Age | Commit message (Collapse) | Author | Files | Lines |
|
The urlencoded string should only be used for href, but it was being
applied and used in printed output, like the keywords textbox.
This fixes a problem that was reported in
FS#12791 - Faster navigation withing package search listing by page numbers
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|
|
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|
|
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|
|
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|
|
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|
|
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|
|
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|
|
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|
|
Adapted from http://wiki.archlinux.org/index.php/AUR_Translation
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|
|
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|
|
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|
|
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|
|
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|
|
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|
|
Don't show extra search results link if there is only one page of results.
Fix indenting and coding style.
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|
|
This is to present things in a bit more logical manner with the
search results page navigation.
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|
|
On the search results page the number of pages is displayed and functions
as a link, for faster navigation (not just the usual "Less/More" buttons.
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|
|
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|
|
The web interface was handling comments in the PKGBUILD variable fields
(such as 'source','depends',etc...) differently from makepkg, because
makepkg ignores the rest of the current line if there is a # character,
while the web interface parsed that as well, and listed the words of the
comment as source files.
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|
|
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|
|
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|
|
Use the APC cache to store all of the counts and the recently updated
package list in a cache, which cuts down on the number of database queries
needed. If the data isn't perfectly up to date we will survive.
This version of the patch will also cache the relevant counts for individual
logged-in users and is more careful about checking whether the value
actually exists in the cache by using the status reference to apc_fetch().
Signed-off-by: Dan McGee <dan@archlinux.org>
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|
|
Also fix a translation string.
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|
|
This makes the site look better in Opera.
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|
|
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|
|
Change layout in the process.
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|
|
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|
|
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|
|
Just like the previous patch for account_from_sid() over-usage.
Signed-off-by: Dan McGee <dan@archlinux.org>
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|
|
No need to call this function way too often, especially when on the package
list page where it could be called up to once per row.
Signed-off-by: Dan McGee <dan@archlinux.org>
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|
|
Signed-off-by: Dan McGee <dan@archlinux.org>
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|
|
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|
|
There is really no need to use persistent connections to the database in
this day and age. Most PHP development guides recommend against it, and the
new mysqli interface doesn't even include the functionality.
Add a matching but currently unused db_disconnect() function while we are at
it. Reference counting will cover us for the most part, however.
Signed-off-by: Dan McGee <dan@archlinux.org>
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|
|
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|
|
Clean up a couple of notices.
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|
|
Changed all the normal variables to their $_GET counterparts so everything is destructively changing the original variables, there should be no issue with this. If there I guess we need to consider making mkurl use a custom array of variables rather than $_GET
Signed-off-by: Callan Barrett <wizzomafizzo@gmail.com>
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|
|
Also changed it around a little bit so it's not tied down the search results function
Signed-off-by: Callan Barrett <wizzomafizzo@gmail.com>
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|
|
Signed-off-by: Callan Barrett <wizzomafizzo@gmail.com>
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|
|
Signed-off-by: Callan Barrett <wizzomafizzo@gmail.com>
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|
|
Quick release to fix the chmod issue. GO TEAM!!
Signed-off-by: Callan Barrett <wizzomafizzo@gmail.com>
|
|
Author: Samed Beyribey <ras0ir@eventualis.org>
Signed-off-by: Callan Barrett <wizzomafizzo@gmail.com>
|
|
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|
|
Removing trailing colons and whitespace makes them more usable.
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
Signed-off-by: Callan Barrett <wizzomafizzo@gmail.com>
|
|
Sorting, and viewing more results resets the search
if you search for '#' in 1.5.3. This should fix that bug.
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
Signed-off-by: Callan Barrett <wizzomafizzo@gmail.com>
|
|
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
Signed-off-by: Callan Barrett <wizzomafizzo@gmail.com>
|
|
I forgot about the forms.
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
Signed-off-by: Callan Barrett <wizzomafizzo@gmail.com>
|
|
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|
|
Logic is separated from html as much as possible, all html in a template
Signed-off-by: Callan Barrett <wizzomafizzo@gmail.com>
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|
|
All we're looking for now is an @ in any email address to fix problem with not being able to have + in addresses and just because I see no advantage to having any sort of stringent validation
Signed-off-by: Callan Barrett <wizzomafizzo@gmail.com>
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|
|
This includes only the requested language for each page and
makes top level language include files obsolete.
Signed-off-by: Loui Chang <louipc.ist@gmail.com>
|