pmbootstrap/pmb/helpers/repo.py

218 lines
8.3 KiB
Python
Raw Permalink Normal View History

2023-01-22 18:11:10 +00:00
# Copyright 2023 Oliver Smith
# SPDX-License-Identifier: GPL-3.0-or-later
2018-11-15 07:36:39 +00:00
"""
Functions that work with binary package repos.
See also:
- pmb/helpers/pmaports.py (work with pmaports)
- pmb/helpers/package.py (work with both)
2018-11-15 07:36:39 +00:00
"""
import os
Properly rebuild/install packages when something changed (Fix #120, #108, #131) (#129) TLDR: Always rebuild/install packages when something changed when executing "pmbootstrap install/initfs/flash", more speed in dependency resolution. --- pmbootstrap has already gotten some support for "timestamp based rebuilds", which modifies the logic for when packages should be rebuilt. It doesn't only consider packages outdated with old pkgver/pkgrel combinations, but also packages, where a source file has a newer timestamp, than the built package has. I've found out, that this can lead to more rebuilds than expected. For example, when you check out the pmbootstrap git repository again into another folder, although you have already built packages. Then all files have the timestamp of the checkout, and the packages will appear to be outdated. While this is not largely a concern now, this will become a problem once we have a binary package repository, because then the packages from the binary repo will always seem to be outdated, if you just freshly checked out the repository. To combat this, git gets asked if the files from the aport we're looking at are in sync with upstream, or not. Only when the files are not in sync with upstream and the timestamps of the sources are newer, a rebuild gets triggered from now on. In case this logic should fail, I've added an option during "pmbootstrap init" where you can enable or disable the "timestamp based rebuilds" option. In addition to that, this commit also works on fixing #120: packages do not get updated in "pmbootstrap install" after they have been rebuilt. For this to work, we specify all packages explicitly for abuild, instead of letting abuild do the resolving. This feature will also work with the "timestamp based rebuilds". This commit also fixes the working_dir argument in pmb.helpers.run.user, which was simply ignored before. Finally, the performance of the dependency resolution is faster again (when compared to the current version in master), because the parsed apkbuilds and finding the aport by pkgname gets cached during one pmbootstrap call (in args.cache, which also makes it easy to put fake data there in testcases). The new dependency resolution code can output lots of verbose messages for debugging by specifying the `-v` parameter. The meaning of that changed, it used to output the file names where log messages come from, but no one seemed to use that anyway.
2017-07-10 15:23:43 +00:00
import hashlib
import logging
import pmb.config.pmaports
import pmb.helpers.http
import pmb.helpers.run
Properly rebuild/install packages when something changed (Fix #120, #108, #131) (#129) TLDR: Always rebuild/install packages when something changed when executing "pmbootstrap install/initfs/flash", more speed in dependency resolution. --- pmbootstrap has already gotten some support for "timestamp based rebuilds", which modifies the logic for when packages should be rebuilt. It doesn't only consider packages outdated with old pkgver/pkgrel combinations, but also packages, where a source file has a newer timestamp, than the built package has. I've found out, that this can lead to more rebuilds than expected. For example, when you check out the pmbootstrap git repository again into another folder, although you have already built packages. Then all files have the timestamp of the checkout, and the packages will appear to be outdated. While this is not largely a concern now, this will become a problem once we have a binary package repository, because then the packages from the binary repo will always seem to be outdated, if you just freshly checked out the repository. To combat this, git gets asked if the files from the aport we're looking at are in sync with upstream, or not. Only when the files are not in sync with upstream and the timestamps of the sources are newer, a rebuild gets triggered from now on. In case this logic should fail, I've added an option during "pmbootstrap init" where you can enable or disable the "timestamp based rebuilds" option. In addition to that, this commit also works on fixing #120: packages do not get updated in "pmbootstrap install" after they have been rebuilt. For this to work, we specify all packages explicitly for abuild, instead of letting abuild do the resolving. This feature will also work with the "timestamp based rebuilds". This commit also fixes the working_dir argument in pmb.helpers.run.user, which was simply ignored before. Finally, the performance of the dependency resolution is faster again (when compared to the current version in master), because the parsed apkbuilds and finding the aport by pkgname gets cached during one pmbootstrap call (in args.cache, which also makes it easy to put fake data there in testcases). The new dependency resolution code can output lots of verbose messages for debugging by specifying the `-v` parameter. The meaning of that changed, it used to output the file names where log messages come from, but no one seemed to use that anyway.
2017-07-10 15:23:43 +00:00
def hash(url, length=8):
r"""Generate the hash that APK adds to the APKINDEX and apk packages in its apk cache folder.
