Programming Forums > Script Programming > Python

Arevos · Nov 25th, 2005, 1:51 PM

Here's a few Python code snippets I've used in several programs, that might, possibly, maybe, be useful for others. Does anyone else have any pieces of code that might make life easier for other Python programmers?

(Toggle Plain Text)

def all(s):
   "Return true if all elements in an iterable sequence are true."
   for x in s:
      if not x:
         return False
   return True

def any(s):
   "Return true if any element in an iterable sequence is true."
   for x in s:
      if x:
         return True
   return False

class Cache:
   """
   Creates a functor that caches the output of a function.
   e.g.
      image_load = Cache(pygame.image.load)
      ball1 = image_load("ball.png")
      ball2 = image_load("ball.png")   # ball.png only loaded from disk once
   """
   def __init__(self, func):
      self.store = {}
      self.func  = func

   def __call__(self, key):
      try:
         return self.store[key]
      except KeyError:
         self.store[key] = self.func(key)
         return self.store[key]

Sane · Nov 25th, 2005, 3:37 PM

Your all/any function could be easily generalised and used in more contexts like so:

(Toggle Plain Text)

def get_instances(list, instance):
    i = 0
    for item in list: i += item==instance
    return i

print get_instances( [0, 1, False, True], True )

As well, that "any" function can be summarized in a simple condition:

(Toggle Plain Text)

if True in list: print "At least one true is in the list."

(and same to your "all" function: if False in list)

Nice Cache function though.

Arevos · Nov 25th, 2005, 4:23 PM

Quote:

Originally Posted by Sane

Your all/any function could be easily generalised and used in more contexts like so:

(Toggle Plain Text)

def get_instances(list, instance):
    i = 0
    for item in list: i += item==instance
    return i

print get_instances( [0, 1, False, True], True )

That's not the point; they're syntax sugar for generators:

(Toggle Plain Text)

assert not any(x > 5 for x in s):
assert all(x <= 5 for x in s)

And get_instances isn't lazy evaluating, either...

Quote:

Originally Posted by Sane

As well, that "any" function can be summarized in a simple condition:

(Toggle Plain Text)

if True in list: print "At least one true is in the list."

(and same to your "all" function: if False in list)

You have a point there. I took the any and all functions from a proposal by Guido (he was arguing against reduce). Because Guido wrote it, I assumed (True in s) was not used because it would cause s to be evaluated fully, but some quick tests show this is not the case. I guess he wasn't thinking of efficiency at the time :o

Still, (True in (x > 5 for x in s)) doesn't seem as intuitive as any(x > 5 for x in s).

(Toggle Plain Text)

def any(s):
   return True in s

def all(s):
   return not False in s

Arevos · Nov 25th, 2005, 6:02 PM

A general-purpose event queue that I sometimes use:

(Toggle Plain Text)

class EventQueue:
   """
   Contains a queue of events. When polled, the event queue calls the
   functions associated with the events' classes.

   e.g.
      class Event:
         pass
      queue = EventQueue()

      @queue.register(Event)
      def foobar(event):
         print event

      queue.push(Event())
      queue.poll()
   """
   def __init__(self):
      self.map = {}
      self.queue = []

   def register(self, event_type, function = None):
      """
      Use as a function: register(event_type, function)
      or as a decorator: @register(event_type)
    
      """
      def decorate(func):
         try:
            self.map[event_type].add(func)
         except KeyError:
            self.map[event_type] = set([func])
         return func
         
      if function != None:
         return decorate(function)
         
      return decorate

   def push(self, event):
      "Push an event onto the queue."
      self.queue.append(event)

   def poll(self):
      "Poll the queue."
      events = self.queue[:]
      self.queue = []
      for event in events:
         for func in self.map[event.__class__]:
            func(event)

Sane · Nov 27th, 2005, 1:42 PM

I've noticed I've used this one quite often in my code:

(Toggle Plain Text)

def get_between(string, start, end):
    return string.split(start)[-1].split(end)[0]

example:

(Toggle Plain Text)

print get_between("blah <a href='http://www.google.com'>blah</a> blah", 
    "href='", "'>")
>>> http://www.google.com

Sane · Nov 27th, 2005, 4:44 PM

This snippet is useful for about every important program that you'd need to distribute to the public, for when needs to be kept updated (and is potentially full of bugs):

(Toggle Plain Text)

def show_error(caption, fileName):

    from tkMessageBox import showwarning          # error dialog box
    from sys import exc_info, exit                # exception info, system exit
    from time import ctime                        # current time and day

    error = '\n'.join(str(i) for i in exc_info()) # change all elements of exc_info to str, then join with newline

    stream = open(fileName, 'a')                  # open stream with append
    stream.write("%s: %s\n\n"%( ctime(), error )) # append error and time to an error log
    stream.close()                                # close stream

    showwarning( caption, error )                 # show error
    exit()                                        # exit program

