-
Notifications
You must be signed in to change notification settings - Fork 43
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Issue with Windows 10/Python 3.5 64 bits #310
Comments
Thanks for the report! We'll do more testing to see if we can come up -tom On 1/14/2016 12:59 PM, G. Ronchi wrote:
|
Right now the python 3.5 version is really new and not available on most of the platforms. We will revisit this when it becomes more widely available on linux platforms were we can more easily debug the issue. For now I would recommend using python 3.4 if possible. |
Hi all, I did some high level investigations (I can't really look into the details because I don't have Visual Studio). Let me say that I was not able to find much, however I have maybe found something that could point in the right direction: it seems that most of the packages of Anaconda Python 3.4 are compiled with Visual C++ 10.0 (VS 2010) while most of the packages of Anaconda Python 3.5 are compiled with Visual C++ 14.0 (VS 2015). Even if not stated explicitly, maybe Python itself within the Anaconda distribution is re-built with the same compiler, and maybe the two compilers have different ABI (seems strange to me). In this case what could be needed to get MDSplus work with Anaconda Python 3.5 is to compile it with VS 2015. Sincerely, |
Try the latest alpha release. I did some high level investigations (I can't really look into the details because I don't have Visual Studio). Let me say that I was not able to find much, however I have maybe found something that could point in the right direction: it seems that most of the packages of Anaconda Python 3.4 are compiled with Visual C++ 10.0 (VS 2010) while most of the packages of Anaconda Python 3.5 are compiled with Visual C++ 14.0 (VS 2015). Even if not state explicitly, maybe Python itself within the Anaconda distribution is re-built with the same compiler, and maybe the two compilers have different ABI (seems srtange to me). In this case what could be needed to get MDSplus work with Anaconda Python 3.5 is to compile it with VS 2015. Sincerely, —You are receiving this because you were assigned.Reply to this email directly, view it on GitHub, or mute the thread. |
Tried both the latest stable and alpha release, still see access violation bug when trying to use openTree on Windows 10 + Anaconda/Python 3.6 |
I just did a clean install of anaconda python IPython 5.1.0 -- An enhanced Interactive Python` and MDSplus:
In [2]: c = Connection('alcdaq.psfc.mit.edu') In [3]: x = c.get('1+3') In [4]: x In [5]: t = Tree('main', -1) In [6]: t In [7]: t.getDefault() |
My problem was with the thin client method: Seems to work fine on Windows 7 64bit, python 3.6, doesn't work on Windows 10 64bit, python 3.6. This is a pretty fresh anaconda installation tested with both the stable and alpha MDSplus install. |
I tested exactly that with our servers and a fresh windows 10 + mdsplus
alpha. For trees that are there and OK, trees that are defined and not
there, and undefined tree names. Non of these seg faulted.
…On 2/28/17 11:36 AM, ktritz wrote:
My problem was with the thin client method:
c = mds.Connection('skylark')
c.openTree(tree, shot)
Seems to work fine on Windows 7 64bit, python 3.6, doesn't work on
Windows 10 64bit, python 3.6. This is a pretty fresh anaconda
installation tested with both the stable and alpha MDSplus install.
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#310 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AHHQZROVflT19TZ8Nq9RJrCwaJpC3Y37ks5rhE0pgaJpZM4HFGjj>.
|
Ok, the code you showed used the Tree instead of the openTree method, so I wasn't sure whether you tested the thin client method. |
Here's the exact output, Windows 10, 64 bit, Anaconda w/ Python 3.6: In [3]: c = mds.Connection('skylark.pppl.gov') In [4]: c.openTree('wf', 141398)OSError Traceback (most recent call last) C:\Anaconda3\lib\site-packages\mdsplus_alpha-7.1.93-py3.6.egg\MDSplus\connection.py in openTree(self, tree, shot, mode) C:\Anaconda3\lib\site-packages\mdsplus_alpha-7.1.93-py3.6.egg\MDSplus\mdsdata.py in and(self, y) C:\Anaconda3\lib\site-packages\mdsplus_alpha-7.1.93-py3.6.egg\MDSplus\mdsdata.py in execute(expr, *args) C:\Anaconda3\lib\site-packages\mdsplus_alpha-7.1.93-py3.6.egg\MDSplus_tdishr.py in TdiExecute(expression, args) C:\Anaconda3\lib\site-packages\mdsplus_alpha-7.1.93-py3.6.egg\MDSplus_tdishr.py in _TdiShrFun(function, errormessage, expression, args) OSError: exception: access violation writing 0x0000000067929020 In [5]: mds.version |
Does the connection object work otherwise?
c = mds.Connection('skylark.pppl.gov')
print c.get('1+2')
print c.get('1 : 10 : .1')
Is the shot 141398 of tree wf accessible ? Does this happen regardless
of the tree ?
I know that Greg set up both older (default) mdsip servers and newer
(beta) mdsip servers. Can you connect to skylark on a port serving a
newer version? Is it the same ? Different ?
