Re: Logistic Regression Define X and Y for Prediction
Hi Jason, I will try it out... Nothing in the documentation tells a person. Thanks From: Python-list on behalf of Jason Friedman Sent: Wednesday, November 13, 2019 7:19 PM Cc: python-list@python.org Subject: Re: Logistic Regression Define X and Y for Prediction > > > When I define the X and Y values for prediction in the train and test > data, should I capture all the columns that has been "OneHotEncoded" (that > is all columns with 0 and 1) for the X and Y values??? > You might have better luck asking on Stackoverflow, per the Pandas instructions: https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fpandas.pydata.org%2Fcommunity.html&data=02%7C01%7C%7Ce4fa0935f3b0443c3b0f08d768988e72%7C84df9e7fe9f640afb435%7C1%7C0%7C637092876737703023&sdata=%2BO6zx05Szg3TeGdtusSaLU1GhXKp7PEL7beHpqg1hcQ%3D&reserved=0. -- https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fmail.python.org%2Fmailman%2Flistinfo%2Fpython-list&data=02%7C01%7C%7Ce4fa0935f3b0443c3b0f08d768988e72%7C84df9e7fe9f640afb435%7C1%7C0%7C637092876737703023&sdata=tDAC3St0kqFfN3rLqBBg9cTsykel5Hhj6MUjzFxZc7I%3D&reserved=0 -- https://mail.python.org/mailman/listinfo/python-list
Logistic Regression Define X and Y for Prediction
Hi All,
I have the below code.
X = df.iloc[:, [4, 403]].values
y = df.iloc[:, 404].values
Dummy Data looks like:
host Mnemonic
12.234.13.6 start
22.22.44.67 something
23.44.44.14 begin
When I define the X and Y values for prediction in the train and test data,
should I capture all the columns that has been "OneHotEncoded" (that is all
columns with 0 and 1) for the X and Y values???
import numpyas np
import pandas as pd
import os
import matplotlib as mpl
mpl.rcParams['figure.dpi'] = 400
import matplotlib.pyplotas plt
# Importing the df
# Importing the df
os.chdir('c:\directory\data')# Location of data files
df = pd.read_csv('blahblahfile.csv')
from sklearn.preprocessing import LabelEncoder
hostip = LabelEncoder()
mnemonic = LabelEncoder()
df['host_encoded'] = hostip.fit_transform(df.reported_hostname)
df['mnemonic_encoded'] = mnemonic.fit_transform(df.mnemonic)
from sklearn.preprocessing import OneHotEncoder
hostip_ohe = OneHotEncoder()
mnemonic_ohe = OneHotEncoder()
X = hostip_ohe.fit_transform(df.host_encoded.values.reshape(-1,1)).toarray()
Y =
mnemonic_ohe.fit_transform(df.mnemonic_encoded.values.reshape(-1,1)).toarray()
## Add back X and Y into the original dataframe
dfOneHot = pd.DataFrame(X, columns = ["host_"+str(int(i)) for i in
range(X.shape[1])])
df = pd.concat([df, dfOneHot], axis=1)
dfOneHot = pd.DataFrame(Y, columns = ["mnemonic_encoded"+str(int(i)) for i in
range(Y.shape[1])])
df = pd.concat([df, dfOneHot], axis=1)
here is where I am not sure if all "host_" and "mnemonic_encoded"
values assigned to X and Y
X = df.iloc[:, [4, 403]].values
y = df.iloc[:, 404].values
# Splitting the dataset into the Training set and Test set
from sklearn.model_selection import train_test_split
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.25,
random_state = 0)
# Feature Scaling
from sklearn.preprocessing import StandardScaler
sc = StandardScaler()
X_train = sc.fit_transform(X_train)
X_test = sc.transform(X_test)
# Fitting Logistic Regression to the Training set
from sklearn.linear_model import LogisticRegression
classifier = LogisticRegression(random_state = 0)
classifier.fit(X_train, y_train)
# Predicting the Test set results
y_pred = classifier.predict(X_test)
# Making the Confusion Matrix
from sklearn.metrics import confusion_matrix
cm = confusion_matrix(y_test, y_pred)
# Visualising the Training set results
from matplotlib.colors import ListedColormap
X_set, y_set = X_train, y_train
X1, X2 = np.meshgrid(np.arange(start = X_set[:, 0].min() - 1, stop = X_set[:,
0].max() + 1, step = 0.01),
np.arange(start = X_set[:, 1].min() - 1, stop = X_set[:,
1].max() + 1, step = 0.01))
plt.contourf(X1, X2, classifier.predict(np.array([X1.ravel(),
X2.ravel()]).T).reshape(X1.shape),
alpha = 0.75, cmap = ListedColormap(('red', 'green')))
plt.xlim(X1.min(), X1.max())
plt.ylim(X2.min(), X2.max())
for i, j in enumerate(np.unique(y_set)):
plt.scatter(X_set[y_set == j, 0], X_set[y_set == j, 1],
c = ListedColormap(('red', 'green'))(i), label = j)
plt.title('Logistic Regression (Training set)')
plt.xlabel('Age')
plt.ylabel('Estimated Salary')
plt.legend()
plt.show()
# Visualising the Test set results
from matplotlib.colors import ListedColormap
X_set, y_set = X_test, y_test
X1, X2 = np.meshgrid(np.arange(start = X_set[:, 0].min() - 1, stop = X_set[:,
0].max() + 1, step = 0.01),
np.arange(start = X_set[:, 1].min() - 1, stop = X_set[:,
1].max() + 1, step = 0.01))
plt.contourf(X1, X2, classifier.predict(np.array([X1.ravel(),
X2.ravel()]).T).reshape(X1.shape),
alpha = 0.75, cmap = ListedColormap(('red', 'green')))
plt.xlim(X1.min(), X1.max())
plt.ylim(X2.min(), X2.max())
for i, j in enumerate(np.unique(y_set)):
plt.scatter(X_set[y_set == j, 0], X_set[y_set == j, 1],
c = ListedColormap(('red', 'green'))(i), label = j)
plt.title('Logistic Regression (Test set)')
plt.xlabel('Host IP')
plt.ylabel('Mnemonic')
plt.legend()
plt.show()
--
https://mail.python.org/mailman/listinfo/python-list
Re: Overwhelmed by the Simplicity of Python. Any Recommendation?
Same here. Debugging in Python is annoying, I like to step through my code line by line, it's impossible to do it with object-oriented programming language. Also, there's no good REPL IDE. Spyder barely works with some basic features. PyCharm, the most popular, takes too long to start, and you have to setup folders and directories EVERY SINGLE TIME at startup. From: Python-list on behalf of songbird Sent: Saturday, November 3, 2018 12:45:57 PM To: python-list@python.org Subject: Re: Overwhelmed by the Simplicity of Python. Any Recommendation? Rhodri James wrote: ... > I completely agree. I too have come from a background in C, and still > do most of my day job in C or assembler. It took a while before I was > writing idiomatic Python, never mind efficient Python (arguably I still > don't, but as Rob says, who cares?). Don't worry about it; at some > point you will discover that the "obvious" Python you are writing looks > a lot like the code you are looking at now and thinking "that's really > clever, I'll never be able to to that." at this stage of my own process in learning, i'm trying to read the FAQs i can find, any tutorials, answers to specific questions on stackoverflow on particular topics to see if i can understand the issues, etc. as for my own code, yes, it's horrible at the moment, but to me working code is always the final arbitor. i much prefer simple and stepwise refinement if speed isn't the issue i think clarity and simplicity is more important. speed is only more important for large projects that process a ton of data. in 3-5yrs i expect to understand more of what the theory and more conceptual things going on as i read more of the history and how the language has developed. i won't consider myself fluent until i start "thinking" in it and can visualise the data structures/objects in my head and such as i currently do for C. songbird -- https://mail.python.org/mailman/listinfo/python-list -- https://mail.python.org/mailman/listinfo/python-list
Re: Why list.reverse() modifies the list, but name.replace() does not
Yes, I forgot that strings are immutable. I can't change anything in the
string. Silly me!
Thank you very much, I appreciate it. I guess sometimes it just take an
outsider to take you outside the box. And all is answered. :)
From: Python-list on behalf
of Mark Lawrence Sent: Monday, September 3, 2018
2:21:36 PM To: python-list@python.org
Subject: Re: Why list.reverse() modifies the list, but name.replace() does not
modify the string?
On 03/09/18 18:49, C W wrote:
> Hello all,
>
> I am learning the basics of Python. How do I know when a method modifies
> the original object, when it does not. I have to exmaples:
> Example 1:
>> L = [3, 6, 1,4]
>> L.reverse()
>> L
> [4, 1, 6, 3]
> This changes the original list.
Lists are mutable, i.e. can be changed, so it makes sense to do this change in
place.
>
> Example 2:
>> name = "John Smith"
>> name.replace("J", j")
>> name
> 'John Smith'
> This does not change the original string.
Strings are immutable, i.e. cannot be changed, so you have to create a new
string. Your call to `replace` will do just that, but as it's not saved `name`
remains the same. You could use
name = name.replace("J", j") or
newname = name.replace("J", j") as you see fit.
>
> Why the two examples produce different results? As a beginner, I find this
> confusing. How do you do it?
>
> Thank you!
>
--
My fellow Pythonistas, ask not what our language can do for you, ask what you
can do for our language.
Mark Lawrence
--
https://mail.python.org/mailman/listinfo/python-list
--
https://mail.python.org/mailman/listinfo/python-list
Re: Why list.reverse() modifies the list, but name.replace() does not modify the string?
Yes, I forgot that strings are immutable. I can't change anything in the
string. Silly me!
Thank you very much, I appreciate it. I guess sometimes it just take an
outsider to take you outside the box. And all is answered. :)
From: Python-list on behalf
of Mark Lawrence
Sent: Monday, September 3, 2018 2:21:36 PM
To: python-list@python.org
Subject: Re: Why list.reverse() modifies the list, but name.replace() does not
modify the string?
On 03/09/18 18:49, C W wrote:
> Hello all,
>
> I am learning the basics of Python. How do I know when a method modifies
> the original object, when it does not. I have to exmaples:
> Example 1:
>> L = [3, 6, 1,4]
>> L.reverse()
>> L
> [4, 1, 6, 3]
> This changes the original list.
Lists are mutable, i.e. can be changed, so it makes sense to do this
change in place.
>
> Example 2:
>> name = "John Smith"
>> name.replace("J", j")
>> name
> 'John Smith'
> This does not change the original string.
Strings are immutable, i.e. cannot be changed, so you have to create a
new string. Your call to `replace` will do just that, but as it's not
saved `name` remains the same. You could use
name = name.replace("J", j") or
newname = name.replace("J", j") as you see fit.
>
> Why the two examples produce different results? As a beginner, I find this
> confusing. How do you do it?
>
> Thank you!
>
--
My fellow Pythonistas, ask not what our language can do for you, ask
what you can do for our language.
Mark Lawrence
--
https://mail.python.org/mailman/listinfo/python-list
--
https://mail.python.org/mailman/listinfo/python-list
Re: Python Object Systems
On 14-08-11 04:26 PM, thequietcen...@gmail.com wrote: ... Hello, has anyone created a survey of Python Object Systems? The two I am aware of are: - elk https://github.com/frasertweedale/elk - Traits http://code.enthought.com/projects/traits/ Here's the ones from my talk at Pycon 2005 (http://www.vrplumber.com/programming/descriptors-pycon2005.pdf): OpenGLContext/PyVRML97 http://bazaar.launchpad.net/~mcfletch/pyvrml97/trunk/files Observable, auto-coercing data properties w/ Numeric/numpy array support, defaults BasicProperty (now largely abandoned): http://bazaar.launchpad.net/~mcfletch/basicproperty/trunk/files Again, observable auto-coercing typed properties, defaults, introspection Zope2 FieldProperty, DublinCore, Validation, Observability PEAK Defaults, delegation, implicit feature loading Traits Delegation, typing, validation, defaults, Observability, introspection PyObjc, ctypes, JythonC, IronPython Function-like things with lots of metadata There's also a listing of the other tasks for which descriptors/object systems were expected to show up at the time, if you look for Python + that key-word you'll likely find a few dozen more "object systems" out there. You'll also likely find about a thousand metaclasses these days. HTH, Mike -- ____ Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- https://mail.python.org/mailman/listinfo/python-list
Re: handling return codes from CTYPES
On 13-01-21 11:52 AM, Steve Simmons wrote: Mike, Thanks for your response - I was puzzled by one part of it though... On 21/01/2013 15:14, Mike C. Fletcher wrote: That's because you've just discarded the object you created I (mis?)understood from the ctypes documentation that '>>> initResult = c_short(0)' would result in the creation of a ctypes 'short' called initResult and that this object would be mutable. On that basis, I expected the following statement '>>> initResult = initScanLib( ... )' would assign the result of the call to initResult. So, I can understand that I am not using the correct approach but I don't understand how I discarded the object I created. Can you clarify please? Sure, the problem isn't here a ctypes issue, but a Python one. When you do the following: >>> initResult = c_short(0) you have bound the name "initResult" (a string key in a "namespace" dictionary) to a ctypes c_short object. The name "initResult" is in no way special, nor is there any "type" associated with that variable name. The fact that it has been assigned to a c_short *at this moment* does not affect any *future* value assigned to that name. >>> initResult = initScanLib( ... ) Here you have assigned whatever object is the result of initScanLib( ... ) to the name "initResult". The old object pointed to by "initResult" is no longer referenced by that name, so the Python intepreter is free to discard that old object immediately. Thus you have "discarded" the object you created by reassigning the single reference to it to another object, which leaves it free to be garbage collected (depending on Python implementation that might be instantly or eventually). Python does not have typed *variables*, every variable is a pointer to an object (PyObject *) under the covers. There is nothing in Python like: int i; short j; nothing which makes a variable type-enforcing or constrained. At least, nothing in core Python. In theory you can create a namespace which *does* allow such things, but that's getting pretty involved. Hope that helps, Mike -- Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: handling return codes from CTYPES
On 13-01-21 05:46 AM, Steve Simmons wrote:
...
>>> from ctypes import *
>>> sLib = cdll.slib
>>> lic_key = c_char_p("asdfghjkl".encode(encoding='utf_8',
errors='strict'))
>>> initResult = sLib.InitScanLib(lic_key.value)
>>> print("InitScanLib Result: ", initResult)
InitScanLib Result: 65535
>>>
I've tried declaring initResult as c_short by: inserting...
>>> initResult = c_short(0)
... before the call to sLib.InitScanLib but I still get the same
response (65535).
That's because you've just discarded the object you created.
What you wanted was, I believe:
initScanLib = sLib.InitScanLib
initScanLib.restype = c_short
initResult = initScanLib( ... )
i.e. you tell the initScanLib function how to coerce its result-type.
*Some* C functions take a pointer to a data-value to fill in their data,
but not *your* function. That pattern looks like:
result = c_short(0)
my_ctypes_function( ..., byref(result) )
print result.value
i.e. you have to pass the variable into the function (as a
reference/pointer).
HTH,
Mike
--
Mike C. Fletcher
Designer, VR Plumber, Coder
http://www.vrplumber.com
http://blog.vrplumber.com
--
http://mail.python.org/mailman/listinfo/python-list
Re: pylint, was Re: pygame - importing GL - very bad...