It is used like so:

(Toggle Plain Text)

def main():
    print "This is our main function"
    print "Oops, this line is an error: %s"%(x)

try:
    main()
except SystemExit: pass
except:
    show_error('Program Error', 'errors.log')

And if your program is not using tkinter, then you just replace the 'showwarning' with a call to
win32api.MessageBox(0, message, caption)

Cerulean · Nov 28th, 2005, 4:21 PM

Quote:

Originally Posted by Sane

And if your program is not using tkinter, then you just replace the 'showwarning' with a call to
win32api.MessageBox(0, message, caption)

And what if it's not on Windows? Trying to show a message box with PyQt/wxPython/PyGTK would be cool, but a fallback of logging to a file is more or less a must.

On the debug topic, I had a generic debug function that worked some magic to find the name of the function that just called it and other interesting things. Will look around to see if I have it still.

Dietrich · Dec 5th, 2005, 9:39 AM

Quote:

On the debug topic, I had a generic debug function that worked some magic to find the name of the function that just called it and other interesting things. Will look around to see if I have it still.
____________________
http://cerulean.pyresoft.com

Yes, we are waiting with bated breath!

Sane · Mar 9th, 2006, 5:06 PM

Here's a semi-useful function I created when I ran into a problem while parsing html.

(Toggle Plain Text)

def splitwithout(text, s, void):
    res = []
    lt = -1
    on = (False,'')
    for t in range(len(text)):
        if text[t] in void:
            if not on[0]:
                on = (True, text[t])
            elif text[t] == on[1]:
                on = (False, '')
        if text[t] == s and not on[0]:
            res += [text[lt+1:t]]
            lt = t
    res += [text[lt+1:]]
    return res

It does what its name says, splits without (the instance of the items passed through void).

e.g.

(Toggle Plain Text)

print splitwithout("""a b "c d" e f 'g h i' j k l""",
                   " ",
                   ('"',"'"))

outputs:

>>> ['a', 'b', '"c d"', 'e', 'f', "'g h i'", 'j', 'k', 'l']

Notice how the spaces in between "c d" and 'g h i' were not split as split() would normally do (since I passed '"', "'" through void)? The point of this function is so I can do split an html tag by its spacing, without taking a space that could be inside the paramaters as well.

I wonder if this is actually faster then split() is.

Edit:

If you still don't see the point, here's what I was using it for.

(Toggle Plain Text)

def parse_tag(text, open = '<', close = '>'):
    first = splitwithout(text.split(open)[1].split(close)[0], ' ', ('"',"'"))
    
    tag = first[0]
    if tag == ' ':
        return False
    
    params = {}
    for p in first[1:]:
        pp = p.split('=')
        try:
            param = pp[0]
            val = pp[1].split(pp[1][0])[1]
        except IndexError: return False
        params[pp[0]] = val

    try:
        content = text.split(open+"/"+tag+close)[0].split(close)[1]
    except IndexError:
        return False

    return tag, params, content

tag, params, content = parse_tag("""<a href='http://google.ca' title="Hyperlink to Google.ca">Hyperlink</a>""")

print tag #'a'
print params #{'href': 'http://google.ca', 'title': 'Hyperlink to Google.ca'}
print content #Hyperlink

Very helpful I think.

hydroxide · Mar 9th, 2006, 7:01 PM

Quote:

Originally Posted by Sane

(Toggle Plain Text)

def parse_tag(text, open = '<', close = '>'):

Were you aware of htmllib? It's in the stdlib and is built for parsing html...
(the documentation for it is fairly awful, though, so you'd have to google for examples of how to use it)

--OH.

Nov 25th, 2005, 1:51 PM	#1
Arevos Programming Guru Join Date: Aug 2005 Location: England Posts: 1,499 Rep Power: 4	Useful code snippets Here's a few Python code snippets I've used in several programs, that might, possibly, maybe, be useful for others. Does anyone else have any pieces of code that might make life easier for other Python programmers? (Toggle Plain Text) def all(s): "Return true if all elements in an iterable sequence are true." for x in s: if not x: return False return True def any(s): "Return true if any element in an iterable sequence is true." for x in s: if x: return True return False class Cache: """ Creates a functor that caches the output of a function. e.g. image_load = Cache(pygame.image.load) ball1 = image_load("ball.png") ball2 = image_load("ball.png") # ball.png only loaded from disk once """ def __init__(self, func): self.store = {} self.func = func def __call__(self, key): try: return self.store[key] except KeyError: self.store[key] = self.func(key) return self.store[key] def all(s): "Return true if all elements in an iterable sequence are true." for x in s: if not x: return False return True def any(s): "Return true if any element in an iterable sequence is true." for x in s: if x: return True return False class Cache: """ Creates a functor that caches the output of a function. e.g. image_load = Cache(pygame.image.load) ball1 = image_load("ball.png") ball2 = image_load("ball.png") # ball.png only loaded from disk once """ def __init__(self, func): self.store = {} self.func = func def __call__(self, key): try: return self.store[key] except KeyError: self.store[key] = self.func(key) return self.store[key]

