From 49f52e29c78b93a8ef6fc7a7f9093a524fe43886 Mon Sep 17 00:00:00 2001 From: Luciano Ramalho Date: Wed, 22 Jan 2020 22:52:23 -0300 Subject: [PATCH] updated chapter 4 and appendix-A files --- 04-text-byte/charfinder/cf.py | 28 + 04-text-byte/charfinder/cf_tests.rst | 36 + 04-text-byte/charfinder/test.sh | 2 + 04-text-byte/numerics_demo.py | 4 +- 04-text-byte/ramanujan.py | 4 +- 04-text-byte/sanitize.py | 12 +- 04-text-byte/skin.py | 12 + 04-text-byte/two_flags.py | 6 + 04-text-byte/zwj_sample.py | 28 + appendix-A/arcfour.py | 55 + appendix-A/arcfour_futures.py | 46 + appendix-A/bottle.py | 3771 ++++++++++++++++++++++ appendix-A/container_perftest.py | 76 + appendix-A/container_perftest_datagen.py | 37 + appendix-A/emoji_families.py | 87 + appendix-A/flags2_common.py | 149 + appendix-A/flags2_sequential.py | 87 + appendix-A/hashdiff.py | 20 + appendix-A/isis2json.py | 263 ++ appendix-A/mem_test.py | 24 + appendix-A/schedule2.py | 132 + appendix-A/sha_futures.py | 38 + appendix-A/taxi_sim.py | 203 ++ appendix-A/test_schedule1.py | 37 + appendix-A/test_schedule2.py | 72 + 25 files changed, 5219 insertions(+), 10 deletions(-) create mode 100755 04-text-byte/charfinder/cf.py create mode 100644 04-text-byte/charfinder/cf_tests.rst create mode 100755 04-text-byte/charfinder/test.sh create mode 100755 04-text-byte/skin.py create mode 100644 04-text-byte/two_flags.py create mode 100644 04-text-byte/zwj_sample.py create mode 100644 appendix-A/arcfour.py create mode 100644 appendix-A/arcfour_futures.py create mode 100644 appendix-A/bottle.py create mode 100644 appendix-A/container_perftest.py create mode 100644 appendix-A/container_perftest_datagen.py create mode 100755 appendix-A/emoji_families.py create mode 100644 appendix-A/flags2_common.py create mode 100644 appendix-A/flags2_sequential.py create mode 100644 appendix-A/hashdiff.py create mode 100644 appendix-A/isis2json.py create mode 100644 appendix-A/mem_test.py create mode 100644 appendix-A/schedule2.py create mode 100644 appendix-A/sha_futures.py create mode 100644 appendix-A/taxi_sim.py create mode 100644 appendix-A/test_schedule1.py create mode 100644 appendix-A/test_schedule2.py diff --git a/04-text-byte/charfinder/cf.py b/04-text-byte/charfinder/cf.py new file mode 100755 index 0000000..5717521 --- /dev/null +++ b/04-text-byte/charfinder/cf.py @@ -0,0 +1,28 @@ +#!/usr/bin/env python3 +import sys +import unicodedata + +FIRST, LAST = ord(' '), sys.maxunicode # <1> + + +def find(*query_words, first=FIRST, last=LAST): # <2> + query = {w.upper() for w in query_words} # <3> + count = 0 + for code in range(first, last + 1): + char = chr(code) # <4> + name = unicodedata.name(char, None) # <5> + if name and query.issubset(name.split()): # <6> + print(f'U+{code:04X}\t{char}\t{name}') # <7> + count += 1 + print(f'({count} found)') + + +def main(words): + if words: + find(*words) + else: + print('Please provide words to find.') + + +if __name__ == '__main__': + main(sys.argv[1:]) diff --git a/04-text-byte/charfinder/cf_tests.rst b/04-text-byte/charfinder/cf_tests.rst new file mode 100644 index 0000000..57028e0 --- /dev/null +++ b/04-text-byte/charfinder/cf_tests.rst @@ -0,0 +1,36 @@ +Doctests for ``cf.py`` +====================== + +How to run the tests +---------------------- + +Run the ``doctest`` module from the command line:: + + $ python3 -m doctest cf_tests.rst + + +Tests +----- + +Import functions for testing:: + + >>> from cf import find, main + +Test ``find`` with single result:: + + >>> find("sign", "registered") # doctest:+NORMALIZE_WHITESPACE + U+00AE ® REGISTERED SIGN + (1 found) + + +Test ``find`` with two results:: + + >>> find("chess", "queen", last=0xFFFF) # doctest:+NORMALIZE_WHITESPACE + U+2655 ♕ WHITE CHESS QUEEN + U+265B ♛ BLACK CHESS QUEEN + (2 found) + +Test ``main`` with no words:: + + >>> main([]) + Please provide words to find. diff --git a/04-text-byte/charfinder/test.sh b/04-text-byte/charfinder/test.sh new file mode 100755 index 0000000..055fa84 --- /dev/null +++ b/04-text-byte/charfinder/test.sh @@ -0,0 +1,2 @@ +#!/bin/bash +python3 -m doctest cf_tests.rst $1 diff --git a/04-text-byte/numerics_demo.py b/04-text-byte/numerics_demo.py index 3dc5c7b..ce43f1b 100644 --- a/04-text-byte/numerics_demo.py +++ b/04-text-byte/numerics_demo.py @@ -1,4 +1,4 @@ -# BEGIN NUMERICS_DEMO +# tag::NUMERICS_DEMO[] import unicodedata import re @@ -15,4 +15,4 @@ for char in sample: format(unicodedata.numeric(char), '5.2f'), # <6> unicodedata.name(char), # <7> sep='\t') -# END NUMERICS_DEMO +# end::NUMERICS_DEMO[] diff --git a/04-text-byte/ramanujan.py b/04-text-byte/ramanujan.py index 09f214d..cafe072 100644 --- a/04-text-byte/ramanujan.py +++ b/04-text-byte/ramanujan.py @@ -1,4 +1,4 @@ -# BEGIN RE_DEMO +# tag::RE_DEMO[] import re re_numbers_str = re.compile(r'\d+') # <1> @@ -18,4 +18,4 @@ print(' bytes:', re_numbers_bytes.findall(text_bytes)) # <7> print('Words') print(' str :', re_words_str.findall(text_str)) # <8> print(' bytes:', re_words_bytes.findall(text_bytes)) # <9> -# END RE_DEMO +# end::RE_DEMO[] diff --git a/04-text-byte/sanitize.py b/04-text-byte/sanitize.py index a511a52..a86cb55 100644 --- a/04-text-byte/sanitize.py +++ b/04-text-byte/sanitize.py @@ -28,7 +28,7 @@ Handling a string with Greek and Latin accented characters: """ -# BEGIN SHAVE_MARKS +# tag::SHAVE_MARKS[] import unicodedata import string @@ -39,9 +39,9 @@ def shave_marks(txt): shaved = ''.join(c for c in norm_txt if not unicodedata.combining(c)) # <2> return unicodedata.normalize('NFC', shaved) # <3> -# END SHAVE_MARKS +# end::SHAVE_MARKS[] -# BEGIN SHAVE_MARKS_LATIN +# tag::SHAVE_MARKS_LATIN[] def shave_marks_latin(txt): """Remove all diacritic marks from Latin base characters""" norm_txt = unicodedata.normalize('NFD', txt) # <1> @@ -56,9 +56,9 @@ def shave_marks_latin(txt): latin_base = c in string.ascii_letters shaved = ''.join(keepers) return unicodedata.normalize('NFC', shaved) # <5> -# END SHAVE_MARKS_LATIN +# end::SHAVE_MARKS_LATIN[] -# BEGIN ASCIIZE +# tag::ASCIIZE[] single_map = str.maketrans("""‚ƒ„†ˆ‹‘’“”•–—˜›""", # <1> """'f"*^<''""---~>""") @@ -84,4 +84,4 @@ def asciize(txt): no_marks = shave_marks_latin(dewinize(txt)) # <5> no_marks = no_marks.replace('ß', 'ss') # <6> return unicodedata.normalize('NFKC', no_marks) # <7> -# END ASCIIZE +# end::ASCIIZE[] diff --git a/04-text-byte/skin.py b/04-text-byte/skin.py new file mode 100755 index 0000000..1247e04 --- /dev/null +++ b/04-text-byte/skin.py @@ -0,0 +1,12 @@ +from unicodedata import name + +SKIN1 = 0x1F3FB # EMOJI MODIFIER FITZPATRICK TYPE-1-2 # <1> +SKINS = [chr(i) for i in range(SKIN1, SKIN1 + 5)] # <2> +THUMB = '\U0001F44d' # THUMBS UP SIGN 👍 + +examples = [THUMB] # <3> +examples.extend(THUMB + skin for skin in SKINS) # <4> + +for example in examples: + print(example, end='\t') # <5> + print(' + '.join(name(char) for char in example)) # <6> diff --git a/04-text-byte/two_flags.py b/04-text-byte/two_flags.py new file mode 100644 index 0000000..feccea5 --- /dev/null +++ b/04-text-byte/two_flags.py @@ -0,0 +1,6 @@ +# REGIONAL INDICATOR SYMBOLS +RIS_A = '\U0001F1E6' # LETTER A +RIS_U = '\U0001F1FA' # LETTER U +print(RIS_A + RIS_U) # AU: Australia +print(RIS_U + RIS_A) # UA: Ukraine +print(RIS_A + RIS_A) # AA: no such country diff --git a/04-text-byte/zwj_sample.py b/04-text-byte/zwj_sample.py new file mode 100644 index 0000000..cdd7a30 --- /dev/null +++ b/04-text-byte/zwj_sample.py @@ -0,0 +1,28 @@ +from unicodedata import name + +zwg_sample = """ +1F468 200D 1F9B0 |man: red hair |E11.0 +1F9D1 200D 1F91D 200D 1F9D1 |people holding hands |E12.0 +1F3CA 1F3FF 200D 2640 FE0F |woman swimming: dark skin tone |E4.0 +1F469 1F3FE 200D 2708 FE0F |woman pilot: medium-dark skin tone |E4.0 +1F468 200D 1F469 200D 1F467 |family: man, woman, girl |E2.0 +1F3F3 FE0F 200D 26A7 FE0F |transgender flag |E13.0 +1F469 200D 2764 FE0F 200D 1F48B 200D 1F469 |kiss: woman, woman |E2.0 +""" + +markers = {'\u200D': 'ZWG', # ZERO WIDTH JOINER + '\uFE0F': 'V16', # VARIATION SELECTOR-16 + } + +for line in zwg_sample.strip().split('\n'): + code, descr, version = (s.strip() for s in line.split('|')) + chars = [chr(int(c, 16)) for c in code.split()] + print(''.join(chars), version, descr, sep='\t', end='') + while chars: + char = chars.pop(0) + if char in markers: + print(' + ' + markers[char], end='') + else: + ucode = f'U+{ord(char):04X}' + print(f'\n\t{char}\t{ucode}\t{name(char)}', end='') + print() diff --git a/appendix-A/arcfour.py b/appendix-A/arcfour.py new file mode 100644 index 0000000..7037f71 --- /dev/null +++ b/appendix-A/arcfour.py @@ -0,0 +1,55 @@ +"""RC4 compatible algorithm""" + +def arcfour(key, in_bytes, loops=20): + + kbox = bytearray(256) # create key box + for i, car in enumerate(key): # copy key and vector + kbox[i] = car + j = len(key) + for i in range(j, 256): # repeat until full + kbox[i] = kbox[i-j] + + # [1] initialize sbox + sbox = bytearray(range(256)) + + # repeat sbox mixing loop, as recommened in CipherSaber-2 + # http://ciphersaber.gurus.com/faq.html#cs2 + j = 0 + for k in range(loops): + for i in range(256): + j = (j + sbox[i] + kbox[i]) % 256 + sbox[i], sbox[j] = sbox[j], sbox[i] + + # main loop + i = 0 + j = 0 + out_bytes = bytearray() + + for car in in_bytes: + i = (i + 1) % 256 + # [2] shuffle sbox + j = (j + sbox[i]) % 256 + sbox[i], sbox[j] = sbox[j], sbox[i] + # [3] compute t + t = (sbox[i] + sbox[j]) % 256 + k = sbox[t] + car = car ^ k + out_bytes.append(car) + + return out_bytes + + +def test(): + from time import time + clear = bytearray(b'1234567890' * 100000) + t0 = time() + cipher = arcfour(b'key', clear) + print('elapsed time: %.2fs' % (time() - t0)) + result = arcfour(b'key', cipher) + assert result == clear, '%r != %r' % (result, clear) + print('elapsed time: %.2fs' % (time() - t0)) + print('OK') + + +if __name__ == '__main__': + test() diff --git a/appendix-A/arcfour_futures.py b/appendix-A/arcfour_futures.py new file mode 100644 index 0000000..a1eea81 --- /dev/null +++ b/appendix-A/arcfour_futures.py @@ -0,0 +1,46 @@ +import sys +import time +from concurrent import futures +from random import randrange +from arcfour import arcfour + +JOBS = 12 +SIZE = 2**18 + +KEY = b"'Twas brillig, and the slithy toves\nDid gyre" +STATUS = '{} workers, elapsed time: {:.2f}s' + + +def arcfour_test(size, key): + in_text = bytearray(randrange(256) for i in range(size)) + cypher_text = arcfour(key, in_text) + out_text = arcfour(key, cypher_text) + assert in_text == out_text, 'Failed arcfour_test' + return size + + +def main(workers=None): + if workers: + workers = int(workers) + t0 = time.time() + + with futures.ProcessPoolExecutor(workers) as executor: + actual_workers = executor._max_workers + to_do = [] + for i in range(JOBS, 0, -1): + size = SIZE + int(SIZE / JOBS * (i - JOBS/2)) + job = executor.submit(arcfour_test, size, KEY) + to_do.append(job) + + for future in futures.as_completed(to_do): + res = future.result() + print('{:.1f} KB'.format(res/2**10)) + + print(STATUS.format(actual_workers, time.time() - t0)) + +if __name__ == '__main__': + if len(sys.argv) == 2: + workers = int(sys.argv[1]) + else: + workers = None + main(workers) diff --git a/appendix-A/bottle.py b/appendix-A/bottle.py new file mode 100644 index 0000000..96ca3e4 --- /dev/null +++ b/appendix-A/bottle.py @@ -0,0 +1,3771 @@ +#!/usr/bin/env python +# -*- coding: utf-8 -*- +""" +Bottle is a fast and simple micro-framework for small web applications. It +offers request dispatching (Routes) with url parameter support, templates, +a built-in HTTP Server and adapters for many third party WSGI/HTTP-server and +template engines - all in a single file and with no dependencies other than the +Python Standard Library. + +Homepage and documentation: http://bottlepy.org/ + +Copyright (c) 2016, Marcel Hellkamp. +License: MIT (see LICENSE for details) +""" + +from __future__ import with_statement + +__author__ = 'Marcel Hellkamp' +__version__ = '0.12.18' +__license__ = 'MIT' + +# The gevent server adapter needs to patch some modules before they are imported +# This is why we parse the commandline parameters here but handle them later +if __name__ == '__main__': + from optparse import OptionParser + _cmd_parser = OptionParser(usage="usage: %prog [options] package.module:app") + _opt = _cmd_parser.add_option + _opt("--version", action="store_true", help="show version number.") + _opt("-b", "--bind", metavar="ADDRESS", help="bind socket to ADDRESS.") + _opt("-s", "--server", default='wsgiref', help="use SERVER as backend.") + _opt("-p", "--plugin", action="append", help="install additional plugin/s.") + _opt("--debug", action="store_true", help="start server in debug mode.") + _opt("--reload", action="store_true", help="auto-reload on file changes.") + _cmd_options, _cmd_args = _cmd_parser.parse_args() + if _cmd_options.server and _cmd_options.server.startswith('gevent'): + import gevent.monkey; gevent.monkey.patch_all() + +import base64, cgi, email.utils, functools, hmac, itertools, mimetypes,\ + os, re, subprocess, sys, tempfile, threading, time, warnings, hashlib + +from datetime import date as datedate, datetime, timedelta +from tempfile import TemporaryFile +from traceback import format_exc, print_exc +from inspect import getargspec +from unicodedata import normalize + + +try: from simplejson import dumps as json_dumps, loads as json_lds +except ImportError: # pragma: no cover + try: from json import dumps as json_dumps, loads as json_lds + except ImportError: + try: from django.utils.simplejson import dumps as json_dumps, loads as json_lds + except ImportError: + def json_dumps(data): + raise ImportError("JSON support requires Python 2.6 or simplejson.") + json_lds = json_dumps + + + +# We now try to fix 2.5/2.6/3.1/3.2 incompatibilities. +# It ain't pretty but it works... Sorry for the mess. + +py = sys.version_info +py3k = py >= (3, 0, 0) +py25 = py < (2, 6, 0) +py31 = (3, 1, 0) <= py < (3, 2, 0) + +# Workaround for the missing "as" keyword in py3k. +def _e(): return sys.exc_info()[1] + +# Workaround for the "print is a keyword/function" Python 2/3 dilemma +# and a fallback for mod_wsgi (resticts stdout/err attribute access) +try: + _stdout, _stderr = sys.stdout.write, sys.stderr.write +except IOError: + _stdout = lambda x: sys.stdout.write(x) + _stderr = lambda x: sys.stderr.write(x) + +# Lots of stdlib and builtin differences. +if py3k: + import http.client as httplib + import _thread as thread + from urllib.parse import urljoin, SplitResult as UrlSplitResult + from urllib.parse import urlencode, quote as urlquote, unquote as urlunquote + urlunquote = functools.partial(urlunquote, encoding='latin1') + from http.cookies import SimpleCookie + if py >= (3, 3, 0): + from collections.abc import MutableMapping as DictMixin + from types import ModuleType as new_module + else: + from collections import MutableMapping as DictMixin + from imp import new_module + import pickle + from io import BytesIO + from configparser import ConfigParser + basestring = str + unicode = str + json_loads = lambda s: json_lds(touni(s)) + callable = lambda x: hasattr(x, '__call__') + imap = map + def _raise(*a): raise a[0](a[1]).with_traceback(a[2]) +else: # 2.x + import httplib + import thread + from urlparse import urljoin, SplitResult as UrlSplitResult + from urllib import urlencode, quote as urlquote, unquote as urlunquote + from Cookie import SimpleCookie + from itertools import imap + import cPickle as pickle + from imp import new_module + from StringIO import StringIO as BytesIO + from ConfigParser import SafeConfigParser as ConfigParser + if py25: + msg = "Python 2.5 support may be dropped in future versions of Bottle." + warnings.warn(msg, DeprecationWarning) + from UserDict import DictMixin + def next(it): return it.next() + bytes = str + else: # 2.6, 2.7 + from collections import MutableMapping as DictMixin + unicode = unicode + json_loads = json_lds + eval(compile('def _raise(*a): raise a[0], a[1], a[2]', '', 'exec')) + +# Some helpers for string/byte handling +def tob(s, enc='utf8'): + return s.encode(enc) if isinstance(s, unicode) else bytes(s) +def touni(s, enc='utf8', err='strict'): + return s.decode(enc, err) if isinstance(s, bytes) else unicode(s) +tonat = touni if py3k else tob + +# 3.2 fixes cgi.FieldStorage to accept bytes (which makes a lot of sense). +# 3.1 needs a workaround. +if py31: + from io import TextIOWrapper + class NCTextIOWrapper(TextIOWrapper): + def close(self): pass # Keep wrapped buffer open. + + +# A bug in functools causes it to break if the wrapper is an instance method +def update_wrapper(wrapper, wrapped, *a, **ka): + try: functools.update_wrapper(wrapper, wrapped, *a, **ka) + except AttributeError: pass + + + +# These helpers are used at module level and need to be defined first. +# And yes, I know PEP-8, but sometimes a lower-case classname makes more sense. + +def depr(message, hard=False): + warnings.warn(message, DeprecationWarning, stacklevel=3) + +def makelist(data): # This is just to handy + if isinstance(data, (tuple, list, set, dict)): return list(data) + elif data: return [data] + else: return [] + + +class DictProperty(object): + ''' Property that maps to a key in a local dict-like attribute. ''' + def __init__(self, attr, key=None, read_only=False): + self.attr, self.key, self.read_only = attr, key, read_only + + def __call__(self, func): + functools.update_wrapper(self, func, updated=[]) + self.getter, self.key = func, self.key or func.__name__ + return self + + def __get__(self, obj, cls): + if obj is None: return self + key, storage = self.key, getattr(obj, self.attr) + if key not in storage: storage[key] = self.getter(obj) + return storage[key] + + def __set__(self, obj, value): + if self.read_only: raise AttributeError("Read-Only property.") + getattr(obj, self.attr)[self.key] = value + + def __delete__(self, obj): + if self.read_only: raise AttributeError("Read-Only property.") + del getattr(obj, self.attr)[self.key] + + +class cached_property(object): + ''' A property that is only computed once per instance and then replaces + itself with an ordinary attribute. Deleting the attribute resets the + property. ''' + + def __init__(self, func): + self.__doc__ = getattr(func, '__doc__') + self.func = func + + def __get__(self, obj, cls): + if obj is None: return self + value = obj.__dict__[self.func.__name__] = self.func(obj) + return value + + +class lazy_attribute(object): + ''' A property that caches itself to the class object. ''' + def __init__(self, func): + functools.update_wrapper(self, func, updated=[]) + self.getter = func + + def __get__(self, obj, cls): + value = self.getter(cls) + setattr(cls, self.__name__, value) + return value + + + + + + +############################################################################### +# Exceptions and Events ######################################################## +############################################################################### + + +class BottleException(Exception): + """ A base class for exceptions used by bottle. """ + pass + + + + + + +############################################################################### +# Routing ###################################################################### +############################################################################### + + +class RouteError(BottleException): + """ This is a base class for all routing related exceptions """ + + +class RouteReset(BottleException): + """ If raised by a plugin or request handler, the route is reset and all + plugins are re-applied. """ + +class RouterUnknownModeError(RouteError): pass + + +class RouteSyntaxError(RouteError): + """ The route parser found something not supported by this router. """ + + +class RouteBuildError(RouteError): + """ The route could not be built. """ + + +def _re_flatten(p): + ''' Turn all capturing groups in a regular expression pattern into + non-capturing groups. ''' + if '(' not in p: return p + return re.sub(r'(\\*)(\(\?P<[^>]+>|\((?!\?))', + lambda m: m.group(0) if len(m.group(1)) % 2 else m.group(1) + '(?:', p) + + +class Router(object): + ''' A Router is an ordered collection of route->target pairs. It is used to + efficiently match WSGI requests against a number of routes and return + the first target that satisfies the request. The target may be anything, + usually a string, ID or callable object. A route consists of a path-rule + and a HTTP method. + + The path-rule is either a static path (e.g. `/contact`) or a dynamic + path that contains wildcards (e.g. `/wiki/`). The wildcard syntax + and details on the matching order are described in docs:`routing`. + ''' + + default_pattern = '[^/]+' + default_filter = 're' + + #: The current CPython regexp implementation does not allow more + #: than 99 matching groups per regular expression. + _MAX_GROUPS_PER_PATTERN = 99 + + def __init__(self, strict=False): + self.rules = [] # All rules in order + self._groups = {} # index of regexes to find them in dyna_routes + self.builder = {} # Data structure for the url builder + self.static = {} # Search structure for static routes + self.dyna_routes = {} + self.dyna_regexes = {} # Search structure for dynamic routes + #: If true, static routes are no longer checked first. + self.strict_order = strict + self.filters = { + 're': lambda conf: + (_re_flatten(conf or self.default_pattern), None, None), + 'int': lambda conf: (r'-?\d+', int, lambda x: str(int(x))), + 'float': lambda conf: (r'-?[\d.]+', float, lambda x: str(float(x))), + 'path': lambda conf: (r'.+?', None, None)} + + def add_filter(self, name, func): + ''' Add a filter. The provided function is called with the configuration + string as parameter and must return a (regexp, to_python, to_url) tuple. + The first element is a string, the last two are callables or None. ''' + self.filters[name] = func + + rule_syntax = re.compile('(\\\\*)'\ + '(?:(?::([a-zA-Z_][a-zA-Z_0-9]*)?()(?:#(.*?)#)?)'\ + '|(?:<([a-zA-Z_][a-zA-Z_0-9]*)?(?::([a-zA-Z_]*)'\ + '(?::((?:\\\\.|[^\\\\>]+)+)?)?)?