sporada/nopaque
1
0
Fork 0
mirror of https://gitlab.ub.uni-bielefeld.de/sfb1288inf/nopaque.git synced 2024-11-14 16:55:42 +00:00
Code Issues Projects Releases Wiki Activity

Merge branch 'development' into public-corpus

Browse Source
This commit is contained in:
Inga Kirschnick 2023-02-09 11:54:58 +01:00
commit 4d227415d9
5 changed files with 301 additions and 182 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
app/corpora/cqi_over_socketio/__init__.py
View File

@ -62,7 +62,9 @@ def connect(auth):
if corpus is None:
# return {'code': 404, 'msg': 'Not Found'}
raise ConnectionRefusedError('Not Found')
if not (corpus.user == current_user or current_user.is_administrator()):
if not (corpus.user == current_user
or current_user.is_following_corpus(corpus)
or current_user.is_administrator()):
# return {'code': 403, 'msg': 'Forbidden'}
raise ConnectionRefusedError('Forbidden')
if corpus.status not in [

65
app/corpora/routes.py
View File

@ -18,18 +18,6 @@ from . import bp
from .forms import ChangeCorpusSettingsForm, CreateCorpusFileForm, CreateCorpusForm, UpdateCorpusFileForm
def user_can_read_corpus(user, corpus):
return corpus.user == user or user.is_administrator() or corpus.is_public
def user_can_update_corpus(user, corpus):
return corpus.user == user or user.is_administrator()
def user_can_delete_corpus(user, corpus):
return user_can_update_corpus(user, corpus)
@bp.route('')
@login_required
def corpora():
@ -70,7 +58,10 @@ def create_corpus():
@login_required
def corpus(corpus_id):
corpus = Corpus.query.get_or_404(corpus_id)
if not user_can_read_corpus(current_user, corpus):
if not (corpus.user == current_user
or current_user.is_administrator()
or current_user.is_following_corpus(corpus)
or corpus.is_public):
abort(403)
corpus_settings_form = ChangeCorpusSettingsForm(
data=corpus.to_json_serializeable(),
@ -81,17 +72,20 @@ def corpus(corpus_id):
db.session.commit()
flash('Your changes have been saved')
return redirect(url_for('.corpus', corpus_id=corpus.id))
# following_users = [
# u.to_json_serializeable() for u
# in corpus.following_users
# ]
return render_template(
'corpora/corpus.html.j2',
corpus_settings_form=corpus_settings_form,
corpus=corpus,
# following_users=following_users,
title='Corpus'
)
if corpus.user == current_user or current_user.is_administrator():
return render_template(
'corpora/corpus.html.j2',
corpus_settings_form=corpus_settings_form,
corpus=corpus,
title='Corpus'
)
else:
print('public')
return render_template(
'corpora/corpus_public.html.j2',
corpus=corpus,
title='Corpus'
)
@ -99,7 +93,7 @@ def corpus(corpus_id):
# @login_required
# def update_corpus(corpus_id):
# corpus = Corpus.query.get_or_404(corpus_id)
# if not user_can_update_corpus(current_user, corpus):
# if not (corpus.user == current_user or current_user.is_administrator()):
# abort(403)
# return render_template(
# 'corpora/update_corpus.html.j2',
@ -118,7 +112,7 @@ def delete_corpus(corpus_id):
db.session.commit()
corpus = Corpus.query.get_or_404(corpus_id)
if not user_can_delete_corpus(current_user, corpus):
if not (corpus.user == current_user or current_user.is_administrator()):
abort(403)
thread = Thread(
target=_delete_corpus,
@ -132,7 +126,10 @@ def delete_corpus(corpus_id):
@login_required
def analyse_corpus(corpus_id):
corpus = Corpus.query.get_or_404(corpus_id)
if not user_can_read_corpus(current_user, corpus):
if not (corpus.user == current_user
or current_user.is_administrator()
or current_user.is_following_corpus(corpus)
or corpus.is_public):
abort(403)
return render_template(
'corpora/analyse_corpus.html.j2',
@ -151,7 +148,7 @@ def build_corpus(corpus_id):
db.session.commit()
corpus = Corpus.query.get_or_404(corpus_id)
if not user_can_update_corpus(current_user, corpus):
if not (corpus.user == current_user or current_user.is_administrator()):
abort(403)
# Check if the corpus has corpus files
if not corpus.files.all():
@ -169,7 +166,7 @@ def build_corpus(corpus_id):
@login_required
def create_corpus_file(corpus_id):
corpus = Corpus.query.get_or_404(corpus_id)
if not user_can_update_corpus(current_user, corpus):
if not (corpus.user == current_user or current_user.is_administrator()):
abort(403)
form = CreateCorpusFileForm()
if form.is_submitted():
@ -334,13 +331,3 @@ def remove_permission(corpus_id, user_id, permission):
db.session.commit()
return 'ok'
@bp.route('/public/<hashid:corpus_id>')
def public_corpus(corpus_id):
corpus = Corpus.query.get_or_404(corpus_id)
corpus_files = [cf.to_json_serializeable() for cf in CorpusFile.query.filter_by(corpus_id = corpus_id).all()]
return render_template(
'corpora/public_corpus.html.j2',
corpus=corpus,
corpus_files=corpus_files,
title=corpus.title
)

