question about XML parsing

[ateale]

Hey Guys,

I am discovering the awesomeness that is XML.

I use an application called Final Cut Pro for editing video. The app is able to export is projects as XML.

I am trying to develop a script to read that script and build a list of the files that are listed in the XML. The projects' imported files are enclosed in the 'pathurl' tag in the XML file.

It is nearly working but I just wanted to see what you guys think of the manner I have approached it.

Currently it spits out a list, but the information is enclosed in the xml element tag - it would be great to get a list without the tags.

example (current output):
<pathurl>file://localhost/Volumes/HD1/FCP_Documents/Projects/Media/Petronas_mediamanaged_07AUG07/research.tif</pathurl>

example of what I'd like:
/Volumes/HD1/FCP_Documents/Projects/Media/Petronas_mediamanaged_07AUG07/research.tif

Is there an 'XML parsing call' (sorry i am making up programming lingo as I go) to do this, or is it a matter of using a python tool like strip/split ?

Thanks for any advice!

Adam

import sys
import os
from xml.dom import minidom  
 
xmldocumentpath = str(sys.argv[1])
 
elementtofind = 'pathurl'
 
 
xmldoc = minidom.parse(xmldocumentpath)
pathlist = xmldoc.getElementsByTagName(elementtofind)
pathlist
 
# All Nodes listed
# AllNode = xmldoc.firstChild
 
itemamount = len (pathlist)
 
print itemamount
 
loop = 0
while loop < itemamount:
    print pathlist[loop].toxml()
    loop = loop + 1
 

[bartonc]

Hi Adam.

I'm currently studying Regular Expressions in my spare (LOL) time. I believe that regex is such an important tool that is perfectly suited to this kind of task that it kills me not to be able to just crank one out for you.

In pure Python, you could use something like this:

>>> s = "<pathurl>file://localhost/Volumes/HD1/FCP_Documents/Projects/Media/Petronas_mediamanaged_07AUG07/research.tif</pathurl>"
>>> token = "pathurl>"
>>> size = len(token)
>>> start = s.find(token)
>>> end = s.find(token, start + size)
>>> s[start + size:end - 2]
'file://localhost/Volumes/HD1/FCP_Documents/Projects/Media/Petronas_mediamanaged_07AUG07/research.tif'
>>> 

[ateale]

Hey BartonC thanks a lot!
That is really really cool!

I'll play around with that for a while!

Cheers mate!

Hi Adam.

I'm currently studying Regular Expressions in my spare (LOL) time. I believe that regex is such an important tool that is perfectly suited to this kind of task that it kills me not to be able to just crank one out for you.

In pure Python, you could use something like this:

>>> s = "<pathurl>file://localhost/Volumes/HD1/FCP_Documents/Projects/Media/Petronas_mediamanaged_07AUG07/research.tif</pathurl>"
>>> token = "pathurl>"
>>> size = len(token)
>>> start = s.find(token)
>>> end = s.find(token, start + size)
>>> s[start + size:end - 2]
'file://localhost/Volumes/HD1/FCP_Documents/Projects/Media/Petronas_mediamanaged_07AUG07/research.tif'
>>> 

[bartonc]

Hey BartonC thanks a lot!
That is really really cool!

I'll play around with that for a while!

Cheers mate!

Actually, that was kind of dumb... If you know the size of the token AND that it exists, simply:

>>> token = "<pathurl>"
>>> size = len(token)
>>> s[size:-size - 1]
'file://localhost/Volumes/HD1/FCP_Documents/Projects/Media/Petronas_mediamanaged_07AUG07/research.tif'
>>> 

[bvdet]

Hi Adam.

I'm currently studying Regular Expressions in my spare (LOL) time. I believe that regex is such an important tool that is perfectly suited to this kind of task that it kills me not to be able to just crank one out for you.

In pure Python, you could use something like this:

>>> s = "<pathurl>file://localhost/Volumes/HD1/FCP_Documents/Projects/Media/Petronas_mediamanaged_07AUG07/research.tif</pathurl>"
>>> token = "pathurl>"
>>> size = len(token)
>>> start = s.find(token)
>>> end = s.find(token, start + size)
>>> s[start + size:end - 2]
'file://localhost/Volumes/HD1/FCP_Documents/Projects/Media/Petronas_mediamanaged_07AUG07/research.tif'
>>> 

Barton, Adam - I am also trying to learn RE. What do you think of this?

import re
 
fn = r'H:\TEMP\temsys\re_parse_string.txt'
 
patt = re.compile(r'<pathurl>file://localhost(.+)</pathurl>')
 
f = open(fn)
data = []
for line in f:
    m = patt.search(line)
    if m:
        data.append(m.group(1))
 
print data

Output:

>>> ['/Volumes/HD1/FCP_Documents/Projects/Media/Petronas_mediamanaged_07AUG07/research.tif']

Interaction:

>>> m.group(0)
'<pathurl>file://localhost/Volumes/HD1/FCP_Documents/Projects/Media/Petronas_mediamanaged_08AUG07/research1.tif</pathurl>'
>>> m.group(1)
'/Volumes/HD1/FCP_Documents/Projects/Media/Petronas_mediamanaged_08AUG07/research1.tif'
>>>

[bartonc]

Barton, Adam - I am also trying to learn RE. What do you think of this?

import re
 
fn = r'H:\TEMP\temsys\re_parse_string.txt'
 
patt = re.compile(r'<pathurl>file://localhost(.+)</pathurl>')
 
f = open(fn)
data = []
for line in f:
    m = patt.search(line)
    if m:
        data.append(m.group(1))
 
print data

Output:

>>> ['/Volumes/HD1/FCP_Documents/Projects/Media/Petronas_mediamanaged_07AUG07/research.tif']

Interaction:

>>> m.group(0)
'<pathurl>file://localhost/Volumes/HD1/FCP_Documents/Projects/Media/Petronas_mediamanaged_08AUG07/research1.tif</pathurl>'
>>> m.group(1)
'/Volumes/HD1/FCP_Documents/Projects/Media/Petronas_mediamanaged_08AUG07/research1.tif'
>>>

That's the one that I was imagining! Thank you, very much.

It's interesting to note that (say) perl regex would not have created group zero, as there are no parentheses (which are the official "group" operators).

[bvdet]

That's the one that I was imagining! Thank you, very much.

It's interesting to note that (say) perl regex would not have created group zero, as there are no parentheses (which are the official "group" operators).

I have been confused about the group() method all along. I learned this recently by experimenting (trial and error, a lot of error!).

[bvdet]

Barton, Adam - I am also trying to learn RE. What do you think of this?

import re
 
fn = r'H:\TEMP\temsys\re_parse_string.txt'
 
patt = re.compile(r'<pathurl>file://localhost(.+)</pathurl>')
 
f = open(fn)
data = []
for line in f:
    m = patt.search(line)
    if m:
        data.append(m.group(1))
 
print data

Output:

>>> ['/Volumes/HD1/FCP_Documents/Projects/Media/Petronas_mediamanaged_07AUG07/research.tif']

Interaction:

>>> m.group(0)
'<pathurl>file://localhost/Volumes/HD1/FCP_Documents/Projects/Media/Petronas_mediamanaged_08AUG07/research1.tif</pathurl>'
>>> m.group(1)
'/Volumes/HD1/FCP_Documents/Projects/Media/Petronas_mediamanaged_08AUG07/research1.tif'
>>>

Just to show good practice, the open file object 'f' should be closed:

f.close()