility layer in order to mitigate
changes in module names, standard library folder structure,
function and class names and so on.
4. Migrate most code to Python 3 syntax.
5. Frequent and regular testing to make sure regressions are not
introduced.



Progress so far:



1. Imports: a compatibility layer is in place to import Python 2 or
3 version of winreg, pickle, io.StringIO and what not.
2. Compiles fine via SCons 2.x.
3. Possible to see the first initialization phase of NVDA launcher
(nvda.pyw) via command line. The biggest showstopper at this time
is the fact that logHandler.StreamHandler constructor uses file
function when Python 3 says it isn't available, which suggests we
need to modernize log handler module.



Benefits:



1. Allows NV Access to look at porting data and make informed
decisions.
2. Allows contributors and Python experts to provide expertise and
prepare the source code for Python 3 transition.
3. Allows add-on authors to look at changes and prepare their
add-ons to run on Python 3.
4. Allows interested users to be involved in a major project and
provide test data.



Disadvantages:



1. We need to part ways with Windows XP and Server 2003.
2. Stress throughout the project, as it is nothing like what we've
seen recently.
3. Some add-ons will need community intervention to get them to run
on Python 3, as some are abandonware (no longer maintained by
authors or the add-ons community don't know who wrote some of them
and lost contact with some authors).
4. Possible issues in code and other ones, especially when looking
at code and attitude conflicts.



Next steps:



1. I'm looking for volunteers and subproject leads: if you are
passionate about Python, NVDA, and want to help NVDA achieve the next
milestone, please help and find folks who can assist in this
project. In
regards to subprojects, mostly have to do with dependency checks and
tests,
testing, documentation, talking to Python Software Foundation (PSF),
recruiting Python experts and what not.
2. Cooperation: as Project Pylennium is a grand-scale project, we
cannot move forward and make it a reality unless cooperation between
developers, add-on authors, users, observers and many others is there.
3. Promote: one way to keep the momentum going is promoting this
project far and wide.
4. Continued work: I'm looking into how to port urllib imports, as this
module has seen huge folder restructuring in Python 3 and other work.



What you'll need for this project:



1. Python 3.6.0 32-bit or later (3.6.2 is recommended).
2. wxPython 4 alpha 3 or later with the wheel for Python 3.5 or 3.6.
3. Six module (from PYPI).
4. ConfigObj 5.0.6.



Guidelines:



1. When sending pull requests, please do NOT send them to NV Access
yet, as folks over there are working on high priority work. Instead,
send

a
pull request against josephsl/nvda branch py3000.
2. Please provide a clear commit message when sending in pull requests
so folks can track what has changed.
3. Before you commit, please make sure to test NVDA in order to catch
regressions and write workarounds.
4. At least once a month, I will merge NV Access master branch to keep
the code up to date and incorporate changes into this project,
sometimes
requiring transforming syntax to fit Python 3 requirements.
5. If you want to port add-ons, please do so at your earliest
convenience or when you are ready. As part of this work, I myself will
port
my own add-ons to Python 3 (one of them is done; throughout this year,
I'll
port rest of my add-ons). Please do not port your add-ons to Python 3
completely - I recommend making it so it'll run happily on both Python
versions.



Thank you.

Sincerely,

Joseph