Revert the unintended change to apk_fileinfo_get on using
fi->xattrs.

fixes commit 1690e131 io, db: refactor struct apk_checksum away from struct apk_db_acl
fixes #11018

fixes commit fb74ef1c db, adb: always use sha256-160 package identity in installeddb

- rename "signing" options group to "generation"
- add --compression as an option there
- enable compression on applets where needed
- update the documentation

as a side effect this now compresses indexes by default,
and allows recompression with 'adbsign' applet

Allows package to specify:
 provider=foo
 depends=!foo

to provide foo as unversioned name and conflict against all
other packages providing the same name (versioned or unversioned).

fixes #10976

Instead of trying to write sha256 hashes to the installeddb,
keep the length compatible with older apk and use sha256-160.
Allow v3 package verification using sha256-160 if its installeddb.

fixes #11017

- adjust encoding of sha256 in such a manner that older apk will
  accept it (but truncate it to sha256-160 upon read)
- always use sha1/sha256-160 for triggers to keep compatiblity
- internally use sha1/sha256-160 as the hash key allow working
  in the mixed setup

This should allow a rollback to older apk if needed, though doing
so will cause truncation of the package hashes to sha256-160. The
truncated package hashes may cause some issues with "apk fix" not
handling the index based trust correctly.

Once the installeddb conversion to adb is complete, these ugly
things will go away.

ref #11017

all arrays need explisit initialization now

If old 'mkndx' is used to generate index from package without
a generated unique-id it would end up in the index without
a hash/unique-id. Inserting such package into the runtime's
database would create various problems. So drop them with
a warning.

ref #11017

Add the unique-id equivalent to allow compatibility with older
apkv3 during transitional time.

reverts part of commit 7972a780

ref #11017

Adjust logic to handle conflicts in various places.

fixes #11011

Resolve first the graph not including install_if triggered
packges, and allow conflicts to prevent install_if triggering.

fixes #10991
ref #11006

fixes #10959

Sertonix authored 7 months ago and Timo Teräs committed 6 months ago

This avoids buffer constrains when printing dependencies.

Also fixes a null pointer dereference in apk_dep_snprintf by removing it.
apk -s add "a><$(printf %254s | tr ' ' '1')"
apk -s add "$(printf %255s | tr ' ' 'a')=1"

[TT: adjusted with s/PKG_DEP_/DEP_/g]

In case the array was already bump allocated, this resets it back
to the empty array. Otherwise the truncate would incorrectly modify
the previous package's dependency array to be zero length.

fixes 037e9140 package: refactor apk_package and apk_dependency_array to use bump allocation

The apk_package * changes on addition always now.

fixes 037e9140 package: refactor apk_package and apk_dependency_array to use bump allocation

fixes c2b4a43f db: refactor apk_checksum away from struct apk_db_file

- mkpkg: no longer calculate the field
- mkndx: unconditionally calculate unique-id as a hash
- implicitly enables v3 package authentication based on index hash

fixes #11008

ref #11008

- make identity a dynamic length array in the end
- add separate package template structure with identity storage

Keep a fixed sha1 sized buffer for the digest to simplify
conversion. The files in installeddb are not expected to
contain longer hashes for now. Eventually the whole struct
will go away so no point in trying to make it dynamically
sized digest at this time.

- unroll the base64 encoding main loop
- rename the base64 chunk functions to tail functions
- fixup push_b64_tail to not read beyond end of length
  (now that it is going to be used with raw data instead of
   struct apk_checksum)
- add (un)likely to optimize for sha1/sha256 length blobs

- allows for static, bump or dynamic allocation of the apk_dependencies
- use bump allocator for packages and dependencies
- adjust package reading to clone apk_package on insert to database