On 13-01-02 09:48 PM, Terry Reedy wrote:
...
2) self.lightDone: Invalid name "lightDone" (should match
[a-z_][a-z0-9_]{2,30}$)
So I can now understand that pylint doesn't like my naming convention
with a capital letter in the middle of the variable name, like:
"lightDone" = a boolean value. I suppose pylint wants me to use (a
little longer method) an underscore to separate words in long variable
names...
That is more conventional in the Python community (and is in pep 8, I
believe) but still a choice.
That seems like a improper error message from the tool. "Invalid name"
does *not* properly describe that situation. The name is *not*
"Invalid" in any sense of the word, and a "checker" that tells you it is
is creating needless false-positives. An error checker should be saying
something like:
"self.lightDone: Does not match PEP8 recommended style"
making it clear that this is *not* an error, it is a *style* related
*warning*.
HTH,
Mike
--
Mike C. Fletcher
Designer, VR Plumber, Coder
http://www.vrplumber.com
http://blog.vrplumber.com
--
http://mail.python.org/mailman/listinfo/python-list
Re: pygame - importing GL - very bad...
On 13-01-02 08:53 PM, someone wrote: On 01/02/2013 10:57 PM, Michael Torrie wrote: On 01/01/2013 04:49 PM, someone wrote: On 01/01/2013 12:13 PM, Chris Angelico wrote: > You could simply > > import OpenGL.GL as GL You're right - but I forgot to write that even though this maybe should/is recommended many places then I've seen a lot of opengl code on the internet and IMHO NOBODY does that and it'll be a lot slower to type that in front of all the opengl commands... So this solution is not something I like too... But I can see some other people came up with good solutions, which I didn't knew about.. Why is this solution not to your liking? Python has namespaces for a Because the amount of opengl-functions is HUGE, many people (at least on the internet) do as I and (IMHO) it takes up too much time to change a lot of code plus sometimes I grab/modify small code pieces from the internet and it makes my development SO MUCH faster just to make an exception here with star-import for opengl-commands. I'd agree on it being rather impractical/pointless/verbose to have every single OpenGL entry point and constant have an extra gl. or glu. or glut. added to the front. OpenGL/GLU/GLUT is already namespaced, but using C-style prefix namespacing (that is gl* glu* glut* and GL_*, GLU_*, GLUT_*), so adding Python style namespacing to the front of that makes it very verbose. OpenGL-using code is *littered* with OpenGL entry points and constants (and yes, I intend the slight slight), so that's going to make it rather annoying to work with. PyOpenGL's current approach is mostly attempting to maintain backward compatibility with the older revisions. wxPython actually rewrote its whole interface to go from * imports into namespaced lookups and then wrote a little migration tool that would attempt to rewrite your code for the new version. They also provided a transitional API so that code could mix-and-match the styles. For PyOpenGL that would look something like this: from OpenGL import gl, glu, glut gl.Rotate(...) gl.Clear(gl.COLOR_BUFFER_BIT) or, if you really needed PEP-8 compliance, and don't mind making the API look nothing like the original, we might even go to: from opengl import gl, glu, glut gl.rotate(...) gl.clear(gl.COLOR_BUFFER_BIT) Either of which would *also* make it possible for us to lazy-load the entry points and symbols (that would save quite a bit of ram). But I'm not actually likely to do this, as it makes it far more annoying to work with C-oriented references (and since PyOpenGL is primarily used by new OpenGL coders who need to lean heavily on references, that's a big deal). Currently you can often copy-and-paste C code into PyOpenGL and have it work properly as far as the OpenGL part is concerned (arrays and the like need to be rewritten, but that's not something I can control, really). People are already confused by the small variations from C OpenGL, making the API look entirely different wouldn't be a good direction to move, IMO. HTH, Mike -- Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: adding a simulation mode
On 12-07-04 05:42 AM, andrea crotti wrote: ... copytree(src, dest) becomes: if not PRETEND_ONLY: copytree(src, dest) import globalsub, unittest class MyTest( unittest.TestCase ): def setUp( self ): globalsub.subs( shutil.copytree ) def tearDown( self ): globalsub.restore( shutil.copytree ) You can also pass a function to subs like so: def setUp( self ): self.copied_trees = [] def fake_copytree( src, dest ): assert os.path.exists( src ) self.copied_trees.append( (src, dest )) return dest # or whatever the thing should return globalsub.subs( shutil.copytree, fake_copytree ) $ pip install globalsub HTH, Mike -- Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: XSLT to Python script conversion?
On 12-02-13 06:20 AM, Matej Cepl wrote: Hi, I am getting more and more discouraged from using XSLT for a transformation from one XML scheme to another one. Does anybody could share any experience with porting moderately complicated XSLT stylesheet (https://gitorious.org/sword/czekms-csp_bible/blobs/master/CEP2OSIS.xsl) into a Python script using ElementTree's interparse or perhaps xml.sax? Any tools for this? Speed differences (currently I am using xsltproc)? Any thoughts? Thank you, Matěj I wound up rewriting the Docbook to XSL transformation for PyOpenGL's docs in Python using lxml.etree and Kid (now reworked to use Genshi). However, that was a fairly direct translation, it has only a handful of strategies for transforming nodes from docbook to xhtml. That said, it took our processing time down from so-long-I-just-didn't-want-to-work-on-the-docs down to regenerate-whenever-I-make-a-trivial-change. http://bazaar.launchpad.net/~mcfletch/pyopengl/directdocs/files <http://bazaar.launchpad.net/%7Emcfletch/pyopengl/directdocs/files> Is the repository where the project lives. It *also* does a lot of other processing, but the generate.py, model.py and templates/section.kid files are all you need to look at to understand the docbook processing. HTH, Mike -- ____ Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: subprocess module and long-lived subprocesses
On 12-01-20 09:42 AM, s...@pobox.com wrote: > I'm converting some os.popen calls to use subprocess.Popen. I had > previously been ignoring stdout and stderr when using os.popen. The primary > motivation to switch to subprocess.Popen now is that I now want to check > stderr, so would have to make code changes to use os.popen[34] anyway. > Might as well go whole hog and switch to the new API. > > The library documentation doesn't talk a lot about long-lived subprocesses > other than the possibility of deadlock when using Popen.wait(). Ideally, I > would write to the subprocess's stdin, check for output on stdout and > stderr, then lather, rinse, repeat. Is it safe to assume that if the stdout > and/or stderr pipes have nothing for me the reads on those file objects (I'm > using PIPE for all three std* files) will return immediately with an empty > string for output? They won't block, will they? Will a broken pipe IOError > get raised as for os.popen() or do I have to call Popen.poll() even in error > situations? > > Thanks, > Definitely *will* block, you have to explicitly set them non-blocking to have non-blocking behaviour: def set_nb( fh ): """Set non-blocking flag on given file handle""" if isinstance( fh, int ) or hasattr( fh, 'fileno' ): flags = fcntl.fcntl(fh, fcntl.F_GETFL) fcntl.fcntl(fh, fcntl.F_SETFL, flags| os.O_NONBLOCK) on each of the 3 buffers, then you need to attempt read/write on each of them periodically, catching the EWOULDBLOCK errors, to prevent deadlocks where the buffers have filled up (e.g. because the subprocess is printing out errors on stderr, or because it is generating output, or because for some reason the process isn't reading your input fast enough). I think everyone winds up with their own wrapper around subprocess after they use it for more than a short period... HTH, Mike -- Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: memory leaks - tools and docs
On 11-11-24 10:00 PM, Aljosa Mohorovic wrote: > i've been trying to find memory leaks in a wsgi application using > gunicorn to run it and after a lot of time invested in research and > testing tools i did find a lot of useful information (most really old) > but i'm left with a feeling that this should be easier, better > documented and with tools that generate better data. > > if anybody can share some tips, links, docs or better tools with > better reports i would really appreciate it. > i'm not against paying for a good tool so any recommendation is > appreciated. > > i mostly used http://guppy-pe.sourceforge.net/#Heapy but found > http://pysizer.8325.org/ and http://code.google.com/p/pympler/ also > interesting. > > Aljosa Meliae is a similar tool wrt collecting memory-usage information. RunSnakeRun can process Meliae dumps to produce visualizations of the memory used in the process. HTH, Mike https://launchpad.net/meliae http://www.vrplumber.com/programming/runsnakerun/ -- Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: logging: warn() methods and function to be deprecated.
On 11-10-26 10:51 AM, Vinay Sajip wrote: ... > auto-translation via 2to3 (because you generally are calling log.warn() >> rather than logging.warning, but sometimes you are doing getattr( log, >> log_level ) and then passing that method around a few times), and it > That doesn't sound like a good usage pattern to me, especially as loggers > have a > log method which takes the logging level. There shouldn't be any need to pass > a > bound method around. Bound methods also pull along to *which* log you are going to send the message, but then I suppose you could just use the logging key as well as a piece of data. I'll withdraw the suggestion that it is not a trivial thing to add to 2to3, though I'll leave the implementation to someone else. Have fun, Mike -- ____ Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: logging: warn() methods and function to be deprecated.
On 11-10-26 05:12 AM, Vinay Sajip wrote: > Mike C. Fletcher vrplumber.com> writes: > >> I actually consider .warning() a nit :) . After all, it's 3 extra >> characters :) , and *who* actually reads documentation instead of just >> poking around and finding the shortest-named method in the instance? > Readability counts :-) Are you saying there's no need to bother documenting > stuff ??? ;-) More: an undocumented entry point is not "deprecated" because, after all, it shows up in PyDoc as a regular method. > >> Anyway, I personally don't see this as worth the breakage. > > What breakage are we really talking about? Remember, Python 2.x will not > change > in this area - the proposed change is for Python 3.3 and later only, and will > not be backported :-) > > As far as I know, Trac doesn't work with Python 3 anyway. Most of the code out > there (which Mark found via Google Code Search) is Python 2.x. When porting > from > 2 to 3.3, it's just one extra little thing to deal with - small compared with > other issues which come up when doing such ports. Sure, but most of *everything* is Python 2.x, and porting to Python 3.x is already enough of a pain that key-stone projects like Trac still aren't there :) . This isn't going to be easily amenable to auto-translation via 2to3 (because you generally are calling log.warn() rather than logging.warning, but sometimes you are doing getattr( log, log_level ) and then passing that method around a few times), and it will often fall into the small percentage of untested code in most projects (coverage is often poor for logging messages in corner cases), so often won't get caught by test suites. Not a 3.x user, but expecting to have to be some day in the future, Mike -- Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: logging: warn() methods and function to be deprecated.
On 11-10-24 04:28 AM, Vinay Sajip wrote: >> I think that is a real shame - it seems to be gratuitous breakage for almost >> zero benefit. That issue shows that Trac makes heavy use of .warn, I've use >> .warn almost exclusively for many years, and code.google.com shows it is >> used >> extensively in the wild. > Okay, but it's easy enough to change, and people are getting a reasonable > amount of time to deal with it. I'd have to echo Mark's sentiment here. There is a *lot* of (fairly old) code around that uses .warn, and I'd wager even a lot of new code written by old programmers (such as myself) that uses it. Breaking it to avoid having an undocumented method seems wrong; suggest just adding documentation: "logger.warn(...) -- an alias to logger.warning provided for backward compatibility" >> Is there still a chance to reconsider? > I'm not dogmatic about things like this; warn() is just a hangover from a > long time ago and bit of a nit, that's all. I suppose I should have removed > it when 3.0 was released, but it went under my radar at that time. > > Hence my post here, to get feedback from logging users about this proposed > change. I actually consider .warning() a nit :) . After all, it's 3 extra characters :) , and *who* actually reads documentation instead of just poking around and finding the shortest-named method in the instance? Anyway, I personally don't see this as worth the breakage. Just my $0.02, Mike -- Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: PyDispatcher on sourceforge doesn't play nice with easy_install
Chris Withers wrote: > Hi All, > > Using easy_install to get PyDispatcher results in: > ... > Who's the maintainer of PyDispatcher nowadays? Would be handy if they > removed the sourceforge link from pypi. ... Thanks for the report. I've released a 2.0.2 version on PyPI. That should now work properly with easy_install. HTH, Mike -- ________ Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: Python for newbies (Pythno users group in Macedonia)
Дамјан Георгиевски wrote: > Hi all, > we are starting with bi-monthly Python User Group meetings in Skopje, > Macedonia. The meetings are targeted for both beginners and more > experienced users. > ... http://us.pycon.org/2010/conference/schedule/event/108/ HTH, Mike -- ____ Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: Library support for Python 3.x
Kevin Walzer wrote: > I'm going to be starting some new Python projects in Python 2.6, but > am concerned that at least three of the libraries I will be > using--pycrypto, paramiko and feedparser--are not currently supported > in Python 3.x. The authors of these programs have not given any > indication that work is underway to support Python 3.x. Eventually I > plan to migrate to Python 3.x. > > I don't want to be stuck using libraries that may not be updated for > the next generation of Python. How are others handling this issue? > > --Kevin Some of us are saying: * When Python 3.x has something compelling to offer: o Users will start asking for Python 3.x with a *reason* to justify the cost o Libraries will begin to see the pain of porting, and most importantly, *maintaining*, as a reasonable trade-off o Libraries will port (and maintain) o Applications should *then* port * or When everyone is already on a (basically) compatible 2.x version (i.e. 2.6+); say in 3 years: o Conversion and maintenance costs will lower to the point where we can justify them for libraries Python 3.x has very little that is a compelling use-case (versus 2.x) for most people I've talked to. My paying customers aren't screaming "we want to spend a week of your time to rewrite, re-test and re-deploy our working code into Python 3.x so that it can do exactly the same thing with no visible benefit and lots of likely subtle failures". Unicode-as-default doesn't really make customers say "wow" when all their Unicode-needing code is already Unicode-using. A few syntax changes here and there... well, no, they certainly don't care (can't say I *really* do either). The initial marked slowdown for 3.0 (which I gather should be mostly fixed in 3.1) didn't do much of a sales-job either. Possible compelling arguments that would get people to convert (with the projects working on them): * 5x (or more) speedup (PyPy, Unladen Swallow) * adaptive parallelization/execution-kernel mechanism as a first-class primitive (Apple's C extensions for OpenCL) * continuation-like mechanisms, anonymous code blocks a-la Ruby (PyPy, Stackless) * (free) threading, GIL removal (IronPython, Jython) * compilation-to-native (PyPy) * compilation to mobile-platform native (PyPy?) None of those in in Python 3.x, though there's talk of merging Unladen Swallow into CPython to provide a carrot for conversions (though AFAIK it's not yet a 5x improvement across the board). To compound the problem, Python 3.x doesn't include any of the syntax changes you'd want to see to support e.g. anonymous blocks, continuations, OpenCL integration, etceteras... so if we want those, we're going to have to either introduce new syntax (which is blocked) or hack it in... which we could do on Python 2.x too. I don't know about other maintainers, but I've started running PyOpenGL tests with -3 on Python 2.6 to get a feel for how many issues are going to come up. Most have been minimal. But when I sit down and actually consider *maintaining* a 3.x release when I'm already *barely* able to keep up with the 2.x maintenance in my tiny amounts of free time... well... I do *that* stuff for *fun* after all, and double the headaches for no noticeable benefit doesn't really sound like fun... oh, and numpy isn't ported, so screw it ;) ... Interestingly, at PyGTA Myles mentioned that he'd found his first-ever Python 3.x-only library! Which he then converted to Python 2.x, because he actually wanted to use it in real code :) . Projects which haven't ported to Python 3.x aren't necessarily dead, they're just not nailed to that particular perch (yet), Mike -- Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: PyOpenGL and graphics card support
jefm wrote:
> these are the imports I use:
>
> from OpenGL.GL import *
> from OpenGL.GLUT import *
> from OpenGL.GLU import *
>
PyOpenGL will use the default OpenGL renderer for your system, however,
before you have an OpenGL context (rendering window) the system can
report whatever the heck it wants to for the various values (in this
case looks like it is reporting the generic microsoft renderer until you
ask for a context). Try this instead:
from OpenGL.GL import *
from OpenGL.GLUT import *
from OpenGL.GLU import *
resX,resY = (400,300 )
if __name__ == "__main__":
glutInit([])
glutInitDisplayMode(GLUT_RGBA | GLUT_DOUBLE | GLUT_DEPTH)
glutInitWindowSize(resX, resY)
glutInitWindowPosition(0, 0)
window = glutCreateWindow("hello")
for name in
(GL_VENDOR,GL_RENDERER,GL_SHADING_LANGUAGE_VERSION,GL_EXTENSIONS):
print name,glGetString(name)
HTH,
Mike
--
____
Mike C. Fletcher
Designer, VR Plumber, Coder
http://www.vrplumber.com
http://blog.vrplumber.com
--
http://mail.python.org/mailman/listinfo/python-list
Re: Twisted 100% event driven or hybrid?
