git-repo/manifest_xml.py

776 lines
25 KiB
Python
Raw Normal View History

2008-10-21 14:00:00 +00:00
#
# Copyright (C) 2008 The Android Open Source Project
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
from __future__ import print_function
import itertools
2008-10-21 14:00:00 +00:00
import os
import re
2008-10-21 14:00:00 +00:00
import sys
import urlparse
2008-10-21 14:00:00 +00:00
import xml.dom.minidom
from git_config import GitConfig
from git_refs import R_HEADS, HEAD
from project import RemoteSpec, Project, MetaProject
2008-10-21 14:00:00 +00:00
from error import ManifestParseError
MANIFEST_FILE_NAME = 'manifest.xml'
LOCAL_MANIFEST_NAME = 'local_manifest.xml'
LOCAL_MANIFESTS_DIR_NAME = 'local_manifests'
2008-10-21 14:00:00 +00:00
urlparse.uses_relative.extend(['ssh', 'git'])
urlparse.uses_netloc.extend(['ssh', 'git'])
2008-10-21 14:00:00 +00:00
class _Default(object):
"""Project defaults within the manifest."""
revisionExpr = None
2008-10-21 14:00:00 +00:00
remote = None
sync_j = 1
sync_c = False
Represent git-submodule as nested projects, take 2 (Previous submission of this change broke Android buildbot due to incorrect regular expression for parsing git-config output. During investigation, we also found that Android, which pulls Chromium, has a workaround for Chromium's submodules; its manifest includes Chromium's submodules. This new change, in addition to fixing the regex, also take this type of workarounds into consideration; it adds a new attribute that makes repo not fetch submodules unless submodules have a project element defined in the manifest, or this attribute is overridden by a parent project element or by the default element.) We need a representation of git-submodule in repo; otherwise repo will not sync submodules, and leave workspace in a broken state. Of course this will not be a problem if all projects are owned by the owner of the manifest file, who may simply choose not to use git-submodule in all projects. However, this is not possible in practice because manifest file owner is unlikely to own all upstream projects. As git submodules are simply git repositories, it is natural to treat them as plain repo projects that live inside a repo project. That is, we could use recursively declared projects to denote the is-submodule relation of git repositories. The behavior of repo remains the same to projects that do not have a sub-project within. As for parent projects, repo fetches them and their sub-projects as normal projects, and then checks out subprojects at the commit specified in parent's commit object. The sub-project is fetched at a path relative to parent project's working directory; so the path specified in manifest file should match that of .gitmodules file. If a submodule is not registered in repo manifest, repo will derive its properties from itself and its parent project, which might not always be correct. In such cases, the subproject is called a derived subproject. To a user, a sub-project is merely a git-submodule; so all tips of working with a git-submodule apply here, too. For example, you should not run `repo sync` in a parent repository if its submodule is dirty. Change-Id: I4b8344c1b9ccad2f58ad304573133e5d52e1faef
2012-01-11 03:28:42 +00:00
sync_s = False
2008-10-21 14:00:00 +00:00
class _XmlRemote(object):
def __init__(self,
name,
Add remote alias support in manifest The `alias` is an optional attribute in element `remote`. It can be used to override attibute `name` to be set as the remote name in each project's .git/config. Its value can be duplicated while attribute `name` has to be unique across the manifest file. This helps each project to be able to have same remote name which actually points to different remote url. It eases some automation scripts to be able to checkout/push to same remote name but actually different remote url, like: repo forall -c "git checkout -b work same_remote/work" repo forall -c "git push same_remote work:work" for example: The manifest with 'alias' will look like: <?xml version='1.0' encoding='UTF-8'?> <manifest> <remote alias="same_alias" fetch="git://git.external1.org/" name="ext1" review="http://review.external1.org"/> <remote alias="same_alias" fetch="git://git.external2.org/" name="ext2" review="http://review.external2.org"/> <remote alias="same_alias" fetch="ssh://git.internal.com:29418" name="int" review="http://review.internal.com"/> <default remote="int" revision="int-branch" sync-j="2"/> <project name="path/to/project1" path="project1" remote="ext1"/> <project name="path/to/project2" path="project2" remote="ext2"/> <project name="path/to/project3" path="project3"/> ... </manifest> In each project, use command "git remote -v" project1: same_alias git://git.external1.org/project1 (fetch) same_alias git://git.external1.org/project1 (push) project2: same_alias git://git.external2.org/project2 (fetch) same_alias git://git.external2.org/project2 (push) project3: same_alias ssh://git.internal.com:29418/project3 (fetch) same_alias ssh://git.internal.com:29418/project3 (push) Change-Id: I2c48263097ff107f0c978f3e83966ae71d06cb90
2012-07-02 14:32:50 +00:00
alias=None,
fetch=None,
manifestUrl=None,
review=None):
self.name = name
self.fetchUrl = fetch
self.manifestUrl = manifestUrl
Add remote alias support in manifest The `alias` is an optional attribute in element `remote`. It can be used to override attibute `name` to be set as the remote name in each project's .git/config. Its value can be duplicated while attribute `name` has to be unique across the manifest file. This helps each project to be able to have same remote name which actually points to different remote url. It eases some automation scripts to be able to checkout/push to same remote name but actually different remote url, like: repo forall -c "git checkout -b work same_remote/work" repo forall -c "git push same_remote work:work" for example: The manifest with 'alias' will look like: <?xml version='1.0' encoding='UTF-8'?> <manifest> <remote alias="same_alias" fetch="git://git.external1.org/" name="ext1" review="http://review.external1.org"/> <remote alias="same_alias" fetch="git://git.external2.org/" name="ext2" review="http://review.external2.org"/> <remote alias="same_alias" fetch="ssh://git.internal.com:29418" name="int" review="http://review.internal.com"/> <default remote="int" revision="int-branch" sync-j="2"/> <project name="path/to/project1" path="project1" remote="ext1"/> <project name="path/to/project2" path="project2" remote="ext2"/> <project name="path/to/project3" path="project3"/> ... </manifest> In each project, use command "git remote -v" project1: same_alias git://git.external1.org/project1 (fetch) same_alias git://git.external1.org/project1 (push) project2: same_alias git://git.external2.org/project2 (fetch) same_alias git://git.external2.org/project2 (push) project3: same_alias ssh://git.internal.com:29418/project3 (fetch) same_alias ssh://git.internal.com:29418/project3 (push) Change-Id: I2c48263097ff107f0c978f3e83966ae71d06cb90
2012-07-02 14:32:50 +00:00
self.remoteAlias = alias
self.reviewUrl = review
self.resolvedFetchUrl = self._resolveFetchUrl()
def __eq__(self, other):
return self.__dict__ == other.__dict__
def __ne__(self, other):
return self.__dict__ != other.__dict__
def _resolveFetchUrl(self):
url = self.fetchUrl.rstrip('/')
manifestUrl = self.manifestUrl.rstrip('/')
p = manifestUrl.startswith('persistent-http')
if p:
manifestUrl = manifestUrl[len('persistent-'):]
# urljoin will get confused if there is no scheme in the base url
# ie, if manifestUrl is of the form <hostname:port>
if manifestUrl.find(':') != manifestUrl.find('/') - 1:
manifestUrl = 'gopher://' + manifestUrl
url = urlparse.urljoin(manifestUrl, url)
url = re.sub(r'^gopher://', '', url)
if p:
url = 'persistent-' + url
return url
def ToRemoteSpec(self, projectName):
url = self.resolvedFetchUrl.rstrip('/') + '/' + projectName
Add remote alias support in manifest The `alias` is an optional attribute in element `remote`. It can be used to override attibute `name` to be set as the remote name in each project's .git/config. Its value can be duplicated while attribute `name` has to be unique across the manifest file. This helps each project to be able to have same remote name which actually points to different remote url. It eases some automation scripts to be able to checkout/push to same remote name but actually different remote url, like: repo forall -c "git checkout -b work same_remote/work" repo forall -c "git push same_remote work:work" for example: The manifest with 'alias' will look like: <?xml version='1.0' encoding='UTF-8'?> <manifest> <remote alias="same_alias" fetch="git://git.external1.org/" name="ext1" review="http://review.external1.org"/> <remote alias="same_alias" fetch="git://git.external2.org/" name="ext2" review="http://review.external2.org"/> <remote alias="same_alias" fetch="ssh://git.internal.com:29418" name="int" review="http://review.internal.com"/> <default remote="int" revision="int-branch" sync-j="2"/> <project name="path/to/project1" path="project1" remote="ext1"/> <project name="path/to/project2" path="project2" remote="ext2"/> <project name="path/to/project3" path="project3"/> ... </manifest> In each project, use command "git remote -v" project1: same_alias git://git.external1.org/project1 (fetch) same_alias git://git.external1.org/project1 (push) project2: same_alias git://git.external2.org/project2 (fetch) same_alias git://git.external2.org/project2 (push) project3: same_alias ssh://git.internal.com:29418/project3 (fetch) same_alias ssh://git.internal.com:29418/project3 (push) Change-Id: I2c48263097ff107f0c978f3e83966ae71d06cb90
