🌐 Mailman 3 August 2004 - Python-list - python.org

asyncore, asynchat and threads
by Joshua Moore-Oliva Aug. 4, 2004

Aug. 4, 2004

Ok.. so here is my situation. I have an application that I would love to be able to use asynchat with for it's handle_close, found_terminator etc abstraction on. However, I must use a separate thread for each connection as the work resulting from the communication can result in blocking for an indefinate period of time. What I would love to do for each thread that is created, is to spin off an asynchat subclass and manually run some kind of loop() function that will do the polling//select etc for me, in a threadsafe manner, only on the one asynchat class/connection for that thread (e.g. I supply variables to keep track of the data) I have come across some sparse documentation that mentioned asyncore.loop() can be supplied a map of some sorts to loop through.. However, I have no idea how to create this map to pass to asyncore.loop, nor what thread-safe precautions (if any) I need to make. A simple example of a thread-safe way to create an asynchat subclass and get it to run would be great... even any pointers, help, ideas on how to make something like this work would be greatly appreciated as well. Thanks, Joshua Moore-Oliva

1 0

Rebinding variable, despite global statement
by telcom_sage＠yahoo.com Aug. 4, 2004

Aug. 4, 2004

I am using "from MyModule import *", (yes, yes, I know) MyModule has a variable "g_my_var" at the "global" scope. In the code that performs the import, I have a function that has the statement "global g_my_var". Despite this, when I try to assign to g_my_var it appears I am rebound to a different object. Beyond philosophical arguments about not using a "global" variable, is there a real reason why I can't assign to the global "g_my_var". I'm using python 2.3.2. One workaround is to place getter/setters in MyModule, but I was still surprised by this behavior. ----------------------- MyModule.py ----------------------------- g_my_var = 142 def UtilityFunction (): print "Inside UtilityFunction", g_my_var print "Inside UtilityFunction Id is", id (g_my_var) return ----------------------- MyProgram.py ----------------------------- from MyModule import * def ProgramFunction (): global g_my_var UtilityFunction () print "In ProgramFunction", g_my_var print "In ProgramFunction Id is", id (g_my_var) print "Now assigning a value of 42 in ProgramFunction." g_my_var = 42 print "After assignment, in ProgramFunction", g_my_var print "After assignment, in ProgramFunction Id is", id (g_my_var) UtilityFunction () return print "Excuting Main Program" UtilityFunction () ProgramFunction () =============== Output ====================== Excuting Main Program Inside UtilityFunction 142 Inside UtilityFunction Id is 7625912 Inside UtilityFunction 142 Inside UtilityFunction Id is 7625912 In ProgramFunction 142 In ProgramFunction Id is 7625912 Now assigning a value of 42 in ProgramFunction. After assignment, in ProgramFunction 42 After assignment, in ProgramFunction Id is 8008553 Inside UtilityFunction 142 Inside UtilityFunction Id is 7625912

2 2

SOAPpy and WSDL and basic authentication
by vivek.bhaskar＠gmail.com Aug. 4, 2004

Aug. 4, 2004

Hi, Can anyone provide me with sample code that will allow me to use SOAPpy with a WSDL file using HTTP basic authentication? The only thing I've been able to find on the net is SOAPpy.URLopener. However, I am unclear what the relationship is between SOAPpy.URLopener and SOAPpy.WSDL is. Is it as straightforward as this? > from SOAPpy import WSDL > url = 'http://user:pass@www.server.com/path/to/wsdl' > ssClient = WSDL.proxy(url) > ssClient.Test() # assuming Test is a SOAP-Action Or do I need to use SOAPpy.URLopener like so? > from SOAPpy import WSDL > from SOAPpy import URLopener > url = 'http://user:pass@www.server.com/path/to/wsdl' > url1 = URLopener.URLopener(username='user',passwd='pass') > ssClient = WSDL.proxy(url1.open(url)) > ssClient.Test() # assumption as above Or do I need to do something completely different? All help is much appreicated. Vivek

