xlwt: version 0.7.1 is now available from PyPI
John Machin
Go to http://pypi.python.org/pypi/xlwt
There's a Windows installer and a .zip source distribution available at
the moment -- a .tar.gz is undergoing quality control.
Please post any problem reports here.
Cheers,
John
Chris Withers
>
> Go to http://pypi.python.org/pypi/xlwt
>
> There's a Windows installer and a .zip source distribution available at
> the moment -- a .tar.gz is undergoing quality control.
And there was much rejoicing! ;-)
Just xlrd to go!
Chris
--
Simplistix - Content Management, Zope & Python Consulting
- http://www.simplistix.co.uk
Michael Schurter
>
> John Machin wrote:
>>
>> Go to http://pypi.python.org/pypi/xlwt
>>
>> There's a Windows installer and a .zip source distribution available at
>> the moment -- a .tar.gz is undergoing quality control.
>
> And there was much rejoicing! ;-)
>
> Just xlrd to go!
Is there a changelog for either? I couldn't find one.
Thanks to all the contributors!
Michael
John Machin
xlrd: "Just xlrd to go!" means the xlrd release has not yet been done.
What were you looking at?
xlwt: There's a file called HISTORY.html but it's only a stub. Below is
a dump of the svn log since the previous release (minus boring details
like which source files were modified, plus the bit I left out about the
audit of built-in functions).
Cheers,
John
8<---
Revision: 3789
Author: chris
Date: 9:46:00 PM, Wednesday, 4 March 2009
Message:
tagging 0.7.1
----
Added : /xlwt/tags/0.7.1 (Copy from path: /xlwt/trunk, Revision, 3788)
Revision: 3788
Author: sjmachin
Date: 12:11:22 AM, Wednesday, 4 March 2009
Message:
Tidy-up and version changes for release 0.7.1
----
Revision: 3787
Author: sjmachin
Date: 2:18:22 PM, Sunday, 1 March 2009
Message:
Generalised the formula references mechanism.
Added support for external functions in formulas.
Added definitions for all the Analysis ToolPak external functions.
Fixed bug in backpatching references in args of CHOOSE built-in function.
Audit of built-in functions: several fixes of name, function number,
minimum number of arguments.
----
Revision: 3750
Author: sjmachin
Date: 11:13:04 AM, Friday, 13 February 2009
Message:
Formulas: Added inter-sheet references, IF() and CHOOSE() functions
regularised, minor bug fixes and enhancements.
Worksheet names: Added checking for invalid and duplicate
(case-insensitive) worksheet names.
----
Revision: 3734
Author: sjmachin
Date: 10:21:42 AM, Tuesday, 27 January 2009
Message:
Fixed bug: If caller does (write some cells, save) more than once on the
same Workbook object, contents of text cells of second and subsequent
files are scrambled. Bug introduced in revision 3715.
----
Revision: 3715
Author: sjmachin
Date: 5:36:01 PM, Monday, 22 December 2008
Message:
Fixed problem with string cell overwriting. When all cells containing a
given string are overwritten, there will no longer be a copy of that
string in the shared string table. Added code to write MULBLANK or MULRK
records instead of multiple adjacent BLANK or RK records.
----
Revision: 3714
Author: sjmachin
Date: 10:48:02 PM, Saturday, 20 December 2008
Message:
Addressed "Overwriting cells" problem: Avoid creating files with more
than one cell record for the same cell. Default is to raise an
exception; override (Workbook.add_sheet(..., cell_overwrite_ok=True))
specifies that the most recent write prevails.
----
Revision: 3633
Author: sjmachin
Date: 9:25:03 AM, Tuesday, 2 December 2008
Message:
Wrong file committed
----
Modified : /xlwt/trunk/xlwt/examples/image_chg_col_wid.py
Revision: 3632
Author: sjmachin
Date: 3:51:31 PM, Monday, 1 December 2008
Message:
Inserting an image file was getting a constant value for column width in
pixels.
Setting a worksheet's zoom/magnification percentage wasn't working.
