Call Java Code from python

[Bidda Gowda]

Hi,
I have a project coming up where i have to integrate our existing
Python based web application with Java Programs. Basically i should be
able to call Java programs which comes in the form of jars. Whats the
best way to call these jars from python ?

I looked at jpype and tried with small program and it works. Are there
any other tools which can do the same or if any how are they compared
to jpype ?

--
B Gowda.
-- 
http://mail.python.org/mailman/listinfo/python-list

Re: How far can stack [LIFO] solve do automatic garbage collection and prevent memory leak ?

[John Nagle]

On 8/17/2010 11:20 AM, Standish P wrote:

On Aug 17, 1:17 am, torb...@diku.dk (Torben Ægidius Mogensen) wrote:

Standish P  writes:

[Q] How far can stack [LIFO] solve do automatic garbage collection and
prevent memory leak ?



Because a stack has push and pop, it is able to release and allocate
memory. We envisage an exogenous stack which has malloc() associated
with a push and free() associated with a pop.


See


How many programmers have applied the ideas of these papers in their
programming practice ? I paste the abstract for convenience


http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.23.5498


Abstract:
This paper describes a memory management discipline for programs that
perform dynamic memory allocation and de-allocation. At runtime, all
values are put into regions. The store consists of a stack of regions.
All points of region allocation and deallocation are inferred
automatically, using a type and effect based program analysis. The
scheme does not assume the presence of a garbage collector.


That's actually an interesting idea.  If you can figure out object
lifetimes at compile time, allocation can be made far more efficient.

One of the basic questions is whether a function will ever "keep"
an object.  That is, will the function ever keep a reference to an
object that outlives the return from the function?  In many cases,
one can easily determine at compile time that a function will never
keep a passed object.  In such a case, you don't have to do reference
count updates on the object.  Most math functions have this property,
even vector and matrix math functions - they have no persistent state.

Python could use a bit of this.  If a function argument can
be identified as "non-kept", then the function doesn't need to
do reference count updates on it.  If a local variable in
a function is used only by "non-keep" functions and operations,
it can be created on the stack and released cheaply at block exit.

One can go much further in lifetime inference than this, as
the papers demonstrate.  There's a big win in the simple optimization
of identifying "non-keep" parameters, especially in mathematical work
where they're very common.  It's not clear that getting fancier than
that is a win.

Does Shed Skin have this optimization?  It should.

John Nagle
--
http://mail.python.org/mailman/listinfo/python-list

Re: Call Java Code from python

[Stefan Schwarzer]

Hi Bidda,

On 2010-08-18 09:19, Bidda Gowda wrote:
> I have a project coming up where i have to integrate our existing
> Python based web application with Java Programs. Basically i should be
> able to call Java programs which comes in the form of jars. Whats the
> best way to call these jars from python ?
> 
> I looked at jpype and tried with small program and it works. Are there
> any other tools which can do the same or if any how are they compared
> to jpype ?

Here are some slides from a talk by Andreas Schreiber,
"Mixing Python and Java", which might help:
http://www.slideshare.net/onyame/mixing-python-and-java

Personally, I've used JCC for accessing Lucene (a search
engine framework implemented in Java). It was a bit
rough in some places, but overall quite usable.

Stefan
-- 
http://mail.python.org/mailman/listinfo/python-list

use of gtk in a nautilus extension

[Nils]

Hi,
I am having some trouble opening a simple message/dialog to the user
from a natilus extension..

I have written a simple nautilus extension using python. It adds one
MenuItem to the context menu.
for testing I wanted to open a simple dialog when the user clicks this
menuitem.
(The code can be found - with some nice formatting here:
http://stackoverflow.com/questions/3325772/use-gtk-in-a-nautilus-extension-using-python)

The code is as follows:
--- cut ---
import gtk
import nautilus
import os
def alert(message):
"""A function to debug"""
dialog = gtk.MessageDialog(None, gtk.DIALOG_MODAL,
gtk.MESSAGE_INFO, gtk.BUTTONS_CLOSE, message)
dialog.run()
dialog.destroy()

class TestExtension(nautilus.MenuProvider):
def __init__(self):
pass

def get_file_items(self, window, files):
items = []
"""Called when the user selects a file in Nautilus."""
item = nautilus.MenuItem("NautilusPython::test_item", "Test",
"Test")
item.connect("activate", self.menu_activate_cb, files)
items.append(item)
return items

def menu_activate_cb(self, menu, files):
"""Called when the user selects the menu."""
for name in files:
alert(name)
--- cut ---

When I click on the menuitem nothing happens...
But when I replace def alert(message) like this:
def alert(message):
"""A function to debug"""
easygui.msgbox(message)
and obviously drop import gtk and add import easygui,

The dialog does appear. Can someone tell me why this is ??

Yours,
Nils
-- 
http://mail.python.org/mailman/listinfo/python-list

Re: 79 chars or more?

[Lie Ryan]

On 08/17/10 12:59, AK wrote:
> On 08/16/2010 10:42 PM, James Mills wrote:
>> On Tue, Aug 17, 2010 at 12:35 PM, AK  wrote:
>>> As monitors are getting bigger, is there a general change in opinion on
>>> the 79 chars limit in source files? I've experimented with 98 characters
>>> per line and I find it quite a bit more comfortable to work with that
>>> length, even though sometimes I have to edit files in 80 width
>>> terminals, it's still easier to adapt to some inconvenience when that
>>> happens than the other way around, since about 95% of time or more, I do
>>> use wider editor window or terminal.
>>>
>>> Is going over 79 still a terrible thing to do?  -andrei
>>
>> My personal opinion (despite monitors being wider) is
>> the horizontal scrolling isn't worth it. Stick to a 80-char width.
> 
> But.. why horizontal scrolling, isn't autowrap much better than that?


Do you seriously use autowrapper when writing code?

-- 
http://mail.python.org/mailman/listinfo/python-list

Re: 79 chars or more?

[Lawrence D'Oliveiro]

In message , Martin Gregorie wrote:

> 1) ssh terminal windows generally come up as 24 x 80

My terminal windows come up by default at something like 40 lines by 100 
characters.

> 2) at 24 x 80 I can get more ssh terminal windows on the desktop with
>minimal overlap than I can do with longer/wider windows.

I have 8 terminal windows open at once, they all fit fine, with room for an 
Emacs window and whatever else I might occasionally squeeze into there.

Hint: I use tabs as well as windows.
-- 
http://mail.python.org/mailman/listinfo/python-list

Re: How far can stack [LIFO] solve do automatic garbage collection and prevent memory leak ?

[Nick Keighley]

On 17 Aug, 18:34, Standish P  wrote:
> On Aug 16, 11:09 am, Elizabeth D Rather  wrote:
> > On 8/15/10 10:33 PM, Standish P wrote:
>
> > >>> If Forth is a general processing language based on stack, is it
> > >>> possible to convert any and all algorithms to stack based ones and
> > >>> thus avoid memory leaks since a pop automatically releases memory when
> > >>> free is an intrinsic part of it.
>
> > Forth uses two stacks.  The "data stack" is used for passing parameters
> > between subroutines ("words") and is completely under the control of the
> > programmer.  Words expect parameters on this stack; they remove them,
> > and leave only explicit results.  The "return stack" is used primarily
> > for return addresses when words are called, although it is also
> > available for auxiliary uses under guidelines which respect the primary
> > use for return addresses.
>
> > Although implementations vary, in most Forths stacks grow from a fixed
> > point (one for each stack) into otherwise-unused memory.  The space
> > involved is allocated when the program is launched, and is not managed
> > as a heap and allocated or deallocated by any complicated mechanism.  On
> > multitasking Forth systems, each task has its own stacks.  Where
> > floating point is implemented (Forth's native arithmetic is
> > integer-based), there is usually a separate stack for floats, to take
> > advantage of hardware FP stacks.
>
> > >>     - is forth a general purpose language? Yes
> > >>     - are all algorithms stack based? No
>
> > > Does Forth uses stack for all algorithms ? Does it use pointers , ie
> > > indirect addressing ? If it can/must use stack then every algorithm
> > > could be made stack based.
>
> > Forth uses its data stack for parameter passing and storage of temporary
> > values.  It is also possible to define variables, strings, and arrays in
> > memory, in which case their addresses may be passed on the data stack.
>
> > Forth is architecturally very simple.  Memory allocations for variables,
> > etc., are normally static, although some implementations include
> > facilities for heaps as needed by applications.
> > although some implementations include facilities for heaps as needed by 
> > applications.
>
> How are these heaps being implemented ? Is there some illustrative
> code or a book showing how to implement these heaps in C for example ?

any book of algorithms I'd have thought

http://en.wikipedia.org/wiki/Dynamic_memory_allocation
http://www.flounder.com/inside_storage_allocation.htm

I've no idea how good either of these is

> Are dictionaries of forth and postscript themselves stacks if we
> consider them as nested two column tables which lisp's lists are in
> essence, but only single row. Multiple rows would just be multiple
> instances of it at the same level inside parens.

I can't make much sense of that. But you seem to see Lisp data
structures in all sorts of strange places. I don't see that Lisp lists
are "nested two column tables"

> we can peek into stacks which is like car.

no.


> if it is not unusually
> costly computation, why not allow it ? there is no need to restrict to
> push and pop.

some stacks have a top() operation.


> roll( stack_name, num)
>
> itself can give all those postfix permutations that push and pop cant
> generate with a single stack. Can we use dictionaries to generate
> multiple stacks inside one global stack ?

I've no idea what you on about


-- 
http://mail.python.org/mailman/listinfo/python-list

Re: Python Developer - HFT Trading firm - Chicago, IL

[Matteo Landi]

Hi Rich,
I think it's better for you to post the message here (
http://www.python.org/community/jobs/ ).

Regards,

On Tue, Aug 17, 2010 at 6:07 PM, Rich Moss  wrote:
>
> Python developer needed for math/trading applications and research at
> leading HFT firm. The person we are searching for will have a strong
> background with python programming and the ability to work with very
> large historical datasets. You should have a very strong math
> background as well. This can involve writing very complicated python
> scripts and programs! You will work very closely with traders and
> quantitative analysts in their equities trading group on state-of-the-
> art trading strategy and execution systems.
>
> Requires:
>
> Strong python programming experience developing applications and
> scripts using complex regular expressions
>
> Strong math knowledge and education
> Experience working with massive datatsets/historical data
>
> This company is a top-tier electronic, algorithmic trading firm,
> located in Chicago, IL. This firm is one of the most advanced high
> frequency electronic trading firms in the world and uses python
> throughout the company, as well as other languages. This firm has a
> culture that rewards creativity and hard work. No third parties,
> please. We will not consider candidates from outside the USA. No
> telecommuting. We offer very generous compensation (best in the
> industry), fantastic benefits and very generous relocation packages.
> Please contact me immediately with a resume!
>
> Send resumes to:
>
> Rich Moss
> r...@mossltd.com
> --
> http://mail.python.org/mailman/listinfo/python-list
>



-- 
Matteo Landi
http://www.matteolandi.net/
-- 
http://mail.python.org/mailman/listinfo/python-list

Re: 79 chars or more?

[Jean-Michel Pichavant]

D'Arcy J.M. Cain wrote:

On Tue, 17 Aug 2010 16:28:02 +0200
Stefan Schwarzer  wrote:
  

I'd probably reformat this to

  self.expiration_date = translate_date(
find(response, 'MPNExpirationDate').text,
'%Y-%m-%d', '%m%d%Y')

or even

  self.expiration_date = translate_date(
find(response, 'MPNExpirationDate').text,
'%Y-%m-%d',
'%m%d%Y')



You can extend this if there are complicated sub-calls.  Probably
overkill for this example but here is the idea.

   self.expiration_date = translate_date(
 find(
   response,
  'MPNExpirationDate',
 ).text,
 '%Y-%m-%d',
 '%m%d%Y'
   )

I also moved the closing brace down to align with the line that opened
that block.

  

If this is supposed to convice 80+ chars users, that's an epic failure :)
This is exactly the kind of layout I'm happy to not use by not caring 
about the line width.


At least, if you still want to stick with 79 chars, do something like

text = find(response, 'MPNExpirationDate', ).text
self.expiration_date = translate_date(text,'%Y-%m-%d', '%m%d%Y')

JM
--
http://mail.python.org/mailman/listinfo/python-list

Re: How far can stack [LIFO] solve do automatic garbage collection and prevent memory leak ?

[Nick Keighley]

On 17 Aug, 21:37, Elizabeth D Rather  wrote:
> On 8/17/10 10:19 AM, Standish P wrote
> > On Aug 17, 12:32 pm, John Passaniti  wrote:

> >>> It is true that the other languages such as F/PS also have borrowed
> >>> lists from lisp in the name of nested-dictionaries and mathematica
> >>> calls them nested-tables as its fundamental data structure.
>
> >> No.
>
> > you are contradicting an earlier poster from forth who admitted the
> > part on dicts.

he's saying a forth dictionary isn't a lisp s-exp. Well it isn't.

> Not at all.  A Forth dictionary is a simple linked list, not the
> complicated kind of nested structures you're referring to.  You really
> seem addicted to very complex structures.

I thought he had the opposite problem! I thought it was trying to
knock in all his programming nails with same stack-based hammer.


>  They really aren't necessary for general programming.

whaever *that* is
-- 
http://mail.python.org/mailman/listinfo/python-list

Re: String substitution VS proper mysql escaping

[Nik Gr]

 Στις 18/8/2010 7:31 πμ, ο/η Cameron Simpson έγραψε:

On 17Aug2010 20:15, Νίκος  wrote:
| ===
| cursor.execute( ''' SELECT host, hits, date FROM visitors WHERE page =
| '%s' ORDER BY date DESC ''' % (page) )
| ===
|
| Someone told me NOT to do string substitution ("%") on SQL statements
| and to let MySQLdb do it
| for me, with proper escaping like the following
|
| ===
| cursor.execute('''SELECT host, hits, date FROM visitors WHERE page=%s
| ORDER BY date DESC''', (page,))
| ===
|
| The difference is that if some external source can control "page",
| and
| they put in a value like
| 100 ; DELETE FROM visitors; SELECT * FROM visitors
| i will be losing my database table data.

That other difference is that the mysql dialect support knows how to
correctly escape a string for insertion into an SQL statement. You may
not, or may forget to pre-escape the string, etc. Using the MySQLdb
stuff do it for you is reliable and robust.


Can you please tell me what escaping means by giving me an example of 
what is escaped and whats isn't?


Also hwo can i delete my data for testing purposes as?

http://webville.gr/index.html?page="100 ; DELETE FROM visitors; SELECT * 
FROM visitors"


I get an error...

| a) I wanted to ask what is proper escaping mean and why after variable
| page syntax has a comma

Because this:

   (page)

means the same thing as:

   page

i.e. the argument to the "%" operator is just the string in page.

This:

   (page,)

is a _tuple_ containing a single element, the page variable.
A bit like:

   [page]

which is a list containing a single element. The trailing comma is
needed to tell python you want to use a tuple, not the bare string.