…On 3/1/17 1:42 PM, ktritz wrote:
Here's the exact output, Windows 10, 64 bit, Anaconda w/ Python 3.6:
In [3]: c = mds.Connection('skylark.pppl.gov')
In [4]: c.openTree('wf', 141398)
OSError Traceback (most recent call last)
in ()
----> 1 c.openTree('wf', 141398)
C:\Anaconda3\lib\site-packages\mdsplus_alpha-7.1.93-py3.6.egg\MDSplus\connection.py
in openTree(self, tree, shot, mode)
202 else: # stay compatible with old TreeOpen
203 status=self.get("TreeOpen($,$)",tree,shot)
--> 204 if not ((status & 1)==1):
205 raise _Exceptions.statusToException(status)
206
C:\Anaconda3\lib\site-packages\mdsplus_alpha-7.1.93-py3.6.egg\MDSplus\mdsdata.py
in *and*(self, y)
251 """And: x.*and*(y) <==> x&y
252 @rtype <https://github.com/rtype>: Data"""
--> 253 return Data.execute('$ & $',self,y)
254
255 def *div*(self,y):
C:\Anaconda3\lib\site-packages\mdsplus_alpha-7.1.93-py3.6.egg\MDSplus\mdsdata.py
in execute(expr, *args)
485 """Execute and expression inserting optional arguments into the
expression before evaluating
486 @rtype <https://github.com/rtype>: Data"""
--> 487 return _tdishr.TdiExecute(expr,args)
488 execute=staticmethod(execute)
489
C:\Anaconda3\lib\site-packages\mdsplus_alpha-7.1.93-py3.6.egg\MDSplus_tdishr.py
in TdiExecute(expression, args)
41 def TdiExecute(expression,args=None):
42 """Compile and execute a TDI expression. Format:
TdiExecute('expression-string')"""
---> 43 return _TdiShrFun(_TdiShr.TdiExecute,"Error
executing",expression,args)
44
45 def TdiDecompile(expression):
C:\Anaconda3\lib\site-packages\mdsplus_alpha-7.1.93-py3.6.egg\MDSplus_tdishr.py
in _TdiShrFun(function, errormessage, expression, args)
29 xd = _descriptor.descriptor_xd()
30 arguments =
[_C.byref(descriptor(expression))]+parseArguments(args)+[_C.byref(xd),_C.c_void_p(-1)]
---> 31 status = function(*arguments)
32 if (status & 1 != 0):
33 return xd.value
OSError: exception: access violation writing 0x0000000067929020
In [5]: mds.*version*
Out[5]: '7.1.93'
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#310 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AHHQZbcpK6L8lBuSvGyelbJg0oLygVi1ks5rhbwggaJpZM4HFGjj>.
|
Both c.get functions also throw an access violation exception in python
3.6, but work fine in python 2.7. The tree/shot is reachable using Python
2.7 on Windows 10, and on 2.7 & 3.6 on Windows 7. This appears to be only a
problem with Windows 10 and Python 3.6 as far as I can tell. I'll check
with Greg regarding connecting to newer servers.
On Thu, Mar 2, 2017 at 3:24 PM, Josh Stillerman <notifications@github.com>
wrote:
… Does the connection object work otherwise?
> c = mds.Connection('skylark.pppl.gov')
>
> print c.get('1+2')
>
> print c.get('1 : 10 : .1')
>
>
Is the shot 141398 of tree wf accessible ? Does this happen regardless
of the tree ?
I know that Greg set up both older (default) mdsip servers and newer
(beta) mdsip servers. Can you connect to skylark on a port serving a
newer version? Is it the same ? Different ?
On 3/1/17 1:42 PM, ktritz wrote:
>
> Here's the exact output, Windows 10, 64 bit, Anaconda w/ Python 3.6:
>
> In [3]: c = mds.Connection('skylark.pppl.gov')
>
>
> In [4]: c.openTree('wf', 141398)
>
> OSError Traceback (most recent call last)
> in ()
> ----> 1 c.openTree('wf', 141398)
>
> C:\Anaconda3\lib\site-packages\mdsplus_alpha-7.1.93-
py3.6.egg\MDSplus\connection.py
> in openTree(self, tree, shot, mode)
> 202 else: # stay compatible with old TreeOpen
> 203 status=self.get("TreeOpen($,$)",tree,shot)
> --> 204 if not ((status & 1)==1):
> 205 raise _Exceptions.statusToException(status)
> 206
>
> C:\Anaconda3\lib\site-packages\mdsplus_alpha-7.1.93-py3.6.egg\MDSplus\mdsdata.py
> in *and*(self, y)
> 251 """And: x.*and*(y) <==> x&y
> 252 @rtype <https://github.com/rtype>: Data"""
> --> 253 return Data.execute('$ & $',self,y)
> 254
> 255 def *div*(self,y):
>
> C:\Anaconda3\lib\site-packages\mdsplus_alpha-7.1.93-py3.6.egg\MDSplus\mdsdata.py
> in execute(expr, *args)
> 485 """Execute and expression inserting optional arguments into the
> expression before evaluating
> 486 @rtype <https://github.com/rtype>: Data"""
> --> 487 return _tdishr.TdiExecute(expr,args)
> 488 execute=staticmethod(execute)
> 489
>
> C:\Anaconda3\lib\site-packages\mdsplus_alpha-7.1.93-py3.6.egg\MDSplus_tdishr.py
> in TdiExecute(expression, args)
> 41 def TdiExecute(expression,args=None):
> 42 """Compile and execute a TDI expression. Format:
> TdiExecute('expression-string')"""
> ---> 43 return _TdiShrFun(_TdiShr.TdiExecute,"Error
> executing",expression,args)
> 44
> 45 def TdiDecompile(expression):
>
> C:\Anaconda3\lib\site-packages\mdsplus_alpha-7.1.93-py3.6.egg\MDSplus_tdishr.py
> in _TdiShrFun(function, errormessage, expression, args)
> 29 xd = _descriptor.descriptor_xd()
> 30 arguments =
> [_C.byref(descriptor(expression))]+parseArguments(
args)+[_C.byref(xd),_C.c_void_p(-1)]
> ---> 31 status = function(*arguments)
> 32 if (status & 1 != 0):
> 33 return xd.value
>
> OSError: exception: access violation writing 0x0000000067929020
>
> In [5]: mds.*version*
> Out[5]: '7.1.93'
>
> —
> You are receiving this because you commented.