It is the "12345678" part in this example:
Properly rebuild/install packages when something changed (Fix #120, #108, #131) (#129) TLDR: Always rebuild/install packages when something changed when executing "pmbootstrap install/initfs/flash", more speed in dependency resolution. --- pmbootstrap has already gotten some support for "timestamp based rebuilds", which modifies the logic for when packages should be rebuilt. It doesn't only consider packages outdated with old pkgver/pkgrel combinations, but also packages, where a source file has a newer timestamp, than the built package has. I've found out, that this can lead to more rebuilds than expected. For example, when you check out the pmbootstrap git repository again into another folder, although you have already built packages. Then all files have the timestamp of the checkout, and the packages will appear to be outdated. While this is not largely a concern now, this will become a problem once we have a binary package repository, because then the packages from the binary repo will always seem to be outdated, if you just freshly checked out the repository. To combat this, git gets asked if the files from the aport we're looking at are in sync with upstream, or not. Only when the files are not in sync with upstream and the timestamps of the sources are newer, a rebuild gets triggered from now on. In case this logic should fail, I've added an option during "pmbootstrap init" where you can enable or disable the "timestamp based rebuilds" option. In addition to that, this commit also works on fixing #120: packages do not get updated in "pmbootstrap install" after they have been rebuilt. For this to work, we specify all packages explicitly for abuild, instead of letting abuild do the resolving. This feature will also work with the "timestamp based rebuilds". This commit also fixes the working_dir argument in pmb.helpers.run.user, which was simply ignored before. Finally, the performance of the dependency resolution is faster again (when compared to the current version in master), because the parsed apkbuilds and finding the aport by pkgname gets cached during one pmbootstrap call (in args.cache, which also makes it easy to put fake data there in testcases). The new dependency resolution code can output lots of verbose messages for debugging by specifying the `-v` parameter. The meaning of that changed, it used to output the file names where log messages come from, but no one seemed to use that anyway.
2017-07-10 15:23:43 +00:00
"APKINDEX.12345678.tar.gz".
:param length: The length of the hash in the output file.
See also: official implementation in apk-tools:
<https://git.alpinelinux.org/cgit/apk-tools/>
blob.c: apk_blob_push_hexdump(), "const char \\*xd"
Properly rebuild/install packages when something changed (Fix #120, #108, #131) (#129) TLDR: Always rebuild/install packages when something changed when executing "pmbootstrap install/initfs/flash", more speed in dependency resolution. --- pmbootstrap has already gotten some support for "timestamp based rebuilds", which modifies the logic for when packages should be rebuilt. It doesn't only consider packages outdated with old pkgver/pkgrel combinations, but also packages, where a source file has a newer timestamp, than the built package has. I've found out, that this can lead to more rebuilds than expected. For example, when you check out the pmbootstrap git repository again into another folder, although you have already built packages. Then all files have the timestamp of the checkout, and the packages will appear to be outdated. While this is not largely a concern now, this will become a problem once we have a binary package repository, because then the packages from the binary repo will always seem to be outdated, if you just freshly checked out the repository. To combat this, git gets asked if the files from the aport we're looking at are in sync with upstream, or not. Only when the files are not in sync with upstream and the timestamps of the sources are newer, a rebuild gets triggered from now on. In case this logic should fail, I've added an option during "pmbootstrap init" where you can enable or disable the "timestamp based rebuilds" option. In addition to that, this commit also works on fixing #120: packages do not get updated in "pmbootstrap install" after they have been rebuilt. For this to work, we specify all packages explicitly for abuild, instead of letting abuild do the resolving. This feature will also work with the "timestamp based rebuilds". This commit also fixes the working_dir argument in pmb.helpers.run.user, which was simply ignored before. Finally, the performance of the dependency resolution is faster again (when compared to the current version in master), because the parsed apkbuilds and finding the aport by pkgname gets cached during one pmbootstrap call (in args.cache, which also makes it easy to put fake data there in testcases). The new dependency resolution code can output lots of verbose messages for debugging by specifying the `-v` parameter. The meaning of that changed, it used to output the file names where log messages come from, but no one seemed to use that anyway.
2017-07-10 15:23:43 +00:00
apk_defines.h: APK_CACHE_CSUM_BYTES
database.c: apk_repo_format_cache_index()
"""
binary = hashlib.sha1(url.encode("utf-8")).digest()
xd = "0123456789abcdefghijklmnopqrstuvwxyz"
csum_bytes = int(length / 2)
ret = ""
for i in range(csum_bytes):
ret += xd[(binary[i] >> 4) & 0xf]
ret += xd[binary[i] & 0xf]
return ret
def urls(args, user_repository=True, postmarketos_mirror=True, alpine=True):
"""Get a list of repository URLs, as they are in /etc/apk/repositories.
:param user_repository: add /mnt/pmbootstrap/packages
:param postmarketos_mirror: add postmarketos mirror URLs
:param alpine: add alpine mirror URLs
:returns: list of mirror strings, like ["/mnt/pmbootstrap/packages",
"http://...", ...]