2012-07-02 14:32:50 +00:00
remoteName = self.name
if self.remoteAlias:
remoteName = self.remoteAlias
return RemoteSpec(remoteName, url, self.reviewUrl)
2008-10-21 14:00:00 +00:00
class XmlManifest(object):
2008-10-21 14:00:00 +00:00
"""manages the repo configuration file"""
def __init__(self, repodir):
self.repodir = os.path.abspath(repodir)
self.topdir = os.path.dirname(self.repodir)
self.manifestFile = os.path.join(self.repodir, MANIFEST_FILE_NAME)
self.globalConfig = GitConfig.ForUser()
self.repoProject = MetaProject(self, 'repo',
gitdir = os.path.join(repodir, 'repo/.git'),
worktree = os.path.join(repodir, 'repo'))
self.manifestProject = MetaProject(self, 'manifests',
gitdir = os.path.join(repodir, 'manifests.git'),
worktree = os.path.join(repodir, 'manifests'))
2008-10-21 14:00:00 +00:00
self._Unload()
def Override(self, name):
"""Use a different manifest, just for the current instantiation.
2008-10-21 14:00:00 +00:00
"""
path = os.path.join(self.manifestProject.worktree, name)
if not os.path.isfile(path):
raise ManifestParseError('manifest %s not found' % name)
old = self.manifestFile
try:
self.manifestFile = path
self._Unload()
self._Load()
finally:
self.manifestFile = old
def Link(self, name):
"""Update the repo metadata to use a different manifest.
"""
self.Override(name)
2008-10-21 14:00:00 +00:00
try:
if os.path.lexists(self.manifestFile):
2008-10-21 14:00:00 +00:00
os.remove(self.manifestFile)
os.symlink('manifests/%s' % name, self.manifestFile)
except OSError as e:
raise ManifestParseError('cannot link manifest %s: %s' % (name, str(e)))
2008-10-21 14:00:00 +00:00
def _RemoteToXml(self, r, doc, root):
e = doc.createElement('remote')
root.appendChild(e)
e.setAttribute('name', r.name)
e.setAttribute('fetch', r.fetchUrl)
if r.reviewUrl is not None:
e.setAttribute('review', r.reviewUrl)
def Save(self, fd, peg_rev=False, peg_rev_upstream=True):
"""Write the current manifest out to the given file descriptor.
"""
mp = self.manifestProject
groups = mp.config.GetString('manifest.groups')
if groups:
groups = [x for x in re.split(r'[,\s]+', groups) if x]
doc = xml.dom.minidom.Document()
root = doc.createElement('manifest')
doc.appendChild(root)
# Save out the notice. There's a little bit of work here to give it the
# right whitespace, which assumes that the notice is automatically indented
# by 4 by minidom.
if self.notice:
notice_element = root.appendChild(doc.createElement('notice'))
notice_lines = self.notice.splitlines()
indented_notice = ('\n'.join(" "*4 + line for line in notice_lines))[4:]
notice_element.appendChild(doc.createTextNode(indented_notice))
d = self.default
sort_remotes = list(self.remotes.keys())
sort_remotes.sort()
for r in sort_remotes:
self._RemoteToXml(self.remotes[r], doc, root)
if self.remotes:
root.appendChild(doc.createTextNode(''))
have_default = False
e = doc.createElement('default')
if d.remote:
have_default = True
e.setAttribute('remote', d.remote.name)
if d.revisionExpr:
have_default = True
e.setAttribute('revision', d.revisionExpr)
if d.sync_j > 1:
have_default = True
e.setAttribute('sync-j', '%d' % d.sync_j)
if d.sync_c:
have_default = True
e.setAttribute('sync-c', 'true')
Represent git-submodule as nested projects, take 2 (Previous submission of this change broke Android buildbot due to incorrect regular expression for parsing git-config output. During investigation, we also found that Android, which pulls Chromium, has a workaround for Chromium's submodules; its manifest includes Chromium's submodules. This new change, in addition to fixing the regex, also take this type of workarounds into consideration; it adds a new attribute that makes repo not fetch submodules unless submodules have a project element defined in the manifest, or this attribute is overridden by a parent project element or by the default element.) We need a representation of git-submodule in repo; otherwise repo will not sync submodules, and leave workspace in a broken state. Of course this will not be a problem if all projects are owned by the owner of the manifest file, who may simply choose not to use git-submodule in all projects. However, this is not possible in practice because manifest file owner is unlikely to own all upstream projects. As git submodules are simply git repositories, it is natural to treat them as plain repo projects that live inside a repo project. That is, we could use recursively declared projects to denote the is-submodule relation of git repositories. The behavior of repo remains the same to projects that do not have a sub-project within. As for parent projects, repo fetches them and their sub-projects as normal projects, and then checks out subprojects at the commit specified in parent's commit object. The sub-project is fetched at a path relative to parent project's working directory; so the path specified in manifest file should match that of .gitmodules file. If a submodule is not registered in repo manifest, repo will derive its properties from itself and its parent project, which might not always be correct. In such cases, the subproject is called a derived subproject. To a user, a sub-project is merely a git-submodule; so all tips of working with a git-submodule apply here, too. For example, you should not run `repo sync` in a parent repository if its submodule is dirty. Change-Id: I4b8344c1b9ccad2f58ad304573133e5d52e1faef
2012-01-11 03:28:42 +00:00
if d.sync_s:
have_default = True
e.setAttribute('sync-s', 'true')
if have_default:
root.appendChild(e)
root.appendChild(doc.createTextNode(''))
if self._manifest_server:
e = doc.createElement('manifest-server')
e.setAttribute('url', self._manifest_server)
root.appendChild(e)
root.appendChild(doc.createTextNode(''))
Represent git-submodule as nested projects, take 2 (Previous submission of this change broke Android buildbot due to incorrect regular expression for parsing git-config output. During investigation, we also found that Android, which pulls Chromium, has a workaround for Chromium's submodules; its manifest includes Chromium's submodules. This new change, in addition to fixing the regex, also take this type of workarounds into consideration; it adds a new attribute that makes repo not fetch submodules unless submodules have a project element defined in the manifest, or this attribute is overridden by a parent project element or by the default element.) We need a representation of git-submodule in repo; otherwise repo will not sync submodules, and leave workspace in a broken state. Of course this will not be a problem if all projects are owned by the owner of the manifest file, who may simply choose not to use git-submodule in all projects. However, this is not possible in practice because manifest file owner is unlikely to own all upstream projects. As git submodules are simply git repositories, it is natural to treat them as plain repo projects that live inside a repo project. That is, we could use recursively declared projects to denote the is-submodule relation of git repositories. The behavior of repo remains the same to projects that do not have a sub-project within. As for parent projects, repo fetches them and their sub-projects as normal projects, and then checks out subprojects at the commit specified in parent's commit object. The sub-project is fetched at a path relative to parent project's working directory; so the path specified in manifest file should match that of .gitmodules file. If a submodule is not registered in repo manifest, repo will derive its properties from itself and its parent project, which might not always be correct. In such cases, the subproject is called a derived subproject. To a user, a sub-project is merely a git-submodule; so all tips of working with a git-submodule apply here, too. For example, you should not run `repo sync` in a parent repository if its submodule is dirty. Change-Id: I4b8344c1b9ccad2f58ad304573133e5d52e1faef
2012-01-11 03:28:42 +00:00
def output_projects(parent, parent_node, projects):
for p in projects:
output_project(parent, parent_node, self.projects[p])
