Reindexing PSS documents to Elasticsearch
Run the pss-json-export.py script against all sets (contents of s3://dpla-pss-json/)
Create a new index with filterable fields for keyword and time period values
curl -XPUT http://[ES_NODE]:9200/[INDEX]\?pretty -H "Content-Type: application/json" -d '
{
"mappings":
{
"properties":
{
"about":
{
"type": "object",
"properties":
{
"disambiguatingDescription":
{
"type": "keyword",
"fields":
{
"not_analyzed":
{
"type": "keyword"
}
}
},
"name":
{
"type": "keyword",
"fields":
{
"not_analyzed":
{
"type": "keyword"
}
}
}
}
}
}
}
}'
Index documents from the ./sets/ directory
find . -name "*updated_*.json" -type f | xargs -I{} sh -c 'echo "$1" "./$(basename ${1%.*}).${1##*.}"' -- {} | xargs -n 2 -P 8 sh -c 'curl -XPOST http://ES_NODE:9200/[INDEX]/doc -H "Content-Type: application/json" -d @"$0"'
Flop the alias to the new index
curl -XPOST http://[ES_NODE]:9200/_aliases -H 'Content-Type: application/json' -d '{"actions":[{"remove" : {"index" : "*", "alias" : "dpla_pss"}},{"add" : { "index" : "[INDEX]", "alias" : "dpla_pss" }}]}'
Test the indexing by getting the facet values for Time Period and Subject filters/facets/dropdowns
curl -XGET http://[NODE]:9200/dpla_pss/_search\?pretty -H "Content-Type: application/json" \
-d '{
"size": "1",
"_source":
{
"includes":
[
"about.disambiguatingDescription", "about.name"
],
"excludes":
[]
},
"query":
{
"match_all":
{}
},
"aggs":
{
"disambiguatingDescriptionAgg":
{
"terms":
{
"field": "about.disambiguatingDescription"
}
}, "nameAgg":
{
"terms":
{
"field": "about.name"
}
}
}
}'
Test sorting by dateCreated chronologically
curl -XGET http://[NODE]:9200/dpla_pss/_search\?pretty -H "Content-Type: application/json" \
-d '{
"size": "500",
"_source":
{
"includes":
[
"dateCreated", "name"
],
"excludes":
[]
},
"query":
{
"match_all":
{}
},
"sort" : [
{ "dateCreated" : {"order" : "desc"}}
]
}'
Prerequisites:
- VirtualBox (Version 4.3)
- Vagrant (Version 1.6)
- vagrant-vbguest (
vagrant plugin install vagrant-vbguest) - Ansible (Version 1.9 or greater; installation instructions)
For installation:
vagrant up
vagrant reload # Because of o/s packages having been upgraded
vagrant ssh
cd /vagrant
To install all gems:
bundle install
Or, to install only the necessary gems:
bundle install --deployment --without dpla_branding production
Finish installation:
bundle exec rake db:migrate
bundle exec rails s
Then access the application at http://localhost:3000/primary-source-sets/
You may re-run the provisioning with vagrant provision. This will upgrade
operating system packages.
Subsequent development:
- Run
bundle updateas necessary
Please note that this information is incomplete and is due to be updated as we flesh out our media functionality:
This project requires an Amazon Web Services account, and uses the S3 cloud container service for media file storage.
The relevant parameters must be defined in settings.local.yml. See
app/config/settings.yml.
The application makes use of one bucket for incoming media files that need to be transcoded, and another for the transcoded derivatives. You should have an IAM account with the necessary permissions on these buckets.
The incoming bucket must have the following CORS configuration, which is managed in Amazon's S3 Management Console.
<?xml version="1.0" encoding="UTF-8"?>
<CORSConfiguration xmlns="http://s3.amazonaws.com/doc/2006-03-01/">
<CORSRule>
<AllowedOrigin>*</AllowedOrigin>
<AllowedMethod>POST</AllowedMethod>
<MaxAgeSeconds>3000</MaxAgeSeconds>
<AllowedHeader>Content-Type</AllowedHeader>
</CORSRule>
</CORSConfiguration>
The destination bucket (for transcoded derivatives and PDFs and images, too) needs the following CORS configuration.
