https://github.com/chatopera/Synonyms/issues/112 优化词向量，扩大词汇表，加快下载速度
This commit is contained in:
Hai Liang Wang
parent
1fdfaaea40
commit
0dbe1ec7cd
2
.gitignore
vendored
2
.gitignore
vendored
@ -12,4 +12,4 @@ synonyms.egg-info
|
||||
.vscode/
|
||||
build/
|
||||
.env
|
||||
synonyms/data/words.vector
|
||||
synonyms/data/words.vector*
|
||||
|
||||
2
LICENSE
2
LICENSE
@ -1,4 +1,4 @@
|
||||
Copyright (2018-2020) Hu Ying Xi<>, Hai Liang Wang<hain@chatopera.com>
|
||||
Copyright (2018-2020) Chatopera Inc. <https://www.chatopera.com>
|
||||
|
||||
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
|
||||
|
||||
|
||||
48
README.md
48
README.md
@ -26,6 +26,8 @@ pip install -U synonyms
|
||||
|
||||
兼容 py2 和 py3，当前稳定版本 [v3.x](https://github.com/chatopera/Synonyms/releases)。
|
||||
|
||||
**提示：安装后初次使用会下载词向量文件，下载速度取决于网络情况。**
|
||||
|
||||
|
||||
|
||||
**Node.js 用户可以使用 [node-synonyms](https://www.npmjs.com/package/node-synonyms)了。**
|
||||
@ -80,7 +82,7 @@ synonyms.nearby(人脸, 10) = (
|
||||
095, 0.525344, 0.524009, 0.523101, 0.516046])
|
||||
```
|
||||
|
||||
在 OOV 的情况下，返回 `([], [])`，目前的字典大小: 125,792。
|
||||
在 OOV 的情况下，返回 `([], [])`，目前的字典大小: 435,729。
|
||||
|
||||
### synonyms#compare
|
||||
|
||||
@ -107,16 +109,16 @@ synonyms.nearby(人脸, 10) = (
|
||||
```
|
||||
>>> synonyms.display("飞机")
|
||||
'飞机'近义词：
|
||||
1. 架飞机:0.837399
|
||||
2. 客机:0.764609
|
||||
3. 直升机:0.762116
|
||||
4. 民航机:0.750519
|
||||
5. 航机:0.750116
|
||||
6. 起飞:0.735736
|
||||
7. 战机:0.734975
|
||||
8. 飞行中:0.732649
|
||||
9. 航空器:0.723945
|
||||
10. 运输机:0.720578
|
||||
1. 飞机:1.0
|
||||
2. 直升机:0.8423391
|
||||
3. 客机:0.8393003
|
||||
4. 滑翔机:0.7872388
|
||||
5. 军用飞机:0.7832081
|
||||
6. 水上飞机:0.77857226
|
||||
7. 运输机:0.7724742
|
||||
8. 航机:0.7664748
|
||||
9. 航空器:0.76592904
|
||||
10. 民航机:0.74209654
|
||||
```
|
||||
|
||||
`SIZE` 是打印词汇表的数量，默认 10。
|
||||
@ -182,7 +184,7 @@ HowNet，也被称为知网，它并不只是一个语义字典，而是一个
|
||||
|
||||
### 对比
|
||||
|
||||
Synonyms 的词表容量是 125,792，下面选择一些在同义词词林、知网和 Synonyms 都存在的几个词，给出其近似度的对比：
|
||||
Synonyms 的词表容量是 435,729，下面选择一些在同义词词林、知网和 Synonyms 都存在的几个词，给出其近似度的对比：
|
||||
|
||||
|
||||
|
||||
@ -190,6 +192,12 @@ Synonyms 的词表容量是 125,792，下面选择一些在同义词词林、知
|
||||
|
||||
更多[比对结果](./VALUATION.md)。
|
||||
|
||||
## Used by
|
||||
|
||||
[Github 关联用户列表](https://github.com/chatopera/Synonyms/network/dependents?package_id=UGFja2FnZS01MjY2NDc1Nw%3D%3D)
|
||||
|
||||
|
||||
|
||||
## Benchmark
|
||||
|
||||
Test with py3, MacBook Pro.