The "%" operator has special knowledge that is it is passed as string instead
of a list or tuple or other sequence then it should act _as_ _if_ it had been
passed a single element tuple containing the string.

%s and %d is behaving the same due to % expecting a string instead of an 
integer?



Otherwise, because a string _is_ a sequence the "%" might want to treat
the string "foo" as the sequence:

   ("f", "o", "o")
cursor.execute('''SELECT host, hits, date FROM visitors WHERE page=%s 
ORDER BY date DESC''',  page)


But it alss might treat it an entity, i mean since 'page' is a variable 
containing a string why not just 'page' as it is expecting 'page' variable to 
give its value when asked?



Run these three loops to see the difference:

   for s in "foo":
 print s
   for s in ("foo"):
 print s
   for s in ("foo",):
 print s

Cheers,

>>> for s in "nikos":
print s


n
i
k
o
s

# this handles the string "nikos" as a series of chars right?

>>> for s in ("nikos"):
print s


n
i
k
o
s

# this handles the string "nikos" as a series of chars too but what si 
the difference with the above in htis with the parentheses? is "nikos" 
is handles still as string here?


>>> for s in ("nikos",):
print s


nikos

# Here yes it handles "nikos" as the 1st item of a tuple

nikos
>>> for s in ["nikos"]:
print s


nikos

# Here? why is it behaving fifferent than the above ("nikos") and is 
proccessign it all chars in one?


>>> for s in ["nikos",]:
print s


nikos

# Here it handles "nikos" as the 1st item of a list right?

--
http://mail.python.org/mailman/listinfo/python-list

Re: 79 chars or more?

[Stefan Schwarzer]

Hi Lie,

On 2010-08-18 12:02, Lie Ryan wrote:
> On 08/17/10 12:59, AK wrote:
>> On 08/16/2010 10:42 PM, James Mills wrote:
>>> My personal opinion (despite monitors being wider) is
>>> the horizontal scrolling isn't worth it. Stick to a 80-char width.
>>
>> But.. why horizontal scrolling, isn't autowrap much better than that?
> 
> Do you seriously use autowrapper when writing code?

I think he means wrapping on the screen, not actually
inserting line breaks.

Stefan
-- 
http://mail.python.org/mailman/listinfo/python-list

Re: Working with PDFs?

[Anssi Saari]

 writes:

> - Pull out text from each PDF page (to search for specific words)
> - Combine separate pdf documents into one document
> - Add bookmarks (with destination settings)

PDF Shuffler is a Python app which does PDF merging and splitting very
well. I don't think it does anything else, though, but maybe that's
where your code comes in?
-- 
http://mail.python.org/mailman/listinfo/python-list

Re: String substitution VS proper mysql escaping

[Cameron Simpson]

On 18Aug2010 12:07, Nik Gr  wrote:
|  Στις 18/8/2010 7:31 πμ, ο/η Cameron Simpson έγραψε:
| >On 17Aug2010 20:15, Νίκος  wrote:
| >| ===
| >| cursor.execute( ''' SELECT host, hits, date FROM visitors WHERE page =
| >| '%s' ORDER BY date DESC ''' % (page) )
| >| ===
| >|
| >| Someone told me NOT to do string substitution ("%") on SQL statements
| >| and to let MySQLdb do it
| >| for me, with proper escaping like the following
| >|
| >| ===
| >| cursor.execute('''SELECT host, hits, date FROM visitors WHERE page=%s
| >| ORDER BY date DESC''', (page,))
| >| ===
| >|
| >| The difference is that if some external source can control "page",
| >| and
| >| they put in a value like
| >| 100 ; DELETE FROM visitors; SELECT * FROM visitors
| >| i will be losing my database table data.
| >
| >That other difference is that the mysql dialect support knows how to
| >correctly escape a string for insertion into an SQL statement. You may
| >not, or may forget to pre-escape the string, etc. Using the MySQLdb
| >stuff do it for you is reliable and robust.
| 
| Can you please tell me what escaping means by giving me an example
| of what is escaped and whats isn't?

In your plain substitution example above:

  cursor.execute( ''' SELECT host, hits, date FROM visitors WHERE page =
  '%s' ORDER BY date DESC ''' % (page) )

Supposing page is the string "100". This will produce the SQL statement:

  SELECT host, hits, date FROM visitors WHERE page = '100' ORDER BY date DESC

which looks ok. But suppose page was the string:

  bill o'reilly

Then your SQL statement looks like this:

  SELECT host, hits, date FROM visitors WHERE page = 'bill o'reilly' ORDER BY 
date DESC

To the SQL engine this looks like the string "bill o" followed by an SQL
instruction named "reilly", and then the opening quote for another string.
Invalid SQL.

The procedure used to avoid this problem (to insert an _arbitrary_
string into the SQL statement) is to "escape" problematic characters in
strings when placing them into SQL statements. In this case, the quote
character in the string is the SQL "end string" character. Therefore the
string must be modified in the SQL statement to be correctly expressed.

IIRC, SQL uses the quote doubling convention for strings, so this:

  SELECT host, hits, date FROM visitors WHERE page = 'bill o''reilly' ORDER BY 
date DESC

is how one would write the literal SQL for that.

The MySQLdb library will do this and a host of other equivalent things
automatically and correctly and consistently when you pass page as a
parameter to the execute() method, needing no special attention or
detailed syntactic knowledge on your part when you write your program.

[...snip...]
| >The "%" operator has special knowledge that is it is passed as string instead
| >of a list or tuple or other sequence then it should act _as_ _if_ it had been
| >passed a single element tuple containing the string.
| >
| %s and %d is behaving the same due to % expecting a string instead
| of an integer?

I haven't checked. I make a point of using the (page,) form (i.e.
always use a tuple, even with just one argument) these days. That way
there is no ambiguity.

| >Otherwise, because a string _is_ a sequence the "%" might want to treat
| >the string "foo" as the sequence:
| >
| >   ("f", "o", "o")
| cursor.execute('''SELECT host, hits, date FROM visitors WHERE
| page=%s ORDER BY date DESC''',  page)
| 
| But it alss might treat it an entity, i mean since 'page' is a
| variable containing a string why not just 'page' as it is expecting
| 'page' variable to give its value when asked?

A string is also a sequence of characters.

| >Run these three loops to see the difference:
| >
| >   for s in "foo":
| > print s
| >   for s in ("foo"):
| > print s
| >   for s in ("foo",):
| > print s
| >
| >Cheers,
| >>> for s in "nikos":
| print s
| 
| 
| n
| i
| k
| o
| s
| 
| # this handles the string "nikos" as a series of chars right?

Yes.

| >>> for s in ("nikos"):
| print s
| 
| 
| n
| i
| k
| o
| s
| 
| # this handles the string "nikos" as a series of chars too but what
| si the difference with the above in htis with the parentheses? is
| "nikos" is handles still as string here?

It is exactly the same as the first loop. Just as:

  1 + 3

is exactly the same as:

  (1) + (3)

| >>> for s in ("nikos",):
| print s
| 
| nikos
| 
| # Here yes it handles "nikos" as the 1st item of a tuple

Yep.

| >>> for s in ["nikos"]:
| print s
| 
| nikos
| 
| # Here? why is it behaving fifferent than the above ("nikos") and is
| proccessign it all chars in one?

("nikos",) is a single element tuple.
["nikos"] is a single element list.
["nikos",] is also a single element list, just written like the tuple.

You don't see the ["nikos",] form very often because ["nikos"] is not
ambiguous. It is only because ("nikos") gets reduced to plain "nikos

Re: How far can stack [LIFO] solve do automatic garbage collection and prevent memory leak ?

[spinoza1111]

On Aug 18, 1:21 am, Standish P  wrote:
> > Garbage collection doesn't use a stack. It uses a "heap", which is in
> > the abstract a collection of memory blocks of different lengths,
> > divided into two lists, generally represented as linked lists:
>
> > 1.  A list of blocks that are free and may be used to store new data
>
> > 2.  A list of blocks that are in use, or haven't been freed (yet)
>
> Is this all that a heap is or is there more to it ? I have been
> looking for simple but complete explanation of heap for a while and
> not gotten to it. I think I am looking for a stack allocation on the
> same pattern. In a disk, a file is fragmented in many contiguous
> blocks and is accessed automatically.
>
>
>
>
>
> > There is no way you could do memory management of all but the most
> > trivial and fixed-length data chunks using a stack. Sure, you could
> > reserve thousands of bytes on the stack for an array but suppose your
> > language allows arrays to grow or shrink. To keep its property of
> > being adjacent, you'd have to do something horrible such as move
> > unrelated data allocated later, which raises all sorts of security
> > issues, doesn't it.
> > A stack, or something which works like a stack (that is, a stack) is a
> > necessary but not sufficient condition for a working C runtime because
> > C functions can call themselves recursively, whether directly or
> > indirectly. If this last condition did not obtain, each function could
> > give the functions it calls some of its own memory and the called
> > function could save a fixed set of non-stacked general registers in
> > that area; this was in fact the practice on IBM 370 and in assembler
> > language at a time when many "data processing managers" though
> > recursion was a Communist plot.
>
> > However, data structures of variable size, or data structures that
> > merely take up a lot of space, don't play nice with others on the
> > stack, so, we place their address on the stack and store them in
> > another place, which was named the heap, probably, as a sort of
> > witticism.
>
> > Gilbert and Sullivan:
>
> > If anyone anything lacks
> > He'll find it all ready in stacks
>
> This you might want to take this to the Forth people because they are
> marketing their language as a cure for all that plagues programming
> today.

No, they're not. Stack based languages have seen better days and Forth
(and the SL/1 language I supported with compilers at Bell-Northern
Research) were last in fashion in the 1970s. Processors seldom could
multitask, so it wasn't recognized that the stack could be a
performance bottleneck, where stack operations cannot be pipelined or
executed in parallel.

John Hennessy of Stanford and MIPS made the stack must die case at ACM
ASPLOS in 1987. Niklaus Wirth was also at this conference at which I
was a fly on the wall, maintaining that the stack was good for
reliability and verifiability of software.

Forth had a snowball's chance because it forces ordinary programmers
to think in Reverse Polish notation and is for the above reasons hard
to pipeline, although of course it can be pipelined.
>
>
>
> > was wrong, and needs to be brought up to date:
>
> > You cannot do everything in a stack
> > Unless you code an almighty hack
> > If you're a coding Knight who says, "Neep",
> > You'll probably need to implement a heap
> > A pile a heap of benefits you'll reap
> > If only my advice in your brain you'll keep
> > And avoid memory leaks from which data doth seep
> > By using a well-implemented, well structured, and well-documented
> > Heap!
>
> > [Chorus of Sailors]
> > We will to heart your advice take, and always use a heap!
>
> > [Soloist]
> > Oh thank you do
> > To this be true
> > And always my sage advice do keep
> > That you always need to use a heap!- Hide quoted text -
>
> > - Show quoted text -

-- 
http://mail.python.org/mailman/listinfo/python-list

Re: 79 chars or more?

[BartC]

"Roy Smith"  wrote in message 
news:roy-319e47.09055017082...@news.panix.com...

In article ,
Lawrence D'Oliveiro  wrote:

In message , AK 
wrote:


> As monitors are getting bigger, is there a general change in opinion on
> the 79 chars limit in source files?

WHAT 79-character limit in source files?

I currently have my Emacs windows set at 100 characters wide, and Iâ?Tm
thinking of going wider.

Remember, the old hardcopy terminals used to produce 132-character-wide
listings.


Those of you who think "old hardcopy terminals" did 132 wide obviously
don't remember the ASR-33 :-)


ASR33s I think might have been 72 columns wide (and punched cards had a 
similar restriction).


However, lineprinter output was more likely to be 132 columns.

--
bartc 


--
http://mail.python.org/mailman/listinfo/python-list

Re: String substitution VS proper mysql escaping

[Tim Chase]

On 08/18/10 04:50, Cameron Simpson wrote:

("nikos",) is a single element tuple.
["nikos"] is a single element list.
["nikos",] is also a single element list, just written like the tuple.

You don't see the ["nikos",] form very often because ["nikos"] is not
ambiguous.


I most frequently see/use the trailing comma in a one-item list 
when I expect additional items will be accrued later, so it makes 
my version-control diffs tidier:


  items = [
   "nikos",
   ]

Then a month later, I add

  items = [
   "nikos",
   "cameron",
   ]

My diff is just

   items = [
"nikos",
+   "cameron",
]

which is quite easy to understand.  However if my original was

  items = [
   "nikos"
   ]

and I later want to make it


  items = [
   "nikos",
   "cameron"
   ]

the diff then becomes

   items = [
-   "nikos"
+   "nikos",
+   "cameron"
]

which isn't nearly as easy to read & understand because I now 
have to notice the inter-linear differences ("did something more 
than the new-comma happen?").


-tkc





--
http://mail.python.org/mailman/listinfo/python-list

Re: 79 chars or more?

[Lawrence D'Oliveiro]

In message <4c6a9c72.4040...@sschwarzer.net>, Stefan Schwarzer wrote:

>   self.expiration_date = translate_date(
> find(response, 'MPNExpirationDate').text,
> '%Y-%m-%d',
> '%m%d%Y')

Might I suggest (guessing at the argument keywords here) :

self.expiration_date = translate_date \
  (
TheText = find(response, 'MPNExpirationDate').text,
ToFormat ='%Y-%m-%d',
FromFormat ='%m%d%Y'
  )

Presumably FromFormat should be localizable, rather than hard-coded.

See, that’s the kind of thing you notice when you think about the code in 
this way.
-- 
http://mail.python.org/mailman/listinfo/python-list

Re: looping through possible combinations of McNuggets packs of 6, 9 and 20

[News123]

On 08/17/2010 11:44 PM, Baba wrote:
> On Aug 16, 6:28 pm, "cbr...@cbrownsystems.com"
>  wrote:
> 
>> First, suppose d = gcd(x, y, z); then for some x', y', z' we have that
>> x = d*x', y = d*y', z = d*z'; and so for any a, b, c:
>>
> 
> 
>could you explain the notation?
> 
>what is the difference btw x and x' ?
> 
>what is x = d*x', y supposed to say?
> 
> 


gcd(x,y,z) determines the greates number by which all three numbers can
be devided.

2,4,6 for example are all divided by 2
thus d=2
now you dived x,y,z by d and call them x' , y' , z'

The point is if
x,y,z have a gcd grater than one, then you know for sure,
that you will never be able to find the a finit greates amount, which
cannot be bought


if xmymz are all divisible by d, then any combination will also be
dividible by d


thas any number not dividible by d ( for d > 1)

for example n*d + 1

can not be bought
-- 
http://mail.python.org/mailman/listinfo/python-list

Python 2.7 support for PyWin32

[paulo.jpi...@gmail.com]

Hi everyone,

does anyone know when PyWin is going to support Python 2.7?

I tried to look for information, but to no avail.

Thanks in advance,
Paulo
-- 
http://mail.python.org/mailman/listinfo/python-list

Re: 79 chars or more?