exar...@twistedmatrix.com wrote: > On 05:55 am, jacopo.pe...@gmail.com wrote: ... >> results to be ready, I collect and process them. From now on I don 19t >> need a the system to be event drive any more, the processing should >> occur only on the master machine, following a deterministic flow. >> As soon as finished I am ready to start again to resubmit the models >> for recalculation and so on. This should go on forever. Jean-Paul is obviously far more authoritative on the "twisted way" than I am, so if he says you can just run your synchronous operation in-situ, that's probably the way to go, but IIRC there's a reactor.deferToThread() function which can run your synchronous code "off to the side", while allowing the twisted code to continue to process incoming operations. Thus you'd do something like: def process( dataset ): dl = [ remote_call( x ) for x in dataset] dl = defer.DeferredList( dl ) def on_all_results( results ): reactor.deferToThread( sync_process, (results,)).addCallback( process ) return dl.addCallback( on_all_results ) (I'm typing all of that from the distance of a few years of memory decay, so take it as "loosely this, with the proper function names and the like"). Threads aren't really part of the "twisted way" in my understanding, but they can be used if necessary AFAIK, and they will let your application remain responsive to network events during the processing. HTH, Mike -- Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: Python and 3d
azrael wrote: > Has anyone any exipience with python and 3d. > > I mean, is there a module to deal with popular 3d formats like 3ds, or > vrml. is it possible to import into python opengl models and then use > it in application for GUI purposes like through WX. I know that WX > supports OpenGL but how to import models from file. > http://www.vrplumber.com/py3d.py There's a few older projects to load 3DS files, but I don't think any of them is currently PyOpenGL 3.x compatible. OpenGLContext loads VRML97 files (with SimpleParse installed). Pyglet, OpenGLContext and Ian Mallet's game engine all load .obj models (I believe Ian's engine is the most advanced there). Pivy should load VRML files as well, and has a very powerful engine (COIN) under the covers (OpenGLContext is more of a demo/testing system). The big engines (Ogre, Soya, OpenSceneGraph, Crystal Space, etc) should all have content loaders, though I haven't played with them enough to be able to say what formats they support. HTH, Mike -- ____ Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: Observer implementations
Tobias Weber wrote: > In article , > "Mike C. Fletcher" wrote: > > >> See PyDispatcher for code to do this. >> > > That was the original problem. Got it now: if used inside the class > definition dispatcher.connect will raise "cannot create weak reference > to 'classmethod' object". Outside (using Class.method) it works fine. > Ah, I think you've got a logic problem there. The classmethod during class creation has no reference to the class being created, so if you were to call *that* thing it would blow up: class x( object ): @classmethod def y( cls, text ): print text y( 'No you do not!' ) if you were to create a reference to *that* thing (the decorated un-bound class method) it would never have the binding for cls, so it wouldn't do anything. The correct way IMO to bind after-class-creation (e.g. by a class decorator or metaclass) where you reference a bound class method. If you *really* want to do it this way, you can do something like this: class x( object ): @classmethod def y( cls, value ): pass dispatcher.connect( lambda *args: x.y( *args ), signal='hello' ) but ick. HTH, Mike -- Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: Observer implementations
Tobias Weber wrote: ... No time to reinvent the wheel I'd still need to know how to weakref a classmethod See PyDispatcher for code to do this. PyDispatcher, at least, is not abandoned, it would be more accurate to say "finished". I use it in OpenGLContext (extensively), but I haven't had to change anything in a rather long time. It pretty much just works. Enjoy, Mike -- ____ Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: PyOpenGL Installation
Roastie wrote: > The web info I have found for downloading/installing PyOpenGL and > its dependencies for Windows is a nightmare. Is there a place where > all of > the endless dependencies are packaged nicely with PyOpenGL? > Hmm, "endless"? * Python (with ctypes, standard on everything >= 2.5) * OpenGL (standard on almost every Win32 machine save old Win95 boxes) Possibly I'm missing something here? If you just want PyOpenGL you only need one file (the PyOpenGL package) + Python. If you want to use it with Numpy, you need to have numpy on the machine, same for GLUT, GLE and the like, but they aren't required unless you want to use them. Heck, if you want you can even easy_install the package last I checked. http://pyopengl.sourceforge.net/documentation/installation.html My dictionary's definition of endless must be broken ;) :) , Mike -- ____ Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: ANN: SuPy 1.6
Greg Ewing wrote: ... What is SuPy? - SuPy is a plugin for the Sketchup 3D modelling application that lets you script it in Python. Hi Greg, Can you review the SuPy writeup on Py3D? I couldn't get through to your server this evening, so I couldn't check the license, if there's anything else you feel should be said about the program, let me know. http://www.vrplumber.com/py3d.py#supy Have fun, Mike -- ________ Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: OpenGL in TK
Almar Klein wrote: Hi list! I am using pyOpenGL to do some 3D rendering. For now I am quite happy with the GLCanvas of wx, but I may want to publish some stuff later, and it would be nice if people would not need wx. I found that there used to be a TK widget called Togl, but it is not (anymore?) included in the TK package that comes with Python. Togl was never included with Python, it used to be included with PyOpenGL, but it was always a royal PITA to build as it had to be built against the dev versions of Tcl and Tk that matched the Python ones. There's now a Togl project on SourceForge that provides a Win32 build of Togl. There's a hacky script in PyOpenGL 3.x's "src/toglinstall" directory that tries to download the package from there and install it for you, but I'd expect it to be rather fragile and broken. You wind up with an external binary download requirement one way or another. Togl is pretty small compared to wxPython, of course, but you're looking a the same basic problem. Have others been able to use Togl, or any other OpenGL widget in TK? I installed it using the above-mentioned hacky script a few months ago, checked that the wrapper (OpenGL/Tk) still worked with it, and then promptly ceased to particularly worry about it any more. As has been the case for years, Togl needs an advocate who wants to work on the support in PyOpenGL, without that I just have no particular interest in maintaining it and the headache of maintenance has vastly outweighed my interest. HTH, Mike -- ________ Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: Python 3D CAD -- need collaborators, or just brave souls :)
nding on your model, you may want to allow for converting extrusions to faces in order to allow tweaking the resulting extrusion. You'll want/need a UI to "open" the extrusion and change the backbone and sweep shapes. measurement measurement == dimensioning (dim*)? Or just on-screen measurement (dist). Both are useful. So there it is. Any takers? :) I can't even begin to commit any real time to another project (I barely have time to work on the ones I already have), but if you need feedback, feel free to email me. Most of the graphics stuff sounds like stuff you could build on top of OpenGLContext or COIN or any of the generic scenegraph libraries. They aren't really CAD-focused, but they've already got the ability to draw the things you're wanting to work on, and have the Transforms and similar support to make the coding a reasonable task. However, they're closer to a Maya/3DSMax model than AutoCAD, so maybe you'll decide you want to go your own way. You may want to check out PythonCAD as well, which IIRC does 2D-only CAD. Anyway, hope this was some help. Good luck, Mike -- Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: PyOpenGL double-buffered hardware surface
Clay Hobbs wrote: How do I create a double-buffered hardware surface with PyOpenGL? I knew how to once, but forgot. Depends on your GUI library, most of them have a flag-set that you pass to the initializer of the OpenGL-holding widget. If you're using Pygame, see Pygame's display module. If wxPython, see their GLcanvas object. If GLUT, see glutInitDisplayMode, etceteras. Good luck, Mike -- ____ Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: Some questions about PyOpenGL and wxPython
Clay Hobbs wrote: I am making a program with wxPython that renders objects in 3D using PyOpenGL, and I am having some problems. For one thing, I forgot how to make a double-buffered hardware surface. http://bazaar.launchpad.net/~mcfletch/openglcontext/trunk/annotate/1?file_id=wxcontext.py-20080920224554-ehwlv3u6uc6sb6e2-55 See method wxFlagsFromDefinition, particularly glcanvas.WX_GL_DOUBLEBUFFER For another thing, glColor(1.0, 0.0, 0.0) just before the rendering doesn't make the object red. Please help, I'm a total noob to OpenGL. Would have to see what you're doing to know what's wrong there. glColor should be working. HTH, Mike -- ____ Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: PyOpenGL Tutorial?
arsyed wrote: ... Also, see: http://www.siafoo.net/browse?keyword_id=245 But note regarding the second tutorial there that the PyOpenGL 3.x *does* supply wrappers for most publicly known extensions, so you shouldn't have to create your own wrappers for them any more. There's also tutorials that are not wxPython specific on the PyOpenGL documentation page. HTH, Mike -- ____ Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: simpleparse - what is wrong with my grammar?
Laszlo Nagy wrote: ... > Is there any way I can specify precedence tables? Or do I have to use > a different parser generator then? You normally specify them as top_precedence/second_precedence/last_precedence sets. SimpleParse' limitation is with ambiguity, not with precedence. SimpleParse will wind up "overparsing" noticeably for each operation, that is, it will parse "expr" and "word" once for each level of precedence. SimpleParse is normally fast enough that you don't really care, but optimising those cases is something that's sat on my "would be nice" list for a long time. Unlike separate lexing/parsing systems, it's a little more involved for SimpleParse to do that without losing the generality of the parsing approach (i.e. the flexibility of a non-lexed approach). HTH, Mike -- http://mail.python.org/mailman/listinfo/python-list
Re: simpleparse - what is wrong with my grammar?
Laszlo Nagy wrote:
> The program below gives me "segmentation fault (core dumped)".
>
> Environment:
>Linux gandalf-desktop 2.6.20-16-generic #2 SMP Tue Feb 12 05:41:34
> UTC 2008 i686 GNU/Linux
>Python 2.5.1
>
> What is wrong with my grammar? Can it be an internal error in simpleparse?
>
You've created an infinitely recursing grammar. SimpleParse is a
straightforward recursive descent parser without look-ahead (unless
explicitly coded) or ambiguity resolution. You are asking it to parse
"expr" to see if "expr,binop,expr" is matched. It will continue
recursing down into "expr" until it runs out of ram or otherwise
encounters a system-level fault.
You need to disambiguate your grammar to have it work with SimpleParse.
How you do that will depend on what you really meant by expr,binop,expr:
a&b&c
could be parsed as either:
((a&b)&c)
or:
(a&(b&c))
the following creates a parser that produces the first option (which is,
I *think* what you wanted):
expr:= (simple_expr,no_binop)/binop_expr
binop_expr := simple_expr,binop_tail
>simple_expr< := (paren_expr/unop_expr/word)
>no_binop< := ?-binop/end
:= EOF
paren_expr := "(",expr,")"
unop_expr := unop,expr
>binop_tail< := binop,expr
unop:= ("+"/"-")
binop := ("|"/"&"/"@")
word:= [a-zA-Z], [a-zA-Z0-9_]*
the ?- lookahead is reasonably inefficient, but lookahead is always
going around discarding partial results anyway, so it's no worse than
most other formulations.
HTH,
Mike
--
Mike C. Fletcher
Designer, VR Plumber, Coder
http://www.vrplumber.com
http://blog.vrplumber.com
--
http://mail.python.org/mailman/listinfo/python-list
Re: IronPython vs CPython: faster in 1.6 times?
dmitrey wrote: > Hi all, > the url http://torquedev.blogspot.com/2008/02/changes-in-air.html > (blog of a game developers) > says IronPython is faster than CPython in 1.6 times. > Is it really true? > On certain platforms, I believe so, for certain types of operations. Not sure if Mono also provides a speedup. Most of the speedup is due to large amounts of (paid) effort being spent creating a high-speed ILM optimizer. Because IronPython can make use of the work that MS has been poring into Dynamic language compilation, it can get quite a few speedups that CPython just doesn't get because they don't have the people to do the work. Optimising code automatically is a reasonably complex process that tends to introduce lots of potential errors. The CPython devs are not AFAIK working on performance much these days, so likely CPython won't improve any time soon, i.e. 3.0 will likely not be any faster than 2.5 from anything I've heard. PyPy is attempting to address this issue via a separate interpreter, but it's currently just playing catch-up on performance most of the time. It does have a JIT, and might one day be fast enough to be a usable replacement for CPython, but it will require a lot of developer-years to get it there, most likely. It would be really nice if PyPy could get Python 2.5 running say 5x faster and then run with that. With that Python would open out into entire new areas of applicability, becoming reasonable as an embedded language, or a systems language. Only 2x slower than C would make Python pretty close to a perfect language... (far more attractive than a slightly tweaked syntax IMO). That's probably 5-10 developer years out, though, not counting any distractions from trying to support Python 3.x. > If yes, what are IronPython drawbacks vs CPython? > Mostly library access from what I understand. Numpy and SciPy, for instance, are not AFAIK ported to IronPython. Those are the people who *really* need speed, and without those APIs having "Python" available faster doesn't really mean much. IronPython has access to the Win32 API, so if you want to use Win32 APIs, rather than the CPython ones, you're golden, but Numpy/SciPy's interface is *really* elegant for working with large arrays of data. If you're trying to write tight numeric loops for gigabyte arrays in raw Python, 1.6 times performance isn't really usable... even 5x is just barely usable. Numpy lets you use the optimized (C) libraries for the heavy lifting and Python friendliness where you interact with humans. If Python were 10x faster you *might* completely rewrite your Numpy in Python code, but I'd expect that naive Python code would still be beat handily by BLAS or what have you under the covers in Numpy. If the implementation is hard to explain, it's a bad idea. If the implementation is easy to explain, it may be a good idea. are the two lines that tend to preclude CPython ever becoming *really* fast. Optimizing code is almost always complex and hard to explain. You need lots and lots of thought to make a compiler smart enough to wring performance out of naive code, and you need a lot of thought to reason about what the compiler is going to do under the covers with your code. IronPython (and Jython, and Parrot) can use the underlying system's complexity without introducing it into their own project. PyPy is trying to create the complexity itself (with the advantage of a smaller problem domain than optimising *every* language). > And is it possible to use IronPython in Linux? > Yes, running on Mono, though again, I don't believe Mono has had the optimisation effort put in to make it competitive with MS's platforms. Just my view from out in the boonies, Mike -- http://mail.python.org/mailman/listinfo/python-list
Re: PyOpenGL
[EMAIL PROTECTED] wrote: > Hi all, > > I apologize if this question was already answered before but I was > unable to find a proper solution to my problem. Anyways, I am trying > to run shaderobjects.py on Windows (Python 2.5.1) by just double- > clicking, and I got the following error: > > [...] > File "/usr/lib/python2.5/site-packages/OpenGL/extensions.py", line 13, > in hasGLExtension > AVAILABLE_GL_EXTENSIONS[:] = glGetString( GL_EXTENSIONS ).split() > AttributeError: 'NoneType' object has no attribute 'split' > > _I think_ I have all requirements and packages properly installed. > > Any clues? Thanks in advance ! > That *looks* like a problem with calling glGetString before you have a valid context. Normally that actually works on GLX platforms, but apparently not in this case. Try delaying your import until after you have the context setup and see if that lets glGetString return a valid pointer. HTH, Mike -- Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: TopSort in Python?
