diff --git a/lib/wttrin_png.py b/lib/wttrin_png.py
new file mode 100644
index 0000000..932c2f9
--- /dev/null
+++ b/lib/wttrin_png.py
@@ -0,0 +1,278 @@
+#!/usr/bin/python
+#vim: encoding=utf-8
+
+import sys
+import os
+import re
+import time
+
+"""
+This module is used to generate png-files for wttr.in queries.
+The only exported function is:
+
+        make_wttr_in_png(filename)
+
+in filename (in the shortname) is coded the weather query.
+The function saves the weather report in the file and returns None.
+"""
+
+import requests
+
+from PIL import Image, ImageFont, ImageDraw
+import pyte.screens
+
+# downloaded from https://gist.github.com/2204527
+# described/recommended here:
+#
+#   http://stackoverflow.com/questions/9868792/find-out-the-unicode-script-of-a-character
+#
+import unicodedata2
+
+MYDIR = os.path.abspath(os.path.dirname( os.path.dirname('__file__')))
+PNG_CACHE = os.path.join(MYDIR, "cache/png")
+
+COLS = 180
+ROWS = 100
+
+CHAR_WIDTH = 7
+CHAR_HEIGHT = 14
+
+# How to find font for non-standard scripts:
+#
+# $ fc-list :lang=ja
+
+FONT_SIZE = 12
+FONT_CAT = {
+    'default':      "/usr/share/fonts/truetype/dejavu/DejaVuSansMono.ttf",
+    'Cyrillic':     "/usr/share/fonts/truetype/dejavu/DejaVuSansMono.ttf",
+    'Greek':        "/usr/share/fonts/truetype/dejavu/DejaVuSansMono.ttf",
+    'Han':          "/usr/share/fonts/truetype/motoya-l-cedar/MTLc3m.ttf",
+    'Hiragana':     "/usr/share/fonts/truetype/motoya-l-cedar/MTLc3m.ttf",
+}
+
+def color_mapping(color):
+    """
+    Convert pyte color to PIL color
+    """
+    if color == 'default':
+        return 'lightgray'
+    if color in ['green', 'black', 'cyan', 'blue', 'brown']:
+        return color
+    try:
+        return (int(color[0:2], 16), int(color[2:4], 16), int(color[4:6], 16))
+    except:
+        # if we do not know this color and it can not be decoded as RGB,
+        # print it and return it as it is (will be displayed as black)
+        # print color
+        return color
+    return color
+
+def strip_buf(buf):
+    """
+    Strips empty spaces from behind and from the right side.
+    (from the right side is not yet implemented)
+    """
+    def empty_line(line):
+        "Returns True if the line consists from spaces"
+        return all(x.data == ' ' for x in line)
+
+    def line_len(line):
+        "Returns len of the line excluding spaces from the right"
+
+        last_pos = len(line)
+        while last_pos > 0 and line[last_pos-1].data == ' ':
+            last_pos -= 1
+        return last_pos
+
+    number_of_lines = 0
+    for line in buf[::-1]:
+        if not empty_line(line):
+            break
+        number_of_lines += 1
+
+    buf = buf[:-number_of_lines]
+
+    max_len = max(line_len(x) for x in buf)
+    buf = [line[:max_len] for line in buf]
+
+    return buf
+
+def script_category(char):
+    """
+    Returns category of a Unicode character
+    Possible values:
+        default, Cyrillic, Greek, Han, Hiragana
+    """
+    cat = unicodedata2.script_cat(char)[0]
+    if char == u'：':
+        return 'Han'
+    if cat in ['Latin', 'Common']:
+        return 'default'
+    else:
+        return cat
+
+def gen_term(filename, buf):
+    buf = strip_buf(buf)
+    cols = len(buf[0])
+    rows = len(buf)
+
+    image = Image.new('RGB', (cols * CHAR_WIDTH, rows * CHAR_HEIGHT))
+
+    buf = buf[-ROWS:]
+
+    draw = ImageDraw.Draw(image)
+    font = {}
+    for cat in FONT_CAT:
+        font[cat] = ImageFont.truetype(FONT_CAT[cat], FONT_SIZE)
+
+    x_pos = 0
+    y_pos = 0
+    for line in buf:
+        x_pos = 0
+        for char in line:
+            current_color = color_mapping(char.fg)
+            if char.bg != 'default':
+                draw.rectangle(
+                    ((x_pos, y_pos),
+                     (x_pos+CHAR_WIDTH, y_pos+CHAR_HEIGHT)),