Represent git-submodule as nested projects We need a representation of git-submodule in repo; otherwise repo will not sync submodules, and leave workspace in a broken state. Of course this will not be a problem if all projects are owned by the owner of the manifest file, who may simply choose not to use git-submodule in all projects. However, this is not possible in practice because manifest file owner is unlikely to own all upstream projects. As git submodules are simply git repositories, it is natural to treat them as plain repo projects that live inside a repo project. That is, we could use recursively declared projects to denote the is-submodule relation of git repositories. The behavior of repo remains the same to projects that do not have a sub-project within. As for parent projects, repo fetches them and their sub-projects as normal projects, and then checks out subprojects at the commit specified in parent's commit object. The sub-project is fetched at a path relative to parent project's working directory; so the path specified in manifest file should match that of .gitmodules file. If a submodule is not registered in repo manifest, repo will derive its properties from itself and its parent project, which might not always be correct. In such cases, the subproject is called a derived subproject. To a user, a sub-project is merely a git-submodule; so all tips of working with a git-submodule apply here, too. For example, you should not run `repo sync` in a parent repository if its submodule is dirty. Change-Id: I541e9e2ac1a70304272dbe09724572aa1004eb5c
2012-01-11 03:28:42 +00:00
Represent git-submodule as nested projects, take 2 (Previous submission of this change broke Android buildbot due to incorrect regular expression for parsing git-config output. During investigation, we also found that Android, which pulls Chromium, has a workaround for Chromium's submodules; its manifest includes Chromium's submodules. This new change, in addition to fixing the regex, also take this type of workarounds into consideration; it adds a new attribute that makes repo not fetch submodules unless submodules have a project element defined in the manifest, or this attribute is overridden by a parent project element or by the default element.) We need a representation of git-submodule in repo; otherwise repo will not sync submodules, and leave workspace in a broken state. Of course this will not be a problem if all projects are owned by the owner of the manifest file, who may simply choose not to use git-submodule in all projects. However, this is not possible in practice because manifest file owner is unlikely to own all upstream projects. As git submodules are simply git repositories, it is natural to treat them as plain repo projects that live inside a repo project. That is, we could use recursively declared projects to denote the is-submodule relation of git repositories. The behavior of repo remains the same to projects that do not have a sub-project within. As for parent projects, repo fetches them and their sub-projects as normal projects, and then checks out subprojects at the commit specified in parent's commit object. The sub-project is fetched at a path relative to parent project's working directory; so the path specified in manifest file should match that of .gitmodules file. If a submodule is not registered in repo manifest, repo will derive its properties from itself and its parent project, which might not always be correct. In such cases, the subproject is called a derived subproject. To a user, a sub-project is merely a git-submodule; so all tips of working with a git-submodule apply here, too. For example, you should not run `repo sync` in a parent repository if its submodule is dirty. Change-Id: I4b8344c1b9ccad2f58ad304573133e5d52e1faef
2012-01-11 03:28:42 +00:00
def output_project(parent, parent_node, p):
if not p.MatchesGroups(groups):
Represent git-submodule as nested projects, take 2 (Previous submission of this change broke Android buildbot due to incorrect regular expression for parsing git-config output. During investigation, we also found that Android, which pulls Chromium, has a workaround for Chromium's submodules; its manifest includes Chromium's submodules. This new change, in addition to fixing the regex, also take this type of workarounds into consideration; it adds a new attribute that makes repo not fetch submodules unless submodules have a project element defined in the manifest, or this attribute is overridden by a parent project element or by the default element.) We need a representation of git-submodule in repo; otherwise repo will not sync submodules, and leave workspace in a broken state. Of course this will not be a problem if all projects are owned by the owner of the manifest file, who may simply choose not to use git-submodule in all projects. However, this is not possible in practice because manifest file owner is unlikely to own all upstream projects. As git submodules are simply git repositories, it is natural to treat them as plain repo projects that live inside a repo project. That is, we could use recursively declared projects to denote the is-submodule relation of git repositories. The behavior of repo remains the same to projects that do not have a sub-project within. As for parent projects, repo fetches them and their sub-projects as normal projects, and then checks out subprojects at the commit specified in parent's commit object. The sub-project is fetched at a path relative to parent project's working directory; so the path specified in manifest file should match that of .gitmodules file. If a submodule is not registered in repo manifest, repo will derive its properties from itself and its parent project, which might not always be correct. In such cases, the subproject is called a derived subproject. To a user, a sub-project is merely a git-submodule; so all tips of working with a git-submodule apply here, too. For example, you should not run `repo sync` in a parent repository if its submodule is dirty. Change-Id: I4b8344c1b9ccad2f58ad304573133e5d52e1faef
2012-01-11 03:28:42 +00:00
return
name = p.name
relpath = p.relpath
if parent:
name = self._UnjoinName(parent.name, name)
relpath = self._UnjoinRelpath(parent.relpath, relpath)
e = doc.createElement('project')
Represent git-submodule as nested projects, take 2 (Previous submission of this change broke Android buildbot due to incorrect regular expression for parsing git-config output. During investigation, we also found that Android, which pulls Chromium, has a workaround for Chromium's submodules; its manifest includes Chromium's submodules. This new change, in addition to fixing the regex, also take this type of workarounds into consideration; it adds a new attribute that makes repo not fetch submodules unless submodules have a project element defined in the manifest, or this attribute is overridden by a parent project element or by the default element.) We need a representation of git-submodule in repo; otherwise repo will not sync submodules, and leave workspace in a broken state. Of course this will not be a problem if all projects are owned by the owner of the manifest file, who may simply choose not to use git-submodule in all projects. However, this is not possible in practice because manifest file owner is unlikely to own all upstream projects. As git submodules are simply git repositories, it is natural to treat them as plain repo projects that live inside a repo project. That is, we could use recursively declared projects to denote the is-submodule relation of git repositories. The behavior of repo remains the same to projects that do not have a sub-project within. As for parent projects, repo fetches them and their sub-projects as normal projects, and then checks out subprojects at the commit specified in parent's commit object. The sub-project is fetched at a path relative to parent project's working directory; so the path specified in manifest file should match that of .gitmodules file. If a submodule is not registered in repo manifest, repo will derive its properties from itself and its parent project, which might not always be correct. In such cases, the subproject is called a derived subproject. To a user, a sub-project is merely a git-submodule; so all tips of working with a git-submodule apply here, too. For example, you should not run `repo sync` in a parent repository if its submodule is dirty. Change-Id: I4b8344c1b9ccad2f58ad304573133e5d52e1faef
2012-01-11 03:28:42 +00:00
parent_node.appendChild(e)
e.setAttribute('name', name)
if relpath != name:
e.setAttribute('path', relpath)
if not d.remote or p.remote.name != d.remote.name:
e.setAttribute('remote', p.remote.name)
if peg_rev:
if self.IsMirror:
value = p.bare_git.rev_parse(p.revisionExpr + '^0')
else:
value = p.work_git.rev_parse(HEAD + '^0')
e.setAttribute('revision', value)
if peg_rev_upstream and value != p.revisionExpr:
