Sign inSign up

coco/factset-reader

By coco

•Updated almost 9 years ago

This application reads files from Factset FTP server and stores them to Amazon S3

Image
0

7.9K

coco/factset-reader repository overview

DECOMISSIONED Using Factsets own data loader instead

⁠Factset Reader (factset-reader)

Circle CI Coverage Status

A service for reading files from Factset FTP (SFTP) server and writing them into an Amazon S3 bucket

⁠Installation

  1. go get github.com/kardianos/govendor
  2. go get github.com/Financial-Times/factset-reader
  3. cd $GOPATH/src/github.com/Financial-Times/factset-reader
  4. govendor sync

⁠Running

$GOPATH/bin/factset-reader
--awsAccessKey=xxx
--awsSecretKey=xxx
--bucketName=com.ft.coco-factset-data
--s3Domain=s3.amazonaws.com
--port=8080
--factsetUser=xxx
--factsetKey=xxx
--factsetFTP=fts-sftp.factset.com
--factsetPort=6671
--resources=/directory/without/version:fileToDownload1.txt;fileToDownload2.txt

The awsAccessKey, awsSecretKey, bucketName, factsetUser, factsetKey arguments are mandatory, and represent authentication credentials for S3 and Factset FTP server.

The resources argument specifies a comma separated list of archives and files within that archive to be downloaded from Factset FTP server. Because every file is inside an archive, the service will first download the archive, unzip the files you specify, zip a collection of daily/weekly files and upload the resulting zips to s3. A resource has the format archive_path:file1.txt;file2.txt, example: /datafeeds/edm/edm_bbg_ids/edm_bbg_ids:edm_bbg_ids.txt, where /datafeeds/edm/edm_bbg_ids/ is the path of the archive, edm_bbg_ids is the prefix of the zip without versions and edm_bbg_ids.txt is the file to be extracted from this archive. On the Factset FTP server the archive name will contain also the data version, but it is enough for this service to provide the archive name without the version and it will download the latest one.

After downloading the zip files from Factset FTP server, the service will write them to the specified Amazon S3 bucket. The zip files written to S3 will be inside of a folder named by the current date. Depending upon the day there may be both a weekly.zip and daily.zip or just a daily.zip

If creating a new daily.zip or weekly.zip, the loader will also update the index files found in the S3 bucket (daily/weelkly respectively). These contain the keys to the latest versions of each zip.

⁠Endpoints

Force-import (initiate importing manually of all most recent files):

http://localhost:8080/force-import -XPOST

Force-import-weekly (initiate importing manually of all most recent weekly files):

http://localhost:8080/force-import-weekly -XPOST

⁠Admin Endpoints

Health checks: http://localhost:8080/__health

Good to go: http://localhost:8080/__gtg

Tag summary

Content type

Image

Digest

Size

72.8 MB

Last updated

almost 9 years ago

docker pull coco/factset-reader