>))') + + def _itertokens(self, rule): + offset, prefix = 0, '' + for match in self.rule_syntax.finditer(rule): + prefix += rule[offset:match.start()] + g = match.groups() + if len(g[0])%2: # Escaped wildcard + prefix += match.group(0)[len(g[0]):] + offset = match.end() + continue + if prefix: + yield prefix, None, None + name, filtr, conf = g[4:7] if g[2] is None else g[1:4] + yield name, filtr or 'default', conf or None + offset, prefix = match.end(), '' + if offset <= len(rule) or prefix: + yield prefix+rule[offset:], None, None + + def add(self, rule, method, target, name=None): + ''' Add a new rule or replace the target for an existing rule. ''' + anons = 0 # Number of anonymous wildcards found + keys = [] # Names of keys + pattern = '' # Regular expression pattern with named groups + filters = [] # Lists of wildcard input filters + builder = [] # Data structure for the URL builder + is_static = True + + for key, mode, conf in self._itertokens(rule): + if mode: + is_static = False + if mode == 'default': mode = self.default_filter + mask, in_filter, out_filter = self.filters[mode](conf) + if not key: + pattern += '(?:%s)' % mask + key = 'anon%d' % anons + anons += 1 + else: + pattern += '(?P<%s>%s)' % (key, mask) + keys.append(key) + if in_filter: filters.append((key, in_filter)) + builder.append((key, out_filter or str)) + elif key: + pattern += re.escape(key) + builder.append((None, key)) + + self.builder[rule] = builder + if name: self.builder[name] = builder + + if is_static and not self.strict_order: + self.static.setdefault(method, {}) + self.static[method][self.build(rule)] = (target, None) + return + + try: + re_pattern = re.compile('^(%s)$' % pattern) + re_match = re_pattern.match + except re.error: + raise RouteSyntaxError("Could not add Route: %s (%s)" % (rule, _e())) + + if filters: + def getargs(path): + url_args = re_match(path).groupdict() + for name, wildcard_filter in filters: + try: + url_args[name] = wildcard_filter(url_args[name]) + except ValueError: + raise HTTPError(400, 'Path has wrong format.') + return url_args + elif re_pattern.groupindex: + def getargs(path): + return re_match(path).groupdict() + else: + getargs = None + + flatpat = _re_flatten(pattern) + whole_rule = (rule, flatpat, target, getargs) + + if (flatpat, method) in self._groups: + if DEBUG: + msg = 'Route <%s %s> overwrites a previously defined route' + warnings.warn(msg % (method, rule), RuntimeWarning) + self.dyna_routes[method][self._groups[flatpat, method]] = whole_rule + else: + self.dyna_routes.setdefault(method, []).append(whole_rule) + self._groups[flatpat, method] = len(self.dyna_routes[method]) - 1 + + self._compile(method) + + def _compile(self, method): + all_rules = self.dyna_routes[method] + comborules = self.dyna_regexes[method] = [] + maxgroups = self._MAX_GROUPS_PER_PATTERN + for x in range(0, len(all_rules), maxgroups): + some = all_rules[x:x+maxgroups] + combined = (flatpat for (_, flatpat, _, _) in some) + combined = '|'.join('(^%s$)' % flatpat for flatpat in combined) + combined = re.compile(combined).match + rules = [(target, getargs) for (_, _, target, getargs) in some] + comborules.append((combined, rules)) + + def build(self, _name, *anons, **query): + ''' Build an URL by filling the wildcards in a rule. ''' + builder = self.builder.get(_name) + if not builder: raise RouteBuildError("No route with that name.", _name) + try: + for i, value in enumerate(anons): query['anon%d'%i] = value + url = ''.join([f(query.pop(n)) if n else f for (n,f) in builder]) + return url if not query else url+'?'+urlencode(query) + except KeyError: + raise RouteBuildError('Missing URL argument: %r' % _e().args[0]) + + def match(self, environ): + ''' Return a (target, url_agrs) tuple or raise HTTPError(400/404/405). ''' + verb = environ['REQUEST_METHOD'].upper() + path = environ['PATH_INFO'] or '/' + target = None + if verb == 'HEAD': + methods = ['PROXY', verb, 'GET', 'ANY'] + else: + methods = ['PROXY', verb, 'ANY'] + + for method in methods: + if method in self.static and path in self.static[method]: + target, getargs = self.static[method][path] + return target, getargs(path) if getargs else {} + elif method in self.dyna_regexes: + for combined, rules in self.dyna_regexes[method]: + match = combined(path) + if match: + target, getargs = rules[match.lastindex - 1] + return target, getargs(path) if getargs else {} + + # No matching route found. Collect alternative methods for 405 response + allowed = set([]) + nocheck = set(methods) + for method in set(self.static) - nocheck: + if path in self.static[method]: + allowed.add(verb) + for method in set(self.dyna_regexes) - allowed - nocheck: + for combined, rules in self.dyna_regexes[method]: + match = combined(path) + if match: + allowed.add(method) + if allowed: + allow_header = ",".join(sorted(allowed)) + raise HTTPError(405, "Method not allowed.", Allow=allow_header) + + # No matching route and no alternative method found. We give up + raise HTTPError(404, "Not found: " + repr(path)) + + + + + + +class Route(object): + ''' This class wraps a route callback along with route specific metadata and + configuration and applies Plugins on demand. It is also responsible for + turing an URL path rule into a regular expression usable by the Router. + ''' + + def __init__(self, app, rule, method, callback, name=None, + plugins=None, skiplist=None, **config): + #: The application this route is installed to. + self.app = app + #: The path-rule string (e.g. ``/wiki/:page``). + self.rule = rule + #: The HTTP method as a string (e.g. ``GET``). + self.method = method + #: The original callback with no plugins applied. Useful for introspection. + self.callback = callback + #: The name of the route (if specified) or ``None``. + self.name = name or None + #: A list of route-specific plugins (see :meth:`Bottle.route`). + self.plugins = plugins or [] + #: A list of plugins to not apply to this route (see :meth:`Bottle.route`). + self.skiplist = skiplist or [] + #: Additional keyword arguments passed to the :meth:`Bottle.route` + #: decorator are stored in this dictionary. Used for route-specific + #: plugin configuration and meta-data. + self.config = ConfigDict().load_dict(config, make_namespaces=True) + + def __call__(self, *a, **ka): + depr("Some APIs changed to return Route() instances instead of"\ + " callables. Make sure to use the Route.call method and not to"\ + " call Route instances directly.") #0.12 + return self.call(*a, **ka) + + @cached_property + def call(self): + ''' The route callback with all plugins applied. This property is + created on demand and then cached to speed up subsequent requests.''' + return self._make_callback() + + def reset(self): + ''' Forget any cached values. The next time :attr:`call` is accessed, + all plugins are re-applied. ''' + self.__dict__.pop('call', None) + + def prepare(self): + ''' Do all on-demand work immediately (useful for debugging).''' + self.call + + @property + def _context(self): + depr('Switch to Plugin API v2 and access the Route object directly.') #0.12 + return dict(rule=self.rule, method=self.method, callback=self.callback, + name=self.name, app=self.app, config=self.config, + apply=self.plugins, skip=self.skiplist) + + def all_plugins(self): + ''' Yield all Plugins affecting this route. ''' + unique = set() + for p in reversed(self.app.plugins + self.plugins): + if True in self.skiplist: break + name = getattr(p, 'name', False) + if name and (name in self.skiplist or name in unique): continue + if p in self.skiplist or type(p) in self.skiplist: continue + if name: unique.add(name) + yield p + + def _make_callback(self): + callback = self.callback + for plugin in self.all_plugins(): + try: + if hasattr(plugin, 'apply'): + api = getattr(plugin, 'api', 1) + context = self if api > 1 else self._context + callback = plugin.apply(callback, context) + else: + callback = plugin(callback) + except RouteReset: # Try again with changed configuration. + return self._make_callback() + if not callback is self.callback: + update_wrapper(callback, self.callback) + return callback + + def get_undecorated_callback(self): + ''' Return the callback. If the callback is a decorated function, try to + recover the original function. ''' + func = self.callback + func = getattr(func, '__func__' if py3k else 'im_func', func) + closure_attr = '__closure__' if py3k else 'func_closure' + while hasattr(func, closure_attr) and getattr(func, closure_attr): + func = getattr(func, closure_attr)[0].cell_contents + return func + + def get_callback_args(self): + ''' Return a list of argument names the callback (most likely) accepts + as keyword arguments. If the callback is a decorated function, try + to recover the original function before inspection. ''' + return getargspec(self.get_undecorated_callback())[0] + + def get_config(self, key, default=None): + ''' Lookup a config field and return its value, first checking the + route.config, then route.app.config.''' + for conf in (self.config, self.app.conifg): + if key in conf: return conf[key] + return default + + def __repr__(self): + cb = self.get_undecorated_callback() + return '<%s %r %r>' % (self.method, self.rule, cb) + + + + + + +############################################################################### +# Application Object ########################################################### +############################################################################### + + +class Bottle(object): + """ Each Bottle object represents a single, distinct web application and + consists of routes, callbacks, plugins, resources and configuration. + Instances are callable WSGI applications. + + :param catchall: If true (default), handle all exceptions. Turn off to + let debugging middleware handle exceptions. + """ + + def __init__(self, catchall=True, autojson=True): + + #: A :class:`ConfigDict` for app specific configuration. + self.config = ConfigDict() + self.config._on_change = functools.partial(self.trigger_hook, 'config') + self.config.meta_set('autojson', 'validate', bool) + self.config.meta_set('catchall', 'validate', bool) + self.config['catchall'] = catchall + self.config['autojson'] = autojson + + #: A :class:`ResourceManager` for application files + self.resources = ResourceManager() + + self.routes = [] # List of installed :class:`Route` instances. + self.router = Router() # Maps requests to :class:`Route` instances. + self.error_handler = {} + + # Core plugins + self.plugins = [] # List of installed plugins. + if self.config['autojson']: + self.install(JSONPlugin()) + self.install(TemplatePlugin()) + + #: If true, most exceptions are caught and returned as :exc:`HTTPError` + catchall = DictProperty('config', 'catchall') + + __hook_names = 'before_request', 'after_request', 'app_reset', 'config' + __hook_reversed = 'after_request' + + @cached_property + def _hooks(self): + return dict((name, []) for name in self.__hook_names) + + def add_hook(self, name, func): + ''' Attach a callback to a hook. Three hooks are currently implemented: + + before_request + Executed once before each request. The request context is + available, but no routing has happened yet. + after_request + Executed once after each request regardless of its outcome. + app_reset + Called whenever :meth:`Bottle.reset` is called. + ''' + if name in self.__hook_reversed: + self._hooks[name].insert(0, func) + else: + self._hooks[name].append(func) + + def remove_hook(self, name, func): + ''' Remove a callback from a hook. ''' + if name in self._hooks and func in self._hooks[name]: + self._hooks[name].remove(func) + return True + + def trigger_hook(self, __name, *args, **kwargs): + ''' Trigger a hook and return a list of results. ''' + return [hook(*args, **kwargs) for hook in self._hooks[__name][:]] + + def hook(self, name): + """ Return a decorator that attaches a callback to a hook. See + :meth:`add_hook` for details.""" + def decorator(func): + self.add_hook(name, func) + return func + return decorator + + def mount(self, prefix, app, **options): + ''' Mount an application (:class:`Bottle` or plain WSGI) to a specific + URL prefix. Example:: + + root_app.mount('/admin/', admin_app) + + :param prefix: path prefix or `mount-point`. If it ends in a slash, + that slash is mandatory. + :param app: an instance of :class:`Bottle` or a WSGI application. + + All other parameters are passed to the underlying :meth:`route` call. + ''' + if isinstance(app, basestring): + depr('Parameter order of Bottle.mount() changed.', True) # 0.10 + + segments = [p for p in prefix.split('/') if p] + if not segments: raise ValueError('Empty path prefix.') + path_depth = len(segments) + + def mountpoint_wrapper(): + try: + request.path_shift(path_depth) + rs = HTTPResponse([]) + def start_response(status, headerlist, exc_info=None): + if exc_info: + try: + _raise(*exc_info) + finally: + exc_info = None + rs.status = status + for name, value in headerlist: rs.add_header(name, value) + return rs.body.append + body = app(request.environ, start_response) + if body and rs.body: body = itertools.chain(rs.body, body) + rs.body = body or rs.body + return rs + finally: + request.path_shift(-path_depth) + + options.setdefault('skip', True) + options.setdefault('method', 'PROXY') + options.setdefault('mountpoint', {'prefix': prefix, 'target': app}) + options['callback'] = mountpoint_wrapper + + self.route('/%s/<:re:.*>' % '/'.join(segments), **options) + if not prefix.endswith('/'): + self.route('/' + '/'.join(segments), **options) + + def merge(self, routes): + ''' Merge the routes of another :class:`Bottle` application or a list of + :class:`Route` objects into this application. The routes keep their + 'owner', meaning that the :data:`Route.app` attribute is not + changed. ''' + if isinstance(routes, Bottle): + routes = routes.routes + for route in routes: + self.add_route(route) + + def install(self, plugin): + ''' Add a plugin to the list of plugins and prepare it for being + applied to all routes of this application. A plugin may be a simple + decorator or an object that implements the :class:`Plugin` API. + ''' + if hasattr(plugin, 'setup'): plugin.setup(self) + if not callable(plugin) and not hasattr(plugin, 'apply'): + raise TypeError("Plugins must be callable or implement .apply()") + self.plugins.append(plugin) + self.reset() + return plugin + + def uninstall(self, plugin): + ''' Uninstall plugins. Pass an instance to remove a specific plugin, a type + object to remove all plugins that match that type, a string to remove + all plugins with a matching ``name`` attribute or ``True`` to remove all + plugins. Return the list of removed plugins. ''' + removed, remove = [], plugin + for i, plugin in list(enumerate(self.plugins))[::-1]: + if remove is True or remove is plugin or remove is type(plugin) \ + or getattr(plugin, 'name', True) == remove: + removed.append(plugin) + del self.plugins[i] + if hasattr(plugin, 'close'): plugin.close() + if removed: self.reset() + return removed + + def reset(self, route=None): + ''' Reset all routes (force plugins to be re-applied) and clear all + caches. If an ID or route object is given, only that specific route + is affected. ''' + if route is None: routes = self.routes + elif isinstance(route, Route): routes = [route] + else: routes = [self.routes[route]] + for route in routes: route.reset() + if DEBUG: + for route in routes: route.prepare() + self.trigger_hook('app_reset') + + def close(self): + ''' Close the application and all installed plugins. ''' + for plugin in self.plugins: + if hasattr(plugin, 'close'): plugin.close() + self.stopped = True + + def run(self, **kwargs): + ''' Calls :func:`run` with the same parameters. ''' + run(self, **kwargs) + + def match(self, environ): + """ Search for a matching route and return a (:class:`Route` , urlargs) + tuple. The second value is a dictionary with parameters extracted + from the URL. Raise :exc:`HTTPError` (404/405) on a non-match.""" + return self.router.match(environ) + + def get_url(self, routename, **kargs): + """ Return a string that matches a named route """ + scriptname = request.environ.get('SCRIPT_NAME', '').strip('/') + '/' + location = self.router.build(routename, **kargs).lstrip('/') + return urljoin(urljoin('/', scriptname), location) + + def add_route(self, route): + ''' Add a route object, but do not change the :data:`Route.app` + attribute.''' + self.routes.append(route) + self.router.add(route.rule, route.method, route, name=route.name) + if DEBUG: route.prepare() + + def route(self, path=None, method='GET', callback=None, name=None, + apply=None, skip=None, **config): + """ A decorator to bind a function to a request URL. Example:: + + @app.route('/hello/:name') + def hello(name): + return 'Hello %s' % name + + The ``:name`` part is a wildcard. See :class:`Router` for syntax + details. + + :param path: Request path or a list of paths to listen to. If no + path is specified, it is automatically generated from the + signature of the function. + :param method: HTTP method (`GET`, `POST`, `PUT`, ...) or a list of + methods to listen to. (default: `GET`) + :param callback: An optional shortcut to avoid the decorator + syntax. ``route(..., callback=func)`` equals ``route(...)(func)`` + :param name: The name for this route. (default: None) + :param apply: A decorator or plugin or a list of plugins. These are + applied to the route callback in addition to installed plugins. + :param skip: A list of plugins, plugin classes or names. Matching + plugins are not installed to this route. ``True`` skips all. + + Any additional keyword arguments are stored as route-specific + configuration and passed to plugins (see :meth:`Plugin.apply`). + """ + if callable(path): path, callback = None, path + plugins = makelist(apply) + skiplist = makelist(skip) + def decorator(callback): + # TODO: Documentation and tests + if isinstance(callback, basestring): callback = load(callback) + for rule in makelist(path) or yieldroutes(callback): + for verb in makelist(method): + verb = verb.upper() + route = Route(self, rule, verb, callback, name=name, + plugins=plugins, skiplist=skiplist, **config) + self.add_route(route) + return callback + return decorator(callback) if callback else decorator + + def get(self, path=None, method='GET', **options): + """ Equals :meth:`route`. """ + return self.route(path, method, **options) + + def post(self, path=None, method='POST', **options): + """ Equals :meth:`route` with a ``POST`` method parameter. """ + return self.route(path, method, **options) + + def put(self, path=None, method='PUT', **options): + """ Equals :meth:`route` with a ``PUT`` method parameter. """ + return self.route(path, method, **options) + + def delete(self, path=None, method='DELETE', **options): + """ Equals :meth:`route` with a ``DELETE`` method parameter. """ + return self.route(path, method, **options) + + def error(self, code=500): + """ Decorator: Register an output handler for a HTTP error code""" + def wrapper(handler): + self.error_handler[int(code)] = handler + return handler + return wrapper + + def default_error_handler(self, res): + return tob(template(ERROR_PAGE_TEMPLATE, e=res)) + + def _handle(self, environ): + path = environ['bottle.raw_path'] = environ['PATH_INFO'] + if py3k: + try: + environ['PATH_INFO'] = path.encode('latin1').decode('utf8') + except UnicodeError: + return HTTPError(400, 'Invalid path string. Expected UTF-8') + + try: + environ['bottle.app'] = self + request.bind(environ) + response.bind() + try: + self.trigger_hook('before_request') + route, args = self.router.match(environ) + environ['route.handle'] = route + environ['bottle.route'] = route + environ['route.url_args'] = args + return route.call(**args) + finally: + self.trigger_hook('after_request') + + except HTTPResponse: + return _e() + except RouteReset: + route.reset() + return self._handle(environ) + except (KeyboardInterrupt, SystemExit, MemoryError): + raise + except Exception: + if not self.catchall: raise + stacktrace = format_exc() + environ['wsgi.errors'].write(stacktrace) + return HTTPError(500, "Internal Server Error", _e(), stacktrace) + + def _cast(self, out, peek=None): + """ Try to convert the parameter into something WSGI compatible and set + correct HTTP headers when possible. + Support: False, str, unicode, dict, HTTPResponse, HTTPError, file-like, + iterable of strings and iterable of unicodes + """ + + # Empty output is done here + if not out: + if 'Content-Length' not in response: + response['Content-Length'] = 0 + return [] + # Join lists of byte or unicode strings. Mixed lists are NOT supported + if isinstance(out, (tuple, list))\ + and isinstance(out[0], (bytes, unicode)): + out = out[0][0:0].join(out) # b'abc'[0:0] -> b'' + # Encode unicode strings + if isinstance(out, unicode): + out = out.encode(response.charset) + # Byte Strings are just returned + if isinstance(out, bytes): + if 'Content-Length' not in response: + response['Content-Length'] = len(out) + return [out] + # HTTPError or HTTPException (recursive, because they may wrap anything) + # TODO: Handle these explicitly in handle() or make them iterable. + if isinstance(out, HTTPError): + out.apply(response) + out = self.error_handler.get(out.status_code, self.default_error_handler)(out) + return self._cast(out) + if isinstance(out, HTTPResponse): + out.apply(response) + return self._cast(out.body) + + # File-like objects. + if hasattr(out, 'read'): + if 'wsgi.file_wrapper' in request.environ: + return request.environ['wsgi.file_wrapper'](out) + elif hasattr(out, 'close') or not hasattr(out, '__iter__'): + return WSGIFileWrapper(out) + + # Handle Iterables. We peek into them to detect their inner type. + try: + iout = iter(out) + first = next(iout) + while not first: + first = next(iout) + except StopIteration: + return self._cast('') + except HTTPResponse: + first = _e() + except (KeyboardInterrupt, SystemExit, MemoryError): + raise + except Exception: + if not self.catchall: raise + first = HTTPError(500, 'Unhandled exception', _e(), format_exc()) + + # These are the inner types allowed in iterator or generator objects. + if isinstance(first, HTTPResponse): + return self._cast(first) + elif isinstance(first, bytes): + new_iter = itertools.chain([first], iout) + elif isinstance(first, unicode): + encoder = lambda x: x.encode(response.charset) + new_iter = imap(encoder, itertools.chain([first], iout)) + else: + msg = 'Unsupported response type: %s' % type(first) + return self._cast(HTTPError(500, msg)) + if hasattr(out, 'close'): + new_iter = _closeiter(new_iter, out.close) + return new_iter + + def wsgi(self, environ, start_response): + """ The bottle WSGI-interface. """ + try: + out = self._cast(self._handle(environ)) + # rfc2616 section 4.3 + if response._status_code in (100, 101, 204, 304)\ + or environ['REQUEST_METHOD'] == 'HEAD': + if hasattr(out, 'close'): out.close() + out = [] + start_response(response._status_line, response.headerlist) + return out + except (KeyboardInterrupt, SystemExit, MemoryError): + raise + except Exception: + if not self.catchall: raise + err = '