|
||||
@ -242,7 +250,7 @@ meminfo 8GB
|
||||
|
||||
# Promotion
|
||||
|
||||
[Chatopera 云服务](https://bot.chatopera.com/dashboard) 是面向企业聊天机器人构建的一站式解决方案，融合信息检索系统、机器学习、聊天机器人脚本语法和语音识别等技术，为定制化聊天机器人和自然语言交互而生！
|
||||
[Chatopera 云服务](https://bot.chatopera.com/dashboard)
|
||||
|
||||
<p align="center">
|
||||
<b>Chatopera 云服务</b><br>
|
||||
@ -251,6 +259,8 @@ meminfo 8GB
|
||||
</a>
|
||||
</p>
|
||||
|
||||
Chatopera 机器人平台包括知识库、多轮对话、意图识别和语音识别等组件，标准化聊天机器人开发，支持企业 OA 智能问答、HR 智能问答、智能客服和网络营销等场景；一站式实现聊天机器人，按量付费，让聊天机器人上线！
|
||||
|
||||
# References
|
||||
|
||||
[wikidata-corpus](https://github.com/Samurais/wikidata-corpus)
|
||||
@ -273,9 +283,9 @@ Google 发布的[word2vec](https://code.google.com/archive/p/word2vec/)，该库
|
||||
|
||||
# Authors
|
||||
|
||||
[Hai Liang Wang](http://blog.chatbot.io/webcv/)
|
||||
[Hai Liang Wang](https://pre-angel.com/peoples/hailiang-wang/)
|
||||
|
||||
[Hu Ying Xi](https://github.com/chatopera/)
|
||||
[Hu Ying Xi](https://github.com/huyingxi)
|
||||
|
||||
# Give credits to
|
||||
|
||||
@ -293,6 +303,14 @@ Google 发布的[word2vec](https://code.google.com/archive/p/word2vec/)，该库
|
||||
|
||||
[MIT](./LICENSE)
|
||||
|
||||
Copyright (2018-2020) Chatopera Inc. <https://www.chatopera.com>
|
||||
|
||||
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
|
||||
|
||||
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
|
||||
|
||||
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.OF
|
||||
|
||||
[![chatoper banner][co-banner-image]][co-url]
|
||||
|
||||
[co-banner-image]: https://user-images.githubusercontent.com/3538629/42383104-da925942-8168-11e8-8195-868d5fcec170.png
|
||||
|
||||
@ -1,6 +1,7 @@
|
||||
# synonyms 分数评测 [(v3.11.0)](https://pypi.python.org/pypi/synonyms/3.11.0)
|
||||
# synonyms 分数评测 [(v3.12.0)](https://pypi.python.org/pypi/synonyms/3.12.0)
|
||||
|
||||
| 词 1 | 词 2 | synonyms | 人工评定 |
|
||||
| --- | --- | --- | --- |
|
||||
| ------ | -------- | -------- | -------- |
|
||||
| 轿车 | 汽车 | 0.892 | 0.98 |
|
||||
| 宝石 | 宝物 | 1.0 | 0.96 |
|
||||
| 旅游 | 游历 | 0.649 | 0.96 |
|
||||
|
||||
BIN
assets/6.png
Normal file
BIN
assets/6.png
Normal file
Binary file not shown.
|
After Width: | Height: | Size: 185 KiB |
@ -15,7 +15,7 @@ export PATH=/opt/miniconda3/envs/venv-py3/bin:$PATH
|
||||
cd $baseDir/..
|
||||
if [ -f .env ]; then
|
||||
echo "load env with" `pwd`"/.env"
|
||||
source .env
|
||||
#source .env
|
||||
fi
|
||||
|
||||
python demo.py
|
||||
|
||||
11
setup.py
11
setup.py
@ -4,20 +4,19 @@ LONGDOC = """
|
||||
Synonyms
|
||||
=====================
|
||||
|
||||
Chinese Synonyms for Natural Language Processing and Understanding.
