normalize .html and .txt urls

[magicmark]

for big json dumps, loading the page can take forever and crash the browser, and i usually just want to curl it anyway

but if folks provide me the .html page, i can't easily map to the .txt url

could they share the same id in the url?

[magicmark]

i see the comment from here #62 (comment) which makes sense, but somewhere in def upload(): can we generate an id and share it with the plaintext and html file in s3?

[chriskuehl]

I think we could make the HTML and TXT files have the same name (other than extension of course).

Just to be clear, technically an upload can produce multiple HTML files and TXT files. For example https://i.fluffy.cc/PPMCfvrXGF1fhGWjs6lKqfLn21xHB7HB.html has five files (not counting the JSON files):

• overview HTML: https://i.fluffy.cc/PPMCfvrXGF1fhGWjs6lKqfLn21xHB7HB.html
• file1 txt: https://i.fluffy.cc/9dj2C8KcLGhcN2zwQms447fkT0jcq6Wg.txt
• file1 HTML: https://i.fluffy.cc/BZFDgV86TRkq48Ks24tTRlk00pZCggcz.html
• file2 txt: https://i.fluffy.cc/C8rltbZk5M2dbRNxLSzMcw44rKF4qCtN.txt
• file2 HTML: https://i.fluffy.cc/4Wb3f6FWrjkZjVP7XGpXbFvBNMHMHwRT.html

If the file1 txt+html and the file2 txt+html had the same name, that would do what you want?

[magicmark]

heck yeah!

so 'direct link' and 'view text' would be the same url, apart from the suffix

[chriskuehl]

I think #177 will do it. I'm traveling at the moment but I'll test this more thoroughly and merge it soon.

[chriskuehl]

Released in v1.40.0.

Here's an example on the new version: https://i.fluffy.cc/pLRkPrVHQdjqXg5697pwcncxs3VG9Bkv.html