Files
odoo_source/addons/website_blog/controllers/main.py
T
Benoit Socias 7559626c54 [IMP] website, *: make a generic search bar snippet available
(*: website_blog, website_event, website_forum, website_sale,
website_slides)

Before this commit the search bar was specific to products.

After this commit a generic search bar is available as a general feature
of website which can be configured to inspect specific models.
The snippet is used to replace the old search bar in blog, courses,
event, forum, page and shop.
The search results of these pages and the autocomplete of the search bar
run through the same search mechanism.
A new hybrid results page has also been created as a target of a search
on "Everything".

In each involved module, `website._search_get_details()` is implemented
to return search metadata for every model related to the `search_type`
parameter.
Search metadata for a single model is returned by `_search_get_detail()`
on that specific model.

The autocomplete runs through the additional
`website._search_render_results()` pre-rendering step that prepares the
data to fit in the autocomplete template.

task-2379555
https://github.com/odoo/odoo/pull/65871

Part-of: odoo/odoo#65871
2021-09-03 06:59:33 +00:00

319 lines
14 KiB
Python

# -*- coding: utf-8 -*-
# Part of Odoo. See LICENSE file for full copyright and licensing details.
import werkzeug
import itertools
import pytz
import babel.dates
from collections import OrderedDict
from odoo import http, fields
from odoo.addons.http_routing.models.ir_http import slug, unslug
from odoo.addons.website.controllers.main import QueryURL
from odoo.addons.portal.controllers.portal import _build_url_w_params
from odoo.http import request
from odoo.osv import expression
from odoo.tools import html2plaintext
from odoo.tools.misc import get_lang
from odoo.tools import sql
class WebsiteBlog(http.Controller):
_blog_post_per_page = 12 # multiple of 2,3,4
_post_comment_per_page = 10
def tags_list(self, tag_ids, current_tag):
tag_ids = list(tag_ids) # required to avoid using the same list
if current_tag in tag_ids:
tag_ids.remove(current_tag)
else:
tag_ids.append(current_tag)
tag_ids = request.env['blog.tag'].browse(tag_ids)
return ','.join(slug(tag) for tag in tag_ids)
def nav_list(self, blog=None):
dom = blog and [('blog_id', '=', blog.id)] or []
if not request.env.user.has_group('website.group_website_designer'):
dom += [('post_date', '<=', fields.Datetime.now())]
groups = request.env['blog.post']._read_group_raw(
dom,
['name', 'post_date'],
groupby=["post_date"], orderby="post_date desc")
for group in groups:
(r, label) = group['post_date']
start, end = r.split('/')
group['post_date'] = label
group['date_begin'] = start
group['date_end'] = end
locale = get_lang(request.env).code
start = pytz.UTC.localize(fields.Datetime.from_string(start))
tzinfo = pytz.timezone(request.context.get('tz', 'utc') or 'utc')
group['month'] = babel.dates.format_datetime(start, format='MMMM', tzinfo=tzinfo, locale=locale)
group['year'] = babel.dates.format_datetime(start, format='yyyy', tzinfo=tzinfo, locale=locale)
return OrderedDict((year, [m for m in months]) for year, months in itertools.groupby(groups, lambda g: g['year']))
def _prepare_blog_values(self, blogs, blog=False, date_begin=False, date_end=False, tags=False, state=False, page=False, search=None):
""" Prepare all values to display the blogs index page or one specific blog"""
BlogPost = request.env['blog.post']
BlogTag = request.env['blog.tag']
# prepare domain
domain = request.website.website_domain()
if blog:
domain += [('blog_id', '=', blog.id)]
if date_begin and date_end:
domain += [("post_date", ">=", date_begin), ("post_date", "<=", date_end)]
active_tag_ids = tags and [unslug(tag)[1] for tag in tags.split(',')] or []
active_tags = BlogTag
if active_tag_ids:
active_tags = BlogTag.browse(active_tag_ids).exists()
fixed_tag_slug = ",".join(slug(t) for t in active_tags)
if fixed_tag_slug != tags:
new_url = request.httprequest.full_path.replace("/tag/%s" % tags, "/tag/%s" % fixed_tag_slug, 1)
if new_url != request.httprequest.full_path: # check that really replaced and avoid loop
return request.redirect(new_url, 301)
domain += [('tag_ids', 'in', active_tags.ids)]
if request.env.user.has_group('website.group_website_designer'):
count_domain = domain + [("website_published", "=", True), ("post_date", "<=", fields.Datetime.now())]
published_count = BlogPost.search_count(count_domain)
unpublished_count = BlogPost.search_count(domain) - published_count
if state == "published":