----
Added : /xlwt/trunk/xlwt/examples/image_chg_col_wid.py
Added : /xlwt/trunk/xlwt/examples/zoom_magnification.py
Revision: 3541
Author: sjmachin
Date: 6:23:27 PM, Tuesday, 14 October 2008
Message:
Added enhanced version of John Harrison's patch for flushing rows to
save memory.
----
Daniel Burke
xlwt 0.7.1 fails installing on win32 python 2.6 at the part where it
compiles the library into pyo files. Specifically....
-----
Runtime Error!
Program: C:\Documents (path of downloaded installer)
R6304
An application has made an attempt to load the C runtime library
incorrectly.
Please contact the applications support team for more information.
----
This error is produced 4 times, once each for compiling the pyc and
pyo of examples/wsprops.py and examples/unicode2.py.
xlwt works fine, and both examples run fine too.
Thinking this was the not unheard of "module built with py2.5 fails
when installing on py2.6" I tried building it with python 2.6 myself
with setup.py bdist_wininst.
The build worked fine, and the resulting executable happily installed
xlwt to py2.5, but py2.6 still fails with the same error.
This looks like a case of mixed compiler versions to me, but I make a
point of not compiling python things. Although having said that I do
have mingw32, cygwin, and vc 2k8 express installed :-/
Any ideas? Can anyone else replicate this on their machines?
On Mar 5, 8:14 am, John Machin <sjmac...@lexicon.net> wrote:
> On 5/03/2009 4:34 AM, Michael Schurter wrote:
>
> > On Wed, Mar 4, 2009 at 8:09 AM, Chris Withers <ch...@simplistix.co.uk> wrote:
> >> John Machin wrote:
Chris Withers
> xlwt 0.7.1 fails installing on win32 python 2.6 at the part where it
> compiles the library into pyo files. Specifically....
What distribution were you using, specifically?
> -----
> Runtime Error!
>
> Program: C:\Documents (path of downloaded installer)
>
> R6304
> An application has made an attempt to load the C runtime library
> incorrectly.
> Please contact the applications support team for more information.
> ----
> This error is produced 4 times, once each for compiling the pyc and
> pyo of examples/wsprops.py and examples/unicode2.py.
How did you install? Can we have the full and unadulterated error
messages please?
> Thinking this was the not unheard of "module built with py2.5 fails
> when installing on py2.6" I tried building it with python 2.6 myself
> with setup.py bdist_wininst.
> The build worked fine, and the resulting executable happily installed
> xlwt to py2.5, but py2.6 still fails with the same error.
This sounds like you're using the windows installer, but I just tried
that with 2.6 here, no problems...
> This looks like a case of mixed compiler versions to me, but I make a
> point of not compiling python things. Although having said that I do
> have mingw32, cygwin, and vc 2k8 express installed :-/
I don't see how this can be a compiler problem, since xlwt contains
nothing but pure python code.
You haven't, by chance, compiled your Python 2.6 yourself?
cheers,
John Machin
Hi Daniel,
When you want to report a problem, please start a new thread.
> I suspect this is a problem with my configuration, but....
>
> xlwt 0.7.1 fails installing on win32 python 2.6 at the part where it
> compiles the library into pyo files. Specifically....
> -----
> Runtime Error!
>
> Program: C:\Documents (path of downloaded installer)
>
> R6304
> An application has made an attempt to load the C runtime library
> incorrectly.
> Please contact the applications support team for more information.
> ----
> This error is produced 4 times, once each for compiling the pyc and
> pyo of examples/wsprops.py and examples/unicode2.py.
> xlwt works fine, and both examples run fine too.
Let me see if I've got the picture:
You downloaded the Windows installer xlwt-0.7.1.win32.exe from PyPI on
or after 26 March, and attempted to install it to Python 2.6 [which
exactly? 2.6.0 or 2.6.1?]. Everything worked OK except that it produced
the above error 4 times. This is somewhat disconcerting, but not what I
would call "fails".
I cannot reproduce this problem, neither with the currently available
installer (built using Python 2.5.4) nor with the one initially put up
on PyPI (built using Python 2.6.1).