[Roy Smith]

In article ,
 "BartC"  wrote:

> >> Remember, the old hardcopy terminals used to produce 132-character-wide
> >> listings.
> >
> > Those of you who think "old hardcopy terminals" did 132 wide obviously
> > don't remember the ASR-33 :-)
> 
> ASR33s I think might have been 72 columns wide (and punched cards had a 
> similar restriction).

Yeah, I was trying to remember if it was 72 or 80.  Hmmm, looks like 
you're right, it *is* 72 (http://www.pdp8.net/asr33/asr33.shtml).  

Punched cards (at least the common ones used by an 029 or 129 punch 
machine) were 80 columns.  The 72 column restriction was an artificial 
one imposed by some programming languages such as Fortran.  Columns 
73-80 could be used to punch a sequence number, so that if you dropped 
your deck, you could re-assemble it by running it through a card sorter.

> However, lineprinter output was more likely to be 132 columns.

Yeah, but I wouldn't call a line printer a "terminal".
-- 
http://mail.python.org/mailman/listinfo/python-list

Re: 79 chars or more?

[Roy Smith]

In article <4c6b9...@dnews.tpgi.com.au>, Lie Ryan  
wrote:

> On 08/17/10 12:59, AK wrote:
> > On 08/16/2010 10:42 PM, James Mills wrote:
> >> On Tue, Aug 17, 2010 at 12:35 PM, AK  wrote:
> >>> As monitors are getting bigger, is there a general change in opinion on
> >>> the 79 chars limit in source files? I've experimented with 98 characters
> >>> per line and I find it quite a bit more comfortable to work with that
> >>> length, even though sometimes I have to edit files in 80 width
> >>> terminals, it's still easier to adapt to some inconvenience when that
> >>> happens than the other way around, since about 95% of time or more, I do
> >>> use wider editor window or terminal.
> >>>
> >>> Is going over 79 still a terrible thing to do?  -andrei
> >>
> >> My personal opinion (despite monitors being wider) is
> >> the horizontal scrolling isn't worth it. Stick to a 80-char width.
> > 
> > But.. why horizontal scrolling, isn't autowrap much better than that?
> 
> 
> Do you seriously use autowrapper when writing code?

I absolutely use auto-indent, parenthesis matching, and syntax coloring.  
>From time to time, I turn on auto-wrap, but generally find it more 
annoying than useful and turn it back off again.
-- 
http://mail.python.org/mailman/listinfo/python-list

Re: Python 2.7 support for PyWin32

[Mark Lawrence]

On 18/08/2010 12:54, paulo.jpi...@gmail.com wrote:

Hi everyone,

does anyone know when PyWin is going to support Python 2.7?

I tried to look for information, but to no avail.

Thanks in advance,
Paulo


It was created on 2009-07-08!!! See:-

http://sourceforge.net/projects/pywin32/files/pywin32/

Cheers.

Mark Lawrence.

--
http://mail.python.org/mailman/listinfo/python-list

when 'myArray * 'myObject' is not equal to 'myObject' * 'myArray'

[Duim]

Although I'm sure somewhere this issue is discussed in this (great)
group, I didn't know the proper search words for it (although I
tried).

I'm using python (2.6) scientifically mostly, and created a simple
class to store time series (my 'Signal' class).
I need this class to have a possibility to get multiplied by an array,
but pre and post multiplication have different mathematical outcomes
( basically A* B != B*A ) .

Post multiplication by an array works fine defining __mul__ in the
Signal class, but pre multiplication does not. It keeps trying to
multiply all elements separately instead to send this array to my
__rmul__ function.

How can I fix this without the need for a separate
'multiplysignal(A,B)' function?
To make things easy I've made a small example:

[code]
import numpy as np

class Signal(object):
def __init__(self,data,dt):
self.data=data
self.dt=dt

def Nch(self):
return self.data.shape[0]

def __mul__(self,other):
print 'mul called! ',other

if isinstance(other,type(np.array([1,2]))):
#it's an array: use dot product:
return Signal(np.dot(self.data,other),self.dt)


if other.__class__.__name__=='Signal':
# do something
pass


def __rmul__(self,other):
print 'rmul called! ',other

if isinstance(other,type(np.array([1,2]))):
#it's an array: use dot product:
return Signal(np.dot(other,self.data),self.dt)


if other.__class__.__name__=='Signal':
# do something
pass

mySignal=Signal(np.array([[1.,2],[4,5]]),1.)
myArray=np.array([[1.,2.],[4.,3.]])

result_mul = mySignal*myArray
result_rmul = myArray*mySignal #called 4 times for all members once!

#result:
#mul called!  [[ 1.  2.]
# [ 4.  3.]]
#rmul called!  1.0
#rmul called!  2.0
#rmul called!  4.0
#rmul called!  3.0
[/code]



-- 
http://mail.python.org/mailman/listinfo/python-list

error Compile libxml2 from python 2.6.4

[Mauricio Martinez Garcia]

Hi!.

 Have the next error on install libxml2.

python/bin/python setup.py install
failed to find headers for libxml2: update includes_dir

This my version of python

==> 
/bscs/bscs/prod/523/WORK/MP/NORTEL/IN/MEXICO/CDRS/ATS/MONITOREO/reportes_Milton/python/bin/python

Python 2.6.5 (r265:79063, Jul 21 2010, 13:05:40) [C] on hp-ux11
Type "help", "copyright", "credits" or "license" for more information.


I'm Installing libxml2 for python 2.6.5, and this error break the install.

The  python path prefix is 
/bscs/bscs/prod/523/WORK/MP/NORTEL/IN/MEXICO/CDRS/ATS/MONITOREO/reportes_Milton/python, 
and i don't privileges of root.


The configuration to be personalized, in prefix path.

My compile line is:
./configure 
--with-python=/bscs/bscs/prod/523/WORK/MP/NORTEL/IN/MEXICO/CDRS/ATS/MONITOREO/reportes_Milton/python 
\
--prefix=/bscs/bscs/prod/523/WORK/MP/NORTEL/IN/MEXICO/CDRS/ATS/MONITOREO/reportes_Milton/libxml 
\
--with-libxml-include-prefix=/bscs/bscs/prod/523/WORK/MP/NORTEL/IN/MEXICO/CDRS/ATS/MONITOREO/reportes_Milton/pythonPruebas/libxml2/libxml2-INC/usr/local/include/libxml2/libxml 





This is the output when finalize compiling this is message:

Enabled Schemas/Relax-NG support
configure: creating ./config.status
config.status: creating libxml2.spec
config.status: creating Makefile
config.status: creating include/Makefile
config.status: creating include/libxml/Makefile
config.status: creating doc/Makefile
config.status: creating doc/examples/Makefile
config.status: creating example/Makefile
config.status: creating python/Makefile
config.status: creating python/tests/Makefile
config.status: creating include/libxml/xmlversion.h
config.status: creating xml2-config
config.status: creating libxml-2.0.pc
config.status: creating libxml-2.0-uninstalled.pc
config.status: creating xml2Conf.sh
config.status: creating python/setup.py
config.status: creating config.h
config.status: config.h is unchanged
config.status: executing depfiles commands

And this is output of make install:

Libraries have been installed in:
   
/bscs/bscs/prod/523/WORK/MP/NORTEL/IN/MEXICO/CDRS/ATS/MONITOREO/reportes_Milton/python/lib/python2.6/site-packages


If you ever happen to want to link against installed libraries
in a given directory, LIBDIR, you must either use libtool, and
specify the full pathname of the library, or use the `-LLIBDIR'
flag during linking and do at least one of the following:
   - add LIBDIR to the `LD_LIBRARY_PATH' environment variable
 during execution
   - use the `-LLIBDIR' linker flag

See any operating system documentation about shared libraries for
more information, such as the ld(1) and ld.so(8) manual pages.

Making install in tests
No suffix list.
No suffix list.
/bin/sh ../../mkinstalldirs 
/bscs/bscs/prod/523/WORK/MP/NORTEL/IN/MEXICO/CDRS/ATS/MONITOREO/reportes_Milton/libxml/share/doc/libxml2-python-2.6.4/examples
mkdir -p -- 
/bscs/bscs/prod/523/WORK/MP/NORTEL/IN/MEXICO/CDRS/ATS/MONITOREO/reportes_Milton/libxml/share/doc/libxml2-python-2.6.4/examples
(for test in build.pyattribs.py  tst.py  
tstxpath.py xpathext.py push.py pushSAX.py  
pushSAXhtml.py  error.pyserialize.py validate.py
tstURI.py   cutnpaste.py xpathret.pyxpath.py
outbuf.py   inbuf.pyresolver.py  regexp.py  reader.py   
reader2.py  reader3.py  reader4.py  reader5.py  
reader6.py  reader7.py  walker.py   ctxterror.py 
readererr.py relaxng.pythread2.py tst.xml  valid.xml   
invalid.xml; \
  do ../.././install-sh -c -m 0644 ./$test 
/bscs/bscs/prod/523/WORK/MP/NORTEL/IN/MEXICO/CDRS/ATS/MONITOREO/reportes_Milton/libxml/share/doc/libxml2-python-2.6.4/examples 
; done)



This an issue?, or bug?



Greetings.




-- 
http://mail.python.org/mailman/listinfo/python-list

Re: 79 chars or more?

[BartC]

"Roy Smith"  wrote in message 
news:roy-181632.07571818082...@news.panix.com...

In article ,
"BartC"  wrote:

>> Remember, the old hardcopy terminals used to produce 
>> 132-character-wide

>> listings.
>
> Those of you who think "old hardcopy terminals" did 132 wide obviously
> don't remember the ASR-33 :-)

ASR33s I think might have been 72 columns wide (and punched cards had a
similar restriction).


Yeah, I was trying to remember if it was 72 or 80.  Hmmm, looks like
you're right, it *is* 72 (http://www.pdp8.net/asr33/asr33.shtml).

Punched cards (at least the common ones used by an 029 or 129 punch
machine) were 80 columns.  The 72 column restriction was an artificial
one imposed by some programming languages such as Fortran.  Columns
73-80 could be used to punch a sequence number, so that if you dropped
your deck, you could re-assemble it by running it through a card sorter.


I'm sure there was a continuation column too. That would mean long lines had 
to be split up, but if the width was longer, that would not be necessary.



However, lineprinter output was more likely to be 132 columns.


Yeah, but I wouldn't call a line printer a "terminal".


Source code tended to be perused and marked up on a printout, then corrected 
at a terminal. So the terminal's width was less important, until fast VDUs 
came in then printouts were used less, and it made sense to adjust to common 
25x80 displays.


(I tend to use 60x100 now, sometimes even wider; editing using 25x80 now is 
like doing keyhole surgery...)


--
bartc 


--
http://mail.python.org/mailman/listinfo/python-list

Re: subprocess.Popen calling httpd reload never finishes

[Nan]

On Aug 17, 8:14 pm, Albert Hopkins  wrote:
> On Tue, 2010-08-17 at 12:55 -0700, Nan wrote:
> > Hi folks --
>
> > I have a Python script running under Apache/mod_wsgi that needs to
> > reload Apache configs as part of its operation.  The script continues
> > to execute after the subprocess.Popen call.  The communicate() method
> > returns the correct text ("Reloading httpd: [  OK  ]"), and I get a
> > returncode of 0.  But the python script (Django) that calls Popen
> > never seems to complete (by returning an HTTP response.
>
> > Any other Popen call I've tried exits properly.  Here's some sample
> > code:
>
> >            args = ['sudo /etc/init.d/httpd reload']
> >            proc = subprocess.Popen(args, stdin=subprocess.PIPE,
> > stdout=subprocess.PIPE, stderr=subprocess.STDOUT, shell=True,
> > close_fds=True)
> >            (stdout_txt, stderr_txt) = proc.communicate("")
> >            proc.wait()
> >            logging.debug('%d %s%s' % (proc.returncode, stdout_txt,
> > stderr_txt))
> >            logging.debug('still executing')
> >            return HttpResponse('done')
>
> > The logging statements are output, but the script doesn't exit.  If
> > you substitute "sudo ls -l" or "sudo /etc/init.d/httpd configtest" for
> > "sudo /etc/init.d/httpd reload", the exits properly.
>
> >  Any idea what I might be doing wrong?
>
> > Thanks!
>
> Django runs inside apache.  It's kinda weird to have an apache process
> restart itself and expect it to return to the caller.
>
> If the init script does like mine, "reload" executes "apachectl -k
> graceful"   What that instructs apache to do is to restart, but only
> kill the process(es) when there are no more connections.  So apache is
> waiting for your connection to close, but you are inside an HTTP request
> waiting for apache to restart.  So you have a race condition here.
>
> It's not advisable to have apache kill itself and expect it to send a
> status back to you telling you it's dead.
>
> See the apache docs[1] for a better explanation.
>
> http://httpd.apache.org/docs/2.0/stopping.html#graceful

Ah, I'd been told that there would be no conflict, and that this was
just reloading the configuration, not restarting Apache.

I do need the web app to instruct Apache to reload because just before
this it's creating new VirtualHosts that need to be recognized.  Is
there a better way to do this (e.g. to say "start doing this once I'm
finished")?

I'm getting a status code and output from the call before the Django
script stops executing... Is there a way to stop waiting for the
process to complete once I have those?
-- 
http://mail.python.org/mailman/listinfo/python-list

Re: 79 chars or more?

[AK]

On 08/18/2010 05:11 AM, Stefan Schwarzer wrote:

Hi Lie,

On 2010-08-18 12:02, Lie Ryan wrote:

On 08/17/10 12:59, AK wrote:

On 08/16/2010 10:42 PM, James Mills wrote:

My personal opinion (despite monitors being wider) is
the horizontal scrolling isn't worth it. Stick to a 80-char width.


But.. why horizontal scrolling, isn't autowrap much better than that?


Do you seriously use autowrapper when writing code?


I think he means wrapping on the screen, not actually
inserting line breaks.

Stefan


Yes, exactly (:set wrap in Vim). -andrei
--
http://mail.python.org/mailman/listinfo/python-list

Re: Simple Problem but tough for me if i want it in linear time

[Frederic Rentsch]

On Mon, 2010-08-16 at 23:17 +, Steven D'Aprano wrote:
> On Mon, 16 Aug 2010 20:40:52 +0200, Frederic Rentsch wrote:
> 
> > How about
> > 
>  [obj for obj in dataList if obj.number == 100]
> > 
> > That should create a list of all objects whose .number is 100. No need
> > to cycle through a loop. 
> 
> What do you think the list comprehension does, if not cycle through a 
> loop?
> 
> 
> -- 
> Steven

What I think is that list comprehensions cycle through a loop a lot
faster than a coded loop (for n in ...:). As at the time of my post only
coded loops had been proposed and the OP was concerned about speed, I
thought I'd propose a list comprehension. I guess my explanation was
poorly phrased. Thanks for the reminder.

