#!/usr/bin/env python3

# http://inamidst.com/saxo/
# Created by Sean B. Palmer

import html.entities
import json
import re
import subprocess
import urllib.parse

import saxo

_regex_entity = re.compile(r"&([^;\s]+);")

def decode_entities(hypertext):
    def entity(match):
        name = match.group(1).lower()

        if name.startswith("#x"):
            return chr(int(name[2:], 16))
        elif name.startswith("#"):
            return chr(int(name[1:]))
        elif name in html.entities.name2codepoint:
            return chr(html.entities.name2codepoint[name])
        return "[" + name + "]"

    def default(match):
        try: return entity(match)
        except: return match.group(1)
    return _regex_entity.sub(default, hypertext)

@saxo.command()
def snippets(arg):
    if not arg:
        return "Retrieve snippets from Google (Ajax via Tor)"
    arg = urllib.parse.quote(arg)
    url = "https://ajax.googleapis.com/ajax/services/search/web?v=1.0&q=%s"
    try: s = subprocess.check_output(["torsocks", "curl", "-s", url % arg])
    except Exception as err:
        return "Error: %s:%s" % (type(err).__name__, err)
    data = json.loads(s.decode("utf-8", "replace"))
    snippets = [res["content"] for res in data["responseData"]["results"]]
    snippets = [decode_entities(res) for res in snippets]
    snippets = [res.replace("<b>", "").replace("</b>", "") for res in snippets]
    snippets = [re.sub(r"\s+", " ", res) for res in snippets]
    snippets = " / ".join(snippets).replace("\n", " ")[:360]
    return snippets