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

Concerning classes (Newb question)
by dontreply＠caramail.com Aug. 2, 2004

Aug. 2, 2004

Hi! I'm very new to the art of programming and I just learn OOP in python. I want to know why the following bit of code crash. Code: class Exemple: def __init__(self): self.list=[] self.var1=3 c=Exemple print c.list, c.var1 Thanks

3 2

Re: Call for Grant Proposals
by Maurice Ling Aug. 2, 2004

Aug. 2, 2004

Anthony Baxter wrote: >>I do agree that open source != hobbyist and I can hear screams already. >>What I'm trying to say is that the wordings of this call for proposal on >>the webpage is probably alright for hobbyist projects but can't face the >>scrutiny of the corporate world. >> >> > >I'm really not sure what you mean. Plenty of people in "the corporate world" >do open source. Plenty of people in "the corporate world" make money from >open source. > No doubt that people are making money from open source but equivalently, the same people may be making money from closed source. I'm arguing that the wordings of this grant is not clear, to the point that we do not know if this grant is applicable to us. >If the terms for the grants are that the project be open source, >how is this a problem? > Then it will means that some projects may not be able to apply for this grant if it is the requirement of the project to be proprietary due to business strategy. What you are saying is that all projects should be open sourced but that may not be the case. Either that or you are indiscriminately lobbying for the cause of open source irregards of whether companies are ready for it or not, and then discriminating against companies who are not ready for it yet. Although I support open source development and had personally donated my work to Sourceforge and Biopython project, I will think that it is tyrannic to assume all work must be open sourced. >It simply means that the person or organisation who >applies for the grants has to factor that element into their calculations. > > Without knowing the elements, how am I going to convince anyone on that? I'm not omnipotent. >Any company in 2004 who is still working in a mode that open source is >some strange thing that isn't used 'in the real world' is probably slow-moving >roadkill. > > > As mentioned above, parts of the work by my organization had been donated to public cause, including Biopython project and large public institutions like NCBI. Even the Online Mendelian Inheritance of Animals (OMIA) database was initially developed by people in my organization. As far as I know, some work precipitating to masters degrees had been made open sourced. So, please refrain from making harsh conclusions. Having said that, there is always a rubber-stamping process before an open sourced decision is made. But the wordings and inexactness of this call for grant proposal makes it hard for me even to get to convince my people that we should put in an application, not mentioning the rubber-stamping. Cheers Maurice

2 1

RE: Call for Grant Proposals
by Tony Meyer Aug. 2, 2004

Aug. 2, 2004

> What you are saying is that all projects should be > open sourced but that may not be the case. Either that or you are > indiscriminately lobbying for the cause of open source irregards of > whether companies are ready for it or not, and then discriminating > against companies who are not ready for it yet. [...] > I will think that it is tyrannic to assume all work must be open > sourced. It seems to me that the idea here is (or might be) that all work benefiting from these grants are open sourced, which is not the same as all work, at all. Given that Python is an open source project, and that the funding for the grant comes from Python users (and some associated open source projects like SpamBayes), this doesn't seem like a reasonable requirement. In fact, it seems that it's a reasonably logical one, one that the people that have contributed the funding would appreciate, and one that would be most likely to advance Python in some way. Given that there is a wide variety of open source licence, and that Python tends towards the least constrictive, it seems even less of a problem. If, for example, a funded project was to create some library for Python, which might help out lots of other users, there would be nothing stopping a closed source project making use of this library. Sure other people could too, but there are always strings attached to grants. > But the wordings and inexactness of this call > for grant proposal makes it hard for me even to get to convince my > people that we should put in an application, not mentioning the > rubber-stamping. Well, it seems to me that the trouble here is really that you haven't managed to hear back officially from the grant committee yet. But, as you probably know, there are always issues that need to be clarified with grants, and especially so with the first time that a grant is offered. Likewise, grant committee are always slow to respond to questions (even those that aren't run by volunteer organisations). =Tony Meyer

1 0

Re: flowcharting is automatic for Python via preprocessor
by Larry Bates Aug. 2, 2004