Known problems with Windows installers:
(1) An installer built with 2.6 requires MSVCR90.DLL, and thus will not
start running on machines which don't have that DLL available. Also
even if that DLL is available, there may be SxS incompatibility ("the
application is not configured correctly") which also prevents it from
starting. This is why I reverted to using the installer built using 2.5,
which requires the commonly-available MSVCR71.DLL. Note: none of this is
specific to xlrd or xlwt.
(2) An installer built with 3.0 will fail (as in: hard error, stop, fall
over, die in a hole) on the first attempt to compile a .py file which
has an encoding other than the default utf-8 -- this is a bug in 3.0,
and doesn't seem related to your problem -- one (examples/wsprops.py) of
the two files you mentioned has no declared encoding and has only ASCII
characters and thus is compatible with utf-8.
> Thinking this was the not unheard of "module built with py2.5 fails
> when installing on py2.6" I tried building it with python 2.6 myself
> with setup.py bdist_wininst.
> The build worked fine, and the resulting executable happily installed
> xlwt to py2.5, but py2.6 still fails with the same error.
>
> This looks like a case of mixed compiler versions to me, but I make a
> point of not compiling python things. Although having said that I do
> have mingw32, cygwin, and vc 2k8 express installed :-/
I have mingw32, the free Borland compilers (old and new), and the MS
freebie monster :-//
> Any ideas? Can anyone else replicate this on their machines?
I'll ask the gurus ... after you've confirmed or amended my translation
of your problem description.
Cheers,
John
Daniel Burke
On 31/03/2009 11:44 AM, Daniel Burke wrote:
Hi Daniel,
When you want to report a problem, please start a new thread.
Sorry, I got out of hospital after an operation and my brain is still having a bit of trouble connecting dots. I saw "post problems here".... At least the pain killers are working.
> I suspect this is a problem with my configuration, but....
>
> xlwt 0.7.1 fails installing on win32 python 2.6 at the part where it
> compiles the library into pyo files. Specifically....
> -----
> Runtime Error!
>
> Program: C:\Documents (path of downloaded installer)
>
> R6304
> An application has made an attempt to load the C runtime library
> incorrectly.
> Please contact the applications support team for more information.
> ----
> This error is produced 4 times, once each for compiling the pyc and
> pyo of examples/wsprops.py and examples/unicode2.py.
> xlwt works fine, and both examples run fine too.
Let me see if I've got the picture:
You downloaded the Windows installer xlwt-0.7.1.win32.exe from PyPI on
or after 26 March, and attempted to install it to Python 2.6 [which
exactly? 2.6.0 or 2.6.1?]. Everything worked OK except that it produced
the above error 4 times. This is somewhat disconcerting, but not what I
would call "fails".
Specifically, was downloaded at 30th March 23:36 UTC.
Python 2.6.1.
My choice of words was influenced by pain killers :-)
Yeah, it's pretty minor, and not entirely unlikely just a case of this box badly needs a reformat. But it's untowards behaviour that I will happily tear this box apart to find the cause of, if it will make xlwt any better.
John Machin
OK, so that makes it the 2.5-built installer that you used.
I think xlwt would be made better if those two examples were replaced by
something more useful, rather than a ceremonial sacrifice of your hard
drive ... I doubt very much that it's anything to do with the file contents.
Cheers,
John
Chris Withers
> I think xlwt would be made better if those two examples were replaced by
> something more useful,
I had a look inside them, what's the problem with them?
(okay, they don't seem to actually make an example of much, but why are
these two and these two alone causing problems?)
John Machin
> John Machin wrote:
>> I think xlwt would be made better if those two examples were replaced by
>> something more useful,
>
> I had a look inside them, what's the problem with them?
> (okay, they don't seem to actually make an example of much, but why are
> these two and these two alone causing problems?)
Yes, Chris, that's a very good question. I've referred it to Mark
Hammond, but my guess (and that of the OP) is that it's something to do
with bitrot on the OP's computer.