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

Re: Reasoning behind nested scope
by Peter Otten Aug. 5, 2004

Aug. 5, 2004

Andy Baker wrote: > (On a side note is there any way to call a nested function from outside > the parent? I was kind of expecting nested functions to be addressable > through dot notation like methods are but I can see why that wouldn't be > quite right. This might be a better question for the tutor list...) When you are nesting functions, you don't get one function sitting inside another function. Instead, the function creation code of the "inner" function is executed every time to the effect that you get a new inner function every time you call the outer one: >>> def make(): ... def f(): return "shoobidoo" ... return f ... >>> f1 = make() >>> f2 = make() >>> f1(), f2() ('shoobidoo', 'shoobidoo') >>> f1 is f2 False You wouldn't do that in cases like the above, when you get nothing in return for the extra overhead, but somtimes nesting is useful - because of the change in the scoping rules you now get readonly-closures: >>> def make(s): ... def f(): return s ... return f ... >>> f1 = make("now") >>> f2 = make("what") >>> f1(), f2() ('now', 'what') Peter

2 3

Re: Timing Difference: insert vs. append & reverse
by Bryan Olson Aug. 5, 2004

Aug. 5, 2004

Tim Peters wrote: > [Bryan Olson] >>[...] I hadn't realized that Python's lists are much slower >>than Perl's in important cases: >> >> http://perlmonks.thepen.com/17890.html >> >>Would this be a PEP-size change, or just a small fix? > > "Futile" is closer -- it's been debated to death repeatedly. Python > 2.4 adds a deque type instead, with O(1) insert/remove at both ends, > regardless of access pattern. That's O(1) per operation (not just > amortized O(1) -- the deque type never needs to move entries). But it breaks (efficient) indexing. Since we can have it all like Perl, why not? Googling up "dequeue" together with "O(1)" in comp.lang.python, (and with the misspelling "deque" which returns more results) I see some discussion, but very little reason against the enhancement. One post by Tim Peters suggests difficulty of implementation as a major reason. -- --Bryan

2 1

Re: cgi script headers - help
by Peter Hansen Aug. 5, 2004

Aug. 5, 2004

Ajay wrote: > hi! > > i am generating a file on the fly and making a user download it. > print "Content-Type: application/model\n\n" > print "Content-Disposition: attachment; filename=modeldef3\n" > print str > > i set the Content-Disposition to force a download dialog box and also give > the file a name. > however when i view the headers sent by my script, i have only > HTTP/1.1 200 OK > Date: Thu, 05 Aug 2004 02:26:48 GMT > Server: Apache/1.3.19 (Unix) > Connection: close > Content-Type: application/model > > why isn't the Content-Disposition header coming up? Probably because the previous line was terminated with three (3) newlines, thus marking the end of the header section... Note that print includes a newline normally, so including any other newlines will result in at least a double newline, which is the delimiter between header and body. -Peter

2 1

Re: The term "Protocol"
by Peter Hansen Aug. 5, 2004

Aug. 5, 2004

Bruce Eckel wrote: > Monday, August 2, 2004, 8:44:38 AM, Aahz wrote: >>Alex Martelli gave an excellent presentation on Design Patterns at OSCON, >>where he made the point that "interface" is roughly equivalent to syntax, >>whereas "protocol" is roughly equivalent to syntax plus semantics. In >>other words, computer langauges rarely (if ever -- although I suppose >>Eiffel comes close) enforce protocols in any meaningful way. > > But what would the syntax be in Python? -- there is none. Perhaps you > mean that in Java a protocol would be an interface plus agreed-upon > semantics, whereas in Python the protocol would be the agreed-upon > (e.g. "latent") interface plus the agreed-upon semantics. In a sense, > both the syntax and semantics would be latent; they would only be > exposed and tested during use, since there is no formalized way to > define them. > > In Python, of course, you could simply use a class with "pass" methods > everywhere to define a protocol. If there were some way to write > assertions about the protocol you could include those in the class. My reading of Alex' description seems to be the opposite. By definining protocol to include the semantics, you are basically ensuring that an implementation with "pass" would not be valid, except for a useless protocol that defined null semantics. But yes, what you say about the "latency" is true, and that's how Python works in many areas. You need decent automated tests to catch certain errors in Python, though tools like PyChecker can help with some of the ones that compilers will find in statically typed languages. -Peter