Properly rebuild/install packages when something changed (Fix #120, #108, #131) (#129) TLDR: Always rebuild/install packages when something changed when executing "pmbootstrap install/initfs/flash", more speed in dependency resolution. --- pmbootstrap has already gotten some support for "timestamp based rebuilds", which modifies the logic for when packages should be rebuilt. It doesn't only consider packages outdated with old pkgver/pkgrel combinations, but also packages, where a source file has a newer timestamp, than the built package has. I've found out, that this can lead to more rebuilds than expected. For example, when you check out the pmbootstrap git repository again into another folder, although you have already built packages. Then all files have the timestamp of the checkout, and the packages will appear to be outdated. While this is not largely a concern now, this will become a problem once we have a binary package repository, because then the packages from the binary repo will always seem to be outdated, if you just freshly checked out the repository. To combat this, git gets asked if the files from the aport we're looking at are in sync with upstream, or not. Only when the files are not in sync with upstream and the timestamps of the sources are newer, a rebuild gets triggered from now on. In case this logic should fail, I've added an option during "pmbootstrap init" where you can enable or disable the "timestamp based rebuilds" option. In addition to that, this commit also works on fixing #120: packages do not get updated in "pmbootstrap install" after they have been rebuilt. For this to work, we specify all packages explicitly for abuild, instead of letting abuild do the resolving. This feature will also work with the "timestamp based rebuilds". This commit also fixes the working_dir argument in pmb.helpers.run.user, which was simply ignored before. Finally, the performance of the dependency resolution is faster again (when compared to the current version in master), because the parsed apkbuilds and finding the aport by pkgname gets cached during one pmbootstrap call (in args.cache, which also makes it easy to put fake data there in testcases). The new dependency resolution code can output lots of verbose messages for debugging by specifying the `-v` parameter. The meaning of that changed, it used to output the file names where log messages come from, but no one seemed to use that anyway.
2017-07-10 15:23:43 +00:00
"""
ret = []
# Get mirrordirs from channels.cfg (postmarketOS mirrordir is the same as
# the pmaports branch of the channel, no need to make it more complicated)
channel_cfg = pmb.config.pmaports.read_config_channel(args)
mirrordir_pmos = channel_cfg["branch_pmaports"]
mirrordir_alpine = channel_cfg["mirrordir_alpine"]
Properly rebuild/install packages when something changed (Fix #120, #108, #131) (#129) TLDR: Always rebuild/install packages when something changed when executing "pmbootstrap install/initfs/flash", more speed in dependency resolution. --- pmbootstrap has already gotten some support for "timestamp based rebuilds", which modifies the logic for when packages should be rebuilt. It doesn't only consider packages outdated with old pkgver/pkgrel combinations, but also packages, where a source file has a newer timestamp, than the built package has. I've found out, that this can lead to more rebuilds than expected. For example, when you check out the pmbootstrap git repository again into another folder, although you have already built packages. Then all files have the timestamp of the checkout, and the packages will appear to be outdated. While this is not largely a concern now, this will become a problem once we have a binary package repository, because then the packages from the binary repo will always seem to be outdated, if you just freshly checked out the repository. To combat this, git gets asked if the files from the aport we're looking at are in sync with upstream, or not. Only when the files are not in sync with upstream and the timestamps of the sources are newer, a rebuild gets triggered from now on. In case this logic should fail, I've added an option during "pmbootstrap init" where you can enable or disable the "timestamp based rebuilds" option. In addition to that, this commit also works on fixing #120: packages do not get updated in "pmbootstrap install" after they have been rebuilt. For this to work, we specify all packages explicitly for abuild, instead of letting abuild do the resolving. This feature will also work with the "timestamp based rebuilds". This commit also fixes the working_dir argument in pmb.helpers.run.user, which was simply ignored before. Finally, the performance of the dependency resolution is faster again (when compared to the current version in master), because the parsed apkbuilds and finding the aport by pkgname gets cached during one pmbootstrap call (in args.cache, which also makes it easy to put fake data there in testcases). The new dependency resolution code can output lots of verbose messages for debugging by specifying the `-v` parameter. The meaning of that changed, it used to output the file names where log messages come from, but no one seemed to use that anyway.
2017-07-10 15:23:43 +00:00
# Local user repository (for packages compiled with pmbootstrap)
if user_repository:
ret.append("/mnt/pmbootstrap/packages")
Properly rebuild/install packages when something changed (Fix #120, #108, #131) (#129) TLDR: Always rebuild/install packages when something changed when executing "pmbootstrap install/initfs/flash", more speed in dependency resolution. --- pmbootstrap has already gotten some support for "timestamp based rebuilds", which modifies the logic for when packages should be rebuilt. It doesn't only consider packages outdated with old pkgver/pkgrel combinations, but also packages, where a source file has a newer timestamp, than the built package has. I've found out, that this can lead to more rebuilds than expected. For example, when you check out the pmbootstrap git repository again into another folder, although you have already built packages. Then all files have the timestamp of the checkout, and the packages will appear to be outdated. While this is not largely a concern now, this will become a problem once we have a binary package repository, because then the packages from the binary repo will always seem to be outdated, if you just freshly checked out the repository. To combat this, git gets asked if the files from the aport we're looking at are in sync with upstream, or not. Only when the files are not in sync with upstream and the timestamps of the sources are newer, a rebuild gets triggered from now on. In case this logic should fail, I've added an option during "pmbootstrap init" where you can enable or disable the "timestamp based rebuilds" option. In addition to that, this commit also works on fixing #120: packages do not get updated in "pmbootstrap install" after they have been rebuilt. For this to work, we specify all packages explicitly for abuild, instead of letting abuild do the resolving. This feature will also work with the "timestamp based rebuilds". This commit also fixes the working_dir argument in pmb.helpers.run.user, which was simply ignored before. Finally, the performance of the dependency resolution is faster again (when compared to the current version in master), because the parsed apkbuilds and finding the aport by pkgname gets cached during one pmbootstrap call (in args.cache, which also makes it easy to put fake data there in testcases). The new dependency resolution code can output lots of verbose messages for debugging by specifying the `-v` parameter. The meaning of that changed, it used to output the file names where log messages come from, but no one seemed to use that anyway.