Critical error while processing request: %s

' \ + % html_escape(environ.get('PATH_INFO', '/')) + if DEBUG: + err += '

Error:

\n
\n%s\n
\n' \ + '

Traceback:

\n
\n%s\n
\n' \ + % (html_escape(repr(_e())), html_escape(format_exc())) + environ['wsgi.errors'].write(err) + headers = [('Content-Type', 'text/html; charset=UTF-8')] + start_response('500 INTERNAL SERVER ERROR', headers, sys.exc_info()) + return [tob(err)] + + def __call__(self, environ, start_response): + ''' Each instance of :class:'Bottle' is a WSGI application. ''' + return self.wsgi(environ, start_response) + + + + + + +############################################################################### +# HTTP and WSGI Tools ########################################################## +############################################################################### + +class BaseRequest(object): + """ A wrapper for WSGI environment dictionaries that adds a lot of + convenient access methods and properties. Most of them are read-only. + + Adding new attributes to a request actually adds them to the environ + dictionary (as 'bottle.request.ext.'). This is the recommended + way to store and access request-specific data. + """ + + __slots__ = ('environ') + + #: Maximum size of memory buffer for :attr:`body` in bytes. + MEMFILE_MAX = 102400 + + def __init__(self, environ=None): + """ Wrap a WSGI environ dictionary. """ + #: The wrapped WSGI environ dictionary. This is the only real attribute. + #: All other attributes actually are read-only properties. + self.environ = {} if environ is None else environ + self.environ['bottle.request'] = self + + @DictProperty('environ', 'bottle.app', read_only=True) + def app(self): + ''' Bottle application handling this request. ''' + raise RuntimeError('This request is not connected to an application.') + + @DictProperty('environ', 'bottle.route', read_only=True) + def route(self): + """ The bottle :class:`Route` object that matches this request. """ + raise RuntimeError('This request is not connected to a route.') + + @DictProperty('environ', 'route.url_args', read_only=True) + def url_args(self): + """ The arguments extracted from the URL. """ + raise RuntimeError('This request is not connected to a route.') + + @property + def path(self): + ''' The value of ``PATH_INFO`` with exactly one prefixed slash (to fix + broken clients and avoid the "empty path" edge case). ''' + return '/' + self.environ.get('PATH_INFO','').lstrip('/') + + @property + def method(self): + ''' The ``REQUEST_METHOD`` value as an uppercase string. ''' + return self.environ.get('REQUEST_METHOD', 'GET').upper() + + @DictProperty('environ', 'bottle.request.headers', read_only=True) + def headers(self): + ''' A :class:`WSGIHeaderDict` that provides case-insensitive access to + HTTP request headers. ''' + return WSGIHeaderDict(self.environ) + + def get_header(self, name, default=None): + ''' Return the value of a request header, or a given default value. ''' + return self.headers.get(name, default) + + @DictProperty('environ', 'bottle.request.cookies', read_only=True) + def cookies(self): + """ Cookies parsed into a :class:`FormsDict`. Signed cookies are NOT + decoded. Use :meth:`get_cookie` if you expect signed cookies. """ + cookies = SimpleCookie(self.environ.get('HTTP_COOKIE','')).values() + return FormsDict((c.key, c.value) for c in cookies) + + def get_cookie(self, key, default=None, secret=None): + """ Return the content of a cookie. To read a `Signed Cookie`, the + `secret` must match the one used to create the cookie (see + :meth:`BaseResponse.set_cookie`). If anything goes wrong (missing + cookie or wrong signature), return a default value. """ + value = self.cookies.get(key) + if secret and value: + dec = cookie_decode(value, secret) # (key, value) tuple or None + return dec[1] if dec and dec[0] == key else default + return value or default + + @DictProperty('environ', 'bottle.request.query', read_only=True) + def query(self): + ''' The :attr:`query_string` parsed into a :class:`FormsDict`. These + values are sometimes called "URL arguments" or "GET parameters", but + not to be confused with "URL wildcards" as they are provided by the + :class:`Router`. ''' + get = self.environ['bottle.get'] = FormsDict() + pairs = _parse_qsl(self.environ.get('QUERY_STRING', '')) + for key, value in pairs: + get[key] = value + return get + + @DictProperty('environ', 'bottle.request.forms', read_only=True) + def forms(self): + """ Form values parsed from an `url-encoded` or `multipart/form-data` + encoded POST or PUT request body. The result is returned as a + :class:`FormsDict`. All keys and values are strings. File uploads + are stored separately in :attr:`files`. """ + forms = FormsDict() + for name, item in self.POST.allitems(): + if not isinstance(item, FileUpload): + forms[name] = item + return forms + + @DictProperty('environ', 'bottle.request.params', read_only=True) + def params(self): + """ A :class:`FormsDict` with the combined values of :attr:`query` and + :attr:`forms`. File uploads are stored in :attr:`files`. """ + params = FormsDict() + for key, value in self.query.allitems(): + params[key] = value + for key, value in self.forms.allitems(): + params[key] = value + return params + + @DictProperty('environ', 'bottle.request.files', read_only=True) + def files(self): + """ File uploads parsed from `multipart/form-data` encoded POST or PUT + request body. The values are instances of :class:`FileUpload`. + + """ + files = FormsDict() + for name, item in self.POST.allitems(): + if isinstance(item, FileUpload): + files[name] = item + return files + + @DictProperty('environ', 'bottle.request.json', read_only=True) + def json(self): + ''' If the ``Content-Type`` header is ``application/json``, this + property holds the parsed content of the request body. Only requests + smaller than :attr:`MEMFILE_MAX` are processed to avoid memory + exhaustion. ''' + ctype = self.environ.get('CONTENT_TYPE', '').lower().split(';')[0] + if ctype == 'application/json': + b = self._get_body_string() + if not b: + return None + return json_loads(b) + return None + + def _iter_body(self, read, bufsize): + maxread = max(0, self.content_length) + while maxread: + part = read(min(maxread, bufsize)) + if not part: break + yield part + maxread -= len(part) + + def _iter_chunked(self, read, bufsize): + err = HTTPError(400, 'Error while parsing chunked transfer body.') + rn, sem, bs = tob('\r\n'), tob(';'), tob('') + while True: + header = read(1) + while header[-2:] != rn: + c = read(1) + header += c + if not c: raise err + if len(header) > bufsize: raise err + size, _, _ = header.partition(sem) + try: + maxread = int(tonat(size.strip()), 16) + except ValueError: + raise err + if maxread == 0: break + buff = bs + while maxread > 0: + if not buff: + buff = read(min(maxread, bufsize)) + part, buff = buff[:maxread], buff[maxread:] + if not part: raise err + yield part + maxread -= len(part) + if read(2) != rn: + raise err + + @DictProperty('environ', 'bottle.request.body', read_only=True) + def _body(self): + body_iter = self._iter_chunked if self.chunked else self._iter_body + read_func = self.environ['wsgi.input'].read + body, body_size, is_temp_file = BytesIO(), 0, False + for part in body_iter(read_func, self.MEMFILE_MAX): + body.write(part) + body_size += len(part) + if not is_temp_file and body_size > self.MEMFILE_MAX: + body, tmp = TemporaryFile(mode='w+b'), body + body.write(tmp.getvalue()) + del tmp + is_temp_file = True + self.environ['wsgi.input'] = body + body.seek(0) + return body + + def _get_body_string(self): + ''' read body until content-length or MEMFILE_MAX into a string. Raise + HTTPError(413) on requests that are to large. ''' + clen = self.content_length + if clen > self.MEMFILE_MAX: + raise HTTPError(413, 'Request to large') + if clen < 0: clen = self.MEMFILE_MAX + 1 + data = self.body.read(clen) + if len(data) > self.MEMFILE_MAX: # Fail fast + raise HTTPError(413, 'Request to large') + return data + + @property + def body(self): + """ The HTTP request body as a seek-able file-like object. Depending on + :attr:`MEMFILE_MAX`, this is either a temporary file or a + :class:`io.BytesIO` instance. Accessing this property for the first + time reads and replaces the ``wsgi.input`` environ variable. + Subsequent accesses just do a `seek(0)` on the file object. """ + self._body.seek(0) + return self._body + + @property + def chunked(self): + ''' True if Chunked transfer encoding was. ''' + return 'chunked' in self.environ.get('HTTP_TRANSFER_ENCODING', '').lower() + + #: An alias for :attr:`query`. + GET = query + + @DictProperty('environ', 'bottle.request.post', read_only=True) + def POST(self): + """ The values of :attr:`forms` and :attr:`files` combined into a single + :class:`FormsDict`. Values are either strings (form values) or + instances of :class:`cgi.FieldStorage` (file uploads). + """ + post = FormsDict() + # We default to application/x-www-form-urlencoded for everything that + # is not multipart and take the fast path (also: 3.1 workaround) + if not self.content_type.startswith('multipart/'): + pairs = _parse_qsl(tonat(self._get_body_string(), 'latin1')) + for key, value in pairs: + post[key] = value + return post + + safe_env = {'QUERY_STRING':''} # Build a safe environment for cgi + for key in ('REQUEST_METHOD', 'CONTENT_TYPE', 'CONTENT_LENGTH'): + if key in self.environ: safe_env[key] = self.environ[key] + args = dict(fp=self.body, environ=safe_env, keep_blank_values=True) + if py31: + args['fp'] = NCTextIOWrapper(args['fp'], encoding='utf8', + newline='\n') + elif py3k: + args['encoding'] = 'utf8' + data = cgi.FieldStorage(**args) + self['_cgi.FieldStorage'] = data #http://bugs.python.org/issue18394#msg207958 + data = data.list or [] + for item in data: + if item.filename: + post[item.name] = FileUpload(item.file, item.name, + item.filename, item.headers) + else: + post[item.name] = item.value + return post + + @property + def url(self): + """ The full request URI including hostname and scheme. If your app + lives behind a reverse proxy or load balancer and you get confusing + results, make sure that the ``X-Forwarded-Host`` header is set + correctly. """ + return self.urlparts.geturl() + + @DictProperty('environ', 'bottle.request.urlparts', read_only=True) + def urlparts(self): + ''' The :attr:`url` string as an :class:`urlparse.SplitResult` tuple. + The tuple contains (scheme, host, path, query_string and fragment), + but the fragment is always empty because it is not visible to the + server. ''' + env = self.environ + http = env.get('HTTP_X_FORWARDED_PROTO') or env.get('wsgi.url_scheme', 'http') + host = env.get('HTTP_X_FORWARDED_HOST') or env.get('HTTP_HOST') + if not host: + # HTTP 1.1 requires a Host-header. This is for HTTP/1.0 clients. + host = env.get('SERVER_NAME', '127.0.0.1') + port = env.get('SERVER_PORT') + if port and port != ('80' if http == 'http' else '443'): + host += ':' + port + path = urlquote(self.fullpath) + return UrlSplitResult(http, host, path, env.get('QUERY_STRING'), '') + + @property + def fullpath(self): + """ Request path including :attr:`script_name` (if present). """ + return urljoin(self.script_name, self.path.lstrip('/')) + + @property + def query_string(self): + """ The raw :attr:`query` part of the URL (everything in between ``?`` + and ``#``) as a string. """ + return self.environ.get('QUERY_STRING', '') + + @property + def script_name(self): + ''' The initial portion of the URL's `path` that was removed by a higher + level (server or routing middleware) before the application was + called. This script path is returned with leading and tailing + slashes. ''' + script_name = self.environ.get('SCRIPT_NAME', '').strip('/') + return '/' + script_name + '/' if script_name else '/' + + def path_shift(self, shift=1): + ''' Shift path segments from :attr:`path` to :attr:`script_name` and + vice versa. + + :param shift: The number of path segments to shift. May be negative + to change the shift direction. (default: 1) + ''' + script = self.environ.get('SCRIPT_NAME','/') + self['SCRIPT_NAME'], self['PATH_INFO'] = path_shift(script, self.path, shift) + + @property + def content_length(self): + ''' The request body length as an integer. The client is responsible to + set this header. Otherwise, the real length of the body is unknown + and -1 is returned. In this case, :attr:`body` will be empty. ''' + return int(self.environ.get('CONTENT_LENGTH') or -1) + + @property + def content_type(self): + ''' The Content-Type header as a lowercase-string (default: empty). ''' + return self.environ.get('CONTENT_TYPE', '').lower() + + @property + def is_xhr(self): + ''' True if the request was triggered by a XMLHttpRequest. This only + works with JavaScript libraries that support the `X-Requested-With` + header (most of the popular libraries do). ''' + requested_with = self.environ.get('HTTP_X_REQUESTED_WITH','') + return requested_with.lower() == 'xmlhttprequest' + + @property + def is_ajax(self): + ''' Alias for :attr:`is_xhr`. "Ajax" is not the right term. ''' + return self.is_xhr + + @property + def auth(self): + """ HTTP authentication data as a (user, password) tuple. This + implementation currently supports basic (not digest) authentication + only. If the authentication happened at a higher level (e.g. in the + front web-server or a middleware), the password field is None, but + the user field is looked up from the ``REMOTE_USER`` environ + variable. On any errors, None is returned. """ + basic = parse_auth(self.environ.get('HTTP_AUTHORIZATION','')) + if basic: return basic + ruser = self.environ.get('REMOTE_USER') + if ruser: return (ruser, None) + return None + + @property + def remote_route(self): + """ A list of all IPs that were involved in this request, starting with + the client IP and followed by zero or more proxies. This does only + work if all proxies support the ```X-Forwarded-For`` header. Note + that this information can be forged by malicious clients. """ + proxy = self.environ.get('HTTP_X_FORWARDED_FOR') + if proxy: return [ip.strip() for ip in proxy.split(',')] + remote = self.environ.get('REMOTE_ADDR') + return [remote] if remote else [] + + @property + def remote_addr(self): + """ The client IP as a string. Note that this information can be forged + by malicious clients. """ + route = self.remote_route + return route[0] if route else None + + def copy(self): + """ Return a new :class:`Request` with a shallow :attr:`environ` copy. """ + return Request(self.environ.copy()) + + def get(self, value, default=None): return self.environ.get(value, default) + def __getitem__(self, key): return self.environ[key] + def __delitem__(self, key): self[key] = ""; del(self.environ[key]) + def __iter__(self): return iter(self.environ) + def __len__(self): return len(self.environ) + def keys(self): return self.environ.keys() + def __setitem__(self, key, value): + """ Change an environ value and clear all caches that depend on it. """ + + if self.environ.get('bottle.request.readonly'): + raise KeyError('The environ dictionary is read-only.') + + self.environ[key] = value + todelete = () + + if key == 'wsgi.input': + todelete = ('body', 'forms', 'files', 'params', 'post', 'json') + elif key == 'QUERY_STRING': + todelete = ('query', 'params') + elif key.startswith('HTTP_'): + todelete = ('headers', 'cookies') + + for key in todelete: + self.environ.pop('bottle.request.'+key, None) + + def __repr__(self): + return '<%s: %s %s>' % (self.__class__.__name__, self.method, self.url) + + def __getattr__(self, name): + ''' Search in self.environ for additional user defined attributes. ''' + try: + var = self.environ['bottle.request.ext.%s'%name] + return var.__get__(self) if hasattr(var, '__get__') else var + except KeyError: + raise AttributeError('Attribute %r not defined.' % name) + + def __setattr__(self, name, value): + if name == 'environ': return object.__setattr__(self, name, value) + self.environ['bottle.request.ext.%s'%name] = value + + +def _hkey(key): + if '\n' in key or '\r' in key or '\0' in key: + raise ValueError("Header names must not contain control characters: %r" % key) + return key.title().replace('_', '-') + + +def _hval(value): + value = tonat(value) + if '\n' in value or '\r' in value or '\0' in value: + raise ValueError("Header value must not contain control characters: %r" % value) + return value + + + +class HeaderProperty(object): + def __init__(self, name, reader=None, writer=None, default=''): + self.name, self.default = name, default + self.reader, self.writer = reader, writer + self.__doc__ = 'Current value of the %r header.' % name.title() + + def __get__(self, obj, cls): + if obj is None: return self + value = obj.get_header(self.name, self.default) + return self.reader(value) if self.reader else value + + def __set__(self, obj, value): + obj[self.name] = self.writer(value) if self.writer else value + + def __delete__(self, obj): + del obj[self.name] + + +class BaseResponse(object): + """ Storage class for a response body as well as headers and cookies. + + This class does support dict-like case-insensitive item-access to + headers, but is NOT a dict. Most notably, iterating over a response + yields parts of the body and not the headers. + + :param body: The response body as one of the supported types. + :param status: Either an HTTP status code (e.g. 200) or a status line + including the reason phrase (e.g. '200 OK'). + :param headers: A dictionary or a list of name-value pairs. + + Additional keyword arguments are added to the list of headers. + Underscores in the header name are replaced with dashes. + """ + + default_status = 200 + default_content_type = 'text/html; charset=UTF-8' + + # Header blacklist for specific response codes + # (rfc2616 section 10.2.3 and 10.3.5) + bad_headers = { + 204: set(('Content-Type',)), + 304: set(('Allow', 'Content-Encoding', 'Content-Language', + 'Content-Length', 'Content-Range', 'Content-Type', + 'Content-Md5', 'Last-Modified'))} + + def __init__(self, body='', status=None, headers=None, **more_headers): + self._cookies = None + self._headers = {} + self.body = body + self.status = status or self.default_status + if headers: + if isinstance(headers, dict): + headers = headers.items() + for name, value in headers: + self.add_header(name, value) + if more_headers: + for name, value in more_headers.items(): + self.add_header(name, value) + + def copy(self, cls=None): + ''' Returns a copy of self. ''' + cls = cls or BaseResponse + assert issubclass(cls, BaseResponse) + copy = cls() + copy.status = self.status + copy._headers = dict((k, v[:]) for (k, v) in self._headers.items()) + if self._cookies: + copy._cookies = SimpleCookie() + copy._cookies.load(self._cookies.output(header='')) + return copy + + def __iter__(self): + return iter(self.body) + + def close(self): + if hasattr(self.body, 'close'): + self.body.close() + + @property + def status_line(self): + ''' The HTTP status line as a string (e.g. ``404 Not Found``).''' + return self._status_line + + @property + def status_code(self): + ''' The HTTP status code as an integer (e.g. 404).''' + return self._status_code + + def _set_status(self, status): + if isinstance(status, int): + code, status = status, _HTTP_STATUS_LINES.get(status) + elif ' ' in status: + status = status.strip() + code = int(status.split()[0]) + else: + raise ValueError('String status line without a reason phrase.') + if not 100 <= code <= 999: raise ValueError('Status code out of range.') + self._status_code = code + self._status_line = str(status or ('%d Unknown' % code)) + + def _get_status(self): + return self._status_line + + status = property(_get_status, _set_status, None, + ''' A writeable property to change the HTTP response status. It accepts + either a numeric code (100-999) or a string with a custom reason + phrase (e.g. "404 Brain not found"). Both :data:`status_line` and + :data:`status_code` are updated accordingly. The return value is + always a status string. ''') + del _get_status, _set_status + + @property + def headers(self): + ''' An instance of :class:`HeaderDict`, a case-insensitive dict-like + view on the response headers. ''' + hdict = HeaderDict() + hdict.dict = self._headers + return hdict + + def __contains__(self, name): return _hkey(name) in self._headers + def __delitem__(self, name): del self._headers[_hkey(name)] + def __getitem__(self, name): return self._headers[_hkey(name)][-1] + def __setitem__(self, name, value): self._headers[_hkey(name)] = [_hval(value)] + + def get_header(self, name, default=None): + ''' Return the value of a previously defined header. If there is no + header with that name, return a default value. ''' + return self._headers.get(_hkey(name), [default])[-1] + + def set_header(self, name, value): + ''' Create a new response header, replacing any previously defined + headers with the same name. ''' + self._headers[_hkey(name)] = [_hval(value)] + + def add_header(self, name, value): + ''' Add an additional response header, not removing duplicates. ''' + self._headers.setdefault(_hkey(name), []).append(_hval(value)) + + def iter_headers(self): + ''' Yield (header, value) tuples, skipping headers that are not + allowed with the current response status code. ''' + return self.headerlist + + @property + def headerlist(self): + """ WSGI conform list of (header, value) tuples. """ + out = [] + headers = list(self._headers.items()) + if 'Content-Type' not in self._headers: + headers.append(('Content-Type', [self.default_content_type])) + if