4
app/templates/corpora/public_corpus.html.j2 → app/templates/corpora/corpus_public.html.j2
View File

@ -67,7 +67,9 @@
followingUserList.add({{ following_users|tojson }}); #}
corpusFollowingRequest.addEventListener('click', () => {
if ({{ current_user.is_following_corpus(corpus)|tojson }}) {
corpusFollowingRequest.innerHTML = '<i class="material-icons left">add</i>Unfollow Corpus';
if ("{{ current_user.is_following_corpus(corpus) }}" === "False") {
corpusFollowingRequest.lastChild.textContent = 'Unfollow Corpus';
return new Promise((resolve, reject) => {
fetch(`/corpora/{{ corpus.hashid }}/unfollow`, {method: 'POST', headers: {Accept: 'application/json'}})
.then(

141
app/templates/main/news.html.j2
View File

@ -9,7 +9,6 @@
</div>
<div class="col s12">
<div id="aggregated-news"></div>
<div class="card" id="april-2022-update">
<div class="card-content">
<span class="card-title">April 2022 update</span>
@ -127,143 +126,3 @@
</div>
</div>
{% endblock page_content %}
{% block scripts %}
{{ super() }}
<script>
function getMastodonStatuses() {
return new Promise((resolve, reject) => {
fetch(`https://fedihum.org/api/v1/accounts/109386364241901080/statuses`, {method: 'GET', headers: {Accept: 'application/json'}})
.then((response) => {
if (!response.ok) {reject(response);}
return response.json();
})
.then((statuses) => {resolve(statuses);})
});
}
function getBisBlogsEntries() {
return new Promise((resolve, reject) => {
fetch(`https://blogs.uni-bielefeld.de/blog/uniintern/feed/entries/atom?cat=%2FAllgemein`, {method: 'GET', headers: {Accept: 'application/xml'}})
.then((response) => {
if (!response.ok) {reject(response);}
return response.text();
})
.then((responseText) => {return new DOMParser().parseFromString(responseText, 'application/xml');})
.then((xmlDocument) => {return xmlDocument.toObject();})
.then((feed) => {resolve(feed);});
});
}
function sortAggregatedNews(a, b) {
let aDate;
let bDate;
switch (a.source) {
case 'mastodon':
aDate = new Date(a.created_at);
break;
case 'big-blogs':
aDate = new Date(a.published);
break;
default:
throw new Error('Unknown source');
}
switch (b.source) {
case 'mastodon':
bDate = new Date(b.created_at);
break;
case 'big-blogs':
bDate = new Date(b.published);
break;
default:
throw new Error('Unknown source');
}
return bDate - aDate;
}
function aggregateNews() {
return new Promise((resolve, reject) => {
Promise.all([getMastodonStatuses(), getBisBlogsEntries()])
.then(
(responses) => {
console.log(responses[1]);
let mastodonStatuses = responses[0].map((obj) => {return { ...obj, source: 'mastodon'}});
let bisBlogsEntries = responses[1].feed.entry.map((obj) => {return { ...obj, source: 'big-blogs'};});
let aggregatedNews = [...mastodonStatuses, ...bisBlogsEntries];
aggregatedNews.sort(sortAggregatedNews);
resolve(aggregatedNews);
},
(error) => {reject(error);}
);
});
}
function mastodonStatusToElement(status) {
let date = new Date(status.created_at).toLocaleString('en-US');
let newsElement = Utils.HTMLToElement(
`
<div class="row">
<div class="col s11">
<div class="card white-text" style="background-color:#5D50E7; border-radius:10px;">
<div class="card-content">
<span class="card-title">New Actitvity on Mastodon</span>
<p><i>Published on ${date}</i></p>
<br>
<p>${status.content}</p>
</div>
</div>
</div>
<div class="col s1">
<img src="https://joinmastodon.org/logos/logo-purple.svg" alt="Mastodon" class="responsive-img hide-on-small-only" style="width:70%; margin-top:30px;">
</div>
</div>
`
);
return newsElement;
}
function bisBlogsEntryToElement(entry) {
let date = new Date(entry.published).toLocaleString('en-US');
let newsElement = Utils.HTMLToElement(
`
<div class="row">
<div class="col s1">
<img src="https://blogs.uni-bielefeld.de/blog/uniintern/resource/themabilder/unilogo-square.svg" alt="Bielefeld University Blogs" class="responsive-img hide-on-small-only" style="width:70%; margin-top:40px;">
</div>
<div class="col s11">
<div class="card" style="background-color: #A5BDCC; border-radius:10px;">
<div class="card-content">
<span class="card-title">${entry.title['#text']}</span>
<p><i>Published on ${date}</i></p>
<br>
<p>${entry.content['#text']}</p>
</div>
</div>
</div>
</div>
`
);
let newsImageElements = newsElement.querySelectorAll('img');
for (let newsImageElement of newsImageElements) {
newsImageElement.classList.add('responsive-img');
}
return newsElement;
}
let aggregatedNewsElement = document.querySelector('#aggregated-news');
aggregateNews().then((aggregatedNews) => {
for (let item of aggregatedNews) {
let newsElement;
switch (item.source) {
case 'mastodon':
newsElement = mastodonStatusToElement(item);
break;
case 'big-blogs':
newsElement = bisBlogsEntryToElement(item);
break;
default:
throw new Error('Unknown source');
}
aggregatedNewsElement.appendChild(newsElement);
}
});
</script>
{% endblock scripts %}