Frederic


-- 
http://mail.python.org/mailman/listinfo/python-list

Re: Python 2.7 support for PyWin32

[Tim Golden]

On 18/08/2010 12:54, paulo.jpi...@gmail.com wrote:

Hi everyone,

does anyone know when PyWin is going to support Python 2.7?

I tried to look for information, but to no avail.

Thanks in advance,
Paulo


It already does and has done for a while:

  http://sourceforge.net/projects/pywin32/files/

TJG
--
http://mail.python.org/mailman/listinfo/python-list

Re: 79 chars or more?

[D'Arcy J.M. Cain]

On Wed, 18 Aug 2010 23:18:06 +1200
Lawrence D'Oliveiro  wrote:
> Might I suggest (guessing at the argument keywords here) :
> 
> self.expiration_date = translate_date \
>   (
> TheText = find(response, 'MPNExpirationDate').text,
> ToFormat ='%Y-%m-%d',
> FromFormat ='%m%d%Y'
>   )

I sometimes use a backslash continuation but it always feels like a
failure to me.  It's an irrational reaction and I'm not sure why I feel
like that.  However, if you are going to do it in order to line up the
parenthese I would prefer this style.

 self.expiration_date = translate_date \
 (
 TheText = find(response, 'MPNExpirationDate').text,
 ToFormat ='%Y-%m-%d',
 FromFormat ='%m%d%Y'
 )

This way you can see not only the block structure at a glance but also
the block starter that it is part of.

> 
> Presumably FromFormat should be localizable, rather than hard-coded.
> 
> See, that’s the kind of thing you notice when you think about the code in 
> this way.

The other thing that jumps out at me is having the input format
different than the output format.  In any case you need a better date
input function.  There's no reason in this day and age to force users
into a particular input form.  You should think about creating a
utility function that converts any date that is unambiguous.  My
scripts generally accept all of the following.

Sep 1 2010
september 1, 2010
2010-9-1 (I have never seen "Y/D/M" format)
2010/9/1
2010 9 1
12/25/2010
25/12/2010
2010/12/25

It fails on the following.

sep 31 2010 (impossible)
2010/25/12 (impossible - Y/D/M never happens)
9/1/2010 (ambiguous - there is no consistiency when year is last field)
foo (not a date)

-- 
D'Arcy J.M. Cain  |  Democracy is three wolves
http://www.druid.net/darcy/|  and a sheep voting on
+1 416 425 1212 (DoD#0082)(eNTP)   |  what's for dinner.
-- 
http://mail.python.org/mailman/listinfo/python-list

Re: 79 chars or more?

[D'Arcy J.M. Cain]

On Wed, 18 Aug 2010 10:57:00 +0200
Jean-Michel Pichavant  wrote:
> D'Arcy J.M. Cain wrote:
> > You can extend this if there are complicated sub-calls.  Probably
> > overkill for this example but here is the idea.
> >
> >self.expiration_date = translate_date(
> >  find(
> >response,
> >   'MPNExpirationDate',
> >  ).text,
> >  '%Y-%m-%d',
> >  '%m%d%Y'
> >)
> >
> > I also moved the closing brace down to align with the line that opened
> > that block.
> >   
> If this is supposed to convice 80+ chars users, that's an epic failure :)
> This is exactly the kind of layout I'm happy to not use by not caring 
> about the line width.

As I said above, this was overkill designed to show the idea.  Of
course I would never split up short calls like that in real code.

-- 
D'Arcy J.M. Cain  |  Democracy is three wolves
http://www.druid.net/darcy/|  and a sheep voting on
+1 416 425 1212 (DoD#0082)(eNTP)   |  what's for dinner.
-- 
http://mail.python.org/mailman/listinfo/python-list

Required Buy Palm Jumeirah aprt AND Springs villa, 050-8320722

[PETER WONG F H (+971 50 8320722)]

Dear

I have a client who is looking to buy apartment and Villa :
1.  the palm jumeirah 3 bed + maid room sea view only (any building)
2.  springs type 3E
3.  springs type 3M
4.  springs type 4E

Please send me your direct availabilities, or call me, viewing
tomorrow, thank you!


Peter Wong F.H   王福兴
(个人注册RERA BRN: 8866)

Mob  手提   : +971   50 83 20 722
Fax传真   : +971   4 32 30 895
E-mail   电邮   : peterwfh2...@gmail.com
Company 公司   : Pinky Real Estate Broker冰奇房地产 (公司注册RERA ORN: 1866)
Website:  http://groups.google.com/group/dubai-property-club
-- 
http://mail.python.org/mailman/listinfo/python-list

Re: Python 2.7 support for PyWin32

[paulo.jpi...@gmail.com]

Thanks for the heads up.

My error was to only look for the green download button. There you
still get 2.6 as
default download.

--
Paulo

On Aug 18, 4:28 pm, Tim Golden  wrote:
> On 18/08/2010 12:54, paulo.jpi...@gmail.com wrote:
>
> > Hi everyone,
>
> > does anyone know when PyWin is going to support Python 2.7?
>
> > I tried to look for information, but to no avail.
>
> > Thanks in advance,
> > Paulo
>
> It already does and has done for a while:
>
>    http://sourceforge.net/projects/pywin32/files/
>
> TJG

-- 
http://mail.python.org/mailman/listinfo/python-list

Re: subprocess.Popen calling httpd reload never finishes

[Albert Hopkins]

On Wed, 2010-08-18 at 06:58 -0700, Nan wrote:
> Ah, I'd been told that there would be no conflict, and that this was
> just reloading the configuration, not restarting Apache.
> 
> I do need the web app to instruct Apache to reload because just before
> this it's creating new VirtualHosts that need to be recognized.  Is
> there a better way to do this (e.g. to say "start doing this once I'm
> finished")?
> 
> I'm getting a status code and output from the call before the Django
> script stops executing... Is there a way to stop waiting for the
> process to complete once I have those? 

I have a wireless router with a built in web server.  Sometimes it needs
to "reload" it's config.  Basically what it's doing is rebooting the
entire router (I can see this if I'm actuall watching the router).  All
it is doing, I'm pretty sure, is calling some program that forks another
process and then exits the main program.  The forked process then
reboots the router.  Meanwhile before that happens the web server sends
a response. Basically in the response it sends an HTTP Refresh with x
number of seconds.  Presumably x is longer than the time it requires for
the router to reboot.  The router reboots, the browser refreshes and
viola.

You probably need to so something similar in that your request calls a
program that forks off and restarts apaches.  It should probably not do
so immediately so that your request has time to send a response with a
refresh header (but that shouldn't take long).  After a second or so,
apache will have restarted and the browser will have refreshed.

so (untested):

def reload(request):
subprocess.call(['my_apache_reloader']) # this should fork and exit
response = HttpResponse()
response['Refresh']='3; url=%s' % reverse(home) # chk back in 3 secs
return response

BTW There is a Django mailing list where this might be more appropriate
to discuss.

-a


-- 
http://mail.python.org/mailman/listinfo/python-list

Problem Creating NewLines in PDF

[Andrew Evans]

Hello I am generating a PDF in web2py but its ignoring my line breaks.

 randname = random.randrange(1, 10001)
styles = getSampleStyleSheet()
title = "My Title"
doc = SimpleDocTemplate("primer.pdf")
story = []
story.append(Paragraph(strftime("%a, %d %b %Y %H:%M:%S",
gmtime()),styles["Heading2"]))
para = ParagraphStyle(name="output", fontName='Helvetica',
fontSize=12)
story.append(Paragraph(str(result_list), para))
doc.build(story)
response.headers['Content-Type']='application/pdf'
response.headers['Content-Disposition'] =
'attachment;filename='+str(randname)+'-.pdf'
return response.stream(open("primer.pdf", 'rb'))


result_list is a generated list. The pdf ignores line breaks and outputs it
as a list so ['  '] are included also. Any idea how I can create line breaks
and get rid of the ['  ']

*cheers
-- 
http://mail.python.org/mailman/listinfo/python-list

Re: Problem Creating NewLines in PDF

[MRAB]

Andrew Evans wrote:

Hello I am generating a PDF in web2py but its ignoring my line breaks.

 randname = random.randrange(1, 10001)
styles = getSampleStyleSheet()
title = "My Title"
doc = SimpleDocTemplate("primer.pdf")
story = []
story.append(Paragraph(strftime("%a, %d %b %Y %H:%M:%S", 
gmtime()),styles["Heading2"]))
para = ParagraphStyle(name="output", fontName='Helvetica', 
fontSize=12)

story.append(Paragraph(str(result_list), para))


str(result_list) is converting the list into a string. Did you mean to
do that?


doc.build(story)
response.headers['Content-Type']='application/pdf'
response.headers['Content-Disposition'] = 
'attachment;filename='+str(randname)+'-.pdf'

return response.stream(open("primer.pdf", 'rb'))


result_list is a generated list. The pdf ignores line breaks and outputs 
it as a list so ['  '] are included also. Any idea how I can create line 
breaks and get rid of the ['  ']
 


--
http://mail.python.org/mailman/listinfo/python-list

Using re.sub with %s

[Brandon Harris]

Having trouble using %s with re.sub

test = '/my/word/whats/wrong'
re.sub('(/)word(/)', r'\1\%s\2'%'1000', test)

return is /my/@0/whats/wrong

however if I cast a value with letters as opposed to numbers

re.sub('(/)word(/)', r'\1\%s\2'%'gosh', test)

return is /my/gosh/whats/wrong


Any help would be good. I've tried passing the value as an int, or 
recasting that value as something else, passing it as a raw string, 
removing the r and just double escaping the groups.


Brandon L. Harris

--
http://mail.python.org/mailman/listinfo/python-list

Re: 79 chars or more?

[Neil Cerutti]

On 2010-08-18, D'Arcy J.M. Cain  wrote:
> The other thing that jumps out at me is having the input format
> different than the output format.  In any case you need a
> better date input function.  There's no reason in this day and
> age to force users into a particular input form.  You should
> think about creating a utility function that converts any date
> that is unambiguous.  My scripts generally accept all of the
> following.

Under the hood, translate_date just use strptime and strftime, so
the formats are whatever those functions support. Moreover, I'm
converting one known format (the one in the XML) to another
required format (the one required by another program). There's no
need to guess the format.

-- 
Neil Cerutti
-- 
http://mail.python.org/mailman/listinfo/python-list

Re: looping through possible combinations of McNuggets packs of 6, 9 and 20

[cbr...@cbrownsystems.com]

On Aug 17, 2:44 pm, Baba  wrote:
> On Aug 16, 6:28 pm, "cbr...@cbrownsystems.com"
>
>  wrote:
> > First, suppose d = gcd(x, y, z); then for some x', y', z' we have that
> > x = d*x', y = d*y', z = d*z'; and so for any a, b, c:
>
>    could you explain the notation?
>
>    what is the difference btw x and x' ?
>
>    what is x = d*x', y supposed to say?

x', y', z' are names for three natural numbers; I could have chosen r,
s, t. "x=d*x'" above simply notes that since x is divisible by d, it
can be written as the product of d and some other natural number, and
the same is true for both y and z. therefore sums of multiples of x, y
and z are always divisible by d.

>
> > To go the other way, if d = 1, then there exists integers (not
> > neccessarily positive) such that
>
> > a*x + b*y + c*z = 1
>
>    what's the link with 6*a+9*b+20*c=n except the similarity?
>

The link is that it shows that if we have some u, v, and w with

6*u + 9*v + 20*w = n,

and we can find some a, b, and c which satisfy

6*a + 9*b + 20*c = 1

then if we let r = u + a, s = v + b, and t = w + c, we get that

6*r + 9*s + 20*t = n+1

although r, s, and t are not neccessarily positive numbers (as they
must be to solve your original problem). However, if u, v, and w are
sufficiently large compared to a, b, and c, then r, s and t WILL all
be positive.

But we can only find such an a,b, and c because the gcd of 6, 9, and
20 is equal to 1; that is why you can't solve this problem for nugget
pack sizes 6, 12, and 21.

Note that if there is one solution (a,b,c) to the gcd equation, there
infinitely many tuples (a,b,c) which satisfy the gcd equation, for
example:

6*0+ 9*9+ 20*(-4) = 1
6*(-5) + 9*(-1) + 20*2= 1
6*2+ 9*1+ 20*(-1) = 1

So the proof I gave regarded the /existence/ of a largest
unobtainable, not an algorithm for obtaining one. However from the
last of those three examples, we can see (details are in my original
proof) that the largest unobtainable must be less than

6*0+ 9*0+ 20*(1*5) = 100

so it is potentially helpful for finding an upper bound to the
problem.

> furthermore i came across this:
>
> For k = 3, efficient algorithms
> have been given by Greenberg and Davison ; if x1 < x2 < x3, these
> algorithms run in
> time bounded by a polynomial in log x3. Kannan  gave a very
> complicated algorithm
> that runs in polynomial time in log xk if k is fixed, but is wildly
> exponential in k. However,
> Ram´ırez Alfons´ın proved that the general problem is NP-hard, under
> Turing reductions,
> by reducing from the integer knapsack problem. So it seems very likely
> that there is no
> simple formula for computing g(x1, x2, . . . , xk) for arbitrary k.
>
> source:http://arxiv.org/PS_cache/arxiv/pdf/0708/0708.3224v1.pdf
>
> i would be interested in the answer to problem 3: explain in English
> why the theorem is true
>

I haven't looked at the link; but to be honest it's unlikely you would
understand it if you are having trouble with the much simpler question
regarding solutions in the case of 6, 12, and 21.

Cheers - Chas
-- 
http://mail.python.org/mailman/listinfo/python-list

Re: Problem Creating NewLines in PDF

[Andrew Evans]

Hello yes

This line doesn't seem to want to accept a list for some strange reason


story.append(Paragraph(str(result_list), para))

*cheers
-- 
http://mail.python.org/mailman/listinfo/python-list

Re: looping through possible combinations of McNuggets packs of 6, 9 and 20

[Baba]

Hi Chas

Thanks for that and i agree on your last remark :)

re the number of required consecutive passes required:

The number of required consecutive passes is equal to the smallest
number because after that you can get any amount of nuggets by just
adding the smallest nugget pack to some other number.

This is only true if gcd(a,b,c)=1.

Thanks to all for the help in getting to the bottom of the exercise. I
have truly enjoyed this and most importantly i have learned some new
things. Hadn't really done any mathematics in a long time and only
starting programming so this was good to get up to speed.

kind regards to everyone!
Baba



-- 
http://mail.python.org/mailman/listinfo/python-list

Re: Using re.sub with %s

[Thomas Jollans]

On Wednesday 18 August 2010, it occurred to Brandon Harris to exclaim:
> Having trouble using %s with re.sub
> 
> test = '/my/word/whats/wrong'
> re.sub('(/)word(/)', r'\1\%s\2'%'1000', test)
> 
> return is /my/@0/whats/wrong
> 

This has nothing to do with %, of course:

>>> re.sub('(/)word(/)', r'\1\%d\2'%1000, test)
'/my/@0/whats/wrong'
>>> re.sub('(/)word(/)', r'\1\1000\2', test)
'/my/@0/whats/wrong'

let's see if we can get rid of that zero:

>>> re.sub('(/)word(/)', r'\1\100\2', test)
'/my/@/whats/wrong'

so '\100' appears to be getting replaced with '@'. Why?