self._status_code in self.bad_headers: + bad_headers = self.bad_headers[self._status_code] + headers = [h for h in headers if h[0] not in bad_headers] + out += [(name, val) for (name, vals) in headers for val in vals] + if self._cookies: + for c in self._cookies.values(): + out.append(('Set-Cookie', _hval(c.OutputString()))) + if py3k: + out = [(k, v.encode('utf8').decode('latin1')) for (k, v) in out] + return out + + content_type = HeaderProperty('Content-Type') + content_length = HeaderProperty('Content-Length', reader=int) + expires = HeaderProperty('Expires', + reader=lambda x: datetime.utcfromtimestamp(parse_date(x)), + writer=lambda x: http_date(x)) + + @property + def charset(self, default='UTF-8'): + """ Return the charset specified in the content-type header (default: utf8). """ + if 'charset=' in self.content_type: + return self.content_type.split('charset=')[-1].split(';')[0].strip() + return default + + def set_cookie(self, name, value, secret=None, **options): + ''' Create a new cookie or replace an old one. If the `secret` parameter is + set, create a `Signed Cookie` (described below). + + :param name: the name of the cookie. + :param value: the value of the cookie. + :param secret: a signature key required for signed cookies. + + Additionally, this method accepts all RFC 2109 attributes that are + supported by :class:`cookie.Morsel`, including: + + :param max_age: maximum age in seconds. (default: None) + :param expires: a datetime object or UNIX timestamp. (default: None) + :param domain: the domain that is allowed to read the cookie. + (default: current domain) + :param path: limits the cookie to a given path (default: current path) + :param secure: limit the cookie to HTTPS connections (default: off). + :param httponly: prevents client-side javascript to read this cookie + (default: off, requires Python 2.6 or newer). + + If neither `expires` nor `max_age` is set (default), the cookie will + expire at the end of the browser session (as soon as the browser + window is closed). + + Signed cookies may store any pickle-able object and are + cryptographically signed to prevent manipulation. Keep in mind that + cookies are limited to 4kb in most browsers. + + Warning: Signed cookies are not encrypted (the client can still see + the content) and not copy-protected (the client can restore an old + cookie). The main intention is to make pickling and unpickling + save, not to store secret information at client side. + ''' + if not self._cookies: + self._cookies = SimpleCookie() + + if secret: + value = touni(cookie_encode((name, value), secret)) + elif not isinstance(value, basestring): + raise TypeError('Secret key missing for non-string Cookie.') + + if len(value) > 4096: raise ValueError('Cookie value to long.') + self._cookies[name] = value + + for key, value in options.items(): + if key == 'max_age': + if isinstance(value, timedelta): + value = value.seconds + value.days * 24 * 3600 + if key == 'expires': + if isinstance(value, (datedate, datetime)): + value = value.timetuple() + elif isinstance(value, (int, float)): + value = time.gmtime(value) + value = time.strftime("%a, %d %b %Y %H:%M:%S GMT", value) + self._cookies[name][key.replace('_', '-')] = value + + def delete_cookie(self, key, **kwargs): + ''' Delete a cookie. Be sure to use the same `domain` and `path` + settings as used to create the cookie. ''' + kwargs['max_age'] = -1 + kwargs['expires'] = 0 + self.set_cookie(key, '', **kwargs) + + def __repr__(self): + out = '' + for name, value in self.headerlist: + out += '%s: %s\n' % (name.title(), value.strip()) + return out + + +def local_property(name=None): + if name: depr('local_property() is deprecated and will be removed.') #0.12 + ls = threading.local() + def fget(self): + try: return ls.var + except AttributeError: + raise RuntimeError("Request context not initialized.") + def fset(self, value): ls.var = value + def fdel(self): del ls.var + return property(fget, fset, fdel, 'Thread-local property') + + +class LocalRequest(BaseRequest): + ''' A thread-local subclass of :class:`BaseRequest` with a different + set of attributes for each thread. There is usually only one global + instance of this class (:data:`request`). If accessed during a + request/response cycle, this instance always refers to the *current* + request (even on a multithreaded server). ''' + bind = BaseRequest.__init__ + environ = local_property() + + +class LocalResponse(BaseResponse): + ''' A thread-local subclass of :class:`BaseResponse` with a different + set of attributes for each thread. There is usually only one global + instance of this class (:data:`response`). Its attributes are used + to build the HTTP response at the end of the request/response cycle. + ''' + bind = BaseResponse.__init__ + _status_line = local_property() + _status_code = local_property() + _cookies = local_property() + _headers = local_property() + body = local_property() + + +Request = BaseRequest +Response = BaseResponse + + +class HTTPResponse(Response, BottleException): + def __init__(self, body='', status=None, headers=None, **more_headers): + super(HTTPResponse, self).__init__(body, status, headers, **more_headers) + + def apply(self, response): + response._status_code = self._status_code + response._status_line = self._status_line + response._headers = self._headers + response._cookies = self._cookies + response.body = self.body + + +class HTTPError(HTTPResponse): + default_status = 500 + def __init__(self, status=None, body=None, exception=None, traceback=None, + **options): + self.exception = exception + self.traceback = traceback + super(HTTPError, self).__init__(body, status, **options) + + + + + +############################################################################### +# Plugins ###################################################################### +############################################################################### + +class PluginError(BottleException): pass + + +class JSONPlugin(object): + name = 'json' + api = 2 + + def __init__(self, json_dumps=json_dumps): + self.json_dumps = json_dumps + + def apply(self, callback, route): + dumps = self.json_dumps + if not dumps: return callback + def wrapper(*a, **ka): + try: + rv = callback(*a, **ka) + except HTTPError: + rv = _e() + + if isinstance(rv, dict): + #Attempt to serialize, raises exception on failure + json_response = dumps(rv) + #Set content type only if serialization succesful + response.content_type = 'application/json' + return json_response + elif isinstance(rv, HTTPResponse) and isinstance(rv.body, dict): + rv.body = dumps(rv.body) + rv.content_type = 'application/json' + return rv + + return wrapper + + +class TemplatePlugin(object): + ''' This plugin applies the :func:`view` decorator to all routes with a + `template` config parameter. If the parameter is a tuple, the second + element must be a dict with additional options (e.g. `template_engine`) + or default variables for the template. ''' + name = 'template' + api = 2 + + def apply(self, callback, route): + conf = route.config.get('template') + if isinstance(conf, (tuple, list)) and len(conf) == 2: + return view(conf[0], **conf[1])(callback) + elif isinstance(conf, str): + return view(conf)(callback) + else: + return callback + + +#: Not a plugin, but part of the plugin API. TODO: Find a better place. +class _ImportRedirect(object): + def __init__(self, name, impmask): + ''' Create a virtual package that redirects imports (see PEP 302). ''' + self.name = name + self.impmask = impmask + self.module = sys.modules.setdefault(name, new_module(name)) + self.module.__dict__.update({'__file__': __file__, '__path__': [], + '__all__': [], '__loader__': self}) + sys.meta_path.append(self) + + def find_module(self, fullname, path=None): + if '.' not in fullname: return + packname = fullname.rsplit('.', 1)[0] + if packname != self.name: return + return self + + def load_module(self, fullname): + if fullname in sys.modules: return sys.modules[fullname] + modname = fullname.rsplit('.', 1)[1] + realname = self.impmask % modname + __import__(realname) + module = sys.modules[fullname] = sys.modules[realname] + setattr(self.module, modname, module) + module.__loader__ = self + return module + + + + + + +############################################################################### +# Common Utilities ############################################################# +############################################################################### + + +class MultiDict(DictMixin): + """ This dict stores multiple values per key, but behaves exactly like a + normal dict in that it returns only the newest value for any given key. + There are special methods available to access the full list of values. + """ + + def __init__(self, *a, **k): + self.dict = dict((k, [v]) for (k, v) in dict(*a, **k).items()) + + def __len__(self): return len(self.dict) + def __iter__(self): return iter(self.dict) + def __contains__(self, key): return key in self.dict + def __delitem__(self, key): del self.dict[key] + def __getitem__(self, key): return self.dict[key][-1] + def __setitem__(self, key, value): self.append(key, value) + def keys(self): return self.dict.keys() + + if py3k: + def values(self): return (v[-1] for v in self.dict.values()) + def items(self): return ((k, v[-1]) for k, v in self.dict.items()) + def allitems(self): + return ((k, v) for k, vl in self.dict.items() for v in vl) + iterkeys = keys + itervalues = values + iteritems = items + iterallitems = allitems + + else: + def values(self): return [v[-1] for v in self.dict.values()] + def items(self): return [(k, v[-1]) for k, v in self.dict.items()] + def iterkeys(self): return self.dict.iterkeys() + def itervalues(self): return (v[-1] for v in self.dict.itervalues()) + def iteritems(self): + return ((k, v[-1]) for k, v in self.dict.iteritems()) + def iterallitems(self): + return ((k, v) for k, vl in self.dict.iteritems() for v in vl) + def allitems(self): + return [(k, v) for k, vl in self.dict.iteritems() for v in vl] + + def get(self, key, default=None, index=-1, type=None): + ''' Return the most recent value for a key. + + :param default: The default value to be returned if the key is not + present or the type conversion fails. + :param index: An index for the list of available values. + :param type: If defined, this callable is used to cast the value + into a specific type. Exception are suppressed and result in + the default value to be returned. + ''' + try: + val = self.dict[key][index] + return type(val) if type else val + except Exception: + pass + return default + + def append(self, key, value): + ''' Add a new value to the list of values for this key. ''' + self.dict.setdefault(key, []).append(value) + + def replace(self, key, value): + ''' Replace the list of values with a single value. ''' + self.dict[key] = [value] + + def getall(self, key): + ''' Return a (possibly empty) list of values for a key. ''' + return self.dict.get(key) or [] + + #: Aliases for WTForms to mimic other multi-dict APIs (Django) + getone = get + getlist = getall + + +class FormsDict(MultiDict): + ''' This :class:`MultiDict` subclass is used to store request form data. + Additionally to the normal dict-like item access methods (which return + unmodified data as native strings), this container also supports + attribute-like access to its values. Attributes are automatically de- + or recoded to match :attr:`input_encoding` (default: 'utf8'). Missing + attributes default to an empty string. ''' + + #: Encoding used for attribute values. + input_encoding = 'utf8' + #: If true (default), unicode strings are first encoded with `latin1` + #: and then decoded to match :attr:`input_encoding`. + recode_unicode = True + + def _fix(self, s, encoding=None): + if isinstance(s, unicode) and self.recode_unicode: # Python 3 WSGI + return s.encode('latin1').decode(encoding or self.input_encoding) + elif isinstance(s, bytes): # Python 2 WSGI + return s.decode(encoding or self.input_encoding) + else: + return s + + def decode(self, encoding=None): + ''' Returns a copy with all keys and values de- or recoded to match + :attr:`input_encoding`. Some libraries (e.g. WTForms) want a + unicode dictionary. ''' + copy = FormsDict() + enc = copy.input_encoding = encoding or self.input_encoding + copy.recode_unicode = False + for key, value in self.allitems(): + copy.append(self._fix(key, enc), self._fix(value, enc)) + return copy + + def getunicode(self, name, default=None, encoding=None): + ''' Return the value as a unicode string, or the default. ''' + try: + return self._fix(self[name], encoding) + except (UnicodeError, KeyError): + return default + + def __getattr__(self, name, default=unicode()): + # Without this guard, pickle generates a cryptic TypeError: + if name.startswith('__') and name.endswith('__'): + return super(FormsDict, self).__getattr__(name) + return self.getunicode(name, default=default) + +class HeaderDict(MultiDict): + """ A case-insensitive version of :class:`MultiDict` that defaults to + replace the old value instead of appending it. """ + + def __init__(self, *a, **ka): + self.dict = {} + if a or ka: self.update(*a, **ka) + + def __contains__(self, key): return _hkey(key) in self.dict + def __delitem__(self, key): del self.dict[_hkey(key)] + def __getitem__(self, key): return self.dict[_hkey(key)][-1] + def __setitem__(self, key, value): self.dict[_hkey(key)] = [_hval(value)] + def append(self, key, value): self.dict.setdefault(_hkey(key), []).append(_hval(value)) + def replace(self, key, value): self.dict[_hkey(key)] = [_hval(value)] + def getall(self, key): return self.dict.get(_hkey(key)) or [] + def get(self, key, default=None, index=-1): + return MultiDict.get(self, _hkey(key), default, index) + def filter(self, names): + for name in (_hkey(n) for n in names): + if name in self.dict: + del self.dict[name] + + +class WSGIHeaderDict(DictMixin): + ''' This dict-like class wraps a WSGI environ dict and provides convenient + access to HTTP_* fields. Keys and values are native strings + (2.x bytes or 3.x unicode) and keys are case-insensitive. If the WSGI + environment contains non-native string values, these are de- or encoded + using a lossless 'latin1' character set. + + The API will remain stable even on changes to the relevant PEPs. + Currently PEP 333, 444 and 3333 are supported. (PEP 444 is the only one + that uses non-native strings.) + ''' + #: List of keys that do not have a ``HTTP_`` prefix. + cgikeys = ('CONTENT_TYPE', 'CONTENT_LENGTH') + + def __init__(self, environ): + self.environ = environ + + def _ekey(self, key): + ''' Translate header field name to CGI/WSGI environ key. ''' + key = key.replace('-','_').upper() + if key in self.cgikeys: + return key + return 'HTTP_' + key + + def raw(self, key, default=None): + ''' Return the header value as is (may be bytes or unicode). ''' + return self.environ.get(self._ekey(key), default) + + def __getitem__(self, key): + return tonat(self.environ[self._ekey(key)], 'latin1') + + def __setitem__(self, key, value): + raise TypeError("%s is read-only." % self.__class__) + + def __delitem__(self, key): + raise TypeError("%s is read-only." % self.__class__) + + def __iter__(self): + for key in self.environ: + if key[:5] == 'HTTP_': + yield key[5:].replace('_', '-').title() + elif key in self.cgikeys: + yield key.replace('_', '-').title() + + def keys(self): return [x for x in self] + def __len__(self): return len(self.keys()) + def __contains__(self, key): return self._ekey(key) in self.environ + + + +class ConfigDict(dict): + ''' A dict-like configuration storage with additional support for + namespaces, validators, meta-data, on_change listeners and more. + + This storage is optimized for fast read access. Retrieving a key + or using non-altering dict methods (e.g. `dict.get()`) has no overhead + compared to a native dict. + ''' + __slots__ = ('_meta', '_on_change') + + class Namespace(DictMixin): + + def __init__(self, config, namespace): + self._config = config + self._prefix = namespace + + def __getitem__(self, key): + depr('Accessing namespaces as dicts is discouraged. ' + 'Only use flat item access: ' + 'cfg["names"]["pace"]["key"] -> cfg["name.space.key"]') #0.12 + return self._config[self._prefix + '.' + key] + + def __setitem__(self, key, value): + self._config[self._prefix + '.' + key] = value + + def __delitem__(self, key): + del self._config[self._prefix + '.' + key] + + def __iter__(self): + ns_prefix = self._prefix + '.' + for key in self._config: + ns, dot, name = key.rpartition('.') + if ns == self._prefix and name: + yield name + + def keys(self): return [x for x in self] + def __len__(self): return len(self.keys()) + def __contains__(self, key): return self._prefix + '.' + key in self._config + def __repr__(self): return '' % self._prefix + def __str__(self): return '' % self._prefix + + # Deprecated ConfigDict features + def __getattr__(self, key): + depr('Attribute access is deprecated.') #0.12 + if key not in self and key[0].isupper(): + self[key] = ConfigDict.Namespace(self._config, self._prefix + '.' + key) + if key not in self and key.startswith('__'): + raise AttributeError(key) + return self.get(key) + + def __setattr__(self, key, value): + if key in ('_config', '_prefix'): + self.__dict__[key] = value + return + depr('Attribute assignment is deprecated.') #0.12 + if hasattr(DictMixin, key): + raise AttributeError('Read-only attribute.') + if key in self and self[key] and isinstance(self[key], self.__class__): + raise AttributeError('Non-empty namespace attribute.') + self[key] = value + + def __delattr__(self, key): + if key in self: + val = self.pop(key) + if isinstance(val, self.__class__): + prefix = key + '.' + for key in self: + if key.startswith(prefix): + del self[prefix+key] + + def __call__(self, *a, **ka): + depr('Calling ConfDict is deprecated. Use the update() method.') #0.12 + self.update(*a, **ka) + return self + + def __init__(self, *a, **ka): + self._meta = {} + self._on_change = lambda name, value: None + if a or ka: + depr('Constructor does no longer accept parameters.') #0.12 + self.update(*a, **ka) + + def load_config(self, filename): + ''' Load values from an *.ini style config file. + + If the config file contains sections, their names are used as + namespaces for the values within. The two special sections + ``DEFAULT`` and ``bottle`` refer to the root namespace (no prefix). + ''' + conf = ConfigParser() + conf.read(filename) + for section in conf.sections(): + for key, value in conf.items(section): + if section not in ('DEFAULT', 'bottle'): + key = section + '.' + key + self[key] = value + return self + + def load_dict(self, source, namespace='', make_namespaces=False): + ''' Import values from a dictionary structure. Nesting can be used to + represent namespaces. + + >>> ConfigDict().load_dict({'name': {'space': {'key': 'value'}}}) + {'name.space.key': 'value'} + ''' + stack = [(namespace, source)] + while stack: + prefix, source = stack.pop() + if not isinstance(source, dict): + raise TypeError('Source is not a dict (r)' % type(key)) + for key, value in source.items(): + if not isinstance(key, basestring): + raise TypeError('Key is not a string (%r)' % type(key)) + full_key = prefix + '.' + key if prefix else key + if isinstance(value, dict): + stack.append((full_key, value)) + if make_namespaces: + self[full_key] = self.Namespace(self, full_key) + else: + self[full_key] = value + return self + + def update(self, *a, **ka): + ''' If the first parameter is a string, all keys are prefixed with this + namespace. Apart from that it works just as the usual dict.update(). + Example: ``update('some.namespace', key='value')`` ''' + prefix = '' + if a and isinstance(a[0], basestring): + prefix = a[0].strip('.') + '.' + a = a[1:] + for key, value in dict(*a, **ka).items(): + self[prefix+key] = value + + def setdefault(self, key, value): + if key not in self: + self[key] = value + return self[key] + + def __setitem__(self, key, value): + if not isinstance(key, basestring): + raise TypeError('Key has type %r (not a string)' % type(key)) + + value = self.meta_get(key, 'filter', lambda x: x)(value) + if key in self and self[key] is value: + return + self._on_change(key, value) + dict.__setitem__(self, key, value) + + def __delitem__(self, key): + dict.__delitem__(self, key) + + def clear(self): + for key in self: + del self[key] + + def meta_get(self, key, metafield, default=None): + ''' Return the value of a meta field for a key. ''' + return self._meta.get(key, {}).get(metafield, default) + + def meta_set(self, key, metafield, value): + ''' Set the meta field for a key to a new value. This triggers the + on-change handler for existing keys. ''' + self._meta.setdefault(key, {})[metafield] = value + if key in self: + self[key] = self[key] + + def meta_list(self, key): + ''' Return an iterable of meta field names defined for a key. ''' + return self._meta.get(key, {}).keys() + + # Deprecated ConfigDict features + def __getattr__(self, key): + depr('Attribute access is deprecated.') #0.12 + if key not in self and key[0].isupper(): + self[key] = self.Namespace(self, key) + if key not in self and key.startswith('__'): + raise AttributeError(key) + return self.get(key) + + def __setattr__(self, key, value): + if key in self.__slots__: + return dict.