Instead of returning a command which gets executed and which should
pass through / stream its output data to the client, we're just
returning an instance of Pluf_HTTP_Response. This is needed, because
some SCMs, most noticable monotone, have no locally executable command
to provide a snapshot archive (and probably never will for our kind
of setup).
We therefor added a little BSD-licensed class "ZipArchive" which allows
the creation of pkzip-compatible archives on the fly by letting it eat
the file contents directly feed from the (remote) stdio instance.
Download performance is ok and lies between 15K/s and 110K/s, but at
least we do no longer block the browser while we pre-generate the zip
file server-side.
Thanks to Patrick Georgi for all his work!
this way additional hooks which need these notifications as well
do not override earlier hooks
* optionally include an additional hooks.lua file at the very end
in which custom hooks can be defined
* refactor out the common menu code from several base*.html classes into main-menu.html
and put the raw links into a unordered list (which makes it easier to create dynamic
menus as we cannot insert block items into inline items in strict mode)
* query the list of available projects on every request and set them for every template
in Middleware.php
* make the popup menu pretty
need of having to start the list with a space, here it was actually
needed code-wise. I rewrote the code so the space was not needed
and the documentation follows the actual behaviour.
has to start with a space - its even a little quirky since the code
expands that to a regex like "/\.(|ext1|ext2...)$/", so in theory files
which end up with a dot and without an extension would be allowed by this.
For now we just fix the comment and indentation of the default
configuration option.
- getTree(), getPathInfo(): use the new extended manifest format
and save the calls to query file sizes from contents as well
as the calls to determine the revision in which a file changed
at last
rest and write the rest in full_message - just like we do it for log and
everything else. This is ugly, really ugly, because it assumes something
on the format of a commit message, which might not be true at all for
some project, but this is something Loic has to decide (see also issue 491
and issue 535)
part of another branch whose certs haven't been pushed into the server
yet, so we need to skip these revisions while going back in time
for the changelog. The initial revision however must carry a branch
cert, otherwise we have nothing to "follow".
to the same signal also for the initial setup, since the memberships
haven't been added at the time the create signal is thrown
* my array references goo was slightly stupid (the usage of foreach
is of course hazardous in cases like this)
* always insert a trailing new line in write-permissions and skip
read-in newlines from being processed
indentifiers in IDF - the SCM function isValidRevision has been replaced
by a validateRevision() method which returns one of three states,
valid, invalid or ambiguous.
The source view can then act accordingly and display disambiguate view
for the latter, so the user can select for which revision he actually
wants to execute the requested action. Also, invalid revisions now lead
to another separate view, telling the user that it is invalid / does
not exist and pointing him optionally to the help page where he can read
further how to access his repository to push the first changes into.
(partially resolves issue 525)