269
app/templates/main/news_new.html.j2 Normal file
View File

@ -0,0 +1,269 @@
{% extends "base.html.j2" %}
{% from "main/_breadcrumbs.html.j2" import breadcrumbs with context %}
{% block page_content %}
<div class="container">
<div class="row">
<div class="col s12">
<h1 id="title">{{ title }}</h1>
</div>
<div class="col s12">
<div id="aggregated-news"></div>
<div class="card" id="april-2022-update">
<div class="card-content">
<span class="card-title">April 2022 update</span>
<p>Dear users</p>
<br>
<p>
with the April 2022 update we have improved nopaque in all places.
We have significantly reworked our backend code to utilize our servers more efficiently,
integrated a new service, updated all previously existing ones, rewrote a lot of code and made a few minor design improvements.
</p>
<br>
<span class="card-title">Where is my Job data?</span>
<p>
At the beginning of the year, we realized that our storage limit had been reached.
This was the time when some users may have noticed system instabilities.
We were fortunately able to temporarily solve this problem without data loss
by deleting some non-nopaque related data on our system (yes we also do <a href="https://digital-history.uni-bielefeld.de">other things then nopaque</a>).
In order to not face the same problem again, we had to dedicate ourselves to a long-term solution.
This consists of deleting all previous job data with this update and henceforth storing new job data
only for three months after job creation (important note: <b>corpora are not affected</b>).
All job data prior to this update has been backed up for you,
feel free to contact us at nopaque@uni-bielefeld.de if you would like to get this data back.
</p>
<br>
<span class="card-title">What's new?</span>
<p>
By partnering up with <a href="https://readcoop.eu/transkribus/?sc=Transkribus">Transkribus</a> we reached one of our long term goals: integrate a HTR service into nopaque.
The <a href="{{ url_for('services.transkribus_htr_pipeline') }}">Transkribus HTR Pipeline</a> service is implemented as a kind of proxied service where the work is split between Transkribus and us.
That means we do the preprocessing, storage and postprocessing, while Transkribus handles the HTR itself.
</p>
<br>
<p>
One of the changes in the background was to fix our performance issues. While implementing the <a href="{{ url_for('services.transkribus_htr_pipeline') }}">Transkribus HTR Pipeline</a> service we
found some optimization potential within different steps of our processing routine. These optimizations are now also
available in our <a href="{{ url_for('services.transkribus_htr_pipeline') }}">Tesseract OCR Pipeline</a> service, resulting in a speed up of about 4x.
For now we are done with the most obvious optimizations but we may include more in the near future, so stay tuned!
</p>
<br>
<p>
The next step was to reorganize our <a href="{{ url_for('services.corpus_analysis') }}">Corpus Analysis</a> code. Unfortunatly it was a bit messy, after a complete rewrite we are
now able to query a corpus without long loading times and with better error handling, resulting in way more stable user experience.
The Corpus Analysis service is now modularized and comes with 2 modules that recreate and extend the functionality of the old service.<br>