__setattr__(self, key, value) + depr('Attribute assignment is deprecated.') #0.12 + if hasattr(dict, key): + raise AttributeError('Read-only attribute.') + if key in self and self[key] and isinstance(self[key], self.Namespace): + raise AttributeError('Non-empty namespace attribute.') + self[key] = value + + def __delattr__(self, key): + if key in self: + val = self.pop(key) + if isinstance(val, self.Namespace): + prefix = key + '.' + for key in self: + if key.startswith(prefix): + del self[prefix+key] + + def __call__(self, *a, **ka): + depr('Calling ConfDict is deprecated. Use the update() method.') #0.12 + self.update(*a, **ka) + return self + + + +class AppStack(list): + """ A stack-like list. Calling it returns the head of the stack. """ + + def __call__(self): + """ Return the current default application. """ + return self[-1] + + def push(self, value=None): + """ Add a new :class:`Bottle` instance to the stack """ + if not isinstance(value, Bottle): + value = Bottle() + self.append(value) + return value + + +class WSGIFileWrapper(object): + + def __init__(self, fp, buffer_size=1024*64): + self.fp, self.buffer_size = fp, buffer_size + for attr in ('fileno', 'close', 'read', 'readlines', 'tell', 'seek'): + if hasattr(fp, attr): setattr(self, attr, getattr(fp, attr)) + + def __iter__(self): + buff, read = self.buffer_size, self.read + while True: + part = read(buff) + if not part: return + yield part + + +class _closeiter(object): + ''' This only exists to be able to attach a .close method to iterators that + do not support attribute assignment (most of itertools). ''' + + def __init__(self, iterator, close=None): + self.iterator = iterator + self.close_callbacks = makelist(close) + + def __iter__(self): + return iter(self.iterator) + + def close(self): + for func in self.close_callbacks: + func() + + +class ResourceManager(object): + ''' This class manages a list of search paths and helps to find and open + application-bound resources (files). + + :param base: default value for :meth:`add_path` calls. + :param opener: callable used to open resources. + :param cachemode: controls which lookups are cached. One of 'all', + 'found' or 'none'. + ''' + + def __init__(self, base='./', opener=open, cachemode='all'): + self.opener = open + self.base = base + self.cachemode = cachemode + + #: A list of search paths. See :meth:`add_path` for details. + self.path = [] + #: A cache for resolved paths. ``res.cache.clear()`` clears the cache. + self.cache = {} + + def add_path(self, path, base=None, index=None, create=False): + ''' Add a new path to the list of search paths. Return False if the + path does not exist. + + :param path: The new search path. Relative paths are turned into + an absolute and normalized form. If the path looks like a file + (not ending in `/`), the filename is stripped off. + :param base: Path used to absolutize relative search paths. + Defaults to :attr:`base` which defaults to ``os.getcwd()``. + :param index: Position within the list of search paths. Defaults + to last index (appends to the list). + + The `base` parameter makes it easy to reference files installed + along with a python module or package:: + + res.add_path('./resources/', __file__) + ''' + base = os.path.abspath(os.path.dirname(base or self.base)) + path = os.path.abspath(os.path.join(base, os.path.dirname(path))) + path += os.sep + if path in self.path: + self.path.remove(path) + if create and not os.path.isdir(path): + os.makedirs(path) + if index is None: + self.path.append(path) + else: + self.path.insert(index, path) + self.cache.clear() + return os.path.exists(path) + + def __iter__(self): + ''' Iterate over all existing files in all registered paths. ''' + search = self.path[:] + while search: + path = search.pop() + if not os.path.isdir(path): continue + for name in os.listdir(path): + full = os.path.join(path, name) + if os.path.isdir(full): search.append(full) + else: yield full + + def lookup(self, name): + ''' Search for a resource and return an absolute file path, or `None`. + + The :attr:`path` list is searched in order. The first match is + returend. Symlinks are followed. The result is cached to speed up + future lookups. ''' + if name not in self.cache or DEBUG: + for path in self.path: + fpath = os.path.join(path, name) + if os.path.isfile(fpath): + if self.cachemode in ('all', 'found'): + self.cache[name] = fpath + return fpath + if self.cachemode == 'all': + self.cache[name] = None + return self.cache[name] + + def open(self, name, mode='r', *args, **kwargs): + ''' Find a resource and return a file object, or raise IOError. ''' + fname = self.lookup(name) + if not fname: raise IOError("Resource %r not found." % name) + return self.opener(fname, mode=mode, *args, **kwargs) + + +class FileUpload(object): + + def __init__(self, fileobj, name, filename, headers=None): + ''' Wrapper for file uploads. ''' + #: Open file(-like) object (BytesIO buffer or temporary file) + self.file = fileobj + #: Name of the upload form field + self.name = name + #: Raw filename as sent by the client (may contain unsafe characters) + self.raw_filename = filename + #: A :class:`HeaderDict` with additional headers (e.g. content-type) + self.headers = HeaderDict(headers) if headers else HeaderDict() + + content_type = HeaderProperty('Content-Type') + content_length = HeaderProperty('Content-Length', reader=int, default=-1) + + def get_header(self, name, default=None): + """ Return the value of a header within the mulripart part. """ + return self.headers.get(name, default) + + @cached_property + def filename(self): + ''' Name of the file on the client file system, but normalized to ensure + file system compatibility. An empty filename is returned as 'empty'. + + Only ASCII letters, digits, dashes, underscores and dots are + allowed in the final filename. Accents are removed, if possible. + Whitespace is replaced by a single dash. Leading or tailing dots + or dashes are removed. The filename is limited to 255 characters. + ''' + fname = self.raw_filename + if not isinstance(fname, unicode): + fname = fname.decode('utf8', 'ignore') + fname = normalize('NFKD', fname).encode('ASCII', 'ignore').decode('ASCII') + fname = os.path.basename(fname.replace('\\', os.path.sep)) + fname = re.sub(r'[^a-zA-Z0-9-_.\s]', '', fname).strip() + fname = re.sub(r'[-\s]+', '-', fname).strip('.-') + return fname[:255] or 'empty' + + def _copy_file(self, fp, chunk_size=2**16): + read, write, offset = self.file.read, fp.write, self.file.tell() + while 1: + buf = read(chunk_size) + if not buf: break + write(buf) + self.file.seek(offset) + + def save(self, destination, overwrite=False, chunk_size=2**16): + ''' Save file to disk or copy its content to an open file(-like) object. + If *destination* is a directory, :attr:`filename` is added to the + path. Existing files are not overwritten by default (IOError). + + :param destination: File path, directory or file(-like) object. + :param overwrite: If True, replace existing files. (default: False) + :param chunk_size: Bytes to read at a time. (default: 64kb) + ''' + if isinstance(destination, basestring): # Except file-likes here + if os.path.isdir(destination): + destination = os.path.join(destination, self.filename) + if not overwrite and os.path.exists(destination): + raise IOError('File exists.') + with open(destination, 'wb') as fp: + self._copy_file(fp, chunk_size) + else: + self._copy_file(destination, chunk_size) + + + + + + +############################################################################### +# Application Helper ########################################################### +############################################################################### + + +def abort(code=500, text='Unknown Error.'): + """ Aborts execution and causes a HTTP error. """ + raise HTTPError(code, text) + + +def redirect(url, code=None): + """ Aborts execution and causes a 303 or 302 redirect, depending on + the HTTP protocol version. """ + if not code: + code = 303 if request.get('SERVER_PROTOCOL') == "HTTP/1.1" else 302 + res = response.copy(cls=HTTPResponse) + res.status = code + res.body = "" + res.set_header('Location', urljoin(request.url, url)) + raise res + + +def _file_iter_range(fp, offset, bytes, maxread=1024*1024): + ''' Yield chunks from a range in a file. No chunk is bigger than maxread.''' + fp.seek(offset) + while bytes > 0: + part = fp.read(min(bytes, maxread)) + if not part: break + bytes -= len(part) + yield part + + +def static_file(filename, root, mimetype='auto', download=False, charset='UTF-8'): + """ Open a file in a safe way and return :exc:`HTTPResponse` with status + code 200, 305, 403 or 404. The ``Content-Type``, ``Content-Encoding``, + ``Content-Length`` and ``Last-Modified`` headers are set if possible. + Special support for ``If-Modified-Since``, ``Range`` and ``HEAD`` + requests. + + :param filename: Name or path of the file to send. + :param root: Root path for file lookups. Should be an absolute directory + path. + :param mimetype: Defines the content-type header (default: guess from + file extension) + :param download: If True, ask the browser to open a `Save as...` dialog + instead of opening the file with the associated program. You can + specify a custom filename as a string. If not specified, the + original filename is used (default: False). + :param charset: The charset to use for files with a ``text/*`` + mime-type. (default: UTF-8) + """ + + root = os.path.abspath(root) + os.sep + filename = os.path.abspath(os.path.join(root, filename.strip('/\\'))) + headers = dict() + + if not filename.startswith(root): + return HTTPError(403, "Access denied.") + if not os.path.exists(filename) or not os.path.isfile(filename): + return HTTPError(404, "File does not exist.") + if not os.access(filename, os.R_OK): + return HTTPError(403, "You do not have permission to access this file.") + + if mimetype == 'auto': + mimetype, encoding = mimetypes.guess_type(filename) + if encoding: headers['Content-Encoding'] = encoding + + if mimetype: + if mimetype[:5] == 'text/' and charset and 'charset' not in mimetype: + mimetype += '; charset=%s' % charset + headers['Content-Type'] = mimetype + + if download: + download = os.path.basename(filename if download == True else download) + headers['Content-Disposition'] = 'attachment; filename="%s"' % download + + stats = os.stat(filename) + headers['Content-Length'] = clen = stats.st_size + lm = time.strftime("%a, %d %b %Y %H:%M:%S GMT", time.gmtime(stats.st_mtime)) + headers['Last-Modified'] = lm + + ims = request.environ.get('HTTP_IF_MODIFIED_SINCE') + if ims: + ims = parse_date(ims.split(";")[0].strip()) + if ims is not None and ims >= int(stats.st_mtime): + headers['Date'] = time.strftime("%a, %d %b %Y %H:%M:%S GMT", time.gmtime()) + return HTTPResponse(status=304, **headers) + + body = '' if request.method == 'HEAD' else open(filename, 'rb') + + headers["Accept-Ranges"] = "bytes" + ranges = request.environ.get('HTTP_RANGE') + if 'HTTP_RANGE' in request.environ: + ranges = list(parse_range_header(request.environ['HTTP_RANGE'], clen)) + if not ranges: + return HTTPError(416, "Requested Range Not Satisfiable") + offset, end = ranges[0] + headers["Content-Range"] = "bytes %d-%d/%d" % (offset, end-1, clen) + headers["Content-Length"] = str(end-offset) + if body: body = _file_iter_range(body, offset, end-offset) + return HTTPResponse(body, status=206, **headers) + return HTTPResponse(body, **headers) + + + + + + +############################################################################### +# HTTP Utilities and MISC (TODO) ############################################### +############################################################################### + + +def debug(mode=True): + """ Change the debug level. + There is only one debug level supported at the moment.""" + global DEBUG + if mode: warnings.simplefilter('default') + DEBUG = bool(mode) + +def http_date(value): + if isinstance(value, (datedate, datetime)): + value = value.utctimetuple() + elif isinstance(value, (int, float)): + value = time.gmtime(value) + if not isinstance(value, basestring): + value = time.strftime("%a, %d %b %Y %H:%M:%S GMT", value) + return value + +def parse_date(ims): + """ Parse rfc1123, rfc850 and asctime timestamps and return UTC epoch. """ + try: + ts = email.utils.parsedate_tz(ims) + return time.mktime(ts[:8] + (0,)) - (ts[9] or 0) - time.timezone + except (TypeError, ValueError, IndexError, OverflowError): + return None + +def parse_auth(header): + """ Parse rfc2617 HTTP authentication header string (basic) and return (user,pass) tuple or None""" + try: + method, data = header.split(None, 1) + if method.lower() == 'basic': + user, pwd = touni(base64.b64decode(tob(data))).split(':',1) + return user, pwd + except (KeyError, ValueError): + return None + +def parse_range_header(header, maxlen=0): + ''' Yield (start, end) ranges parsed from a HTTP Range header. Skip + unsatisfiable ranges. The end index is non-inclusive.''' + if not header or header[:6] != 'bytes=': return + ranges = [r.split('-', 1) for r in header[6:].split(',') if '-' in r] + for start, end in ranges: + try: + if not start: # bytes=-100 -> last 100 bytes + start, end = max(0, maxlen-int(end)), maxlen + elif not end: # bytes=100- -> all but the first 99 bytes + start, end = int(start), maxlen + else: # bytes=100-200 -> bytes 100-200 (inclusive) + start, end = int(start), min(int(end)+1, maxlen) + if 0 <= start < end <= maxlen: + yield start, end + except ValueError: + pass + +def _parse_qsl(qs): + r = [] + for pair in qs.replace(';','&').split('&'): + if not pair: continue + nv = pair.split('=', 1) + if len(nv) != 2: nv.append('') + key = urlunquote(nv[0].replace('+', ' ')) + value = urlunquote(nv[1].replace('+', ' ')) + r.append((key, value)) + return r + +def _lscmp(a, b): + ''' Compares two strings in a cryptographically safe way: + Runtime is not affected by length of common prefix. ''' + return not sum(0 if x==y else 1 for x, y in zip(a, b)) and len(a) == len(b) + + +def cookie_encode(data, key): + ''' Encode and sign a pickle-able object. Return a (byte) string ''' + msg = base64.b64encode(pickle.dumps(data, -1)) + sig = base64.b64encode(hmac.new(tob(key), msg, digestmod=hashlib.md5).digest()) + return tob('!') + sig + tob('?') + msg + + +def cookie_decode(data, key): + ''' Verify and decode an encoded string. Return an object or None.''' + data = tob(data) + if cookie_is_encoded(data): + sig, msg = data.split(tob('?'), 1) + if _lscmp(sig[1:], base64.b64encode(hmac.new(tob(key), msg, digestmod=hashlib.md5).digest())): + return pickle.loads(base64.b64decode(msg)) + return None + + +def cookie_is_encoded(data): + ''' Return True if the argument looks like a encoded cookie.''' + return bool(data.startswith(tob('!')) and tob('?') in data) + + +def html_escape(string): + ''' Escape HTML special characters ``&<>`` and quotes ``'"``. ''' + return string.replace('&','&').replace('<','<').replace('>','>')\ + .replace('"','"').replace("'",''') + + +def html_quote(string): + ''' Escape and quote a string to be used as an HTTP attribute.''' + return '"%s"' % html_escape(string).replace('\n',' ')\ + .replace('\r',' ').replace('\t',' ') + + +def yieldroutes(func): + """ Return a generator for routes that match the signature (name, args) + of the func parameter. This may yield more than one route if the function + takes optional keyword arguments. The output is best described by example:: + + a() -> '/a' + b(x, y) -> '/b//' + c(x, y=5) -> '/c/' and '/c//' + d(x=5, y=6) -> '/d' and '/d/' and '/d//' + """ + path = '/' + func.__name__.replace('__','/').lstrip('/') + spec = getargspec(func) + argc = len(spec[0]) - len(spec[3] or []) + path += ('/<%s>' * argc) % tuple(spec[0][:argc]) + yield path + for arg in spec[0][argc:]: + path += '/<%s>' % arg + yield path + + +def path_shift(script_name, path_info, shift=1): + ''' Shift path fragments from PATH_INFO to SCRIPT_NAME and vice versa. + + :return: The modified paths. + :param script_name: The SCRIPT_NAME path. + :param script_name: The PATH_INFO path. + :param shift: The number of path fragments to shift. May be negative to + change the shift direction. (default: 1) + ''' + if shift == 0: return script_name, path_info + pathlist = path_info.strip('/').split('/') + scriptlist = script_name.strip('/').split('/') + if pathlist and pathlist[0] == '': pathlist = [] + if scriptlist and scriptlist[0] == '': scriptlist = [] + if shift > 0 and shift <= len(pathlist): + moved = pathlist[:shift] + scriptlist = scriptlist + moved + pathlist = pathlist[shift:] + elif shift < 0 and shift >= -len(scriptlist): + moved = scriptlist[shift:] + pathlist = moved + pathlist + scriptlist = scriptlist[:shift] + else: + empty = 'SCRIPT_NAME' if shift < 0 else 'PATH_INFO' + raise AssertionError("Cannot shift. Nothing left from %s" % empty) + new_script_name = '/' + '/'.join(scriptlist) + new_path_info = '/' + '/'.join(pathlist) + if path_info.endswith('/') and pathlist: new_path_info += '/' + return new_script_name, new_path_info + + +def auth_basic(check, realm="private", text="Access denied"): + ''' Callback decorator to require HTTP auth (basic). + TODO: Add route(check_auth=...) parameter. ''' + def decorator(func): + def wrapper(*a, **ka): + user, password = request.auth or (None, None) + if user is None or not check(user, password): + err = HTTPError(401, text) + err.add_header('WWW-Authenticate', 'Basic realm="%s"' % realm) + return err + return func(*a, **ka) + return wrapper + return decorator + + +# Shortcuts for common Bottle methods. +# They all refer to the current default application. + +def make_default_app_wrapper(name): + ''' Return a callable that relays calls to the current default app. ''' + @functools.wraps(getattr(Bottle, name)) + def wrapper(*a, **ka): + return getattr(app(), name)(*a, **ka) + return wrapper + +route = make_default_app_wrapper('route') +get = make_default_app_wrapper('get') +post = make_default_app_wrapper('post') +put = make_default_app_wrapper('put') +delete = make_default_app_wrapper('delete') +error = make_default_app_wrapper('error') +mount = make_default_app_wrapper('mount') +hook = make_default_app_wrapper('hook') +install = make_default_app_wrapper('install') +uninstall = make_default_app_wrapper('uninstall') +url = make_default_app_wrapper('get_url') + + + + + + + +############################################################################### +# Server Adapter ############################################################### +############################################################################### + + +class ServerAdapter(object): + quiet = False + def __init__(self, host='127.0.0.1', port=8080, **options): + self.options = options + self.host = host + self.port = int(port) + + def run(self, handler): # pragma: no cover + pass + + def __repr__(self): + args = ', '.join(['%s=%s'%(k,repr(v)) for k, v in self.options.items()]) + return "%s(%s)" % (self.__class__.__name__, args) + + +class CGIServer(ServerAdapter): + quiet = True + def run(self, handler): # pragma: no cover + from wsgiref.handlers import CGIHandler + def fixed_environ(environ, start_response): + environ.setdefault('PATH_INFO', '') + return handler(environ, start_response) + CGIHandler().run(fixed_environ) + + +class FlupFCGIServer(ServerAdapter): + def run(self, handler): # pragma: no cover + import flup.server.fcgi + self.options.setdefault('bindAddress', (self.host, self.port)) + flup.server.fcgi.WSGIServer(handler, **self.options).run() + + +class WSGIRefServer(ServerAdapter): + def run(self, app): # pragma: no cover + from wsgiref.simple_server import WSGIRequestHandler, WSGIServer + from wsgiref.simple_server import make_server + import socket + + class FixedHandler(WSGIRequestHandler): + def address_string(self): # Prevent reverse DNS lookups please. + return self.client_address[0] + def log_request(*args, **kw): + if not self.quiet: + return WSGIRequestHandler.log_request(*args, **kw) + + handler_cls = self.options.get('handler_class', FixedHandler) + server_cls = self.options.get('server_class', WSGIServer) + + if ':' in self.host: # Fix wsgiref for IPv6 addresses. + if getattr(server_cls, 'address_family') == socket.AF_INET: + class server_cls(server_cls): + address_family = socket.AF_INET6 + + srv = make_server(self.host, self.port, app, server_cls, handler_cls) + srv.serve_forever() + + +class CherryPyServer(ServerAdapter): + def run(self, handler): # pragma: no cover + from cherrypy import wsgiserver + self.options['bind_addr'] = (self.host, self.port) + self.options['wsgi_app'] = handler + + certfile = self.options.get('certfile') + if certfile: + del self.options['certfile'] + keyfile = self.options.get('keyfile') + if keyfile: + del self.options['keyfile'] + + server = wsgiserver.CherryPyWSGIServer(**self.options) + if certfile: + server.ssl_certificate = certfile + if keyfile: + server.ssl_private_key = keyfile + + try: + server.start() + finally: + server.stop() + + +class WaitressServer(ServerAdapter): + def run(self, handler): + from waitress import serve + serve(handler, host=self.host, port=self.port) + + +class PasteServer(ServerAdapter): + def run(self, handler): # pragma: no cover + from paste import httpserver + from paste.translogger import TransLogger + handler = TransLogger(handler, setup_console_handler=(not self.quiet)) + httpserver.serve(handler, host=self.host, port=str(self.port), + **self.options) + + +class MeinheldServer(ServerAdapter): + def run(self, handler): + from meinheld import server + server.listen((self.host, self.port)) + server.run(handler) + + +class FapwsServer(ServerAdapter): + """ Extremely fast webserver using libev. See http://www.fapws.org/ """ + def run(self, handler): # pragma: no cover + import fapws._evwsgi as evwsgi + from fapws import base, config + port = self.port + if float(config.SERVER_IDENT[-2:]) > 0.4: + # fapws3 silently changed its API in 0.5 + port = str(port) + evwsgi.start(self.host, port) + # fapws3 never releases the GIL. Complain upstream. I tried. No luck. + if 'BOTTLE_CHILD' in os.environ and not self.quiet: + _stderr("WARNING: Auto-reloading does not work with Fapws3.