PublicInbox::WWW (PSGI interface) design notes

URL and anchor naming

### Unstable endpoints
/$INBOX/?r=$GIT_COMMIT                 -> HTML only
/$INBOX/new.atom                       -> Atom feed

#### Optional, relies on Search::Xapian (or Xapian SWIG binding)
/$INBOX/$MESSAGE_ID/t/                 -> HTML content of thread (nested)
/$INBOX/$MESSAGE_ID/T/                 -> HTML content of thread (flat)
	#u          location of $MESSAGE_ID in URL
	#m<SHA-1>   per-message links, where <SHA-1> is of the Message-ID
	            of each message (stable)
	#s<NUM>     relative numeric position of message in thread (unstable)
	#i<...>     diffstat location for patch emails
	#Z?<...>    per-file diff header location for patch emails

/$INBOX/$MESSAGE_ID/t.atom             -> Atom feed for thread
/$INBOX/$MESSAGE_ID/t.mbox.gz          -> gzipped mbox of thread

/$INBOX/$GIT_OID/s/                    -> "git show" (via "git apply")
	This endpoint requires "coderepo" entries configured for
	a given inbox.  It can recreate ("solve") blobs from
	patch emails using Xapian and git-apply(1).  It can also
	display non-blob content, but that remains a

/$INBOX/$GIT_OID/s/$FILENAME           -> "git show", raw output
	As above, but shows the raw (usually text/plain) output.

### Stable endpoints
/$INBOX/$MESSAGE_ID/                   -> HTML content
	#r          location of the current message in thread skeleton
	            (requires Xapian search)
	#b          start of the message body (linked from thread skeleton)

/$INBOX/$MESSAGE_ID                    -> 301 to /$INBOX/$MESSAGE_ID/
/$INBOX/$MESSAGE_ID/raw                -> raw mbox
/$INBOX/$MESSAGE_ID/#R                 -> HTML reply instructions

# Covering up a pre-1.0 design mistake:
/$INBOX/$MESSAGE_ID/f/                 -> 301 to /$INBOX/$MESSAGE_ID/

### Legacy endpoints (may be ambiguous given Message-IDs with similar suffixes)
/$INBOX/m/$MESSAGE_ID/                 -> 301 to /$INBOX/$MESSAGE_ID/
/$INBOX/m/$MESSAGE_ID.html             -> 301 to /$INBOX/$MESSAGE_ID/
/$INBOX/m/$MESSAGE_ID.txt              -> 301 to /$INBOX/$MESSAGE_ID/raw
/$INBOX/f/$MESSAGE_ID.html             -> 301 to /$INBOX/$MESSAGE_ID/
/$INBOX/f/$MESSAGE_ID.txt [1]          -> 301 to /$INBOX/$MESSAGE_ID/raw

/$INBOX/atom.xml [2]                   -> identical to /$INBOX/new.atom

Additionally, we support git clone/fetch over HTTP (dumb and smart):

	git clone --mirror http://$HOSTNAME/$INBOX

FIXME: we must refactor/cleanup/add tests for most of our CGI before
adding more endpoints and features.

[1] These URLs were never linked, but only exist as a convenience to folks
    who edit existing URLs

[2] Do not make this into a 301 since feed readers may not follow them as well
    as normal browsers do.

Encoding notes

Raw HTML and XML should only contain us-ascii characters which render
to UTF-8.  We must not rely on users having the necessary fonts
installed to render uncommon characters.

Plain text (raw message) endpoints display in the original encoding(s)
of the original email.

Offline friendly

The "/t/", "/T/", "t.mbox.gz" endpoints are designed to be
useful for reading long threads for users with intermittent
connections or saved for offline viewing.

Date displays are always absolute, not the "X hours ago"
pattern commonly seen because readers may be reading a
previously-saved or cached copy.

HTML URLs end with '/' or "$FILENAME.html".  The reason many
URLs end with the '/' character is so it can trivially be saved
to a directory via wget or similar tools as "index.html", making
it easy to mirror all files ending in ".html" using any static
web server.

Guidelines for using limited HTML

We mainly use HTML for linking pages together with <a>.
We also set <title> to make window management easier.

We favor <pre>-formatted text since public-inbox is intended as a place
to share and discuss patches and code.  Unfortunately, long paragraphs
tends to be less readable with fixed-width serif fonts which GUI
browsers default to.

* No graphics, images, or icons at all.  We tolerate, but do not
  encourage the use of GUIs.

* No setting font sizes, power to users to decide those.
  We will include and document <span class=?> to support colors
  for user-supplied CSS.

* Only one font type: fixed.  This is for accessibility, we must
  not blow certain elements out-of-proportion with different
  fonts on the page when a reader increases font size.

* Bold and underline elements are OK since they should render fine
  regardless of chosen font and gracefully degrade if a display does
  not support them.  Italics and strike-through elements must be
  avoided as they do not render well with some displays or user-chosen

* No JavaScript. JS is historically too buggy and insecure, and we will
  never expect our readers to do either of the following:
  a) read and audit all our code for on every single page load
  b) trust us and and run code without reading it

* We only use CSS for one reason: wrapping pre-formatted text
  This is necessary because unfortunate GUI browsers tend to be
  prone to layout widening from unwrapped mailers.
  Do not expect CSS to be enabled, especially with scary things like:

  However, we will try to make it easy for users to supply their
  own colors via user-side CSS.

CSS classes (for user-supplied CSS)

See examples in contrib/css/ and lib/PublicInbox/
(or soon)