about summary refs log tree commit homepage
DateCommit message (Collapse)
2014-04-15scripts/import_gmane_spool: preserve delivery order
Unfortunately, this means we get rid of parallelization, as we need to preserve delivery order so HTML indices look chronological. Order may also affect spam filtering and training, too.
2014-04-15doc: add notes on scalability
Fortunately, most mailing lists will never grow too large.
2014-04-14doc: fold philosophy into the README
Hopefully this makes the scope and intent of the project clearer.
2014-04-14rename list from "bugs" to "meta"
"bugs" might confuse and limit the discussion, so "meta" it is!
2014-04-14mda: add most RFC 2919 and 2369 headers
These probably make sense even though we do not handle delivery ourselves. It can aid in searching/filtering/tagging of messages.
2014-04-14derive -primary_address in config
This may be useful for generating List-Id headers and HTML pages.
2014-04-14cgi: fix up top-level index
We do not have all messages in the top-level index (and we need to adjust the test while we're at it).
2014-04-14cgi: 301 for list-indices without trailing slash
It is common to type upper-level URLs without the slash, redirect users to the correct page for usability.
2014-04-14t/cgi: cleanup: no need for additional block
Not sure what I was thinking...
2014-04-12cgi: ensure we unescape MIDs correctly in URLs
MIDs may have strange characters in them, so we need to handle escaping/unescaping properly to avoid broken links or worse.
2014-04-12cgi: avoid parsing ENV directly for PATH_INFO
This might make it easier to go to non-CGI things.
2014-04-12cgi: relax path restriction for top-level
We may have something like /foo.cgi/m/$MID.html in there.
2014-04-12cgi: rename to have .cgi suffix
This makes it easier to configure for systems which determine a script is a CGI script based on suffix.
2014-04-11scripts/import_gmane_spool: misc updates
We may promote this to be a real script, since public-inbox-mda is idempotent.
2014-04-11view: trim_message_id consistently
Message-IDs could have embedded '<' and '>'
2014-04-11config: support multiple addresses for a inbox
This makes it possible to gradually migrate to new address in case of list name changes, and is one step closer to operating in "stealth hijack mode" :)
2014-04-11add spam/ham learning wrapper script
This is essential for integrating into my inotify-based spam training setup.
2014-04-11Documentation/design_notes: more updates
Laziness \o/ (Then impatience and hubris :)
2014-04-11filter: clarify regular expression
I often forget the subtleties of Perl regexps and newlines, so I suspect others do, too. Use explicit capture so it's more familiar to users of non-Perl regexps.
2014-04-11cgi: wire up HTML pages for messages
These need better tests and verification, but it's something for now.
2014-04-11cgi: update feed/view and tests for shorter URLs
Code should be consistent with the design docs (and we will need better tests).
2014-04-11cgi: add one-line descriptions for subroutines
Hopefully it's slightly easier-to-follow this way.
2014-04-11cgi: /$LISTNAME/ and /$LISTNAME/index.html are equal
This prevents ambiguity when switching URLs between static file servers and CGI. The /$LISTNAME/index.html URL appearing in the wild is inevitable because of our static file server support. Worst yet, there's no easy/consistent way to get all installations detect and 301 them to the shorter /$LISTNAME/. So we make the CGI support /$LISTNAME/index.html. The downside of this is the potential duplicate entry in all caches.
2014-04-11cgi: be strict about UTF-8 encoding in HTML and XML
Hopefully this forces us to generate valid UTF-8 data.
2014-04-11filter: use IPC::Run and improve lynx error handling
We may occasionally encounter horrid HTML which lynx cannot handle, so improve error reporting.
2014-04-11INSTALL: add lynx to install requirements
We need it for converting HTML, unfortunately.
2014-04-10INSTALL: update with Mail::Thread dependency
While we're at it, sort Makefile.PL and add a note to update INSTALL, too.
2014-04-10cgi: implement get_mid_txt
This is essential when telling people to use something like: curl $URL | git am
2014-04-10doc/design_www: add 301 URLs
We'll probably support these so they're easier-to-type and share.
2014-04-10cgi: wire up index + tests
Remove the specified /all.html while we're at it, we only have /all.atom.xml because it's convenient for feed readers.
2014-04-10cgi: do not specify charset in Atom HTTP header
The feed itself already specifies it in XML, and we risk confusing clients if XML::Atom::SimpleFeed changes in the future. This also increases consistency for CGI vs static-file serving.
2014-04-10cgi: remove some redundant logic
We'll be reusing more validation logic for per-message and per-thread pages.
2014-04-09doc/design_www: shorten URLs as much as possible
Message-IDs are extremely long already, so try to keep them short here.
2014-04-09t/mda: additional precheck tests
We did not test as well as we should have.
2014-04-09mda: set GIT_AUTHOR_DATE in commits as well
While we're at it, write some quick tests.
2014-04-09mda: set GIT_{COMMITTER,AUTHOR}_{NAME,EMAIL} env
This can make it easy to query via "git log --author=..." without extracting each message.
2014-04-09config: include listname on lookup
We will be using it when setting GIT_COMMITTER_NAME
2014-04-09mda: prevent duplicate Message-IDs from appearing
For practical purposes, Message-IDs are unique and duplicates do not appear unless client software is broken.
2014-04-09Makefile.PL: add parallel test target
These tests were designed to run in parallel.
2014-04-09doc: split out philosophy to a different page
Hopefully a little easier to find for clients and not admins running servers. While we're at it, expand design_notes.
2014-04-09preliminary HTML index generation
Using JWZ threading might work decently for this. Haven't checked in lynx, yet.
2014-04-09precheck: stricter checks including min length
We should reject values which are too short to be useful or sane.
2014-04-08precheck: reject messages with no subject
Composers may screw up and leave the subject out, so reject those messages.
2014-04-08feed: filter out each_recent_blob wrapper
We will need it for HTML indices, too.
2014-04-08design_notes: various updates, including "why git?"
Things to keep in mind when working on this.
2014-04-08scripts/report-spam: explain design decisions
Trying my best to not forget things I wrote this years ago.
2014-04-08doc: various cleanups all around
Most notably, the INSTALL is geared towards potential server admins, whereas the README is also for interested "drive-by" readers.
2014-04-08cgi: cleanup dependencies
We do not need to use CGI::Util internals here.
2014-04-07cgi: make internal interface more Plack-like
This should make it easier to support non-CGI uses, as well as making it easier to generate static sites.
2014-04-07feed: generate takes a hashref for args
Passing a giant argument list is to error prone and hard-to-document.