Re: making executables smaller
Carter Temm wrote: > I’m writing a couple different projects at the moment, and when I compile > it into a single executable using pyinstaller, it becomes extremely large. > I’m guessing this is because of the modules used. Because I’m not that > skilled at python, I put stuff like for example, import sys. I imagine the > final project could be made smaller by specifying from something import > something_else. but the thing is, I don’t know what smaller I could import > with these set of modules. Is there a program that could tell me this. I recommend to comment out all “import” statements (for later reference) and then use a Python editor like PyDev to generate step by step “from … import …” statements for all used symbols that are not yet defined. If it is a rather simple program where you can easily make no-ops out of modifying statements by defining mockups, you can just run it before you compile it to see where you get errors with commented-out “import” statements. In theory, you could look for code of the form “foo.bar” using the regular expression "(?:"")?(?:[^"\\]|\\"(?:"")?)+"(?:"")?|'(?:'')?(?:[^'\\]| \\'(?:'')?)+'(?:'')?)|((?:\w+\.)+)(\w+)¹. If $1 (e.g., “foo”) is not defined in the file, then it is likely that $2 (e.g., “bar”) is a symbol from module $1 and you could write #- from $1 import $2 … $2 #- instead of #- import $1 … $1.$2 #- But there can be false positives and duplicate symbols this way. _ ¹ the first two alternatives exclude Python string literals -- PointedEars Twitter: @PointedEars2 Please do not cc me. / Bitte keine Kopien per E-Mail. -- https://mail.python.org/mailman/listinfo/python-list
Re: Python Print Error
On Wednesday 27 July 2016 13:45, Cai Gengyang wrote:
> How to debug this error message ?
Start by reading the message:
invalid literal for int() with base 10: ''
Now try to experiment at the interactive interpreter:
int('45') # works
int('xyz') # ValueError: invalid literal for int() with base 10: 'xyz'
int('QWE') # ValueError: invalid literal for int() with base 10: 'QWE'
int('!@#') # ValueError: invalid literal for int() with base 10: '!@#'
What do you think int('') will do? Try it and see. Were you right?
> print('You will be ' + str(int(myAge) + 1) + ' in a year.')
> Traceback (most recent call last):
> File "", line 1, in
> print('You will be ' + str(int(myAge) + 1) + ' in a year.')
> ValueError: invalid literal for int() with base 10: ''
What value do you think myAge has?
Hint: you call int(myAge). That raises ValueError, and says that '' is an
invalid literal for int. What value do you think myAge must have?
Where does myAge get its value from?
--
Steve
--
https://mail.python.org/mailman/listinfo/python-list
Re: making executables smaller
On Wednesday 27 July 2016 14:52, Thomas 'PointedEars' Lahn wrote:
> Carter Temm wrote:
>
>> I’m writing a couple different projects at the moment, and when I compile
>> it into a single executable using pyinstaller, it becomes extremely large.
>> I’m guessing this is because of the modules used. Because I’m not that
>> skilled at python, I put stuff like for example, import sys. I imagine the
>> final project could be made smaller by specifying from something import
>> something_else. but the thing is, I don’t know what smaller I could import
>> with these set of modules. Is there a program that could tell me this.
>
> I recommend to comment out all “import” statements (for later reference) and
> then use a Python editor like PyDev to generate step by step “from … import
> …” statements for all used symbols that are not yet defined.
What benefit do you think you will gain from changing (let's say):
import collections
x = collections.deque(foo)
to:
from collections import deque
x = deque(foo)
as far as executable size goes? Do you think that by using from...import... the
module, and all of its dependencies, won't need to be included?
Or are you just trying to save a handful of bytes ("collections.deque" in UTF-8
is 17 bytes, compared to "deque" is just 5 bytes)?
Since the OP is talking about saving space, how much space do you expect to be
able to save using this technique?
--
Steve
--
https://mail.python.org/mailman/listinfo/python-list
Re: making executables smaller
A couple of things you can try:
* Generate a directory rather than onefile, on the directory you can apply du
-hs * | sort -h -r (or treesize if you are using windows
https://www.jam-software.com/treesize_free) to see which folders / files
are taking up a lot of space. Then once you see what is taking up a lot of
space you can try and figure out why it is being included, maybe you have a
load of unused imports? With pyinstaller you can explicitly exclude modules
you know you won't need with --exclude-module once you've optimised the
directory build you can of course switch back to onefile
* If all else fails you can use upx to compress your binary files (dlls +
exe) which can help reduce the overall size:
https://pythonhosted.org/PyInstaller/usage.html#using-upx
Hope this helps.
2016-07-27 10:20 GMT+02:00 Steven D'Aprano <
steve+comp.lang.pyt...@pearwood.info>:
> On Wednesday 27 July 2016 14:52, Thomas 'PointedEars' Lahn wrote:
>
> > Carter Temm wrote:
> >
> >> I’m writing a couple different projects at the moment, and when I
> compile
> >> it into a single executable using pyinstaller, it becomes extremely
> large.
> >> I’m guessing this is because of the modules used. Because I’m not that
> >> skilled at python, I put stuff like for example, import sys. I imagine
> the
> >> final project could be made smaller by specifying from something import
> >> something_else. but the thing is, I don’t know what smaller I could
> import
> >> with these set of modules. Is there a program that could tell me this.
> >
> > I recommend to comment out all “import” statements (for later reference)
> and
> > then use a Python editor like PyDev to generate step by step “from …
> import
> > …” statements for all used symbols that are not yet defined.
>
> What benefit do you think you will gain from changing (let's say):
>
> import collections
> x = collections.deque(foo)
>
> to:
>
> from collections import deque
> x = deque(foo)
>
>
> as far as executable size goes? Do you think that by using
> from...import... the
> module, and all of its dependencies, won't need to be included?
>
> Or are you just trying to save a handful of bytes ("collections.deque" in
> UTF-8
> is 17 bytes, compared to "deque" is just 5 bytes)?
>
>
> Since the OP is talking about saving space, how much space do you expect
> to be
> able to save using this technique?