> Reply to this email directly, view it on GitHub
> <#310 (comment)>,
> or mute the thread
> <https://github.com/notifications/unsubscribe-auth/
AHHQZbcpK6L8lBuSvGyelbJg0oLygVi1ks5rhbwggaJpZM4HFGjj>.
>
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#310 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AFEfYUrhchL06d1tBIwMwklPKZUoc4jsks5rhyV8gaJpZM4HFGjj>
.
|
Can you verify that you are actually have the MDSplus alpha version
7.1.93 software installed on your system? I don't mean the python module
version but the MDSplus windows installed version. A simple check is to
just run mdstcl from a cmd shell and type the command "SHOW VERSION". It
is somewhat important that the python module version match the version
of the libraries you have installed. When the MDSplus python support is
included in the MDSplus windows installer it should put the matching
python code in:
C:\Program Files\MDSplus\mdsobjects\python
You can add that version of the python to your python application by
cd'ing to that directory and running:
<python-prog-you-want-to-use> setup.py install
While you can use pip to install the MDSplus module, pip will install
the latest version and this may not be compatible with the MDSplus
libraries you may have installed on your system.
…-tom
On 3/2/2017 4:40 PM, ktritz wrote:
Both c.get functions also throw an access violation exception in python
3.6, but work fine in python 2.7. The tree/shot is reachable using Python
2.7 on Windows 10, and on 2.7 & 3.6 on Windows 7. This appears to be
only a
problem with Windows 10 and Python 3.6 as far as I can tell. I'll check
with Greg regarding connecting to newer servers.
On Thu, Mar 2, 2017 at 3:24 PM, Josh Stillerman ***@***.***>
wrote:
> Does the connection object work otherwise?
>
> > c = mds.Connection('skylark.pppl.gov')
> >
> > print c.get('1+2')
> >
> > print c.get('1 : 10 : .1')
> >
> >
> Is the shot 141398 of tree wf accessible ? Does this happen regardless
> of the tree ?
>
>
> I know that Greg set up both older (default) mdsip servers and newer
> (beta) mdsip servers. Can you connect to skylark on a port serving a
> newer version? Is it the same ? Different ?
> On 3/1/17 1:42 PM, ktritz wrote:
> >
> > Here's the exact output, Windows 10, 64 bit, Anaconda w/ Python 3.6:
> >
> > In [3]: c = mds.Connection('skylark.pppl.gov')
> >
> >
> > In [4]: c.openTree('wf', 141398)
> >
> > OSError Traceback (most recent call last)
> > in ()
> > ----> 1 c.openTree('wf', 141398)
> >
> > C:\Anaconda3\lib\site-packages\mdsplus_alpha-7.1.93-
> py3.6.egg\MDSplus\connection.py
> > in openTree(self, tree, shot, mode)
> > 202 else: # stay compatible with old TreeOpen
> > 203 status=self.get("TreeOpen($,$)",tree,shot)
> > --> 204 if not ((status & 1)==1):
> > 205 raise _Exceptions.statusToException(status)
> > 206
> >
> >
C:\Anaconda3\lib\site-packages\mdsplus_alpha-7.1.93-py3.6.egg\MDSplus\mdsdata.py
>
> > in *and*(self, y)
> > 251 """And: x.*and*(y) <==> x&y
> > 252 @rtype <https://github.com/rtype>: Data"""
> > --> 253 return Data.execute('$ & $',self,y)
> > 254
> > 255 def *div*(self,y):
> >
> >
C:\Anaconda3\lib\site-packages\mdsplus_alpha-7.1.93-py3.6.egg\MDSplus\mdsdata.py
>
> > in execute(expr, *args)
> > 485 """Execute and expression inserting optional arguments into the
> > expression before evaluating
> > 486 @rtype <https://github.com/rtype>: Data"""
> > --> 487 return _tdishr.TdiExecute(expr,args)
> > 488 execute=staticmethod(execute)
> > 489
> >
> >
C:\Anaconda3\lib\site-packages\mdsplus_alpha-7.1.93-py3.6.egg\MDSplus_tdishr.py
>
> > in TdiExecute(expression, args)
> > 41 def TdiExecute(expression,args=None):
> > 42 """Compile and execute a TDI expression. Format:
> > TdiExecute('expression-string')"""
> > ---> 43 return _TdiShrFun(_TdiShr.TdiExecute,"Error
> > executing",expression,args)
> > 44
> > 45 def TdiDecompile(expression):
> >
> >
C:\Anaconda3\lib\site-packages\mdsplus_alpha-7.1.93-py3.6.egg\MDSplus_tdishr.py
>
> > in _TdiShrFun(function, errormessage, expression, args)
> > 29 xd = _descriptor.descriptor_xd()
> > 30 arguments =
> > [_C.byref(descriptor(expression))]+parseArguments(
> args)+[_C.byref(xd),_C.c_void_p(-1)]
> > ---> 31 status = function(*arguments)
> > 32 if (status & 1 != 0):
> > 33 return xd.value
> >
> > OSError: exception: access violation writing 0x0000000067929020
> >
> > In [5]: mds.*version*
> > Out[5]: '7.1.93'
> >
> > —
> > You are receiving this because you commented.
