Evangelist

Evangelist converts PDF to JPEGs quickly using Golang, Ghostscript, and epeg.

To use it, first run the server:

$ go run server.go
# => Serving on 0.0.0.0:8000

Then, make a POST request to the server's root path with two parameters:

1. s3PDFPath: The S3 path that refers to the PDF to convert.
2. s3JPEGPath: The S3 path to upload JPEGs to. This should include a %d symbol that will be replaced by the page number.

Three JPEGs will be uploaded for each page: one small, one normal, and one large. The %d in s3JPEGPath will actually be replaced by "{pageNumber}-small" for the small JPEG, "{pageNumber}" for the normal JPEG and "{pageNumber}-large" for the large JPEG.

For instance, if s3JPEGPath is "split-pages/page%d.jpg", then:

• A small JPEG for page 1 will be uploaded to "split-pages/page1-small.jpg"
• A normal JPEG for page 1 will be uploaded to "split-pages/page1.jpg"
• A large JPEG for page 1 will be uploaded to "split-pages/page1-large.jpg"

Here's an example request:

$ curl -d
"s3PDFPath=exam-pdf/0mzvCOQXkGixnu0LrUGjPuagFevCQ120140203000038.pdf&s3JPEGPath=split-pages/gotest%25d.jpg"
localhost:8000
# => Done

Note that the '%' sign in s3JPEGPath must be escaped as '%25' due to the rules of parsing multipart form content.

The above command downloads the given PDF, converts all 6 pages into 18 JPEGs, and uploads the JPEGs to S3 in ~7.5 seconds.