Redirect old url structure to new structure

[epicfaace]

https://stanforddailyarchive.com/cgi-bin/stanford?a=q&hs=1&r=1&results=1&txf=txIN&ssnip=txt&o=20&dafdq=&dafmq=&dafyq=&datdq=&datmq=&datyq=

[epicfaace]

https://stanforddailyarchive.com/cgi-bin/stanford?a=d&d=stanford19920219-01.2.16#

[epicfaace]

URL structures:

Need to redirect

https://stanforddailyarchive.com/cgi-bin/stanford?a=q&r=1&results=1&txf=txIN&txq=test - query "test"
https://stanforddailyarchive.com/cgi-bin/stanford?a=d&d=stanford19770519-01 - view for stanford19770519-01

Don't need to redirect

https://stanforddailyarchive.com/cgi-bin/stanford?a=d&d=stanford19770519-01.2.25 - view for stanford19770519-01, zoomed in to article # 2.25 (probably can just redirect to the page in general)
https://stanforddailyarchive.com/cgi-bin/imageserver/imageserver.pl?oid=stanford19820429-01&getpdf=true - issue pdf for stanford19820429-01

We need to also make sure that:

• Searching for a particular query with filters saves that state in the query string, and changes the page title accordingly.
• Viewing a particular article saves that state in the query string, and changes the page title accordingly.
• All redirects should be a HTTP 301.

Ways to redirect

• Serve stanforddailyarchive.com from an S3 bucket with routing rules (https://stackoverflow.com/questions/43835731/301-redirect-for-specific-page-with-cloudfront and https://docs.aws.amazon.com/AmazonS3/latest/dev/how-to-page-redirect.html)
• Serve stanforddailyarchive.com behind our wordpress site, and have a wordpress plugin that parses the query string and then redirects.
• Serve stanforddailyarchive.com with an AWS Lambda function which parses the query string and redirects.

thoughts @hesyifei ?

[hesyifei]

I think WordPress is overkill and it might complicate things (I'm sure a simple PHP will work as well/better). I'll look into the rest of the options.

Also, are we using archive.stanforddaily.com or stanforddailyarchive.com for the new site?

[epicfaace]

I think we should use archive.stanforddaily.com for the new site and keep stanforddailyarchive.com as the old site. It makes more sense to make archives on a subdomain of our site.

Not sure about a simple PHP because then it adds complexity to deployment (i.e. one more thing to deploy)

[hesyifei]

We also need to search some most common results for stanforddailyarchive.com in Google and make sure those are redirected correctly.

---

Structure:

Search page

Old: https://stanforddailyarchive.com/cgi-bin/stanford?a=q&hs=1&r=1&results=1&txq=test&txf=txIN&ssnip=txt&o=50&dafdq=03&dafmq=02&dafyq=1892&datdq=27&datmq=09&datyq=2001&e=01-01-1892-27-09-2001--en-50--1--txt-txIN-test------

New: https://archive.stanforddaily.com/search?page=1&pagelen=50&q=test&year_end=2001&year_start=1892

o -> pagelen
datyq -> year_end
dafyq -> year_start
txq -> q

Paper page

Old: https://stanforddailyarchive.com/cgi-bin/stanford?a=d&d=stanford19770519-01

New: https://archive.stanforddaily.com/paper/1977-05-19

d=stanford[year][month][day]-01 -> paper/[year]-[month]-[day]

Cannot be converted

Old: https://stanforddailyarchive.com/cgi-bin/stanford?a=d&d=stanford19770519-01.2.5&e=-------en-20--1--txt-txIN-------

New: https://archive.stanforddaily.com/paper/1977-05-19#page=1&section[]=MODSMD_ARTICLE4

It is a link that highlights a specific section of the paper. The ID system is different (e.g., old is stanford19770519-01.2.5 but new is MODSMD_ARTICLE4).