# Only save the origin if the origin is not a sha1, and the default
# isn't our value, and the if the default doesn't already have that
# covered.
e.setAttribute('upstream', p.revisionExpr)
elif not d.revisionExpr or p.revisionExpr != d.revisionExpr:
e.setAttribute('revision', p.revisionExpr)
for c in p.copyfiles:
ce = doc.createElement('copyfile')
ce.setAttribute('src', c.src)
ce.setAttribute('dest', c.dest)
e.appendChild(ce)
default_groups = ['all', 'name:%s' % p.name, 'path:%s' % p.relpath]
egroups = [g for g in p.groups if g not in default_groups]
if egroups:
e.setAttribute('groups', ','.join(egroups))
for a in p.annotations:
if a.keep == "true":
ae = doc.createElement('annotation')
ae.setAttribute('name', a.name)
ae.setAttribute('value', a.value)
e.appendChild(ae)
if p.sync_c:
e.setAttribute('sync-c', 'true')
Represent git-submodule as nested projects, take 2 (Previous submission of this change broke Android buildbot due to incorrect regular expression for parsing git-config output. During investigation, we also found that Android, which pulls Chromium, has a workaround for Chromium's submodules; its manifest includes Chromium's submodules. This new change, in addition to fixing the regex, also take this type of workarounds into consideration; it adds a new attribute that makes repo not fetch submodules unless submodules have a project element defined in the manifest, or this attribute is overridden by a parent project element or by the default element.) We need a representation of git-submodule in repo; otherwise repo will not sync submodules, and leave workspace in a broken state. Of course this will not be a problem if all projects are owned by the owner of the manifest file, who may simply choose not to use git-submodule in all projects. However, this is not possible in practice because manifest file owner is unlikely to own all upstream projects. As git submodules are simply git repositories, it is natural to treat them as plain repo projects that live inside a repo project. That is, we could use recursively declared projects to denote the is-submodule relation of git repositories. The behavior of repo remains the same to projects that do not have a sub-project within. As for parent projects, repo fetches them and their sub-projects as normal projects, and then checks out subprojects at the commit specified in parent's commit object. The sub-project is fetched at a path relative to parent project's working directory; so the path specified in manifest file should match that of .gitmodules file. If a submodule is not registered in repo manifest, repo will derive its properties from itself and its parent project, which might not always be correct. In such cases, the subproject is called a derived subproject. To a user, a sub-project is merely a git-submodule; so all tips of working with a git-submodule apply here, too. For example, you should not run `repo sync` in a parent repository if its submodule is dirty. Change-Id: I4b8344c1b9ccad2f58ad304573133e5d52e1faef
2012-01-11 03:28:42 +00:00
if p.sync_s:
e.setAttribute('sync-s', 'true')
if p.subprojects:
sort_projects = [subp.name for subp in p.subprojects]
sort_projects.sort()
output_projects(p, e, sort_projects)
sort_projects = [key for key in self.projects.keys()
if not self.projects[key].parent]
sort_projects.sort()
output_projects(None, root, sort_projects)
Support repo-level pre-upload hook and prep for future hooks. All repo-level hooks are expected to live in a single project at the top level of that project. The name of the hooks project is provided in the manifest.xml. The manifest also lists which hooks are enabled to make it obvious if a file somehow failed to sync down (or got deleted). Before running any hook, we will prompt the user to make sure that it is OK. A user can deny running the hook, allow once, or allow "forever" (until hooks change). This tries to keep with the git spirit of not automatically running anything on the user's computer that got synced down. Note that individual repo commands can add always options to avoid these prompts as they see fit (see below for the 'upload' options). When hooks are run, they are loaded into the current interpreter (the one running repo) and their main() function is run. This mechanism is used (instead of using subprocess) to make it easier to expand to a richer hook interface in the future. During loading, the interpreter's sys.path is updated to contain the directory containing the hooks so that hooks can be split into multiple files. The upload command has two options that control hook behavior: - no-verify=False, verify=False (DEFAULT): If stdout is a tty, can prompt about running upload hooks if needed. If user denies running hooks, the upload is cancelled. If stdout is not a tty and we would need to prompt about upload hooks, upload is cancelled. - no-verify=False, verify=True: Always run upload hooks with no prompt. - no-verify=True, verify=False: Never run upload hooks, but upload anyway (AKA bypass hooks). - no-verify=True, verify=True: Invalid Sample bit of manifest.xml code for enabling hooks (assumes you have a project named 'hooks' where hooks are stored): <repo-hooks in-project="hooks" enabled-list="pre-upload" /> Sample main() function in pre-upload.py in hooks directory: def main(project_list, **kwargs): print ('These projects will be uploaded: %s' % ', '.join(project_list)) print ('I am being a good boy and ignoring anything in kwargs\n' 'that I don\'t understand.') print 'I fail 50% of the time. How flaky.' if random.random() <= .5: raise Exception('Pre-upload hook failed. Have a nice day.') Change-Id: I5cefa2cd5865c72589263cf8e2f152a43c122f70
2011-03-04 19:54:18 +00:00
if self._repo_hooks_project:
root.appendChild(doc.createTextNode(''))
e = doc.createElement('repo-hooks')
e.setAttribute('in-project', self._repo_hooks_project.name)
e.setAttribute('enabled-list',
' '.join(self._repo_hooks_project.enabled_repo_hooks))
root.appendChild(e)
doc.writexml(fd, '', ' ', '\n', 'UTF-8')
2008-10-21 14:00:00 +00:00
@property
def projects(self):
self._Load()
return self._projects
@property
def remotes(self):
self._Load()
return self._remotes
@property
def default(self):
self._Load()
return self._default
Support repo-level pre-upload hook and prep for future hooks. All repo-level hooks are expected to live in a single project at the top level of that project. The name of the hooks project is provided in the manifest.xml. The manifest also lists which hooks are enabled to make it obvious if a file somehow failed to sync down (or got deleted). Before running any hook, we will prompt the user to make sure that it is OK. A user can deny running the hook, allow once, or allow "forever" (until hooks change). This tries to keep with the git spirit of not automatically running anything on the user's computer that got synced down. Note that individual repo commands can add always options to avoid these prompts as they see fit (see below for the 'upload' options). When hooks are run, they are loaded into the current interpreter (the one running repo) and their main() function is run. This mechanism is used (instead of using subprocess) to make it easier to expand to a richer hook interface in the future. During loading, the interpreter's sys.path is updated to contain the directory containing the hooks so that hooks can be split into multiple files. The upload command has two options that control hook behavior: - no-verify=False, verify=False (DEFAULT): If stdout is a tty, can prompt about running upload hooks if needed. If user denies running hooks, the upload is cancelled. If stdout is not a tty and we would need to prompt about upload hooks, upload is cancelled. - no-verify=False, verify=True: Always run upload hooks with no prompt. - no-verify=True, verify=False: Never run upload hooks, but upload anyway (AKA bypass hooks). - no-verify=True, verify=True: Invalid Sample bit of manifest.xml code for enabling hooks (assumes you have a project named 'hooks' where hooks are stored): <repo-hooks in-project="hooks" enabled-list="pre-upload" /> Sample main() function in pre-upload.py in hooks directory: def main(project_list, **kwargs): print ('These projects will be uploaded: %s' % ', '.join(project_list)) print ('I am being a good boy and ignoring anything in kwargs\n' 'that I don\'t understand.') print 'I fail 50% of the time. How flaky.' if random.random() <= .5: raise Exception('Pre-upload hook failed. Have a nice day.') Change-Id: I5cefa2cd5865c72589263cf8e2f152a43c122f70