> > Reply to this email directly, view it on GitHub
> >
<#310 (comment)>,
> > or mute the thread
> > <https://github.com/notifications/unsubscribe-auth/
> AHHQZbcpK6L8lBuSvGyelbJg0oLygVi1ks5rhbwggaJpZM4HFGjj>.
>
> >
>
> —
> You are receiving this because you commented.
> Reply to this email directly, view it on GitHub
> <#310 (comment)>,
> or mute the thread
>
<https://github.com/notifications/unsubscribe-auth/AFEfYUrhchL06d1tBIwMwklPKZUoc4jsks5rhyV8gaJpZM4HFGjj>
> .
>
—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
<#310 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AI44gdB2lI1x8BFNYqb3jsX9GOmayv2Zks5rhzc5gaJpZM4HFGjj>.
|
I had installed the .93 package and did the setup.py install from the
python directory. The DLL files in the MDSplus bin_x86_x64 directory have a
2/23/2017 date stamp. The 'show version' doesn't seem to work for me:
C:\Users\ktritz>mdstcl
TCL> show version
Error: %LIB-F-KEYNOTFOU, Key not found
TCL> show server skylark.pppl.gov::
Checking server: skylark.pppl.gov::
Fri Mar 3 12:40:54 2017, skylark:mdsip_standard, logging enabled, Inactive
TCL>
On Fri, Mar 3, 2017 at 12:02 PM, MDSplusBuilder <notifications@github.com>
wrote:
… Can you verify that you are actually have the MDSplus alpha version
7.1.93 software installed on your system? I don't mean the python module
version but the MDSplus windows installed version. A simple check is to
just run mdstcl from a cmd shell and type the command "SHOW VERSION". It
is somewhat important that the python module version match the version
of the libraries you have installed. When the MDSplus python support is
included in the MDSplus windows installer it should put the matching
python code in:
C:\Program Files\MDSplus\mdsobjects\python
You can add that version of the python to your python application by
cd'ing to that directory and running:
<python-prog-you-want-to-use> setup.py install
While you can use pip to install the MDSplus module, pip will install
the latest version and this may not be compatible with the MDSplus
libraries you may have installed on your system.
-tom
On 3/2/2017 4:40 PM, ktritz wrote:
> Both c.get functions also throw an access violation exception in python
> 3.6, but work fine in python 2.7. The tree/shot is reachable using Python
> 2.7 on Windows 10, and on 2.7 & 3.6 on Windows 7. This appears to be
> only a
> problem with Windows 10 and Python 3.6 as far as I can tell. I'll check
> with Greg regarding connecting to newer servers.
>
> On Thu, Mar 2, 2017 at 3:24 PM, Josh Stillerman <
***@***.***>
> wrote:
>
> > Does the connection object work otherwise?
> >
> > > c = mds.Connection('skylark.pppl.gov')
> > >
> > > print c.get('1+2')
> > >
> > > print c.get('1 : 10 : .1')
> > >
> > >
> > Is the shot 141398 of tree wf accessible ? Does this happen regardless
> > of the tree ?
> >
> >
> > I know that Greg set up both older (default) mdsip servers and newer
> > (beta) mdsip servers. Can you connect to skylark on a port serving a
> > newer version? Is it the same ? Different ?
> > On 3/1/17 1:42 PM, ktritz wrote:
> > >
> > > Here's the exact output, Windows 10, 64 bit, Anaconda w/ Python 3.6:
> > >
> > > In [3]: c = mds.Connection('skylark.pppl.gov')
> > >
> > >
> > > In [4]: c.openTree('wf', 141398)
> > >
> > > OSError Traceback (most recent call last)
> > > in ()
> > > ----> 1 c.openTree('wf', 141398)
> > >
> > > C:\Anaconda3\lib\site-packages\mdsplus_alpha-7.1.93-
> > py3.6.egg\MDSplus\connection.py
> > > in openTree(self, tree, shot, mode)
> > > 202 else: # stay compatible with old TreeOpen
> > > 203 status=self.get("TreeOpen($,$)",tree,shot)
> > > --> 204 if not ((status & 1)==1):
> > > 205 raise _Exceptions.statusToException(status)
> > > 206
> > >
> > >
> C:\Anaconda3\lib\site-packages\mdsplus_alpha-7.1.93-
py3.6.egg\MDSplus\mdsdata.py
> >
> > > in *and*(self, y)
> > > 251 """And: x.*and*(y) <==> x&y
> > > 252 @rtype <https://github.com/rtype>: Data"""
> > > --> 253 return Data.execute('$ & $',self,y)
> > > 254
> > > 255 def *div*(self,y):
> > >
> > >
> C:\Anaconda3\lib\site-packages\mdsplus_alpha-7.1.93-
py3.6.egg\MDSplus\mdsdata.py
> >
> > > in execute(expr, *args)
> > > 485 """Execute and expression inserting optional arguments into the
> > > expression before evaluating
> > > 486 @rtype <https://github.com/rtype>: Data"""
> > > --> 487 return _tdishr.TdiExecute(expr,args)
> > > 488 execute=staticmethod(execute)
> > > 489
> > >
> > >
> C:\Anaconda3\lib\site-packages\mdsplus_alpha-7.1.93-
py3.6.egg\MDSplus_tdishr.py
> >
> > > in TdiExecute(expression, args)
> > > 41 def TdiExecute(expression,args=None):
> > > 42 """Compile and execute a TDI expression. Format:
> > > TdiExecute('expression-string')"""
> > > ---> 43 return _TdiShrFun(_TdiShr.TdiExecute,"Error
> > > executing",expression,args)
> > > 44
> > > 45 def TdiDecompile(expression):
> > >
> > >
> C:\Anaconda3\lib\site-packages\mdsplus_alpha-7.1.93-
py3.6.egg\MDSplus_tdishr.py
> >
> > > in _TdiShrFun(function, errormessage, expression, args)
> > > 29 xd = _descriptor.descriptor_xd()
> > > 30 arguments =
> > > [_C.byref(descriptor(expression))]+parseArguments(
> > args)+[_C.byref(xd),_C.c_void_p(-1)]
> > > ---> 31 status = function(*arguments)
> > > 32 if (status & 1 != 0):
> > > 33 return xd.value
> > >
> > > OSError: exception: access violation writing 0x0000000067929020
> > >
> > > In [5]: mds.*version*
> > > Out[5]: '7.1.93'
> > >
> > > —
> > > You are receiving this because you commented.
