make-tree-html

#! /usr/bin/env python3

from __future__ import print_function
from os.path import join, basename
from glob import glob
import re


section_nr = [1]


def normalize_nodes(nodes, text_icon):
    for i in range(len(nodes)):
        node = nodes[i]
        if isinstance(node, str):
            node = {"text": node, "icon": text_icon}

        assign_section_number = False

        if "section_nr" in node and node["section_nr"] is not False:
            if node["section_nr"] is True:
                assign_section_number = True
            else:
                section_nr[0] = node["section_nr"]

        if "section" in node:
            if node.get("section_nr") is not False:
                assign_section_number = True

        if assign_section_number:
            node["section_nr"] = section_nr[0]
            section_nr[0] += 1

        nodes[i] = node

        if "nodes" in node:
            normalize_nodes(node["nodes"], text_icon)


def find_section_nodes(section_dict, node):
    if "section" in node:
        section_dict[node["section"]] = node

    for subnode in node.get("nodes", []):
        find_section_nodes(section_dict, subnode)


class RenderSettings:
    def __init__(self, default_icon, number_sections):
        self.default_icon = default_icon
        self.number_sections = number_sections


def render(settings, outf, node, indent=0, skip=1):
    attrs = {}

    icon = settings.default_icon
    if "icon" in node:
        icon = node["icon"]
    attrs["icon"] = icon

    if "opened" in node and int(bool(node["opened"])):
        attrs["opened"] = True

    text = node["text"]
    if (settings.number_sections
            and "section_nr" in node
            and node["section_nr"] is not False):
        text = "%d. %s" % (node["section_nr"], text)
    if "section" in node:
        text = "<b>%s</b>" % text

    if not skip:
        from json import dumps
        print(
            indent * " ",
            "<li data-jstree='%s'>" % dumps(attrs, sort_keys=True).replace("'", '"'),
            file=outf, sep="")
        indent += 2
        if "link" in node:
            print(
                indent * " ",
                "<a href=\"%s\">%s</a>" % (
                    node["link"],
                    text),
                file=outf, sep="")
        else:
            print(indent * " ", text, file=outf, sep="")

    subnodes = node.get("nodes", [])
    if subnodes:
        print(indent * " ", "<ul>", file=outf, sep="")
        indent += 2
        for subnode in subnodes:
            render(settings, outf, subnode, indent, skip=max(0, skip - 1))
        indent -= 2
        print(indent * " ", "</ul>", file=outf, sep="")

    if not skip:
        indent -= 2
        print(indent * " ", "</li>", file=outf, sep="")


FN_REGEX = re.compile(r"^([0-9]+)-(.*)(\.[a-z]+)$")


def get_section_id_and_display_name(trunk, include_extension):
    from os.path import splitext

    fn_match = FN_REGEX.match(trunk)
    if fn_match is not None:
        section_id = fn_match.group(1)
        if include_extension:
            fname = fn_match.group(2) + fn_match.group(3)
        else:
            fname = fn_match.group(2)
        display_name = basename(fname)
    else:
        section_id, display_name = trunk.split("/")

        if not include_extension:
            display_name, _ = splitext(display_name)

    try:
        section_id = int(section_id)
    except ValueError:
        pass

    return section_id, display_name


def blacklisted_glob(basedir, pattern, blacklist_regexps):
    return sorted(
            name
            for name in glob(join(basedir, pattern))
            if not any(bl_re.match(name[len(basedir)+1:])
                for bl_re in blacklist_regexps))


def main():
    import argparse

    parser = argparse.ArgumentParser(
            description='Turn a YAML file into a jsTree-compatible data file')

    parser.add_argument("-o", "--output-file", metavar="FILE", required=True)

    parser.add_argument("--ipynb-dir", metavar="DIRECTORY")
    parser.add_argument("--ipynb-urlroot", metavar="URL",
            help="(without the trailing slash)")
    parser.add_argument("--ipynb-as-py", action="store_true")
    parser.add_argument("--ipynb-as-ipynb", action="store_true")

    parser.add_argument("--binder-urlroot", metavar="URL",
            help="(without the trailing slash)")

    parser.add_argument("--py-dir", metavar="DIRECTORY")
    parser.add_argument("--py-urlroot", metavar="URL",
            help="(without the trailing slash)")