|
||||
中文近义词
|
||||
|
||||
Welcome
|
||||
-------
|
||||
https://github.com/chatopera/Synonyms
|
||||
|
||||
"""
|
||||
|
||||
setup(
|
||||
name='synonyms',
|
||||
version='3.11.0',
|
||||
version='3.12.0',
|
||||
description='中文近义词：聊天机器人，智能问答工具包；Chinese Synonyms for Natural Language Processing and Understanding',
|
||||
long_description=LONGDOC,
|
||||
author='Hai Liang Wang, Hu Ying Xi',
|
||||
author_email='hailiang.hl.wang@gmail.com',
|
||||
author_email='hain@chatopera.com',
|
||||
url='https://github.com/chatopera/Synonyms',
|
||||
license="MIT",
|
||||
classifiers=[
|
||||
@ -32,6 +31,7 @@ setup(
|
||||
'Programming Language :: Python :: 3',
|
||||
'Programming Language :: Python :: 3.5',
|
||||
'Programming Language :: Python :: 3.6',
|
||||
'Programming Language :: Python :: 3.7',
|
||||
'Topic :: Text Processing',
|
||||
'Topic :: Text Processing :: Indexing',
|
||||
'Topic :: Text Processing :: Linguistic'],
|
||||
@ -48,5 +48,4 @@ setup(
|
||||
'synonyms': [
|
||||
'**/*.gz',
|
||||
'**/*.txt',
|
||||
'**/*.vector',
|
||||
'LICENSE']})
|
||||
|
||||
@ -20,7 +20,7 @@ from __future__ import division
|
||||
__copyright__ = "Copyright (c) (2017-2020) Chatopera Inc. All Rights Reserved"
|
||||
__author__ = "Hu Ying Xi<>, Hai Liang Wang<hailiang.hl.wang@gmail.com>"
|
||||
__date__ = "2017-09-27"
|
||||
__version__ = "3.11.0"
|
||||
__version__ = "3.12.0"
|
||||
|
||||
import os
|
||||
import sys
|
||||
@ -56,6 +56,7 @@ from .utils import cosine
|
||||
from .utils import is_digit
|
||||
import jieba
|
||||
from .jieba import posseg as _tokenizer
|
||||
import wget
|
||||
|
||||
'''
|
||||
globals
|
||||
@ -119,19 +120,28 @@ def _segment_words(sen):
|
||||
word embedding
|
||||
'''
|
||||
# vectors
|
||||
_f_model = os.path.join(curdir, 'data', 'words.vector')
|
||||
_f_url = os.environ.get("SYNONYMS_WORD2VEC_BIN_URL_ZH_CN", "https://static-public.chatopera.com/ml/synonyms/words.vector.gz")
|
||||
_f_model = os.path.join(curdir, 'data', 'words.vector.gz')
|
||||
_download_model = not os.path.exists(_f_model)
|
||||
if "SYNONYMS_WORD2VEC_BIN_MODEL_ZH_CN" in ENVIRON:
|
||||
_f_model = ENVIRON["SYNONYMS_WORD2VEC_BIN_MODEL_ZH_CN"]
|
||||
_download_model = False
|
||||
|
||||
def _load_w2v(model_file=_f_model, binary=True):
|
||||
'''
|
||||
load word2vec model
|
||||
'''
|
||||
if not os.path.exists(model_file):
|
||||
print("os.path : ", os.path)
|
||||
if not os.path.exists(model_file) and _download_model:
|
||||
print("\n[Synonyms] downloading data from %s to %s ... \n this only happens if SYNONYMS_WORD2VEC_BIN_URL_ZH_CN is not present and Synonyms initialization for the first time. \n It would take minutes that depends on network." % (_f_url, model_file))
|
||||
wget.download(_f_url, out = model_file)
|
||||
print("\n[Synonyms] download is done.\n")
|
||||
elif not os.path.exists(model_file):
|
||||
print("[Synonyms] os.path : ", os.path)
|
||||
raise Exception("Model file [%s] does not exist." % model_file)
|
||||
|
||||
return KeyedVectors.load_word2vec_format(
|
||||
model_file, binary=binary, unicode_errors='ignore')
|
||||
print(">> Synonyms on loading vectors [%s] ..." % _f_model)
|
||||
print("[Synonyms] on loading vectors [%s] ..." % _f_model)
|
||||
_vectors = _load_w2v(model_file=_f_model)
|
||||
|
||||
def _get_wv(sentence, ignore=False):
|
||||
|
||||
402
synonyms/wget.py
Normal file
402
synonyms/wget.py
Normal file
@ -0,0 +1,402 @@
|
||||
#!/usr/bin/env python
|
||||
"""
|
||||
Download utility as an easy way to get file from the net
|
||||
|
||||
python -m wget <URL>
|
||||
python wget.py <URL>
|
||||
|
||||
Downloads: http://pypi.python.org/pypi/wget/
|
||||
Development: http://bitbucket.org/techtonik/python-wget/
|
||||
|
||||
wget.py is not option compatible with Unix wget utility,
|
||||
to make command line interface intuitive for new people.