Paddy wrote: ... > I searched for dependancy sort, and later dependency sort (cos I > couldn't spell). I had convinced that I was using the right term and > was flummoxed by the lack of hits. Even today the term topological > sort means far less than what it describes: sorting items based on > their interdependencies. > "dependency sort python" typed into Google today gives a post pointing to http://www.vrplumber.com/programming/ (which has Tim and my algorithms (toposort.py)) as the second link... vagaries of Google I suppose. > Is this a case of something being named after its mathematical/ > technical description and so obscuring its wider practical use cases? > Could be, I tried to make sure that the word dependency was in the description on the download page (since I had the same problem starting out (I implemented the algorithm before I knew the name IIRC)). > P.S. we have revived a thread started in 1999! > For some of us 1999 is well into our Pythonic life-cycle :) Have fun, Mike -- ________ Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: GUI development with 3D view
Diez B. Roggisch wrote: ... > I was under the impression that PyOpenGL is ctypes-based in the new > versions? > It is, but I haven't had the time to do a new release and check it on a Windows box. There are minor fixes in CVS that *should* IIRC make us run better on those Windows machines that have problems with the 3.x alphas so far. Tempus fugit, Mike -- ____ Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: Screen Shots?
katie smith wrote: > UGH so frustrating. In my game there is a minimap. On top of the > minimap is a window showing the part of the map your viewing. The > minimap is composed of other little pictures piled of top of eachother. > > I want to know how to basically take a picture from ex.((50,50),to > (150,150)) and blit the mini window image on top of it. hard to > explain but everytime I move the window box I don't want to redo the > window with its couple hundred compiled image. I need to copy picture > and make it into a surface??.. I dont even know. > > Any help would be greatly appreciated If you are using Pygame (you don't say which GUI library you are using), you can draw the mini-map into a surface and then re-blit the surface to the screen each time you want to display it. HTH, Mike -- ____ Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: dependency algorithm
Tom Jones wrote: > Hi, > > Consider tuples of the above numbers in the form: >(a,b) > > Suppose this relation means: >a depends on b > > Given a list of tuples, I would like an algorithm to return the proper > ordering of the elements...and if the ordering has a loop (which in this > case, prevents a proper ordering), then it should return None. > You want what's called a topological sort, see: http://blog.vrplumber.com/scripts/toposort.py for a pair of old algorithms from Tim Peters and I. I believe we raise errors on loops, but that's pretty trivial to change. Have fun, Mike -- ____ Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: TwistedMatrix missing OpenSSL?
Lamonte Harris wrote: > Where can I get it? Anyone got any idea? http://www.voidspace.org.uk/python/modules.shtml#pycrypto Might help. Mike > > Where do I get it? I'm currently running python 2.5 > > On 10/4/07, *Lamonte Harris* <[EMAIL PROTECTED] > <mailto:[EMAIL PROTECTED]>> wrote: -- ____ Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: Question: wxpython and 3d engine example with model load ?
OpenPavilion wrote: ... > I would like to create an application, which uses wxpython tree, menu > and grid elements and embedds a 3d view of some of the listed objects > in an own detail window, so showing the object as a 3d model. > I know there is PyOpenGL as well, but I need to load complete models > (meshes done with cinema4d or blender) into the 3d view. > There's a bit of sample code in OpenGLContext/bin (yes, yes, I know, part of PyOpenGL, which you don't want) that does a VRML-97-scenegraph-rendering window + a shell (you'll need the latest CVS version to work with modern wxPythons). There's commented out code that inserted a tree (to show the scenegraph as a tree, but I haven't got that finished). Other options: * I think Pivy (Coin/Inventor wrapper) could likely be embedded in PyQt (it has a Qt-based window operation IIRC) if that's an acceptable compromise. * http://www.python-ogre.org/wiki/DownloadsPage seems to suggest there's an example for using Ogre with wxPython/wxWidgets. * http://panda3d.org/phpbb2/viewtopic.php?=&p=11555 suggests there's a way to get panda3d and wxPython working together I don't see a soya + wxPython page in a few seconds of googling, so I guess finding that is left as an exercise for the reader. HTH, Mike -- Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: Call for advice on how to start PyOpenGL!
math2life wrote: > I work with python for two years, are familiar with image processing, > but beginner on PyOpenGL and OpenGL. > > Any advice appreciated! > As has been mentioned, the NeHe tutorials[1] are a good starting point. There are (multiple) translations of the first 8 or so and then some cherry-picked translations further along the set. There is a set of OpenGL demo/sample code available[2] as a separate download from the 3.x series (these files are mostly included in the 2.x series, though there is new material in the 3.x set). OpenGLContext also has a large number of simple samples, though I've still got "release OpenGLContext" on my seemingly infinitely long and growing todo list, so that might not be the easiest way to start. The "Red Book" is an older standard manual for getting started with (C) OpenGL code, often you can find older editions online for free. There are hundreds of C tutorials hanging around the net that normally translate easily into Python/PyOpenGL. The PyOpenGL manual pages *try* to have links to source code, but unfortunately the (java, ick!) based system that generates the base files has been broken for a while and I haven't had time to fix it properly so I can re-run the generator to update the links. In the meantime, Google's code search helps a lot when you restrict to Python code.[3] HTH, Mike [1] http://nehe.gamedev.net [2] http://sourceforge.net/project/showfiles.php?group_id=5988&package_id=221827 [3] http://www.google.com/codesearch?q=lang%3Apython -- ____ Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: Where do they tech Python officialy ?
NicolasG wrote: ... > I'm planning to save some money and attend a course in any of the > universities that teach hard core Python. > > Does some one have any suggestions on which University to attend ? > In Canada, the University of Toronto is planning to switch all first-year Comp-Sci courses to Python this September. Last I heard the University of Waterloo allowed Python submissions for most assignments in most courses. But those are "learn hard-core computer science using Python" solutions, not "learn hard-core Python" solutions. If you really want to learn hard-core Python, probably your best bet is: * read everything Tim Peters has ever written in comp.lang.python (this will take a few months), start with "import this" * read everything the PyPy guys have ever written (particularly Christian and Armin) * read and try to beat the more exotic recipes in the Python cookbook * read the papers from the various PyCons on metaclasses and the like, build a couple of dozen metaclasses and descriptors But jumping into "hardcore" first might not be the best approach. Many would suggest just learning "normal" Python first, *then* moving onto the hardcore stuff. HTH, Mike -- Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: Understanding Python's interpreter
Steve Holden wrote: > Rafael Almeida wrote: > ... > Because they aer smarter than you, without wishing to be too rude. > Replace that with "more experienced", please. Otherwise it is a bit rude, despite your wishes. We've always been newbie positive on Python-list, and new compiler writers are just as important educational targets as new Python programmers. Have fun, Mike -- ____ Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: PyDispatcher question
Jorgen Bodde wrote: > Hi all, > > Hopefully someone can help me. I am fairly new to Python, and I am > looking into PyDispatcher. I am familiar with the C++ sigslot variant, > and I wonder how similar PyDispatches is. I run in to the following > 'problem' (pseudo code, untested here) > Here's some real code... from pydispatch import dispatcher import gc class X(object): def __init__( self ): dispatcher.connect( self.someSignal, signal=1 ) def someSignal( self ): print 'hello world' obj = X() dispatcher.send( signal= 1 ) del obj #gc.collect() dispatcher.send( signal= 1 ) This will print out only one "hello world" on my Python 2.5 Gentoo machine (it should work the same on any recent Python). Basically your python shell will tend to keep around an extra copy of the X instance until you get rid of it explicitly, and that's what keeps the object "live" and receiving signals if you try the code in the shell. PyDispatcher is designed so that, by default, when the object goes away the registration is removed. It uses the weakref module to do this, rather than __del__ methods, to avoid garbage cycles, btw. HTH, Mike -- Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: interpreting glyph outlines from ttfquery?
Wojciech Muła wrote:
> swiftset wrote:
>
>> I'm try to convert a glyph into a format I can easily numerically
>> manipulate. So far I've figured out how to use ttfquery to get a list
>> that represents the outline of a contour in a glyph:
>>
>> from ttfquery import describe, glyphquery, glyph
>> f = describe.openFont("/usr/share/fonts/truetype/freefont/
>> FreeSans.ttf")
>> n = glyphquery.glyphName(f, 'D')
>> g = glyph.Glyph(n)
>> c = g.calculateContours(f)
>> o = glyph.decomposeOutline(c[1])
>>
>> o looks like:
>>
>> [array([182, 82],'s'),
>> (354, 82),
>> (420.29, 90.014),
>> (474.91, 114.0), ...,
>> array([182, 82],'s'),
>> array([182, 82],'s')]
>>
>> Is this a polyline?
>>
>
> decomposeOutline docstring confirms -- it's a polyline. I think
> elements marked with 's' starts new subpath.
>
> w.
>
Examples of rendering self.contours and self.outlines (created by
self.compile on the Glyph) using OpenGL operations are in the toolsfont
module in OpenGLContext:
http://pyopengl.cvs.sourceforge.net/pyopengl/OpenGLContext/scenegraph/text/toolsfont.py?view=markup
The outlines are simple 2D line-loops, the things with 's' are array
objects (not start coordinates), you can convert the whole thing to a
consistent array with numpy.array (or Numeric.array). That is, the 's'
is just an artefact of how the calculation was done, the values are all
2-element coordinates, some as Python tuples, some as 2-element
Numeric/numpy "short" arrays.
HTH,
Mike
--
Mike C. Fletcher
Designer, VR Plumber, Coder
http://www.vrplumber.com
http://blog.vrplumber.com
--
http://mail.python.org/mailman/listinfo/python-list
Re: Python GUI + OpenGL
Achim Domma wrote: > Hi, > > I'm developing a GUI app in Python/C++ to visualize numerical results. > Currently I'm using Python 2.4 with wx and PyOpenGLContext, but there > are no windows binaries for Python 2.5 for quite some time now. > > I need a OpenGL context without restrictions and some settings dialogs. > Is wx + PyOpenGL the way to go? Or could somebody recommend a better set > of tools/libs? > > regards, > Achim > PyOpenGL 3.x (currently in alpha state, but reasonably usable) works on Python 2.5, there are no binaries because the system no longer requires binary versions. Install the setuptools package, then run easy_install PyOpenGL and the egg file should be downloaded and installed to your machine. The current version doesn't package GLE along with the code, however, so you'll have to find a DLL for that if you need it. HTH, Mike -- Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: Dlls
Jason Ward wrote: > Hi. I am interested to know why python can't access DLL files directly. > It seems to me that because python can't access DLL's directly we have to > waste our time and write wrappers around libraries that have already > been written. > > So if the python developers were to implement say this. > > import MYDLL.dll > > Then we would be able to do everything with that library that we can > do in other languages. > > For eg. I want to use PyOpenGL. But the problem is the library hasn't > got all the opengl functions implemented. > So I can either develop quickly with an incomplete library or develop > slowly in assembler. > I mean really, in asm we just call the dll function directly. > > Why must python be different? Hi again, Jason, As we pointed out to you, ctypes *does* allow you to just load the DLL and start calling functions. In fact, that's how development tends to happen in the ctypes implementation. Someone just loads the DLL, hacks up a demo, I look at it and factor their code into the code-base. If there's some function missing that you need, do something like this: from OpenGL.platform import GL GL.myFunction( myParameter, myOtherParameter ) and if the parameters are of simple types, it will often "just work". If you want to pass arrays or similar data-types you'll need to make them ctypes arrays or pointers to use those "raw" functions, but they should work perfectly well. That is, you have to pass the right data-type, but then you'd have to do that in assembler too. Have fun, Mike -- Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: Saving PyOpenGl figures as ps
Frank wrote: > Hi, > > I installed pyopengl (opengl for python) on my linux box and > everything works fine. But now I want to save the generated images as, > e.g., ps or eps. How can I do that and how can I adjust the resolution > (if necessary)? This is probably simple but for some reason I can not > find out how to do that. > > I appreciate every hint! > > Thanks, Frank > Hi Frank, Take a look at the demos in OpenGL, particularly: OpenGL-ctypes/OpenGL/Demo/GLUT/tom/conesave.py which is a little GLUT application that renders into a GLUT windows and then allows you to save it. That's raster operation though (i.e. produces a PNG or JPG). Raster operations can only save a portion of the screen as rendered, so if you want a higher resolution image you need to make a larger window. If you want to get more advanced you can try using an off-screen buffer (pbuffer or MESA) for the rendering, which may (depending on implementation) allow for higher resolutions. If you want a true vector graphic (PS, EPS) you'll need to use something like GL2PS. http://www.geuz.org/gl2ps/ HTH, Mike -- ____ Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: pygame and python 2.5
[EMAIL PROTECTED] wrote: > Ben> Python extensions written in C require recompilation for each new > Ben> version of Python, due to Python limitations. > > Can you propose a means to eliminate this limitation? > Sure, write your wrapper-style extensions in ctypes :) . For example, pygame-ctypes[1] should work on Python 2.5. Of course, you need to get the PyGame dependencies (SDL) installed via some external mechanism, but the ctypes-based code should run in Python 2.5 today (with the caveat that it's not finished software). [1] http://www.pygame.org/ctypes/ Have fun, Mike -- ____ Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: Traceback of hanged process
Klaas wrote: > Hynek Hanke wrote: > >> Hello, >> >> please, how do I create a pythonic traceback from a python process that >> hangs and is not running in an interpreter that I executed manually >> or it is but doesn't react on CTRL-C etc? I'm trying to debug a server >> implemented in Python, so I need some analog of 'gdb attach' for C. >> ... > In python2.5, you can run a background thread that listens on a port or > unix socket, and prints a formatted version of sys._current_frames() to > stderr. > You can also use the signal module to similar effect. Works well in Twisted, at least: http://blog.vrplumber.com/835 HTH, Mike -- Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: Metaclass uses?
