husk/husk_helpers.py

import glob
import os
# from collections import deque
# import toml

# base_url = toml.load('settings.toml')["general"]["base_url"]

# def sort_branch( lst ):
#     '''
#     put directories in front of files and
#     sorts both alphabetically
#     '''
#     files = []
#     dirs = []
#     for path in lst:
#         if path.endswith(".md"):
#             files.append(path)
#         else:
#             dirs.append(path)
#     files.sort()
#     dirs.sort()
#     return [*dirs, *files]


def sort_branch(lst):
    """
    Puts directories in front of files and sorts both alphabetically.
    """
    files = [path for path in lst if path.endswith(".md")]
    dirs = [path for path in lst if not path.endswith(".md")]
    files.sort()
    dirs.sort()
    return dirs + files


# def make_tree(path="templates/content"):
#     '''
#     creates a dictionary of the directories and files of the doc repository
#     '''
#     if '/' in path:
#         tree = dict(name=path, children=[], type="directory")
#     else:
#         pass
#     try: lst = os.listdir(path)
#     except OSError:
#         pass #ignore errors
#     else:
#         lst = sort_branch(lst)
#         for name in lst:
#             fn = os.path.join(path, name)
#             if os.path.isdir(fn) and not fn.endswith(".git"):
#                 if os.listdir(fn): # this line is experimental and needs to be tested
#                     tree['children'].append(make_tree(fn))
#             else:
#                 if fn.endswith('.md'):
#                     tree['children'].append(dict(name=fn))
#     return tree

def make_tree(path="templates/content"):
    """
    Creates a dictionary of the directories and files of the doc repository.
    """
    tree = {'name': path, 'children': [], 'type': 'directory'}

    try:
        lst = os.listdir(path)
        lst = sort_branch(lst)
        for name in lst:
            fn = os.path.join(path, name)
            if os.path.isdir(fn) and not fn.endswith(".git") and os.listdir(fn):
                tree['children'].append(make_tree(fn))
            elif fn.endswith('.md'):
                tree['children'].append({'name': fn})
    except OSError:
        pass  # Ignore errors

    return tree


def rem_readme(path, tree):
    '''
    This functions puts out the path of the README.md and
    which will be index.html
    path can be dynamically configured and may have a / at the end nor not
    '''
    index = f"{path.rstrip('/')}/README.md"
    for item in tree['children']:
        if isinstance(item, dict):
            for k, v in item.items():
                if k == "name":
                    if v == index:
                        del tree['children'][-1][k]
    return tree


def cut_path_tree(tree, subdir, file_ending):
    '''
    pruning of the tree structure from make_tree()
    '''
    if subdir.endswith('/'):
        subdir = subdir[:-1]
    for key, val in tree.items():
        if isinstance(val, list):
            for item in val:
                item["name"] = item["name"][len(subdir):]
                if item["name"].endswith(file_ending):
                    item["name"] = item["name"][:-len(file_ending)]
                if isinstance(item, dict):
                    cut_path_tree(item, subdir, file_ending)
    return tree


def cut_filetype_tree(tree, filetype):
    '''
    removes file type of the links stored in make_tree()
    '''
    for key, val in tree.items():
        if isinstance(val, list):
            for item in val:
                if item["name"].endswith(filetype):
                    item["name"] = item["name"][:-len(filetype)]
                    if isinstance(item, dict):
                        cut_filetype_tree(item, filetype)
    return tree


# def list_files(path):
#     '''
#     creates a simple, one dimensional list of the doc repository
#     filters markdown files only
#     '''
#     doc_files = []
#     for root, dirs, files in os.walk(path):
#         for file in files:
#             if file.endswith(".md") and not ".git" in root:
#                 doc_files.append(os.path.join(root, file))
#     return doc_files


def list_files(path):
    """
    Creates a simple, one-dimensional list of the doc repository.
    Filters markdown files only.
    """
    return [
        file for file in glob.glob(
            os.path.join(path, '**/*.md'), recursive=True
            )
        if '.git' not in file
    ]


def build_index(path, file_ending):
    """
    Builds the searchable JSON object containing
    all markdown files with metadata.
    """
    searchable = {'index': []}
    file_list = list_files(path)
    for item in file_list:
        if item.endswith(file_ending):
            with open(item, 'r') as file:
                data = file.readlines()
                title = data[0].strip('# \n')
                uri = '/' + item[len(path):-len(file_ending)] + '.html'
                content = ''.join(data)
                searchable['index'].append({
                    'uri': uri,
                    'title': title,
                    'tags': [],
                    'content': content,
                    'description': ''
                })
        elif isinstance(item, dict):
            build_index(item, file_ending)
    return searchable