2011-03-04 19:54:18 +00:00
@property
def repo_hooks_project(self):
self._Load()
return self._repo_hooks_project
@property
def notice(self):
self._Load()
return self._notice
@property
def manifest_server(self):
self._Load()
return self._manifest_server
@property
def IsMirror(self):
return self.manifestProject.config.GetBoolean('repo.mirror')
2008-10-21 14:00:00 +00:00
def _Unload(self):
self._loaded = False
self._projects = {}
self._remotes = {}
self._default = None
Support repo-level pre-upload hook and prep for future hooks. All repo-level hooks are expected to live in a single project at the top level of that project. The name of the hooks project is provided in the manifest.xml. The manifest also lists which hooks are enabled to make it obvious if a file somehow failed to sync down (or got deleted). Before running any hook, we will prompt the user to make sure that it is OK. A user can deny running the hook, allow once, or allow "forever" (until hooks change). This tries to keep with the git spirit of not automatically running anything on the user's computer that got synced down. Note that individual repo commands can add always options to avoid these prompts as they see fit (see below for the 'upload' options). When hooks are run, they are loaded into the current interpreter (the one running repo) and their main() function is run. This mechanism is used (instead of using subprocess) to make it easier to expand to a richer hook interface in the future. During loading, the interpreter's sys.path is updated to contain the directory containing the hooks so that hooks can be split into multiple files. The upload command has two options that control hook behavior: - no-verify=False, verify=False (DEFAULT): If stdout is a tty, can prompt about running upload hooks if needed. If user denies running hooks, the upload is cancelled. If stdout is not a tty and we would need to prompt about upload hooks, upload is cancelled. - no-verify=False, verify=True: Always run upload hooks with no prompt. - no-verify=True, verify=False: Never run upload hooks, but upload anyway (AKA bypass hooks). - no-verify=True, verify=True: Invalid Sample bit of manifest.xml code for enabling hooks (assumes you have a project named 'hooks' where hooks are stored): <repo-hooks in-project="hooks" enabled-list="pre-upload" /> Sample main() function in pre-upload.py in hooks directory: def main(project_list, **kwargs): print ('These projects will be uploaded: %s' % ', '.join(project_list)) print ('I am being a good boy and ignoring anything in kwargs\n' 'that I don\'t understand.') print 'I fail 50% of the time. How flaky.' if random.random() <= .5: raise Exception('Pre-upload hook failed. Have a nice day.') Change-Id: I5cefa2cd5865c72589263cf8e2f152a43c122f70
2011-03-04 19:54:18 +00:00
self._repo_hooks_project = None
self._notice = None
2008-10-21 14:00:00 +00:00
self.branch = None
self._manifest_server = None
2008-10-21 14:00:00 +00:00
def _Load(self):
if not self._loaded:
m = self.manifestProject
b = m.GetBranch(m.CurrentBranch).merge
if b is not None and b.startswith(R_HEADS):
b = b[len(R_HEADS):]
self.branch = b
nodes = []
nodes.append(self._ParseManifestXml(self.manifestFile,
self.manifestProject.worktree))
local = os.path.join(self.repodir, LOCAL_MANIFEST_NAME)
if os.path.exists(local):
print('warning: %s is deprecated; put local manifests in `%s` instead'
% (LOCAL_MANIFEST_NAME, os.path.join(self.repodir, LOCAL_MANIFESTS_DIR_NAME)),
file=sys.stderr)
nodes.append(self._ParseManifestXml(local, self.repodir))
local_dir = os.path.abspath(os.path.join(self.repodir, LOCAL_MANIFESTS_DIR_NAME))
try:
for local_file in sorted(os.listdir(local_dir)):
if local_file.endswith('.xml'):
local = os.path.join(local_dir, local_file)
nodes.append(self._ParseManifestXml(local, self.repodir))
except OSError:
pass
try:
self._ParseManifest(nodes)
except ManifestParseError as e:
# There was a problem parsing, unload ourselves in case they catch
# this error and try again later, we will show the correct error
self._Unload()
raise e
if self.IsMirror:
self._AddMetaProjectMirror(self.repoProject)
self._AddMetaProjectMirror(self.manifestProject)
2008-10-21 14:00:00 +00:00
self._loaded = True
def _ParseManifestXml(self, path, include_root):
try:
root = xml.dom.minidom.parse(path)
except (OSError, xml.parsers.expat.ExpatError) as e:
raise ManifestParseError("error parsing manifest %s: %s" % (path, e))
2008-10-21 14:00:00 +00:00
if not root or not root.childNodes:
ManifestXml: add include support Having the ability to include other manifests is a very practical feature to ease the managment of manifest. It allows to divide a manifest into separate files, and create different environment depending on what we want to release You can have unlimited recursion of include, the manifest configs will simply be concatenated as if it was in a single file. command "repo manifest" will create a single manifest, and not recreate the manifest hierarchy for example: Our developement manifest will look like: <?xml version='1.0' encoding='UTF-8'?> <manifest> <default revision="platform/android/main" remote="intel"/> <include name="server.xml"/> <!-- The Server configuration --> <include name="aosp.xml" /> <!-- All the AOSP projects --> <include name="bsp.xml" /> <!-- The BSP projects that we release in source form --> <include name="bsp-priv.xml" /> <!-- The source of the BSP projects we release in binary form --> </manifest> Our release manifest will look like: <?xml version='1.0' encoding='UTF-8'?> <manifest> <default revision="platform/android/release-ext" remote="intel"/> <include name="server.xml"/> <!-- The Server configuration --> <include name="aosp.xml" /> <!-- All the AOSP projects --> <include name="bsp.xml" /> <!-- The BSP projects that we release in source form --> <include name="bsp-ext.xml" /> <!-- The PREBUILT version of the BSP projects we release in binary form --> </manifest> And it is also easy to create and maintain feature branch with a manifest that looks like: <?xml version='1.0' encoding='UTF-8'?> <manifest> <default revision="feature_branch_foobar" remote="intel"/> <include name="server.xml"/> <!-- The Server configuration --> <include name="aosp.xml" /> <!-- All the AOSP projects --> <include name="bsp.xml" /> <!-- The BSP projects that we release in source form --> <include name="bsp-priv.xml" /> <!-- The source of the BSP projects we release in binary form --> </manifest> Signed-off-by: Brian Harring <brian.harring@intel.com> Signed-off-by: Pierre Tardy <pierre.tardy@intel.com> Change-Id: I833a30d303039e485888768e6b81561b7665e89d
2011-04-28 12:04:41 +00:00
raise ManifestParseError("no root node in %s" % (path,))
2008-10-21 14:00:00 +00:00
for manifest in root.childNodes:
if manifest.nodeName == 'manifest':
break
else:
ManifestXml: add include support Having the ability to include other manifests is a very practical feature to ease the managment of manifest. It allows to divide a manifest into separate files, and create different environment depending on what we want to release You can have unlimited recursion of include, the manifest configs will simply be concatenated as if it was in a single file. command "repo manifest" will create a single manifest, and not recreate the manifest hierarchy for example: Our developement manifest will look like: <?xml version='1.0' encoding='UTF-8'?> <manifest> <default revision="platform/android/main" remote="intel"/> <include name="server.xml"/> <!-- The Server configuration --> <include name="aosp.xml" /> <!-- All the AOSP projects --> <include name="bsp.xml" /> <!-- The BSP projects that we release in source form --> <include name="bsp-priv.xml" /> <!-- The source of the BSP projects we release in binary form --> </manifest> Our release manifest will look like: <?xml version='1.0' encoding='UTF-8'?> <manifest> <default revision="platform/android/release-ext" remote="intel"/> <include name="server.xml"/> <!-- The Server configuration --> <include name="aosp.xml" /> <!-- All the AOSP projects --> <include name="bsp.xml" /> <!-- The BSP projects that we release in source form --> <include name="bsp-ext.xml" /> <!-- The PREBUILT version of the BSP projects we release in binary form --> </manifest> And it is also easy to create and maintain feature branch with a manifest that looks like: <?xml version='1.0' encoding='UTF-8'?> <manifest> <default revision="feature_branch_foobar" remote="intel"/> <include name="server.xml"/> <!-- The Server configuration --> <include name="aosp.xml" /> <!-- All the AOSP projects --> <include name="bsp.xml" /> <!-- The BSP projects that we release in source form --> <include name="bsp-priv.xml" /> <!-- The source of the BSP projects we release in binary form --> </manifest> Signed-off-by: Brian Harring <brian.harring@intel.com> Signed-off-by: Pierre Tardy <pierre.tardy@intel.com> Change-Id: I833a30d303039e485888768e6b81561b7665e89d