Nathan Harmston wrote: > Hi, > > Recently I posted a question about the factory pattern in python. I ve > implemented this following one of the ideas posted. After some reading > I ve seem a lot of Aspect Oriented Programming mentioned but I m not > really sure what it is. > > Can anyone help me understand how metaclasses can improve the quality > of my code and let me do better things. > My presentation from three pycons ago was an attempt to address this issue (what metaclasses are useful for): http://www.vrplumber.com/programming/metaclasses-pycon.pdf HTH, Mike -- ____ Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: pyopengl glShaderSourceARB error
joroy wrote: > Hi all, > > I think this is ctypes related but how can I call the glShaderSourceARB > function? > > The function have this header: > > glShaderSourceARB( GLhandleARB(shaderObj), GLsizei(count), > POINTER(arrays.GLcharARBArray)(string), GLintArray(length) ) -> None > > I call the function with someting like: glShaderSourceARB(self._object, > 1, sourceString, 1) > > The error is > "expected LP_GLcharArray instance instead of str" > > In fact I don't have any information on how to use this function. > This is the last version of pyopengl available on the CVS > (PyOpenGL-3.0.0a5-py2.5.egg) > Sorry about the lack of information available. There's a sample of usage here (known to work on Win32 and Linux with latest CVS HEAD): http://pyopengl.cvs.sourceforge.net/pyopengl/OpenGLContext/tests/shaderobjects.py?view=markup The key information you seem to be missing are that the Python version has a simpler API and that you have to pass an array (list) of strings, not just a single string to the compilation function. The count and length (array of lengths) parameters are pulled from the list-of-strings you pass. HTH, Mike -- Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: "fork and exit" needed?
Vincent Delporte wrote: > On Thu, 30 Nov 2006 15:38:11 +0100, Vincent Delporte <[EMAIL PROTECTED]> > wrote: > >> Here's my Python rewrite: >> http://codecomplete.free.fr/asterisk/python_cid.txt >> First line of your script should be: #! /usr/bin/env python which is how Linux knows what interpreter to use for the script. HTH, Mike -- ____ Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: 3D Vector Type Line-Drawing Program
James Stroud wrote: > James Stroud wrote: > >> Hello All, >> >> I'm looking for a program to do line-drawings in 3d, with output to >> postscript or svg or pdf, etc. I would like to describe a scene with >> certain 1-3d elements oriented in 3d space with dashed or colored lines >> and filled or transparent surfaces (or maybe semitransparent). >> >> I would then like to set the camera and have the scene depicted as a >> line drawing (not ray-traced solid body, etc). >> >> Does anyone know of a library to do this? >> > > > I'm really looking for a python library. The title of the post is kind > of misleading. > At one point I created a gl2ps wrapper for use with PyOpenGL, that, combined with any of the PyOpenGL-based renderers would allow much of what you're describing, though IIRC it didn't properly support transparent or textured surfaces. You could probably recreate the wrapper using ctypes in a few hours and then hook it up in a few more. I believe there are similar OpenGL-to-SVG libraries here and there, you could likely hook one of them up to get a reasonable result that would support more graphics features (maybe). Generally what you get out of these things, however, is a quite fragmented view of your objects (that is, you get the triangles that would be rendered). That *can* be coded around, but generally people are either focused on making it look good or making it editable. Good luck, Mike -- Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: does anybody earn a living programming in python?
walterbyrd wrote: > If so, I doubt there are many. > > I wonder why that is? > I've now used Python in every job I've had for the last 10 years. Started off with web-sites for a few months, then writing VRML-processing libraries to piece together and massage virtual worlds (not a *lot* of jobs there). After that worked on a piece of Cable-system SNMP monitoring software (again, not a *lot* of jobs in that). After that billing/provisioning systems for VoIP (you really only need one). The last two companies (one of which I own, one in which I was a partner) were python-only shops. PyGTA (Toronto Python User's Group), which is a fairly large user-group, seems to be about 60% (off-the-cuff estimate) paid Pythonistas, with some people picking it up for filling in corners and others spending all day working on it. The paid positions tend (like most programming positions in less-well-known general-purpose languages) to be programming + X for values of X which are a problem domain having an obscure nature. That is, boutique/bespoke development that requires some specialisation such that the customer is willing to go for whatever language is suitable and gets the job done. Regarding Java from one of the previous comments (and this is a cynical moment here, I'm really a far more positive person, just ignore this): There are still lots of jobs in Java, just as there are still lots of jobs in COBOL. Kids in the know (i.e. who *love* programming, rather than just wanting a job) going into university are minimising their Java exposure so that they don't box themselves into the big-old-company environments that have locked themselves into Java while everyone else moves onto the more dynamic languages/platforms. Job security and easy availability is not the be-all and end-all of happiness in life. That said, if you know anyone who "just wants a job", please, push them at Java, someone has to spend the next 30 years maintaining the Struts and J*EE sites people are putting together today in all those big institutions. Have fun, Mike ____ Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: simpleparse parsing problem
David Hirschfield wrote: > Anyone out there use simpleparse? If so, I have a problem that I can't > seem to solve...I need to be able to parse this line: > > """Cen2 = Cen(OUT, "Cep", "ies", wh, 544, (wh/ht));""" > ... > expr:= termlist, ( operator, termlist )+ > I *believe* you wanted that to be *, not + Best way to write a SimpleParse grammar is from the ground up, using test cases, so test each thing you think you're parsing. Makes it much easier to spot situations like this. HTH, Mike -- Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: Beginner Python OpenGL difficulties
[EMAIL PROTECTED] wrote: > I'm beginning learning Python and OpenGL in Python. > > Python fine. But difficulties with OpenGL; presumably with the > installation of OpenGL. > > OS = Linux FC5. > > Python program gl_test.py: > > from OpenGL.GLUT import * > from OpenGL.GLU import * > from OpenGL.GL import * > > name = "Hello, World" > height = 400 > etc. > > [EMAIL PROTECTED]/etc/python>$ python2 gl_test.py > > Traceback (most recent call last): > File "gl_test.py", line 1, in ? > from OpenGL.GLUT import * > ImportError: No module named OpenGL.GLUT > > [EMAIL PROTECTED]/etc/python>$ echo $PYTHONPATH > /usr/lib/python2.2/site-packages/OpenGL > You should already have site-packages in your PythonPath. You want the directory *above* OpenGL in the path, not OpenGL itself. I'm unsure why you're running a Python 2.2 instance on a modern Linux. I'd suspect that you're using an RPM for an older Linux? Not sure there, I run Gentoo, so everything builds from source for the packages that are installed. There are some Fedora Core build patches in CVS that are waiting for me to get my posterior in gear with Win32 testing to be released. Not sure if that would change anything for you, though. Good luck, Mike -- Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: Using python for a CAD program
. That said, threading is always tricky, particularly during debugging. If you can avoid it for most operations and only put things in threads that are largely disconnected from other processes you'll likely be happier. The big win would be if you are doing all of your simulation in C with an event-based update such that the simulation generates changes as a queued list of updates to the model. You could then have the simulation yield the interpreter lock and seldom block the rest of the process (save when it touches a Python object (i.e. the updates coming from or going to the queue)). Have fun, Mike -- Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: State of the "Vaults of Parnassus"
Christoph Zwerschke wrote: > Does anybody know what happened to the "Vaults of Parnassus" site at > http://www.vex.net/~x/parnassus/? > > "Dr Dobb's weekly Python news" still claims that it "ambitiously > collects references to all sorts of Python resources." But I have the > impression that it does not do this any more since April 2005, probably > because we have the Cheese Shop now. It's strange that the site does not > mention that it is not maintained any more, and it is still possible to > submit and update entries, but these changes seem to be discarded. > That's not very nice. Also, I found no contact address on the whole > website. Who is in charge of it? > > -- Christoph > The Vaults are run by one of my colleagues, who is (as it seems we all are), rather overworked these days. Last I heard he had someone who was helping him every once in a while to review the new submissions, but that was quite a while ago. I don't *think* there's any discarding, but rather a delay so long in publishing that it would feel like a discard... Take care, Mike -- Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: Can I control Video Card by using Python under linux?
LUK wrote: > Hi, I have a video card based on cx2388 chip to catch video and do the > other thing. > There's already a V4L2 driver for it, but it is too hard for me to > program in C. > Can I use Python do the job? > Does Python has simpler APIs? > Please gvie me some suggestion. > Thanks! > You are probably looking for the "gstreamer" package. It has Python bindings available. HTH, Mike -- ____ Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: Cheese Shop: some history for the new-comers
A.M. Kuchling wrote: > On Sat, 11 Mar 2006 16:50:26 +1100, > richard <[EMAIL PROTECTED]> wrote: > >> So I did what people always do in this situation, I asked Barry Warsaw to >> name. it. And he did, "Cheese Shop". I liked the name, so it was done. When >> the new pydotorg machines went live last year, so too did the name >> cheeseshop.python.org >> > > Given the endless whiny complaints about the name, though, I think we > should just give up and go back to PyPI (pronounced 'Pippy'). > I realise it's an incredibly boring name, but what about pronouncing it as "the package index" or "the Python Package Index". Spelling it "PyPI" if one insists on a very short name in some written context is fine, but try not to use it when talking to a new user. That is, link to the page as "The Package Index" from www.python.org, refer to it as "you can find that in the package index", or "The Python Package Index has modules for X, Y, and Z" in a non-Python context. Use pippy if you're talking with someone deeply embedded in Python culture (maybe), but don't expect people to know what you're talking about. Use "the Cheese Shop" similarly. Luckily, www.python.org already does this, so no change is really required, it's only the server-name that says "cheeseshop", everything else says "Package Index" (that I could find)... Which boils down to "don't call it the cheese shop or pippy when you're talking to new users", maybe don't even call it that when you're talking to *anyone* who's not deeply pythonified, but feel free to think of it as the cheese shop or pippy. The familiar names, cute as they may be, don't really help users much. That is, make it an in-joke if you like, but assume that users will refer to it via the obvious, simple description of what it is and target documentation and communication appropriately. When/if the user "gets" that the Package Index is the Cheese Shop they can feel great having pierced that bubble, but don't force them to pierce it to get the software they need. Just my two cents, Mike -- Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: Speed of data structures in python
Dave wrote: > Hi. I am learning PyOpenGL and I am working with a largish fixed scene > composed of several thousand GLtriangles. I plan to store the coords and > normals in a NumPy array. > > Is this the fastest solution in python? would i be significantly better > off (timewise or otherwise) using some other data structure to hold this > information? no other data structure appears as convenient, but am i > loosing valuable time? > > I am new to 3d programming so please feel free to state the obvious, > thanks. > There are two major approaches you can take: * array-based geometry -- this relies on fast memory-transfer rates operating a C speeds * display-list-based geometry -- this relies on moving all of the data over to the card If you can reduce your entire (static) scene to a single display-list call, you'll likely find that it's faster than using array-based geometry. If you have special effects (e.g. transparency), then you'll likely find that coding as array-based geometry is the more practical approach. By way of reference, OpenGLContext uses array-based geometry throughout these days. I had (at one point) code to make individual display-lists out of particular indexed face sets, but it didn't provide any performance boost when applied at the single-object level (i.e. thousands of display lists with all the traversal code still required), so the code is now disabled. Simpler (static) scenegraphs where there's lots of traversal between nodes should be a much bigger win for converting the whole thing to a single display list. HTH, Mike -- Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: Collecting snmp data - threads, processes, twisted?