2017-07-10 15:23:43 +00:00
# Upstream postmarketOS binary repository
if postmarketos_mirror:
for mirror in args.mirrors_postmarketos:
# Remove "master" mirrordir to avoid breakage until bpo is adjusted
# (build.postmarketos.org#63) and to give potential other users of
# this flag a heads up.
if mirror.endswith("/master"):
logging.warning("WARNING: 'master' at the end of"
" --mirror-pmOS is deprecated, the branch gets"
" added automatically now!")
mirror = mirror[:-1 * len("master")]
ret.append(f"{mirror}{mirrordir_pmos}")
Properly rebuild/install packages when something changed (Fix #120, #108, #131) (#129) TLDR: Always rebuild/install packages when something changed when executing "pmbootstrap install/initfs/flash", more speed in dependency resolution. --- pmbootstrap has already gotten some support for "timestamp based rebuilds", which modifies the logic for when packages should be rebuilt. It doesn't only consider packages outdated with old pkgver/pkgrel combinations, but also packages, where a source file has a newer timestamp, than the built package has. I've found out, that this can lead to more rebuilds than expected. For example, when you check out the pmbootstrap git repository again into another folder, although you have already built packages. Then all files have the timestamp of the checkout, and the packages will appear to be outdated. While this is not largely a concern now, this will become a problem once we have a binary package repository, because then the packages from the binary repo will always seem to be outdated, if you just freshly checked out the repository. To combat this, git gets asked if the files from the aport we're looking at are in sync with upstream, or not. Only when the files are not in sync with upstream and the timestamps of the sources are newer, a rebuild gets triggered from now on. In case this logic should fail, I've added an option during "pmbootstrap init" where you can enable or disable the "timestamp based rebuilds" option. In addition to that, this commit also works on fixing #120: packages do not get updated in "pmbootstrap install" after they have been rebuilt. For this to work, we specify all packages explicitly for abuild, instead of letting abuild do the resolving. This feature will also work with the "timestamp based rebuilds". This commit also fixes the working_dir argument in pmb.helpers.run.user, which was simply ignored before. Finally, the performance of the dependency resolution is faster again (when compared to the current version in master), because the parsed apkbuilds and finding the aport by pkgname gets cached during one pmbootstrap call (in args.cache, which also makes it easy to put fake data there in testcases). The new dependency resolution code can output lots of verbose messages for debugging by specifying the `-v` parameter. The meaning of that changed, it used to output the file names where log messages come from, but no one seemed to use that anyway.
2017-07-10 15:23:43 +00:00
# Upstream Alpine Linux repositories
if alpine:
directories = ["main", "community"]
if mirrordir_alpine == "edge":
directories.append("testing")
for dir in directories:
ret.append(f"{args.mirror_alpine}{mirrordir_alpine}/{dir}")
Properly rebuild/install packages when something changed (Fix #120, #108, #131) (#129) TLDR: Always rebuild/install packages when something changed when executing "pmbootstrap install/initfs/flash", more speed in dependency resolution. --- pmbootstrap has already gotten some support for "timestamp based rebuilds", which modifies the logic for when packages should be rebuilt. It doesn't only consider packages outdated with old pkgver/pkgrel combinations, but also packages, where a source file has a newer timestamp, than the built package has. I've found out, that this can lead to more rebuilds than expected. For example, when you check out the pmbootstrap git repository again into another folder, although you have already built packages. Then all files have the timestamp of the checkout, and the packages will appear to be outdated. While this is not largely a concern now, this will become a problem once we have a binary package repository, because then the packages from the binary repo will always seem to be outdated, if you just freshly checked out the repository. To combat this, git gets asked if the files from the aport we're looking at are in sync with upstream, or not. Only when the files are not in sync with upstream and the timestamps of the sources are newer, a rebuild gets triggered from now on. In case this logic should fail, I've added an option during "pmbootstrap init" where you can enable or disable the "timestamp based rebuilds" option. In addition to that, this commit also works on fixing #120: packages do not get updated in "pmbootstrap install" after they have been rebuilt. For this to work, we specify all packages explicitly for abuild, instead of letting abuild do the resolving. This feature will also work with the "timestamp based rebuilds". This commit also fixes the working_dir argument in pmb.helpers.run.user, which was simply ignored before. Finally, the performance of the dependency resolution is faster again (when compared to the current version in master), because the parsed apkbuilds and finding the aport by pkgname gets cached during one pmbootstrap call (in args.cache, which also makes it easy to put fake data there in testcases). The new dependency resolution code can output lots of verbose messages for debugging by specifying the `-v` parameter. The meaning of that changed, it used to output the file names where log messages come from, but no one seemed to use that anyway.