For now we had to disable the Query Result viewer, the code was based on the old Corpus Analysis service and will be reintegrated as a module to the Corpus Analysis.
</p>
<br>
<p>
The <a href="{{ url_for('services.spacy_nlp_pipeline') }}">spaCy NLP Pipeline</a> service got some love in the form of smaller updates too.
This is important preliminary work to support more models/languages that does not provide the full set of linguistic features (lemma, ner, pos, simple_pos). It still needs some testing and tweaking but will be ready soon!
</p>
<br>
<p>
Last but not least we made some design changes. Now you can find colors in places where we had just black and white before.
Nothing big but the new colors will help you identify ressources more efficient!
</p>
<br>
<span class="card-title">Database cleanup</span>
<p>
We may be a bit late with our spring cleaning but with this update we tidied up within our database system.
This means we deleted old corpora with no corpus files, unconfirmed user accounts and in general unnecessary data fields.
</p>
<br>
<p>
That's it, thank you for using nopaque! We hope you like the update and appreciate all your past and future feedback.
</p>
</div>
</div>
</div>
<div class="col s12">
<div class="card" id="maintenance">
<div class="card-content">
<span class="card-title">Maintenance</span>
<p>Dear users</p>
<br>
<p>Currently we are rewriting big parts of our project infrastructure. Due to this the following features are not available:</p>
<ul>
<li>Corpus export and import</li>
<li>Query result export, import and view</li>
</ul>
<p>We hope to add these features back in the near future, until then check out our updated corpus analysis.</p>
</div>
</div>
</div>
<div class="col s12">
<div class="card" id="nlp-removed-language-support">
<div class="card-content">
<span class="card-title">Natural Language Processing removed language support</span>
<p>Dear users</p>
<br>
<p>Not all language models support all features we utizlize in our NLP service. Thats why we had to drop them, as soon as they meet our requirements we will add them back!</p>
</div>
</div>
</div>
<div class="col s12">
<div class="card" id="beta-launch">
<div class="card-content">
<span class="card-title">nopaque's beta launch</span>
<p>Dear users</p>
<br>
<p>A few days ago we went live with nopaque. Right now nopaque is still in its Beta phase. So some bugs are to be expected. If you encounter any bugs or some feature is not working as expected please send as an email using the feedback button at the botton of the page in the footer!</p>
<p>We are happy to help you with any issues and will use the feedback to fix all mentioned bugs!</p>
</div>
</div>
</div>
</div>
</div>
{% endblock page_content %}
{% block scripts %}
{{ super() }}
<script>
function getMastodonStatuses() {
return new Promise((resolve, reject) => {
fetch(`https://fedihum.org/api/v1/accounts/109386364241901080/statuses`, {method: 'GET', headers: {Accept: 'application/json'}})
.then((response) => {
if (!response.ok) {reject(response);}
return response.json();
})
.then((statuses) => {resolve(statuses);})
});
}
function getBisBlogsEntries() {
return new Promise((resolve, reject) => {
fetch(`https://blogs.uni-bielefeld.de/blog/uniintern/feed/entries/atom?cat=%2FAllgemein`, {method: 'GET', headers: {Accept: 'application/xml'}})