|
||||
|
||||
Public domain by anatoly techtonik <techtonik@gmail.com>
|
||||
Also available under the terms of MIT license
|
||||
Copyright (c) 2010-2014 anatoly techtonik
|
||||
"""
|
||||
|
||||
|
||||
import sys, shutil, os
|
||||
import tempfile
|
||||
import math
|
||||
|
||||
PY3K = sys.version_info >= (3, 0)
|
||||
if PY3K:
|
||||
import urllib.request as urllib
|
||||
import urllib.parse as urlparse
|
||||
else:
|
||||
import urllib
|
||||
import urlparse
|
||||
|
||||
|
||||
__version__ = "2.3-beta1"
|
||||
|
||||
|
||||
def filename_from_url(url):
|
||||
""":return: detected filename or None"""
|
||||
fname = os.path.basename(urlparse.urlparse(url).path)
|
||||
if len(fname.strip(" \n\t.")) == 0:
|
||||
return None
|
||||
return fname
|
||||
|
||||
def filename_from_headers(headers):
|
||||
"""Detect filename from Content-Disposition headers if present.
|
||||
http://greenbytes.de/tech/tc2231/
|
||||
|
||||
:param: headers as dict, list or string
|
||||
:return: filename from content-disposition header or None
|
||||
"""
|
||||
if type(headers) == str:
|
||||
headers = headers.splitlines()
|
||||
if type(headers) == list:
|
||||
headers = dict([x.split(':', 1) for x in headers])
|
||||
cdisp = headers.get("Content-Disposition")
|
||||
if not cdisp:
|
||||
return None
|
||||
cdtype = cdisp.split(';')
|
||||
if len(cdtype) == 1:
|
||||
return None
|
||||
if cdtype[0].strip().lower() not in ('inline', 'attachment'):
|
||||
return None
|
||||
# several filename params is illegal, but just in case
|
||||
fnames = [x for x in cdtype[1:] if x.strip().startswith('filename=')]
|
||||
if len(fnames) > 1:
|
||||
return None
|
||||
name = fnames[0].split('=')[1].strip(' \t"')
|
||||
name = os.path.basename(name)
|
||||
if not name:
|
||||
return None
|
||||
return name
|
||||
|
||||
def filename_fix_existing(filename):
|
||||
"""Expands name portion of filename with numeric ' (x)' suffix to
|
||||
return filename that doesn't exist already.
|
||||
"""
|
||||
dirname = '.'
|
||||
name, ext = filename.rsplit('.', 1)
|
||||
names = [x for x in os.listdir(dirname) if x.startswith(name)]
|
||||
names = [x.rsplit('.', 1)[0] for x in names]
|
||||
suffixes = [x.replace(name, '') for x in names]
|
||||
# filter suffixes that match ' (x)' pattern
|
||||
suffixes = [x[2:-1] for x in suffixes
|
||||
if x.startswith(' (') and x.endswith(')')]
|
||||
indexes = [int(x) for x in suffixes
|
||||
if set(x) <= set('0123456789')]
|
||||
idx = 1
|
||||
if indexes:
|
||||
idx += sorted(indexes)[-1]
|
||||
return '%s (%d).%s' % (name, idx, ext)
|
||||
|
||||
|
||||
# --- terminal/console output helpers ---
|
||||
|
||||
def get_console_width():
|
||||
"""Return width of available window area. Autodetection works for
|
||||
Windows and POSIX platforms. Returns 80 for others
|
||||
|
||||
Code from http://bitbucket.org/techtonik/python-pager
|
||||
"""
|
||||
|
||||
if os.name == 'nt':
|
||||
STD_INPUT_HANDLE = -10
|
||||
STD_OUTPUT_HANDLE = -11
|
||||
STD_ERROR_HANDLE = -12
|
||||
|
||||
# get console handle
|
||||
from ctypes import windll, Structure, byref
|
||||
try:
|
||||
from ctypes.wintypes import SHORT, WORD, DWORD
|
||||
except ImportError:
|
||||
# workaround for missing types in Python 2.5
|
||||
from ctypes import (
|
||||
c_short as SHORT, c_ushort as WORD, c_ulong as DWORD)
|
||||