2017-07-10 15:23:43 +00:00
return ret
def apkindex_files(args, arch=None, user_repository=True, pmos=True,
alpine=True):
"""Get a list of outside paths to all resolved APKINDEX.tar.gz files for a specific arch.
:param arch: defaults to native
:param user_repository: add path to index of locally built packages
:param pmos: add paths to indexes of postmarketos mirrors
:param alpine: add paths to indexes of alpine mirrors
:returns: list of absolute APKINDEX.tar.gz file paths
Properly rebuild/install packages when something changed (Fix #120, #108, #131) (#129) TLDR: Always rebuild/install packages when something changed when executing "pmbootstrap install/initfs/flash", more speed in dependency resolution. --- pmbootstrap has already gotten some support for "timestamp based rebuilds", which modifies the logic for when packages should be rebuilt. It doesn't only consider packages outdated with old pkgver/pkgrel combinations, but also packages, where a source file has a newer timestamp, than the built package has. I've found out, that this can lead to more rebuilds than expected. For example, when you check out the pmbootstrap git repository again into another folder, although you have already built packages. Then all files have the timestamp of the checkout, and the packages will appear to be outdated. While this is not largely a concern now, this will become a problem once we have a binary package repository, because then the packages from the binary repo will always seem to be outdated, if you just freshly checked out the repository. To combat this, git gets asked if the files from the aport we're looking at are in sync with upstream, or not. Only when the files are not in sync with upstream and the timestamps of the sources are newer, a rebuild gets triggered from now on. In case this logic should fail, I've added an option during "pmbootstrap init" where you can enable or disable the "timestamp based rebuilds" option. In addition to that, this commit also works on fixing #120: packages do not get updated in "pmbootstrap install" after they have been rebuilt. For this to work, we specify all packages explicitly for abuild, instead of letting abuild do the resolving. This feature will also work with the "timestamp based rebuilds". This commit also fixes the working_dir argument in pmb.helpers.run.user, which was simply ignored before. Finally, the performance of the dependency resolution is faster again (when compared to the current version in master), because the parsed apkbuilds and finding the aport by pkgname gets cached during one pmbootstrap call (in args.cache, which also makes it easy to put fake data there in testcases). The new dependency resolution code can output lots of verbose messages for debugging by specifying the `-v` parameter. The meaning of that changed, it used to output the file names where log messages come from, but no one seemed to use that anyway.
2017-07-10 15:23:43 +00:00
"""
if not arch:
arch = pmb.config.arch_native
Properly rebuild/install packages when something changed (Fix #120, #108, #131) (#129) TLDR: Always rebuild/install packages when something changed when executing "pmbootstrap install/initfs/flash", more speed in dependency resolution. --- pmbootstrap has already gotten some support for "timestamp based rebuilds", which modifies the logic for when packages should be rebuilt. It doesn't only consider packages outdated with old pkgver/pkgrel combinations, but also packages, where a source file has a newer timestamp, than the built package has. I've found out, that this can lead to more rebuilds than expected. For example, when you check out the pmbootstrap git repository again into another folder, although you have already built packages. Then all files have the timestamp of the checkout, and the packages will appear to be outdated. While this is not largely a concern now, this will become a problem once we have a binary package repository, because then the packages from the binary repo will always seem to be outdated, if you just freshly checked out the repository. To combat this, git gets asked if the files from the aport we're looking at are in sync with upstream, or not. Only when the files are not in sync with upstream and the timestamps of the sources are newer, a rebuild gets triggered from now on. In case this logic should fail, I've added an option during "pmbootstrap init" where you can enable or disable the "timestamp based rebuilds" option. In addition to that, this commit also works on fixing #120: packages do not get updated in "pmbootstrap install" after they have been rebuilt. For this to work, we specify all packages explicitly for abuild, instead of letting abuild do the resolving. This feature will also work with the "timestamp based rebuilds". This commit also fixes the working_dir argument in pmb.helpers.run.user, which was simply ignored before. Finally, the performance of the dependency resolution is faster again (when compared to the current version in master), because the parsed apkbuilds and finding the aport by pkgname gets cached during one pmbootstrap call (in args.cache, which also makes it easy to put fake data there in testcases). The new dependency resolution code can output lots of verbose messages for debugging by specifying the `-v` parameter. The meaning of that changed, it used to output the file names where log messages come from, but no one seemed to use that anyway.