Aug. 2, 2004

Small changes eliminate the need for flow diagram (IMHO). def f(): """Demonstration of flow diagrams.""" for i in range(100): r=i % 3 if r == 2: print 'Remainder two, stopping loop' break if r == 1: if i < 10: print 'Less than ten' else: print 'Greater than ten' continue print 'Multiple of three' While I realize your example function was just that and example (because you could never even get to some of the branches because it exits when i=5). I see a lot of code with nested if/elif constructs that are unnecessary which add to the difficulty of "deciphering" what is going on. Larry Bates Syscon, Inc. <barnesc(a)onid.orst.edu> wrote in message news:mailman.1069.1091487166.5135.python-list@python.org... > > I like your idea. > > It is easy to get lost inside in eg a multiply nested if statement in a couple > multiply nested for loops, inside a try block. > > Sometimes one can split such code into separate functions, and this makes the > code cleaner. Sometimes not. > > However, the real issue is how the eye parses code. > > It's really inefficient to scan a whole page of code line by line to understand > where loops, branches, continues, and breaks occur. > > I think your visual diagrams could help, if they are: > > 1. Semantically cleaner. > IMO, there is too much detail in the flow diagrams. > (See http://www.getcet.org/images/ex3.png). > > A minimal amount of detail is easier to parse with the eye. > > See http://oregonstate.edu/~barnesc/temp/flow.pdf for a cleaner > example I came up with. This doesn't have arrows all over the > place (and hence may not be a flow diagram), but it does help > point out the relevant control structures. > > 2. Integrated into a code editor. > > I could see this being a new code editor fad, like collapsable > functions and classes. > > It could be quite useful, if the amount of visual information > is kept minimal, so that the coder is not distracted during > coding. > > If the flow diagrams are not integrated into an editor, I don't > think many coders will use them. > > Perhaps editors already do this. > > - Connelly

1 0

Re: New to Python/Programming
by Alan Gauld Aug. 2, 2004

Aug. 2, 2004

On Mon, 2 Aug 2004 19:30:21 +0100, "Tim Williams" <listserver(a)tdw.net> wrote: > As a recent Newbie, I found this to be a lot of help. (It covers both your > requirements) > > "Learn to Program Using Python" > http://www.amazon.com/exec/obidos/tg/detail/-/0201709384/102-1047927-021455… Thanks for the advocacy Tim. > would treat it as getting you to an intermediate level in-between "knowing > nothing" and "understanding what the hell the other books, online-docs and > examples are on about" :-) That's exactly what it's intended for. By the end of it you should breeze through the official tutor and understand most of the discussions here. > You'll be able to write in Python by the end of it. You would still need > something a bit more in depth afterwards (or maybe the online Docs will be > sufficient for you) but it won't seem as daunting. FWIW I am also updating the online version of the tutor(see .sig) and will be adding a few more pragmatic chapters - how to use the os module and do basic network and web client programming for example. But its slow work, the basic tutor has been updated to Python 2.3, all except the case study. (Tcl and QBASIC have been replaced by JavaScript and VBScript too) At the risk of losing sales the OP might try the web version first since it's now more up to date than the book (although not by much!) but the book is still better on fundamental principles (lots sidebars and footnotes for example). And in the interests of impartiality, there are several other tutorials for non programmers linked from the Python web site. Alan G. http://www.freenetpages.co.uk/hp/alan.gauld/tutor2/

1 0

Re: flowcharting is automatic for Python via preprocessor
by barnesc＠onid.orst.edu Aug. 2, 2004

Aug. 2, 2004

I like your idea. It is easy to get lost inside in eg a multiply nested if statement in a couple multiply nested for loops, inside a try block. Sometimes one can split such code into separate functions, and this makes the code cleaner. Sometimes not. However, the real issue is how the eye parses code. It's really inefficient to scan a whole page of code line by line to understand where loops, branches, continues, and breaks occur. I think your visual diagrams could help, if they are: 1. Semantically cleaner. IMO, there is too much detail in the flow diagrams. (See http://www.getcet.org/images/ex3.png). A minimal amount of detail is easier to parse with the eye. See http://oregonstate.edu/~barnesc/temp/flow.pdf for a cleaner example I came up with. This doesn't have arrows all over the place (and hence may not be a flow diagram), but it does help point out the relevant control structures. 2. Integrated into a code editor. I could see this being a new code editor fad, like collapsable functions and classes. It could be quite useful, if the amount of visual information is kept minimal, so that the coder is not distracted during coding. If the flow diagrams are not integrated into an editor, I don't think many coders will use them. Perhaps editors already do this. - Connelly