>>> '\100'
'@'

This is Python's way of escaping characters using octal numbers.

>>> chr(int('100', 8))
'@'

How to avoid this? Well, if you wanted the literal backslash, you'll need to 
escape it properly:

>>> print(re.sub('(/)word(/)', r'\1\\1000\2', test))
/my/\1000/whats/wrong
>>> 

If you didn't want the backslash, then why on earth did you put it there? You 
have to be careful with backslashes, they bite ;-)

Anyway, you can simply do the formatting after the match.

>>> re.sub('(/)word(/)', r'\1%d\2', test) % 1000
'/my/1000/whats/wrong'
>>> 

Or work with match objects to construct the resulting string by hand.

 - Thomas
-- 
http://mail.python.org/mailman/listinfo/python-list

expression in an if statement

[ernest]

Hi,

In this code:

if set(a).union(b) == set(a): pass

Does Python compute set(a) twice?
Thanks in advance.

Ernest
-- 
http://mail.python.org/mailman/listinfo/python-list

Re: expression in an if statement

[Peter Otten]

ernest wrote:

> In this code:
> 
> if set(a).union(b) == set(a): pass
> 
> Does Python compute set(a) twice?

>>> a = "abc"
>>> b = "def"
>>> _set = set
>>> def set(x):
... print "computing set(%r)" % x
... return _set(x)
...
>>> if set(a).union(b) == set(a): pass
...
computing set('abc')
computing set('abc')

So yes, set(a) is computed twice.

Peter
-- 
http://mail.python.org/mailman/listinfo/python-list

Re: looping through possible combinations of McNuggets packs of 6, 9 and 20

[cbr...@cbrownsystems.com]

On Aug 18, 10:52 am, Baba  wrote:
> Hi Chas
>
> Thanks for that and i agree on your last remark :)
>
> re the number of required consecutive passes required:
>
> The number of required consecutive passes is equal to the smallest
> number because after that you can get any amount of nuggets by just
> adding the smallest nugget pack to some other number.
>
> This is only true if gcd(a,b,c)=1.
>
> Thanks to all for the help in getting to the bottom of the exercise. I
> have truly enjoyed this and most importantly i have learned some new
> things. Hadn't really done any mathematics in a long time and only
> starting programming so this was good to get up to speed.
>
> kind regards to everyone!
> Baba

Happy to be of service!

Cheers - Chas
-- 
http://mail.python.org/mailman/listinfo/python-list

Re: How far can stack [LIFO] solve do automatic garbage collection and prevent memory leak ?

[Alex McDonald]

On 18 Aug, 11:09, spinoza  wrote:
> On Aug 18, 1:21 am, Standish P  wrote:

>
> > This you might want to take this to the Forth people because they are
> > marketing their language as a cure for all that plagues programming
> > today.
>
> No, they're not.

That I agree with.

> Stack based languages have seen better days and Forth
> (and the SL/1 language I supported with compilers at Bell-Northern
> Research) were last in fashion in the 1970s. Processors seldom could
> multitask, so it wasn't recognized that the stack could be a
> performance bottleneck, where stack operations cannot be pipelined or
> executed in parallel.
>
> John Hennessy of Stanford and MIPS made the stack must die case at ACM
> ASPLOS in 1987. Niklaus Wirth was also at this conference at which I
> was a fly on the wall, maintaining that the stack was good for
> reliability and verifiability of software.
>
> Forth had a snowball's chance because it forces ordinary programmers
> to think in Reverse Polish notation and is for the above reasons hard
> to pipeline, although of course it can be pipelined.

I really don't understand much of what you're saying here; Forth can
be implemented on processors that have several hardware assisted
stacks, 1 stack or even no stack at all. Multitasking? Why's that a
problem? And why is it hard to pipeline? Are you thinking of a
specific processor?





-- 
http://mail.python.org/mailman/listinfo/python-list

Re: Using re.sub with %s

[MRAB]

Thomas Jollans wrote:

On Wednesday 18 August 2010, it occurred to Brandon Harris to exclaim:

Having trouble using %s with re.sub

test = '/my/word/whats/wrong'
re.sub('(/)word(/)', r'\1\%s\2'%'1000', test)

return is /my/@0/whats/wrong



This has nothing to do with %, of course:


re.sub('(/)word(/)', r'\1\%d\2'%1000, test)

'/my/@0/whats/wrong'

re.sub('(/)word(/)', r'\1\1000\2', test)

'/my/@0/whats/wrong'

let's see if we can get rid of that zero:


re.sub('(/)word(/)', r'\1\100\2', test)

'/my/@/whats/wrong'

so '\100' appears to be getting replaced with '@'. Why?


'\100'

'@'

This is Python's way of escaping characters using octal numbers.


chr(int('100', 8))

'@'

How to avoid this? Well, if you wanted the literal backslash, you'll need to 
escape it properly:



print(re.sub('(/)word(/)', r'\1\\1000\2', test))

/my/\1000/whats/wrong

If you didn't want the backslash, then why on earth did you put it there? You 
have to be careful with backslashes, they bite ;-)


Anyway, you can simply do the formatting after the match.


re.sub('(/)word(/)', r'\1%d\2', test) % 1000

'/my/1000/whats/wrong'

Or work with match objects to construct the resulting string by hand.


You can stop group references which are followed by digits from turning
into octal escapes in the replacement template by using \g instead:

>>> print r'\1%s' % '00'
\100
>>> print r'\g<1>%s' % '00'
\g<1>00
--
http://mail.python.org/mailman/listinfo/python-list

Re: looping through possible combinations of McNuggets packs of 6, 9 and 20

[John Posner]

On 8/18/2010 1:38 PM, cbr...@cbrownsystems.com wrote:


To go the other way, if d = 1, then there exists integers (not
neccessarily positive) such that



a*x + b*y + c*z = 1


That fact is non-trivial, although the proof isn't *too* hard [1]. I 
found it interesting to demonstrate the simpler case (a*x + b*y = 1) by 
"instrumenting" the classic Python implementation of Euclid's Algorithm:


def show_gcd(a,b):
"""
find GCD of two integers, showing intermediate steps
in both remainder and linear-combination forms
"""
while b:
if a%b > 0:
rem_form = "%d == %d*(%d), rem %d" % (a, b, a/b, a%b)
equ_form = "%d == %d*(1) + %d*(-%d)" % (a%b, a, b, a/b)
print "%3d %3d %-30s %s" % (a,b, rem_form, equ_form)
a,b = b, a%b
print "\nGCD is", a


>>> show_gcd(124, 39)
124  39 124 == 39*(3), rem 7   7 == 124*(1) + 39*(-3)
 39   7 39 == 7*(5), rem 4 4 == 39*(1) + 7*(-5)
  7   4 7 == 4*(1), rem 3  3 == 7*(1) + 4*(-1)
  4   3 4 == 3*(1), rem 1  1 == 4*(1) + 3*(-1)


Performing successive substitutions, bottom to top, using the equations 
in the right-hand column:


  1 == 4*(1) + 3*(-1)

== 4*(1) + (7*(1) + 4*(-1))*(-1)

== 4*(2) + 7*(-1)

== (39*(1) + 7*(-5))*(2) + 7*(-1)

== 39*(2) + 7*(-11)

== 39*(2) + (124*(1) + 39*(-3))*(-11)

== 39*(35) + 124*(-11)

What could be simpler!  :-)

-John


[1] http://math453fall2008.wikidot.com/lecture-3 ("GCD as a linear 
combonation" [sic])


--
http://mail.python.org/mailman/listinfo/python-list

Re: Problem Creating NewLines in PDF

[MRAB]

Andrew Evans wrote:

Hello yes

This line doesn't seem to want to accept a list for some strange reason


story.append(Paragraph(str(result_list), para))


From the documentation it appears that you need to pass a string.

You're just passing the result of str(result_list), which isn't giving
you what you want:

>>> result_list = ['hello', 'world']
>>> print str(result_list)
['hello', 'world']

But what do you want? Do you just want to concatenate the entries into a
single string (assuming they're all strings)?

>>> print "".join(result_list)
helloworld

Or with some kind of separator between them?

>>> print " ".join(result_list)
hello world
--
http://mail.python.org/mailman/listinfo/python-list

Re: Problem Creating NewLines in PDF

[Andrew Evans]

Hello ty for the fast replies

This is the string I am using for the PDF I was able to create new lines
using the HTML  "br" tag which is what I wanted a method to create new lines

search_str="Position: (%d) - Keyword: (%s) - Domain (%s) " %
(idx+1, target_keyword, session.target_domain)
result_list.append(search_str)

 however it maintains these characters in the text of the generated PDF

['
', '
 '] with the string in between those

I just want to be able to out put the string with no extra characters I
assume these characters are there because its a list



..
>From the documentation it appears that you need to pass a string.

You're just passing the result of str(result_list), which isn't giving
you what you want:
-- 
http://mail.python.org/mailman/listinfo/python-list

Re: Python 2.7 support for PyWin32

[Terry Reedy]

On 8/18/2010 8:33 AM, Mark Lawrence wrote:

On 18/08/2010 12:54, paulo.jpi...@gmail.com wrote:

Hi everyone,

does anyone know when PyWin is going to support Python 2.7?

I tried to look for information, but to no avail.

Thanks in advance,
Paulo


It was created on 2009-07-08!!! See:-

http://sourceforge.net/projects/pywin32/files/pywin32/


Considering that that is months before the first alpha release and a 
year before the final of 2.7, truly amazing.



--
Terry Jan Reedy

--
http://mail.python.org/mailman/listinfo/python-list

Re: Problem Creating NewLines in PDF

[MRAB]

Andrew Evans wrote:

Hello ty for the fast replies

This is the string I am using for the PDF I was able to create new lines 
using the HTML  "br" tag which is what I wanted a method to create new lines


search_str="Position: (%d) - Keyword: (%s) - Domain (%s) " % 
(idx+1, target_keyword, session.target_domain)

result_list.append(search_str)

 however it maintains these characters in the text of the generated PDF

['
', '
 '] with the string in between those

I just want to be able to out put the string with no extra characters I 
assume these characters are there because its a list



As I said, if you just want to join a list of strings into one string,
then use "".join(result_list):

story.append(Paragraph("".join(result_list), para))
--
http://mail.python.org/mailman/listinfo/python-list

Re: How far can stack [LIFO] solve do automatic garbage collection and prevent memory leak ?

[Elizabeth D Rather]

On 8/18/10 12:09 AM, spinoza wrote:

On Aug 18, 1:21 am, Standish P  wrote:

Garbage collection doesn't use a stack. It uses a "heap", which is in
the abstract a collection of memory blocks of different lengths,
divided into two lists, generally represented as linked lists:



1.  A list of blocks that are free and may be used to store new data



2.  A list of blocks that are in use, or haven't been freed (yet)


Is this all that a heap is or is there more to it ? I have been
looking for simple but complete explanation of heap for a while and
not gotten to it. I think I am looking for a stack allocation on the
same pattern. In a disk, a file is fragmented in many contiguous
blocks and is accessed automatically.


Stacks (at least as far as Forth uses them) and heaps are fundamentally 
different things.


...

However, data structures of variable size, or data structures that
merely take up a lot of space, don't play nice with others on the
stack, so, we place their address on the stack and store them in
another place, which was named the heap, probably, as a sort of
witticism.


In Forth, they go in "data space", which might or might not be in the 
dictionary, and is almost never in a dynamically managed heap; certainly 
not on a stack.

...



No, they're not. Stack based languages have seen better days and Forth
(and the SL/1 language I supported with compilers at Bell-Northern
Research) were last in fashion in the 1970s. Processors seldom could
multitask, so it wasn't recognized that the stack could be a
performance bottleneck, where stack operations cannot be pipelined or
executed in parallel.


Lol.  Forth supported multitasking on every processor it was implemented 
on in the 70's, with blazing speed compared to competitive techniques. 
I have never seen stack operations to be a bottleneck.


...

Forth had a snowball's chance because it forces ordinary programmers
to think in Reverse Polish notation and is for the above reasons hard
to pipeline, although of course it can be pipelined.


Mostly it had a "snowball's chance" because it was never picked up by 
the CS gurus who, AFAIK, never really took a serious look at it.


Cheers,
Elizabeth

--
==
Elizabeth D. Rather   (US & Canada)   800-55-FORTH
FORTH Inc. +1 310.999.6784
5959 West Century Blvd. Suite 700
Los Angeles, CA 90045
http://www.forth.com

"Forth-based products and Services for real-time
applications since 1973."
==
--
http://mail.python.org/mailman/listinfo/python-list

Re: scipy / stats : quantiles using sample weights from survey data

[Aahz]

In article <94bb6313-1b09-4eeb-9969-07d76048a...@m35g2000prn.googlegroups.com>,
Christopher Barrington-Leigh   wrote:
>
>There is a function scipy.stats.mstats.mquantiles  that returns
>quantiles for a vector of data.
>But my data should not be uniformly weighted in an estimate of the
>distribution, since they are from a survey and come with estimated
>sampling weights based on the stratification used in sampling.
>
>Is there a routine to calculate these quantiles taking into account
>the survey weights?  I can find nothing, so maybe you have had the
>same problem and written something.

You should ask on the scipy mailing list.
-- 
Aahz (a...@pythoncraft.com)   <*> http://www.pythoncraft.com/

"...if I were on life-support, I'd rather have it run by a Gameboy than a
Windows box."  --Cliff Wells
-- 
http://mail.python.org/mailman/listinfo/python-list

Re: assigning variables from list data

[Aahz]

In article ,
Chris Hare   wrote:
>
>cursor.execute('select * from net where NetNumber > 0')

Unless your table is guaranteed to never change layout, I suggest that
instead listing fields is a Good Idea:

cursor.execute('select netNumber, netType,  from net where NetNumber > 0')

Then the tuple unpack suggested by other posters will never fail.
-- 
Aahz (a...@pythoncraft.com)   <*> http://www.pythoncraft.com/

"...if I were on life-support, I'd rather have it run by a Gameboy than a
Windows box."  --Cliff Wells
-- 
http://mail.python.org/mailman/listinfo/python-list

Re: looping through possible combinations of McNuggets packs of 6, 9 and 20

[cbr...@cbrownsystems.com]

On Aug 18, 11:50 am, John Posner  wrote:
> On 8/18/2010 1:38 PM, cbr...@cbrownsystems.com wrote:
>
> >>> To go the other way, if d = 1, then there exists integers (not
> >>> neccessarily positive) such that
>
> >>> a*x + b*y + c*z = 1
>
> That fact is non-trivial, although the proof isn't *too* hard [1]. I
> found it interesting to demonstrate the simpler case (a*x + b*y = 1)...

And to get the more general case, if we write (a,b) for gcd of and b,
we can think of the "," as a binary operator that you can show is
associative:

((a,b), c) = (a, (b,c)) = (a, b, c)

and so a proof that exists x,y with a*x + b*y = (a,b) can then be
extended to a proof for an arbitrary number of elements.