\n") + _stderr(" (Fapws3 breaks python thread support)\n") + evwsgi.set_base_module(base) + def app(environ, start_response): + environ['wsgi.multiprocess'] = False + return handler(environ, start_response) + evwsgi.wsgi_cb(('', app)) + evwsgi.run() + + +class TornadoServer(ServerAdapter): + """ The super hyped asynchronous server by facebook. Untested. """ + def run(self, handler): # pragma: no cover + import tornado.wsgi, tornado.httpserver, tornado.ioloop + container = tornado.wsgi.WSGIContainer(handler) + server = tornado.httpserver.HTTPServer(container) + server.listen(port=self.port,address=self.host) + tornado.ioloop.IOLoop.instance().start() + + +class AppEngineServer(ServerAdapter): + """ Adapter for Google App Engine. """ + quiet = True + def run(self, handler): + from google.appengine.ext.webapp import util + # A main() function in the handler script enables 'App Caching'. + # Lets makes sure it is there. This _really_ improves performance. + module = sys.modules.get('__main__') + if module and not hasattr(module, 'main'): + module.main = lambda: util.run_wsgi_app(handler) + util.run_wsgi_app(handler) + + +class TwistedServer(ServerAdapter): + """ Untested. """ + def run(self, handler): + from twisted.web import server, wsgi + from twisted.python.threadpool import ThreadPool + from twisted.internet import reactor + thread_pool = ThreadPool() + thread_pool.start() + reactor.addSystemEventTrigger('after', 'shutdown', thread_pool.stop) + factory = server.Site(wsgi.WSGIResource(reactor, thread_pool, handler)) + reactor.listenTCP(self.port, factory, interface=self.host) + reactor.run() + + +class DieselServer(ServerAdapter): + """ Untested. """ + def run(self, handler): + from diesel.protocols.wsgi import WSGIApplication + app = WSGIApplication(handler, port=self.port) + app.run() + + +class GeventServer(ServerAdapter): + """ Untested. Options: + + * `fast` (default: False) uses libevent's http server, but has some + issues: No streaming, no pipelining, no SSL. + * See gevent.wsgi.WSGIServer() documentation for more options. + """ + def run(self, handler): + from gevent import pywsgi, local + if not isinstance(threading.local(), local.local): + msg = "Bottle requires gevent.monkey.patch_all() (before import)" + raise RuntimeError(msg) + if self.options.pop('fast', None): + depr('The "fast" option has been deprecated and removed by Gevent.') + if self.quiet: + self.options['log'] = None + address = (self.host, self.port) + server = pywsgi.WSGIServer(address, handler, **self.options) + if 'BOTTLE_CHILD' in os.environ: + import signal + signal.signal(signal.SIGINT, lambda s, f: server.stop()) + server.serve_forever() + + +class GeventSocketIOServer(ServerAdapter): + def run(self,handler): + from socketio import server + address = (self.host, self.port) + server.SocketIOServer(address, handler, **self.options).serve_forever() + + +class GunicornServer(ServerAdapter): + """ Untested. See http://gunicorn.org/configure.html for options. """ + def run(self, handler): + from gunicorn.app.base import Application + + config = {'bind': "%s:%d" % (self.host, int(self.port))} + config.update(self.options) + + class GunicornApplication(Application): + def init(self, parser, opts, args): + return config + + def load(self): + return handler + + GunicornApplication().run() + + +class EventletServer(ServerAdapter): + """ Untested """ + def run(self, handler): + from eventlet import wsgi, listen + try: + wsgi.server(listen((self.host, self.port)), handler, + log_output=(not self.quiet)) + except TypeError: + # Fallback, if we have old version of eventlet + wsgi.server(listen((self.host, self.port)), handler) + + +class RocketServer(ServerAdapter): + """ Untested. """ + def run(self, handler): + from rocket import Rocket + server = Rocket((self.host, self.port), 'wsgi', { 'wsgi_app' : handler }) + server.start() + + +class BjoernServer(ServerAdapter): + """ Fast server written in C: https://github.com/jonashaag/bjoern """ + def run(self, handler): + from bjoern import run + run(handler, self.host, self.port) + + +class AutoServer(ServerAdapter): + """ Untested. """ + adapters = [WaitressServer, PasteServer, TwistedServer, CherryPyServer, WSGIRefServer] + def run(self, handler): + for sa in self.adapters: + try: + return sa(self.host, self.port, **self.options).run(handler) + except ImportError: + pass + +server_names = { + 'cgi': CGIServer, + 'flup': FlupFCGIServer, + 'wsgiref': WSGIRefServer, + 'waitress': WaitressServer, + 'cherrypy': CherryPyServer, + 'paste': PasteServer, + 'fapws3': FapwsServer, + 'tornado': TornadoServer, + 'gae': AppEngineServer, + 'twisted': TwistedServer, + 'diesel': DieselServer, + 'meinheld': MeinheldServer, + 'gunicorn': GunicornServer, + 'eventlet': EventletServer, + 'gevent': GeventServer, + 'geventSocketIO':GeventSocketIOServer, + 'rocket': RocketServer, + 'bjoern' : BjoernServer, + 'auto': AutoServer, +} + + + + + + +############################################################################### +# Application Control ########################################################## +############################################################################### + + +def load(target, **namespace): + """ Import a module or fetch an object from a module. + + * ``package.module`` returns `module` as a module object. + * ``pack.mod:name`` returns the module variable `name` from `pack.mod`. + * ``pack.mod:func()`` calls `pack.mod.func()` and returns the result. + + The last form accepts not only function calls, but any type of + expression. Keyword arguments passed to this function are available as + local variables. Example: ``import_string('re:compile(x)', x='[a-z]')`` + """ + module, target = target.split(":", 1) if ':' in target else (target, None) + if module not in sys.modules: __import__(module) + if not target: return sys.modules[module] + if target.isalnum(): return getattr(sys.modules[module], target) + package_name = module.split('.')[0] + namespace[package_name] = sys.modules[package_name] + return eval('%s.%s' % (module, target), namespace) + + +def load_app(target): + """ Load a bottle application from a module and make sure that the import + does not affect the current default application, but returns a separate + application object. See :func:`load` for the target parameter. """ + global NORUN; NORUN, nr_old = True, NORUN + try: + tmp = default_app.push() # Create a new "default application" + rv = load(target) # Import the target module + return rv if callable(rv) else tmp + finally: + default_app.remove(tmp) # Remove the temporary added default application + NORUN = nr_old + +_debug = debug +def run(app=None, server='wsgiref', host='127.0.0.1', port=8080, + interval=1, reloader=False, quiet=False, plugins=None, + debug=None, **kargs): + """ Start a server instance. This method blocks until the server terminates. + + :param app: WSGI application or target string supported by + :func:`load_app`. (default: :func:`default_app`) + :param server: Server adapter to use. See :data:`server_names` keys + for valid names or pass a :class:`ServerAdapter` subclass. + (default: `wsgiref`) + :param host: Server address to bind to. Pass ``0.0.0.0`` to listens on + all interfaces including the external one. (default: 127.0.0.1) + :param port: Server port to bind to. Values below 1024 require root + privileges. (default: 8080) + :param reloader: Start auto-reloading server? (default: False) + :param interval: Auto-reloader interval in seconds (default: 1) + :param quiet: Suppress output to stdout and stderr? (default: False) + :param options: Options passed to the server adapter. + """ + if NORUN: return + if reloader and not os.environ.get('BOTTLE_CHILD'): + try: + lockfile = None + fd, lockfile = tempfile.mkstemp(prefix='bottle.', suffix='.lock') + os.close(fd) # We only need this file to exist. We never write to it + while os.path.exists(lockfile): + args = [sys.executable] + sys.argv + environ = os.environ.copy() + environ['BOTTLE_CHILD'] = 'true' + environ['BOTTLE_LOCKFILE'] = lockfile + p = subprocess.Popen(args, env=environ) + while p.poll() is None: # Busy wait... + os.utime(lockfile, None) # I am alive! + time.sleep(interval) + if p.poll() != 3: + if os.path.exists(lockfile): os.unlink(lockfile) + sys.exit(p.poll()) + except KeyboardInterrupt: + pass + finally: + if os.path.exists(lockfile): + os.unlink(lockfile) + return + + try: + if debug is not None: _debug(debug) + app = app or default_app() + if isinstance(app, basestring): + app = load_app(app) + if not callable(app): + raise ValueError("Application is not callable: %r" % app) + + for plugin in plugins or []: + app.install(plugin) + + if server in server_names: + server = server_names.get(server) + if isinstance(server, basestring): + server = load(server) + if isinstance(server, type): + server = server(host=host, port=port, **kargs) + if not isinstance(server, ServerAdapter): + raise ValueError("Unknown or unsupported server: %r" % server) + + server.quiet = server.quiet or quiet + if not server.quiet: + _stderr("Bottle v%s server starting up (using %s)...\n" % (__version__, repr(server))) + _stderr("Listening on http://%s:%d/\n" % (server.host, server.port)) + _stderr("Hit Ctrl-C to quit.\n\n") + + if reloader: + lockfile = os.environ.get('BOTTLE_LOCKFILE') + bgcheck = FileCheckerThread(lockfile, interval) + with bgcheck: + server.run(app) + if bgcheck.status == 'reload': + sys.exit(3) + else: + server.run(app) + except KeyboardInterrupt: + pass + except (SystemExit, MemoryError): + raise + except: + if not reloader: raise + if not getattr(server, 'quiet', quiet): + print_exc() + time.sleep(interval) + sys.exit(3) + + + +class FileCheckerThread(threading.Thread): + ''' Interrupt main-thread as soon as a changed module file is detected, + the lockfile gets deleted or gets to old. ''' + + def __init__(self, lockfile, interval): + threading.Thread.__init__(self) + self.lockfile, self.interval = lockfile, interval + #: Is one of 'reload', 'error' or 'exit' + self.status = None + + def run(self): + exists = os.path.exists + mtime = lambda path: os.stat(path).st_mtime + files = dict() + + for module in list(sys.modules.values()): + path = getattr(module, '__file__', '') or '' + if path[-4:] in ('.pyo', '.pyc'): path = path[:-1] + if path and exists(path): files[path] = mtime(path) + + while not self.status: + if not exists(self.lockfile)\ + or mtime(self.lockfile) < time.time() - self.interval - 5: + self.status = 'error' + thread.interrupt_main() + for path, lmtime in list(files.items()): + if not exists(path) or mtime(path) > lmtime: + self.status = 'reload' + thread.interrupt_main() + break + time.sleep(self.interval) + + def __enter__(self): + self.start() + + def __exit__(self, exc_type, exc_val, exc_tb): + if not self.status: self.status = 'exit' # silent exit + self.join() + return exc_type is not None and issubclass(exc_type, KeyboardInterrupt) + + + + + +############################################################################### +# Template Adapters ############################################################ +############################################################################### + + +class TemplateError(HTTPError): + def __init__(self, message): + HTTPError.__init__(self, 500, message) + + +class BaseTemplate(object): + """ Base class and minimal API for template adapters """ + extensions = ['tpl','html','thtml','stpl'] + settings = {} #used in prepare() + defaults = {} #used in render() + + def __init__(self, source=None, name=None, lookup=[], encoding='utf8', **settings): + """ Create a new template. + If the source parameter (str or buffer) is missing, the name argument + is used to guess a template filename. Subclasses can assume that + self.source and/or self.filename are set. Both are strings. + The lookup, encoding and settings parameters are stored as instance + variables. + The lookup parameter stores a list containing directory paths. + The encoding parameter should be used to decode byte strings or files. + The settings parameter contains a dict for engine-specific settings. + """ + self.name = name + self.source = source.read() if hasattr(source, 'read') else source + self.filename = source.filename if hasattr(source, 'filename') else None + self.lookup = [os.path.abspath(x) for x in lookup] + self.encoding = encoding + self.settings = self.settings.copy() # Copy from class variable + self.settings.update(settings) # Apply + if not self.source and self.name: + self.filename = self.search(self.name, self.lookup) + if not self.filename: + raise TemplateError('Template %s not found.' % repr(name)) + if not self.source and not self.filename: + raise TemplateError('No template specified.') + self.prepare(**self.settings) + + @classmethod + def search(cls, name, lookup=[]): + """ Search name in all directories specified in lookup. + First without, then with common extensions. Return first hit. """ + if not lookup: + depr('The template lookup path list should not be empty.') #0.12 + lookup = ['.'] + + if os.path.isabs(name) and os.path.isfile(name): + depr('Absolute template path names are deprecated.') #0.12 + return os.path.abspath(name) + + for spath in lookup: + spath = os.path.abspath(spath) + os.sep + fname = os.path.abspath(os.path.join(spath, name)) + if not fname.startswith(spath): continue + if os.path.isfile(fname): return fname + for ext in cls.extensions: + if os.path.isfile('%s.%s' % (fname, ext)): + return '%s.%s' % (fname, ext) + + @classmethod + def global_config(cls, key, *args): + ''' This reads or sets the global settings stored in class.settings. ''' + if args: + cls.settings = cls.settings.copy() # Make settings local to class + cls.settings[key] = args[0] + else: + return cls.settings[key] + + def prepare(self, **options): + """ Run preparations (parsing, caching, ...). + It should be possible to call this again to refresh a template or to + update settings. + """ + raise NotImplementedError + + def render(self, *args, **kwargs): + """ Render the template with the specified local variables and return + a single byte or unicode string. If it is a byte string, the encoding + must match self.encoding. This method must be thread-safe! + Local variables may be provided in dictionaries (args) + or directly, as keywords (kwargs). + """ + raise NotImplementedError + + +class MakoTemplate(BaseTemplate): + def prepare(self, **options): + from mako.template import Template + from mako.lookup import TemplateLookup + options.update({'input_encoding':self.encoding}) + options.setdefault('format_exceptions', bool(DEBUG)) + lookup = TemplateLookup(directories=self.lookup, **options) + if self.source: + self.tpl = Template(self.source, lookup=lookup, **options) + else: + self.tpl = Template(uri=self.name, filename=self.filename, lookup=lookup, **options) + + def render(self, *args, **kwargs): + for dictarg in args: kwargs.update(dictarg) + _defaults = self.defaults.copy() + _defaults.update(kwargs) + return self.tpl.render(**_defaults) + + +class CheetahTemplate(BaseTemplate): + def prepare(self, **options): + from Cheetah.Template import Template + self.context = threading.local() + self.context.vars = {} + options['searchList'] = [self.context.vars] + if self.source: + self.tpl = Template(source=self.source, **options) + else: + self.tpl = Template(file=self.filename, **options) + + def render(self, *args, **kwargs): + for dictarg in args: kwargs.update(dictarg) + self.context.vars.update(self.defaults) + self.context.vars.update(kwargs) + out = str(self.tpl) + self.context.vars.clear() + return out + + +class Jinja2Template(BaseTemplate): + def prepare(self, filters=None, tests=None, globals={}, **kwargs): + from jinja2 import Environment, FunctionLoader + if 'prefix' in kwargs: # TODO: to be removed after a while + raise RuntimeError('The keyword argument `prefix` has been removed. ' + 'Use the full jinja2 environment name line_statement_prefix instead.') + self.env = Environment(loader=FunctionLoader(self.loader), **kwargs) + if filters: self.env.filters.update(filters) + if tests: self.env.tests.update(tests) + if globals: self.env.globals.update(globals) + if self.source: + self.tpl = self.env.from_string(self.source) + else: + self.tpl = self.env.get_template(self.filename) + + def render(self, *args, **kwargs): + for dictarg in args: kwargs.update(dictarg) + _defaults = self.defaults.copy() + _defaults.update(kwargs) + return self.tpl.render(**_defaults) + + def loader(self, name): + fname = self.search(name, self.lookup) + if not fname: return + with open(fname, "rb") as f: + return f.read().decode(self.encoding) + + +class SimpleTemplate(BaseTemplate): + + def prepare(self, escape_func=html_escape, noescape=False, syntax=None, **ka): + self.cache = {} + enc = self.encoding + self._str = lambda x: touni(x, enc) + self._escape = lambda x: escape_func(touni(x, enc)) + self.syntax = syntax + if noescape: + self._str, self._escape = self._escape, self._str + + @cached_property + def co(self): + return compile(self.code, self.filename or '', 'exec') + + @cached_property + def code(self): + source = self.source + if not source: + with open(self.filename, 'rb') as f: + source = f.read() + try: + source, encoding = touni(source), 'utf8' + except UnicodeError: + depr('Template encodings other than utf8 are no longer supported.') #0.11 + source, encoding = touni(source, 'latin1'), 'latin1' + parser = StplParser(source, encoding=encoding, syntax=self.syntax) + code = parser.translate() + self.encoding = parser.encoding + return code + + def _rebase(self, _env, _name=None, **kwargs): + if _name is None: + depr('Rebase function called without arguments.' + ' You were probably looking for {{base}}?', True) #0.12 + _env['_rebase'] = (_name, kwargs) + + def _include(self, _env, _name=None, **kwargs): + if _name is None: + depr('Rebase function called without arguments.' + ' You were probably looking for {{base}}?', True) #0.12 + env = _env.copy() + env.update(kwargs) + if _name not in self.cache: + self.cache[_name] = self.__class__(name=_name, lookup=self.lookup) + return self.cache[_name].execute(env['_stdout'], env) + + def execute(self, _stdout, kwargs): + env = self.defaults.copy() + env.update(kwargs) + env.update({'_stdout': _stdout, '_printlist': _stdout.extend, + 'include': functools.partial(self._include, env), + 'rebase': functools.partial(self._rebase, env), '_rebase': None, + '_str': self._str, '_escape': self._escape, 'get': env.get, + 'setdefault': env.setdefault, 'defined': env.__contains__ }) + eval(self.co, env) + if env.get('_rebase'): + subtpl, rargs = env.pop('_rebase') + rargs['base'] = ''.join(_stdout) #copy stdout + del _stdout[:] # clear stdout + return self._include(env, subtpl, **rargs) + return env + + def render(self, *args, **kwargs): + """ Render the template using keyword arguments as local variables. """ + env = {}; stdout = [] + for dictarg in args: env.update(dictarg) + env.update(kwargs) + self.execute(stdout, env) + return ''.join(stdout) + + +class StplSyntaxError(TemplateError): pass + + +class StplParser(object): + ''' Parser for stpl templates. ''' + _re_cache = {} #: Cache for compiled re patterns + # This huge pile of voodoo magic splits python code into 8 different tokens. + # 1: All kinds of python strings (trust me, it works) + _re_tok = '([urbURB]?(?:\'\'(?!\')|""(?!")|\'{6}|"{6}' \ + '|\'(?:[^\\\\\']|\\\\.)+?\'|"(?:[^\\\\"]|\\\\.)+?"' \ + '|\'{3}(?:[^\\\\]|\\\\.|\\n)+?\'{3}' \ + '|"{3}(?:[^\\\\]|\\\\.|\\n)+?"{3}))' + _re_inl = _re_tok.replace('|\\n','') # We re-use this string pattern later + # 2: Comments (until end of line, but not the newline itself) + _re_tok += '|(#.*)' + # 3,4: Open and close grouping tokens + _re_tok += '|([\\[\\{\\(])' + _re_tok += '|([\\]\\}\\)])' + # 5,6: Keywords that start or continue a python block (only start of line) + _re_tok += '|^([ \\t]*(?:if|for|while|with|try|def|class)\\b)' \ + '|^([ \\t]*(?:elif|else|except|finally)\\b)' + # 7: Our special 'end' keyword (but only if it stands alone) + _re_tok += '|((?:^|;)[ \\t]*end[ \\t]*(?=(?:%(block_close)s[ \\t]*)?\\r?$|;|#))' + # 8: A customizable end-of-code-block template token (only end of line) + _re_tok += '|(%(block_close)s[ \\t]*(?=\\r?$))' + # 9: And finally, a single newline. The 10th token is 'everything else' + _re_tok += '|(\\r?\\n)' + + # Match the start tokens of code areas in a template + _re_split = '(?m)^[ \t]*(\\\\?)((%(line_start)s)|(%(block_start)s))(%%?)' + # Match inline statements (may contain python strings) + _re_inl = '(?m)%%(inline_start)s((?:%s|[^\'"\n]*?)