domain += [("website_published", "=", True), ("post_date", "<=", fields.Datetime.now())]
elif state == "unpublished":
domain += ['|', ("website_published", "=", False), ("post_date", ">", fields.Datetime.now())]
else:
domain += [("post_date", "<=", fields.Datetime.now())]
use_cover = request.website.is_view_active('website_blog.opt_blog_cover_post')
fullwidth_cover = request.website.is_view_active('website_blog.opt_blog_cover_post_fullwidth_design')
# if blog, we show blog title, if use_cover and not fullwidth_cover we need pager + latest always
offset = (page - 1) * self._blog_post_per_page
if not blog:
if use_cover and not fullwidth_cover:
offset += 1
options = {
'displayDescription': True,
'displayDetail': False,
'displayExtraDetail': False,
'displayExtraLink': False,
'displayImage': False,
'allowFuzzy': not request.params.get('noFuzzy'),
'blog': str(blog.id) if blog else None,
'tag': ','.join([str(id) for id in active_tags.ids]),
'date_begin': date_begin,
'date_end': date_end,
'state': state,
}
total, details, fuzzy_search_term = request.website._search_with_fuzzy("blog_posts_only", search,
limit=page * self._blog_post_per_page, order="is_published desc, post_date desc, id asc", options=options)
posts = details[0].get('results', BlogPost)
first_post = BlogPost
if posts and not blog and posts[0].website_published:
first_post = posts[0]
posts = posts[offset:offset + self._blog_post_per_page]
pager = request.website.pager(
url=request.httprequest.path.partition('/page/')[0],
total=total,
page=page,
step=self._blog_post_per_page,
)
if not blogs:
all_tags = request.env['blog.tag']
else:
all_tags = blogs.all_tags(join=True) if not blog else blogs.all_tags().get(blog.id, request.env['blog.tag'])
tag_category = sorted(all_tags.mapped('category_id'), key=lambda category: category.name.upper())
other_tags = sorted(all_tags.filtered(lambda x: not x.category_id), key=lambda tag: tag.name.upper())
# for performance prefetch the first post with the others
post_ids = (first_post | posts).ids
# and avoid accessing related blogs one by one
posts.blog_id
return {
'date_begin': date_begin,
'date_end': date_end,
'first_post': first_post.with_prefetch(post_ids),
'other_tags': other_tags,
'tag_category': tag_category,
'nav_list': self.nav_list(),
'tags_list': self.tags_list,
'pager': pager,
'posts': posts.with_prefetch(post_ids),
'tag': tags,
'active_tag_ids': active_tags.ids,
'domain': domain,
'state_info': state and {"state": state, "published": published_count, "unpublished": unpublished_count},
'blogs': blogs,
'blog': blog,
'search': fuzzy_search_term or search,
'search_count': total,
'original_search': fuzzy_search_term and search,
}
@http.route([
'/blog',
'/blog/page/<int:page>',
'/blog/tag/<string:tag>',
'/blog/tag/<string:tag>/page/<int:page>',
'''/blog/<model("blog.blog"):blog>''',
'''/blog/<model("blog.blog"):blog>/page/<int:page>''',
'''/blog/<model("blog.blog"):blog>/tag/<string:tag>''',
'''/blog/<model("blog.blog"):blog>/tag/<string:tag>/page/<int:page>''',
], type='http', auth="public", website=True, sitemap=True)
def blog(self, blog=None, tag=None, page=1, search=None, **opt):
Blog = request.env['blog.blog']
blogs = Blog.search(request.website.website_domain(), order="create_date asc, id asc")
if not blog and len(blogs) == 1:
return request.redirect('/blog/%s' % slug(blogs[0]), code=302)
date_begin, date_end, state = opt.get('date_begin'), opt.get('date_end'), opt.get('state')
if tag and request.httprequest.method == 'GET':
# redirect get tag-1,tag-2 -> get tag-1
tags = tag.split(',')
if len(tags) > 1:
url = QueryURL('' if blog else '/blog', ['blog', 'tag'], blog=blog, tag=tags[0], date_begin=date_begin, date_end=date_end, search=search)()
return request.redirect(url, code=302)
values = self._prepare_blog_values(blogs=blogs, blog=blog, date_begin=date_begin, date_end=date_end, tags=tag, state=state, page=page, search=search)
# in case of a redirection need by `_prepare_blog_values` we follow it
if isinstance(values, werkzeug.wrappers.Response):
return values
if blog:
values['main_object'] = blog
values['edit_in_backend'] = True
values['blog_url'] = QueryURL('', ['blog', 'tag'], blog=blog, tag=tag, date_begin=date_begin, date_end=date_end, search=search)
else:
values['blog_url'] = QueryURL('/blog', ['tag'], date_begin=date_begin, date_end=date_end, search=search)
return request.render("website_blog.blog_post_short", values)
@http.route(['''/blog/<model("blog.blog"):blog>/feed'''], type='http', auth="public", website=True, sitemap=True)
