def upgrade(request):
doc_list = request.db.docs
for doc in doc_list.find():
if "version" not in doc:
doc["version"] = 1
doc["created"] = datetime.utcnow()
doc_list.save(doc)
for doc in doc_list.find({"version": 1}):
if "searchable_text" not in doc.keys():
doc["version"] = 2
doc_list.save(doc)
continue
searchable_text = doc.pop("searchable_text")
lang = guessLanguage(searchable_text)
search_terms = index(searchable_text + " " + doc["title"], [lang])
doc["search_terms"] = search_terms
doc["language"] = lang
doc["version"] = 2
doc_list.save(doc)
for doc in doc_list.find({"version": 2}):
doc["version"] = 3
doc["search_terms"] = [x.lower() for x in doc["search_terms"]]
doc_list.save(doc)
for doc in doc_list.find({"version": 3}):
doc["version"] = 4
doc["keywords"] = []
doc_list.save(doc)
for doc in doc_list.find({"version": 4}):
doc["version"] = 5
doc["already_scanned"] = True
doc_list.save(doc)
return {"success": 1}
def list_view(request):
query_args = {}
if "filter" in request.params:
keys = list(index(request.params["filter"]))
query_args.update({"search_terms": {"$in": keys}})
if "keyword" in request.params:
query_args.update({"keywords": {"$in": [request.params["keyword"]]}})
docs = request.db.docs.find(spec=query_args)
docs.sort("created", DESCENDING)
item_count = docs.count()
page = int(request.params.get("page", 1))
items_per_page = 10
url_maker = PageURL_WebOb(request)
docs = Page(
list(docs[(page - 1) * items_per_page:page * items_per_page]),
url=url_maker,
page=page,
items_per_page=items_per_page,
item_count=item_count,
presliced_list=True,
)
distinct_keywords = request.db.docs.distinct("keywords")
distinct_keywords.sort()
return {"docs": docs, "distinct_keywords": distinct_keywords}
def list_view(request):
query_args = {}
if "filter" in request.params:
keys = list(index(request.params["filter"]))
query_args.update({"search_terms": {"$in": keys}})
if "keyword" in request.params:
query_args.update({"keywords": {"$in": [request.params["keyword"]]}})
docs = request.db.docs.find(spec=query_args)
docs.sort("created", DESCENDING)
item_count = docs.count()
page = int(request.params.get("page", 1))
items_per_page = 10
url_maker = PageURL_WebOb(request)
docs = Page(
list(docs[(page - 1) * items_per_page:page * items_per_page]),
url=url_maker,
page=page,
items_per_page=items_per_page,
item_count=item_count,
presliced_list=True,
)
distinct_keywords = request.db.docs.distinct("keywords")
distinct_keywords.sort()
return {"docs": docs, "distinct_keywords": distinct_keywords}
def reindex(self):
'''
Reindex the document
'''
text_to_index = [self.doc['title']]
for field in self.doc.get('fulltext_fields', []):
text_to_index.append(self.doc[field])
indexed_text = list(index(' '.join(text_to_index),
accepted_languages=self.accepted_languages))
self.doc['search_terms'] = indexed_text
self.db.docs.save(self.doc)
def reindex(self):
'''
Reindex the document
'''
text_to_index = [self.doc['title']]
for field in self.doc.get('fulltext_fields', []):
text_to_index.append(self.doc[field])
indexed_text = list(
index(' '.join(text_to_index),
accepted_languages=self.accepted_languages))
self.doc['search_terms'] = indexed_text
self.db.docs.save(self.doc)
def edit_post(request):
doc = request.db.docs.find({"_id": ObjectId(request.matchdict["id"])})[0]
doc["title"] = request.params["title"]
doc["created"] = datetime.strptime(request.params["created"],
"%Y-%m-%dT%H:%MZ")
description = request.params["description"]
if description:
doc["search_terms"] += index(request.params["description"])
doc["keywords"] = request.params["keywords"].split()
request.db.docs.update({"_id": doc["_id"]}, doc)
url = request.route_url("edit", id=doc["_id"])
request.session.flash("Changes saved", "success")
