Python Tips

File Access

• If you are building string records by s='' followed by s+='chunk' or the like many times, this is a typical newbie performance killer. Try accumulating your chunks in a list, like sl=[] followed by sl.append('chunk') instead, and then do f.write(''.join(sl)) instead of the f.write(s) you would have done. -- Advice from Bengt Ricter on comp.lang.python.
• A minor correction, if I may add: f.writelines(sl) might be quite more efficient than f.write(''.join(sl)) AFAIK the output is the same, as long as one remembers to insert the necessary '\n' for line breaks where needed :) Advice from Christos Georgiou on comp.lang.python.

Redefining special operators

• To redefine the multiplication(*) operator:
  

  class a(str):
       def __mul__(self, other):
       return self.startswith(other) or other.startswith(self)
  
  a1 = a('Hi')
  a2 = a('Hallo')
  a1*a2
  (returns 1)  -- Advice from Holger Krekel on comp.lang.python.
  

Name / Value Techniques

• Converting Name / Value pairs to a Dictionary of values
  

  >>> original = [('L0','S2'), ('S2','S0'), ('S2','S1'),  ('S1','Q2'), ('S0','Q1'), ('S2','S3')]
  >>> tree = {}
  >>> for key, value in original:
   tree.setdefault(key,[]).append(value)
  
  >>> tree
  {'S2': ['S0', 'S1', 'S3'], 'S1': ['Q2'], 'S0': ['Q1'], 'L0': ['S2']}
  
  -- Advice from Raymond Hettinger on comp.lang.python.
  
  Also, see Guido's discussion on Graphs: 
  http://www.python.org/doc/essays/graphs.html
  

Printing Techniques

• Using ReportLab's PDF tool:
  

  Perhaps you can draw a PDF file with Reportlab (http://www.reportlab.com/) 
  and printing it. Because you are using Outlook Express, I assume you need 
  it for Windows. An example:
  
  from reportlab.pdfgen import canvas
  from reportlab.lib.pagesizes import A4
  c = canvas.Canvas('c:\\tmp\\test.pdf', pagesize=A4, bottomup=0)
  c.setFont('Helvetica', 14)
  c.drawString(10, 20, 'Hello World!')
  c.save()
  
  If you want direct printing, without user interaction and you know the 
  path of the Acrobat Reader, you can use this line:
  
  system('"C:\\Programme\\Adobe\\Acrobat 5.0\\Reader\\AcroRd32.exe" /p /h c:
  \\tmp\\test.pdf')
  
  Perhaps it's easier with the ShellExecute command, because then you don't 
  need to know the path of AcroRd32.exe, but then you need this, if you 
  don't have ActiveState Python installed:
  http://starship.python.net/crew/mhammond/win32/Downloads.html
  
  If you just want to show the user the page and let the user decide to 
  print or close the document, this lines are sufficient:
  
  from os import startfile
  startfile('c:\\tmp\\test.pdf')
  
  Advice from Frank Buss on comp.lang.python
  

• Printing to the printer port
  

  >>> p=file('LPT1:','w')
  >>> p.write('Test Line\n\f')
  >>> p.flush() # p.close() also flushes
  
  Note: \f is linefeed and critical for this to work properly.
  
   Advice from Terry J. Reedy on comp.lang.python
  

• Printing on Unix
  

  Simple:
  import os
  p = os.popen('lpr', 'w')
  p.write("Hello world!\n")
  p.close()
  
  Complicated / using Postscript:
  import os
  p = os.popen('lpr', 'w')
  p.write('%!PS\n')
  p.write('/Helvetica findfont 12 scalefont setfont\n')
  p.write('/cm { 72 mul 2.54 div } def\n')
  p.write('1 cm 28 cm moveto\n')
  p.write('(Hello world!) show\n')
  p.write('showpage\n')
  p.close()
  