>
>
> --
> Steve
>
> --
> https://mail.python.org/mailman/listinfo/python-list
>
--
https://mail.python.org/mailman/listinfo/python-list
Two constructive reviewers sought
To promote the use of Python and formalise Python approach, I decided to publish a paper. I used geodata as a showcase. Geodata lies in the heart of geographical information science. The management and processing of such data is of great importance. I got an email from International Journal of Information Management. I was asked to suggest two reviewers for the following article. "Key solutions for automated data linkage, processing and generating output for big data analytics over the Internet". Nitrogen input from fertilizer use data was used as a showcase. If you like to be one of reviewers, please email drshishaozh...@gmail.com with your full contact details. Regards. David -- https://mail.python.org/mailman/listinfo/python-list
Re: pyinstaller
On Wed, Jul 27, 2016 at 2:23 AM, Christian Gollwitzer wrote: > Am 27.07.16 um 03:15 schrieb Larry Martell: >> >> On Tue, Jul 26, 2016 at 8:49 PM, Tom Brown wrote: >>> >>> I used pyinstaller quite a bit 3 years ago. I could brush off the cobwebs >>> and see if I can help if you have not solved it already. >>> >>> What is the issue you are having? >> >> >> If I import the requests module, then when I run the executable I get: >> >> ImportError: No module named 'requests.packages.chardet' > > > That's a classic issue. pyinstaller does static analysis of the program, > which modules must be included. If the code computes a module dynamically, > it can not always succeed. The solution is to tell pyinstaller to add this > module. In previous versions, you could add these by "pyinstaller > --hidden-import=requests.packages.chardet" or similar. If this doesn't work, > you need to edit the spec file. See here: > > https://pythonhosted.org/PyInstaller/when-things-go-wrong.html#listing-hidden-imports Yes, I had seen that and I tried it with: --hidden-import=requests.packages.chardet and I got the same error at run time: $ dist/pyi_test/pyi_test Traceback (most recent call last): File "pyi_test.py", line 1, in import requests File "/usr/lib/python2.7/site-packages/PyInstaller/loader/pyimod03_importers.py", line 389, in load_module exec(bytecode, module.dict) File "requests/init.py", line 58, in File "/usr/lib/python2.7/site-packages/PyInstaller/loader/pyimod03_importers.py", line 389, in load_module exec(bytecode, module.dict) File "requests/utils.py", line 26, in File "/usr/lib/python2.7/site-packages/PyInstaller/loader/pyimod03_importers.py", line 389, in load_module exec(bytecode, module.dict) File "requests/compat.py", line 7, in File "requests/packages/init.py", line 95, in load_module ImportError: No module named 'requests.packages.chardet' -- https://mail.python.org/mailman/listinfo/python-list
logging: getLogger() or getLogger(__name__)?
I've read that best practice for logging is to place the following line at the top of all modules: logger = getLogger(__name__) I'm curious why the following technique wouldn't be a better choice: logger = getLogger() Are there scenarios that favor one style over another? Thank you, Malcolm -- https://mail.python.org/mailman/listinfo/python-list
Re: reshape with xyz ordering
Thanks for your replies. Let me explain my problem a little bit more. I have the following data which i read from a file using numpy.loadtxt and then i sort it using np.lexsort: x=f[:,0] # XColumn y=f[:,1] # YColumn z=f[:,2] # ZColumn val=f[:,3] # Val Column xcoord=np.sort(np.unique(f[:,0])) # XCoordinates ycoord=np.sort(np.unique(f[:,1])) # YCoordinates zcoord=np.sort(np.unique(f[:,2])) # ZCoordinates ind = np.lexsort((val,z,y,x)) val_sorted=np.array(val[ind]) I know that the val column has data sorted first by x, then by y, then by z which means that column x changes slowest and column z changes fastest. x,y,z, val 0,0,0,val1 0,0,1,val2 0,0,2,val3 0,0,zn,valn ... xn,yn,zn,valfin I want to reshape val_sorted in to a 3d numpy array of (nx,ny,nz). which of the following is the correct way and why? #1 val_sorted_reshaped=val_sorted.reshape((xcoord.size,ycoord.size,zcoord.size)) #2 #val_sorted_reshaped=val_sorted.reshape((xcoord.size,ycoord.size,zcoord.size)).transpose() #3 #val_sorted_reshaped=val_sorted.reshape((zcoord.size,ycoord.size,xcoord.size)) #4 #val_sorted_reshaped=val_sorted.reshape((zcoord.size,ycoord.size,xcoord.size)).transpose() Thanks, -- https://mail.python.org/mailman/listinfo/python-list
Re: Python environment on mac
Yep, I agree. virtualenv is the best way to go so far. It solves my wishes to use python completely without root access too. Thank's a lot. Leonid -- https://mail.python.org/mailman/listinfo/python-list
Re: pyinstaller
2016-07-27 3:15 GMT+02:00 Larry Martell :
> On Tue, Jul 26, 2016 at 8:49 PM, Tom Brown wrote:
>> I used pyinstaller quite a bit 3 years ago. I could brush off the cobwebs
>> and see if I can help if you have not solved it already.
>>
>> What is the issue you are having?
>
> If I import the requests module, then when I run the executable I get:
>
> ImportError: No module named 'requests.packages.chardet'
>
> I tried to post to the pyinstaller group, but it said my post had to
> be approved by the moderator, and it apparently never was. I have no
> idea who the moderator is, so there was no one I could contact about
> that. I posted an issue to github
> (https://github.com/pyinstaller/pyinstaller/issues/2060) and some
> suggestions were made, but none fixed the problem. I am on RHEL 7.2
> with Python 2.7.5, and it's reproducible, just by having a 1 line
> script that has "import requests". Thanks for any help you could
> provide.
>
>
>
>
>>
>> On Jun 21, 2016 16:57, "Larry Martell" wrote:
>>>
>>> Anyone here have any experience with pyinstaller? I am trying to use
>>> it, but I'm not having much success. I tried posting to the
>>> pyinstaller ML but it said my post had to be approved first, and that
>>> hasn't happened in a while. I'll post details if someone here thinks
>>> they can help.
>>> --
Hi,
is there a direct reference to that function in your code?
On win7, python 3.5, if I use just a trivial stub source file like:
import requests
print(requests.packages.chardet.detect(b"qwe"))
and freeze it with
_path_to_\Python3\Scripts\pyinstaller.exe _path_to_\test_chardet.py
--clean --noconfirm --onedir
the resulting executable works ok, (it prints {'encoding': 'ascii',
'confidence': 1.0} just like the source version).
However, I remember, that I had problems in the past with freezing
(sub)modules or (sub)packages, that were not actually used in the
source but should be made available for interactive usage on runtime
within the app. I believe, manually referencing such objects helped in
such cases, but I can't remember the details.
Otherwise, it may be some version or platform dependent issue, of course.
regards,
vbr
--
https://mail.python.org/mailman/listinfo/python-list
Behavior of tempfile temp files when scripts killed, interpreter crashes, server crashes?
Can someone share their OS specific experience in working with tempfile generated temp files under these conditions? 1. Script killed by another process 2. Interpreter crashes 3. Server crashes (sudden loss of power) 4. Other application termination conditions ??? I'm curious which scenarios result in temp files not being automatically deleted after use and what technique you're using to cleanup temp files left over after these conditions (without affecting legitimate temp files present from the current session)? Do any OS's support a type of temp file that truly gets automatically deleted in all of the above scenarios? Thank you, Malcolm -- https://mail.python.org/mailman/listinfo/python-list
TypeError: '_TemporaryFileWrapper' object is not an iterator
I am rewriting a program so it can work in python3.