    parser.add_argument("--pdf-dir", metavar="DIRECTORY")
    parser.add_argument("--pdf-urlroot", metavar="URL",
            help="(without the trailing slash)")

    parser.add_argument("--default-icon", metavar="ICON_STR",
            default="fa fa-file-o")
    parser.add_argument("--text-icon", metavar="ICON_STR",
            default="fa fa-file-o")
    parser.add_argument("--number-sections", action="store_true")
    parser.add_argument("--blacklist-file", metavar="PATTERN_FILE")

    parser.add_argument("input_file", metavar="FILE")

    args = parser.parse_args()

    if args.blacklist_file is not None:
        blacklist_regexps = []
        with open(args.blacklist_file, "rt") as bl_file:
            import fnmatch
            for pattern in bl_file:
                blacklist_regexps.append(
                        re.compile(fnmatch.translate(pattern.strip())))

    from yaml import load
    with open(args.input_file, "rb") as inf:
        root_node = load(inf)

    normalize_nodes([root_node], args.text_icon)

    section_dict = {}
    find_section_nodes(section_dict, root_node)

    # {{{ demos

    if args.ipynb_dir is not None:
        for fn in blacklisted_glob(args.ipynb_dir, join("*", "*.ipynb"),
                blacklist_regexps):
            trunk = fn[len(args.ipynb_dir)+1:]

            section_id, display_name = get_section_id_and_display_name(
                    trunk, include_extension=False)

            link_ipynb = args.ipynb_urlroot + "/" + trunk
            link_html = link_ipynb.replace(".ipynb", ".html")
            main_link = link_html

            sub_nodes = [{
                        "text": "View on the web",
                        "link": link_html,
                        "icon": "fa fa-newspaper-o",
                        }]

            if args.binder_urlroot:
                binder_url = args.binder_urlroot + "/" + trunk

                sub_nodes.append({
                    "text": "Run interactively (on mybinder.org)",
                    "link": binder_url,
                    "icon": "fa fa-keyboard-o",
                    })
                main_link = binder_url

            if args.ipynb_as_py:
                link_py = link_ipynb.replace(".ipynb", ".py")
                sub_nodes.append({
                    "text": "Download Python script",
                    "link": link_py,
                    "icon": "fa fa-terminal",
                    })

            if args.ipynb_as_ipynb:
                sub_nodes.append({
                    "text": "Download Jupyter notebook",
                    "link": link_ipynb,
                    "icon": "fa fa-download",
                    })

            demo_node = {
                "text": "Demo: " + display_name,
                "link": main_link,
                "icon": "fa fa-keyboard-o",
                "nodes": sub_nodes,
                }

            if section_id in section_dict:
                section_dict[section_id]["nodes"].append(demo_node)

    # }}}

    # {{{ python source

    if args.py_dir is not None:
        for fn in blacklisted_glob(args.py_dir, join("*", "*.py"),
                blacklist_regexps):
            trunk = fn[len(args.py_dir)+1:]
            section_id, display_name = get_section_id_and_display_name(
                    trunk, include_extension=True)

            src_node = {
                "text": "Code: " + display_name,
                "link": args.py_urlroot + "/" + trunk,
                "icon": "fa fa-file-text-o",
            }
            if section_id in section_dict:
                section_dict[section_id]["nodes"].append(src_node)

    # }}}

    # {{{ notes

    if args.pdf_dir is not None:
        for fn in blacklisted_glob(args.pdf_dir, join("*.pdf"),
                blacklist_regexps):
            if "autosave" in fn:
                continue

            trunk = fn[len(args.pdf_dir)+1:]
            section_id, display_name = get_section_id_and_display_name(
                    trunk, include_extension=True)

            notes_node = {
                "text": "PDF: " + basename(display_name),
                "link": args.pdf_urlroot + "/" + trunk,
                "icon": "fa fa-book",
                }

            if section_id in section_dict:
                section_dict[section_id]["nodes"].insert(0, notes_node)

    # }}}

    with open(args.output_file, "wt", encoding="utf-8") as outf:
        render(
                RenderSettings(
                    default_icon=args.default_icon,
                    number_sections=args.number_sections),
                outf, root_node)


if __name__ == "__main__":
    main()

# vim: foldmethod=marker