<?xml version="1.0" encoding="UTF-8"?>
<CORSConfiguration xmlns="http://s3.amazonaws.com/doc/2006-03-01/">
<CORSRule>
<AllowedOrigin>*</AllowedOrigin>
<AllowedMethod>POST</AllowedMethod>
<AllowedMethod>GET</AllowedMethod>
<AllowedMethod>HEAD</AllowedMethod>
<MaxAgeSeconds>3000</MaxAgeSeconds>
<AllowedHeader>Content-Type</AllowedHeader>
</CORSRule>
</CORSConfiguration>
The IAM account should have a Policy Document that makes the following allowances. Policy Documents are managed in Amazon's IAM Management Console.
{
"Version": "2012-10-17",
"Statement": [
{
"Sid": "1",
"Effect": "Allow",
"Action": [
"s3:DeleteObject",
"s3:GetObject",
"s3:PutObject",
"s3:GetBucketCORS"
],
"Resource": [
"arn:aws:s3:::incoming-bucket-name/*"
]
},
{
"Sid": "2",
"Effect": "Allow",
"Action": [
"s3:DeleteObject",
"s3:GetObject",
"s3:PutObject",
"s3:GetBucketCORS"
],
"Resource": [
"arn:aws:s3:::destination-bucket-name/*"
]
},
{
"Sid": "3",
"Effect": "Allow",
"Action": [
"s3:ListBucket",
"s3:GetBucketLocation"
],
"Resource": "arn:aws:s3:::incoming-bucket-name"
}
]
}
Note that the "Sid" values can be whatever you want them to be, as long as they
are unique within the Policy Document. The strings incoming-bucket-name and
destination-bucket-name need to be replaced with the actual names of your
buckets.
Your destination bucket should have a CloudFront Distribution, i.e. should be
served up with Amazon's content delivery network. The Domain Name property of
that CloudFront Distribution is what goes into your settings.local.yml as
aws.s3_cloudfront_domain.
The zencoder-fetcher gem can be used during development to complete each upload-and-transcoding operation. It's necessary when your development environment is behind a NAT firewall (the usual case) and Zencoder can't access your notification endpoint.
In the shell that you'll be using during development that has access to the application (e.g. http://localhost:3000/primary-source-sets/), do this:
The zencoder-fetcher gem should be installed with the :development group in
the Gemfile. This will allow its zencoder-fetcher command to be run like
one of your rake or rails commands. You could also install it somewhere
else that has access to your app by running gem install zencoder-fetcher.
Then you'll be able to run the command as follows:
zencoder_fetcher -u http://USER:PASSWORD@localhost:3000/primary-source-sets/video_notifications <API KEY>
... Where USER is zencoder.notification_user and PASSWORD is
zencoder.notification_pass in your settings.yml.
Substitute video_notifications for audio_notifications as necessary.
Note the existence of
our automation project and its pss role
for this application.
This project's VM is best when you need to make quick iterative developments to
the code and want to re-run the test suite. For a better representation of how
the whole stack will function in production, including how deployments will
work, use automation.
The application's configuration files will be managed by you in your working
copy, if you're using the local VM. If you're using automation, and you want
it to "use local source," in other words, use this working copy as the source of
deployments, please be aware that the working copy's configuration files will be
copied into the automation VM, by design.
This project uses ActionMailer to send account-creation confirmation emails. We
use the mailcatcher gem in development to simulate a
mail server, rather than running a real one. Per its
documentation, we do not include it in our Gemfile, and
you need to install it with gem install mailcatcher.
The mail settings for production are ActionMailer smtp_settings key/value
pairs. There are defaults in settings.yml that can be overridden in
settings.local.yml.
If you're running in the development VM (with our Vagrantfile), you'll want to
invoke mailcatcher this way:
mailcatcher --http-ip 0.0.0.0
This ensures that the HTTP interface will be listening on the network interface
that is being forwarded to localhost on your host machine.
To run rspec tests in your console:
rake ci
To run jasmine tests in your console:
rake jasmine:ci
To run a jasmine Webrick server on port 3001, from the console do:
rake jasmine
and then go to http://localhost:3001/ in your browser. Whenever you reload the page, the tests are run again.
To create an initial manager admin account, from the console do:
Admin.new({ email: '[EMAIL]', password: '[PASSWORD]', password_confirmation: '[PASSWORD]', status: 2 }).save
Example:
Admin.new({ email: '[email protected]', password 'secret', password_confirmation: 'secret', status: 2 }).save
The output in the console will contain a web address to visit in order to activate the account. Go to [ROOT]/admins/sign_in to log in.
- Copyright Digital Public Library of America, 2015 -- 2017
- License: MIT