console_handle = windll.kernel32.GetStdHandle(STD_OUTPUT_HANDLE)
|
||||
|
||||
# CONSOLE_SCREEN_BUFFER_INFO Structure
|
||||
class COORD(Structure):
|
||||
_fields_ = [("X", SHORT), ("Y", SHORT)]
|
||||
|
||||
class SMALL_RECT(Structure):
|
||||
_fields_ = [("Left", SHORT), ("Top", SHORT),
|
||||
("Right", SHORT), ("Bottom", SHORT)]
|
||||
|
||||
class CONSOLE_SCREEN_BUFFER_INFO(Structure):
|
||||
_fields_ = [("dwSize", COORD),
|
||||
("dwCursorPosition", COORD),
|
||||
("wAttributes", WORD),
|
||||
("srWindow", SMALL_RECT),
|
||||
("dwMaximumWindowSize", DWORD)]
|
||||
|
||||
sbi = CONSOLE_SCREEN_BUFFER_INFO()
|
||||
ret = windll.kernel32.GetConsoleScreenBufferInfo(console_handle, byref(sbi))
|
||||
if ret == 0:
|
||||
return 0
|
||||
return sbi.srWindow.Right+1
|
||||
|
||||
elif os.name == 'posix':
|
||||
from fcntl import ioctl
|
||||
from termios import TIOCGWINSZ
|
||||
from array import array
|
||||
|
||||
winsize = array("H", [0] * 4)
|
||||
try:
|
||||
ioctl(sys.stdout.fileno(), TIOCGWINSZ, winsize)
|
||||
except IOError:
|
||||
pass
|
||||
return (winsize[1], winsize[0])[0]
|
||||
|
||||
return 80
|
||||
|
||||
|
||||
def bar_thermometer(current, total, width=80):
|
||||
"""Return thermometer style progress bar string. `total` argument
|
||||
can not be zero. The minimum size of bar returned is 3. Example:
|
||||
|
||||
[.......... ]
|
||||
|
||||
Control and trailing symbols (\r and spaces) are not included.
|
||||
See `bar_adaptive` for more information.
|
||||
"""
|
||||
# number of dots on thermometer scale
|
||||
avail_dots = width-2
|
||||
shaded_dots = int(math.floor(float(current) / total * avail_dots))
|
||||
return '[' + '.'*shaded_dots + ' '*(avail_dots-shaded_dots) + ']'
|
||||
|
||||
def bar_adaptive(current, total, width=80):
|
||||
"""Return progress bar string for given values in one of three
|
||||
styles depending on available width:
|
||||
|
||||
[.. ] downloaded / total
|
||||
downloaded / total
|
||||
[.. ]
|
||||
|
||||
if total value is unknown or <= 0, show bytes counter using two
|
||||
adaptive styles:
|
||||
|
||||
%s / unknown
|
||||
%s
|
||||
|
||||
if there is not enough space on the screen, do not display anything
|
||||
|
||||
returned string doesn't include control characters like \r used to
|
||||
place cursor at the beginning of the line to erase previous content.
|
||||
|
||||
this function leaves one free character at the end of string to
|
||||
avoid automatic linefeed on Windows.
|
||||
"""
|
||||
|
||||
# process special case when total size is unknown and return immediately
|
||||
if not total or total < 0:
|
||||
msg = "%s / unknown" % current
|
||||
if len(msg) < width: # leaves one character to avoid linefeed
|
||||
return msg
|
||||
if len("%s" % current) < width:
|
||||
return "%s" % current
|
||||
|
||||
# --- adaptive layout algorithm ---
|
||||
#
|
||||
# [x] describe the format of the progress bar
|
||||
# [x] describe min width for each data field
|
||||
# [x] set priorities for each element
|
||||
# [x] select elements to be shown
|
||||
# [x] choose top priority element min_width < avail_width
|
||||
# [x] lessen avail_width by value if min_width
|
||||
# [x] exclude element from priority list and repeat
|
||||
|
||||
# 10% [.. ] 10/100
|
||||
# pppp bbbbb sssssss
|
||||
|
||||
min_width = {
|
||||
'percent': 4, # 100%
|
||||
'bar': 3, # [.]
|
||||
'size': len("%s" % total)*2 + 3, # 'xxxx / yyyy'
|
||||
}
|
||||
priority = ['percent', 'bar', 'size']
|
||||
|
||||
# select elements to show
|
||||
selected = []
|
||||
avail = width
|
||||
for field in priority:
|
||||
if min_width[field] < avail:
|
||||
selected.append(field)
|
||||
avail -= min_width[field]+1 # +1 is for separator or for reserved space at
|
||||
# the end of line to avoid linefeed on Windows
|
||||
# render
|
||||
output = ''
|
||||
for field in selected:
|
||||
|
||||
if field == 'percent':
|
||||
# fixed size width for percentage
|
||||
output += ('%s%%' % (100 * current // total)).rjust(min_width['percent'])
|
||||
elif field == 'bar': # [. ]
|
||||
# bar takes its min width + all available space
|
||||
output += bar_thermometer(current, total, min_width['bar']+avail)
|
||||
elif field == 'size':
|
||||
# size field has a constant width (min == max)
|
||||
output += ("%s / %s" % (current, total)).rjust(min_width['size'])
|
||||
|
||||
selected = selected[1:]
|
||||
if selected:
|
||||
output += ' ' # add field separator
|
||||
|
||||
return output
|
||||
|
||||
# --/ console helpers
|
||||
|
||||
|
||||
__current_size = 0 # global state variable, which exists solely as a
|
||||
# workaround against Python 3.3.0 regression
|
||||
# http://bugs.python.org/issue16409
|
||||
# fixed in Python 3.3.1
|
||||
def callback_progress(blocks, block_size, total_size, bar_function):
|
||||
"""callback function for urlretrieve that is called when connection is
|
||||
created and when once for each block
|
||||
|
||||
draws adaptive progress bar in terminal/console
|
||||
|
||||
use sys.stdout.write() instead of "print,", because it allows one more
|
||||
symbol at the line end without linefeed on Windows
|
||||
|
||||
:param blocks: number of blocks transferred so far
|
||||
:param block_size: in bytes
|
||||
:param total_size: in bytes, can be -1 if server doesn't return it
|
||||
:param bar_function: another callback function to visualize progress
|
||||
"""
|
||||
global __current_size
|
||||
|
||||
width = min(100, get_console_width())
|
||||
|
||||
if sys.version_info[:3] == (3, 3, 0): # regression workaround
|
||||
if blocks == 0: # first call
|
||||
__current_size = 0
|
||||
else:
|
||||
__current_size += block_size
|
||||
current_size = __current_size
|
||||
else:
|
||||
current_size = min(blocks*block_size, total_size)
|
||||
progress = bar_function(current_size, total_size, width)
|
||||
if progress:
|
||||
sys.stdout.write("\r" + progress)
|
||||
|
||||
class ThrowOnErrorOpener(urllib.FancyURLopener):
|
||||
def http_error_default(self, url, fp, errcode, errmsg, headers):
|
||||
raise Exception("%s: %s" % (errcode, errmsg))
|
||||
|
||||
def download(url, out=None, bar=bar_adaptive):
|
||||
"""High level function, which downloads URL into tmp file in current
|
||||
directory and then renames it to filename autodetected from either URL
|
||||
or HTTP headers.
|
||||
|
||||
:param bar: function to track download progress (visualize etc.)
|
||||
:param out: output filename or directory
|
||||
:return: filename where URL is downloaded to
|
||||
"""
|
||||
names = dict()
|
||||
names["out"] = out or ''
|
||||
names["url"] = filename_from_url(url)
|
||||
# get filename for temp file in current directory
|
||||
prefix = (names["url"] or names["out"] or ".") + "."
|
||||
(fd, tmpfile) = tempfile.mkstemp(".tmp", prefix=prefix, dir=".")
|
||||
os.close(fd)
|
||||
os.unlink(tmpfile)
|
||||
|
||||
# set progress monitoring callback
|
||||
def callback_charged(blocks, block_size, total_size):
|
||||
# 'closure' to set bar drawing function in callback
|
||||
callback_progress(blocks, block_size, total_size, bar_function=bar)
|
||||
if bar:
|
||||
callback = callback_charged
|
||||
else:
|
||||
callback = None
|
||||
|
||||
(tmpfile, headers) = ThrowOnErrorOpener().retrieve(url, tmpfile, callback)
|
||||
names["header"] = filename_from_headers(headers)
|
||||
if os.path.isdir(names["out"]):
|
||||
filename = names["header"] or names["url"]
|
||||
filename = names["out"] + "/" + filename
|
||||
else:
|
||||
filename = names["out"] or names["header"] or names["url"]
|
||||
# add numeric ' (x)' suffix if filename already exists
|
||||
if os.path.exists(filename):
|
||||
filename = filename_fix_existing(filename)
|
||||
shutil.move(tmpfile, filename)
|
||||
|
||||
#print headers
|
||||
return filename
|
||||
|
||||
|
||||
usage = """\
|
||||
usage: wget.py [options] URL
|
||||
|
||||
options:
|
||||
-o --output FILE|DIR output filename or directory
|
||||
-h --help
|
||||
--version
|
||||
"""
|
||||
|
||||
if __name__ == "__main__":
|
||||
if len(sys.argv) < 2 or "-h" in sys.argv or "--help" in sys.argv:
|
||||
sys.exit(usage)
|
||||
if "--version" in sys.argv:
|
||||
sys.exit("wget.py " + __version__)
|
||||
|
||||
from optparse import OptionParser
|
||||
parser = OptionParser()
|
||||
parser.add_option("-o", "--output", dest="output")
|
||||
(options, args) = parser.parse_args()
|
||||
|
||||
url = sys.argv[1]
|
||||
filename = download(args[0], out=options.output)
|
||||
|
||||
print("")
|
||||
print("Saved under %s" % filename)
|
||||
|
||||
r"""
|
||||
features that require more tuits for urlretrieve API
|
||||
http://www.python.org/doc/2.6/library/urllib.html#urllib.urlretrieve
|
||||
|
||||
[x] autodetect filename from URL
|
||||
[x] autodetect filename from headers - Content-Disposition
|
||||
http://greenbytes.de/tech/tc2231/
|
||||
[ ] make HEAD request to detect temp filename from Content-Disposition
|
||||
[ ] process HTTP status codes (i.e. 404 error)
|
||||
http://ftp.de.debian.org/debian/pool/iso-codes_3.24.2.orig.tar.bz2
|
||||
[ ] catch KeyboardInterrupt
|
||||
[ ] optionally preserve incomplete file
|
||||
[x] create temp file in current directory
|
||||
[ ] resume download (broken connection)
|
||||
[ ] resume download (incomplete file)
|
||||
[x] show progress indicator
|
||||
http://mail.python.org/pipermail/tutor/2005-May/038797.html
|
||||
[x] do not overwrite downloaded file
|
||||
[x] rename file automatically if exists
|
||||
[x] optionally specify path for downloaded file
|
||||
|
||||
[ ] options plan
|
||||
[x] -h, --help, --version (CHAOS speccy)
|
||||
[ ] clpbar progress bar style
|
||||
_ 30.0Mb at 3.0 Mbps eta: 0:00:20 30% [===== ]
|
||||
[ ] test "bar \r" print with \r at the end of line on Windows
|
||||
[ ] process Python 2.x urllib.ContentTooShortError exception gracefully
|
||||
(ideally retry and continue download)
|
||||
|
||||
(tmpfile, headers) = urllib.urlretrieve(url, tmpfile, callback_progress)
|
||||
File "C:\Python27\lib\urllib.py", line 93, in urlretrieve
|
||||
return _urlopener.retrieve(url, filename, reporthook, data)
|
||||
File "C:\Python27\lib\urllib.py", line 283, in retrieve
|
||||
"of %i bytes" % (read, size), result)
|
||||
urllib.ContentTooShortError: retrieval incomplete: got only 15239952 out of 24807571 bytes
|
||||
|
||||
[ ] find out if urlretrieve may return unicode headers
|
||||
[ ] test suite for unsafe filenames from url and from headers
|
||||
|
||||
[ ] security checks
|
||||
[ ] filename_from_url
|
||||
[ ] filename_from_headers
|
||||
[ ] MITM redirect from https URL
|
||||
[ ] https certificate check
|
||||
[ ] size+hash check helpers
|
||||
[ ] fail if size is known and mismatch
|
||||
[ ] fail if hash mismatch
|
||||
"""
|
||||
Loading…
Reference in New Issue
Block a user