I am making progress, but now I stumble on the exception mentioned in
the subject.
The code still works in python2.
What I'm doing is to write the results to temporary file that needs to be
sorted afterwards. For as far as I understand after done writing all to
the file and seeking it to the beginning I get the following when I start
reading from it.
Traceback (most recent call last):
File "/local/home/apardon/src/fietsroutes/pymain.py", line 132, in main
Exit_Nr = process(sys.argv)
File "/local/home/apardon/src/fietsroutes/route.py", line 491, in program
make_routes(map)
File "/local/home/apardon/src/fietsroutes/route.py", line 465, in make_routes
for rt in sortfile(allroutes, load, dump):
File "/local/home/apardon/src/fietsroutes/filesort.py", line 129, in sortfile
prv = next(itr)
File "/local/home/apardon/src/fietsroutes/filesort.py", line 67, in Queuer
buf = take(21, itr)
File "/local/home/apardon/src/fietsroutes/filesort.py", line 21, in take
ls.append(next(iterable))
File "/local/home/apardon/src/fietsroutes/filesort.py", line 35, in loading
ln = load(fl)
File "/local/home/apardon/src/fietsroutes/route.py", line 366, in load
return greep(next(fl))
TypeError: '_TemporaryFileWrapper' object is not an iterator
Locals by frame, innermost last
Frame main in /local/home/apardon/src/fietsroutes/pymain.py at line 149
pn = 'route'
process =
Frame program in /local/home/apardon/src/fietsroutes/route.py at line 491
argv = ['route', '-r', '45']
index = 3
map = {'84Bh': {'83Bh': 17, '01Wo': 42, '85Bm': 37}, '70Rb':
{'18Rb': 26, '03Zu': 91, "43Gd'": 33}, "43Gd'": { ...
Frame make_routes in /local/home/apardon/src/fietsroutes/route.py at line 465
Dists = {'35HS': 204, '91Rt': 208, '79Oi': 492, "61Ch'": 347,
'78AP': 479, '33Bt': 444, '241L': 429, '54Rh': 385 ...
MinLength = 249
OldMin = 216
allroutes =
dest = '16Pl'
dist = 58
last = '86Ll'
last_time = 1469626452.5767715
map = {'84Bh': {'83Bh': 17, '01Wo': 42, '85Bm': 37}, '70Rb':
{'18Rb': 26, '03Zu': 91, "43Gd'": 33}, "43Gd'": { ...
nr = 29
nwrt = ('16Pl', '86Ll', '80Wz', '79Bs', '59Bs', '69Vs',
"70Vs'", '70Vs"', "78Th'", '78Th"', '48Th', '45Th', '01 ...
nwtot = 274
prv = '80Wz'
route = ('86Ll', '80Wz', '79Bs', '59Bs', '69Vs', "70Vs'",
'70Vs"', "78Th'", '78Th"', '48Th', '45Th', '01Th', '04 ...
start = '00Th'
total = 216
vizor = ''
Frame sortfile in /local/home/apardon/src/fietsroutes/filesort.py at line 129
fl =
in1 =
in2 =
itr =
ou1 =
ou2 =
swaps = 0
warn =
Frame Queuer in /local/home/apardon/src/fietsroutes/filesort.py at line 67
fl =
itr =
Frame take in /local/home/apardon/src/fietsroutes/filesort.py at line 23
_ = 0
iterable =
ls = []
n = 21
Frame loading in /local/home/apardon/src/fietsroutes/filesort.py at line 40
fl =
Frame load in /local/home/apardon/src/fietsroutes/route.py at line 366
fl =
--
https://mail.python.org/mailman/listinfo/python-list
python and open office
I try to create some scripts that will help me to open and manipulate OpenOffice documents. Calc in particular. But I have some problems finding right packages or libraries that offer such interface. So far I was trying uno and unotools but the first step is to import them failed. Here is the output: UNO tools are installed: $ pip list | grep uno uno (0.3.3) unotools (0.3.3) Try to import them: $ python Python 2.7.12 (default, Jun 29 2016, 12:53:15) [GCC 4.2.1 Compatible Apple LLVM 6.1.0 (clang-602.0.53)] on darwin Type "help", "copyright", "credits" or "license" for more information. >>> import uno >>> import unotools Traceback (most recent call last): File "", line 1, in File "/Volumes/home/lshanin/Dropbox/Python/ve/accounting/lib/python2.7/site-packages/unotools/__init__.py", line 16 def __init__(self, name: str): ^ SyntaxError: invalid syntax >>> I would appreciate is somebody help me to find what is wrong with untools package. Or point me to some other available libraries. I expect to work with OpenOffice (LibreOffice) files only not with MS Excel files. Regards Leonid -- https://mail.python.org/mailman/listinfo/python-list
Re: python and open office
def __init__(self, name: str): That "name: str" syntax is called function annotations, and was added in Python 3, and you are trying to use the module in Python 2.7. There may be another variation of the module compatible with Python 2, or you'll need to upgrade your Python to a version of Python 3. Chris On Wed, Jul 27, 2016 at 9:28 AM, Crane Ugly wrote: > I try to create some scripts that will help me to open and manipulate > OpenOffice documents. Calc in particular. But I have some problems finding > right packages or libraries that offer such interface. > So far I was trying uno and unotools but the first step is to import them > failed. Here is the output: > > UNO tools are installed: > $ pip list | grep uno > uno (0.3.3) > unotools (0.3.3) > > Try to import them: > $ python > Python 2.7.12 (default, Jun 29 2016, 12:53:15) > [GCC 4.2.1 Compatible Apple LLVM 6.1.0 (clang-602.0.53)] on darwin > Type "help", "copyright", "credits" or "license" for more information. > >>> import uno > >>> import unotools > Traceback (most recent call last): > File "", line 1, in > File > "/Volumes/home/lshanin/Dropbox/Python/ve/accounting/lib/python2.7/site-packages/unotools/__init__.py", > line 16 > def __init__(self, name: str): >^ > SyntaxError: invalid syntax > >>> > > I would appreciate is somebody help me to find what is wrong with untools > package. > Or point me to some other available libraries. I expect to work with > OpenOffice (LibreOffice) files only not with MS Excel files. > > Regards > Leonid > -- > https://mail.python.org/mailman/listinfo/python-list > -- https://mail.python.org/mailman/listinfo/python-list
Re: logging: getLogger() or getLogger(__name__)?