2017-07-10 15:23:43 +00:00
ret = []
# Local user repository (for packages compiled with pmbootstrap)
if user_repository:
channel = pmb.config.pmaports.read_config(args)["channel"]
ret = [f"{args.work}/packages/{channel}/{arch}/APKINDEX.tar.gz"]
# Resolve the APKINDEX.$HASH.tar.gz files
for url in urls(args, False, pmos, alpine):
Properly rebuild/install packages when something changed (Fix #120, #108, #131) (#129) TLDR: Always rebuild/install packages when something changed when executing "pmbootstrap install/initfs/flash", more speed in dependency resolution. --- pmbootstrap has already gotten some support for "timestamp based rebuilds", which modifies the logic for when packages should be rebuilt. It doesn't only consider packages outdated with old pkgver/pkgrel combinations, but also packages, where a source file has a newer timestamp, than the built package has. I've found out, that this can lead to more rebuilds than expected. For example, when you check out the pmbootstrap git repository again into another folder, although you have already built packages. Then all files have the timestamp of the checkout, and the packages will appear to be outdated. While this is not largely a concern now, this will become a problem once we have a binary package repository, because then the packages from the binary repo will always seem to be outdated, if you just freshly checked out the repository. To combat this, git gets asked if the files from the aport we're looking at are in sync with upstream, or not. Only when the files are not in sync with upstream and the timestamps of the sources are newer, a rebuild gets triggered from now on. In case this logic should fail, I've added an option during "pmbootstrap init" where you can enable or disable the "timestamp based rebuilds" option. In addition to that, this commit also works on fixing #120: packages do not get updated in "pmbootstrap install" after they have been rebuilt. For this to work, we specify all packages explicitly for abuild, instead of letting abuild do the resolving. This feature will also work with the "timestamp based rebuilds". This commit also fixes the working_dir argument in pmb.helpers.run.user, which was simply ignored before. Finally, the performance of the dependency resolution is faster again (when compared to the current version in master), because the parsed apkbuilds and finding the aport by pkgname gets cached during one pmbootstrap call (in args.cache, which also makes it easy to put fake data there in testcases). The new dependency resolution code can output lots of verbose messages for debugging by specifying the `-v` parameter. The meaning of that changed, it used to output the file names where log messages come from, but no one seemed to use that anyway.
2017-07-10 15:23:43 +00:00
ret.append(args.work + "/cache_apk_" + arch + "/APKINDEX." +
hash(url) + ".tar.gz")
return ret
def update(args, arch=None, force=False, existing_only=False):
"""Download the APKINDEX files for all URLs depending on the architectures.
:param arch: * one Alpine architecture name ("x86_64", "armhf", ...)
* None for all architectures
:param force: even update when the APKINDEX file is fairly recent
:param existing_only: only update the APKINDEX files that already exist,
this is used by "pmbootstrap update"
:returns: True when files have been downloaded, False otherwise
"""
# Skip in offline mode, only show once
cache_key = "pmb.helpers.repo.update"
if args.offline:
if not pmb.helpers.other.cache[cache_key]["offline_msg_shown"]:
logging.info("NOTE: skipping package index update (offline mode)")
pmb.helpers.other.cache[cache_key]["offline_msg_shown"] = True
return False
# Architectures and retention time
architectures = [arch] if arch else pmb.config.build_device_architectures
retention_hours = pmb.config.apkindex_retention_time
retention_seconds = retention_hours * 3600
# Find outdated APKINDEX files. Formats:
# outdated: {URL: apkindex_path, ... }
# outdated_arches: ["armhf", "x86_64", ... ]
outdated = {}
outdated_arches = []
for url in urls(args, False):
for arch in architectures:
# APKINDEX file name from the URL
url_full = url + "/" + arch + "/APKINDEX.tar.gz"
cache_apk_outside = args.work + "/cache_apk_" + arch
apkindex = cache_apk_outside + "/APKINDEX." + hash(url) + ".tar.gz"
# Find update reason, possibly skip non-existing or known 404 files
reason = None
if url_full in pmb.helpers.other.cache[cache_key]["404"]:
# We already attempted to download this file once in this
# session
continue
elif not os.path.exists(apkindex):
if existing_only:
continue
reason = "file does not exist yet"
elif force:
reason = "forced update"
elif pmb.helpers.file.is_older_than(apkindex, retention_seconds):
reason = "older than " + str(retention_hours) + "h"
if not reason:
continue
# Update outdated and outdated_arches
logging.debug("APKINDEX outdated (" + reason + "): " + url_full)
outdated[url_full] = apkindex
if arch not in outdated_arches:
outdated_arches.append(arch)
# Bail out or show log message
if not len(outdated):
return False
logging.info("Update package index for " + ", ".join(outdated_arches) +
" (" + str(len(outdated)) + " file(s))")
# Download and move to right location
for (i, (url, target)) in enumerate(outdated.items()):
pmb.helpers.cli.progress_print(args, i / len(outdated))
temp = pmb.helpers.http.download(args, url, "APKINDEX", False,
logging.DEBUG, True)
if not temp:
pmb.helpers.other.cache[cache_key]["404"].append(url)
continue
target_folder = os.path.dirname(target)
if not os.path.exists(target_folder):
pmb.helpers.run.root(args, ["mkdir", "-p", target_folder])
pmb.helpers.run.root(args, ["cp", temp, target])
pmb.helpers.cli.progress_flush(args)
return True