2011-04-28 12:04:41 +00:00
raise ManifestParseError("no <manifest> in %s" % (path,))
nodes = []
for node in manifest.childNodes: # pylint:disable=W0631
# We only get here if manifest is initialised
if node.nodeName == 'include':
name = self._reqatt(node, 'name')
fp = os.path.join(include_root, name)
if not os.path.isfile(fp):
raise ManifestParseError, \
"include %s doesn't exist or isn't a file" % \
(name,)
try:
nodes.extend(self._ParseManifestXml(fp, include_root))
# should isolate this to the exact exception, but that's
# tricky. actual parsing implementation may vary.
except (KeyboardInterrupt, RuntimeError, SystemExit):
raise
except Exception as e:
raise ManifestParseError(
"failed parsing included manifest %s: %s", (name, e))
else:
nodes.append(node)
return nodes
ManifestXml: add include support Having the ability to include other manifests is a very practical feature to ease the managment of manifest. It allows to divide a manifest into separate files, and create different environment depending on what we want to release You can have unlimited recursion of include, the manifest configs will simply be concatenated as if it was in a single file. command "repo manifest" will create a single manifest, and not recreate the manifest hierarchy for example: Our developement manifest will look like: <?xml version='1.0' encoding='UTF-8'?> <manifest> <default revision="platform/android/main" remote="intel"/> <include name="server.xml"/> <!-- The Server configuration --> <include name="aosp.xml" /> <!-- All the AOSP projects --> <include name="bsp.xml" /> <!-- The BSP projects that we release in source form --> <include name="bsp-priv.xml" /> <!-- The source of the BSP projects we release in binary form --> </manifest> Our release manifest will look like: <?xml version='1.0' encoding='UTF-8'?> <manifest> <default revision="platform/android/release-ext" remote="intel"/> <include name="server.xml"/> <!-- The Server configuration --> <include name="aosp.xml" /> <!-- All the AOSP projects --> <include name="bsp.xml" /> <!-- The BSP projects that we release in source form --> <include name="bsp-ext.xml" /> <!-- The PREBUILT version of the BSP projects we release in binary form --> </manifest> And it is also easy to create and maintain feature branch with a manifest that looks like: <?xml version='1.0' encoding='UTF-8'?> <manifest> <default revision="feature_branch_foobar" remote="intel"/> <include name="server.xml"/> <!-- The Server configuration --> <include name="aosp.xml" /> <!-- All the AOSP projects --> <include name="bsp.xml" /> <!-- The BSP projects that we release in source form --> <include name="bsp-priv.xml" /> <!-- The source of the BSP projects we release in binary form --> </manifest> Signed-off-by: Brian Harring <brian.harring@intel.com> Signed-off-by: Pierre Tardy <pierre.tardy@intel.com> Change-Id: I833a30d303039e485888768e6b81561b7665e89d
2011-04-28 12:04:41 +00:00
def _ParseManifest(self, node_list):
for node in itertools.chain(*node_list):
2008-10-21 14:00:00 +00:00
if node.nodeName == 'remote':
remote = self._ParseRemote(node)
if remote:
if remote.name in self._remotes:
if remote != self._remotes[remote.name]:
raise ManifestParseError(
'remote %s already exists with different attributes' %
(remote.name))
else:
self._remotes[remote.name] = remote
2008-10-21 14:00:00 +00:00
for node in itertools.chain(*node_list):
2008-10-21 14:00:00 +00:00
if node.nodeName == 'default':
if self._default is not None:
Support repo-level pre-upload hook and prep for future hooks. All repo-level hooks are expected to live in a single project at the top level of that project. The name of the hooks project is provided in the manifest.xml. The manifest also lists which hooks are enabled to make it obvious if a file somehow failed to sync down (or got deleted). Before running any hook, we will prompt the user to make sure that it is OK. A user can deny running the hook, allow once, or allow "forever" (until hooks change). This tries to keep with the git spirit of not automatically running anything on the user's computer that got synced down. Note that individual repo commands can add always options to avoid these prompts as they see fit (see below for the 'upload' options). When hooks are run, they are loaded into the current interpreter (the one running repo) and their main() function is run. This mechanism is used (instead of using subprocess) to make it easier to expand to a richer hook interface in the future. During loading, the interpreter's sys.path is updated to contain the directory containing the hooks so that hooks can be split into multiple files. The upload command has two options that control hook behavior: - no-verify=False, verify=False (DEFAULT): If stdout is a tty, can prompt about running upload hooks if needed. If user denies running hooks, the upload is cancelled. If stdout is not a tty and we would need to prompt about upload hooks, upload is cancelled. - no-verify=False, verify=True: Always run upload hooks with no prompt. - no-verify=True, verify=False: Never run upload hooks, but upload anyway (AKA bypass hooks). - no-verify=True, verify=True: Invalid Sample bit of manifest.xml code for enabling hooks (assumes you have a project named 'hooks' where hooks are stored): <repo-hooks in-project="hooks" enabled-list="pre-upload" /> Sample main() function in pre-upload.py in hooks directory: def main(project_list, **kwargs): print ('These projects will be uploaded: %s' % ', '.join(project_list)) print ('I am being a good boy and ignoring anything in kwargs\n' 'that I don\'t understand.') print 'I fail 50% of the time. How flaky.' if random.random() <= .5: raise Exception('Pre-upload hook failed. Have a nice day.') Change-Id: I5cefa2cd5865c72589263cf8e2f152a43c122f70
2011-03-04 19:54:18 +00:00
raise ManifestParseError(
'duplicate default in %s' %
(self.manifestFile))
2008-10-21 14:00:00 +00:00
self._default = self._ParseDefault(node)
if self._default is None:
self._default = _Default()
for node in itertools.chain(*node_list):
if node.nodeName == 'notice':
if self._notice is not None:
Support repo-level pre-upload hook and prep for future hooks. All repo-level hooks are expected to live in a single project at the top level of that project. The name of the hooks project is provided in the manifest.xml. The manifest also lists which hooks are enabled to make it obvious if a file somehow failed to sync down (or got deleted). Before running any hook, we will prompt the user to make sure that it is OK. A user can deny running the hook, allow once, or allow "forever" (until hooks change). This tries to keep with the git spirit of not automatically running anything on the user's computer that got synced down. Note that individual repo commands can add always options to avoid these prompts as they see fit (see below for the 'upload' options). When hooks are run, they are loaded into the current interpreter (the one running repo) and their main() function is run. This mechanism is used (instead of using subprocess) to make it easier to expand to a richer hook interface in the future. During loading, the interpreter's sys.path is updated to contain the directory containing the hooks so that hooks can be split into multiple files. The upload command has two options that control hook behavior: - no-verify=False, verify=False (DEFAULT): If stdout is a tty, can prompt about running upload hooks if needed. If user denies running hooks, the upload is cancelled. If stdout is not a tty and we would need to prompt about upload hooks, upload is cancelled. - no-verify=False, verify=True: Always run upload hooks with no prompt. - no-verify=True, verify=False: Never run upload hooks, but upload anyway (AKA bypass hooks). - no-verify=True, verify=True: Invalid Sample bit of manifest.xml code for enabling hooks (assumes you have a project named 'hooks' where hooks are stored): <repo-hooks in-project="hooks" enabled-list="pre-upload" /> Sample main() function in pre-upload.py in hooks directory: def main(project_list, **kwargs): print ('These projects will be uploaded: %s' % ', '.join(project_list)) print ('I am being a good boy and ignoring anything in kwargs\n' 'that I don\'t understand.') print 'I fail 50% of the time. How flaky.' if random.random() <= .5: raise Exception('Pre-upload hook failed. Have a nice day.') Change-Id: I5cefa2cd5865c72589263cf8e2f152a43c122f70