Malcolm Greene wrote: > I've read that best practice for logging is to place the following line > at the top of all modules: > > logger = getLogger(__name__) > > I'm curious why the following technique wouldn't be a better choice: > > logger = getLogger() > > Are there scenarios that favor one style over another? With __name__ you will have one logger per source file (module), with corresponding filtering possibilities, and organized hierarchically as are packages (logging use . to built its loggers hierarchy). Without __name__, you have one global default logger. > > Thank you, > Malcolm -- https://mail.python.org/mailman/listinfo/python-list
working with OpenOffice Calc
I am looking for a library that will allow me to work with Calc documents from
Python.
But so far I was not able to build properly working environment for that.
Here is what I already tried.
Installed uno and unotools for Python 2.7, but importing unotools gives an
error:
UNO tools are installed:
(accounting) $ pip list | grep uno
uno (0.3.3)
unotools (0.3.3)
Try to import them:
(accounting) $ python
Python 2.7.12 (default, Jun 29 2016, 12:53:15)
[GCC 4.2.1 Compatible Apple LLVM 6.1.0 (clang-602.0.53)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> import uno
>>> import unotools
Traceback (most recent call last):
File "", line 1, in
File
"/Python/ve/accounting/lib/python2.7/site-packages/unotools/__init__.py", line
16
def __init__(self, name: str):
^
SyntaxError: invalid syntax
>>>
I also tried to install pyoo, but it is also failing on import:
(accounting) $ pip list |grep pyoo
pyoo (1.1)
(accounting) $ python
Python 2.7.12 (default, Jun 29 2016, 12:53:15)
[GCC 4.2.1 Compatible Apple LLVM 6.1.0 (clang-602.0.53)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> import pyoo
Traceback (most recent call last):
File "", line 1, in
File "/Python/ve/accounting/lib/python2.7/site-packages/pyoo.py", line 26, in
FORMAT_TEXT =
uno.getConstantByName('com.sun.star.i18n.NumberFormatIndex.TEXT')
AttributeError: 'module' object has no attribute 'getConstantByName'
>>>
Could someone point me to the right direction to solve the issue.
Regards
Leonid
--
https://mail.python.org/mailman/listinfo/python-list
Re: python and open office
On 7/27/2016 12:37 PM, Chris Kaynor wrote: def __init__(self, name: str): That "name: str" syntax is called function annotations, and was added in Python 3, and you are trying to use the module in Python 2.7. There may be another variation of the module compatible with Python 2, or you'll need to upgrade your Python to a version of Python 3. Chris On Wed, Jul 27, 2016 at 9:28 AM, Crane Ugly wrote: I try to create some scripts that will help me to open and manipulate OpenOffice documents. Calc in particular. But I have some problems finding right packages or libraries that offer such interface. So far I was trying uno and unotools but the first step is to import them failed. Here is the output: UNO tools are installed: $ pip list | grep uno uno (0.3.3) unotools (0.3.3) Try to import them: $ python Python 2.7.12 (default, Jun 29 2016, 12:53:15) [GCC 4.2.1 Compatible Apple LLVM 6.1.0 (clang-602.0.53)] on darwin Type "help", "copyright", "credits" or "license" for more information. import uno import unotools Traceback (most recent call last): File "", line 1, in File "/Volumes/home/lshanin/Dropbox/Python/ve/accounting/lib/python2.7/site-packages/unotools/__init__.py", line 16 def __init__(self, name: str): ^ SyntaxError: invalid syntax I would appreciate is somebody help me to find what is wrong with untools package. Or point me to some other available libraries. I expect to work with OpenOffice (LibreOffice) files only not with MS Excel files. Are you working with OpenOffice or LibreOffice? There are *different programs*. Last I know, current LibreOffice comes with python 3.3.3 in its program directory and you need at least Python 3.3.3 for its UNO bridge, as it used the FSR unicode representation introduced in 3.3. -- Terry Jan Reedy -- https://mail.python.org/mailman/listinfo/python-list
Re: working with OpenOffice Calc
On 7/27/2016 1:54 PM, id23...@gmail.com wrote:
I am looking for a library that will allow me to work with Calc documents from
Python.
But so far I was not able to build properly working environment for that.
You posted this same question 1 1/2 hours before under a different name.
Please don't repost, especially when a correct answer was posted 9
minutes after the first.
I also tried to install pyoo, but it is also failing on import:
A different question, but the answer is still the same. Use compatible
versions of python and 3rd party packages.
(accounting) $ pip list |grep pyoo
pyoo (1.1)
(accounting) $ python
Python 2.7.12 (default, Jun 29 2016, 12:53:15)
[GCC 4.2.1 Compatible Apple LLVM 6.1.0 (clang-602.0.53)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
import pyoo
Traceback (most recent call last):
File "", line 1, in
File "/Python/ve/accounting/lib/python2.7/site-packages/pyoo.py", line 26, in
FORMAT_TEXT =
uno.getConstantByName('com.sun.star.i18n.NumberFormatIndex.TEXT')
AttributeError: 'module' object has no attribute 'getConstantByName'
Could someone point me to the right direction to solve the issue.
Regards
Leonid
--
Terry Jan Reedy
--
https://mail.python.org/mailman/listinfo/python-list
Re: python and open office
I use LibreOffice. Yes, it looks like I have to switch to Python 3 in this case. I'll try v.3 then Thank you Leonid > On 27-07-2016, at 21:01, Terry Reedy wrote: > > On 7/27/2016 12:37 PM, Chris Kaynor wrote: >> def __init__(self, name: str): >> >> That "name: str" syntax is called function annotations, and was added in >> Python 3, and you are trying to use the module in Python 2.7. >> >> There may be another variation of the module compatible with Python 2, or >> you'll need to upgrade your Python to a version of Python 3. >> >> Chris >> >> On Wed, Jul 27, 2016 at 9:28 AM, Crane Ugly wrote: >> >>> I try to create some scripts that will help me to open and manipulate >>> OpenOffice documents. Calc in particular. But I have some problems finding >>> right packages or libraries that offer such interface. >>> So far I was trying uno and unotools but the first step is to import them >>> failed. Here is the output: >>> >>> UNO tools are installed: >>> $ pip list | grep uno >>> uno (0.3.3) >>> unotools (0.3.3) >>> >>> Try to import them: >>> $ python >>> Python 2.7.12 (default, Jun 29 2016, 12:53:15) >>> [GCC 4.2.1 Compatible Apple LLVM 6.1.0 (clang-602.0.53)] on darwin >>> Type "help", "copyright", "credits" or "license" for more information. >> import uno >> import unotools >>> Traceback (most recent call last): >>> File "", line 1, in >>> File >>> "/Volumes/home/lshanin/Dropbox/Python/ve/accounting/lib/python2.7/site-packages/unotools/__init__.py", >>> line 16 >>>def __init__(self, name: str): >>> ^ >>> SyntaxError: invalid syntax >> >>> >>> I would appreciate is somebody help me to find what is wrong with untools >>> package. >>> Or point me to some other available libraries. I expect to work with >>> OpenOffice (LibreOffice) files only not with MS Excel files. > > Are you working with OpenOffice or LibreOffice? There are *different > programs*. Last I know, current LibreOffice comes with python 3.3.3 in its > program directory and you need at least Python 3.3.3 for its UNO bridge, as > it used the FSR unicode representation introduced in 3.3. > > -- > Terry Jan Reedy > > -- > https://mail.python.org/mailman/listinfo/python-list -- https://mail.python.org/mailman/listinfo/python-list
Is it possible to draw a BUTTON?