Antal Rutz wrote: ... > I'd like to collect snmp data from varoius network devices parallel. > ... > Could you give me some advice how can I make my script really parallel? > > My options maybe: > 1. pySNMP (as it's full in python) > 2. subprocess (I'd like to make (find) a solution as threadpool.py) > (running yapsnmp, pySNMP or even net-snmp?...) > 3. TwistedSNMP (seems too weird for me.. cannot imagine a simple > solution for my problem) > This kind of parallel mass-query operation is TwistedSNMP's primary focus (it was created for a commercial product that scans thousands of agents). If you really want all of the queries to run in parallel (i.e. send out all the queries within milliseconds of each other): agents = [ ... ] dl = [] for agent in agents: dl.append( agent.get( (oid,oid,oid) ) ) return defer.DeferredList( dl ) add a callback to the DeferredList which processes the (success,result_or_failure) list you get back. If you want to process individual results/failures before the whole set returns, add the callback for that to the agent.get() result. If you have thousands of devices you'll likely want to batch the queries into ~ 200 simultaneous queries and/or use multiple protocols (gives you more OS-level buffering). Have fun, Mike Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: OpenGL
[EMAIL PROTECTED] wrote: > Does that mean PyOpenGL is based on ctypes ? > PyOpenGL, the project with releases, binaries on most platforms, and lots of history is currently written in SWIG. It was originally written in pure Python C-extension code, then Tarn rewrote it using SWIG a few years ago. OpenGL-ctypes is in the PyOpenGL CVS repository, it's a ctypes re-re-implementation of the same Python API to OpenGL. My intention is that the ctypes implementation will become the 3.0.0 release when it is finished and optimised. If there's time and enough interest I'll try to produce another maintenance release for PyOpenGL-SWIG, but I'm really quite tired of spelunking through dozens of levels of C-code macro expansions, so I'm not all that interested in working with the codebase. > I thought is was using SWIG ? > And on that note: I have some (old) SWIG typemaps for numarray arrays > that I'm using for many years. > > I was always wondering how difficult it would be for me to add them > into PyOpenGL ? > Likely difficult, for the SWIG version. You might be able to *substitute* them quite readily for the Numpy mapping, but having the APIs support two different numeric systems requires some API to switch between them, which requires rewriting the (really-deeply-embedded) typemaps to support that. The difficulty of doing that kind of change is part of what prompted me to rewrite the system in Python (using ctypes). > So far I'm just using my own (one !) OpenGL function with that typemap > to draw 6 vertices in a vector linestrip. But I would like to not > have to resort to "string conversion" when for example drawing 2D > typemaps... > > Anyway, thanks a lot for PytOpenGL - my program (image analysis > platform and microscope control) would not be possible without it. > Sebastian Haase > OpenGL-ctypes is designed with a fairly well abstracted array-handling API. Basically any array type can be registered with handlers that let you tell the system how to do basic operations; get an array's size, shape, data-type, convert to a given data-format, build an "empty" array of a given size, that kind of thing. I haven't written a numarray one yet, but it should be a fairly trivial change from the Numpy/scipy-core module. Have fun, Mike -- Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: OpenGL
NaeRey wrote: > If you notice the project died, being latest release Jan2 2005. Thats > over a year old. > Died, no. We have a rather small developer community, basically myself and a few people who contribute a module here or there. Most (really, all) of the time being spent on PyOpenGL these days is focusing on the OpenGL-ctypes implementation, which hopefully will let far more people work on the source-code (and add features and coverage of more extensions). Have fun, Mike -- ____ Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: Extracting results from a large hotshot profile
Brian Cole wrote: >Thanks for the tip. I got the following error message when trying to >run your profiler. > > ... > File "c:\Documents and Settings\coleb2\My > Documents\software\Python24\Lib\site >-packages\runsnakerun\hotshotreader.py", line 95, in loadHotshot >localDeltas[depth] = 0 >IndexError: index out of bounds > > ... >As the traceback shows I'm running this on Windows. I'm gonna go >wrestle with installing wxPython and Numeric on my Linux box and see >if that makes a difference. > > Not likely. The hotshotreader module is creating an array of sys.getrecursionlimit() long integers to store the accumulated local times. It does the same to track the number of frames. Most likely your main program is setting sys.getrecursionlimit explicitly and as a result it winds up with a larger possible stack. You can try replacing the sys.getrecursionlimit() calls with some big number (say 40,000, or just multiply the recursion limit by 10 or 20 if you like) and see if that solves the problem. Probably should allow for setting the value on the command-line I guess. I suppose there may also be cases where a recursive call to the interpreter gets a whole separate stack... or maybe there's something going on with threads intermixing to create greater depth. Let me know if that helps, Mike -- Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: Extracting results from a large hotshot profile
Brian Cole wrote: ... >I did a hotshot profile. To make this profile took >about a day. However, I've been trying to extract results from the >profile for the past two days. It's eating up all the memory (2G >memory) on the machine and is slowing sucking up the swap space >(another 2G). > >I'm hesitant to kill it for I might get results out of it some day >(even if I may have to bequeath them to my grand-children), but is >there a better way to do this? > > ... >The profile is 355 MB large. Is there anything I can do to extract that data? > > You might want to try my RunSnakeRun utility: http://www.vrplumber.com/programming/runsnakerun/ It's designed to load large hotspot runs. I've never had a 355 MB one, but I regularly load 140+MB ones on a desktop 1GB machine. It does an incremental load, so you start to see results as soon as it starts running and continues loading the data-set in the background (it can take quite a while for the full data-set to load, but I don't *think* it should run out of RAM). In case you don't want the GUI, the package includes the core hotshotreader module that actually does the incremental loading, you could use that to produce a view by writing code to display the results statically. HTH, Mike -- Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: Limiting the size of List and making it Homogeneous
Little class that does restrictions/checks on adding to a list (simply subclass and put your restrictions in the beforeAdd or beforeMultipleAdd methods). http://pytable.sourceforge.net/pydoc/basictypes.rlist.html http://cvs.sourceforge.net/viewcvs.py/basicproperty/properties/basictypes/rlist.py?view=markup Note, however, that if you goal is to create a simple C-level pointer of machine ints you'll need to use Numpy or a similar system that implements such a type. Have fun, Mike ankit wrote: >Is it possible to limit the size of list in python. >I want to make list of 5 elements. How can I achieve this thing in >python. And one more thing can we declare list to store elements of >same type as in c, C++ we can declare an >array which can have 5 elements of type int. >C, C++: > int intarr[5] >How can I achieve this kind of behavior ? > >Thanks > > -- ____ Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: PyOpengl text rendering with autocad font
Marco wrote: >I must use text in Opengl, Python >This text is for architecture design then I need to use AutoCad fonts >(extension .shx). >I have two troubles: > >I don't find a guide, a good program, something for use any kind of >fonts in PyOpengl. The nehe tutorial has some bugs in translation to >Python. >I need to scale rotate the text. > > For general font rendering you can choose two major approaches, either use a built-in platform-Dependant system (such as seen in the NeHe tutorial) or code your own to render the individual characters using either one of the cross-platform GUI libraries (2D text) or TTFQuery (3D/polygonal text (i.e. you can rotate in all 3 dimensions)). There is a NeHe tutorial translation for the font-rendering demo in OpenGLContext's tests directory. To render text in general, see the OpenGLContext/scenegraph/text/* modules. However, none of those knows how to handle .shx files, they load TTF fonts instead. >I don't know anything about handling fonts, I must use this .shx but I >don't know. In which way can I take this file and put them in my >programs? > > If you can find a format specification you may be able to write an importer (IIRC SHX fonts are just series of lines, so would be very easy to render once you have the format read). You might find it easier to redefine the process as "tell autocad to use truetype, then render truetype" if whoever is asking for .shx will allow it. >If somebody wants to help me :-) >Thanks. > > HTH, Mike -- Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: new in programing
Python iterates over "things" (objects), of which integer numbers are
just one possible choice. The range built-in command produces ranges of
integers which are useful for tasks such as this.
lim = 3
for i in range( 1, lim+1 ):
for j in range( i+1, lim+2):
for k in range( j+1, lim+3):
for l in range( k+1, lim+4):
for m in range( l+1, lim+5):
for n in range( m+1, lim+6):
print i,j,k,l,m,n
Would be a direct translation of your code (with a few lines to make it
actually do something and a fix for the last variable name).
HTH,
Mike
Efrain Marrero wrote:
>i want to now how to do this in python
>this is java
>
>
>for(int i=1 ; i<=lim ; i++){
>
> for(int j=i+1; j<=lim+1; j++){
>
>
...
--
____
Mike C. Fletcher
Designer, VR Plumber, Coder
http://www.vrplumber.com
http://blog.vrplumber.com
--
http://mail.python.org/mailman/listinfo/python-list
Re: Favorite flavor of Linux? (for python or anything else)
Brett Hoerner wrote: >I have to add another vote for Gentoo. > And another here. Portage (the Python-coded package-management system) does a very good job. I tend to use fairly conservative settings as well, Gentoo's just nice and stable as a general rule, I don't care about ultimate speed or tweaking the code much, I just find that building everything from source (but without the headaches) tends to make things work together well. As far as Python support goes, Gentoo tends to work very well for Python packages, Pygame and wxPython on AMD64, for instance, install flawlessly out of the box (where others seem to have problems)... in fact most of the major packages install reliably. There's a key note that you might miss regarding the Python updating process, namely that you have to run python-updater to rebuild all of your packages for the new Python version, but once you know that, Pythonicity in Gentoo is pretty straightforward. Anyway, just a vote that's trying very hard to keep on-topic for the python list, Mike -- ____ Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: 3d Transformation
Tyler Eaves wrote: >Got a 3d algorithmns question. Algorithmn will be implemented in python, >so a library would be great if it works on windows. Basically my function >would be defined as such. > >Problem is one of coordinate transformation. Give a 3d vector, which >reprents the +z axis in the source coordinate system, and a rotation value >around that axis for determinging the x and y axises, I wish to translate >a point from that geometry into "world" geometry, or in other words >absolute coordinates. Input data is a 3d path with banking data that will >be sampled at regular points for purposes of building other geometry. > > You can find code for doing this in the OpenGLContext project. The lowest-level code is in the vrml.vrml97.transformmatrix module, which just generates matrices for the various transformations, rotations and scales. You can find the module here: http://cvs.sourceforge.net/viewcvs.py/pyopengl/vrml/vrml97/transformmatrix.py?view=markup the module also allows for generating inverse arrays (what you use for transforming from outer coordinates to inner, instead of inner to outer), and has an accelerator C module for much of the functionality. The higher levels in OpenGLContext support compositing multiple transformations in a containment hierarchy for a scenegraph. See e.g. OpenGLContext/scenegraph/transform and OpenGLContext/scenegraph/nodepath module for that higher-level stuff. Anyway, hope this helps, Mike -- ____ Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: System tray Icon
I have a sample up here: http://blog.vrplumber.com/scripts/recordingdevices.py using wxPython. The sample application is Linux-specific, but it should give you a fairly good idea of how to use the system tray icon control in wxPython. HTH, Mike Mike Pippin wrote: > How would I have an app run with just a system tray Icon??? any help > would be greatly appreciated. I have no clue where to start. -- ____ Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: graphic memory & animation
Peres wrote: > Thanks a lot for your answer, Fredrik, > Slow means more than 20ms to erase the screen. After double buffering > it improved a lot , of course (16 ms) but I'll need a faster speed. > I program 2D animated sequences on a PC. Do you think OpenGL is the > correct direction to take? If so is it easy to interface with Python > (it seems PyOpenGL is not freeware...) PyOpenGL has a BSD-style license, that's about as 'free' as one can get in the freeware sense. Just a note, Mike -- ____ Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: Merging sorted lists/iterators/generators into one stream of values...
Alex Martelli wrote: >George Sakkis <[EMAIL PROTECTED]> wrote: > ... > > >>>manipulation of a heap to place an item in the right spot, but with 4-5 >>>or a few more sources might not make an impact at all. >>> >>> >>Unless you're talking about hundreds or thousands sources, it probably >>won't. I would still go for the heap solution since IMO the resulting >>code it's more readable and easier to understand. >> >> ... >i.e., a heap solution may be over 4 times faster than a sort-based one >(in the following implementations). Readability seems quite comparable >(skipping the rest of the infrastructure, which generates random sorted >streams and ensures a stream is exhausted and verifies it etc etc): > > One thing to keep in mind (if you care about performance) is that you one could use bisect, instead of sort, as the sorted list of streams is already in order save for the one element you are processing. Btw, nice trick with reverse to reduce memory copying, when did that get introduced? Wonder if bisect can deal with reverse-sorted elements. Anyway, should reduce the O-complexity of that part of the operation, though you'll still have to do a memcpy to shift the rest of the source list's array, and if it can't deal with reverse-sorted lists it would move you back to front-of-list popping. Oh, we're still missing use of a comparison function in both versions. I'd think you'd want that functionality *available* if you're going to make this a general tool. You'll also need to check for StopIteration on creation of sources for null sequences. Finally, why the 'i' element? It's never used AFAICS. >def merge_by_sort(streams): > sources = [[s.next(), i, s.next] for i, s in enumerate(streams)] > while sources: >sources.sort(reverse=True) >best_source = sources[-1] > yield best_source[0] >try: best_source[0] = best_source[-1]() >except StopIteration: sources.pop() > > ... Have fun, Mike -- Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: Merging sorted lists/iterators/generators into one stream of values...
Lasse Vågsæther Karlsen wrote:
>Ok, that one looks more sleak than what I came up with.
>
>Couple of things I learn from your solution, please correct me if I
>misunderstood something:
>
>1. list containing other lists will sort itself based on first element
>on lists inside ?
>
>
Correct, comparison is recursive for lists/tuples.
>2. sort(), pop() is not costly operations
>
>
They *can* be costly, but the algorithm reduces the number of times they
are called to less-than-linear for all but pathological cases (i.e. they
are only called when you need to switch streams). It also only requires
sorting only the number of streams, rather than the whole set.
>Other than that you seem to not use the cmp operator but that's easily
>fixed.
>
>
Sorry about that, revised version below:
def firstIter( value ):
it = iter( value )
try:
return it.next(), it
except StopIteration, err:
return None, None
def cmpFirstWith( comparison ):
def compare( a,b ):
return comparison(a[0],b[0])
return compare
def inorder( comparison, *args ):
iterators = [
[value,it]
for (value,it) in [ firstIter( arg ) for arg in args ]
if it is not None
]
iterators.sort()
while iterators:
yield iterators[0][0]
try:
value = iterators[0][0] = iterators[0][1].next()
except StopIteration, err:
iterators.pop(0)
else:
if len(iterators) > 1 and comparison( value,
iterators[1][0]) == 1:
iterators.sort( cmpFirstWith( comparison ) )
continue
if __name__ == "__main__":
s1 = [10, 20, 30, 40, 50]
s2 = [15, 25]
s3 = [17, 27, 37]
s4 = []
for value in inorder(cmp, s1, s2, s3, s4):
print value
s1 = [{'a':'b'}, {'a':'e'}]
s2 = [{'a':'d'}, {'a':'z'}]
def comp( a,b ):
return cmp( a['a'],b['a'])
for value in inorder(cmp, s1, s2 ):
print value
Have fun,
Mike
--
Mike C. Fletcher
Designer, VR Plumber, Coder
http://www.vrplumber.com
http://blog.vrplumber.com
--
http://mail.python.org/mailman/listinfo/python-list
Re: Merging sorted lists/iterators/generators into one stream of values...
Lasse Vågsæther Karlsen wrote:
>I need to merge several sources of values into one stream of values. All
>of the sources are sorted already and I need to retrieve the values from
>them all in sorted order.
>
>In other words:
>s1 = [10, 20, 30, 40, 50]
>s2 = [15, 25]
>s3 = [17, 27, 37]
>
>for value in ???(s1, s2, s3):
> print value
>
>will print out 10, 15, 17, 20, 25, 27, 30, 37, 40, 50 in that order.
>
>The sources are cursors retrieving data from several databases, not from
>the same server, and there's a potential for a large number of rows from
>several of the sources. As such, any method that would load it all into
>memory and sort it is no good as it would too much memory.
>
>Is there a function or whatnot in Python that will do what I want? I
>have come up with my own method but since it uses generators I'm not
>sure how much overhead there is. Additionally, since the values
>retrieved from the cursors will be dictionaries of "fieldname":value
>pairs, the method would either need to handle that construct (and be
>told what fieldname to sort on), or be able to take a function object to
>use for the comparison operation.
>
>Basically, I'll use my own function for this unless someone can point me
>to something that is already available. Couldn't seem to find anything
>in the builtin modules but I didn't find glob.glob when I was looking
>for how to find filenames either so who knows what it's called :)
>
>Since I need to deal with a variable list of sources (that is, 2 in one
>application, 3 in another and 4-5 in a third), a simple 2-source method
>isn't enough but if it's better than what I do for 2 sources then I can
>make a wrapper for it, since that's what I do anyway.
>
>
I doubt you'll find a prebuilt one, it's fairly easy to create your own,
after all. Generators are fairly fast constructs in Python, btw.
Here's what I whipped up in a few minutes:
def firstIter( value ):
it = iter( value )
try:
return it.next(), it
except StopIteration, err:
return None, None
def inorder( comparision, *args ):
iterators = [
[value,it]
for (value,it) in [ firstIter( arg ) for arg in args ]
if it is not None
]
iterators.sort()
while iterators:
yield iterators[0][0]
try:
value = iterators[0][0] = iterators[0][1].next()
except StopIteration, err:
iterators.pop(0)
else:
if len(iterators) > 1 and comparision( value,
iterators[1][0]) == 1:
iterators.sort()
continue
if __name__ == "__main__":
s1 = [10, 20, 30, 40, 50]
s2 = [15, 25]
s3 = [17, 27, 37]
s4 = []
for value in inorder(cmp, s1, s2, s3, s4):
print value
Anyway, have fun,
Mike
--
Mike C. Fletcher
Designer, VR Plumber, Coder
http://www.vrplumber.com
http://blog.vrplumber.com
--
http://mail.python.org/mailman/listinfo/python-list
Re: Teenagers of busses do *what*?