(Oddly, "," is also distributive over itself: ((a,b), c) = ((a,c),
(b,c))...)

Cheers - Chas
-- 
http://mail.python.org/mailman/listinfo/python-list

Re: when 'myArray * 'myObject' is not equal to 'myObject' * 'myArray'

[Nobody]

On Wed, 18 Aug 2010 05:56:27 -0700, Duim wrote:

> Although I'm sure somewhere this issue is discussed in this (great)
> group, I didn't know the proper search words for it (although I
> tried).
> 
> I'm using python (2.6) scientifically mostly, and created a simple
> class to store time series (my 'Signal' class).
> I need this class to have a possibility to get multiplied by an array,
> but pre and post multiplication have different mathematical outcomes
> ( basically A* B != B*A ) .
> 
> Post multiplication by an array works fine defining __mul__ in the
> Signal class, but pre multiplication does not. It keeps trying to
> multiply all elements separately instead to send this array to my
> __rmul__ function.
> 
> How can I fix this without the need for a separate
> 'multiplysignal(A,B)' function?

Make Signal a subclass of numpy.ndarray. If one operand is a subclass of
the other, its __rmul__ will be preferred to the parent's __mul__.

In the absence of a subclass-superclass relationship, the LHS's __mul__ is
preferred to the RHS's __rmul__, so the RHS's __rmul__ is only called if
the LHS lacks a __mul__ method or if the method refuses its argument
(returns NotImplemented).

Likewise for other "reflected" methods.

-- 
http://mail.python.org/mailman/listinfo/python-list

Re: subprocess.Popen calling httpd reload never finishes

[Nan]

On Aug 18, 12:37 pm, Albert Hopkins  wrote:
> On Wed, 2010-08-18 at 06:58 -0700, Nan wrote:
> > Ah, I'd been told that there would be no conflict, and that this was
> > just reloading the configuration, not restarting Apache.
>
> > I do need the web app to instruct Apache to reload because just before
> > this it's creating new VirtualHosts that need to be recognized.  Is
> > there a better way to do this (e.g. to say "start doing this once I'm
> > finished")?
>
> > I'm getting a status code and output from the call before the Django
> > script stops executing... Is there a way to stop waiting for the
> > process to complete once I have those?
>
> I have a wireless router with a built in web server.  Sometimes it needs
> to "reload" it's config.  Basically what it's doing is rebooting the
> entire router (I can see this if I'm actuall watching the router).  All
> it is doing, I'm pretty sure, is calling some program that forks another
> process and then exits the main program.  The forked process then
> reboots the router.  Meanwhile before that happens the web server sends
> a response. Basically in the response it sends an HTTP Refresh with x
> number of seconds.  Presumably x is longer than the time it requires for
> the router to reboot.  The router reboots, the browser refreshes and
> viola.
>
> You probably need to so something similar in that your request calls a
> program that forks off and restarts apaches.  It should probably not do
> so immediately so that your request has time to send a response with a
> refresh header (but that shouldn't take long).  After a second or so,
> apache will have restarted and the browser will have refreshed.
>
> so (untested):
>
> def reload(request):
>     subprocess.call(['my_apache_reloader']) # this should fork and exit
>     response = HttpResponse()
>     response['Refresh']='3; url=%s' % reverse(home) # chk back in 3 secs
>     return response
>
> BTW There is a Django mailing list where this might be more appropriate
> to discuss.
>
> -a

Sadly, questions about subprocess spawning on the Django lists seem to
get referred back to the Python lists when they get an answer at all.
You've been very helpful, though, so thank you.

I think I'll have to mark in the database when the server needs
restarting, and run a cron job to check that and reload Apache from a
non-Apache process.  It will mean delayed creation of the Virtual
Hosts and no immediate feedback to indicate success or failure, but it
doesn't look like there's much alternative.
-- 
http://mail.python.org/mailman/listinfo/python-list

q.join() is probably the wrong method for you

[Raymond Hettinger]

The join() method is all about waiting for all the tasks to be done.  If you 
don't care whether the tasks have actually finished, you can periodically poll 
the unfinished task count:

  stop = time() + timeout
  while q.unfinished_tasks and time() < stop:
sleep(1)

This loop will exist either when the tasks are done or when the timeout period 
has elapsed.


Raymond

 

> On Thursday, June 17, 2010 5:52 PM pacopyc wrote:

> Hi, I am trying to work with threads and I need your help. This is
> code:
> 
> from threading import Thread
> from Queue import Queue
> import time
> import random
> 
> def test_fun (k,q,t):
> time.sleep(t)
> print "hello world from thread " + str(q.get()) + " (sleep time =
> " + str(t) + " sec.)"
> q.task_done()
> 
> queue = Queue()
> for i in range (1,10):
> queue.put(i)
> for j in range(queue.qsize()):
> num = random.randint(1,30)
> worker = Thread(target=test_fun, args=(j,queue,num))
> worker.setDaemon(True)
> worker.start()
> queue.join()
> 
> 
> Execution:
> 
> hello world from thread 1 (sleep time = 5 sec.)
> hello world from thread 2 (sleep time = 5 sec.)
> hello world from thread 3 (sleep time = 6 sec.)
> hello world from thread 4 (sleep time = 8 sec.)
> hello world from thread 5 (sleep time = 10 sec.)
> hello world from thread 6 (sleep time = 13 sec.)
> hello world from thread 7 (sleep time = 18 sec.)
> hello world from thread 8 (sleep time = 19 sec.)
> hello world from thread 9 (sleep time = 20 sec.)
> 
> Some questions for you:
> 
> 1) Why order is always the same (thread 1, thread 2, thread 3 
> thread 9) and also seconds are always increasing? I do not understand.
> 2) I'd like to decide a max time for each thread. If max time = 7 sec.
> I want to print only threads with sleep time <= 7 sec. How can I do?
> Can you modify my code?
> 
> Thank you very much


>> On Thursday, June 17, 2010 7:04 PM MRAB wrote:

>> pacopyc wrote:
>> 
>> 1)
>> 
>> First it puts the numbers 1..9 into 'queue', then it starts 9 threads,
>> giving each a number 'num'.
>> 
>> Each thread waits for 'num' seconds ('t' in the thread).
>> 
>> The thread with the lowest value of 'num' wakes first, gets the first
>> entry from 'queue' (the value 1), and therefore prints "thread 1".
>> 
>> The thread with the second-lowest value of 'num' wakes next, gets the
>> second entry from 'queue' (the value 2), and therefore prints "thread
>> 2".
>> 
>> And so on.
>> 
>> 2)
>> 
>> If a thread is given a value of 'num' of more than a maximum, that
>> thread should not print its output, but it should still get the entry
>> from the queue (assuming that you want it to still behave the same
>> otherwise).


>>> On Friday, June 18, 2010 5:45 PM pacopyc wrote:

>>> =3D
>>> .
>>> 
>>> Ok, the problem is that I want fix a time max for each thread. For
>>> example run 10 threads (each can terminate its work in 10 sec. max
>>> fixed time) and wait them. If they finish its work in < 10 sec. (for
>>> example 2 sec.) very good ... go on immediately (do not wait
>>> unnecessary time), but if a thread use more than 10 sec. stop wait (I
>>> kill it)  when all threads have finished their work or when they
>>> have used all their available time (10 sec.) the program must go on
>>> (do not wait).


 On Friday, June 18, 2010 6:27 PM MRAB wrote:

 pacopyc wrote:
 [snip]
 
 it is not possible to kill a thread. If you want a thread to have a
 maximum time, the thread must check occasionally how long it has been
 running and terminate if necessary.
 
 Another programming language (Java) originally had the ability to kill
 threads, but that was later deprecated because it caused problems due to
 not knowing what the thread was doing when it was killed (it might have
 been in the middle of updating something at the time, for example,
 leaving the system in an inconsistent state).


> On Friday, June 18, 2010 6:58 PM pacopyc wrote:

> Ok, I understand. But is possible fix max wait time (for example 60
> sec.) in main thread and check queue for task done and at the same
> time the remaining time. When all task have done or wait time has
> expired main thread go on. Is it possible? Can you write code?
> Thank you very much.


>> On Friday, June 18, 2010 7:45 PM MRAB wrote:

>> pacopyc wrote:
>> 
>> The documentation says that queue.join() cannot have a timeout, so you
>> might have to think of another way to achieve the same effect.
>> 
>> You could, for example, have a results queue into which a thread puts
>> something to indicate when it has finished, and then use the .get method
>> on it in the main thread (the .get method can have a timeout).


>> Submitted via EggHeadCafe - Software Developer Portal of Choice 
>> Custom Favorites Web Site with MongoDb and NoRM
>> http://www.eggheadcafe.com/tutorials/aspnet/7fbc7a01-5d30-4cd3-b373-51d4a0e1afa8/custom-favorites-web-site-with-mongodb-and-norm.aspx
-- 
http

Re: How far can stack [LIFO] solve do automatic garbage collection and prevent memory leak ?

[Paul Rubin]

Elizabeth D Rather  writes:
>> Processors seldom could multitask, so it wasn't recognized that the
>> stack could be a performance bottleneck
> Lol.  Forth supported multitasking on every processor it was
> implemented on in the 70's, with blazing speed compared to competitive
> techniques. I have never seen stack operations to be a bottleneck.

I think "multitasking" in that post refers to superscalar execution,
which wasn't done in the 1970's except on supercomputers.  That the
stack is a bottleneck is the precise reason that optimizing Forth
compilers do complicated flow analysis to translate stack operations
into register operations.
-- 
http://mail.python.org/mailman/listinfo/python-list

Re: Python "why" questions

[AK]

On 08/17/2010 10:15 PM, Russ P. wrote:

On Aug 7, 5:54 am, "D'Arcy J.M. Cain"  wrote:


Would said beginner also be surprised that a newborn baby is zero years
old or would it be more natural to call them a one year old?  Zero
based counting is perfectly natural.


You're confusing continuous and discrete variables. Time is a
continuous variable, but a list index is discrete.

Take a look at any numbered list, such as the top ten football teams
or the top ten software companies. Have you ever seen such a list
start with zero? If so, where? I sure haven't.

When I studied linear algebra way back, vector and matrix indices also
always started with one, and I assume they still do.

The convention of starting with zero may have had some slight
performance advantage in the early days of computing, but the huge
potential for error that it introduced made it a poor choice in the
long run, at least for high-level languages.


I have to agree, there's innumerable number of examples where sequential
number of an item in a series is counted starting with one. Second loaf
of bread; third day of vacation, first cup of tea today, first gray
hair, 50th anniversary, 2nd century AD, and approximately a gazillion
other examples.

Contrast this with _one_ example that was repeated in this thread of
there being ground floor, 1st floor, 2nd, and so on. However! Consider
that ground floor is kind of different from the other floors. It's the
floor that's not built up over ground, but is already there -- in case
of the most primitive dwelling, you can put some sawdust over the
ground, put a few boards overhead and it's a "home", although probably
not a "house". But does it really have what can be officially called a
"floor"?

On a more practical angle, ground floors usually have the lobby,
receptionists, storefronts and stores, etc; while 1st floor and up are
business/residential.

I think different numbering from pretty much all other things out there
gives you a hint that the ground floor is a different animal.

 -andrei
--
http://mail.python.org/mailman/listinfo/python-list

Re: Python "why" questions

[AK]

On 08/17/2010 10:15 PM, Russ P. wrote:

On Aug 7, 5:54 am, "D'Arcy J.M. Cain"  wrote:


Would said beginner also be surprised that a newborn baby is zero years
old or would it be more natural to call them a one year old?  Zero
based counting is perfectly natural.


You're confusing continuous and discrete variables. Time is a
continuous variable, but a list index is discrete.

Take a look at any numbered list, such as the top ten football teams
or the top ten software companies. Have you ever seen such a list
start with zero? If so, where? I sure haven't.

When I studied linear algebra way back, vector and matrix indices also
always started with one, and I assume they still do.

The convention of starting with zero may have had some slight
performance advantage in the early days of computing, but the huge
potential for error that it introduced made it a poor choice in the
long run, at least for high-level languages.


Besides that, the way things are now, it's almost an Abbot & Costello
routine:

- How many folders are there?
- 5
- Ok, give me the fourth one.
- Here.
- No, that's the last one!
- That's what you said!
- No, I said, fourth one!
- That's what I did!
- How many are there in all?
- I already said, five!
- You gave me the last one!!
- Just like you said - fourth
--
http://mail.python.org/mailman/listinfo/python-list

Re: expression in an if statement

[John Nagle]

On 8/18/2010 11:24 AM, ernest wrote:

Hi,

In this code:

if set(a).union(b) == set(a): pass

Does Python compute set(a) twice?


   CPython does.  Shed Skin might optimize.  Don't know
about Iron Python.

John Nagle
--
http://mail.python.org/mailman/listinfo/python-list

Re: How far can stack [LIFO] solve do automatic garbage collection and prevent memory leak ?

[John Nagle]

On 8/18/2010 1:32 PM, Paul Rubin wrote:

Elizabeth D Rather  writes:

Processors seldom could multitask, so it wasn't recognized that the
stack could be a performance bottleneck

Lol.  Forth supported multitasking on every processor it was
implemented on in the 70's, with blazing speed compared to competitive
techniques. I have never seen stack operations to be a bottleneck.


I think "multitasking" in that post refers to superscalar execution,
which wasn't done in the 1970's except on supercomputers.  That the
stack is a bottleneck is the precise reason that optimizing Forth
compilers do complicated flow analysis to translate stack operations
into register operations.


   Some small FORTH machines had dedicated stack hardware.  On each
CPU cycle, the CPU could do one stack access, one main memory access,
and one return stack access.  This was before cacheing; those CPUs
were slow relative to their memory, so a non-cached 
one-instruction-per-clock machine made sense.


   In the superscalar era, there's not much of an advantage to avoiding
stack accesses.  x86 superscalar machines have many registers not
visible to the program, as the fastest level of cache.  In practice,
the top of the stack is usually in CPU registers.   The "huge number
of programmer-visible register" machines like SPARCs turned out to be
a dead end.  So did making all the instructions the same width; it
makes the CPU simpler, but not faster, and it bulks up the program
by 2x or so.

John Nagle

--
http://mail.python.org/mailman/listinfo/python-list

Re: Python "why" questions

[Russ P.]