Hi Folks, It is common to put a BUTTON on a canvas by the means of coding. However, in my application, I need to draw a circle on canvas, and then make this circle to work as if it is a button. When the circle is clicked, it triggers a new image to be displayed. Somebody can help? Thanks! -- https://mail.python.org/mailman/listinfo/python-list
ImportError: Import by filename is not supported when unpickleing
When I try and unpickle an object with pickle.loads it fails with: ImportError: Import by filename is not supported when unpickleing I've never used pickle before. Why do I get this and how can I fix it? -- https://mail.python.org/mailman/listinfo/python-list
Re: pyinstaller
On Wed, Jul 27, 2016 at 9:12 AM, Vlastimil Brom
wrote:
> 2016-07-27 3:15 GMT+02:00 Larry Martell :
>> On Tue, Jul 26, 2016 at 8:49 PM, Tom Brown wrote:
>>> I used pyinstaller quite a bit 3 years ago. I could brush off the cobwebs
>>> and see if I can help if you have not solved it already.
>>>
>>> What is the issue you are having?
>>
>> If I import the requests module, then when I run the executable I get:
>>
>> ImportError: No module named 'requests.packages.chardet'
>>
>> I tried to post to the pyinstaller group, but it said my post had to
>> be approved by the moderator, and it apparently never was. I have no
>> idea who the moderator is, so there was no one I could contact about
>> that. I posted an issue to github
>> (https://github.com/pyinstaller/pyinstaller/issues/2060) and some
>> suggestions were made, but none fixed the problem. I am on RHEL 7.2
>> with Python 2.7.5, and it's reproducible, just by having a 1 line
>> script that has "import requests". Thanks for any help you could
>> provide.
>>
>>
>>
>>
>>>
>>> On Jun 21, 2016 16:57, "Larry Martell" wrote:
Anyone here have any experience with pyinstaller? I am trying to use
it, but I'm not having much success. I tried posting to the
pyinstaller ML but it said my post had to be approved first, and that
hasn't happened in a while. I'll post details if someone here thinks
they can help.
--
>
> Hi,
> is there a direct reference to that function in your code?
No there is no reference to that function. As I said, for testing I
have just a single line script with
import requests
>
> On win7, python 3.5, if I use just a trivial stub source file like:
>
> import requests
> print(requests.packages.chardet.detect(b"qwe"))
>
> and freeze it with
> _path_to_\Python3\Scripts\pyinstaller.exe _path_to_\test_chardet.py
> --clean --noconfirm --onedir
>
> the resulting executable works ok, (it prints {'encoding': 'ascii',
> 'confidence': 1.0} just like the source version).
Yes, people have reported on git that it works on Windows.
> However, I remember, that I had problems in the past with freezing
> (sub)modules or (sub)packages, that were not actually used in the
> source but should be made available for interactive usage on runtime
> within the app. I believe, manually referencing such objects helped in
> such cases, but I can't remember the details.
I will try that.
> Otherwise, it may be some version or platform dependent issue, of course.
--
https://mail.python.org/mailman/listinfo/python-list
Re: ImportError: Import by filename is not supported when unpickleing
On Wed, 27 Jul 2016 17:25:43 -0400, Larry Martell wrote: > When I try and unpickle an object with pickle.loads it fails with: > > ImportError: Import by filename is not supported when unpickleing > > I've never used pickle before. Why do I get this and how can I fix it? Try using *pickle.load* instead of *pickle.loads*. pickle.loads is for strings. Retrieved from documentation: help(pickle) after importing pickle. -- [The Computer] was the first machine man built that assisted the power of his brain instead of the strength of his arm. - Grace Hopper -- https://mail.python.org/mailman/listinfo/python-list
Re: Is it possible to draw a BUTTON?
On Wed, 27 Jul 2016 13:18:16 -0700, huey.y.jiang wrote: > Hi Folks, > > It is common to put a BUTTON on a canvas by the means of coding. > However, in my application, I need to draw a circle on canvas, and then > make this circle to work as if it is a button. When the circle is > clicked, it triggers a new image to be displayed. Somebody can help? > Thanks! It depends on what GUI toolkit you use. Can you give more specific information? Since a canvas is refreshed (updated) every time the function for redraw is called, it is as simple as changing the variable for what should be drawn when a click event is sent to the canvas and the area hot-spot is the same as the area of the button. Buffered drawing is similar. -- [The Computer] was the first machine man built that assisted the power of his brain instead of the strength of his arm. - Grace Hopper -- https://mail.python.org/mailman/listinfo/python-list
ImportError: Import by filename is not supported when unpickleing
On Wednesday, July 27, 2016, Jason Benjamin > wrote: > On Wed, 27 Jul 2016 17:25:43 -0400, Larry Martell wrote: > > > When I try and unpickle an object with pickle.loads it fails with: > > > > ImportError: Import by filename is not supported when unpickleing > > > > I've never used pickle before. Why do I get this and how can I fix it? > > Try using *pickle.load* instead of *pickle.loads*. pickle.loads is for > strings. Retrieved from documentation: help(pickle) after importing > pickle. > I am unpickling a string created with pickle.dumps -- https://mail.python.org/mailman/listinfo/python-list
Re: Is it possible to draw a BUTTON?
