Sort the messages retrieved from an Mnesia archive before selecting the
subset limited by the <max/> value. This makes sure the desired subset
of messages is sent to the client.
If the client doesn't specify a maximum number of messages to retrieve
per page, set a limit of 50 messages. If the client specifies a limit
larger than 250, cap the number to 250 messages.
These limits aren't enforced for MAM v0.2 requests though, as that
version of the XEP doesn't require clients to support RSM. The newer
revisions say that "a server MAY place a reasonable limit on how many
stanzas may be pushed to a client in one request. Whether or not the
client query included a <set/> element, the server MAY simply return its
limited results, modifying the <set/> element it returns appropriately."
If an <index/> is specified in the MAM request, reject the request
rather than ignoring the desired index and returning wrong results.
XEP-0059 says that the server "MAY return a <feature-not-implemented/>
error."
Make sure the binary comparison performed when clients use message UIDs
to page through Mnesia archives yields correct results even if the
specified UIDs don't have the same number of digits as the UIDs of the
stored messages. This way, MAM will continue to work as expected after
migrating from mod_mam_mnesia to mod_mam.
The new "delete_old_mam_messages" command allows for purging all MAM
messages of the specified type older than the specified number of days.
(Currently only implemented for Mnesia archives.)
Enabling "request_activates_archiving" tells mod_mam not to store any
messages for a user until his client issued a MAM request, regardless of
mod_mam's "default" option. Once a MAM request is issued, messages are
archived as usual.
If a message stanza is blocked as per XEP-0016 or XEP-0191, return an
error only if the type of the blocked message is "normal" or "chat".
This makes sure users won't be kicked from MUC rooms when blocking other
participants.
Closes#897.
Don't just use the "put_url" domain name, but also any path components
of the specified URL, to generate a mod_http_upload process name. This
way, a single domain name can be used for multiple virtual hosts by
specifying a "put_url" such as "https://example.com/@HOST@/".
Use <stanza-id/> elements instead of <delay/> tags to check for messages
resent by the stream management code. The <stanza-id/> element is
preferable, as it is added by mod_mam itself.
As per XEP-0313 version 0.2 and newer, advertise the MAM feature in the
service discovery information for the bare account (or MUC room) JID.
Some clients check the server's discovery information instead, so we'll
continue to advertise the feature there as well.