+)%%(inline_end)s' % _re_inl + _re_tok = '(?m)' + _re_tok + + default_syntax = '<% %> % {{ }}' + + def __init__(self, source, syntax=None, encoding='utf8'): + self.source, self.encoding = touni(source, encoding), encoding + self.set_syntax(syntax or self.default_syntax) + self.code_buffer, self.text_buffer = [], [] + self.lineno, self.offset = 1, 0 + self.indent, self.indent_mod = 0, 0 + self.paren_depth = 0 + + def get_syntax(self): + ''' Tokens as a space separated string (default: <% %> % {{ }}) ''' + return self._syntax + + def set_syntax(self, syntax): + self._syntax = syntax + self._tokens = syntax.split() + if not syntax in self._re_cache: + names = 'block_start block_close line_start inline_start inline_end' + etokens = map(re.escape, self._tokens) + pattern_vars = dict(zip(names.split(), etokens)) + patterns = (self._re_split, self._re_tok, self._re_inl) + patterns = [re.compile(p%pattern_vars) for p in patterns] + self._re_cache[syntax] = patterns + self.re_split, self.re_tok, self.re_inl = self._re_cache[syntax] + + syntax = property(get_syntax, set_syntax) + + def translate(self): + if self.offset: raise RuntimeError('Parser is a one time instance.') + while True: + m = self.re_split.search(self.source[self.offset:]) + if m: + text = self.source[self.offset:self.offset+m.start()] + self.text_buffer.append(text) + self.offset += m.end() + if m.group(1): # New escape syntax + line, sep, _ = self.source[self.offset:].partition('\n') + self.text_buffer.append(m.group(2)+m.group(5)+line+sep) + self.offset += len(line+sep)+1 + continue + elif m.group(5): # Old escape syntax + depr('Escape code lines with a backslash.') #0.12 + line, sep, _ = self.source[self.offset:].partition('\n') + self.text_buffer.append(m.group(2)+line+sep) + self.offset += len(line+sep)+1 + continue + self.flush_text() + self.read_code(multiline=bool(m.group(4))) + else: break + self.text_buffer.append(self.source[self.offset:]) + self.flush_text() + return ''.join(self.code_buffer) + + def read_code(self, multiline): + code_line, comment = '', '' + while True: + m = self.re_tok.search(self.source[self.offset:]) + if not m: + code_line += self.source[self.offset:] + self.offset = len(self.source) + self.write_code(code_line.strip(), comment) + return + code_line += self.source[self.offset:self.offset+m.start()] + self.offset += m.end() + _str, _com, _po, _pc, _blk1, _blk2, _end, _cend, _nl = m.groups() + if (code_line or self.paren_depth > 0) and (_blk1 or _blk2): # a if b else c + code_line += _blk1 or _blk2 + continue + if _str: # Python string + code_line += _str + elif _com: # Python comment (up to EOL) + comment = _com + if multiline and _com.strip().endswith(self._tokens[1]): + multiline = False # Allow end-of-block in comments + elif _po: # open parenthesis + self.paren_depth += 1 + code_line += _po + elif _pc: # close parenthesis + if self.paren_depth > 0: + # we could check for matching parentheses here, but it's + # easier to leave that to python - just check counts + self.paren_depth -= 1 + code_line += _pc + elif _blk1: # Start-block keyword (if/for/while/def/try/...) + code_line, self.indent_mod = _blk1, -1 + self.indent += 1 + elif _blk2: # Continue-block keyword (else/elif/except/...) + code_line, self.indent_mod = _blk2, -1 + elif _end: # The non-standard 'end'-keyword (ends a block) + self.indent -= 1 + elif _cend: # The end-code-block template token (usually '%>') + if multiline: multiline = False + else: code_line += _cend + else: # \n + self.write_code(code_line.strip(), comment) + self.lineno += 1 + code_line, comment, self.indent_mod = '', '', 0 + if not multiline: + break + + def flush_text(self): + text = ''.join(self.text_buffer) + del self.text_buffer[:] + if not text: return + parts, pos, nl = [], 0, '\\\n'+' '*self.indent + for m in self.re_inl.finditer(text): + prefix, pos = text[pos:m.start()], m.end() + if prefix: + parts.append(nl.join(map(repr, prefix.splitlines(True)))) + if prefix.endswith('\n'): parts[-1] += nl + parts.append(self.process_inline(m.group(1).strip())) + if pos < len(text): + prefix = text[pos:] + lines = prefix.splitlines(True) + if lines[-1].endswith('\\\\\n'): lines[-1] = lines[-1][:-3] + elif lines[-1].endswith('\\\\\r\n'): lines[-1] = lines[-1][:-4] + parts.append(nl.join(map(repr, lines))) + code = '_printlist((%s,))' % ', '.join(parts) + self.lineno += code.count('\n')+1 + self.write_code(code) + + def process_inline(self, chunk): + if chunk[0] == '!': return '_str(%s)' % chunk[1:] + return '_escape(%s)' % chunk + + def write_code(self, line, comment=''): + line, comment = self.fix_backward_compatibility(line, comment) + code = ' ' * (self.indent+self.indent_mod) + code += line.lstrip() + comment + '\n' + self.code_buffer.append(code) + + def fix_backward_compatibility(self, line, comment): + parts = line.strip().split(None, 2) + if parts and parts[0] in ('include', 'rebase'): + depr('The include and rebase keywords are functions now.') #0.12 + if len(parts) == 1: return "_printlist([base])", comment + elif len(parts) == 2: return "_=%s(%r)" % tuple(parts), comment + else: return "_=%s(%r, %s)" % tuple(parts), comment + if self.lineno <= 2 and not line.strip() and 'coding' in comment: + m = re.match(r"#.*coding[:=]\s*([-\w.]+)", comment) + if m: + depr('PEP263 encoding strings in templates are deprecated.') #0.12 + enc = m.group(1) + self.source = self.source.encode(self.encoding).decode(enc) + self.encoding = enc + return line, comment.replace('coding','coding*') + return line, comment + + +def template(*args, **kwargs): + ''' + Get a rendered template as a string iterator. + You can use a name, a filename or a template string as first parameter. + Template rendering arguments can be passed as dictionaries + or directly (as keyword arguments). + ''' + tpl = args[0] if args else None + adapter = kwargs.pop('template_adapter', SimpleTemplate) + lookup = kwargs.pop('template_lookup', TEMPLATE_PATH) + tplid = (id(lookup), tpl) + if tplid not in TEMPLATES or DEBUG: + settings = kwargs.pop('template_settings', {}) + if isinstance(tpl, adapter): + TEMPLATES[tplid] = tpl + if settings: TEMPLATES[tplid].prepare(**settings) + elif "\n" in tpl or "{" in tpl or "%" in tpl or '$' in tpl: + TEMPLATES[tplid] = adapter(source=tpl, lookup=lookup, **settings) + else: + TEMPLATES[tplid] = adapter(name=tpl, lookup=lookup, **settings) + if not TEMPLATES[tplid]: + abort(500, 'Template (%s) not found' % tpl) + for dictarg in args[1:]: kwargs.update(dictarg) + return TEMPLATES[tplid].render(kwargs) + +mako_template = functools.partial(template, template_adapter=MakoTemplate) +cheetah_template = functools.partial(template, template_adapter=CheetahTemplate) +jinja2_template = functools.partial(template, template_adapter=Jinja2Template) + + +def view(tpl_name, **defaults): + ''' Decorator: renders a template for a handler. + The handler can control its behavior like that: + + - return a dict of template vars to fill out the template + - return something other than a dict and the view decorator will not + process the template, but return the handler result as is. + This includes returning a HTTPResponse(dict) to get, + for instance, JSON with autojson or other castfilters. + ''' + def decorator(func): + @functools.wraps(func) + def wrapper(*args, **kwargs): + result = func(*args, **kwargs) + if isinstance(result, (dict, DictMixin)): + tplvars = defaults.copy() + tplvars.update(result) + return template(tpl_name, **tplvars) + elif result is None: + return template(tpl_name, defaults) + return result + return wrapper + return decorator + +mako_view = functools.partial(view, template_adapter=MakoTemplate) +cheetah_view = functools.partial(view, template_adapter=CheetahTemplate) +jinja2_view = functools.partial(view, template_adapter=Jinja2Template) + + + + + + +############################################################################### +# Constants and Globals ######################################################## +############################################################################### + + +TEMPLATE_PATH = ['./', './views/'] +TEMPLATES = {} +DEBUG = False +NORUN = False # If set, run() does nothing. Used by load_app() + +#: A dict to map HTTP status codes (e.g. 404) to phrases (e.g. 'Not Found') +HTTP_CODES = httplib.responses +HTTP_CODES[418] = "I'm a teapot" # RFC 2324 +HTTP_CODES[422] = "Unprocessable Entity" # RFC 4918 +HTTP_CODES[428] = "Precondition Required" +HTTP_CODES[429] = "Too Many Requests" +HTTP_CODES[431] = "Request Header Fields Too Large" +HTTP_CODES[511] = "Network Authentication Required" +_HTTP_STATUS_LINES = dict((k, '%d %s'%(k,v)) for (k,v) in HTTP_CODES.items()) + +#: The default template used for error pages. Override with @error() +ERROR_PAGE_TEMPLATE = """ +%%try: + %%from %s import DEBUG, HTTP_CODES, request, touni + + + + Error: {{e.status}} + + + +

Error: {{e.status}}

+

Sorry, the requested URL {{repr(request.url)}} + caused an error:

+ 
{{e.body}}
+ %%if DEBUG and e.exception: +

Exception:

+ 
{{repr(e.exception)}}
+ %%end + %%if DEBUG and e.traceback: +

Traceback:

+ 
{{e.traceback}}
+ %%end + + +%%except ImportError: + ImportError: Could not generate the error page. Please add bottle to + the import path. +%%end +""" % __name__ + +#: A thread-safe instance of :class:`LocalRequest`. If accessed from within a +#: request callback, this instance always refers to the *current* request +#: (even on a multithreaded server). +request = LocalRequest() + +#: A thread-safe instance of :class:`LocalResponse`. It is used to change the +#: HTTP response for the *current* request. +response = LocalResponse() + +#: A thread-safe namespace. Not used by Bottle. +local = threading.local() + +# Initialize app stack (create first empty Bottle app) +# BC: 0.6.4 and needed for run() +app = default_app = AppStack() +app.push() + +#: A virtual package that redirects import statements. +#: Example: ``import bottle.ext.sqlite`` actually imports `bottle_sqlite`. +ext = _ImportRedirect('bottle.ext' if __name__ == '__main__' else __name__+".ext", 'bottle_%s').module + +if __name__ == '__main__': + opt, args, parser = _cmd_options, _cmd_args, _cmd_parser + if opt.version: + _stdout('Bottle %s\n'%__version__) + sys.exit(0) + if not args: + parser.print_help() + _stderr('\nError: No application specified.\n') + sys.exit(1) + + sys.path.insert(0, '.') + sys.modules.setdefault('bottle', sys.modules['__main__']) + + host, port = (opt.bind or 'localhost'), 8080 + if ':' in host and host.rfind(']') < host.rfind(':'): + host, port = host.rsplit(':', 1) + host = host.strip('[]') + + run(args[0], host=host, port=int(port), server=opt.server, + reloader=opt.reload, plugins=opt.plugin, debug=opt.debug) + + + + +# THE END diff --git a/appendix-A/container_perftest.py b/appendix-A/container_perftest.py new file mode 100644 index 0000000..8f8176d --- /dev/null +++ b/appendix-A/container_perftest.py @@ -0,0 +1,76 @@ +""" +Container ``in`` operator performance test +""" +import sys +import timeit + +SETUP = ''' +import array +selected = array.array('d') +with open('selected.arr', 'rb') as fp: + selected.fromfile(fp, {size}) +if {test_type} is dict: + haystack = dict.fromkeys(selected, 1) +else: + haystack = {test_type}(selected) +if {verbose}: + print(type(haystack), end=' ') + print('haystack: %10d' % len(haystack), end=' ') +needles = array.array('d') +with open('not_selected.arr', 'rb') as fp: + needles.fromfile(fp, 500) +needles.extend(selected[::{size}//500]) +if {verbose}: + print(' needles: %10d' % len(needles), end=' ') +''' + +SETUP_NEEDLE_SET = ''' +needle_set = set(needles) +if {verbose}: + print(' needle_set: %10d' % len(needle_set), end=' ') +''' + + +TEST = ''' +found = 0 +for n in needles: + if n in haystack: + found += 1 +if {verbose}: + print(' found: %10d' % found) +''' + +TEST_INTERSECT = ''' +found = len(needle_set & haystack) +if {verbose}: + print(' found: %10d' % found) +''' + +def test(test_type, verbose): + MAX_EXPONENT = 7 + if test_type == 'intersect': + test_type = 'set' + setup_template = SETUP + SETUP_NEEDLE_SET + test_template = TEST_INTERSECT + else: + setup_template = SETUP + test_template = TEST + + for n in range(3, MAX_EXPONENT + 1): + size = 10**n + setup = setup_template.format(test_type=test_type, + size=size, verbose=verbose) + test = test_template.format(verbose=verbose) + tt = timeit.repeat(stmt=test, setup=setup, repeat=5, number=1) + print('|{:{}d}|{:f}'.format(size, MAX_EXPONENT + 1, min(tt))) + +if __name__=='__main__': + if '-v' in sys.argv: + sys.argv.remove('-v') + verbose = True + else: + verbose = False + if len(sys.argv) != 2: + print('Usage: %s ' % sys.argv[0]) + else: + test(sys.argv[1], verbose) diff --git a/appendix-A/container_perftest_datagen.py b/appendix-A/container_perftest_datagen.py new file mode 100644 index 0000000..0f01ddd --- /dev/null +++ b/appendix-A/container_perftest_datagen.py @@ -0,0 +1,37 @@ +""" +Generate data for container performance test +""" + +import random +import array + +MAX_EXPONENT = 7 +HAYSTACK_LEN = 10 ** MAX_EXPONENT +NEEDLES_LEN = 10 ** (MAX_EXPONENT - 1) +SAMPLE_LEN = HAYSTACK_LEN + NEEDLES_LEN // 2 + +needles = array.array('d') + +sample = {1/random.random() for i in range(SAMPLE_LEN)} +print('initial sample: %d elements' % len(sample)) + +# complete sample, in case duplicate random numbers were discarded +while len(sample) < SAMPLE_LEN: + sample.add(1/random.random()) + +print('complete sample: %d elements' % len(sample)) + +sample = array.array('d', sample) +random.shuffle(sample) + +not_selected = sample[:NEEDLES_LEN // 2] +print('not selected: %d samples' % len(not_selected)) +print(' writing not_selected.arr') +with open('not_selected.arr', 'wb') as fp: + not_selected.tofile(fp) + +selected = sample[NEEDLES_LEN // 2:] +print('selected: %d samples' % len(selected)) +print(' writing selected.arr') +with open('selected.arr', 'wb') as fp: + selected.tofile(fp) diff --git a/appendix-A/emoji_families.py b/appendix-A/emoji_families.py new file mode 100755 index 0000000..d4d3a4c --- /dev/null +++ b/appendix-A/emoji_families.py @@ -0,0 +1,87 @@ +#!/usr/bin/env python3.8 + +import sys +import io +from itertools import product +from unicodedata import name, unidata_version + +from bottle import route, run, response, template + + +ZWJ = '\u200D' # ZERO WIDTH JOINER + +ADULTS = '\N{MAN}\N{WOMAN}' +CHILDREN = '\N{BOY}\N{GIRL}' + +adults = list(ADULTS) + list(product(ADULTS, repeat=2)) +children = list(CHILDREN) + list(product(CHILDREN, repeat=2)) + + +def matrix(): + for kids in children: + row = [] + suffix = ZWJ + ZWJ.join(kids) + for p in adults: + row.append(ZWJ.join(p)+suffix) + yield row + + +def report(): + out = io.StringIO() + print('Python', '.'.join(str(n) for n in sys.version_info[:3]), file=out) + print('Unicode', unidata_version, file=out) + + print(end='\t', file=out) + for parents in adults: + print('+'.join(parents), end='\t', file=out) + + print(file=out) + + for kids, row in zip(children, matrix()): + print('+'.join(kids), end='\t', file=out) + for cell in row: + print(cell, end='\t', file=out) + + print(file=out) + + return out.getvalue() + +TABLE = """ + + + + + + % for parents in adults: + + % end + + % for kids, row in zip(children, matrix()): + + + % for cell in row: + + % end + + % end +
{{'+'.join(parents)}}
{{'+'.join(kids)}}{{cell}}
+ + +""" + + +@route('/') +def index(): + response.content_type = 'text/html; charset=utf-8' + return template(TABLE, **globals()) + + +if __name__ == '__main__': + run(host='localhost', port=8080) + + diff --git a/appendix-A/flags2_common.py b/appendix-A/flags2_common.py new file mode 100644 index 0000000..bfa40fb --- /dev/null +++ b/appendix-A/flags2_common.py @@ -0,0 +1,149 @@ +"""Utilities for second set of flag examples. +""" + +import os +import time +import sys +import string +import argparse +from collections import namedtuple +from enum import Enum + + +Result = namedtuple('Result', 'status data') + +HTTPStatus = Enum('Status', 'ok not_found error') + +POP20_CC = ('CN IN US ID BR PK NG BD RU JP ' + 'MX PH VN ET EG DE IR TR CD FR').split() + +DEFAULT_CONCUR_REQ = 1 +MAX_CONCUR_REQ = 1 + +SERVERS = { + 'REMOTE': 'http://flupy.org/data/flags', + 'LOCAL': 'http://localhost:8001/flags', + 'DELAY': 'http://localhost:8002/flags', + 'ERROR': 'http://localhost:8003/flags', +} +DEFAULT_SERVER = 'LOCAL' + +DEST_DIR = 'downloads/' +COUNTRY_CODES_FILE = 'country_codes.txt' + + +def save_flag(img, filename): + path = os.path.join(DEST_DIR, filename) + with open(path, 'wb') as fp: + fp.write(img) + + +def initial_report(cc_list, actual_req, server_label): + if len(cc_list) <= 10: + cc_msg = ', '.join(cc_list) + else: + cc_msg = 'from {} to {}'.format(cc_list[0], cc_list[-1]) + print('{} site: {}'.format(server_label, SERVERS[server_label])) + msg = 'Searching for {} flag{}: {}' + plural = 's' if len(cc_list) != 1 else '' + print(msg.format(len(cc_list), plural, cc_msg)) + plural = 's' if actual_req != 1 else '' + msg = '{} concurrent connection{} will be used.' + print(msg.format(actual_req, plural)) + + +def final_report(cc_list, counter, start_time): + elapsed = time.time() - start_time + print('-' * 20) + msg = '{} flag{} downloaded.' + plural = 's' if counter[HTTPStatus.ok] != 1 else '' + print(msg.format(counter[HTTPStatus.ok], plural)) + if counter[HTTPStatus.not_found]: + print(counter[HTTPStatus.not_found], 'not found.') + if counter[HTTPStatus.error]: + plural = 's' if counter[HTTPStatus.error] != 1 else '' + print('{} error{}.'.format(counter[HTTPStatus.error], plural)) + print('Elapsed time: {:.2f}s'.format(elapsed)) + + +def expand_cc_args(every_cc, all_cc, cc_args, limit): + codes = set() + A_Z = string.ascii_uppercase + if every_cc: + codes.update(a+b for a in A_Z for b in A_Z) + elif all_cc: + with open(COUNTRY_CODES_FILE) as fp: + text = fp.read() + codes.update(text.split()) + else: + for cc in (c.upper() for c in cc_args): + if len(cc) == 1 and cc in A_Z: + codes.update(cc+c for c in A_Z) + elif len(cc) == 2 and all(c in A_Z for c in cc): + codes.add(cc) + else: + msg = 'each CC argument must be A to Z or AA to ZZ.' + raise ValueError('*** Usage error: '+msg) + return sorted(codes)[:limit] + + +def process_args(default_concur_req): + server_options = ', '.join(sorted(SERVERS)) + parser = argparse.ArgumentParser( + description='Download flags for country codes. ' + 'Default: top 20 countries by population.') + parser.add_argument('cc', metavar='CC', nargs='*', + help='country code or 1st letter (eg. B for BA...BZ)') + parser.add_argument('-a', '--all', action='store_true', + help='get all available flags (AD to ZW)') + parser.add_argument('-e', '--every', action='store_true', + help='get flags for every possible code (AA...ZZ)') + parser.add_argument('-l', '--limit', metavar='N', type=int, + help='limit to N first codes', default=sys.maxsize) + parser.add_argument('-m', '--max_req', metavar='CONCURRENT', type=int, + default=default_concur_req, + help='maximum concurrent requests (default={})' + .format(default_concur_req)) + parser.add_argument('-s', '--server', metavar='LABEL', + default=DEFAULT_SERVER, + help='Server to hit; one of {} (default={})' + .format(server_options, DEFAULT_SERVER)) + parser.add_argument('-v', '--verbose', action='store_true', + help='output detailed progress info') + args = parser.parse_args() + if args.max_req < 1: + print('*** Usage error: --max_req CONCURRENT must be >= 1') + parser.print_usage() + sys.exit(1) + if args.limit < 1: + print('*** Usage error: --limit N must be >= 1') + parser.print_usage() + sys.exit(1) + args.server = args.server.upper() + if args.server not in SERVERS: + print('*** Usage error: --server LABEL must be one of', + server_options) + parser.print_usage() + sys.exit(1) + try: + cc_list = expand_cc_args(args.every, args.all, args.cc, args.limit) + except ValueError as exc: + print(exc.args[0]) + parser.print_usage() + sys.exit(1) + + if not cc_list: + cc_list = sorted(POP20_CC) + return args, cc_list + + +def main(download_many, default_concur_req, max_concur_req): + args, cc_list = process_args(default_concur_req) + actual_req = min(args.max_req, max_concur_req, len(cc_list)) + initial_report(cc_list, actual_req, args.server) + base_url = SERVERS[args.server] + t0 = time.time() + counter = download_many(cc_list, base_url, args.verbose, actual_req) + assert sum(counter.values()) == len(cc_list), \ + 'some downloads are unaccounted for' + final_report(cc_list, counter, t0) diff --git a/appendix-A/flags2_sequential.py b/appendix-A/flags2_sequential.py new file mode 100644 index 0000000..05ec302 --- /dev/null +++ b/appendix-A/flags2_sequential.py @@ -0,0 +1,87 @@ +"""Download flags of countries (with error handling). + +Sequential version + +Sample run:: + + $ python3 flags2_sequential.py -s DELAY b + DELAY site: http://localhost:8002/flags + Searching for 26 flags: from BA to BZ + 1 concurrent connection will be used. + -------------------- + 17 flags downloaded. + 9 not found. + Elapsed time: 13.36s + +""" + +import collections + +import requests +import tqdm + +from flags2_common import main, save_flag, HTTPStatus, Result + + +DEFAULT_CONCUR_REQ = 1 +MAX_CONCUR_REQ = 1 + +# BEGIN FLAGS2_BASIC_HTTP_FUNCTIONS +def get_flag(base_url, cc): + url = '{}/{cc}/{cc}.gif'.format(base_url, cc=cc.lower()) + resp = requests.get(url) + if resp.status_code != 200: + resp.raise_for_status() + return resp.content + + +def download_one(cc, base_url, verbose=False): + try: + image = get_flag(base_url, cc) + except requests.exceptions.HTTPError as exc: + res = exc.response + if res.status_code == 404: + status = HTTPStatus.not_found + msg = 'not found' + else: + raise + else: + save_flag(image, cc.lower() + '.gif') + status = HTTPStatus.ok + msg = 'OK' + + if verbose: + print(cc, msg) + + return Result(status, cc) +# END FLAGS2_BASIC_HTTP_FUNCTIONS + +# BEGIN FLAGS2_DOWNLOAD_MANY_SEQUENTIAL +def download_many(cc_list, base_url, verbose, max_req): + counter = collections.Counter() + cc_iter = sorted(cc_list) + if not verbose: + cc_iter = tqdm.tqdm(cc_iter) + for cc in cc_iter: + try: + res = download_one(cc, base_url, verbose) + except requests.exceptions.HTTPError as exc: + error_msg = 'HTTP error {res.status_code} - {res.reason}' + error_msg = error_msg.format(res=exc.response) + except requests.exceptions.ConnectionError as exc: + error_msg = 'Connection error' + else: + error_msg = '' + status = res.status + + if error_msg: + status = HTTPStatus.error + counter[status] += 1 + if verbose and error_msg: + print('*** Error for {}: {}'.format(cc, error_msg)) + + return counter +# END FLAGS2_DOWNLOAD_MANY_SEQUENTIAL + +if __name__ == '__main__': + main(download_many, DEFAULT_CONCUR_REQ, MAX_CONCUR_REQ) diff --git a/appendix-A/hashdiff.py b/appendix-A/hashdiff.py new file mode 100644 index 0000000..163e300 --- /dev/null +++ b/appendix-A/hashdiff.py @@ -0,0 +1,20 @@ +import sys + +MAX_BITS = len(format(sys.maxsize, 'b')) +print('%s-bit Python build' % (MAX_BITS + 1)) + +def hash_diff(o1, o2): + h1 = '{:>0{}b}'.format(hash(o1), MAX_BITS) + h2 = '{:>0{}b}'.format(hash(o2), MAX_BITS) + diff = ''.join('!' if b1 != b2 else ' ' for b1, b2 in zip(h1, h2)) + count = '!= {}'.format(diff.count('!')) + width = max(len(repr(o1)), len(repr(o2)), 8) + sep = '-' * (width * 2 + MAX_BITS) + return '{!r:{width}} {}\n{:{width}} {} {}\n{!r:{width}} {}\n{}'.format( + o1, h1, ' ' * width, diff, count, o2, h2, sep, width=width) + +if __name__ == '__main__': + print(hash_diff(1, 1.0)) + print(hash_diff(1.0, 1.0001)) + print(hash_diff(1.0001, 1.0002)) + print(hash_diff(1.0002, 1.0003)) diff --git a/appendix-A/isis2json.py b/appendix-A/isis2json.py new file mode 100644 index 0000000..5572766 --- /dev/null +++ b/appendix-A/isis2json.py @@ -0,0 +1,263 @@ +#!