On Wednesday, July 27, 2016 at 4:18:29 PM UTC-4, huey.y...@gmail.com wrote: > Hi Folks, > > It is common to put a BUTTON on a canvas by the means of coding. However, in > my application, I need to draw a circle on canvas, and then make this circle > to work as if it is a button. When the circle is clicked, it triggers a new > image to be displayed. Somebody can help? Thanks! ---> By the way, the GUI is TK. -- https://mail.python.org/mailman/listinfo/python-list
Re: ImportError: Import by filename is not supported when unpickleing
On Wed, Jul 27, 2016 at 6:54 PM, Jason Benjamin wrote: > If it has and 's' on the end it will only work on strings. *dumps* refers > to a string too. Yes, I know. I have an object, which I pickle with dumps, which turns it into a string. Then I try to unpickle it with loads and I get that error. (Pdb) type(args.target) (Pdb) pickle.loads(args.target) *** ImportError: Import by filename is not supported. > On Wed, Jul 27, 2016 at 3:44 PM, Larry Martell > wrote: > > > > On Wednesday, July 27, 2016, Jason Benjamin wrote: >> >> On Wed, 27 Jul 2016 17:25:43 -0400, Larry Martell wrote: >> >> > When I try and unpickle an object with pickle.loads it fails with: >> > >> > ImportError: Import by filename is not supported when unpickleing >> > >> > I've never used pickle before. Why do I get this and how can I fix it? >> >> Try using *pickle.load* instead of *pickle.loads*. pickle.loads is for >> strings. Retrieved from documentation: help(pickle) after importing >> pickle. > > > I am unpickling a string created with pickle.dumps -- https://mail.python.org/mailman/listinfo/python-list
Re: Is it possible to draw a BUTTON?
On 2016-07-28 00:13, huey.y.ji...@gmail.com wrote:
On Wednesday, July 27, 2016 at 4:18:29 PM UTC-4, huey.y...@gmail.com wrote:
Hi Folks,
It is common to put a BUTTON on a canvas by the means of coding. However, in my
application, I need to draw a circle on canvas, and then make this circle to
work as if it is a button. When the circle is clicked, it triggers a new image
to be displayed. Somebody can help? Thanks!
---> By the way, the GUI is TK.
Here's a simple example:
#! python3.5
# -*- coding: utf-8 -*-
import tkinter as tk
def mouse_clicked(event):
dist_sq = (event.x - circle['x']) ** 2 + (event.y - circle['y']) ** 2
if dist_sq <= circle['radius'] ** 2:
print('Clicked inside circle')
root = tk.Tk()
canvas = tk.Canvas(root, width=400, height=200)
canvas.pack()
circle = dict(x=60, y=60, radius=20)
left = circle['x'] - circle['radius']
top = circle['y'] - circle['radius']
right = circle['x'] + circle['radius']
bottom = circle['y'] + circle['radius']
canvas.create_oval((left, top, right, bottom), outline='red', fill='red')
canvas.bind('', mouse_clicked)
root.mainloop()
--
https://mail.python.org/mailman/listinfo/python-list
Re: Is it possible to draw a BUTTON?
On Wednesday, July 27, 2016 at 7:15:20 PM UTC-5, MRAB wrote:
> On 2016-07-28 00:13, huey.y.ji...@gmail.com wrote:
> > On Wednesday, July 27, 2016 at 4:18:29 PM UTC-4, huey.y...@gmail.com wrote:
> >> Hi Folks,
> >>
> >> It is common to put a BUTTON on a canvas by the means of coding. However,
> >> in my application, I need to draw a circle on canvas, and then make this
> >> circle to work as if it is a button. When the circle is clicked, it
> >> triggers a new image to be displayed. Somebody can help? Thanks!
> >
> > ---> By the way, the GUI is TK.
> >
> Here's a simple example:
>
>
> #! python3.5
> # -*- coding: utf-8 -*-
> import tkinter as tk
>
> def mouse_clicked(event):
> dist_sq = (event.x - circle['x']) ** 2 + (event.y - circle['y']) ** 2
>
> if dist_sq <= circle['radius'] ** 2:
> print('Clicked inside circle')
>
> root = tk.Tk()
>
> canvas = tk.Canvas(root, width=400, height=200)
> canvas.pack()
>
> circle = dict(x=60, y=60, radius=20)
>
> left = circle['x'] - circle['radius']
> top = circle['y'] - circle['radius']
> right = circle['x'] + circle['radius']
> bottom = circle['y'] + circle['radius']
>
> canvas.create_oval((left, top, right, bottom), outline='red', fill='red')
> canvas.bind('', mouse_clicked)
>
> root.mainloop()
I didn't try your code, but you can simply it by using some of the core
functionality provided via "canvas.tag_bind(...)"
## START CODE ##
import Tkinter as tk
def cb_canvasButton(event):
print 'You clicked me using button {0}'.format(event.num)
canvas.move('button', 10, 10) # Little extra surprise!
root = tk.Tk()
canvas = tk.Canvas(root)
canvas.create_oval(10,10,50,50, outline='red', fill='blue',
tags=('button',)
)
canvas.tag_bind('button', '', cb_canvasButton)
canvas.pack()
root.mainloop()
## END CODE ##
--
https://mail.python.org/mailman/listinfo/python-list
Re: Is it possible to draw a BUTTON?