aportgen: Gracefully handle old aports_upstream (#1291) In order to get cross-compilers, we generate a few aports (e.g. binutils-armhf, gcc-armhf) automatically from Alpine's aports. pmbootstrap was already able to perform a git checkout of Alpine's aports repository. But it needed to be manually updated. Otherwise the `pmbootstrap aportgen` command could actually downgrade the aport instead of updating it to the current version. After thinking about adding a dedicated pmbootstrap command for updating git repositories, I thought it would be better to not open that can of worms (pmbootstrap as general git wrapper? no thanks). The solution implemented here compares the upstream aport version of the git checkout of a certain package (e.g. gcc for gcc-armhf) with the version in Alpine's binary package APKINDEX. When the aport version is lower than the binary package version, it shows the user how to update the git repository with just one command: pmbootstrap chroot --add=git --user -- \ git -C /mnt/pmbootstrap-git/aports_upstream pull Changes: * `pmb.aportgen.core.get_upstream_aport()`: new function, that returns the absolute path to the upstream aport on disk, after checking the version of the aport against the binary package. * Use that new function in pmb.aportgen.gcc and pmb.aportgen.binutils * New function `pmb.helpers.repo.alpine_apkindex_path()`: updates the APKINDEX if necessary and returns the absolute path to the APKINDEX. This code was basically present already, but not as function, so now we have a bit less overhead there. * `pmbootstrap chroot`: new `--user` argument * `pmb.parse.apkbuild`: make pkgname check optional, as it fails with the official gcc APKBUILD before we modify it (the current APKBUILD parser is not meant to be perfect, as this would require a full shell parsing implementation). * Extended `test_aportgen.py` and enabled it by default in `testcases_fast.sh`. Previously it was disabled due to traffic concerns (cloning the aports repo, but then again we do a full KDE plasma mobile installation in Travis now, so that shouldn't matter too much). * `testcases_fast.sh`: With "test_aport_in_sync_with_git" removed from the disabled-by-default list (left over from timestamp based rebuilds), there were no more test cases disabled by default. I've changed it, so now the qemu_running_processes test case is disabled, and added an `--all` parameter to the script to disable no test cases. Travis runs with the `--all` parameter while it's useful to do a quick local test without `--all` in roughly 2 minutes instead of 10. * `aports/cross/binutils-*`: Fix `_mirror` variable to point to current default Alpine mirror (so the aportgen testcase runs through).
2018-03-11 14:18:21 +00:00
def alpine_apkindex_path(args, repo="main", arch=None):
"""Get the path to a specific Alpine APKINDEX file on disk and download it if necessary.
aportgen: Gracefully handle old aports_upstream (#1291) In order to get cross-compilers, we generate a few aports (e.g. binutils-armhf, gcc-armhf) automatically from Alpine's aports. pmbootstrap was already able to perform a git checkout of Alpine's aports repository. But it needed to be manually updated. Otherwise the `pmbootstrap aportgen` command could actually downgrade the aport instead of updating it to the current version. After thinking about adding a dedicated pmbootstrap command for updating git repositories, I thought it would be better to not open that can of worms (pmbootstrap as general git wrapper? no thanks). The solution implemented here compares the upstream aport version of the git checkout of a certain package (e.g. gcc for gcc-armhf) with the version in Alpine's binary package APKINDEX. When the aport version is lower than the binary package version, it shows the user how to update the git repository with just one command: pmbootstrap chroot --add=git --user -- \ git -C /mnt/pmbootstrap-git/aports_upstream pull Changes: * `pmb.aportgen.core.get_upstream_aport()`: new function, that returns the absolute path to the upstream aport on disk, after checking the version of the aport against the binary package. * Use that new function in pmb.aportgen.gcc and pmb.aportgen.binutils * New function `pmb.helpers.repo.alpine_apkindex_path()`: updates the APKINDEX if necessary and returns the absolute path to the APKINDEX. This code was basically present already, but not as function, so now we have a bit less overhead there. * `pmbootstrap chroot`: new `--user` argument * `pmb.parse.apkbuild`: make pkgname check optional, as it fails with the official gcc APKBUILD before we modify it (the current APKBUILD parser is not meant to be perfect, as this would require a full shell parsing implementation). * Extended `test_aportgen.py` and enabled it by default in `testcases_fast.sh`. Previously it was disabled due to traffic concerns (cloning the aports repo, but then again we do a full KDE plasma mobile installation in Travis now, so that shouldn't matter too much). * `testcases_fast.sh`: With "test_aport_in_sync_with_git" removed from the disabled-by-default list (left over from timestamp based rebuilds), there were no more test cases disabled by default. I've changed it, so now the qemu_running_processes test case is disabled, and added an `--all` parameter to the script to disable no test cases. Travis runs with the `--all` parameter while it's useful to do a quick local test without `--all` in roughly 2 minutes instead of 10. * `aports/cross/binutils-*`: Fix `_mirror` variable to point to current default Alpine mirror (so the aportgen testcase runs through).
2018-03-11 14:18:21 +00:00
:param repo: Alpine repository name (e.g. "main")
:param arch: Alpine architecture (e.g. "armhf"), defaults to native arch.