1 0

cgi script headers - help
by Ajay Aug. 5, 2004

Aug. 5, 2004

hi! i am generating a file on the fly and making a user download it. print "Content-Type: application/model\n\n" print "Content-Disposition: attachment; filename=modeldef3\n" print str i set the Content-Disposition to force a download dialog box and also give the file a name. however when i view the headers sent by my script, i have only HTTP/1.1 200 OK Date: Thu, 05 Aug 2004 02:26:48 GMT Server: Apache/1.3.19 (Unix) Connection: close Content-Type: application/model why isn't the Content-Disposition header coming up? thanks cheers -- Ajay Brar, CS Honours 2004 Smart Internet Technology Research Group ---------------------------------------------------------------- This message was sent using IMP, the Internet Messaging Program.

2 2

Can decorator syntax do this ? (annotated results' names)
by Sakesun Roykiattisak Aug. 5, 2004

Aug. 5, 2004

I have developed a framework to ease my database programming job, called "DescribedData Framework" which use inspection on function/method parameter-name to simplify my code. for example: -------- connection = dbi.connect('database-connection-string') cursor = connection.cursor() cursor.execute('select TITLE, NAME, AGE from PEOPLE') descdata = DescribedData(cursor.description, cursor.fetchall()) def show(TITLE, NAME, AGE): print '%s %s %s' % (TITLE, NAME, AGE) # print all records print 'Show all data' descdata.perform(show) # filter def teenage(AGE): return AGE <= 19 fdata = FilteredData(teenage, descdata) print 'Show teenages' fdata.perform(show) #sort def compare_age(AGE1, AGE2): return cmp(AGE1, AGE2) sdata = SortedData(compare_age, fdata) print 'Show sorted teenages' sdata.perform(show) -------- FilteredData & SortedData are handy decorators for DescribedData But when I implement more kind of decorators like "JoinedData", parameter-name inspection is no longer sufficient, function's *result-name* also have to be inspected. Currently, I use class instead of function, like this: -------- class ExtraNameJoiner(object): def join(self, NAME): return NAME.upper(), NAME.lower(), NAME.capitalize() def get_joining_description(self): return 'UPPER_NAME', 'LOWER_NAME', 'CAPITALIZED_NAME' joiner = ExtrasNameJoiner() jdata = JoinedData(joiner, sdata) # join extra fields def show_extra(NAME, UPPER_NAME, LOWER_NAME, CAPITALIZED_NAME): print '%s %s %s %s' % (NAME, UPPER_NAME, LOWER_NAME, CAPITALIZED_NAME) jdata.perform(show_extra) -------- The question is: Can the upcomming decorator-syntax be applied in this situation ? Can I do something like this : -------- @resultnames('UPPER_NAME', 'LOWER_NAME', 'CAPITALIZED_NAME') def join_extra_name(NAME): return NAME.upper(), NAME.lower(), NAME.capitalize() jdata = JoinedData(join_extra_name, sdata) -------- I stopped follow decorator-syntax discussion for a while, because it seems everlasting. But now look like it's settle, so I wonder how far can I go with this new feature ?