1 0

Re: Call for Grant Proposals
by Maurice LING Aug. 4, 2004

Aug. 4, 2004

Hi, At the same time, there are other questions besides an issue with open source or close source development. These 4 questions in my original email are probably more important to me, personally... 2. Will there be any form of intellectual property claims, including, copyrights, trade marks, patents, on any form of work generated from the PSF Grant by the Python Software Foundation (PSF) or any of its associates? If so, what are the precise nature of it? This is important as it may injure the research work and subsequently, the resultant thesis and publications directing towards the award of my degree. 3. On the website (http://www.python.org/psf/call-2004.html), it is stated that the project of which the grant is granted for must be completed by October 30, 2005. In event of failure to meet, what will be the penalities? 4. From Question 3, what constitutes partial failure or complete failure of the project? 5. From Question 4, it is stated on the website (http://www.python.org/psf/call-2004.html) that a delivery plan is essential in the proposal. How will delay in delivery affect the payment plan and the grant on the whole? The answers to these questions are something that the grant committee can answer now as it involves internal policy. As for Question 2, I cannot be in a project that will take up my time and cannot be directed towards the award of my degree. The next 3 questions deal closely with my lab's financial budgetting and without any answers to these questions, I can be in dire straits when things doesn't go as planned. Maurice Tim Peters wrote: > [Maurice Ling, wants to know whether PSF grants require funded work to be > open-sourced, etc] > > While I'm a PSF Director, I'm not on the Grants Committee and don't > speak for them. > > I'll just note that since the PSF is a public charity (under US tax > law), there are a lot of rules the PSF has to comply with when > distributing its funds. I expect that answering your questions will > require paying a lawyer to spell out what's possible. I don't know > what they'll say. In general, the PSF is required to act in the > public interest. > > That said, the PSF doesn't have a ton of money, and there will almost > certainly be more worthy requests for funding than there are funds to > pass out. Proposals will therefore compete. If I were on the Grants > Committee, whether a proposal would benefit the public at large would > be a major consideration for me. IOW, I think it's predictable-- and > appropriate --that proposals to fund open-source work will "score > points" for that reason alone. -- Maurice Han Tong LING, BSc(Hons)(Biochem), AdvDipComp, SN Doctor of Philosophy (Science) Candidate, The University of Melbourne mobile: +61 4 22781753 +65 96669233 mailing address: Department of Zoology, The University of Melbourne Royal Parade, Parkville, Victoria 3010, Australia residential address: 9/41 Dover Street Flemington, Victoria 3031, Australia email: mauriceling(a)acm.org resume: http://maurice.vodien.com/maurice_resume.pdf www: http://www.geocities.com/beldin79/ The information contained in this message, including its attachment(s), is CONFIDENTIAL and solely intended to its addressee(s) only. The content of this message, including its attachment(s), may be subjected to copyright and privacy laws. If you have received this email in error, please let me know by returning this email, and then destroy all copies. "I cannot discover anyone knows enough to say definitely what is and what is not possible" -Henry Ford "The difference between the impossible and the possible lies in a person's determination" -Tommy Charles Lasorda

2 2

Re: Call for Grant Proposals
by Maurice Ling Aug. 4, 2004

Aug. 4, 2004

Thanks for the reminder that it will stay in mail folders for a "LOOONG" time. I think I need to apologize for it. The fact here is that the wordings of the call for proposal is not concise and instead on looking at the wordings clearly and suggesting better wordings, everyone seems to be only interested in attacking the open source vs. closed source of the problem, which might just be a minor problem here. If only open sourced projects are logically funded, then by changing the words, "The Python Software Foundation is seeking grant proposals for projects related to the further development of Python, Python-related technology, and educational resources." to "The Python Software Foundation is seeking grant proposals for projects related to the further open-sourced development of Python, Python-related technology, and educational resources." will solve the problem. The crucial issues here are in my other questions that everyone convienently forgot, such as, what constitutes complete failure or partial failure of the project? Is not delivering a single deliverable account for complete failure? And what are the penalties? I can't imagine that if you had been granted a grant for an open sourced development in python and was not told way in advanced that a delay in delivering a deliverable will result in you being forfeited the rest of the grant money and will have to pay back 150% of all the money you had received from the grant, until you had accepted the grant and working on the project? There are more important issues here to solve than an argument between open and close source development. Maurice Michael Sparks wrote: >>In either sense, I hate to think that I may eventually force >>PSF to give me a grant for close sourced project. >> >> > >*Force* a registered charity to give you money ? > >*FORCE* ? > >I doubt I'm the only one **utterly disgusted** by that apparent attitude. > >Consider what you say before you send things - you've just stated the >moral equivalent of mugging a nun, and now it's on record in google and >mailing lists and in everyone's mail folders for a LOOONG time to come. > >The only positive things you've done here is remind me that I should >contribute to the PSF, and I sincerely hope 2 things a) that only OSS >projects get financed, since they have very few ways of getting financed >otherwise b) that some project like Star Killer gets financed to finish >off what's necessary - open or closed that project will be a boost to >many, many other projects. (I really hope they've applied) > >If it's closed however in the long term its net value is zero. (Much like >much closed software I've paid for over the years) > > >Michael. > > > >