> > > Reply to this email directly, view it on GitHub
> > >
> <#310 (comment)>,
> > > or mute the thread
> > > <https://github.com/notifications/unsubscribe-auth/
> > AHHQZbcpK6L8lBuSvGyelbJg0oLygVi1ks5rhbwggaJpZM4HFGjj>.
> >
> > >
> >
> > —
> > You are receiving this because you commented.
> > Reply to this email directly, view it on GitHub
> > <#310 (comment)
>,
> > or mute the thread
> >
> <https://github.com/notifications/unsubscribe-auth/
AFEfYUrhchL06d1tBIwMwklPKZUoc4jsks5rhyV8gaJpZM4HFGjj>
> > .
> >
>
> —
> You are receiving this because you are subscribed to this thread.
> Reply to this email directly, view it on GitHub
> <#310 (comment)>,
> or mute the thread
> <https://github.com/notifications/unsubscribe-auth/
AI44gdB2lI1x8BFNYqb3jsX9GOmayv2Zks5rhzc5gaJpZM4HFGjj>.
>
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#310 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AFEfYalBVG7lor9Wwkcu7n_X5e4evUOvks5riEfAgaJpZM4HFGjj>
.
|
Success! Your line of inquiry got me thinking. I inspected the datestamps
of the DLLs in the Windows/System32 directory, and they were from 11/2016,
so somehow they weren't updated with one of the recent installs. So, I
manually replaced the DLLs in the Windows/System32 directory with the
version in the bin_x86_x64 directory, and now I can open the tree.
show version now works properly:
c:\Windows\System32>mdstcl
TCL> show version
MDSplus version: 7.1.93
Git branch: alpha
Git tag: alpha_release-7-1-93
View:
https://github.com/MDSplus/mdsplus/tree/alpha_release-7-1-93
Download:
https://github.com/MDSplus/mdsplus/archive/alpha_release-7-1-93.tar.gz
Git commit: dd3f02d
Build date: Thu Feb 23 08:06:07 EST 2017
Thanks for your assistance and patience.
…On Fri, Mar 3, 2017 at 12:44 PM, Kevin ***@***.***> wrote:
I had installed the .93 package and did the setup.py install from the
python directory. The DLL files in the MDSplus bin_x86_x64 directory have a
2/23/2017 date stamp. The 'show version' doesn't seem to work for me:
C:\Users\ktritz>mdstcl
TCL> show version
Error: %LIB-F-KEYNOTFOU, Key not found
TCL> show server skylark.pppl.gov::
Checking server: skylark.pppl.gov::
Fri Mar 3 12:40:54 2017, skylark:mdsip_standard, logging enabled, Inactive
TCL>
On Fri, Mar 3, 2017 at 12:02 PM, MDSplusBuilder ***@***.***>
wrote:
> Can you verify that you are actually have the MDSplus alpha version
> 7.1.93 software installed on your system? I don't mean the python module
> version but the MDSplus windows installed version. A simple check is to
> just run mdstcl from a cmd shell and type the command "SHOW VERSION". It
> is somewhat important that the python module version match the version
> of the libraries you have installed. When the MDSplus python support is
> included in the MDSplus windows installer it should put the matching
> python code in:
>
> C:\Program Files\MDSplus\mdsobjects\python
>
> You can add that version of the python to your python application by
> cd'ing to that directory and running:
>
> <python-prog-you-want-to-use> setup.py install
>
> While you can use pip to install the MDSplus module, pip will install
> the latest version and this may not be compatible with the MDSplus
> libraries you may have installed on your system.
>
> -tom
>
>
>
> On 3/2/2017 4:40 PM, ktritz wrote:
> > Both c.get functions also throw an access violation exception in python
> > 3.6, but work fine in python 2.7. The tree/shot is reachable using
> Python
> > 2.7 on Windows 10, and on 2.7 & 3.6 on Windows 7. This appears to be
> > only a
> > problem with Windows 10 and Python 3.6 as far as I can tell. I'll check
> > with Greg regarding connecting to newer servers.