  Advice from Eric Brunel on comp.lang.python
  

• Printing on Windows
  

  import win32ui
  
  dc = win32ui.CreateDC()
  dc.CreatePrinterDC()
  dc.SetMapMode(4)        # This is UI_MM_LOENGLISH
  ## With this map mode, 12 points is 12*100/72 units = 16
  font = win32ui.CreateFont({'name' : 'Arial', 'height' : 16})
  dc.SelectObject(font)
  
  dc.StartDoc("HelloWorld")
  dc.StartPage()
  dc.TextOut(40, 40, "Hello world!")
  dc.EndPage()
  dc.EndDoc()
  
  For more information on the Windows version, see the documenattion for 
  win32ui @ 
  http://aspn.activestate.com//ASPN/Python/Reference/Products/ActivePython/PythonWin/modules.html 
  and Micro$oft MFC documenation @ 
  http://msdn.microsoft.com/library/default.asp
  
  Advice from Eric Brunel on comp.lang.python
  

Math Tricks

• Using Regular Expressions to find prime numbers:
  

  def isprime(num):
  	if num == 1 :return 1
  	prog=re.compile(r"^1?$|^(11+?)\1+$")
  	return prog.match('1'*num) is None
  

String Manipulation/Tricks

• Getting the longest string in a list:

  def lengthOfLongestStr(list):
    l=len(list[0])
    for count in range(1,len(list)):
      if len(list[count])>l:
        l=len(list[count])
    return l
  
  OR
  
  def lengthOfLongestStr(myList):
    return max([len(s) for s in myList])
  
   Advice by Matthew Cowles from comp.lang.python
  

  
  

  OR
      def lengthOfLongestStr(myList):
        return max (map (len, myList))
  
  Advice by Paul Rubin from comp.lang.python
  
  

Dynamically Evaluating Python Code (perhaps someone elses)

• * Avoid using exec, execfile, eval, and input unless you are deliberately asking your user to supply Python code, and take serious precautions if the program is running at a higher privledge level than the user. All other uses of them should be thoroughly thought out, and the implications understood.-- Advice on comp.lang.python.

Speed Optimizations

•  Instead of doing string concatentation like this:
  s = s1 + ' ' + s2 + ' ' + s3
  do:
  s = "%s %s %s" % (s1, s2, s3)
  

• If a function or variable from another module is being accessed a lot (e.g. in a loop, or multiple times in a flat function), then assigning it to a local variable will make the scope lookup cost (my term) occur only once instead of repeatedly.
• Instead of import library specify the functions you need, such as from library import fxn1,fxn2

• I recently wanted to match among a list of keywords, repeatedly, and want
  ed
  to help the Python `re' module a bit, speed-wise.  I wrote the following
  helper function (I stole the idea of this from Markku Rossi's `enscript')
  :
  
  
  def build regexp(tokens):
      # Build an optimised regular expression able to recognise all TOKENS.
      tokens by first = {}
      empty = False
      for token in tokens:
          if token == '':
              empty = True
          else:
              sequence = tokens by first.get(token[0])
              if sequence is None:
                  sequence = tokens by first[token[0]] = []
              sequence.append(token[1:])
      if not tokens by first:
          return ''
      fragments = [re.escape(letter) + build regexp(tokens)
                   for letter, tokens in tokens by first.iteritems()]
      if empty:
          fragments.append('')
      if len(fragments) == 1:
          return fragments[0]
      return '(?:%s)' % '|'.join(fragments)
  
  
  Given the above,
  
  
      build regexp(['this', 'that', 'the-other'])
  
  
  yields the string 'th(?:is|at|e\\-other)', which one may choose to
  `re.compile' before use.  By Francois Pinard

Dynamically loading modules in a directory

• import sys
  sys.path.append(additionaldirectory)
  

  Advice by Tim Delaney on comp.lang.python

Using Slots

• You can also make it even more general:
  
  class MyData:
  
       def __init__(self, **args):
           self.__dict__.update(args)
  
  d = MyData(firstName='Max M', lastName='Rasmussen')
  
  print d.firstName, d.lastName
  -- Advice on comp.lang.python.