On Aug 18, 2:01 pm, AK  wrote:
> On 08/17/2010 10:15 PM, Russ P. wrote:
>
>
>
> > On Aug 7, 5:54 am, "D'Arcy J.M. Cain"  wrote:
>
> >> Would said beginner also be surprised that a newborn baby is zero years
> >> old or would it be more natural to call them a one year old?  Zero
> >> based counting is perfectly natural.
>
> > You're confusing continuous and discrete variables. Time is a
> > continuous variable, but a list index is discrete.
>
> > Take a look at any numbered list, such as the top ten football teams
> > or the top ten software companies. Have you ever seen such a list
> > start with zero? If so, where? I sure haven't.
>
> > When I studied linear algebra way back, vector and matrix indices also
> > always started with one, and I assume they still do.
>
> > The convention of starting with zero may have had some slight
> > performance advantage in the early days of computing, but the huge
> > potential for error that it introduced made it a poor choice in the
> > long run, at least for high-level languages.
>
> Besides that, the way things are now, it's almost an Abbot & Costello
> routine:
>
> - How many folders are there?
> - 5
> - Ok, give me the fourth one.
> - Here.
> - No, that's the last one!
> - That's what you said!
> - No, I said, fourth one!
> - That's what I did!
> - How many are there in all?
> - I already said, five!
> - You gave me the last one!!
> - Just like you said - fourth

Yes, it's confusing. Which element of a list is the "first" element?
Wait, "first" is sometimes abbreviated as "1st". So is the 1st element
the 0 element or the 1 element? I honestly don't know.

Is the top team in the league the number 1 team -- or the number 0
team? I have yet to hear anyone call the best team the number 0 team!

Unfortunately, we're stuck with this goofy numbering system in many
languages. Fortunately, the trend is away from explicit indexing and
toward "for" loops when possible.

-- 
http://mail.python.org/mailman/listinfo/python-list

Re: Python "why" questions

[Mark Lawrence]

On 18/08/2010 22:47, Russ P. wrote:

On Aug 18, 2:01 pm, AK  wrote:

On 08/17/2010 10:15 PM, Russ P. wrote:




On Aug 7, 5:54 am, "D'Arcy J.M. Cain"wrote:



Would said beginner also be surprised that a newborn baby is zero years
old or would it be more natural to call them a one year old?  Zero
based counting is perfectly natural.



You're confusing continuous and discrete variables. Time is a
continuous variable, but a list index is discrete.



Take a look at any numbered list, such as the top ten football teams
or the top ten software companies. Have you ever seen such a list
start with zero? If so, where? I sure haven't.



When I studied linear algebra way back, vector and matrix indices also
always started with one, and I assume they still do.



The convention of starting with zero may have had some slight
performance advantage in the early days of computing, but the huge
potential for error that it introduced made it a poor choice in the
long run, at least for high-level languages.


Besides that, the way things are now, it's almost an Abbot&  Costello
routine:

- How many folders are there?
- 5
- Ok, give me the fourth one.
- Here.
- No, that's the last one!
- That's what you said!
- No, I said, fourth one!
- That's what I did!
- How many are there in all?
- I already said, five!
- You gave me the last one!!
- Just like you said - fourth


Yes, it's confusing. Which element of a list is the "first" element?
Wait, "first" is sometimes abbreviated as "1st". So is the 1st element
the 0 element or the 1 element? I honestly don't know.

Is the top team in the league the number 1 team -- or the number 0
team? I have yet to hear anyone call the best team the number 0 team!

Unfortunately, we're stuck with this goofy numbering system in many
languages. Fortunately, the trend is away from explicit indexing and
toward "for" loops when possible.



Bring back Coral 66, all is forgiven.

http://www.xgc.com/manuals/xgc-c66-rm/x357.html

Cheers.

Mark Lawrence.

--
http://mail.python.org/mailman/listinfo/python-list

Re: exception handling with sqlite db errors

[CM]

On Aug 12, 3:31 pm, a...@pythoncraft.com (Aahz) wrote:
> In article 
> <2a47b306-45d1-474a-9f8e-5b71eba62...@p11g2000prf.googlegroups.com>,
>
> CM  wrote:
>
> >Maybe it's not much of an issue, but I think it would be a shame if
> >occasional hangs/crashes could be caused by these (rare?) database
> >conflicts if there is a good approach for avoiding them.  I guess I
> >could put every last write to the db in a try/except block but I
> >thought there should be a more general solution, since that will
> >require many such exceptions and seems inelegant.
>
> Wrap all your uses of sqlite into a function that does the try/except;
> you only write the code once, then.  As you progress, you can also
> change the code to retry operations.  Here's some ugly code I wrote on
> top of SQLObject:
>
> from sqlobject.dbconnection import registerConnection
> from sqlobject.sqlite.sqliteconnection import SQLiteConnection
>
> class RetrySQLiteConnection(SQLiteConnection):
>     """
>     Because SQLite is not really concurrent, having multiple processes
>     read/write can result in locked DB failures.  In addition, SQLObject
>     doesn't properly protect operations in transations, so you can get
>     spurious DB errors claiming that the DB is corrupt because of
>     foreign key integrity failures.
>
>     This subclass retries DatabaseError and OperationalError
>     exceptions.
>     """
>     MAX_RETRIES = 4
>     SAFE_DB_ERROR = [
>         'database disk image is malformed',
>         'file is encrypted or is not a database',
>         ]
>
>     def _safe_db_error(self,exception):
>         err = str(exception).lower()
>         for safe_err in self.SAFE_DB_ERROR:
>             if safe_err in err:
>                 return True
>         return False
>
>     def _check_integrity(self):
>         conn = self.getConnection()
>         try:
>             i = 0
>             while True:
>                 i += 1
>                 try:
>                     cursor = conn.cursor()
>                     query = "pragma integrity_check"
>                     SQLiteConnection._executeRetry(self, conn, cursor, query)
>                     result = cursor.fetchall()
>                     if result == [('ok',)]:
>                         return True
>                     else:
>                         logging.error("Bad integrity result: %s", result)
>                         return False
>                 except DatabaseError, e:
>                     if i < self.MAX_RETRIES:
>                         logging.info('integrity_check, try #%s: %s', i, e)
>                         time.sleep(2)
>                     else:
>                         logging.error('integrity_check, try #%s: %s', i, e)
>                         raise
>         finally:
>             self.releaseConnection(conn)
>
>     def _executeRetry(self, conn, cursor, query):
>         i = 0
>         while True:
>             i += 1
>             try:
>                 return SQLiteConnection._executeRetry(self, conn, cursor, 
> query)
>             except OperationalError, e:
>                 if i < self.MAX_RETRIES:
>                     logging.warn('OperationalError, try #%s: %s', i, e)
>                     time.sleep(10)
>                 else:
>                     logging.error('OperationalError, try #%s: %s', i, e)
>                     raise
>             except DatabaseError, e:
>                 if e.__class__ is not DatabaseError:
>                     # Don't retry e.g. IntegrityError
>                     raise
>                 if not self._safe_db_error(e):
>                     # Only retry specific errors
>                     raise
>                 if not self._check_integrity():
>                     raise
>                 if i < self.MAX_RETRIES:
>                     logging.warn('DatabaseError, try #%s: %s', i, e)
>                     time.sleep(0.5)
>                 else:
>                     logging.error('DatabaseError, try #%s: %s', i, e)
>                     raise
>
> def conn_builder():
>     return RetrySQLiteConnection
>
> registerConnection(['retrysqlite'], conn_builder)
>
> def init():
>     dbpath = os.path.join(common.getSyncDataPath(), app.dbname)
>     connection_string = "retrysqlite:" + dbpath
>     global _connection
>     _connection = connectionForURI(connection_string)
> --
> Aahz (a...@pythoncraft.com)           <*>        http://www.pythoncraft.com/
>
> "...if I were on life-support, I'd rather have it run by a Gameboy than a
> Windows box."  --Cliff Wells

Thanks, Aahz, I'll try to implement something along these lines.

Che
-- 
http://mail.python.org/mailman/listinfo/python-list

Re: expression in an if statement

[Thomas Jollans]

On Wednesday 18 August 2010, it occurred to John Nagle to exclaim:
> On 8/18/2010 11:24 AM, ernest wrote:
> > Hi,
> > 
> > In this code:
> > 
> > if set(a).union(b) == set(a): pass
> > 
> > Does Python compute set(a) twice?
> 
> CPython does.  Shed Skin might optimize.  Don't know
> about Iron Python.

I doubt any actual Python implementation optimizes this -- how could it? The 
object "set" is clearly being called twice, and it happens to be called with 
the object "a" as a sole argument twice. What if "set" has side effects? A 
compiler could only exclude this possibility if it knew exactly what "set" 
will be at run time, which it can't.

I expect that "set" and "a" have to be looked up twice, actually: 
"set(a).union(b)" might rebind either one of them. This would be considered a 
very rude and inappropriate thing to do, but Python usually guarantees to 
allow bad taste and behaviour.

I might be wrong on some points here, but this is what I expect the expression 
(set(a).union(b) == set(a)) has to do, in any conforming implementation of 
Python. Please correct me if I'm wrong.

  1. find out which object "set" refers to
  2. find out which object "a" refers to
  3. call (set) with the single positional argument (a), no keyword arguments
  4. get the attribute "union" of the return value of [3]
  5. find out which object "b" refers to
  6. call (.union) with the single positional argument (b).
  7. look up __eq__ in the __class__ of the return value of [6]
  8. find out which object "set" refers to
  9. find out which object "a" refers to
 10. call (set) with the single positional argument (a), no keyword arguments
 11. call [7] with two positional arguments: the return values [6] & [10]
 
I'm not 100% sure if there are any guarantees as to when (5) is taken care of 
-- what would happen if set(a) or even set(a).__getattr__ changed the global 
"b"?
My list there is obviously referring to Python 3.x, so there is no __cmp__ to 
worry about.
-- 
http://mail.python.org/mailman/listinfo/python-list

Re: Unsupported Format Character '&' (0x26)

[Andrew Evans]

nvm I got it by adding s and d respectively after each value eg %(query)s

thank you all

On Wed, Aug 18, 2010 at 4:35 PM, Andrew Evans  wrote:

> I get an error message "Unsupported Format Character '&' (0x26)" I narrowed
> it down to these two variables
>
> any idea how to fix it?
>
>   SEARCH_URL_0 = "
> http://search.yahoo.com/search;_ylt=A0oGdEf1XGxMJRoAUdml87UF;_ylc=X1MDMjE0MjQ3ODk0OARfcgMyBGZyA3NmcARuX2dwcwMxMARvcmlnaW4Dc3ljBHF1ZXJ5A3Rlc3QEc2FvAzE-?p=%(query)&fr=sfp&fr2=&iscqry=
> "
> NEXT_PAGE_0 = "
> http://search.yahoo.com/search;_ylt=A0oGdEEMXWxMnBAAiWhXNyoA?p=%(query)&fr=sfp&xargs=12KPjg1qNyy4-MkfqnfKqLCLLAhlMFta2E&pstart=%(start)&b=11
> "
>
>
>
> Not sure why its doing it the value in parenthesis should work %(query) etc
>
> Any ideas cheeers
>
-- 
http://mail.python.org/mailman/listinfo/python-list

Re: How far can stack [LIFO] solve do automatic garbage collection and prevent memory leak ?

[Standish P]

On Aug 18, 12:30 pm, Elizabeth D Rather  wrote:
> On 8/18/10 12:09 AM, spinoza wrote:
>
> > On Aug 18, 1:21 am, Standish P  wrote:
> >>> Garbage collection doesn't use a stack. It uses a "heap", which is in
> >>> the abstract a collection of memory blocks of different lengths,
> >>> divided into two lists, generally represented as linked lists:
>
> >>> 1.  A list of blocks that are free and may be used to store new data
>
> >>> 2.  A list of blocks that are in use, or haven't been freed (yet)
>
> >> Is this all that a heap is or is there more to it ? I have been
> >> looking for simple but complete explanation of heap for a while and
> >> not gotten to it. I think I am looking for a stack allocation on the
> >> same pattern. In a disk, a file is fragmented in many contiguous
> >> blocks and is accessed automatically.
>
> Stacks (at least as far as Forth uses them) and heaps are fundamentally
> different things.
>
> ...
>
> >>> However, data structures of variable size, or data structures that
> >>> merely take up a lot of space, don't play nice with others on the
> >>> stack, so, we place their address on the stack and store them in
> >>> another place, which was named the heap, probably, as a sort of
> >>> witticism.
>
> In Forth, they go in "data space", which might or might not be in the
> dictionary, and is almost never in a dynamically managed heap; certainly
> not on a stack.
> ...
>
>
>
> > No, they're not. Stack based languages have seen better days and Forth
> > (and the SL/1 language I supported with compilers at Bell-Northern
> > Research) were last in fashion in the 1970s. Processors seldom could
> > multitask, so it wasn't recognized that the stack could be a
> > performance bottleneck, where stack operations cannot be pipelined or
> > executed in parallel.
>

> Lol.  Forth supported multitasking on every processor it was implemented
> on in the 70's, with blazing speed compared to competitive techniques.
> I have never seen stack operations to be a bottleneck.
>

>
> > Forth had a snowball's chance because it forces ordinary programmers
> > to think in Reverse Polish notation and is for the above reasons hard
> > to pipeline, although of course it can be pipelined.
>
> Mostly it had a "snowball's chance" because it was never picked up by
> the CS gurus who, AFAIK, never really took a serious look at it.

Its quite possible that the criticism is unfair, but dont you think
that in part some responsibility must be borne by your organization in
not doing a good job of education ? I have looked at this book you
authored in the past few weeks and found a link for your convenience
now. This is entitled Advanced .

http://www.amazon.com/Forth-Application-Techniques-5th-Notebook/dp/1419685767/ref=sr_1_1?ie=UTF8&s=books&qid=1282175842&sr=8-1#reader_1419685767

Show me on what page does it explain how Forth implements dynamic
binding or lexical binding and takes care of the scope of definition
of the "nouns" ?

Provide me with a link, if you kindly would, that can take me to a
tutorial of Forth internals or discusses this issue.

> Cheers,
> Elizabeth

She is quite humble. Take a look at this page,

http://www.forth.com/resources/evolution/index.html

She is currently the number 1 in the forth world and if there was a
nobel prize in forth, it would go to these three.


Authors

Elizabeth D. Rather
FORTH, Inc.
5959 W. Century Blvd.
Suite 700
Los Angeles, CA 90045

Elizabeth Rather is the co-founder of FORTH, Inc. and is a leading
expert in the Forth programming language. Elizabeth was a colleague of
Chuck Moore back when he worked at NRAO in the early 1970s. During his
development of Forth, she became the second ever Forth programmer.
Since then, she has become a leading expert in the language and one of
its main proponents. Elizabeth was the chair of the ANSI Technical
Committee that produced the ANSI Standard for Forth (1994).  She is an
author of several books on Forth and gives regular training seminars
on its usage.

Donald R. Colburn

c/o Digital Media Magic
14712 Westbury Rd.
Rockville, MD 20853

Don Colburn was one of the earliest Forth users. He was one of the
founders of the Forth Interest Group, and contributed to the
development of the first public-domain figForth.  Subsequently, he
founded Creative Solutions, Inc. (CSI), which introduced MacForth™ in
1984. MacForth was the first programming language capable of running
on the Macintosh when it was first introduced.  Don was a member of
the ANSI Technical Committee that produced the ANSI Standard for Forth
(1994). He died in 2009.