Steven D'Aprano wrote: ... >He is the equivalent of one of those bored, spoiled teenagers who urinate >on public transport just to see the shocked reactions of other people. You >can't engage him in rational debate. Until we find a way to send electric >shocks through the Internet, all we can do is ignore him. To argue with >him just gives him the sick entertainment he wants. > > Rational debate? They urinate on your buses and you worry about rational debate? Call the transit police. Have them hauled home in a paddy wagon to say hello to their parents after a few dozen hours of picking up litter and scrubbing bus floors. We may not be able to do anything particularly effective to the morons who light *online* fires; but then they are easily ignored when their signal/noise ratio drops to nil. But if you're seeing this kind of behaviour in the real world, for criminney's sake, fix it with the judicious use of a billy club upside the head. No society thrives while letting people piss in the common well. The public transit system is a commons to be protected like any other. If you see people pissing on your public transit, get scared and fix your society. Entirely off-topic, Mike -- ____ Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: py2exe and OpenGL problem
Line 13:17 of OpenGL.__init__.py, replace with: try: filename = os.path.join(os.path.dirname(__file__), 'version') __version__ = string.strip(open(filename).read()) except Exception, err: __version__ = '2.0.2.02' HTH, Mike [EMAIL PROTECTED] wrote: >I am using the new py2exe and python 24. When I run my setup.py, the >dist seems to generate just fine. However run I run the resulting exe, >I get an error: > >IOError: [Errno 2] No such file or directory >'c:\\app\\dist\\library.zip\\OpenGL\\version' > >Now I've tried the recommended route with older py2exes, where I >exclude OpenGL and copy the entire directory over to my dist, but that >didn't work. I included "version" into the "OpenGL" directory via my >setup.py script, and it is there right in the library.zip, but I still >get the same error. > >Anyone have any ideas or suggections? > > -- Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: improvements for the logging package
[EMAIL PROTECTED] wrote: >Vinay> Well, it seems a little too late now, for unittest, threading, > logging >Vinay> and probably a few more. > >Correct, as I indicated. > > ... >Vinay> What would you suggest for threading, unittest etc. in terms of >Vinay> binding more unix_like_names and deprecating existing ones? It >Vinay> seems a lot of work for not very much benefit, beyond consistency >Vinay> for its own sake. > >No, while "consistency is the hobgoblin of little minds" I think we should >strive for consistency within the core code, simply because it tends to be >used as an informal educational tool. As for the others (and logging as >well, for that matter), maybe it's just been too long to make changes and >they should only be changed in Py3K. > > I would strongly support the assertion that changing the interfaces on deployed modules just to match the style guide is not something that should happen before Py3K (that is to say, it should not happen until we simply don't care about being able to run old Python code, which is to say, a very long time). There's no reason whatsoever to break *every* piece of code that ever used a service just to support consistency in naming. Especially when you're dealing with these modules that are used in probably the majority of projects it's just a *huge* make-work project to go removing the original service (API). If someone really wants the newly-named version, create a different entry-point (module) and allow them to use that. See how popular the renamed version is as a stand-alone module where the barrier to entry is copying it in (which is far less than revising hundreds of projects to deal with a deprecation). If it flies, consider adding it to the standard library and eventually deprecating the current version. Consistency in the standard library is good. We *should* strive for it as we improve the library. But reworking elements in the standard library *just* to support consistency, at the expense of complete API incompatibility, just doesn't have a good cost/benefit ratio for the *users* of Python. A note in the docstring and documentation noting the inconsistency in naming due to historic factors would seem to be more appropriate if we're worried about emulation by users. Anyway, I realise Skip probably was using "in Py3K" in the "some unimaginably far-off time" sense, but just in case he wasn't I felt I should pipe up... Mike -- Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: Has anybody tried to make a navigation like Google?
Lad wrote:
>Hi,
>I would like to make in my web application a similar navigation like
>Google uses, I mean at the bottom of each page there are numbers of
>returned pages after you search query.
>Has anyone tried that? Is there an algorithm for that? I do not want to
>re-invent the wheel.
>Thank you for help
>La.
>
>
First the assumptions:
* We'll assume you're using SQL queries against a database
* You don't want to have to produce the whole result-set across the
database interface, but you don't mind producing them on the
database side
* You want precise values (Google's aren't precise measures of
number of records)
o If you don't, you can use whatever estimation mechanism you
like to get the total instead of a second query
* You have a complete ordering of records (if not then your results
will randomly shuffle themselves as the user pages through)
Given that:
* You need to construct a query that produces (just) a count of all
records in the set
* You then need a query that is parameterised to return a subset of
the records in the set
o offset -- count of records from the start of the set
o limit -- number of records to display on any given page
Now, when you want to retrieve the records:
def search( self ):
"""Perform the search, retrieve records and total"""
self.records = self.SEARCH_QUERY(
connection,
whatToSelect = self.whatToSelectSubset,
limits = self.calculateLimits(),
orders = self.calculateOrders(),
wheres = self.calculateWheres(),
**self.queryParameters
)
for record in self.SEARCH_QUERY(
connection,
whatToSelect = self.whatToSelectCount,
limits = "",
orders = "",
wheres = self.calculateWheres(),
**self.queryParameters
):
self.total = record[0]
In that code the SEARCH_QUERY is a PyTable SQLQuery object, it just
takes care of the cross-database substitution operations. The code to
handle the whatToSelect determination looks like this (by default):
keyColumnName = common.StringProperty(
"keyColumnName", """Column name used to count total number of
columns""",
#define keyColumnName on Search!,
)
whatToSelectSubset = common.StringProperty(
"whatToSelectSubset", """What to select in the subset-of-records
query""",
defaultValue = "*",
)
whatToSelectCount = common.StringProperty(
"whatToSelectCount", """What to select in the count-of-records
query""",
defaultFunction = lambda prop,client: """COUNT(DISTINCT( %s
))"""%(client.keyColumnName),
)
the wheres are the natural WHERE clauses you want to apply to the query
(i.e. the search terms). The orders are normally a default set of
fields with the ability for the user to move any field to the front of
the set via UI interactions. The "limits" are actually both limits and
orders in this example, since they are tied together as the paging
functionality:
def calculateLimits( self ):
"""Calculate the limit/offset clauses for a select"""
return """LIMIT %s OFFSET %s"""%( self.limit, self.offset )
Just for completeness, here's an example of a SEARCH_QUERY:
SEARCH_QUERY = sqlquery.SQLQuery("""SELECT
%(whatToSelect)s
FROM
voip.voicemail JOIN voip.voipfeature USING (voipfeature_id)
JOIN voip.voipaccount USING (voipaccount_id)
%(wheres)s
%(orders)s
%(limits)s
""")
The UI then offers the user the ability to increase offset (page
forward), decrease offset (page backward), and re-sort (change the
ordering fields). You disable the paging if you've reached either end
of the record-set (offset<0 offset >= total-1), obviously.
Anyway, hope that helps,
Mike
--
Mike C. Fletcher
Designer, VR Plumber, Coder
http://www.vrplumber.com
http://blog.vrplumber.com
--
http://mail.python.org/mailman/listinfo/python-list
Re: File parser
infidel wrote: >Angelic Devil wrote: > > ... >Some time ago I was toying around with writing a tool in python to >parse our VB6 code (the original idea was to write our own .NET >conversion tool because the Wizard that comes with VS.NET sucks hard on >some things). I tried various parsing tools and EBNF grammars but VB6 >isn't really an EBNF-esque syntax in all cases, so I needed something >else. > ... You may find this project interesting to play with: http://vb2py.sourceforge.net/index.html Have fun, Mike -- ____ Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: Pointers and ctypes
F. Petitjean wrote: >Le 29 Aug 2005 06:19:17 -0700, [EMAIL PROTECTED] a écrit : > > ... >You can use the ctypes.byref() function (as it is in an argulent list): > > ReadOPort.argtypes = (c_ushort, c_ushort, ctypes.POINTER(c_ulong) ) > ReadOPort.restype = c_short > status = c_ulong() # status value to be read > number = c_ushort(1) # CardNumber = 1 > port = c_ushort(11) > rc = ReadOPort(number, port, ctypes.byref(status)) > print rc, ststus > > Just as an FYI, Thomas has recently added code which does the byref automatically. The version of ctypes in CVS HEAD will allow you to pass in a c_ulong where the argtype is ctypes.POINTER(c_ulong). Have fun, Mike -- ____ Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: python image thumbnail generator?
Chris Dewin wrote: >Hi. I run a website for my band, and the other guys want an image gallery. > >I'm thinking it would be nice and easy, if we could just upload a jpg into >a dir called "gallery/". When the client clicks the "gallery" link, a >cgi script could search the gallery/ dir, and create thumbnails of any >jpeg images that don't already have a thumbnail associated with them. The >script could then generate a page of clickable thumbnails. > >A few questions: > >(1) Can this be done with python? If so, what module do I need to look up? > > PIL can certainly handle the functionality. >(2) If it can't be done with python, is there some small utility that will >do it for me? Something I could easily install locally in my own >"public_html" dir on my website? > > The core function looks something like this: import Image # this is PIL def getThumbnail( filename, size = (32,32) ): '''Get a thumbnail image of filename''' image = Image.open(filename) rx, ry = image.size[0]/float(size[0]), image.size[1]/float(size[1]) if rx > ry: resize = int(size[0]), int(round(image.size[1]*(1.0/rx), 0)) else: resize = int(round(image.size[0]*(1.0/ry), 0)), int(size[1]) image = image.resize( resize, Image.BILINEAR ) newimage = Image.new( 'RGB', size, ) x,y = image.size newimage.paste( image, ((size[0]-x)/2, (size[1]-y)/2), ) return newimage then you call newImage.save( filename, format ) to save the result. >(3) Is this the sort of thing which, if done regularly, would hog far far >too much of my webhosts system resources? > > As long as you *only* generate images for sources that don't yet have (or are newer than) their thumbnail you should be fine. >(4) Am I talking about re inventing the wheel? > > Probably. ZPhotoSlides (a Zope product) provides thumnailing, slideshows and the like, probably closer to what you'd want for this kind of application. You'll probably find other non-Zope versions as well. Good luck, Mike -- Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: Decorator and Metaclasses Documentation
bruno modulix wrote: >Mike C. Fletcher wrote: >(snip) > > >>Though the don't go into extreme detail on decorators (they are >>basically syntactic sugar for a particular type of descriptor). >> >> >> >Err... Could you elaborate on this ? Decorators are syntactic sugar for >function wrapping, while descriptors are a 'protocol' to hook into >attribute lookup, so I don't really see how one can describe the first >in terms of the second... > > There are two major types of descriptors, the elven and the dwarven. * Elven descriptors are things like property, BasicProperty, VRML97 fields, Zope Field Properties, or PEAK's mechanisms. They mediate access to an instance's attributes, (both setting and getting of values) and are used primarily for domain modeling. * Dwarven descriptors are things like staticmethod or classmethod; function-like things that tend to look like a function and quack like a function, but have some special property or functionality attached when accessed as an attribute of a class/instance (functions themselves return instance methods or class methods depending on how they are retrieved). As you'll read in the PyCon 2005 paper pointed to, the whole set basically grew out of a generalisation of what functions were already doing (conceptually). Functions are, in short, the proto-descriptor. That is, objects which control what result is returned by attribute access (e.g. an instance or unbound instance method). Many (most?) dwarven descriptors are implemented as simple wrapper functions around the base function to use the function's already-present descriptor hooks. Decorators are syntactic sugar for defining dwarven descriptors. In Python 2.2 and 2.3 a dwarven descriptor was instantiated like this: def function( cls ): """Do something simple with the class""" function = classmethod( function ) in Python 2.4+, they can be instantiated like this: @classmethod def function( cls ): """Do something simple with the class""" technically you *could* return something that's not a descriptor from the decorator (e.g. classmethod), but it's unlikely you'd do that very often. HTH, Mike -- Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: Decorator and Metaclasses Documentation
sysfault wrote: >Does anyone know of any good documentation on these topics, doesn't look >like the official python tutorial covers them. Thanks in advance. > > PyCon 2005, PyCon 2004 and PyGTA November 2004 presentations here are all on these topics: http://www.vrplumber.com/programming/ Though the don't go into extreme detail on decorators (they are basically syntactic sugar for a particular type of descriptor). HTH, Mike -- ____ Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: PyOpenGL
[EMAIL PROTECTED] wrote: >Hey I'm a programmer looking to port some of my opengl ...although >limited into a python app I've made... I'd like to know where to find >any python/opengl source or a tutorial etc.. whatever I'd like to get a >series of points that represent a 3d slope presented to the user. > >Thanks >Matt > > That'd a fairly simple demo; you'll find a "big brother" of it in the "dots" demo with PyOpenGL (which is a particle system composed of dots (points)). OpenGLContext has a number of demos for displaying pointsets (dots), including the basic pointset test that just creates a set of 20 points or so. If what you mean is that you want to create a *mesh* from a point-set, then you need to add some more information, particularly you need a way to figure out the topological connections between points. If it's a regular x,y grid, you can readily construct the polygons, if not, you get into something a little more fuzzy. Good luck, Mike -- Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: Metaclasses and class variables
Jan-Ole Esleben wrote: >Yes, that works, but it is unfortunately not an option (at least not a >good one). > >Is there no way to create a class variable that exists during >definition of the class? (I cannot imagine there isn't, since >technically it's possible and manually it can be done...) > >Ole > > The metaclass hook occurs *after* class definition, anything using a side-effect of a metaclass hook then, *must* occur after the execution of the metaclass hook. At the time you want to write classvar.append the "class" is only a namespace, so, if you really need this feature you'll need to look elsewhere for at least *part* of the solution. A global "classvar" that, when appended to, caches values until your metaclass is called and transfers the cache to the class should *work*, but egads that's ugly compared to just classvar = [] . I guess what I'd ask is *why* is avoiding that single line so important. It could be there's a reasonable answer, but the amount of machinery required to avoid it is going to be significant. class meta( type ): newClassVar = [] def __new__( cls, name, bases, dictionary ): dictionary[ 'classvar' ] = cls.newClassVar[:] del cls.newClassVar[:] return super( meta, cls ).__new__( cls, name, bases, dictionary ) __metaclass__ = meta classvar = meta.newClassVar or something along those lines... Um, ick, but HTH, Mike -- Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: Selection, picking with PyOpenGL?
