The Protocol Informatics Project

FreeMind

Yesterday I got distracted again for a bit (Oct 14). I was playing with a thing called freemind (partially as my wife wants a package that can do work-breakdown-structures). FreeMind It occured to me that the FreeMind display lookes very like a phylogenetic tree. And the freemind file format is xml - and very simple xml too - simple enough that I can understand & generate it easily.

I hacked a dot file I had generated into freemind's xml form and it loaded nicely. It could be quite useful I think. You can browse the tree quickly, ... also hide or open branches, add notes/colours/whatever to nodes, export the tree to html.

Hacking dot files is hard work if you have to do it every time so I took a look into PI's code tree.py and I got sucked in and suddently I was attempting to program in python! eek. I succeded in generating freemind xml.

What's this all about huh? Here. A screenshot: (thanks to ImageMagick)

Here are generated files:

• icmp.mm generated file (you would need FreeMind to look at this)
• icmp-withnotes.mm can then view and edit file, add notes, hide branches
• icmp-withnotes.mm.html and can export to html, cute. not quite as pretty as freemind

Here be diff and source: (this main.py and The Protocol Informatics Project are Copyright © 2004 Marshall Beddoe)

• tree.py - edited by me
• diff -u /usr/local/lib/python2.3/site-packages/PI/tree.py /home/me/src/PI/tree.py > tree.patch

FreeMind comes precompiled for Windows/MacOSX/linux though I used the source package to install. It had no other dependancies. My machine here is getting a bit old and slow but freemind runs nice & zippily.

See here my train of thought starting to hack python

# warning: I don't know what I'm doing when it comes to python
#
   www.python.org/moin/BeginnersGuide_2fProgrammers
#
   diveintopython.org/toc/index.html

#
   www.zvon.org/other/python/doc21/lib/python.html
#
   docs.python.org/lib/lib.html
#
   rgruet.free.fr/PQR2.3.html


### I needed to look at data structures I didn't understand

#
   www.informatik.hu-berlin.de/Themen/manuals/python/python-texinfo/marshal.html
#
   www.sourcekeg.co.uk/www.python.org/doc/2.2.3/lib/module-marshal.html
import marshal,sys
#print marshal.dump( (1, 2.0, 'name', [2,3,5,7]), sys.stdout )
print "\ndumps", marshal.dumps( (1, 2.0, 'name', [2,3,5,7]) )

#[jamesc@betty] ~/src/PI/data/$ python pyxml.py >pyxml.out
#[jamesc@betty] ~/src/PI/data/$ xxd pyxml.
#xxd: pyxml.: No such file or directory
#[jamesc@betty] ~/src/PI/data/$ xxd pyxml.out 
#0000000: 0a64 756d 7073 2028 0400 0000 6901 0000  .dumps (....i...
#0000010: 0066 0332 2e30 7304 0000 006e 616d 655b  .f.2.0s....name[
#0000020: 0400 0000 6902 0000 0069 0300 0000 6905  ....i....i....i.
#0000030: 0000 0069 0700 0000 0a0a 786d 6c20 6475  ...i......xml du
#0000040: 6d70 203c 3f78 6d6c 2076 6572 7369 6f6e  mp <?xml version
#0000050: 3d22 312e 3022 3f3e 3c6d 6172 7368 616c  ="1.0"?><marshal
#0000060: 3e3c 7475 706c 653e 3c69 6e74 3e31 3c2f  ><tuple><int>1</
#0000070: 696e 743e 3c66 6c6f 6174 3e32 2e30 3c2f  int><float>2.0</
#0000080: 666c 6f61 743e 3c73 7472 696e 673e 6e61  float><string>na
#0000090: 6d65 3c2f 7374 7269 6e67 3e3c 6c69 7374  me</string><list
#00000a0: 2069 643d 2269 3222 3e3c 696e 743e 323c   id="i2"><int>2<
#00000b0: 2f69 6e74 3e3c 696e 743e 333c 2f69 6e74  /int><int>3</int
#00000c0: 3e3c 696e 743e 353c 2f69 6e74 3e3c 696e  ><int>5</int><in
#00000d0: 743e 373c 2f69 6e74 3e3c 2f6c 6973 743e  t>7</int></list>
#00000e0: 3c2f 7475 706c 653e 3c2f 6d61 7273 6861  </tuple></marsha
#00000f0: 6c3e 0a                                  l>.

#
   www.amk.ca/python/simple/ns-filter.html


### using xml.marshal I could just print the dumped xml to screen and look at it

#
   pyxml.sourceforge.net/
#
   prdownloads.sourceforge.net/pyxml/PyXML-0.8.3.tar.gz?download
#
   pyxml.sourceforge.net/topics/howto/node26.html
import xml.marshal.generic;

# The interface is the same as the standard Python marshal module:
# dump(value, file) and dumps(value)
# convert value into XML and either write it to the given file or return
# it as a string, while load(file) and loads(string) perform the reverse conversion. For example:

print "\nxml dump", xml.marshal.generic.dumps( (1, 2.0, 'name', [2,3,5,7]) )

# okay. xml is lovely but we're better off writing data <map version .... <node .... by hand (print)
# but the dumper is handy

#import xml.marshal.generic;
#print "\nxml dump", xml.marshal.generic.dumps( (1, 2.0, 'name', [2,3,5,7]) )


### how to open and write files in python
### also formatted output

logfile = open('test.log', 'w') 
logfile.write('test succeeded') 
indent = "iiiiiiiiiiiiii"
v1 = 1234
l = "56.04%"
#logfile.write( indent + '<node TEXT="', v1, '-' + l + '">' ) 
#logfile.write( indent + '<node TEXT="', v1, '-' + l + '">' ) 
s = "<node TEXT=\"%(node)d-%(percent)s\">\n" %  { 'node':v1, 'percent':l}
logfile.write( indent + s )
logfile.close()
print file('test.log').read()  

level = 0
#level++ doesn't work
#++level doesn't work
level=level+1
level+=1

# don't know how to work the formatted print directive * in python
print "\n"+'%*s', level*3, " "; # indent nodes by level in tree                  
print "</node>\n";

print '%s has %03d quote types.' % ('Python', 2)  == 'Python has 002 quote types.'
a = '%(lang)s has %(c)03d quote types.' % {'c':2, 'lang':'Python'}
print a

level = 0
print "\n"
print "mmm" * level
print "</node>\n";
level += 1
print "\n"
print "mmm" * level
print "</node>\n";
level += 1
print "\n"
print "mmm" * level
print "</node>\n";

#
   www.awprofessional.com/articles/article.asp?p=28790&seqNum=2


### for loop

print '\nlevel is', level
for n in range(2, level):
    print ' n is', n
print "\n"

### string repeat using *
indent = "\n" + "   " * level # indent nodes by level in tree                  
print indent, level, "\n"

### if not

if not 0:
    print "! 0"

if not 1:
    print "! 1"

home