2011-03-04 19:54:18 +00:00
raise ManifestParseError(
'duplicate notice in %s' %
(self.manifestFile))
self._notice = self._ParseNotice(node)
for node in itertools.chain(*node_list):
if node.nodeName == 'manifest-server':
url = self._reqatt(node, 'url')
if self._manifest_server is not None:
raise ManifestParseError(
'duplicate manifest-server in %s' %
(self.manifestFile))
self._manifest_server = url
Represent git-submodule as nested projects, take 2 (Previous submission of this change broke Android buildbot due to incorrect regular expression for parsing git-config output. During investigation, we also found that Android, which pulls Chromium, has a workaround for Chromium's submodules; its manifest includes Chromium's submodules. This new change, in addition to fixing the regex, also take this type of workarounds into consideration; it adds a new attribute that makes repo not fetch submodules unless submodules have a project element defined in the manifest, or this attribute is overridden by a parent project element or by the default element.) We need a representation of git-submodule in repo; otherwise repo will not sync submodules, and leave workspace in a broken state. Of course this will not be a problem if all projects are owned by the owner of the manifest file, who may simply choose not to use git-submodule in all projects. However, this is not possible in practice because manifest file owner is unlikely to own all upstream projects. As git submodules are simply git repositories, it is natural to treat them as plain repo projects that live inside a repo project. That is, we could use recursively declared projects to denote the is-submodule relation of git repositories. The behavior of repo remains the same to projects that do not have a sub-project within. As for parent projects, repo fetches them and their sub-projects as normal projects, and then checks out subprojects at the commit specified in parent's commit object. The sub-project is fetched at a path relative to parent project's working directory; so the path specified in manifest file should match that of .gitmodules file. If a submodule is not registered in repo manifest, repo will derive its properties from itself and its parent project, which might not always be correct. In such cases, the subproject is called a derived subproject. To a user, a sub-project is merely a git-submodule; so all tips of working with a git-submodule apply here, too. For example, you should not run `repo sync` in a parent repository if its submodule is dirty. Change-Id: I4b8344c1b9ccad2f58ad304573133e5d52e1faef
2012-01-11 03:28:42 +00:00
def recursively_add_projects(project):
if self._projects.get(project.name):
raise ManifestParseError(
'duplicate project %s in %s' %
(project.name, self.manifestFile))
self._projects[project.name] = project
for subproject in project.subprojects:
recursively_add_projects(subproject)
for node in itertools.chain(*node_list):
2008-10-21 14:00:00 +00:00
if node.nodeName == 'project':
project = self._ParseProject(node)
Represent git-submodule as nested projects, take 2 (Previous submission of this change broke Android buildbot due to incorrect regular expression for parsing git-config output. During investigation, we also found that Android, which pulls Chromium, has a workaround for Chromium's submodules; its manifest includes Chromium's submodules. This new change, in addition to fixing the regex, also take this type of workarounds into consideration; it adds a new attribute that makes repo not fetch submodules unless submodules have a project element defined in the manifest, or this attribute is overridden by a parent project element or by the default element.) We need a representation of git-submodule in repo; otherwise repo will not sync submodules, and leave workspace in a broken state. Of course this will not be a problem if all projects are owned by the owner of the manifest file, who may simply choose not to use git-submodule in all projects. However, this is not possible in practice because manifest file owner is unlikely to own all upstream projects. As git submodules are simply git repositories, it is natural to treat them as plain repo projects that live inside a repo project. That is, we could use recursively declared projects to denote the is-submodule relation of git repositories. The behavior of repo remains the same to projects that do not have a sub-project within. As for parent projects, repo fetches them and their sub-projects as normal projects, and then checks out subprojects at the commit specified in parent's commit object. The sub-project is fetched at a path relative to parent project's working directory; so the path specified in manifest file should match that of .gitmodules file. If a submodule is not registered in repo manifest, repo will derive its properties from itself and its parent project, which might not always be correct. In such cases, the subproject is called a derived subproject. To a user, a sub-project is merely a git-submodule; so all tips of working with a git-submodule apply here, too. For example, you should not run `repo sync` in a parent repository if its submodule is dirty. Change-Id: I4b8344c1b9ccad2f58ad304573133e5d52e1faef
2012-01-11 03:28:42 +00:00
recursively_add_projects(project)
Support repo-level pre-upload hook and prep for future hooks. All repo-level hooks are expected to live in a single project at the top level of that project. The name of the hooks project is provided in the manifest.xml. The manifest also lists which hooks are enabled to make it obvious if a file somehow failed to sync down (or got deleted). Before running any hook, we will prompt the user to make sure that it is OK. A user can deny running the hook, allow once, or allow "forever" (until hooks change). This tries to keep with the git spirit of not automatically running anything on the user's computer that got synced down. Note that individual repo commands can add always options to avoid these prompts as they see fit (see below for the 'upload' options). When hooks are run, they are loaded into the current interpreter (the one running repo) and their main() function is run. This mechanism is used (instead of using subprocess) to make it easier to expand to a richer hook interface in the future. During loading, the interpreter's sys.path is updated to contain the directory containing the hooks so that hooks can be split into multiple files. The upload command has two options that control hook behavior: - no-verify=False, verify=False (DEFAULT): If stdout is a tty, can prompt about running upload hooks if needed. If user denies running hooks, the upload is cancelled. If stdout is not a tty and we would need to prompt about upload hooks, upload is cancelled. - no-verify=False, verify=True: Always run upload hooks with no prompt. - no-verify=True, verify=False: Never run upload hooks, but upload anyway (AKA bypass hooks). - no-verify=True, verify=True: Invalid Sample bit of manifest.xml code for enabling hooks (assumes you have a project named 'hooks' where hooks are stored): <repo-hooks in-project="hooks" enabled-list="pre-upload" /> Sample main() function in pre-upload.py in hooks directory: def main(project_list, **kwargs): print ('These projects will be uploaded: %s' % ', '.join(project_list)) print ('I am being a good boy and ignoring anything in kwargs\n' 'that I don\'t understand.') print 'I fail 50% of the time. How flaky.' if random.random() <= .5: raise Exception('Pre-upload hook failed. Have a nice day.') Change-Id: I5cefa2cd5865c72589263cf8e2f152a43c122f70
2011-03-04 19:54:18 +00:00
if node.nodeName == 'repo-hooks':