1 0

Re: Lambdas and variables
by jfouhy＠paradise.net.nz Aug. 2, 2004

Aug. 2, 2004

James Henderson <james(a)logicalprogression.net> wrote in message news:<mailman.917.1091100184.5135.python-list(a)python.org>... > I agree it's a gotcha. Interestingly you would have been forced to pass > x in as a default argument and avoided your problem until Python > introduced nested scopes in versions 2.1 and 2.2. You may find PEP 227 > on statically nested scopes illuminating: > > http://www.python.org/peps/pep-0227.html Heh. """ The proposed solution, in crude terms, implements the default argument approach automatically. The "root=root" argument can be omitted. """ Ok, I've got it now (famous last words). Thanks for your help :-) -- John.

1 0

parsing tar-style dates
by ddrucker＠gmail.com Aug. 2, 2004

Aug. 2, 2004

Is there any way to parse tar-style dates like "tomorrow" or "next wednesday" in python, other than doing a system call to date --date="string" and then parsing its output? Daniel Drucker ddrucker(a)gmail.com

2 1

RE: MySQLdb select
by Raaijmakers, Vincent (GE Infrastructure) Aug. 2, 2004

Aug. 2, 2004

I'm so glad that this issue is raised here. Can someone help me in this thread explaining me how to use multiple parameters. Jaco helped me already a lot and perhaps someone in this thread can give me the final answer: My table contains a blob and a bigint for storing images. The table name is also a variable, so the query looks like: SQL = "INSERT INTO %s (number, image) VALUES (%s, %s)" Well, I get a mysql exception if the execution looks like: cursor.execute(SQL, (tableName, aValue, jpgImage)) The execution only works when I fill out the data until the blob value is a 'left over' as a %s: Yes, this looks ugly.. but it works... SQL = "INSERT INTO %s (number, image) VALUES (%s," % (tableName, aValue) ## table name and number SQL += "%s)" ## left over... only the image cursor.execute(SQL, jpgImage) Can someone explain this to me? Vincent -----Original Message----- From: python-list-bounces+vincent.raaijmakers=ge.com(a)python.org [mailto:python-list-bounces+vincent.raaijmakers=ge.com@python.org]On Behalf Of Harald Massa Sent: Sunday, August 01, 2004 5:41 PM To: python-list(a)python.org Subject: Re: MySQLdb select Ruben, it is considered bad style wihtin c.l.p. to have aggressive emotions apart from mails about a "ternary operator" The correct way would have been: > > cursor.execute("SELECT * FROM mytest where clientID = %d" % numb) >> >> That's particularly BAD STYLE. It's best to keep to letting the >> DB-API do the proper quoting for all parameters. "Gerhard, why is this BAD STYLE? And how can I make the DB-API do the proper quoting for all parameters?" And probably the answer would have been: with doing "SELECT * FROM mytest where clientID = %d" % numb you are using the standard Python String-Formatting Functions. That may or may not lead to success. Especially when using a mixture of strings and numbers, they have to be quoted differently. Also special characters within strings have to be quoted - for example, the single ' has to be quoted within SQL because it is usually used as string delimiter cursor.execute("SELECT * FROM mytest where clientID = %d" , numb) The "," is the magic - with "Bla " % sth you are passing ONE argument --- the %s gets substituted, and one argument gets passed to the function. With "," two arguments get passed to the execute-"function" so that the DB-API has the challenge to integrate the parameters into the SQL-String. And it is more bad style to write "SELECT * FROM mytest where clientID = %d" % numb because % is defined with using a Tuple as second parameter, that would be "clientID = %d" % (numb,) Using % with sth. else than a tuple may succeed or give you very funny errors if numb happens to be a List in a later run. Best wishes, Harald -- http://mail.python.org/mailman/listinfo/python-list

2 1

where is yaml ?
by Wilk Aug. 2, 2004

Aug. 2, 2004

Hi, It seems that the link for python-yaml on the http://yaml.org main page is loose (http://iron.cx/yaml) :-( Anybody knows what's append ? What do you use ? pure python-yaml or syck ? -- Wilk - http://flibuste.net

2 1