index
:
apt.git
1.4.y+ios
1.8.2.z+ios
1.8.2.z+ios+json
1.8.y+ios
1.8.y+ios+json
cydia
master
sille-apt
Saurik's version of apt managed for tvOS/ARM64
Sam Bingner
summary
refs
log
tree
commit
diff
log msg
author
committer
range
Age
Commit message (
Collapse
)
Author
2018-09-30
Use access instead of faccessat for iOS5 compatibility
Sam Bingner
2018-09-28
Remove errors that we don't need
Sam Bingner
2018-02-08
build changes to work with our new http
Jaywalker
2018-02-07
Added kirb's TLS1.2 patch
Jaywalker
2018-02-07
Lets get those sizes right
Jaywalker
2018-02-07
Reverting to kb's gpgv version for now
Jaywalker
2018-02-07
Fixed last few errors. APT is strict...
Jaywalker
2018-02-07
Missed a semicolon
Jaywalker
2018-02-06
Fixed system() using coolstar's patch and added other required patches
Jaywalker
2018-02-06
Added seemingly missing files
Jaywalker
2018-02-06
Build system changes
Jaywalker
2017-02-15
Don't download "optional" files not in Release :/.
Jay Freeman (saurik)
2017-02-01
For ReMap to work, S has to be marked volatile :/.
Jay Freeman (saurik)
2017-02-01
Tags should use StoreString, not WriteStringInMap.
Jay Freeman (saurik)
2017-01-31
Port TagIterator to correctly support ReMap (doh).
Jay Freeman (saurik)
2017-01-31
Repeat after me: IMS-Hit is really "I am shit" :/.
Jay Freeman (saurik)
2017-01-30
It is NOT OK to just munmap memory from malloc :/.
Jay Freeman (saurik)
2017-01-30
Symlinking final file failed is a common issue :/.
Jay Freeman (saurik)
2017-01-30
This is realloc, not reallocf: be more careful :/.
Jay Freeman (saurik)
2017-01-30
Bill is consistent. Bill is correct. Be like Bill.
Jay Freeman (saurik)
2017-01-30
You can't just assume the start is always zero :/.
Jay Freeman (saurik)
2017-01-30
The length given to msync was calculated wrong :/.
Jay Freeman (saurik)
2017-01-30
Not /not/ immediately mapping a file is INSANE :/.
Jay Freeman (saurik)
2017-01-29
The entire concept of PendingError() is flawed :/.
Jay Freeman (saurik)
2017-01-27
Bug #807012 also involves package dependencies :/.
Jay Freeman (saurik)
2016-12-31
On IMS-Hit, you can't assume local compression :/.
Jay Freeman (saurik)
2016-12-31
It is *never* a good idea to throw away stderr :/.
Jay Freeman (saurik)
2016-12-30
It is just so absolutely critical that this works.
Jay Freeman (saurik)
2016-12-30
My "tolerance" patch was a tad bit overzealous :(.
Jay Freeman (saurik)
2016-12-30
You have to do the bounds check before the access.
Jay Freeman (saurik)
2016-12-30
Fix field-without-values appearing at end of file.
Jay Freeman (saurik)
2016-12-29
Of *course* I managed to get this check backwards.
Jay Freeman (saurik)
2016-12-29
In the C locale, let's avoid any crazy transforms.
Jay Freeman (saurik)
2016-12-29
It's just *mean* to complain if they already know.
Jay Freeman (saurik)
2016-12-26
Wreck validation until we can assess ecosystem :/.
Jay Freeman (saurik)
2016-12-25
This is 2016 and APT already requires use of mmap.
Jay Freeman (saurik)
2016-12-23
Only check for valid Date if checking Valid-Until.
Jay Freeman (saurik)
2016-11-30
I know this is "bad", but a "full wedge" is worse.
Jay Freeman (saurik)
2016-11-30
For speed, you *need* multiple downloads per host.
Jay Freeman (saurik)
2016-11-30
The fastest way to get field values is as a range.
Jay Freeman (saurik)
2016-11-29
Avoid wedging the entire system if recoverable :/.
Jay Freeman (saurik)
2016-11-29
Most interfaces (Maemo) need a high-level name :/.
Jay Freeman (saurik)
2016-11-29
Store tags in the cache (they are very useful :/).
Jay Freeman (saurik)
2016-11-29
unlinkat is way way too modern to not autoconf :/.
Jay Freeman (saurik)
2016-11-29
__deprecated is already defined by sys/cdefs.h :/.
Jay Freeman (saurik)
2016-11-29
arpa/nameser.h, unlike nameser.h, uses NS_ prefix.
Jay Freeman (saurik)
2016-11-29
std::map from std::initializer_list is "explicit".
Jay Freeman (saurik)
2016-11-25
Release 1.4~beta1
1.4_beta1
Julian Andres Klode
2016-11-25
Add apt-pkg/tagfile-keys.cc to the gitignore file
Julian Andres Klode
This is output of triehash. Gbp-Dch: ignore
2016-11-25
gpgv: Untrust SHA1, RIPE-MD/160, but allow downgrading to weak
Julian Andres Klode
Change the trust level check to allow downgrading an Untrusted option to weak (APT::Hashes::SHA1::Weak "yes";), so it prints a warning instead of an error; and change the default values for SHA1 and RIPE-MD/160 from Weak to Untrusted.
[next]