Charles H. Moore
Computer Cowboys
40 Cedar Lane
P.O. Box 127
Sierra City, CA 96125

Chuck Moore is Chairman and CTO of Green Arrays, Inc. He co-founded
FORTH, Inc., in 1971 and went on to develop a Forth-based chip
(RTX2000) in the mid 1980s, derivatives of which are still being used
widely by NASA. At Computer Cowboys, Mr. Moore designed the Sh-Boom
microprocessor and then co-founded iTv, an Internet Applia

Re: How far can stack [LIFO] solve do automatic garbage collection and prevent memory leak ?

[Standish P]

On Aug 17, 6:38 pm, John Passaniti  wrote:

> You asked if Forth "borrowed" lists from Lisp.  It did not.  In Lisp,
> lists are constructed with pair of pointers called a "cons cell".
> That is the most primitive component that makes up a list.  Forth has
> no such thing; in Forth, the dictionary (which is traditionally, but
> not necessarily a list) is a data structure that links to the previous
> word with a pointer.  

Would you show me a picture, ascii art or whatever for Forth ? I know
what lisp lists look like so I dont need that for comparison. Forth
must have a convention and a standard or preferred practice for its
dicts. However, let me tell you that in postscript the dictionaries
can be nested inside other dictionaries and any such hiearchical
structure is a nested associative list, which is what linked list,
nested dictionaries, nested tables are.

> This is in fact one of the nice things about
> Lisp; because all lists are created out of the same primitive cons
> cell, you can consistently process any list in the system.  In Forth,
> any lists (such as the dictionary, if it is a list) are specific to
> their purpose and have to be treated individually.
>
> I don't know what you mean by "nested-dictionaries."  There is no such
> thing in Forth.  Dictionaries don't nest.  You can create wordlists,
> but each wordlist is flat.  When most people think of a nested
> dictionary, they would think of a structure that would allow any
> arbitrary level of nesting, not a string of flat wordlists.

-- 
http://mail.python.org/mailman/listinfo/python-list

Re: How far can stack [LIFO] solve do automatic garbage collection and prevent memory leak ?

[Keith Thompson]

Standish P  writes:
> On Aug 18, 12:30 pm, Elizabeth D Rather  wrote:
[...]
>> Mostly it had a "snowball's chance" because it was never picked up by
>> the CS gurus who, AFAIK, never really took a serious look at it.
>
> Its quite possible that the criticism is unfair, but dont you think
> that in part some responsibility must be borne by your organization in
> not doing a good job of education ?
[snip]
> Show me on what page does it explain how Forth implements dynamic
> binding or lexical binding and takes care of the scope of definition
> of the "nouns" ?
[...]

Show me how this is relevant to comp.lang.c, comp.lang.c++, comp.theory,
or comp.lang.python.  Please trim the Newsgroups line.

-- 
Keith Thompson (The_Other_Keith) ks...@mib.org  
Nokia
"We must do something.  This is something.  Therefore, we must do this."
-- Antony Jay and Jonathan Lynn, "Yes Minister"
-- 
http://mail.python.org/mailman/listinfo/python-list

How to see intermediate fail results from unittest as tests are running?

[Margie Roginski]

Hi,

I am using unittest in a fairly basic way, where I have a single file
that simply defines a class that inherits from unittest.TestCase and
then within that class I have a bunch of methods that start with
"test".  Within that file, at the bottom I have:

if __name__ == "__main__":
unittest.main()

This works fine and it runs all of the testxx() methods in my file.
As it runs it prints if the tests passed or failed, but if they fail,
it does not print the details of the assert that made them fail.  It
collects this info up and prints it all at the end.

Ok - my question: Is there any way to get unittest to print the
details of the assert that made a test fail, as the tests are
running?  IE, after a test fails, I would like to see why, rather than
waiting until all the tests are done.

I've searched the doc and even looked at the code, and it seems the
answer is no, but I'm just wondering if I'm missing something.

Thanks!

Margie
-- 
http://mail.python.org/mailman/listinfo/python-list

Re: Python "why" questions

[Dan Sommers]

On Wed, 18 Aug 2010 16:56:22 -0400, AK wrote:

> Contrast this with _one_ example that was repeated in this thread of
> there being ground floor, 1st floor, 2nd, and so on. However! Consider
> that ground floor is kind of different from the other floors. It's the
> floor that's not built up over ground, but is already there -- in case
> of the most primitive dwelling, you can put some sawdust over the
> ground, put a few boards overhead and it's a "home", although probably
> not a "house". But does it really have what can be officially called a
> "floor"?

That's the perfect example, although perhaps for an [apparently] 
unintended reason :  I think that the notion of a qualitatively 
different "ground floor" is European, or at least that's the way I 
remember it from my high school French class way back in the late 1970s.  
In the U.S., when you walk into a building (even a very tall commercial 
building), that's the first floor, and when you go up a level, that's the 
second floor, and all the room/suite/office numbers are two hundred and 
something.  I also seem to recall that some European buildings have a 
mezzanine floor between the ground floor and the floor whose reference 
number is 1, but again, high school was a long time ago.

Dan

-- 
http://mail.python.org/mailman/listinfo/python-list

Re: Simple Problem but tough for me if i want it in linear time

[Steven D'Aprano]

On Wed, 18 Aug 2010 16:03:58 +0200, Frederic Rentsch wrote:

> On Mon, 2010-08-16 at 23:17 +, Steven D'Aprano wrote:
>> On Mon, 16 Aug 2010 20:40:52 +0200, Frederic Rentsch wrote:
>> 
>> > How about
>> > 
>>  [obj for obj in dataList if obj.number == 100]
>> > 
>> > That should create a list of all objects whose .number is 100. No
>> > need to cycle through a loop.
>> 
>> What do you think the list comprehension does, if not cycle through a
>> loop?
>> 
>> 
>> --
>> Steven
> 
> What I think is that list comprehensions cycle through a loop a lot
> faster than a coded loop (for n in ...:). 

I think measurement beats intuition:


[st...@wow-wow ~]$ python -m timeit '[str(i) for i in xrange(10)]'
10 loops, best of 3: 84 msec per loop

[st...@wow-wow ~]$ python -m timeit 'L=[]
> for i in xrange(10):
> L.append(str(i))
> '
10 loops, best of 3: 105 msec per loop


But wait... we're not comparing apples with apples. There's an extra name 
lookup in the for-loop that the list comp doesn't have. We can fix that:


[st...@wow-wow ~]$ python -m timeit 'L=[]; append = L.append
for i in xrange(10):
append(str(i))
'
10 loops, best of 3: 86.7 msec per loop


The difference between 84 and 86 msec is essentially measurement error. 
Hell, the difference between 84 and 104 msec is not terribly significant 
either.



> As at the time of my post only
> coded loops had been proposed and the OP was concerned about speed, I
> thought I'd propose a list comprehension.

Yes, but the OP was concerned with asymptotic speed (big-oh notation), 
and both a for-loop and a list-comp are both O(N).

Frankly, I think the OP doesn't really know what he wants, other than 
premature optimization. It's amazing how popular that is :)



-- 
Steven
-- 
http://mail.python.org/mailman/listinfo/python-list

Re: Python "why" questions

[Steven D'Aprano]

On Wed, 18 Aug 2010 14:47:08 -0700, Russ P. wrote:

> Is the top team in the league the number 1 team -- or the number 0 team?
> I have yet to hear anyone call the best team the number 0 team!

Why is the top team the one with the lowest number?


> Unfortunately, we're stuck with this goofy numbering system in many
> languages. Fortunately, the trend is away from explicit indexing and
> toward "for" loops when possible.

Agreed on the second sentence there, but not on the first. There's 
nothing "goofy" about indexing items from 0. Yes, it does lead to slight 
more difficulty when discussing which item you want in *human* languages, 
but not in *programming* languages. The nth item is always the nth item. 
The only difference is whether n starts at 0 or 1, and frankly, if you 
(generic you, not you personally) can't learn which to use, you have no 
business pretending to be a programmer.



-- 
Steven
-- 
http://mail.python.org/mailman/listinfo/python-list

Re: How far can stack [LIFO] solve do automatic garbage collection and prevent memory leak ?

[Elizabeth D Rather]

On 8/18/10 2:23 PM, Standish P wrote:

On Aug 17, 6:38 pm, John Passaniti  wrote:


You asked if Forth "borrowed" lists from Lisp.  It did not.  In Lisp,
lists are constructed with pair of pointers called a "cons cell".
That is the most primitive component that makes up a list.  Forth has
no such thing; in Forth, the dictionary (which is traditionally, but
not necessarily a list) is a data structure that links to the previous
word with a pointer.


Would you show me a picture, ascii art or whatever for Forth ? I know
what lisp lists look like so I dont need that for comparison. Forth
must have a convention and a standard or preferred practice for its
dicts. However, let me tell you that in postscript the dictionaries
can be nested inside other dictionaries and any such hiearchical
structure is a nested associative list, which is what linked list,
nested dictionaries, nested tables are.


You indicated that you have a copy of Forth Application Techniques. 
Sections 8.1 and 8.2 cover this topic, with some drawings.


Cheers,
Elizabeth

--
==
Elizabeth D. Rather   (US & Canada)   800-55-FORTH
FORTH Inc. +1 310.999.6784
5959 West Century Blvd. Suite 700
Los Angeles, CA 90045
http://www.forth.com

"Forth-based products and Services for real-time
applications since 1973."
==
--
http://mail.python.org/mailman/listinfo/python-list

Re: Simple Problem but tough for me if i want it in linear time

[Tim Chase]

On 08/18/10 21:47, Steven D'Aprano wrote:

Frankly, I think the OP doesn't really know what he wants, other than
premature optimization. It's amazing how popular that is :)


You see, the trick to prematurely optimizing is to have a good 
algorithm for prematurely optimizing...the real question them 
becomes "How can I optimize my premature-optimization algorithms 
to O(1) instead of O(newsgroup)?"


:-)

-tkc


PS: I'm not positive, but O(newsgroup) may asymptotically 
approach O(log n) if the question is well formed, but O(2^n) if 
flaming, indentation/line-length preferences, the meaning of OOP, 
SQL-parameter escaping, McNugget combinations, or suggestions 
that Python is "just a scripting language" are involved...






--
http://mail.python.org/mailman/listinfo/python-list

Re: expression in an if statement

[Daniel Kluev]

On Thu, Aug 19, 2010 at 9:12 AM, Thomas Jollans  wrote:

> I doubt any actual Python implementation optimizes this -- how could it?
> The
> object "set" is clearly being called twice, and it happens to be called
> with
> the object "a" as a sole argument twice. What if "set" has side effects? A
> compiler could only exclude this possibility if it knew exactly what "set"
> will be at run time, which it can't.
>
> I expect that "set" and "a" have to be looked up twice, actually:
> "set(a).union(b)" might rebind either one of them. This would be considered
> a
> very rude and inappropriate thing to do, but Python usually guarantees to
> allow bad taste and behaviour.
>
>
Yep.

>>> def test():
... a = [1]
... b = [1]
... if set(a).union(b) == set(a): pass
...
>>> dis.dis(test)
  2   0 LOAD_CONST   1 (1)
  3 BUILD_LIST   1
  6 STORE_FAST   0 (a)

  3   9 LOAD_CONST   1 (1)
 12 BUILD_LIST   1
 15 STORE_FAST   1 (b)

  4  18 LOAD_GLOBAL  0 (set)
 21 LOAD_FAST0 (a)
 24 CALL_FUNCTION1
 27 LOAD_ATTR1 (union)
 30 LOAD_FAST1 (b)
 33 CALL_FUNCTION1
 36 LOAD_GLOBAL  0 (set)
 39 LOAD_FAST0 (a)
 42 CALL_FUNCTION1
 45 COMPARE_OP   2 (==)
 48 JUMP_IF_FALSE4 (to 55)
 51 POP_TOP
 52 JUMP_FORWARD 1 (to 56)
>>   55 POP_TOP
>>   56 LOAD_CONST   0 (None)
 59 RETURN_VALUE



> I might be wrong on some points here, but this is what I expect the
> expression
> (set(a).union(b) == set(a)) has to do, in any conforming implementation of
> Python. Please correct me if I'm wrong.
>

You can use dis module to let Python do compiling and explaining for you

-- 
With best regards,
Daniel Kluev
-- 
http://mail.python.org/mailman/listinfo/python-list

Re: Python "why" questions

[Russ P.]

On Aug 18, 7:58 pm, Steven D'Aprano  wrote:
> On Wed, 18 Aug 2010 14:47:08 -0700, Russ P. wrote:
> > Is the top team in the league the number 1 team -- or the number 0 team?
> > I have yet to hear anyone call the best team the number 0 team!
>
> Why is the top team the one with the lowest number?

How could it be otherwise? What is the highest number?

Here's a couple of things I'd like to see just once before I die:

1. The winner of the championship game chanting, "We're number zero!
We're number zero!

2. The loser of the championship game chanting, "We're number one!
We're number one!

>
> > Unfortunately, we're stuck with this goofy numbering system in many
> > languages. Fortunately, the trend is away from explicit indexing and
> > toward "for" loops when possible.
>
> Agreed on the second sentence there, but not on the first. There's
> nothing "goofy" about indexing items from 0. Yes, it does lead to slight
> more difficulty when discussing which item you want in *human* languages,
> but not in *programming* languages. The nth item is always the nth item.
> The only difference is whether n starts at 0 or 1, and frankly, if you
> (generic you, not you personally) can't learn which to use, you have no
> business pretending to be a programmer.

Maybe "goofy" was too derogatory, but I think you are rationalizing a
bad decision, at least for high-level languages. I don't think
programming languages should always mimic human languages, but this is
one case where there is no advantage to doing otherwise.

Why do you think "off by one" errors are so common? Because the darn
indexing convention is off by one!

And I'd still like to know if the "1st" element of aList is aList[0]
or aList[1].
-- 
http://mail.python.org/mailman/listinfo/python-list

Simple hack to get $5000 to your Paypal account

[Hot sex]

Simple hack to get $5000 to your Paypal account At http://simplelivevideos.tk

i have hidden the Paypal Form link in an image. in that website on
Right Side below search box, click on image and enter your name and
Paypal ID.
-- 
http://mail.python.org/mailman/listinfo/python-list

Re: expression in an if statement

[Frederic Rentsch]

On Thu, 2010-08-19 at 00:12 +0200, Thomas Jollans wrote:
> On Wednesday 18 August 2010, it occurred to John Nagle to exclaim:
> > On 8/18/2010 11:24 AM, ernest wrote:
> > > Hi,
> > > 
> > > In this code:
> > > 
> > > if set(a).union(b) == set(a): pass
> > > 
> > > Does Python compute set(a) twice?
> > 
> > CPython does.  Shed Skin might optimize.  Don't know
> > about Iron Python.
> 
> I doubt any actual Python implementation optimizes this -- how could it? 

And why should it if a programmer uses its facilities inefficiently. I
would write

>>> if set(a).issuperset (b): pass

Frederic



-- 
http://mail.python.org/mailman/listinfo/python-list