> >
> > On Thu, Mar 2, 2017 at 3:24 PM, Josh Stillerman <
> ***@***.***>
> > wrote:
> >
> > > Does the connection object work otherwise?
> > >
> > > > c = mds.Connection('skylark.pppl.gov')
> > > >
> > > > print c.get('1+2')
> > > >
> > > > print c.get('1 : 10 : .1')
> > > >
> > > >
> > > Is the shot 141398 of tree wf accessible ? Does this happen regardless
> > > of the tree ?
> > >
> > >
> > > I know that Greg set up both older (default) mdsip servers and newer
> > > (beta) mdsip servers. Can you connect to skylark on a port serving a
> > > newer version? Is it the same ? Different ?
> > > On 3/1/17 1:42 PM, ktritz wrote:
> > > >
> > > > Here's the exact output, Windows 10, 64 bit, Anaconda w/ Python 3.6:
> > > >
> > > > In [3]: c = mds.Connection('skylark.pppl.gov')
> > > >
> > > >
> > > > In [4]: c.openTree('wf', 141398)
> > > >
> > > > OSError Traceback (most recent call last)
> > > > in ()
> > > > ----> 1 c.openTree('wf', 141398)
> > > >
> > > > C:\Anaconda3\lib\site-packages\mdsplus_alpha-7.1.93-
> > > py3.6.egg\MDSplus\connection.py
> > > > in openTree(self, tree, shot, mode)
> > > > 202 else: # stay compatible with old TreeOpen
> > > > 203 status=self.get("TreeOpen($,$)",tree,shot)
> > > > --> 204 if not ((status & 1)==1):
> > > > 205 raise _Exceptions.statusToException(status)
> > > > 206
> > > >
> > > >
> > C:\Anaconda3\lib\site-packages\mdsplus_alpha-7.1.93-py3.6.
> egg\MDSplus\mdsdata.py
> > >
> > > > in *and*(self, y)
> > > > 251 """And: x.*and*(y) <==> x&y
> > > > 252 @rtype <https://github.com/rtype>: Data"""
> > > > --> 253 return Data.execute('$ & $',self,y)
> > > > 254
> > > > 255 def *div*(self,y):
> > > >
> > > >
> > C:\Anaconda3\lib\site-packages\mdsplus_alpha-7.1.93-py3.6.
> egg\MDSplus\mdsdata.py
> > >
> > > > in execute(expr, *args)
> > > > 485 """Execute and expression inserting optional arguments into the
> > > > expression before evaluating
> > > > 486 @rtype <https://github.com/rtype>: Data"""
> > > > --> 487 return _tdishr.TdiExecute(expr,args)
> > > > 488 execute=staticmethod(execute)
> > > > 489
> > > >
> > > >
> > C:\Anaconda3\lib\site-packages\mdsplus_alpha-7.1.93-py3.6.
> egg\MDSplus_tdishr.py
> > >
> > > > in TdiExecute(expression, args)
> > > > 41 def TdiExecute(expression,args=None):
> > > > 42 """Compile and execute a TDI expression. Format:
> > > > TdiExecute('expression-string')"""
> > > > ---> 43 return _TdiShrFun(_TdiShr.TdiExecute,"Error
> > > > executing",expression,args)
> > > > 44
> > > > 45 def TdiDecompile(expression):
> > > >
> > > >
> > C:\Anaconda3\lib\site-packages\mdsplus_alpha-7.1.93-py3.6.
> egg\MDSplus_tdishr.py
> > >
> > > > in _TdiShrFun(function, errormessage, expression, args)
> > > > 29 xd = _descriptor.descriptor_xd()
> > > > 30 arguments =
> > > > [_C.byref(descriptor(expression))]+parseArguments(
> > > args)+[_C.byref(xd),_C.c_void_p(-1)]
> > > > ---> 31 status = function(*arguments)
> > > > 32 if (status & 1 != 0):
> > > > 33 return xd.value
> > > >
> > > > OSError: exception: access violation writing 0x0000000067929020
> > > >
> > > > In [5]: mds.*version*
> > > > Out[5]: '7.1.93'
> > > >
> > > > —
> > > > You are receiving this because you commented.
> > > > Reply to this email directly, view it on GitHub
> > > >
> > <#310 (comment)>,
> > > > or mute the thread
> > > > <https://github.com/notifications/unsubscribe-auth/
> > > AHHQZbcpK6L8lBuSvGyelbJg0oLygVi1ks5rhbwggaJpZM4HFGjj>.
> > >
> > > >
> > >
> > > —
> > > You are receiving this because you commented.
> > > Reply to this email directly, view it on GitHub
> > > <#310 (comment)
> >,
> > > or mute the thread
> > >
> > <https://github.com/notifications/unsubscribe-auth/AFEfYUrhc
> hL06d1tBIwMwklPKZUoc4jsks5rhyV8gaJpZM4HFGjj>
> > > .
> > >
> >
> > —
> > You are receiving this because you are subscribed to this thread.
> > Reply to this email directly, view it on GitHub
> > <#310 (comment)>,
>
> > or mute the thread
> > <https://github.com/notifications/unsubscribe-auth/AI44gdB2l
> I1x8BFNYqb3jsX9GOmayv2Zks5rhzc5gaJpZM4HFGjj>.