.then((response) => {
if (!response.ok) {reject(response);}
return response.text();
})
.then((responseText) => {return new DOMParser().parseFromString(responseText, 'application/xml');})
.then((xmlDocument) => {return xmlDocument.toObject();})
.then((feed) => {resolve(feed);});
});
}
function sortAggregatedNews(a, b) {
let aDate;
let bDate;
switch (a.source) {
case 'mastodon':
aDate = new Date(a.created_at);
break;
case 'big-blogs':
aDate = new Date(a.published);
break;
default:
throw new Error('Unknown source');
}
switch (b.source) {
case 'mastodon':
bDate = new Date(b.created_at);
break;
case 'big-blogs':
bDate = new Date(b.published);
break;
default:
throw new Error('Unknown source');
}
return bDate - aDate;
}
function aggregateNews() {
return new Promise((resolve, reject) => {
Promise.all([getMastodonStatuses(), getBisBlogsEntries()])
.then(
(responses) => {
console.log(responses[1]);
let mastodonStatuses = responses[0].map((obj) => {return { ...obj, source: 'mastodon'}});
let bisBlogsEntries = responses[1].feed.entry.map((obj) => {return { ...obj, source: 'big-blogs'};});
let aggregatedNews = [...mastodonStatuses, ...bisBlogsEntries];
aggregatedNews.sort(sortAggregatedNews);
resolve(aggregatedNews);
},
(error) => {reject(error);}
);
});
}
function mastodonStatusToElement(status) {
let date = new Date(status.created_at).toLocaleString('en-US');
let newsElement = Utils.HTMLToElement(
`
<div class="row">
<div class="col s11">
<div class="card white-text" style="background-color:#5D50E7; border-radius:10px;">
<div class="card-content">
<span class="card-title">New Actitvity on Mastodon</span>
<p><i>Published on ${date}</i></p>
<br>
<p>${status.content}</p>
</div>
</div>
</div>
<div class="col s1">
<img src="https://joinmastodon.org/logos/logo-purple.svg" alt="Mastodon" class="responsive-img hide-on-small-only" style="width:70%; margin-top:30px;">
</div>
</div>
`
);
return newsElement;
}
function bisBlogsEntryToElement(entry) {
let date = new Date(entry.published).toLocaleString('en-US');
let newsElement = Utils.HTMLToElement(
`
<div class="row">
<div class="col s1">
<img src="https://blogs.uni-bielefeld.de/blog/uniintern/resource/themabilder/unilogo-square.svg" alt="Bielefeld University Blogs" class="responsive-img hide-on-small-only" style="width:70%; margin-top:40px;">
</div>
<div class="col s11">
<div class="card" style="background-color: #A5BDCC; border-radius:10px;">
<div class="card-content">
<span class="card-title">${entry.title['#text']}</span>
<p><i>Published on ${date}</i></p>
<br>
<p>${entry.content['#text']}</p>
</div>
</div>
</div>
</div>
`
);
let newsImageElements = newsElement.querySelectorAll('img');
for (let newsImageElement of newsImageElements) {
newsImageElement.classList.add('responsive-img');
}
return newsElement;
}
let aggregatedNewsElement = document.querySelector('#aggregated-news');
aggregateNews().then((aggregatedNews) => {
for (let item of aggregatedNews) {
let newsElement;
switch (item.source) {
case 'mastodon':
newsElement = mastodonStatusToElement(item);
break;
case 'big-blogs':
newsElement = bisBlogsEntryToElement(item);
break;
default:
throw new Error('Unknown source');
}
aggregatedNewsElement.appendChild(newsElement);
}
});
</script>
{% endblock scripts %}