def blog_feed(self, blog, limit='15', **kwargs):
v = {}
v['blog'] = blog
v['base_url'] = blog.get_base_url()
v['posts'] = request.env['blog.post'].search([('blog_id', '=', blog.id)], limit=min(int(limit), 50), order="post_date DESC")
v['html2plaintext'] = html2plaintext
r = request.render("website_blog.blog_feed", v, headers=[('Content-Type', 'application/atom+xml')])
return r
@http.route([
'''/blog/<model("blog.blog"):blog>/post/<model("blog.post", "[('blog_id','=',blog.id)]"):blog_post>''',
], type='http', auth="public", website=True, sitemap=False)
def old_blog_post(self, blog, blog_post, tag_id=None, page=1, enable_editor=None, **post):
# Compatibility pre-v14
return request.redirect(_build_url_w_params("/blog/%s/%s" % (slug(blog), slug(blog_post)), request.params), code=301)
@http.route([
'''/blog/<model("blog.blog"):blog>/<model("blog.post", "[('blog_id','=',blog.id)]"):blog_post>''',
], type='http', auth="public", website=True, sitemap=True)
def blog_post(self, blog, blog_post, tag_id=None, page=1, enable_editor=None, **post):
""" Prepare all values to display the blog.
:return dict values: values for the templates, containing
- 'blog_post': browse of the current post
- 'blog': browse of the current blog
- 'blogs': list of browse records of blogs
- 'tag': current tag, if tag_id in parameters
- 'tags': all tags, for tag-based navigation
- 'pager': a pager on the comments
- 'nav_list': a dict [year][month] for archives navigation
- 'next_post': next blog post, to direct the user towards the next interesting post
"""
BlogPost = request.env['blog.post']
date_begin, date_end = post.get('date_begin'), post.get('date_end')
domain = request.website.website_domain()
blogs = blog.search(domain, order="create_date, id asc")
tag = None
if tag_id:
tag = request.env['blog.tag'].browse(int(tag_id))
blog_url = QueryURL('', ['blog', 'tag'], blog=blog_post.blog_id, tag=tag, date_begin=date_begin, date_end=date_end)
if not blog_post.blog_id.id == blog.id:
return request.redirect("/blog/%s/%s" % (slug(blog_post.blog_id), slug(blog_post)), code=301)
tags = request.env['blog.tag'].search([])
# Find next Post
blog_post_domain = [('blog_id', '=', blog.id)]
if not request.env.user.has_group('website.group_website_designer'):
blog_post_domain += [('post_date', '<=', fields.Datetime.now())]
all_post = BlogPost.search(blog_post_domain)
if blog_post not in all_post:
return request.redirect("/blog/%s" % (slug(blog_post.blog_id)))
# should always return at least the current post
all_post_ids = all_post.ids
current_blog_post_index = all_post_ids.index(blog_post.id)
nb_posts = len(all_post_ids)
next_post_id = all_post_ids[(current_blog_post_index + 1) % nb_posts] if nb_posts > 1 else None
next_post = next_post_id and BlogPost.browse(next_post_id) or False
values = {
'tags': tags,
'tag': tag,
'blog': blog,
'blog_post': blog_post,
'blogs': blogs,
'main_object': blog_post,
'nav_list': self.nav_list(blog),
'enable_editor': enable_editor,
'next_post': next_post,
'date': date_begin,
'blog_url': blog_url,
}
response = request.render("website_blog.blog_post_complete", values)
if blog_post.id not in request.session.get('posts_viewed', []):
if sql.increment_field_skiplock(blog_post, 'visits'):
if not request.session.get('posts_viewed'):
request.session['posts_viewed'] = []
request.session['posts_viewed'].append(blog_post.id)
request.session.modified = True
return response
@http.route('/blog/<int:blog_id>/post/new', type='http', auth="user", website=True)
def blog_post_create(self, blog_id, **post):
# Use sudo so this line prevents both editor and admin to access blog from another website
# as browse() will return the record even if forbidden by security rules but editor won't
# be able to access it
if not request.env['blog.blog'].browse(blog_id).sudo().can_access_from_current_website():
raise werkzeug.exceptions.NotFound()
new_blog_post = request.env['blog.post'].create({
'blog_id': blog_id,
'is_published': False,
})
return request.redirect("/blog/%s/%s?enable_editor=1" % (slug(new_blog_post.blog_id), slug(new_blog_post)))
@http.route('/blog/post_duplicate', type='http', auth="user", website=True, methods=['POST'])
def blog_post_copy(self, blog_post_id, **post):
""" Duplicate a blog.
:param blog_post_id: id of the blog post currently browsed.
:return redirect to the new blog created
"""
new_blog_post = request.env['blog.post'].with_context(mail_create_nosubscribe=True).browse(int(blog_post_id)).copy()
return request.redirect("/blog/%s/%s?enable_editor=1" % (slug(new_blog_post.blog_id), slug(new_blog_post)))