Erik Max Francis wrote: >I was interesting in adding selection and hit testing to ZOE, and was >looking at the PyOpenGL wrappers' handling of selection and picking. I >see glSelectBuffer to specify the size of the buffer, and glRenderMode >properly returns the number of hits when put back into GL_RENDER mode, >but I don't see any way within PyOpenGL itself to access the select >buffer to actually process the hits. Is there any way to do this >without using OpenGLContext? (If not, does this make sense, given that >OpenGLContext is supposed to be an additional helper library for a >gentle introduction to OpenGL?) > > I think you're missing what's returned from the glRenderMode code (which is non-standard OpenGL, to avoid the need for Python code using and decoding structured pointers), from OpenGLContext: nameStack = list(glRenderMode(GL_RENDER)) that is, you get a stack of name records (i.e. your glSelectBuffer with (near,far,names) records) back from the glRenderMode call, *not* just a count of the records produced. In general, you shouldn't *need* OpenGLContext to do anything, as all of its OpenGL code is written directly in Python to PyOpenGL's API (that was its original purpose in life, to test those APIs). It is, however, intended to provide a source for sample code showing how to accomplish effects when using PyOpenGL, so from renderpass.py: event.setObjectPaths([ nodepath.NodePath(filter(None,[ self.selectable.get(name) for name in names ])) for (near,far,names) in nameStack ]) shows you how to deal with the glSelectBuffer result, namely that it's a near value, a far value, and a set of integer names (the name stack) for each record in the set. OpenGLContext is here creating a node-path of all selectable objects from the scenegraph root to the final selected object using the nameStack and a registered set of name:node values (self.selectable). It will use those to populate the event objects that show up in the OpenGLContext mouse-event-handling APIs. HTH, Mike -- Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: Fire event when variable is Set/Get
Varghjärta wrote: ... But there is something that keeps bugging me, and that keeps me from embracing Python even more as a serious alternative to C#(for me). In C# I make heavy use of Get & Set, mainly to fire an event that some property has changed so that one can act on that _if one would need to. I've grown very used to doing it and it feels like the best way to make very OO and reusuable and easy to use Classes. It's fairly straightforward to implement observability for any of the various "modeling" descriptor systems out there. Attached you'll find a quick demo of how one could use BasicProperty with PyDispatcher to produce the effect. The biggest problem is that normally the particular events you want to fire are application specific. Heck, it's a toss-up whether the property or the client should be the sender of a message (basically depends on whether you normally want to filter by one or the other). The OpenGLContext vrml scenegraph has PyDispatcher-sending fields built-in if you want to see a real-world use. (We have others which use BasicProperty + PyDispatcher, but they're not Open-Source). Have fun, Mike -- ________ Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com """Demonstration of observable properties""" from dispatch import dispatcher from basicproperty import basic class Observable( basic.BasicProperty ): """Sample of creating an observable BasicProperty""" SETTING_SIGNAL = "basicproperty.observable.set" def __set__( self, client, value ): """Override to send dispatcher messages on setting""" value = super( Observable, self ).__set__( client, value ) try: dispatcher.send( signal = self.SETTING_SIGNAL, sender = self, value = value, client = client, ) except Exception, err: # you'd actually want to log here... pass return value if __name__ == "__main__": from basicproperty import propertied import random class Model( propertied.Propertied ): name = Observable( "name", """The name of the object, for some purpose""", defaultValue = "Uncle Tim", ) age = Observable( "age", """How old the object is, often just a guess""", defaultFunction = lambda prop, client: random.randint( 1, 100), ) def onName( sender, client, value ): """Process a change to a Model Object""" print 'Name', sender, client, value def onAge( client, value ): """Process a change to a Model Object""" print 'Age', client, value dispatcher.connect( onName, sender = Model.name, signal = Observable.SETTING_SIGNAL, ) dispatcher.connect( onAge, sender = Model.age, signal = Observable.SETTING_SIGNAL, ) m = Model() m.name = 'Peters' m.age # note we send a message, as the default is __set__ on retrieval m.age = 18 -- http://mail.python.org/mailman/listinfo/python-list
Re: Frankenstring
Thomas Lotze wrote: >Hi, > >I think I need an iterator over a string of characters pulling them out >one by one, like a usual iterator over a str does. At the same time the >thing should allow seeking and telling like a file-like object: > > Okay, first off, this is never going to be *fast* compared to something coded in C and wrapped with Python. You are dealing with every single character as a Python object, so let's forget fast for the moment and do a straightforward implementation: class Franken( str ): frankenIndex = 0 def __iter__( self ): while self.frankenIndex < len(self): yield self[ self.frankenIndex ] self.frankenIndex += 1 self.frankenIndex = 0 def seek( self, index ): self.frankenIndex = index def tell( self, index ): return self.frankenIndex if __name__ == "__main__": f = Franken( 'abcdefg' ) for c in f: print 'char', c if c == 'c': break f.seek( 5 ) l1 = list( f ) l2 = list( f ) assert l1 == [ 'f','g' ] assert l2 == list(str(f)) print 'first list', l1 print 'second list', l2 If you want to speed it up, you can optimise for various string sizes (eg using a slice of the string and the built-in iterator when appropriate), but in the end, it's not going to be anywhere near as fast as a C-engine tokeniser, so I'd personally spend more time on elegance than on speed... Anywho, have fun, Mike -- Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: Proposal: reducing self.x=x; self.y=y; self.z=z boilerplate code
Ralf W. Grosse-Kunstleve wrote: ... >class grouping: > >def __init__(self, x, y, z): >self.x = x >self.y = y >self.z = z ># real code, finally > >This becomes a serious nuisance in complex applications with long >argument lists, especially if long variable names are essential for >managing the complexity. Therefore I propose that Python includes >built-in support for reducing the ``self.x=x`` clutter. Below are >arguments for the following approach (*please* don't get too agitated >about the syntax right here, it really is a secondary consideration):: > > ... This can be dealt with quite nicely if you take a step back and consider what you're trying to do: * define a set of properties/fields for a class, potentially with default values for any given property * define an initialisation function which, given an instance of a class with defined properties (potentially with defaults) and a mapping of name:value pairs, assigns the named values to the named properties... class Grouping( propertied.Propertied ): x = FloatProperty( "x", """Holder for the grouping node's x coordinate""", defaultValue = 0, ) y = FloatProperty( "y", """Holder for the grouping node's y coordinate""", defaultValue = 0, ) z = FloatProperty( "z", """Holder for the grouping node's z coordinate""", defaultValue = 0, ) def __init__( self, **args ): # code here to do whatever you like to args (or whatever)... super( Grouping, self ).__init__( **args ) # extra code here... where the propertied.Propertied class defines a simple init that iterates through the defined properties pulling those which are present in the **args of the __init__ function and assigning them to the appropriate property. The advantage here is that your class is now documented properly, with the information about a given property all localised in a definition of the property. Your properties can also implement the observer pattern, automatic type checking and coercion, and can provide default values that appear when you delete current values (not just during initialisation). You can also do useful things like using the defined properties of a class to auto-generate __repr__ and __str__ values. You can have your properties delegate to other objects, or have them delay loading/resolving a property's value until the last possible second (lazy loading from an object database, for instance). The fact that it's all doable in Python 2.2+ without any need for new syntax... well... that's cool too. The disadvantage is that the properties tend to be unordered (unless you explicitly order them, and that's awkward looking), and you have to pass them to the constructor as keyword arguments. But then, if you're defining a class with 50 properties, you likely aren't really wanting to pass all 50 as positional arguments anyway. You'll find patterns like this in my own OpenGLContext (3D scenegraph viewer) and BasicProperty packages, as well as in Traits (originally part of a 2D graphics package), and Zope's FieldProperty (and PEAK, though as always with PEAK, it's somewhat weird in there ;) ). All of those are largish systems (or used in largish systems), btw. Just my thoughts, Mike -- Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Any movement on FreeBSD segfaults when using threads?
There was a post a few months ago from "snacktime" in which they described a failure when running under Twisted: http://mail.python.org/pipermail/python-list/2005-February/265137.html which they eventually narrowed down to an apparent problem with pycrypto: http://mail.python.org/pipermail/python-list/2005-February/265146.html and which generated this sourceforge bug report for Python: http://twistedmatrix.com/pipermail/twisted-python/2005-February/009475.html http://sourceforge.net/tracker/index.php?func=detail&aid=1080660&group_id=5470&atid=105470 I am seeing a similar effect, (though only after the program has been running a few days), in one of our products running on FreeBSD 5.3-RC2 with Python 2.3.4, PyOpenSSL 0.6, Twisted 2.0, and PyPgSQL 2.4 (which is somewhat different than snacktime's library collection (different OpenSSL and PostgreSQL bindings)). Note particularly, that the PyCrypto library to which the problem was apparently localised is *not* installed on the failing machine. As with snacktime, I see the failure coming from pthread_testcancel() (gdb) bt #0 0x28222f17 in pthread_testcancel () from /usr/lib/libpthread.so.1 #1 0x2821b39f in pthread_mutexattr_init () from /usr/lib/libpthread.so.1 #2 0x in ?? () I haven't been able to isolate the failures to a particular piece of code (this is showing up on a live server, so I'm hesitant to run under gdb for a couple of days to try to catch the failure). I'm *not* using RSA encryption (which was the area to which snacktime localised their problem). I *do* use SSL sockets from Twisted, however. I am running a background thread for database access, with a Queue mediating the set of items to process and reactor.callFromThread returning control to Twisted when the items are finished. Anyway, my question is really "has this been solved already"? If so, what was the solution? If not, I'll have to work at building a minimal failing test case so I can begin to debug it. Thanks for any information, Mike -- Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: preallocate list
Jim wrote: > Thanks for the suggestions. I guess I must ensure that this is my > bottle neck. ... > for line in f: > factor = [] > tokens = line.split() > for i in tokens: > factor.append(float(i)) > factors.append(loadFactor) > ... You might try: factors = [ [float(item) for item in line.split()] for line in f ] avoiding the extra statements for appending to the lists. Also might try: factors = [ map(float, line.split()) for line in f ] though it uses the out-of-favour functional form for the mapping. Good luck, Mike ____ Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: Extracting Font Outline informations
TTFQuery, built on top of FontTools, allows you to do this. You can see use of the extraction in the OpenGLContext/scenegraph/text package, which uses TTFQuery to implement 3D text rendering. http://ttfquery.sourceforge.net/ http://pyopengl.sourceforge.net/context/ HTH, Mike Gabriele *Darkbard* Farina wrote: > Hi, > > there is a Python library that makes me able to extract outline > informations from font files? I'd like to manage TrueType and FreeType > fonts ... > > I searched for som wrappers, but I didn't find anything ... > > bye ____ Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: Newbiw - PypenGL and OpenGLContext
Steve T wrote: >I have been searching for a language to help with a product I am >developing - last week I discovered PYOpenGL - looks really useful. I >found some demos which I studied to create part of what I need (a heap >of boxes) but when I try and add an image to the box faces it looks as >if I have to delve into OpenGLContext. Now from here drawing boxes is >no longer as simple. My main driver is speed of _development_ (not >graphics speed) so I would really like to be able to say >box(position_vector, size_vector,texture_file). > >Am I on the wrong track entirely here, should I pursue OpenGLContext >or am I misusing the visual.* parts of PYOpenGL? > > If you are really just drawing boxes, then raw PyOpenGL *may* be appropriate, but you'll need to figure out the wonders of texturing. You can find some pretty good code for that in the NeHe demos (I believe around the 6th or 7th). They create a rotating textured, lit box, so if that's all you need, you're golden. If your goal is to eventually produce a more involved scene, you will likely want one of the scenegraph engines. Obviously I'm biased toward OpenGLContext, but it's visual-alike package (i.e. a simplified scenegraph model) is not yet ready for prime time, so you'd have to use the full VRML97 model to create your boxes. That would look something like this: from OpenGLContext import testingcontext BaseContext, MainFunction = testingcontext.getInteractive() from OpenGLContext.scenegraph.basenodes import * class TestContext( BaseContext ): def OnInit( self ): """Initialise and set up the scenegraph""" self.sg = sceneGraph( children = [ Transform( rotation = (0,1,0,.8), children = [ Shape( geometry = Box( size=(4,4,2) ), appearance = Appearance( material = Material( diffuseColor =(1,1,1) ), texture = ImageTexture( url = "http://blog.vrplumber.com/images/ruminations.jpg";, ), ), ), ], ), ], ) def getSceneGraph( self ): """With older OpenGLContext versions need to explicitly return this""" return self.sg if __name__ == "__main__": MainFunction( TestContext ) There are other retained-mode/scenegraph engines available besides OpenGLContext, you can find a listing of some of them here: http://www.vrplumber.com/py3d.py?category=retained HTH, Mike Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: twistedSnmp and hexadecimal values ...
Francesco Ciocchetti wrote: >I'm tryng to use TwistedSnmp to make query and walk directly inside my >python code. > >The problem i'm facing is that any time there is an hexadecimal value to >be returned by the snmpwalk it is returened in a weird and useless >way... does anyone had any (successfull) experience with twistedsnmp and >hexadecimal values (as Mac Addresses)? > > TwistedSNMP returns the raw (binary) values as strings. Hexadecimal "values" are just human readable presentation logic for a binary value (stored as a string). For instance, for a MAC address, we currently display it to our users using: ":".join([ '%02x'%(ord(v)) for v in mac ]) where mac is the "weird and useful" ;) value you got back. That will give you a colon-separated hexadecimal display value, which may or may not be exactly what you want. Feel free to play with the display logic to get what you feel to be a good display format. We (Cinemon) often use string sub-classes that provide extra formatting logic and apply those to the values when we get the data-values in, but that's just an artefact of how we store and process the results of the TwistedSNMP queries, not TwistedSNMP itself. BTW, more detail in a question (what result you got, what result you expected, what you did to get the result) is generally a good idea. HTH, Mike Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com -- http://mail.python.org/mailman/listinfo/python-list
Re: Python becoming less Lisp-like
Thomas Bellman wrote: Torsten Bronger <[EMAIL PROTECTED]> wrote: Just to amplify Thomas' statements... ... And inflexibility will always make some situations horribly daunting to get out of. Powerful constructs like these can, in some cases, enable a skilled package writer to design an API that reduces the amount of boiler plate code needed for using the package. ... The following is from the (draft of) my paper for my pycon presentation (upcoming). We see similar patterns throughout the class-type redesign, the introduction of the __new__ method which allowed for “hooking” the creation (as distinct from initialization) of an object, the __getattribute__ method for hooking all attribute access. In each case, the operations were previously “part of the interpreter” and unavailable for customization by the Python programmer, the protocols generalized the behavior, making objects themselves responsible for what was previously implemented by the interpreter, and thus was always exactly the same. That sameness was arguably a strong advantage for Python in the early days of its life-cycle. When you were handed an object (particularly, any object defined by a class), it had simple, predictable behavior. The (one) class implementation was straightforward enough that most new users didn't particularly get lost in it. Of course, the flip side was the presence of “types”, which could behave very differently, and the need for people who wanted to customize the core behavior of an object (for instance to change how attribute traversal was handled) to write C extensions. And then there was the strange exception for attribute access for functions, but that was so natural and practical it seldom bit anyone who wasn't doing meta-programming. Effectively, there were a few bits of “magic” that the Python programmer needed to understand, but the variety and complexity of the magical patterns was limited. The magic incantations could be learned readily by the acolyte, but there was a hard cap on what could be done without becoming a “wizard” and programming in C. The limits on functionality made Python a smaller and less complex language, but it also tended to frustrate the “wizards”, who, after all, might know C, but would rather either program in Python (or spend more of their time looking at really deep C, rather than creating yet another customization to support something just beyond the reach of the acolytes). So, with Python 2.2, the wizards of Python made it possible to override and customize the behavior of objects in ways that were previously impossible. They brought things that were previously “magic” (for example, the order of attribute lookup that somehow created bound instance methods) into the realm of the understandable by rationalizing the process. metaclasses and descriptors are primarily of interest to metaprogrammers, people whose target audience is other programmers. Most programmers can get along perfectly fine ignoring them save for knowing what the particular ones they *use* do. Metaprogramming is not *necessarily* a *dark* art, but it can certainly lead the way to darkness if done improperly or wantonly. When done properly, you can build systems that seem to perfectly map a problem domain into the familiar embrace of everyday Python with just the right "lived in" touches to make it seem as though the language was designed from the start to support that domain [leaving aside any questions of syntactic variants, which is its own involved debate]. When you have metaprogramming to do, it is *such* a relief when you can get it done in Python without resorting to C. Have fun all, Mike ________ Mike C. Fletcher Designer, VR Plumber, Coder http://www.vrplumber.com http://blog.vrplumber.com PyCon is coming... -- http://mail.python.org/mailman/listinfo/python-list