>
> >
>
> —
> You are receiving this because you commented.
> Reply to this email directly, view it on GitHub
> <#310 (comment)>,
> or mute the thread
> <https://github.com/notifications/unsubscribe-auth/AFEfYalBVG7lor9Wwkcu7n_X5e4evUOvks5riEfAgaJpZM4HFGjj>
> .
>
|
Oh! That explains it. You must have had an old MDSplus installation on
the PC. A couple years ago we switched from using Visual Studio
installers to a different package manager and the new package manager
could not remove the 6.n VS based MDSplus. On the MDSplus windows
download page there is an explanation just about the links to the kits:
"The following are the latest builds. If you are upgrading from a 6.x
release of MDSplus you will need to first manually deinstall your
installed version before installing a 7.x version. This is because the
installation utility used has changed in 7.0 and does not recognize
pre-7.0 installations."
I don't know if your system still thinks it has a 6.x version installed
and if you can remove it at this time or not. If you still have problems
you might consider deinstalling any MDSplus windows packages your system
thinks it has installed and after that remove the c:\Program
Files\MDSplus directory tree if anything remains and then try installing
the 7.1 package again.
…On 3/3/2017 12:50 PM, ktritz wrote:
Success! Your line of inquiry got me thinking. I inspected the datestamps
of the DLLs in the Windows/System32 directory, and they were from 11/2016,
so somehow they weren't updated with one of the recent installs. So, I
manually replaced the DLLs in the Windows/System32 directory with the
version in the bin_x86_x64 directory, and now I can open the tree.
show version now works properly:
c:\Windows\System32>mdstcl
TCL> show version
MDSplus version: 7.1.93
Git branch: alpha
Git tag: alpha_release-7-1-93
View:
https://github.com/MDSplus/mdsplus/tree/alpha_release-7-1-93
Download:
https://github.com/MDSplus/mdsplus/archive/alpha_release-7-1-93.tar.gz
Git commit: dd3f02d
Build date: Thu Feb 23 08:06:07 EST 2017
Thanks for your assistance and patience.
On Fri, Mar 3, 2017 at 12:44 PM, Kevin ***@***.***> wrote:
> I had installed the .93 package and did the setup.py install from the
> python directory. The DLL files in the MDSplus bin_x86_x64 directory
have a
> 2/23/2017 date stamp. The 'show version' doesn't seem to work for me:
>
> C:\Users\ktritz>mdstcl
> TCL> show version
> Error: %LIB-F-KEYNOTFOU, Key not found
> TCL> show server skylark.pppl.gov::
> Checking server: skylark.pppl.gov::
> Fri Mar 3 12:40:54 2017, skylark:mdsip_standard, logging enabled,
Inactive
> TCL>
>
>
>
> On Fri, Mar 3, 2017 at 12:02 PM, MDSplusBuilder
***@***.***>
> wrote:
>
>> Can you verify that you are actually have the MDSplus alpha version
>> 7.1.93 software installed on your system? I don't mean the python
module
>> version but the MDSplus windows installed version. A simple check is to
>> just run mdstcl from a cmd shell and type the command "SHOW
VERSION". It
>> is somewhat important that the python module version match the version
>> of the libraries you have installed. When the MDSplus python support is
>> included in the MDSplus windows installer it should put the matching
>> python code in:
>>
>> C:\Program Files\MDSplus\mdsobjects\python
>>
>> You can add that version of the python to your python application by
>> cd'ing to that directory and running:
>>
>> <python-prog-you-want-to-use> setup.py install
>>
>> While you can use pip to install the MDSplus module, pip will install
>> the latest version and this may not be compatible with the MDSplus
>> libraries you may have installed on your system.
>>
>> -tom
>>
>>
>>
>> On 3/2/2017 4:40 PM, ktritz wrote:
>> > Both c.get functions also throw an access violation exception in
python
>> > 3.6, but work fine in python 2.7. The tree/shot is reachable using
>> Python
>> > 2.7 on Windows 10, and on 2.7 & 3.6 on Windows 7. This appears to be
>> > only a
>> > problem with Windows 10 and Python 3.6 as far as I can tell. I'll
check
>> > with Greg regarding connecting to newer servers.
>> >
>> > On Thu, Mar 2, 2017 at 3:24 PM, Josh Stillerman <
>> ***@***.***>
>> > wrote:
>> >
>> > > Does the connection object work otherwise?
>> > >
>> > > > c = mds.Connection('skylark.pppl.gov')
>> > > >
>> > > > print c.get('1+2')
>> > > >
>> > > > print c.get('1 : 10 : .1')
>> > > >
>> > > >
>> > > Is the shot 141398 of tree wf accessible ? Does this happen
regardless
>> > > of the tree ?
>> > >
>> > >
>> > > I know that Greg set up both older (default) mdsip servers and
newer
>> > > (beta) mdsip servers. Can you connect to skylark on a port
serving a
>> > > newer version? Is it the same ? Different ?