/usr/bin/env python +# -*- encoding: utf-8 -*- + +# isis2json.py: convert ISIS and ISO-2709 files to JSON +# +# Copyright (C) 2010 BIREME/PAHO/WHO +# +# This program is free software: you can redistribute it and/or modify +# it under the terms of the GNU Lesser General Public License as published +# by the Free Software Foundation, either version 2.1 of the License, or +# (at your option) any later version. + +# This program is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU Lesser General Public License for more details. + +# You should have received a copy of the GNU Lesser General Public License +# along with this program. If not, see . + +############################ +# tag::ISIS2JSON[] +# this script works with Python or Jython (versions >=2.5 and <3) + +import sys +import argparse +from uuid import uuid4 +import os + +try: + import json +except ImportError: + if os.name == 'java': # running Jython + from com.xhaus.jyson import JysonCodec as json + else: + import simplejson as json + +SKIP_INACTIVE = True +DEFAULT_QTY = 2**31 +ISIS_MFN_KEY = 'mfn' +ISIS_ACTIVE_KEY = 'active' +SUBFIELD_DELIMITER = '^' +INPUT_ENCODING = 'cp1252' + + +def iter_iso_records(iso_file_name, isis_json_type): # <1> + from iso2709 import IsoFile + from subfield import expand + + iso = IsoFile(iso_file_name) + for record in iso: + fields = {} + for field in record.directory: + field_key = str(int(field.tag)) # remove leading zeroes + field_occurrences = fields.setdefault(field_key, []) + content = field.value.decode(INPUT_ENCODING, 'replace') + if isis_json_type == 1: + field_occurrences.append(content) + elif isis_json_type == 2: + field_occurrences.append(expand(content)) + elif isis_json_type == 3: + field_occurrences.append(dict(expand(content))) + else: + raise NotImplementedError('ISIS-JSON type %s conversion ' + 'not yet implemented for .iso input' % isis_json_type) + + yield fields + iso.close() + + +def iter_mst_records(master_file_name, isis_json_type): # <2> + try: + from bruma.master import MasterFactory, Record + except ImportError: + print('IMPORT ERROR: Jython 2.5 and Bruma.jar ' + 'are required to read .mst files') + raise SystemExit + mst = MasterFactory.getInstance(master_file_name).open() + for record in mst: + fields = {} + if SKIP_INACTIVE: + if record.getStatus() != Record.Status.ACTIVE: + continue + else: # save status only there are non-active records + fields[ISIS_ACTIVE_KEY] = (record.getStatus() == + Record.Status.ACTIVE) + fields[ISIS_MFN_KEY] = record.getMfn() + for field in record.getFields(): + field_key = str(field.getId()) + field_occurrences = fields.setdefault(field_key, []) + if isis_json_type == 3: + content = {} + for subfield in field.getSubfields(): + subfield_key = subfield.getId() + if subfield_key == '*': + content['_'] = subfield.getContent() + else: + subfield_occurrences = content.setdefault(subfield_key, []) + subfield_occurrences.append(subfield.getContent()) + field_occurrences.append(content) + elif isis_json_type == 1: + content = [] + for subfield in field.getSubfields(): + subfield_key = subfield.getId() + if subfield_key == '*': + content.insert(0, subfield.getContent()) + else: + content.append(SUBFIELD_DELIMITER + subfield_key + + subfield.getContent()) + field_occurrences.append(''.join(content)) + else: + raise NotImplementedError('ISIS-JSON type %s conversion ' + 'not yet implemented for .mst input' % isis_json_type) + yield fields + mst.close() + + +def write_json(input_gen, file_name, output, qty, skip, id_tag, # <3> + gen_uuid, mongo, mfn, isis_json_type, prefix, + constant): + start = skip + end = start + qty + if id_tag: + id_tag = str(id_tag) + ids = set() + else: + id_tag = '' + for i, record in enumerate(input_gen): + if i >= end: + break + if not mongo: + if i == 0: + output.write('[') + elif i > start: + output.write(',') + if start <= i < end: + if id_tag: + occurrences = record.get(id_tag, None) + if occurrences is None: + msg = 'id tag #%s not found in record %s' + if ISIS_MFN_KEY in record: + msg = msg + (' (mfn=%s)' % record[ISIS_MFN_KEY]) + raise KeyError(msg % (id_tag, i)) + if len(occurrences) > 1: + msg = 'multiple id tags #%s found in record %s' + if ISIS_MFN_KEY in record: + msg = msg + (' (mfn=%s)' % record[ISIS_MFN_KEY]) + raise TypeError(msg % (id_tag, i)) + else: # ok, we have one and only one id field + if isis_json_type == 1: + id = occurrences[0] + elif isis_json_type == 2: + id = occurrences[0][0][1] + elif isis_json_type == 3: + id = occurrences[0]['_'] + if id in ids: + msg = 'duplicate id %s in tag #%s, record %s' + if ISIS_MFN_KEY in record: + msg = msg + (' (mfn=%s)' % record[ISIS_MFN_KEY]) + raise TypeError(msg % (id, id_tag, i)) + record['_id'] = id + ids.add(id) + elif gen_uuid: + record['_id'] = unicode(uuid4()) + elif mfn: + record['_id'] = record[ISIS_MFN_KEY] + if prefix: + # iterate over a fixed sequence of tags + for tag in tuple(record): + if str(tag).isdigit(): + record[prefix+tag] = record[tag] + del record[tag] # this is why we iterate over a tuple + # with the tags, and not directly on the record dict + if constant: + constant_key, constant_value = constant.split(':') + record[constant_key] = constant_value + output.write(json.dumps(record).encode('utf-8')) + output.write('\n') + if not mongo: + output.write(']\n') + + +def main(): # <4> + # create the parser + parser = argparse.ArgumentParser( + description='Convert an ISIS .mst or .iso file to a JSON array') + + # add the arguments + parser.add_argument( + 'file_name', metavar='INPUT.(mst|iso)', + help='.mst or .iso file to read') + parser.add_argument( + '-o', '--out', type=argparse.FileType('w'), default=sys.stdout, + metavar='OUTPUT.json', + help='the file where the JSON output should be written' + ' (default: write to stdout)') + parser.add_argument( + '-c', '--couch', action='store_true', + help='output array within a "docs" item in a JSON document' + ' for bulk insert to CouchDB via POST to db/_bulk_docs') + parser.add_argument( + '-m', '--mongo', action='store_true', + help='output individual records as separate JSON dictionaries, one' + ' per line for bulk insert to MongoDB via mongoimport utility') + parser.add_argument( + '-t', '--type', type=int, metavar='ISIS_JSON_TYPE', default=1, + help='ISIS-JSON type, sets field structure: 1=string, 2=alist,' + ' 3=dict (default=1)') + parser.add_argument( + '-q', '--qty', type=int, default=DEFAULT_QTY, + help='maximum quantity of records to read (default=ALL)') + parser.add_argument( + '-s', '--skip', type=int, default=0, + help='records to skip from start of .mst (default=0)') + parser.add_argument( + '-i', '--id', type=int, metavar='TAG_NUMBER', default=0, + help='generate an "_id" from the given unique TAG field number' + ' for each record') + parser.add_argument( + '-u', '--uuid', action='store_true', + help='generate an "_id" with a random UUID for each record') + parser.add_argument( + '-p', '--prefix', type=str, metavar='PREFIX', default='', + help='concatenate prefix to every numeric field tag' + ' (ex. 99 becomes "v99")') + parser.add_argument( + '-n', '--mfn', action='store_true', + help='generate an "_id" from the MFN of each record' + ' (available only for .mst input)') + parser.add_argument( + '-k', '--constant', type=str, metavar='TAG:VALUE', default='', + help='Include a constant tag:value in every record (ex. -k type:AS)') + + ''' + # TODO: implement this to export large quantities of records to CouchDB + parser.add_argument( + '-r', '--repeat', type=int, default=1, + help='repeat operation, saving multiple JSON files' + ' (default=1, use -r 0 to repeat until end of input)') + ''' + # parse the command line + args = parser.parse_args() + if args.file_name.lower().endswith('.mst'): + input_gen_func = iter_mst_records # <5> + else: + if args.mfn: + print('UNSUPORTED: -n/--mfn option only available for .mst input.') + raise SystemExit + input_gen_func = iter_iso_records # <6> + input_gen = input_gen_func(args.file_name, args.type) # <7> + if args.couch: + args.out.write('{ "docs" : ') + write_json(input_gen, args.file_name, args.out, args.qty, # <8> + args.skip, args.id, args.uuid, args.mongo, args.mfn, + args.type, args.prefix, args.constant) + if args.couch: + args.out.write('}\n') + args.out.close() + + +if __name__ == '__main__': + main() +# end::ISIS2JSON[] diff --git a/appendix-A/mem_test.py b/appendix-A/mem_test.py new file mode 100644 index 0000000..ecf6c04 --- /dev/null +++ b/appendix-A/mem_test.py @@ -0,0 +1,24 @@ +import importlib +import sys +import resource + +NUM_VECTORS = 10**7 + +if len(sys.argv) == 2: + module_name = sys.argv[1].replace('.py', '') + module = importlib.import_module(module_name) +else: + print('Usage: {} '.format()) + sys.exit(1) + +fmt = 'Selected Vector2d type: {.__name__}.{.__name__}' +print(fmt.format(module, module.Vector2d)) + +mem_init = resource.getrusage(resource.RUSAGE_SELF).ru_maxrss +print('Creating {:,} Vector2d instances'.format(NUM_VECTORS)) + +vectors = [module.Vector2d(3.0, 4.0) for i in range(NUM_VECTORS)] + +mem_final = resource.getrusage(resource.RUSAGE_SELF).ru_maxrss +print('Initial RAM usage: {:14,}'.format(mem_init)) +print(' Final RAM usage: {:14,}'.format(mem_final)) diff --git a/appendix-A/schedule2.py b/appendix-A/schedule2.py new file mode 100644 index 0000000..d1d3019 --- /dev/null +++ b/appendix-A/schedule2.py @@ -0,0 +1,132 @@ +""" +schedule2.py: traversing OSCON schedule data + + >>> import shelve + >>> db = shelve.open(DB_NAME) + >>> if CONFERENCE not in db: load_db(db) + +# BEGIN SCHEDULE2_DEMO + + >>> DbRecord.set_db(db) + >>> event = DbRecord.fetch('event.33950') + >>> event + + >>> event.venue + + >>> event.venue.name + 'Portland 251' + >>> for spkr in event.speakers: + ... print('{0.serial}: {0.name}'.format(spkr)) + ... + speaker.3471: Anna Martelli Ravenscroft + speaker.5199: Alex Martelli + +# END SCHEDULE2_DEMO + + >>> db.close() + +""" + +# BEGIN SCHEDULE2_RECORD +import warnings +import inspect + +import osconfeed + +DB_NAME = 'data/schedule2_db' +CONFERENCE = 'conference.115' + + +class Record: + def __init__(self, **kwargs): + self.__dict__.update(kwargs) + + def __eq__(self, other): + if isinstance(other, Record): + return self.__dict__ == other.__dict__ + else: + return NotImplemented +# END SCHEDULE2_RECORD + + +# BEGIN SCHEDULE2_DBRECORD +class MissingDatabaseError(RuntimeError): + """Raised when a database is required but was not set.""" + + +class DbRecord(Record): + + __db = None + + @staticmethod + def set_db(db): + DbRecord.__db = db + + @staticmethod + def get_db(): + return DbRecord.__db + + @classmethod + def fetch(cls, ident): + db = cls.get_db() + try: + return db[ident] + except TypeError: + if db is None: + msg = "database not set; call '{}.set_db(my_db)'" + raise MissingDatabaseError(msg.format(cls.__name__)) + else: + raise + + def __repr__(self): + if hasattr(self, 'serial'): + cls_name = self.__class__.__name__ + return '<{} serial={!r}>'.format(cls_name, self.serial) + else: + return super().__repr__() +# END SCHEDULE2_DBRECORD + + +# BEGIN SCHEDULE2_EVENT +class Event(DbRecord): + + @property + def venue(self): + key = 'venue.{}'.format(self.venue_serial) + return self.__class__.fetch(key) + + @property + def speakers(self): + if not hasattr(self, '_speaker_objs'): + spkr_serials = self.__dict__['speakers'] + fetch = self.__class__.fetch + self._speaker_objs = [fetch('speaker.{}'.format(key)) + for key in spkr_serials] + return self._speaker_objs + + def __repr__(self): + if hasattr(self, 'name'): + cls_name = self.__class__.__name__ + return '<{} {!r}>'.format(cls_name, self.name) + else: + return super().__repr__() +# END SCHEDULE2_EVENT + + +# BEGIN SCHEDULE2_LOAD +def load_db(db): + raw_data = osconfeed.load() + warnings.warn('loading ' + DB_NAME) + for collection, rec_list in raw_data['Schedule'].items(): + record_type = collection[:-1] + cls_name = record_type.capitalize() + cls = globals().get(cls_name, DbRecord) + if inspect.isclass(cls) and issubclass(cls, DbRecord): + factory = cls + else: + factory = DbRecord + for record in rec_list: + key = '{}.{}'.format(record_type, record['serial']) + record['serial'] = key + db[key] = factory(**record) +# END SCHEDULE2_LOAD diff --git a/appendix-A/sha_futures.py b/appendix-A/sha_futures.py new file mode 100644 index 0000000..594c6aa --- /dev/null +++ b/appendix-A/sha_futures.py @@ -0,0 +1,38 @@ +import sys +import time +import hashlib +from concurrent import futures +from random import randrange + +JOBS = 12 +SIZE = 2**20 +STATUS = '{} workers, elapsed time: {:.2f}s' + + +def sha(size): + data = bytearray(randrange(256) for i in range(size)) + algo = hashlib.new('sha256') + algo.update(data) + return algo.hexdigest() + + +def main(workers=None): + if workers: + workers = int(workers) + t0 = time.time() + + with futures.ProcessPoolExecutor(workers) as executor: + actual_workers = executor._max_workers + to_do = (executor.submit(sha, SIZE) for i in range(JOBS)) + for future in futures.as_completed(to_do): + res = future.result() + print(res) + + print(STATUS.format(actual_workers, time.time() - t0)) + +if __name__ == '__main__': + if len(sys.argv) == 2: + workers = int(sys.argv[1]) + else: + workers = None + main(workers) diff --git a/appendix-A/taxi_sim.py b/appendix-A/taxi_sim.py new file mode 100644 index 0000000..63eef88 --- /dev/null +++ b/appendix-A/taxi_sim.py @@ -0,0 +1,203 @@ + +""" +Taxi simulator +============== + +Driving a taxi from the console:: + + >>> from taxi_sim import taxi_process + >>> taxi = taxi_process(ident=13, trips=2, start_time=0) + >>> next(taxi) + Event(time=0, proc=13, action='leave garage') + >>> taxi.send(_.time + 7) + Event(time=7, proc=13, action='pick up passenger') + >>> taxi.send(_.time + 23) + Event(time=30, proc=13, action='drop off passenger') + >>> taxi.send(_.time + 5) + Event(time=35, proc=13, action='pick up passenger') + >>> taxi.send(_.time + 48) + Event(time=83, proc=13, action='drop off passenger') + >>> taxi.send(_.time + 1) + Event(time=84, proc=13, action='going home') + >>> taxi.send(_.time + 10) + Traceback (most recent call last): + File "", line 1, in + StopIteration + +Sample run with two cars, random seed 10. This is a valid doctest:: + + >>> main(num_taxis=2, seed=10) + taxi: 0 Event(time=0, proc=0, action='leave garage') + taxi: 0 Event(time=5, proc=0, action='pick up passenger') + taxi: 1 Event(time=5, proc=1, action='leave garage') + taxi: 1 Event(time=10, proc=1, action='pick up passenger') + taxi: 1 Event(time=15, proc=1, action='drop off passenger') + taxi: 0 Event(time=17, proc=0, action='drop off passenger') + taxi: 1 Event(time=24, proc=1, action='pick up passenger') + taxi: 0 Event(time=26, proc=0, action='pick up passenger') + taxi: 0 Event(time=30, proc=0, action='drop off passenger') + taxi: 0 Event(time=34, proc=0, action='going home') + taxi: 1 Event(time=46, proc=1, action='drop off passenger') + taxi: 1 Event(time=48, proc=1, action='pick up passenger') + taxi: 1 Event(time=110, proc=1, action='drop off passenger') + taxi: 1 Event(time=139, proc=1, action='pick up passenger') + taxi: 1 Event(time=140, proc=1, action='drop off passenger') + taxi: 1 Event(time=150, proc=1, action='going home') + *** end of events *** + +See longer sample run at the end of this module. + +""" + +import random +import collections +import queue +import argparse +import time + +DEFAULT_NUMBER_OF_TAXIS = 3 +DEFAULT_END_TIME = 180 +SEARCH_DURATION = 5 +TRIP_DURATION = 20 +DEPARTURE_INTERVAL = 5 + +Event = collections.namedtuple('Event', 'time proc action') + + +# BEGIN TAXI_PROCESS +def taxi_process(ident, trips, start_time=0): + """Yield to simulator issuing event at each state change""" + time = yield Event(start_time, ident, 'leave garage') + for i in range(trips): + time = yield Event(time, ident, 'pick up passenger') + time = yield Event(time, ident, 'drop off passenger') + + yield Event(time, ident, 'going home') + # end of taxi process +# END TAXI_PROCESS + + +# BEGIN TAXI_SIMULATOR +class Simulator: + + def __init__(self, procs_map): + self.events = queue.PriorityQueue() + self.procs = dict(procs_map) + + def run(self, end_time): + """Schedule and display events until time is up""" + # schedule the first event for each cab + for _, proc in sorted(self.procs.items()): + first_event = next(proc) + self.events.put(first_event) + + # main loop of the simulation + sim_time = 0 + while sim_time < end_time: + if self.events.empty(): + print('*** end of events ***') + break + + current_event = self.events.get() + sim_time, proc_id, previous_action = current_event + print('taxi:', proc_id, proc_id * ' ', current_event) + active_proc = self.procs[proc_id] + next_time = sim_time + compute_duration(previous_action) + try: + next_event = active_proc.send(next_time) + except StopIteration: + del self.procs[proc_id] + else: + self.events.put(next_event) + else: + msg = '*** end of simulation time: {} events pending ***' + print(msg.format(self.events.qsize())) +# END TAXI_SIMULATOR + + +def compute_duration(previous_action): + """Compute action duration using exponential distribution""" + if previous_action in ['leave garage', 'drop off passenger']: + # new state is prowling + interval = SEARCH_DURATION + elif previous_action == 'pick up passenger': + # new state is trip + interval = TRIP_DURATION + elif previous_action == 'going home': + interval = 1 + else: + raise ValueError('Unknown previous_action: %s' % previous_action) + return int(random.expovariate(1/interval)) + 1 + + +def main(end_time=DEFAULT_END_TIME, num_taxis=DEFAULT_NUMBER_OF_TAXIS, + seed=None): + """Initialize random generator, build procs and run simulation""" + if seed is not None: + random.seed(seed) # get reproducible results + + taxis = {i: taxi_process(i, (i+1)*2, i*DEPARTURE_INTERVAL) + for i in range(num_taxis)} + sim = Simulator(taxis) + sim.run(end_time) + + +if __name__ == '__main__': + + parser = argparse.ArgumentParser( + description='Taxi fleet simulator.') + parser.add_argument('-e', '--end-time', type=int, + default=DEFAULT_END_TIME, + help='simulation end time; default = %s' + % DEFAULT_END_TIME) + parser.add_argument('-t', '--taxis', type=int, + default=DEFAULT_NUMBER_OF_TAXIS, + help='number of taxis running; default = %s' + % DEFAULT_NUMBER_OF_TAXIS) + parser.add_argument('-s', '--seed', type=int, default=None, + help='random generator seed (for testing)') + + args = parser.parse_args() + main(args.end_time, args.taxis, args.seed) + + +""" + +Sample run from the command line, seed=3, maximum elapsed time=120:: + +# BEGIN TAXI_SAMPLE_RUN +$ python3 taxi_sim.py -s 3 -e 120 +taxi: 0 Event(time=0, proc=0, action='leave garage') +taxi: 0 Event(time=2, proc=0, action='pick up passenger') +taxi: 1 Event(time=5, proc=1, action='leave garage') +taxi: 1 Event(time=8, proc=1, action='pick up passenger') +taxi: 2 Event(time=10, proc=2, action='leave garage') +taxi: 2 Event(time=15, proc=2, action='pick up passenger') +taxi: 2 Event(time=17, proc=2, action='drop off passenger') +taxi: 0 Event(time=18, proc=0, action='drop off passenger') +taxi: 2 Event(time=18, proc=2, action='pick up passenger') +taxi: 2 Event(time=25, proc=2, action='drop off passenger') +taxi: 1 Event(time=27, proc=1, action='drop off passenger') +taxi: 2 Event(time=27, proc=2, action='pick up passenger') +taxi: 0 Event(time=28, proc=0, action='pick up passenger') +taxi: 2 Event(time=40, proc=2, action='drop off passenger') +taxi: 2 Event(time=44, proc=2, action='pick up passenger') +taxi: 1 Event(time=55, proc=1, action='pick up passenger') +taxi: 1 Event(time=59, proc=1, action='drop off passenger') +taxi: 0 Event(time=65, proc=0, action='drop off passenger') +taxi: 1 Event(time=65, proc=1, action='pick up passenger') +taxi: 2 Event(time=65, proc=2, action='drop off passenger') +taxi: 2 Event(time=72, proc=2, action='pick up passenger') +taxi: 0 Event(time=76, proc=0, action='going home') +taxi: 1 Event(time=80, proc=1, action='drop off passenger') +taxi: 1 Event(time=88, proc=1, action='pick up passenger') +taxi: 2 Event(time=95, proc=2, action='drop off passenger') +taxi: 2 Event(time=97, proc=2, action='pick up passenger') +taxi: 2 Event(time=98, proc=2, action='drop off passenger') +taxi: 1 Event(time=106, proc=1, action='drop off passenger') +taxi: 2 Event(time=109, proc=2, action='going home') +taxi: 1 Event(time=110, proc=1, action='going home') +*** end of events *** +# END TAXI_SAMPLE_RUN + +""" diff --git a/appendix-A/test_schedule1.py b/appendix-A/test_schedule1.py new file mode 100644 index 0000000..dbaacc9 --- /dev/null +++ b/appendix-A/test_schedule1.py @@ -0,0 +1,37 @@ +import shelve +import pytest + +import schedule1 as schedule + + +@pytest.yield_fixture +def db(): + with shelve.open(schedule.DB_NAME) as the_db: + if schedule.CONFERENCE not in the_db: + schedule.load_db(the_db) + yield the_db + + +def test_record_class(): + rec = schedule.Record(spam=99, eggs=12) + assert rec.spam == 99 + assert rec.eggs == 12 + + +def test_conference_record(db): + assert schedule.CONFERENCE in db + + +def test_speaker_record(db): + speaker = db['speaker.3471'] + assert speaker.name == 'Anna Martelli Ravenscroft' + + +def test_event_record(db): + event = db['event.33950'] + assert event.name == 'There *Will* Be Bugs' + + +def test_event_venue(db): + event = db['event.33950'] + assert event.venue_serial == 1449 diff --git a/appendix-A/test_schedule2.py b/appendix-A/test_schedule2.py new file mode 100644 index 0000000..de09d32 --- /dev/null +++ b/appendix-A/test_schedule2.py @@ -0,0 +1,72 @@ +import shelve +import pytest + +import schedule2 as schedule + + +@pytest.yield_fixture +def db(): + with shelve.open(schedule.DB_NAME) as the_db: + if schedule.CONFERENCE not in the_db: + schedule.load_db(the_db) + yield the_db + + +def test_record_attr_access(): + rec = schedule.Record(spam=99, eggs=12) + assert rec.spam == 99 + assert rec.eggs == 12 + + +def test_record_repr(): + rec = schedule.DbRecord(spam=99, eggs=12) + assert 'DbRecord object at 0x' in repr(rec) + rec2 = schedule.DbRecord(serial=13) + assert repr(rec2) == "" + + +def test_conference_record(db): + assert schedule.CONFERENCE in db + + +def test_speaker_record(db): + speaker = db['speaker.3471'] + assert speaker.name == 'Anna Martelli Ravenscroft' + + +def test_missing_db_exception(): + with pytest.raises(schedule.MissingDatabaseError): + schedule.DbRecord.fetch('venue.1585') + + +def test_dbrecord(db): + schedule.DbRecord.set_db(db) + venue = schedule.DbRecord.fetch('venue.1585') + assert venue.name == 'Exhibit Hall B' + + +def test_event_record(db): + event = db['event.33950'] + assert repr(event) == "" + + +def test_event_venue(db): + schedule.Event.set_db(db) + event = db['event.33950'] + assert event.venue_serial == 1449 + assert event.venue == db['venue.1449'] + assert event.venue.name == 'Portland 251' + + +def test_event_speakers(db): + schedule.Event.set_db(db) + event = db['event.33950'] + assert len(event.speakers) == 2 + anna_and_alex = [db['speaker.3471'], db['speaker.5199']] + assert event.speakers == anna_and_alex + + +def test_event_no_speakers(db): + schedule.Event.set_db(db) + event = db['event.36848'] + assert len(event.speakers) == 0