return HTTPFound(location=url)
def edit_post(request):
doc = request.db.docs.find({"_id": ObjectId(request.matchdict["id"])})[0]
doc["title"] = request.params["title"]
doc["created"] = datetime.strptime(request.params["created"],
"%Y-%m-%dT%H:%MZ")
description = request.params["description"]
if description:
doc["search_terms"] += index(request.params["description"])
doc["keywords"] = request.params["keywords"].split()
request.db.docs.update({"_id": doc["_id"]}, doc)
url = request.route_url("edit", id=doc["_id"])
request.session.flash("Changes saved", "success")
return HTTPFound(location=url)
def collection_get(self):
limit = 2
page = self.request.validated.get('page', 0)
if "filter" in self.request.params:
keys = list(index(self.request.params["filter"]))
spec = {"search_terms": {"$in": keys}}
else:
spec = {}
if page < 0:
pages = int(self.db.count(spec) / limit)
page = pages +1 +page
return [x.to_jsonable_dict() for x in self.db.find(spec, limit=limit,
skip=page * limit, sort=[('created', 1)])]
def collection_get(self):
limit = 2
page = self.request.validated.get('page', 0)
if "filter" in self.request.params:
keys = list(index(self.request.params["filter"]))
spec = {"search_terms": {"$in": keys}}
else:
spec = {}
if page < 0:
pages = int(self.db.count(spec) / limit)
page = pages + 1 + page
return [
x.to_jsonable_dict() for x in self.db.find(
spec, limit=limit, skip=page * limit, sort=[('created', 1)])
]
def add(request):
title = request.params['title']
created = request.params.get('created', datetime.utcnow()) or \
datetime.utcnow()
description = request.params.get('description', '')
force_detection = request.params.get("force_detection",
'true').lower() == 'true'
accepted_languages = request.registry.settings['accepted_languages']
imgstream = request.params['file'].file
doc_list = request.db.docs
for image in get_images_from_stream(imgstream):
try:
lang, img, text = recognize(image, accepted_languages,
force_detection)
except TypeError, e:
err_msg = "Error: " + str(e)
request.session.flash(err_msg, 'failure')
return HTTPServerError(explanation=err_msg,
detail='Go back, unselect force '\
'detection, and try again')
text += " " + description + " " + title
if text:
search_terms = list(index(text, [lang]))
else:
search_terms = ''
thumb = get_thumbnail(image)
doc_list.insert({
'img': Binary(image),
'thumb': Binary(thumb),
'created': created,
'version': 4,
'forced_detection': force_detection,
'language': lang,
'keywords': [],
'search_terms': search_terms,
'title': title
})
def add(request):
title = request.params['title']
created = request.params.get('created', datetime.utcnow()) or \
datetime.utcnow()
description = request.params.get('description', '')
force_detection = request.params.get("force_detection",
'true').lower() == 'true'
accepted_languages = request.registry.settings['accepted_languages']
imgstream = request.params['file'].file
doc_list = request.db.docs
for image in get_images_from_stream(imgstream):
try:
lang, img, text = recognize(image, accepted_languages,
force_detection)
except TypeError, e:
err_msg = "Error: " + str(e)
request.session.flash(err_msg, 'failure')
return HTTPServerError(explanation=err_msg,
detail='Go back, unselect force '\
'detection, and try again')
text += " " + description + " " + title
if text:
search_terms = list(index(text, [lang]))
else:
search_terms = ''
thumb = get_thumbnail(image)
doc_list.insert({'img': Binary(image),
'thumb': Binary(thumb),
'created': created,
'version': 4,
'forced_detection': force_detection,
'language': lang,
'keywords': [],
'search_terms': search_terms,
'title': title})
def test_index():
from lembrar import index
retval = index.index("These are some tests", ['en'])
assert set([u'these', u'are', u'some', u'test']) == retval