Nov 25th, 2005, 3:37 PM	#2
Sane Programming Guru Join Date: Apr 2005 Posts: 1,785 Rep Power: 5	Your all/any function could be easily generalised and used in more contexts like so: (Toggle Plain Text) def get_instances(list, instance): i = 0 for item in list: i += item==instance return i print get_instances( [0, 1, False, True], True ) def get_instances(list, instance): i = 0 for item in list: i += item==instance return i print get_instances( [0, 1, False, True], True ) As well, that "any" function can be summarized in a simple condition: (Toggle Plain Text) if True in list: print "At least one true is in the list." if True in list: print "At least one true is in the list." (and same to your "all" function: if False in list) Nice Cache function though. __________________ Waterloo's Canadian Computing Competition (CCC) - Stage 2 Problems, Solutions, and Test Data Last edited by Sane; Nov 25th, 2005 at 3:53 PM.

Nov 25th, 2005, 6:02 PM	#4
Arevos Programming Guru Join Date: Aug 2005 Location: England Posts: 1,499 Rep Power: 4	A general-purpose event queue that I sometimes use: (Toggle Plain Text) class EventQueue: """ Contains a queue of events. When polled, the event queue calls the functions associated with the events' classes. e.g. class Event: pass queue = EventQueue() @queue.register(Event) def foobar(event): print event queue.push(Event()) queue.poll() """ def __init__(self): self.map = {} self.queue = [] def register(self, event_type, function = None): """ Use as a function: register(event_type, function) or as a decorator: @register(event_type) """ def decorate(func): try: self.map[event_type].add(func) except KeyError: self.map[event_type] = set([func]) return func if function != None: return decorate(function) return decorate def push(self, event): "Push an event onto the queue." self.queue.append(event) def poll(self): "Poll the queue." events = self.queue[:] self.queue = [] for event in events: for func in self.map[event.__class__]: func(event) class EventQueue: """ Contains a queue of events. When polled, the event queue calls the functions associated with the events' classes. e.g. class Event: pass queue = EventQueue() @queue.register(Event) def foobar(event): print event queue.push(Event()) queue.poll() """ def __init__(self): self.map = {} self.queue = [] def register(self, event_type, function = None): """ Use as a function: register(event_type, function) or as a decorator: @register(event_type) """ def decorate(func): try: self.map[event_type].add(func) except KeyError: self.map[event_type] = set([func]) return func if function != None: return decorate(function) return decorate def push(self, event): "Push an event onto the queue." self.queue.append(event) def poll(self): "Poll the queue." events = self.queue[:] self.queue = [] for event in events: for func in self.map[event.__class__]: func(event)

Nov 27th, 2005, 1:42 PM	#5
Sane Programming Guru Join Date: Apr 2005 Posts: 1,785 Rep Power: 5	I've noticed I've used this one quite often in my code: (Toggle Plain Text) def get_between(string, start, end): return string.split(start)[-1].split(end)[0] def get_between(string, start, end): return string.split(start)[-1].split(end)[0] example: (Toggle Plain Text) print get_between("blah <a href='http://www.google.com'>blah</a> blah", "href='", "'>") >>> http://www.google.com print get_between("blah <a href='http://www.google.com'>blah</a> blah", "href='", "'>") >>> http://www.google.com __________________ Waterloo's Canadian Computing Competition (CCC) - Stage 2 Problems, Solutions, and Test Data