+                    fill=color_mapping(char.bg))
+
+            cat = script_category(char.data)
+            if cat not in font:
+                print "Unknown font category: %s" % cat
+            draw.text(
+                (x_pos, y_pos),
+                char.data,
+                font=font.get(cat, font.get('default')),
+                fill=current_color)
+            #sys.stdout.write(c.data)
+
+            x_pos += CHAR_WIDTH
+        y_pos += CHAR_HEIGHT
+        #sys.stdout.write('\n')
+
+    image.save(filename)
+
+def typescript_to_one_frame(png_file, text):
+    """
+    Render text (terminal sequence) in png_file
+    """
+
+    # fixing some broken characters because of bug #... in pyte 6.0
+    text = text.replace('Н', 'H').replace('Ν', 'N')
+
+    screen = pyte.screens.Screen(COLS, ROWS)
+    #screen.define_charset("B", "(")
+
+    stream = pyte.streams.ByteStream()
+    stream.attach(screen)
+
+    stream.feed(text)
+
+    gen_term(png_file, screen.buffer)
+
+#
+# wttr.in related functions
+#
+
+def parse_wttrin_png_name(name):
+    """
+    Parse the PNG filename and return the result as a dictionary.
+    For example:
+        input = City_200x_lang=ru.png
+        output = {
+            "lang": "ru",
+            "width": "200",
+            "filetype": "png",
+            "location": "City"
+        }
+    """
+
+    parsed = {}
+
+    if name.lower()[-4:] == '.png':
+        parsed['filetype'] = 'png'
+        name = name[:-4]
+
+    parts = name.split('_')
+    parsed['location'] = parts[0]
+
+    for part in parts:
+        if re.match('(?:[0-9]+)x', part):
+            parsed['width'] = part[:-1]
+        elif re.match('x(?:[0-9]+)', part):
+            parsed['height'] = part[1:]
+        elif re.match(part, '(?:[0-9]+)x(?:[0-9]+)'):
+            parsed['width'], parsed['height'] = part.split('x', 1)
+        elif '=' in part:
+            arg, val = part.split('=', 1)
+            parsed[arg] = val
+
+    return parsed
+
+def make_wttrin_query(parsed):
+    """
+    Convert parsed data into query name
+    """
+
+    for key in ['width', 'height', 'filetype']:
+        if key in parsed:
+            del parsed[key]
+
+    location = parsed['location']
+    del parsed['location']
+
+    args = []
+    if 'options' in parsed:
+        args = [parsed['options']]
+        del parsed['options']
+    else:
+        args = []
+
+    for key, val in parsed.items():
+        args.append('%s=%s' % (key, val))
+
+    url = "http://wttr.in/%s" % location
+    if args != []:
+        url += "?%s" % ("&".join(args))
+
+    return url
+
+
+def make_wttr_in_png(png_name, options=None):
+    """
+    The function saves the weather report in the file and returns None.
+    The weather query is coded in filename (in the shortname).
+    """
+
+    parsed = parse_wttrin_png_name(png_name)
+
+    # if location is MyLocation it should be overriden 
+    # with autodetected location (from options)
+    if parsed.get('location', 'MyLocation') == 'MyLocation':
+        del parsed['location']
+
+    if options is not None:
+        for key, val in options.items():
+            if key not in parsed:
+                parsed[key] = val
+    url = make_wttrin_query(parsed)
+
+    timestamp = time.strftime("%Y%m%d%H", time.localtime())
+    cached_basename = url[14:].replace('/','_')
+
+    cached_png_file = "%s/%s/%s.png" % (PNG_CACHE, timestamp, cached_basename)
+
+    dirname = os.path.dirname(cached_png_file)
+    if not os.path.exists(dirname):
+        os.makedirs(dirname)
+
+    if os.path.exists(cached_png_file):
+        return cached_png_file
+
+    text = requests.get(url).text.replace('\n', '\r\n')
+    curl_output = text.encode('utf-8')
+
+    typescript_to_one_frame(cached_png_file, curl_output)
+
+    return cached_png_file