On Wednesday, July 27, 2016 at 4:18:29 PM UTC-4, huey.y...@gmail.com wrote: > Hi Folks, > > It is common to put a BUTTON on a canvas by the means of coding. However, in > my application, I need to draw a circle on canvas, and then make this circle > to work as if it is a button. When the circle is clicked, it triggers a new > image to be displayed. Somebody can help? Thanks! The example works. It is smart to bind a button to a widget, I'v learned. Thanks so much! -- https://mail.python.org/mailman/listinfo/python-list
Re: ImportError: Import by filename is not supported when unpickleing
On Wed, Jul 27, 2016 at 7:45 PM, Jason Benjamin wrote: > Look at this: https://wiki.python.org/moin/UsingPickle > It uses *pickle.dump* not *pickle.dumps* Yes that uses a file. I do not want to use a file. I want to pass the object as a string. > If you still don't get it send me the code for the function with a working > code example the calls that function revealing the error. > > I've had a lot of experience with Python and it looks like its the sequence > of functions that causes the problem and not just one Python function call > on its own. I have an object of type Target: (Pdb) type(target) And I pickle it like this: (Pdb) type(pickle.dumps(target)) And then it looks like this: (Pdb) pickle.dumps(target) "ccopy_reg\n_reconstructor\np0\n(cworkitem\nTarget\np1\nc__builtin__\nobject\np2\nNtp3\nRp4\n(dp5\nS'histologySections'\np6\n(lp7\nsS'registrationTransforms'\np8\nV\np9\nsS'valueMaps'\np10\n(dp11\nsS'targetID'\np12\nVRight-CarotidArtery\np13\nsS'targetInitializer'\np14\ng9\nsS'regions'\np15\n(dp16\nsS'bodySite'\np17\nVRightCarotid\np18\nsS'targetLocalFolderName'\np19\nV21597135/wi-54976537/Right-CarotidArtery\np20\nsS'readingsLocalFileName'\np21\ng9\nsS'probabilityMaps'\np22\n(dp23\nsS'targetPath'\np24\ng9\nsb." And I pass it into a subprocess.Popen call. Then in the program that is called, it comes in as a string: (Pdb) type(args.target) With the same content: (Pdb) args.target "ccopy_reg\n_reconstructor\np0\n(cworkitem\nTarget\np1\nc__builtin__\nobject\np2\nNtp3\nRp4\n(dp5\nS'histologySections'\np6\n(lp7\nsS'registrationTransforms'\np8\nV\np9\nsS'valueMaps'\np10\n(dp11\nsS'targetID'\np12\nVRight-CarotidArtery\np13\nsS'targetInitializer'\np14\ng9\nsS'regions'\np15\n(dp16\nsS'bodySite'\np17\nVRightCarotid\np18\nsS'targetLocalFolderName'\np19\nV21597135/wi-54976537/Right-CarotidArtery\np20\nsS'readingsLocalFileName'\np21\ng9\nsS'probabilityMaps'\np22\n(dp23\nsS'targetPath'\np24\ng9\nsb." And when I unpickle it I get the error: (Pdb) pickle.loads(args.target) ***ImportError: Import by filename is not supported. Also, please keep the discussion on the list, and please don't top post. > > > > On Wed, Jul 27, 2016 at 4:29 PM, Larry Martell > wrote: > > On Wed, Jul 27, 2016 at 6:54 PM, Jason Benjamin > wrote: > > If it has and 's' on the end it will only work on strings. *dumps* refers to > a string too. > > Yes, I know. I have an object, which I pickle with dumps, which turns it > into a string. Then I try to unpickle it with loads and I get that error. > (Pdb) type(args.target) (Pdb) pickle.loads(args.target) *** > ImportError: Import by filename is not supported. > > On Wed, Jul 27, 2016 at 3:44 PM, Larry Martell > wrote: On Wednesday, July 27, 2016, Jason Benjamin > wrote: > > On Wed, 27 Jul 2016 17:25:43 -0400, Larry Martell wrote: > When I try and > unpickle an object with pickle.loads it fails with: > > ImportError: Import > by filename is not supported when unpickleing > > I've never used pickle > before. Why do I get this and how can I fix it? Try using *pickle.load* > instead of *pickle.loads*. pickle.loads is for strings. Retrieved from > documentation: help(pickle) after importing pickle. > > I am unpickling a string created with pickle.dumps -- https://mail.python.org/mailman/listinfo/python-list
Re: ImportError: Import by filename is not supported when unpickleing
On Wed, Jul 27, 2016 at 10:39 PM, Larry Martell wrote: > On Wed, Jul 27, 2016 at 7:45 PM, Jason Benjamin wrote: >> Look at this: https://wiki.python.org/moin/UsingPickle >> It uses *pickle.dump* not *pickle.dumps* > > Yes that uses a file. I do not want to use a file. I want to pass the > object as a string. > >> If you still don't get it send me the code for the function with a working >> code example the calls that function revealing the error. >> >> I've had a lot of experience with Python and it looks like its the sequence >> of functions that causes the problem and not just one Python function call >> on its own. > > I have an object of type Target: > > (Pdb) type(target) > > > And I pickle it like this: > > (Pdb) type(pickle.dumps(target)) > > > And then it looks like this: > > (Pdb) pickle.dumps(target) > "ccopy_reg\n_reconstructor\np0\n(cworkitem\nTarget\np1\nc__builtin__\nobject\np2\nNtp3\nRp4\n(dp5\nS'histologySections'\np6\n(lp7\nsS'registrationTransforms'\np8\nV\np9\nsS'valueMaps'\np10\n(dp11\nsS'targetID'\np12\nVRight-CarotidArtery\np13\nsS'targetInitializer'\np14\ng9\nsS'regions'\np15\n(dp16\nsS'bodySite'\np17\nVRightCarotid\np18\nsS'targetLocalFolderName'\np19\nV21597135/wi-54976537/Right-CarotidArtery\np20\nsS'readingsLocalFileName'\np21\ng9\nsS'probabilityMaps'\np22\n(dp23\nsS'targetPath'\np24\ng9\nsb." > > And I pass it into a subprocess.Popen call. Then in the program that > is called, it comes in as a string: > > (Pdb) type(args.target) > > > With the same content: > > (Pdb) args.target > "ccopy_reg\n_reconstructor\np0\n(cworkitem\nTarget\np1\nc__builtin__\nobject\np2\nNtp3\nRp4\n(dp5\nS'histologySections'\np6\n(lp7\nsS'registrationTransforms'\np8\nV\np9\nsS'valueMaps'\np10\n(dp11\nsS'targetID'\np12\nVRight-CarotidArtery\np13\nsS'targetInitializer'\np14\ng9\nsS'regions'\np15\n(dp16\nsS'bodySite'\np17\nVRightCarotid\np18\nsS'targetLocalFolderName'\np19\nV21597135/wi-54976537/Right-CarotidArtery\np20\nsS'readingsLocalFileName'\np21\ng9\nsS'probabilityMaps'\np22\n(dp23\nsS'targetPath'\np24\ng9\nsb." > > And when I unpickle it I get the error: > > (Pdb) pickle.loads(args.target) > ***ImportError: Import by filename is not supported. > > > Also, please keep the discussion on the list, and please don't top post. Also let me add that initially I was calling Popen with shell=False and the arguments in a list, and that was failing with: arg 2 must contain only strings And when I debugged I found that this was happening in the execvp call in subprocess and arg 2 was a str. So then I changed it using shell=False with the command line in a string, and that is when I get the ImportError error. >> On Wed, Jul 27, 2016 at 4:29 PM, Larry Martell >> wrote: >> >> On Wed, Jul 27, 2016 at 6:54 PM, Jason Benjamin >> wrote: >> >> If it has and 's' on the end it will only work on strings. *dumps* refers to >> a string too. >> >> Yes, I know. I have an object, which I pickle with dumps, which turns it >> into a string. Then I try to unpickle it with loads and I get that error. >> (Pdb) type(args.target) (Pdb) pickle.loads(args.target) *** >> ImportError: Import by filename is not supported. >> >> On Wed, Jul 27, 2016 at 3:44 PM, Larry Martell >> wrote: On Wednesday, July 27, 2016, Jason Benjamin >> wrote: >> >> On Wed, 27 Jul 2016 17:25:43 -0400, Larry Martell wrote: > When I try and >> unpickle an object with pickle.loads it fails with: > > ImportError: Import >> by filename is not supported when unpickleing > > I've never used pickle >> before. Why do I get this and how can I fix it? Try using *pickle.load* >> instead of *pickle.loads*. pickle.loads is for strings. Retrieved from >> documentation: help(pickle) after importing pickle. >> >> I am unpickling a string created with pickle.dumps -- https://mail.python.org/mailman/listinfo/python-list
Python text file fetch specific part of line
I am writing Imdb scrapper, and getting available list of titles from IMDB website which provide txt file in very raw format, Here is the one part of file(http://pastebin.com/fpMgBAjc) as the file provides tags like Distribution Votes,Rank,Title I want to parse title names, I tried with readlines() method but it returns only list which is quite heterogeneous, is it possible that I can parse each value comes under title section? -- https://mail.python.org/mailman/listinfo/python-list
Re: Behavior of tempfile temp files when scripts killed, interpreter crashes, server crashes?