Nov 27th, 2005, 4:44 PM	#6
Sane Programming Guru Join Date: Apr 2005 Posts: 1,785 Rep Power: 5	This snippet is useful for about every important program that you'd need to distribute to the public, for when needs to be kept updated (and is potentially full of bugs): (Toggle Plain Text) def show_error(caption, fileName): from tkMessageBox import showwarning # error dialog box from sys import exc_info, exit # exception info, system exit from time import ctime # current time and day error = '\n'.join(str(i) for i in exc_info()) # change all elements of exc_info to str, then join with newline stream = open(fileName, 'a') # open stream with append stream.write("%s: %s\n\n"%( ctime(), error )) # append error and time to an error log stream.close() # close stream showwarning( caption, error ) # show error exit() # exit program def show_error(caption, fileName): from tkMessageBox import showwarning # error dialog box from sys import exc_info, exit # exception info, system exit from time import ctime # current time and day error = '\n'.join(str(i) for i in exc_info()) # change all elements of exc_info to str, then join with newline stream = open(fileName, 'a') # open stream with append stream.write("%s: %s\n\n"%( ctime(), error )) # append error and time to an error log stream.close() # close stream showwarning( caption, error ) # show error exit() # exit program It is used like so: (Toggle Plain Text) def main(): print "This is our main function" print "Oops, this line is an error: %s"%(x) try: main() except SystemExit: pass except: show_error('Program Error', 'errors.log') def main(): print "This is our main function" print "Oops, this line is an error: %s"%(x) try: main() except SystemExit: pass except: show_error('Program Error', 'errors.log') And if your program is not using tkinter, then you just replace the 'showwarning' with a call to win32api.MessageBox(0, message, caption) __________________ Waterloo's Canadian Computing Competition (CCC) - Stage 2 Problems, Solutions, and Test Data Last edited by Sane; Nov 27th, 2005 at 5:00 PM.

Mar 9th, 2006, 5:06 PM	#9
Sane Programming Guru Join Date: Apr 2005 Posts: 1,785 Rep Power: 5	Here's a semi-useful function I created when I ran into a problem while parsing html. (Toggle Plain Text) def splitwithout(text, s, void): res = [] lt = -1 on = (False,'') for t in range(len(text)): if text[t] in void: if not on[0]: on = (True, text[t]) elif text[t] == on[1]: on = (False, '') if text[t] == s and not on[0]: res += [text[lt+1:t]] lt = t res += [text[lt+1:]] return res def splitwithout(text, s, void): res = [] lt = -1 on = (False,'') for t in range(len(text)): if text[t] in void: if not on[0]: on = (True, text[t]) elif text[t] == on[1]: on = (False, '') if text[t] == s and not on[0]: res += [text[lt+1:t]] lt = t res += [text[lt+1:]] return res It does what its name says, splits without (the instance of the items passed through void). e.g. (Toggle Plain Text) print splitwithout("""a b "c d" e f 'g h i' j k l""", " ", ('"',"'")) print splitwithout("""a b "c d" e f 'g h i' j k l""", " ", ('"',"'")) outputs: >>> ['a', 'b', '"c d"', 'e', 'f', "'g h i'", 'j', 'k', 'l'] Notice how the spaces in between "c d" and 'g h i' were not split as split() would normally do (since I passed '"', "'" through void)? The point of this function is so I can do split an html tag by its spacing, without taking a space that could be inside the paramaters as well. I wonder if this is actually faster then split() is. Edit: If you still don't see the point, here's what I was using it for. (Toggle Plain Text) def parse_tag(text, open = '<', close = '>'): first = splitwithout(text.split(open)[1].split(close)[0], ' ', ('"',"'")) tag = first[0] if tag == ' ': return False params = {} for p in first[1:]: pp = p.split('=') try: param = pp[0] val = pp[1].split(pp[1][0])[1] except IndexError: return False params[pp[0]] = val try: content = text.split(open+"/"+tag+close)[0].split(close)[1] except IndexError: return False return tag, params, content tag, params, content = parse_tag("""<a href='http://google.ca' title="Hyperlink to Google.ca">Hyperlink</a>""") print tag #'a' print params #{'href': 'http://google.ca', 'title': 'Hyperlink to Google.ca'} print content #Hyperlink def parse_tag(text, open = '<', close = '>'): first = splitwithout(text.split(open)[1].split(close)[0], ' ', ('"',"'")) tag = first[0] if tag == ' ': return False params = {} for p in first[1:]: pp = p.split('=') try: param = pp[0] val = pp[1].split(pp[1][0])[1] except IndexError: return False params[pp[0]] = val try: content = text.split(open+"/"+tag+close)[0].split(close)[1] except IndexError: return False return tag, params, content tag, params, content = parse_tag("""<a href='http://google.ca' title="Hyperlink to Google.ca">Hyperlink</a>""") print tag #'a' print params #{'href': 'http://google.ca', 'title': 'Hyperlink to Google.ca'} print content #Hyperlink Very helpful I think. Last edited by Sane; Mar 9th, 2006 at 5:17 PM.

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)

Thread Tools
Show Printable Version Email this Page
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

Programming Forums > Script Programming > Python

Useful code snippets