Question

Here’s what I’ve got so far:

project_dir = '/my/project/dir'
project_depth = len(project_dir.split(os.path.sep))

xml_files = []
for dirpath, dirnames, filenames in os.walk(project_dir):
    for filename in fnmatch.filter(filenames, '*.xml'):
        dirs = dirpath.split(os.path.sep)[project_depth:]
        print(dirs)
        xml_files.append(os.path.join(dirpath,filename))

Essentially what I want to do is spit out my project directory structure with all the XML files as an HTML tree (using <ul>). I can get all the files this way, but I can’t seem to figure out how to organize them into a tree.

With the way this os.walk works, I don’t know when I’ve gone in a level deeper, or if I’m still traversing the same directory.

---

for dirpath, dirnames, filenames in os.walk(project_dir):
    xml_files = fnmatch.filter(filenames, '*.xml')
    if len(xml_files) > 0:
        out.write('<li>{0}<ul>'.format(dirpath))
        for f in xml_files:
            out.write('<li>{0}</li>'.format(f))
        out.write('</ul></li>')
out.write('</ul>')  

This gives me a list of directories and all the files underneath them, but I still can’t figure out how to split the directory path so that it’s nested too.

Answer 1

os.walk might not be the best solution if you care about the hierarchy. A simpler solution would probably be just do use os.listdir with os.path.isdir to traverse your tree recursively.

import os

def traverse(dir):
    print '<ul>'
    for item in os.listdir(dir):
        print '<li>%s</li>' % item
        fullpath = os.path.join(dir, item)
        if os.path.isdir(fullpath):
            traverse(fullpath)
    print '</ul>'

projectdir = '.'
traverse(projectdir)