:returns: full path to the APKINDEX file
"""
# Repo sanity check
if repo not in ["main", "community", "testing", "non-free"]:
raise RuntimeError("Invalid Alpine repository: " + repo)
# Download the file
arch = arch or pmb.config.arch_native
aportgen: Gracefully handle old aports_upstream (#1291) In order to get cross-compilers, we generate a few aports (e.g. binutils-armhf, gcc-armhf) automatically from Alpine's aports. pmbootstrap was already able to perform a git checkout of Alpine's aports repository. But it needed to be manually updated. Otherwise the `pmbootstrap aportgen` command could actually downgrade the aport instead of updating it to the current version. After thinking about adding a dedicated pmbootstrap command for updating git repositories, I thought it would be better to not open that can of worms (pmbootstrap as general git wrapper? no thanks). The solution implemented here compares the upstream aport version of the git checkout of a certain package (e.g. gcc for gcc-armhf) with the version in Alpine's binary package APKINDEX. When the aport version is lower than the binary package version, it shows the user how to update the git repository with just one command: pmbootstrap chroot --add=git --user -- \ git -C /mnt/pmbootstrap-git/aports_upstream pull Changes: * `pmb.aportgen.core.get_upstream_aport()`: new function, that returns the absolute path to the upstream aport on disk, after checking the version of the aport against the binary package. * Use that new function in pmb.aportgen.gcc and pmb.aportgen.binutils * New function `pmb.helpers.repo.alpine_apkindex_path()`: updates the APKINDEX if necessary and returns the absolute path to the APKINDEX. This code was basically present already, but not as function, so now we have a bit less overhead there. * `pmbootstrap chroot`: new `--user` argument * `pmb.parse.apkbuild`: make pkgname check optional, as it fails with the official gcc APKBUILD before we modify it (the current APKBUILD parser is not meant to be perfect, as this would require a full shell parsing implementation). * Extended `test_aportgen.py` and enabled it by default in `testcases_fast.sh`. Previously it was disabled due to traffic concerns (cloning the aports repo, but then again we do a full KDE plasma mobile installation in Travis now, so that shouldn't matter too much). * `testcases_fast.sh`: With "test_aport_in_sync_with_git" removed from the disabled-by-default list (left over from timestamp based rebuilds), there were no more test cases disabled by default. I've changed it, so now the qemu_running_processes test case is disabled, and added an `--all` parameter to the script to disable no test cases. Travis runs with the `--all` parameter while it's useful to do a quick local test without `--all` in roughly 2 minutes instead of 10. * `aports/cross/binutils-*`: Fix `_mirror` variable to point to current default Alpine mirror (so the aportgen testcase runs through).
2018-03-11 14:18:21 +00:00
update(args, arch)
# Find it on disk
channel_cfg = pmb.config.pmaports.read_config_channel(args)
repo_link = f"{args.mirror_alpine}{channel_cfg['mirrordir_alpine']}/{repo}"
aportgen: Gracefully handle old aports_upstream (#1291) In order to get cross-compilers, we generate a few aports (e.g. binutils-armhf, gcc-armhf) automatically from Alpine's aports. pmbootstrap was already able to perform a git checkout of Alpine's aports repository. But it needed to be manually updated. Otherwise the `pmbootstrap aportgen` command could actually downgrade the aport instead of updating it to the current version. After thinking about adding a dedicated pmbootstrap command for updating git repositories, I thought it would be better to not open that can of worms (pmbootstrap as general git wrapper? no thanks). The solution implemented here compares the upstream aport version of the git checkout of a certain package (e.g. gcc for gcc-armhf) with the version in Alpine's binary package APKINDEX. When the aport version is lower than the binary package version, it shows the user how to update the git repository with just one command: pmbootstrap chroot --add=git --user -- \ git -C /mnt/pmbootstrap-git/aports_upstream pull Changes: * `pmb.aportgen.core.get_upstream_aport()`: new function, that returns the absolute path to the upstream aport on disk, after checking the version of the aport against the binary package. * Use that new function in pmb.aportgen.gcc and pmb.aportgen.binutils * New function `pmb.helpers.repo.alpine_apkindex_path()`: updates the APKINDEX if necessary and returns the absolute path to the APKINDEX. This code was basically present already, but not as function, so now we have a bit less overhead there. * `pmbootstrap chroot`: new `--user` argument * `pmb.parse.apkbuild`: make pkgname check optional, as it fails with the official gcc APKBUILD before we modify it (the current APKBUILD parser is not meant to be perfect, as this would require a full shell parsing implementation). * Extended `test_aportgen.py` and enabled it by default in `testcases_fast.sh`. Previously it was disabled due to traffic concerns (cloning the aports repo, but then again we do a full KDE plasma mobile installation in Travis now, so that shouldn't matter too much). * `testcases_fast.sh`: With "test_aport_in_sync_with_git" removed from the disabled-by-default list (left over from timestamp based rebuilds), there were no more test cases disabled by default. I've changed it, so now the qemu_running_processes test case is disabled, and added an `--all` parameter to the script to disable no test cases. Travis runs with the `--all` parameter while it's useful to do a quick local test without `--all` in roughly 2 minutes instead of 10. * `aports/cross/binutils-*`: Fix `_mirror` variable to point to current default Alpine mirror (so the aportgen testcase runs through).
2018-03-11 14:18:21 +00:00
cache_folder = args.work + "/cache_apk_" + arch
return cache_folder + "/APKINDEX." + hash(repo_link) + ".tar.gz"