1 0

PyProtocols 0.9.3 Final
by Phillip J. Eby Aug. 5, 2004

Aug. 5, 2004

As there were no further bugs reported for PyProtocols 0.9.3rc2, PyProtocols 0.9.3 final is now available for download. There are no practical changes from rc2, except for file and documentation revision numbers and dates. What is PyProtocols? -------------------- PyProtocols is an extended implementation of PEP 246, adding a new "declaration API" that lets you easily define your own interfaces and adapters, and declare what adapters should be used to adapt what types, objects, or interfaces. Using PyProtocols, you can easily make flexible frameworks that you or other developers can extend without needing to modify the base framework. PyProtocols interfaces can interoperate with those of Twisted and Zope, or can be used entirely standalone. PyProtocols may be used, modified, and distributed under the same terms and conditions as Python or Zope. What's new in version 0.9.3? (Highlights) -------------------------------------------- * Adapter factories can now accept just one argument, the way Twisted and Zope adapters do. * Interface and protocol objects can be called, as a shortcut for 'adapt()' (as Zope and Twisted interfaces do) * You can now more easily install PyProtocols without a C compiler, using the '--without-speedups' option to 'setup.py' (see the README.txt file for details.) * Numerous other bug fixes and enhancements - see CHANGES.txt for details. IMPORTANT: If you are upgrading from a previous version of PyProtocols, please read UPGRADING.txt for important information. Certain rarely-used features have been deprecated, and others have changed slightly. Most users should not experience any problems (except perhaps for DeprecationWarnings), but please be sure to verify this before you upgrade any production code from 0.9.2 to 0.9.3. PyProtocols Resources --------------------- * Upgrading to PyProtocols 0.9.3 (and a look ahead to 1.0) http://peak.telecommunity.com/protocol_api/UPGRADING.txt.html * Detailed Changes for all releases: http://peak.telecommunity.com/protocol_api/CHANGES.txt.html * Release notes, installation instructions, and browsable API docs: http://peak.telecommunity.com/protocol_api/ * Source and Binary Releases: http://peak.telecommunity.com/dist/ * Reference Manual (HTML): http://peak.telecommunity.com/protocol_ref/module-protocols.html * Reference Manual (PDF): http://peak.telecommunity.com/protocol_ref.pdf * Browsable CVS Repository: http://cvs.eby-sarna.com/PyProtocols/

1 0

on the fly document generation
by Ajay Aug. 5, 2004

Aug. 5, 2004

hi! i generate a document on the fly and then print it from my script. to force the browser to show the 'save file' dialog rather than display the content in the browser, i have added the header Content-Type print "Content-Type: application\n" however when the save file dialog box comes up, it gives the script name (in this case sfile.cgi) as the filename. how can i change that to a different filename? print "Content-Type: application; name="test.xml\n" sorry its a little off-topic thanks cheers -- Ajay Brar, CS Honours 2004 Smart Internet Technology Research Group ---------------------------------------------------------------- This message was sent using IMP, the Internet Messaging Program.

1 0

Decorator syntax
by C. Barnes Aug. 5, 2004

Aug. 5, 2004

I vote for def f(): [synchronized, classmethod] (body of function) This is backwards compatible (Python <= 2.3 raise SyntaxError), and looks much nicer than @. The only problem is that you can't one-line a decorated function. You can't do that with the @ notation either, so it's not a big deal. - Connelly __________________________________ Do you Yahoo!? New and Improved Yahoo! Mail - 100MB free storage! http://promotions.yahoo.com/new_mail

1 0

optparse object population
by Eric O. Angell Aug. 5, 2004

Aug. 5, 2004

Say I have class Header: def __init__(self): self.foo = 0 # ... more fields and I have some commandline options that can specify better values for things in the header, such as foo. If I try the obvious (to me): parser = optparse.OptionParser() parser.add_option('--secret', type='string', dest='header', default=Header()) parser.add_option('--foo', type='int', dest='header.foo') (options, args) = parser.parse_args() when I run the program with "--foo 3", I'm sadly left with something like this: >>> print options <Values at 0xdeadbeef: {'header': <Header instance at 0xdeadbeef>, 'header.foo': 3}> So when optparse first initializes the variables to None, it's creating a 'header.foo' that has nothing to do with header=Header(). Is there a way to coerce optparse into populating an object for me? Yes, I could just do something like this: (options, args) = parser.parse_args() header = Header() header.foo = options.foo # ... more population of header but I'd rather just end up with options.header that I could then use somewhere else. Am I going to have any success with this, or am I just SOL? Thanks. -E

2 1