>> > > On 3/1/17 1:42 PM, ktritz wrote:
>> > > >
>> > > > Here's the exact output, Windows 10, 64 bit, Anaconda w/
Python 3.6:
>> > > >
>> > > > In [3]: c = mds.Connection('skylark.pppl.gov')
>> > > >
>> > > >
>> > > > In [4]: c.openTree('wf', 141398)
>> > > >
>> > > > OSError Traceback (most recent call last)
>> > > > in ()
>> > > > ----> 1 c.openTree('wf', 141398)
>> > > >
>> > > > C:\Anaconda3\lib\site-packages\mdsplus_alpha-7.1.93-
>> > > py3.6.egg\MDSplus\connection.py
>> > > > in openTree(self, tree, shot, mode)
>> > > > 202 else: # stay compatible with old TreeOpen
>> > > > 203 status=self.get("TreeOpen($,$)",tree,shot)
>> > > > --> 204 if not ((status & 1)==1):
>> > > > 205 raise _Exceptions.statusToException(status)
>> > > > 206
>> > > >
>> > > >
>> > C:\Anaconda3\lib\site-packages\mdsplus_alpha-7.1.93-py3.6.
>> egg\MDSplus\mdsdata.py
>> > >
>> > > > in *and*(self, y)
>> > > > 251 """And: x.*and*(y) <==> x&y
>> > > > 252 @rtype <https://github.com/rtype>: Data"""
>> > > > --> 253 return Data.execute('$ & $',self,y)
>> > > > 254
>> > > > 255 def *div*(self,y):
>> > > >
>> > > >
>> > C:\Anaconda3\lib\site-packages\mdsplus_alpha-7.1.93-py3.6.
>> egg\MDSplus\mdsdata.py
>> > >
>> > > > in execute(expr, *args)
>> > > > 485 """Execute and expression inserting optional arguments
into the
>> > > > expression before evaluating
>> > > > 486 @rtype <https://github.com/rtype>: Data"""
>> > > > --> 487 return _tdishr.TdiExecute(expr,args)
>> > > > 488 execute=staticmethod(execute)
>> > > > 489
>> > > >
>> > > >
>> > C:\Anaconda3\lib\site-packages\mdsplus_alpha-7.1.93-py3.6.
>> egg\MDSplus_tdishr.py
>> > >
>> > > > in TdiExecute(expression, args)
>> > > > 41 def TdiExecute(expression,args=None):
>> > > > 42 """Compile and execute a TDI expression. Format:
>> > > > TdiExecute('expression-string')"""
>> > > > ---> 43 return _TdiShrFun(_TdiShr.TdiExecute,"Error
>> > > > executing",expression,args)
>> > > > 44
>> > > > 45 def TdiDecompile(expression):
>> > > >
>> > > >
>> > C:\Anaconda3\lib\site-packages\mdsplus_alpha-7.1.93-py3.6.
>> egg\MDSplus_tdishr.py
>> > >
>> > > > in _TdiShrFun(function, errormessage, expression, args)
>> > > > 29 xd = _descriptor.descriptor_xd()
>> > > > 30 arguments =
>> > > > [_C.byref(descriptor(expression))]+parseArguments(
>> > > args)+[_C.byref(xd),_C.c_void_p(-1)]
>> > > > ---> 31 status = function(*arguments)
>> > > > 32 if (status & 1 != 0):
>> > > > 33 return xd.value
>> > > >
>> > > > OSError: exception: access violation writing 0x0000000067929020
>> > > >
>> > > > In [5]: mds.*version*
>> > > > Out[5]: '7.1.93'
>> > > >
>> > > > —
>> > > > You are receiving this because you commented.
>> > > > Reply to this email directly, view it on GitHub
>> > > >
>> >
<#310 (comment)>,
>> > > > or mute the thread
>> > > > <https://github.com/notifications/unsubscribe-auth/
>> > > AHHQZbcpK6L8lBuSvGyelbJg0oLygVi1ks5rhbwggaJpZM4HFGjj>.
>> > >
>> > > >
>> > >
>> > > —
>> > > You are receiving this because you commented.
>> > > Reply to this email directly, view it on GitHub
>> > >
<#310 (comment)
>> >,
>> > > or mute the thread
>> > >
>> > <https://github.com/notifications/unsubscribe-auth/AFEfYUrhc
>> hL06d1tBIwMwklPKZUoc4jsks5rhyV8gaJpZM4HFGjj>
>> > > .
>> > >
>> >
>> > —
>> > You are receiving this because you are subscribed to this thread.
>> > Reply to this email directly, view it on GitHub
>> >
<#310 (comment)>,
>>
>> > or mute the thread
>> > <https://github.com/notifications/unsubscribe-auth/AI44gdB2l
>> I1x8BFNYqb3jsX9GOmayv2Zks5rhzc5gaJpZM4HFGjj>.
>>
>> >
>>
>> —
>> You are receiving this because you commented.
>> Reply to this email directly, view it on GitHub
>> <#310 (comment)>,
>> or mute the thread
>>
<https://github.com/notifications/unsubscribe-auth/AFEfYalBVG7lor9Wwkcu7n_X5e4evUOvks5riEfAgaJpZM4HFGjj>
>> .
>>
>
>
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#310 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AI44gQkPQTD-jx_FW9BmODrhseYdbXQzks5riFLogaJpZM4HFGjj>.
|
Using Windows 10 (or Windows Server 2012) with Python 3.5 64 bits, the ctypes seems to generate some memory access violation as shown below.
It's working fine with older windows versions and/or older python versions as well as with Windows 10 + python 3.5 32 bits. It probably should be reported to the Python bug list, but I'm not able to produce a simple example (like https://bugs.python.org/issue17310) that could reproduce this bug.
The text was updated successfully, but these errors were encountered: