def test_render_attribute_to_document():
document = PyQuery('<a attribute="{ value }" data-riot-id="0"></a>')
expression = {'expression': '{ value }', 'attribute': 'attribute', 'type': 'attribute', 'node': document}
render_document([expression], {'value': 'value'})
assert document.outer_html() == '<a attribute="value" data-riot-id="0" data-riot-dirty="true"></a>'
render_document([expression], {'value': 1})
assert document.outer_html() == '<a attribute="1" data-riot-id="0" data-riot-dirty="true"></a>'
def test_render_attribute_to_document():
document = PyQuery('<a attribute="{ value }" data-riot-id="0"></a>')
expression = {
'expression': '{ value }',
'attribute': 'attribute',
'type': 'attribute',
'node': document
}
render_document([expression], {'value': 'value'})
assert document.outer_html(
) == '<a attribute="value" data-riot-id="0" data-riot-dirty="true"></a>'
render_document([expression], {'value': 1})
assert document.outer_html(
) == '<a attribute="1" data-riot-id="0" data-riot-dirty="true"></a>'
def fetch_comment_list(self, broadcast_url, broadcast_douban_id):
url = broadcast_url
comments = []
while True:
response = self.fetch_url_content(url)
dom = PyQuery(response.text)
comment_items = dom('#comments>.comment-item')
for comment_item in comment_items:
item_div = PyQuery(comment_item)
comments.append({
'content':
item_div.outer_html(),
'target_type':
'broadcast',
'target_douban_id':
broadcast_douban_id,
'douban_id':
item_div.attr('data-cid'),
'user':
self.fetch_user(
PyQuery(item_div('.pic>a')).attr('data-uid')),
'text':
PyQuery(item_div('.content>p.text')).text(),
'created':
PyQuery(item_div('.content>.author>.created_at')).text(),
})
next_page = dom('#comments>.paginator>.next>a')
if next_page:
url = broadcast_url + next_page.attr('href')
else:
break
return comments
def render_svg(image,
width=None,
height=None,
request=None,
css_class='',
img_class='',
alt=''):
"""Render SVG file"""
# pylint: disable=too-many-arguments
options = {}
if width or height:
options['style'] = 'width: {0}{1}; height: {2}{3};'.format(
width, 'px' if isinstance(width, int) else '', height,
'px' if isinstance(height, int) else '')
else:
options['style'] = ''
options['css_class'] = css_class
if alt or img_class:
svg = PyQuery(image.data)
if alt:
group = PyQuery('<g></g>')
group.append(PyQuery('<title />').text(alt))
for child in svg.children():
group.append(child)
svg.empty().append(group)
if img_class:
svg.attr('class', img_class)
options['svg'] = svg.outer_html()
else:
options['svg'] = image.data
return render('templates/svg-render.pt', options, request)
def render(self):
self._generate_slides()
self._render_slides()
press_wrapper = Pq('<div></div>')
press_wrapper.attr('id', 'pres_wrapper')
press_wrapper.append(self.body)
self.html = press_wrapper.outer_html()
def test_render_markup_to_document():
document = PyQuery('<custom data-riot-id="0"><text data-riot-id="0.0"><span class="name">{ name}</span><span class="greet">{ greet }</span></text></custom>')
expressions = parse_document_expressions(document)
render_document(expressions, {'name': '@ainesmile', 'greet': 'I love you.'})
assert document.outer_html() == '<custom data-riot-id="0" data-riot-dirty="true"><text data-riot-id="0.0" data-riot-dirty="true"></text></custom>'
assert expressions[0]['value'] == [(u'name', '@ainesmile'), (u'greet', 'I love you.')]
render_document(expressions, {'name': '@soasme', 'greet': 'I love you, too.'})
assert expressions[0]['value'] == [(u'name', '@soasme'), (u'greet', 'I love you, too.')]
def rich_text_export(html_str, request=None, id_prefix=''):
if not (html_str and request and id_prefix):
return html_str
pq = PyQuery(html_str)
# Footnote labels have an :after css pseudo element with '.' content
# To prevent editing of the label in a way that will be lost
pq(".footnote-label").append(" ")
# IDs that unique within a document may not be unique within multiple documents
# so we add a prefix
def prefix_id(_, el):
original_id = el.attrib['id']
el.attrib['id'] = f"{id_prefix}-{original_id}"
def prefix_href(_, el):
original_target = el.attrib['href'][1:]
el.attrib['href'] = f"#{id_prefix}-{original_target}"
pq("[id]").each(prefix_id)
pq("[href^='#']").each(prefix_href)
# Images need to have absolute urls
def absolute_src(_, el):
original_src = el.attrib['src']
el.attrib['src'] = request.build_absolute_uri(original_src)
pq("img[src]").each(absolute_src)
# Add Doc styling wrappers to images
def replace_in_parent(style, el):
original_html = el.parent().html(method="html")
src = el.outer_html()
replacement = f"</p><div data-custom-style='{style}'>{el.outer_html()}</div><p>"
el.parent().html(original_html.replace(src, replacement))
def attach_id_to_style(el):
el.attrib['data-custom-style'] += f"-{id_prefix}"
for el in pq("img.image-center-large").items():
replace_in_parent("Image Centered Large", el)
for el in pq("img.image-center-medium").items():
replace_in_parent("Image Centered Medium", el)
for el in pq("img.image-left-medium").items():
replace_in_parent("Image Left Medium", el)
for el in pq("img.image-right-medium").items():
replace_in_parent("Image Right Medium", el)
for el in pq("[data-custom-style='Footnote Reference']"):
attach_id_to_style(el)
for el in pq("[data-custom-style='Footnote Text']"):
attach_id_to_style(el)
# Insert a non-breaking space if the paragraph after an image is empty
# to prevent it from potentially overlapping with a following image
pq("div[data-custom-style]+p:empty").text(" \xa0 ")
return pq.outer_html()
def get_html_page(self, task):
'''
抓取目标页面
'''
msg = 'get_html_page(): task: {task:s}'
logger.debug(msg.format(task = str(task)))
## 更新记录状态
update_record_status(self.db_conn, task['url'], 'pending')
if 0 < self.config['max_depth'] < task['depth']:
msg = '已超过最大深度: task: {task:s}'
logger.warning(msg.format(task = str(task)))
return
if task['failed_times'] > self.config['max_retry_times']:
msg = '失败次数过多, 不再重试: task: {task:s}'
logger.warning(msg.format(task = str(task)))
update_record_status(self.db_conn, task['url'], 'failed')
return
code, resp = request_get_async(task, self.config)
if not code:
msg = '请求页面失败, 重新入队列: task: {task:s}, err: {err:s}'
logger.error(msg.format(task = str(task), err = resp))
## 出现异常, 则失败次数加1
## 不需要调用enqueue(), 直接入队列.
task['failed_times'] += 1
self.enqueue_page(task)
return
elif resp.status_code == 404:
## 抓取失败一般是5xx或403, 405等, 出现404基本上就没有重试的意义了, 可以直接放弃
update_record_status(self.db_conn, task['url'], 'failed')
return
msg = '请求页面成功, 准备解析并保存: task: {task:s}'
logger.debug(msg.format(task = str(task)))
try:
charset = get_page_charset(resp.content)
resp.encoding = charset
pq_selector = PyQuery(resp.text)
## 超过最大深度的页面不再抓取, 在入队列前就先判断.
## 但静态资源无所谓深度, 所以还是要抓取的.
if 0 < self.config['max_depth'] < task['depth'] + 1:
msg = '当前页面已达到最大深度, 不再抓取新页面: task {task:s}'
logger.warning(msg.format(task = str(task)))
else:
parse_linking_pages(pq_selector, task, self.config, callback = self.enqueue_page)
parse_linking_assets(pq_selector, task, self.config, callback = self.enqueue_asset)
logger.debug('修改页面元素链接完成')
## 抓取此页面上的静态资源
self.asset_worker_pool.start(task)
byte_content = pq_selector.outer_html().encode('utf-8')
file_path, file_name = trans_to_local_path(task['url'], 'page', self.main_site)
code, data = save_file_async(self.config['site_path'], file_path, file_name, byte_content)
if code: update_record_status(self.db_conn, task['url'], 'success')
except Exception as err:
msg = '保存页面文件失败: task: {task:s}, err: {err:s}'
logger.error(msg.format(task = str(task), err = repr(err)))
def format_footnotes_for_export(html_str):
"""
Footnotes are stored in TextBlocks as
`<span class="footnote footnote-ref" data-custom-style="Footnote Reference" id="footnote-${data.id}-ref">${data.mark}</span>`
`<div class="footnote footnote-body" id="footnote-${data.id}"><p><span class="footnote-label" contenteditable="false">${data.mark}</span>${data.footnote}</p></div>`
"""
if not html_str:
return html_str
pq = PyQuery(html_str)
pq(".footnote-label").append(". ")
return pq.outer_html()
def _convert_svg_to_resource(self, svg_element):
"""
Converts svg element to image.
"""
name = uuid.uuid4().hex
element = PyQuery(svg_element)
svg = element.outer_html().replace('viewbox', 'viewBox')
element.replace_with(PyQuery('<img src="name:{0}" />'.format(name)))
return Resource(name, convert_svg_to_png(svg), 'image/png')
def refineLocal():
with open(os.path.join(savefolder, "downloads.html"), "r", encoding="utf8") as f:
html = PyQuery(f.read())
cmd = ""
for item in html("li button"):
dl_url = item.get("data-clipboard-text").split(" ")[1]
cmd += "{}\n".format(item.get("data-clipboard-text"))
if os.path.exists(os.path.join(savefolder, os.path.basename(urllib.parse.urlsplit(dl_url).path))):
item.set("class", "btn downloaded")
else:
item.set("class", "btn")
with open(os.path.join(savefolder, "downloads-refined.html"), "w+", encoding="utf-8") as wf:
wf.write(html.outer_html())
subprocess.Popen(["explorer", os.path.join(savefolder, "downloads-refined.html")])
with open(os.path.join(savefolder, "wwdc2016.sh"), "w+", encoding="utf8") as f:
f.write(cmd[:-2])
def prefix_ids_hrefs(html_str, prefix):
if not html_str:
return html_str
def prefix_id(_, el):
original_id = el.attrib['id']
el.attrib['id'] = f"{prefix}-{original_id}"
def prefix_href(_, el):
original_target = el.attrib['href'][1:]
el.attrib['href'] = f"#{prefix}-{original_target}"
pq = PyQuery(html_str)
pq("[id]").each(prefix_id)
pq("[href^='#']").each(prefix_href)
return pq.outer_html()
def fetch_note_comments(self, url, dom, douban_id):
comments = []
strip_username = lambda el: re.findall(
r'^http(?:s?)://www\.douban\.com/people/(.+)/$', el.attr('href')
).pop(0)
while True:
comment_items = dom('#comments .comment-item')
for comment_item in comment_items:
item_div = PyQuery(comment_item)
quote_user_link = item_div('.content>.reply-quote>.pubdate>a')
if quote_user_link:
quote_user_name = quote_user_link.text()
quote_user_id = strip_username(quote_user_link)
quote_text = item_div('.content>.reply-quote>.all').text()
blockquote = '{0}({1}):{2}'.format(quote_user_name,
quote_user_id,
quote_text)
else:
blockquote = None
comments.append({
'douban_id':
item_div.attr('data-cid'),
'content':
item_div.outer_html(),
'target_type':
'note',
'target_douban_id':
douban_id,
'user':
self.fetch_user(strip_username(item_div('.pic>a'))),
'text':
item_div('.content>p').text(),
'created':
item_div('.content>.author>span').text(),
'quote':
blockquote,
})
next_page = dom('#comments>.paginator>.next>a')
if next_page:
url = next_page.attr('href')
else:
break
response = self.fetch_url_content(url)
dom = PyQuery(response.text)
return comments
def test_render_markup_to_document():
document = PyQuery(
'<custom data-riot-id="0"><text data-riot-id="0.0"><span class="name">{ name}</span><span class="greet">{ greet }</span></text></custom>'
)
expressions = parse_document_expressions(document)
render_document(expressions, {
'name': '@ainesmile',
'greet': 'I love you.'
})
assert document.outer_html(
) == '<custom data-riot-id="0" data-riot-dirty="true"><text data-riot-id="0.0" data-riot-dirty="true"></text></custom>'
assert expressions[0]['value'] == [(u'name', '@ainesmile'),
(u'greet', 'I love you.')]
render_document(expressions, {
'name': '@soasme',
'greet': 'I love you, too.'
})
assert expressions[0]['value'] == [(u'name', '@soasme'),
(u'greet', 'I love you, too.')]
def convert_image_links(self, url: str, html: str):
"""
replace image src from relative url to absolute url
"""
doc = PyQuery(html)
images = doc.find("img").items()
for image in images:
src = image.attr("src")
abs_url = urllib.parse.urljoin(url, src)
if self.exclude_image_url_pattern and re.match(self.exclude_image_url_pattern, abs_url):
self.logger.info("skiping %s", abs_url)
image.remove_attr("src")
image.attr("alt", "excluded image")
else:
ext = get_url_ext(abs_url)
save_to = "{}{}".format(md5_digest(abs_url), ext)
image.attr("src", save_to.replace(".svg", ".png"))
self.images.add((abs_url, save_to))
return doc.outer_html()
def parse_status(status_div):
"""
关于object_kind说明：
1001: 图书
1002: 电影
1003: 音乐
1005: 关注好友
1011: 活动
1012: 评论
1013: 小组话题
1014: （电影）讨论
1015: 日记
1018: 图文广播
1019: 小组
1020: 豆列
1021: 九点文章
1022: 网页
1025: 相册照片
1026: 相册
1043: 影人
1044: 艺术家
1062: board(???)
2001: 线上活动
2004: 小站视频
3043: 豆瓣FM单曲
3049: 读书笔记
3065: 条目
3072: 豆瓣FM兆赫
3090: 东西
3114: 游戏
5021: 豆瓣阅读的图片
5022: 豆瓣阅读的作品
"""
if not isinstance(status_div, PyQuery):
status_div = PyQuery(status_div)
reshared_count = 0
like_count = 0
comments_count = 0
created_at = None
is_noreply = False
status_url = None
target_type = None
object_kind = None
object_id = None
reshared_detail = None
blockquote = None
douban_user_id = status_div.attr('data-uid')
douban_id = status_div.attr('data-sid')
is_saying = status_div.has_class('saying')
is_reshared = status_div.has_class('status-reshared-wrapper')
try:
created_span = status_div.find('.actions>.created_at')[0]
except:
is_noreply = True
try:
"""
获取广播链接
"""
exactly_link = PyQuery(status_div.find('.actions a').eq(0))
status_url = exactly_link.attr('href')
except:
pass
try:
"""
获取关于广播类型的属性
"""
status_item_div = PyQuery(
status_div.find('.status-item').eq(0))
target_type = status_item_div.attr('data-target-type')
object_kind = status_item_div.attr('data-object-kind')
object_id = status_item_div.attr('data-object-id')
if not douban_user_id:
douban_user_id = status_item_div.attr('data-uid')
if not douban_id:
douban_id = status_div.attr('data-sid')
blockquote = PyQuery(status_item_div.find('blockquote')).html()
except:
pass
if not is_noreply:
"""
获取创建时间、回复、点赞、转播数
"""
try:
created_at = PyQuery(created_span).attr('title')
reply_link = PyQuery(
status_item_div.find('.actions>.new-reply'))
comments_count = reply_link.attr('data-count')
like_span = PyQuery(
status_item_div.find('.actions>.like-count'))
like_count = like_span.attr('data-count')
if like_count is None:
try:
like_count = int(
re.match(r'赞\((.*)\)',
like_span.text().strip())[1])
except:
like_count = 0
reshared_span = PyQuery(
status_item_div.find('.actions>.reshared-count'))
reshared_count = reshared_span.attr('data-count')
if reshared_count is None:
reshared_count = 0
except:
pass
if not douban_id or douban_id == 'None':
"""
原广播已被删除
"""
return None, None
detail = {
'douban_id': douban_id,
'douban_user_id': douban_user_id,
'content': status_div.outer_html(),
'created': created_at,
'is_reshared': is_reshared,
'is_saying': is_saying,
'is_noreply': is_noreply,
'updated_at': now,
'reshared_count': reshared_count,
'like_count': like_count,
'comments_count': comments_count,
'status_url': status_url,
'target_type': target_type,
'object_kind': object_kind,
'object_id': object_id,
'user': self.fetch_user_by_id(douban_user_id),
'blockquote': blockquote,
}
if is_reshared:
reshared_status_div = PyQuery(
status_div.find('.status-real-wrapper').eq(0))
reshared_detail, _ = parse_status(reshared_status_div)
if reshared_detail:
detail['reshared_id'] = reshared_detail['douban_id']
if target_type == 'sns':
attachments = []
images = status_div.find(
'.attachments-saying.group-pics a.view-large')
for img_lnk in images:
attachments.append({
'type': 'image',
'url': PyQuery(img_lnk).attr('href'),
})
images = status_div.find(
'.attachments-saying.attachments-pic img')
for img in images:
img_lnk = PyQuery(img).attr('data-raw-src')
if img_lnk:
attachments.append({
'type': 'image',
'url': img_lnk,
})
if attachments:
self.save_attachments(attachments)
detail['attachments'] = attachments
elif target_type == 'movie' and object_kind == '1002':
self.fetch_movie(object_id)
elif target_type == 'book' and object_kind == '1001':
self.fetch_book(object_id)
elif target_type == 'music' and object_kind == '1003':
self.fetch_music(object_id)
return detail, reshared_detail
def test_identify_document(document, result):
root = PyQuery(document)
identify_document(root)
assert root.outer_html() == result
def scanRemote():
html = PyQuery("""<!doctype html>
<html lang="en">
<head>
<title>WWDC 2016</title>
<script src="data:text/javascript;base64,LyohCiAqIGNsaXBib2FyZC5qcyB2MS41LjEwCiAqIGh0dHBzOi8vemVub3JvY2hhLmdpdGh1Yi5p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">
</script>
<style type="text/css">
*{
font-family: Consolas, Serif;
}
body{
margin:0;
padding: 0;
}
h1{
text-align: center;
}
ul{
column-count: 3;
column-gap: 20px;
}
li{
list-style: none;
margin-bottom: 0.2em;
}
.btn{
height : 32px;
background-color: white;
border-radius: 5px;
}
.downloaded{
background-color:yellow;
}
</style>
</head>
<body>
<h1>WWDC 2016 DOWNLOADS</h2>
<ul id="downloadlist">
</ul>
<script type="text/javascript">
var clipboard = new Clipboard('.btn');
</script>
</body>
</html>""")
item = """<li><button class="btn downloaded" data-clipboard-text="{}">{}</button></li>"""
not_downloaded_item = """<li><button class="btn" data-clipboard-text="{}">{}</button></li>"""
entrance = PyQuery(url=url)
videos = entrance('li')
for v in videos:
v = PyQuery(v)
if v.attr("data-released") == "true":
h5 = v.find("h5")
a = v.find("a[href]")
(title, page) = (h5.text(), url + a.attr("href"))
subpage = PyQuery(page)
dl_url = urllib.parse.urlparse(subpage("a:contains('HD')").attr.href).path
print("[{}]({})".format(title, dl_url))
if os.path.exists(os.path.join(savefolder, os.path.basename(urllib.parse.urlsplit(dl_url).path))):
html("#downloadlist").append(item.format("curl {} -O -C -".format(dl_url),title))
else:
html("#downloadlist").append(not_downloaded_item.format("curl {} -O -C -".format(dl_url), title))
with open(os.path.join(savefolder, "downloads.html"), "w+", encoding="utf8") as f:
print(html.outer_html())
f.write(html.outer_html())
subprocess.Popen(["explorer", os.path.join(savefolder, "downloads.html")])
def test_identify_document(document, result):
root = PyQuery(document)
identify_document(root)
assert root.outer_html() == result