On Wed, Jul 27, 2016 at 1:58 PM, Malcolm Greene wrote:
> Can someone share their OS specific experience in working with tempfile
> generated temp files under these conditions?
>
> 1. Script killed by another process
> 2. Interpreter crashes
> 3. Server crashes (sudden loss of power)
> 4. Other application termination conditions ???
>
> I'm curious which scenarios result in temp files not being automatically
> deleted after use and what technique you're using to cleanup temp files
> left over after these conditions (without affecting legitimate temp
> files present from the current session)?
>
> Do any OS's support a type of temp file that truly gets automatically
> deleted in all of the above scenarios?
Linux has the O_TMPFILE open() flag [1]. This creates an anonymous
file that gets automatically deleted when the last open file
descriptor is closed. If the file isn't opened O_EXCL, then you can
make it permanent by linking it back into the filesystem. For example:
>>> fd = os.open('/tmp', os.O_TMPFILE | os.O_RDWR, 0o600)
>>> tmp_dir_fd = os.open('/tmp', os.O_DIRECTORY)
>>> os.write(fd, b'spam')
4
>>> os.link('/proc/self/fd/%d' % fd, 'tempfile', dst_dir_fd=tmp_dir_fd)
>>> os.close(fd); os.close(tmp_dir_fd)
>>> open('/tmp/tempfile').read()
'spam'
tempfile.TemporaryFile uses the O_TMPFILE flag when it's available.
Since it also uses O_EXCL, the files it creates cannot be made
permanent via os.link.
[1]: http://man7.org/linux/man-pages/man2/open.2.html
Windows CreateFile [2] has two related flags:
FILE_FLAG_DELETE_ON_CLOSE to ensure that a temporary file gets deleted
and FILE_ATTRIBUTE_TEMPORARY, which avoids writing the file to disk as
long as there's enough available cache memory. The Windows C runtime
and Python make these available as O_TEMPORARY and O_SHORT_LIVED,
respectively [3].
tempfile.NamedTemporaryFile uses O_TEMPORARY. tempfile.TemporaryFile
is an alias for NamedTemporary, since files can't be anonymous on
Windows.
You can delete the temporary file, because it's opened with delete
sharing, but it won't actually get unlinked from the directory until
the last handle is closed. The only benefit to deleting the file is
getting exclusive access, but you can already control that in Windows
via the dwShareMode parameter of CreateFile or the shflag parameter of
_wsopen_s. Unfortunately Python doesn't provide a high-level API for
setting the sharing mode, so if you need to set it you're stuck with
using _winapi.CreateFile and msvcrt.open_osfhandle, PyWin32, or
ctypes.
It's possible to work around the delete-on-close flag by opening the
file a 2nd time, closing the original handle, and then calling
SetFileInformationByHandle [4] to unset the file's delete disposition.
[2]: https://msdn.microsoft.com/en-us/library/aa363858
[3]: https://msdn.microsoft.com/en-us/library/w64k0ytk
[4]: https://msdn.microsoft.com/en-us/library/aa365539
--
https://mail.python.org/mailman/listinfo/python-list
Re: ImportError: Import by filename is not supported when unpickleing
On Thursday 28 July 2016 12:39, Larry Martell wrote: > I have an object of type Target: > > (Pdb) type(target) > > > And I pickle it like this: > > (Pdb) type(pickle.dumps(target)) > > > And then it looks like this: > > (Pdb) pickle.dumps(target) > "ccopy_reg\n_reconstructor\np0\n(cworkitem\nTarget\np1\nc__builtin__\nobject\np2\nNtp3\nRp4\n(dp5\nS'histologySections'\np6\n(lp7\nsS'registrationTransforms'\np8\nV\np9\nsS'valueMaps'\np10\n(dp11\nsS'targetID'\np12\nVRight- CarotidArtery\np13\nsS'targetInitializer'\np14\ng9\nsS'regions'\np15\n(dp16\nsS'bodySite'\np17\nVRightCarotid\np18\nsS'targetLocalFolderName'\np19\nV21597135/wi-54976537/Right- CarotidArtery\np20\nsS'readingsLocalFileName'\np21\ng9\nsS'probabilityMaps'\np22\n(dp23\nsS'targetPath'\np24\ng9\nsb." Gag me with a spoon! What happens if you do s = pickle.dumps(target) pickle.loads(s) Do you get the same error? > And I pass it into a subprocess.Popen call. Then in the program that > is called, it comes in as a string: > > (Pdb) type(args.target) > > > With the same content: > > (Pdb) args.target > "ccopy_reg\n_reconstructor\np0\n(cworkitem\nTarget\np1\nc__builtin__\nobject\np2\nNtp3\nRp4\n(dp5\nS'histologySections'\np6\n(lp7\nsS'registrationTransforms'\np8\nV\np9\nsS'valueMaps'\np10\n(dp11\nsS'targetID'\np12\nVRight- CarotidArtery\np13\nsS'targetInitializer'\np14\ng9\nsS'regions'\np15\n(dp16\nsS'bodySite'\np17\nVRightCarotid\np18\nsS'targetLocalFolderName'\np19\nV21597135/wi-54976537/Right- CarotidArtery\np20\nsS'readingsLocalFileName'\np21\ng9\nsS'probabilityMaps'\np22\n(dp23\nsS'targetPath'\np24\ng9\nsb." Are you sure it's the same content, and not just the same to the naked eye? > And when I unpickle it I get the error: > > (Pdb) pickle.loads(args.target) > ***ImportError: Import by filename is not supported. Does the target process already have workitem imported? What happens if you import workitem first? By the way, just in case this is relevant to you... pickle is an insecure format. If your target process is running in a context where it can receive pickles from untrusted clients, you're vulnerable to them running arbitrary code on your machine. If you are trying to do a remote procedure call, rather than invent your own, you should use a reliable, trusted library. I've used both rpyc and Pyro and can recommend them both: https://pypi.python.org/pypi/rpyc/ https://pypi.python.org/pypi/Pyro4/ -- Steve -- https://mail.python.org/mailman/listinfo/python-list
