Merge branch 'develop' into refresh

plugins/help.py

@@ -46,3 +46,5 @@ def help(inp, notice=None, input=None, conn=None, bot=None):
     else:
         if inp in commands:
             notice(conn.conf["command_prefix"] + commands[inp].__doc__)
+        else:
+            notice("Command {}{} not found".format(conn.conf["command_prefix"], inp))

plugins/minecraft_wiki.py

@@ -1,8 +1,8 @@
 from util import hook, http, text
 import re
 
-api_url = "http://minecraftwiki.net/api.php?action=opensearch"
-mc_url = "http://minecraftwiki.net/wiki/"
+api_url = "http://minecraft.gamepedia.com/api.php?action=opensearch"
+mc_url = "http://minecraft.gamepedia.com/"
 
 
 @hook.command
@@ -10,20 +10,40 @@ def mcwiki(inp):
     """mcwiki <phrase> -- Gets the first paragraph of
     the Minecraft Wiki article on <phrase>."""
 
-    j = http.get_json(api_url, search=inp)
+    try:
+        j = http.get_json(api_url, search=inp)
+    except (http.HTTPError, http.URLError) as e:
+        return "Error fetching search results: {}".format(e)
+    except ValueError as e:
+        return "Error reading search results: {}".format(e)
 
     if not j[1]:
         return "No results found."
-    article_name = j[1][0].replace(' ', '_').encode('utf8')
+
+    # we remove items with a '/' in the name, because
+    # gamepedia uses subpages for different languages
+    # for some stupid reason
+    items = [item for item in j[1] if not "/" in item]
+
+    if items:
+        article_name = items[0].replace(' ', '_').encode('utf8')
+    else:
+        # there are no items without /, just return a / one
+        article_name = j[1][0].replace(' ', '_').encode('utf8')
 
     url = mc_url + http.quote(article_name, '')
-    page = http.get_html(url)
+
+    try:
+        page = http.get_html(url)
+    except (http.HTTPError, http.URLError) as e:
+        return "Error fetching wiki page: {}".format(e)
 
     for p in page.xpath('//div[@class="mw-content-ltr"]/p'):
         if p.text_content():
             summary = " ".join(p.text_content().splitlines())
             summary = re.sub("\[\d+\]", "", summary)
             summary = text.truncate_str(summary, 200)
-            return "{} :: {}".format(summary, url)
+            return u"{} :: {}".format(summary, url)
 
+    # this shouldn't happen
     return "Unknown Error."

plugins/steam.py

@@ -1,28 +1,62 @@
 import re
-from util import hook, http, web, text
-from bs4 import BeautifulSoup
+from util import hook, http, web
+from util.text import truncate_str
+from bs4 import BeautifulSoup, NavigableString, Tag
 
 
 steam_re = (r'(.*:)//(store.steampowered.com)(:[0-9]+)?(.*)', re.I)
 
 
 def get_steam_info(url):
-    # we get the soup manually because the steam pages have some odd encoding troubles
     page = http.get(url)
     soup = BeautifulSoup(page, 'lxml', from_encoding="utf-8")
 
-    name = soup.find('div', {'class': 'apphub_AppName'}).text
-    desc = ": " + text.truncate_str(soup.find('div', {'class': 'game_description_snippet'}).text.strip())
+    data = {}
 
-    # the page has a ton of returns and tabs
-    details = soup.find('div', {'class': 'glance_details'}).text.strip().split(u"\n\n\r\n\t\t\t\t\t\t\t\t\t")
-    genre = " - Genre: " + details[0].replace(u"Genre: ", u"")
-    date = " - Release date: " + details[1].replace(u"Release Date: ", u"")
-    price = ""
-    if not "Free to Play" in genre:
-        price = " - Price: " + soup.find('div', {'class': 'game_purchase_price price'}).text.strip()
+    data["name"] = soup.find('div', {'class': 'apphub_AppName'}).text
+    data["desc"] = truncate_str(soup.find('meta', {'name': 'description'})['content'].strip(), 80)
 
-    return name + desc + genre + date + price
+    # get the element details_block
+    details = soup.find('div', {'class': 'details_block'})
+
+    # loop over every <b></b> tag in details_block
+    for b in details.findAll('b'):
+        # get the contents of the <b></b> tag, which is our title
+        title = b.text.lower().replace(":", "")
+        if title == "languages":
+            # we have all we need!
+            break
+
+        # find the next element directly after the <b></b> tag
+        next = b.nextSibling
+        if next:
+            # if the element is some text
+            if isinstance(next, NavigableString):
+                text = next.string.strip()
+                if text:
+                    # we found valid text, save it and continue the loop
+                    data[title] = text
+                    continue
+                else:
+                    # the text is blank - sometimes this means there are
+                    # useless spaces or tabs between the <b> and <a> tags.
+                    # so we find the next <a> tag and carry on to the next
+                    # bit of code below
+                    next = next.find_next('a', href=True)
+
+            # if the element is an <a></a> tag
+            if isinstance(next, Tag) and next.name == 'a':
+                text = next.string.strip()
+                if text:
+                    # we found valid text (in the <a></a> tag),
+                    # save it and continue the loop
+                    data[title] = text
+                    continue
+
+
+    data["price"] = soup.find('div', {'class': 'game_purchase_price price'}).text.strip()
+
+    return u"\x02{name}\x02: {desc}, \x02Genre\x02: {genre}, \x02Release Date\x02: {release date}, \x02Price\x02: {price}".format(**data)
 
 
 @hook.regex(*steam_re)