2 1

Disappearing objects in shelve
by RayOsborn＠mac.com Aug. 4, 2004

Aug. 4, 2004

I have read a number of threads relating to problems with shelve, but I couldn't find one directly related my experience, so I would appreciate some advice. The closest one ended with a recommendation to upgrade to Python 2.3, but that, unfortunately, is out of my control. The system I use runs Python 2.2.2, and whichdb gives dbhash. The problem only occurs when serializing class instances; lists don't seem to be affected. Every time I attempt to read the instance from the shelve file, it seems to delete it from the file even though I am not performing any write operations. Attempts to use flag='r' in the shelve.open generated error messages, as did restricting write permissions to the database. This is some example code. 'Editor.py' defines the class whose instance is stored in db["john"], etc. The file, "editors.db", originally contained keys "nick and "john". By the end, the file has been emptied. import shelve import Editor def main(): db = shelve.open("editors.db") value = db["john"] db.close() db=shelve.open("editors.db") print db.keys() db.close() db = shelve.open("editors.db") value = db["nick"] db.close() db=shelve.open("editors.db") print db.keys() db.close() main() >> import test [ 'john','nick'] ['nick'] I got the same result when I attempted a deep copy instead of the assignment to value. Anyone know what I'm missing?

1 0

Re: python + byte array
by Peter Hansen Aug. 4, 2004

Aug. 4, 2004

Ruslan wrote: > I use win32com package to call WinHTTP COM methods. > One of them is 'Send'. I can pass string parameter to 'Send' method from > python - it's OK, but to send binary data i need somehow pass in 'Send' > method data of _byte array_ type(Visual Basic type). Does anybody know > how to mimic that 'byte array' type of Visual Basic in Python (i need to > pass big chunks of binary data in 'Send')? > I tried struct.pack, but that doesn't seem to be an option. Strings are probably what you are looking for... in Python a string is basically an array of bytes.

2 1

Re: flowcharting is automatic for Python via preprocessor
by Neil Hodgson Aug. 4, 2004

Aug. 4, 2004

Connelly: > I think your visual diagrams could help, if they are: > ... > 2. Integrated into a code editor. > > I could see this being a new code editor fad, like collapsable > functions and classes. http://www.eng.auburn.edu/grasp/index.html Neil

2 1

python + byte array
by Ruslan Aug. 4, 2004

Aug. 4, 2004

Hi, everybody. I use win32com package to call WinHTTP COM methods. One of them is 'Send'. I can pass string parameter to 'Send' method from python - it's OK, but to send binary data i need somehow pass in 'Send' method data of _byte array_ type(Visual Basic type). Does anybody know how to mimic that 'byte array' type of Visual Basic in Python (i need to pass big chunks of binary data in 'Send')? I tried struct.pack, but that doesn't seem to be an option. Thanks in advance. Your help is very appreciated. Best regards, Ruslan

1 0

PyQt - clear widget for redraw
by Peter Aug. 4, 2004

Aug. 4, 2004

I want to draw some lines on a widget. This works ok, but when I want to redraw, the old lines are still there. How do I clear or refresh the widget, so I can draw a new set of lines? Code snip below. TIA Peter ------------------------- def paintLines(self, e): p = QPainter(e) mar = e.width()/100 # margin vsp = 2 # vertical spacing hrz = (e.width() - 2*mar)/100 # horizontal scalar for i in range(100): p.drawLine(mar,mar+vsp*i,hrz*(l1[i]+1),mar+vsp*i) -------------------------

5 9