# Get the name of the project and the (space-separated) list of enabled.
repo_hooks_project = self._reqatt(node, 'in-project')
enabled_repo_hooks = self._reqatt(node, 'enabled-list').split()
# Only one project can be the hooks project
if self._repo_hooks_project is not None:
raise ManifestParseError(
'duplicate repo-hooks in %s' %
(self.manifestFile))
# Store a reference to the Project.
try:
self._repo_hooks_project = self._projects[repo_hooks_project]
except KeyError:
raise ManifestParseError(
'project %s not found for repo-hooks' %
(repo_hooks_project))
# Store the enabled hooks in the Project object.
self._repo_hooks_project.enabled_repo_hooks = enabled_repo_hooks
if node.nodeName == 'remove-project':
name = self._reqatt(node, 'name')
try:
del self._projects[name]
except KeyError:
raise ManifestParseError('remove-project element specifies non-existent '
'project: %s' % name)
# If the manifest removes the hooks project, treat it as if it deleted
# the repo-hooks element too.
if self._repo_hooks_project and (self._repo_hooks_project.name == name):
self._repo_hooks_project = None
Support repo-level pre-upload hook and prep for future hooks. All repo-level hooks are expected to live in a single project at the top level of that project. The name of the hooks project is provided in the manifest.xml. The manifest also lists which hooks are enabled to make it obvious if a file somehow failed to sync down (or got deleted). Before running any hook, we will prompt the user to make sure that it is OK. A user can deny running the hook, allow once, or allow "forever" (until hooks change). This tries to keep with the git spirit of not automatically running anything on the user's computer that got synced down. Note that individual repo commands can add always options to avoid these prompts as they see fit (see below for the 'upload' options). When hooks are run, they are loaded into the current interpreter (the one running repo) and their main() function is run. This mechanism is used (instead of using subprocess) to make it easier to expand to a richer hook interface in the future. During loading, the interpreter's sys.path is updated to contain the directory containing the hooks so that hooks can be split into multiple files. The upload command has two options that control hook behavior: - no-verify=False, verify=False (DEFAULT): If stdout is a tty, can prompt about running upload hooks if needed. If user denies running hooks, the upload is cancelled. If stdout is not a tty and we would need to prompt about upload hooks, upload is cancelled. - no-verify=False, verify=True: Always run upload hooks with no prompt. - no-verify=True, verify=False: Never run upload hooks, but upload anyway (AKA bypass hooks). - no-verify=True, verify=True: Invalid Sample bit of manifest.xml code for enabling hooks (assumes you have a project named 'hooks' where hooks are stored): <repo-hooks in-project="hooks" enabled-list="pre-upload" /> Sample main() function in pre-upload.py in hooks directory: def main(project_list, **kwargs): print ('These projects will be uploaded: %s' % ', '.join(project_list)) print ('I am being a good boy and ignoring anything in kwargs\n' 'that I don\'t understand.') print 'I fail 50% of the time. How flaky.' if random.random() <= .5: raise Exception('Pre-upload hook failed. Have a nice day.') Change-Id: I5cefa2cd5865c72589263cf8e2f152a43c122f70
2011-03-04 19:54:18 +00:00
def _AddMetaProjectMirror(self, m):
name = None
m_url = m.GetRemote(m.remote.name).url
if m_url.endswith('/.git'):
raise ManifestParseError, 'refusing to mirror %s' % m_url
if self._default and self._default.remote:
url = self._default.remote.resolvedFetchUrl
if not url.endswith('/'):
url += '/'
if m_url.startswith(url):
remote = self._default.remote
name = m_url[len(url):]
if name is None:
s = m_url.rindex('/') + 1
manifestUrl = self.manifestProject.config.GetString('remote.origin.url')
remote = _XmlRemote('origin', fetch=m_url[:s], manifestUrl=manifestUrl)
name = m_url[s:]
if name.endswith('.git'):
name = name[:-4]
if name not in self._projects:
m.PreSync()
gitdir = os.path.join(self.topdir, '%s.git' % name)
project = Project(manifest = self,
name = name,
remote = remote.ToRemoteSpec(name),
gitdir = gitdir,
worktree = None,
relpath = None,
revisionExpr = m.revisionExpr,
revisionId = None)
self._projects[project.name] = project
2008-10-21 14:00:00 +00:00
def _ParseRemote(self, node):
"""
reads a <remote> element from the manifest file
"""
name = self._reqatt(node, 'name')
Add remote alias support in manifest The `alias` is an optional attribute in element `remote`. It can be used to override attibute `name` to be set as the remote name in each project's .git/config. Its value can be duplicated while attribute `name` has to be unique across the manifest file. This helps each project to be able to have same remote name which actually points to different remote url. It eases some automation scripts to be able to checkout/push to same remote name but actually different remote url, like: repo forall -c "git checkout -b work same_remote/work" repo forall -c "git push same_remote work:work" for example: The manifest with 'alias' will look like: <?xml version='1.0' encoding='UTF-8'?> <manifest> <remote alias="same_alias" fetch="git://git.external1.org/" name="ext1" review="http://review.external1.org"/> <remote alias="same_alias" fetch="git://git.external2.org/" name="ext2" review="http://review.external2.org"/> <remote alias="same_alias" fetch="ssh://git.internal.com:29418" name="int" review="http://review.internal.com"/> <default remote="int" revision="int-branch" sync-j="2"/> <project name="path/to/project1" path="project1" remote="ext1"/> <project name="path/to/project2" path="project2" remote="ext2"/> <project name="path/to/project3" path="project3"/> ... </manifest> In each project, use command "git remote -v" project1: same_alias git://git.external1.org/project1 (fetch) same_alias git://git.external1.org/project1 (push) project2: same_alias git://git.external2.org/project2 (fetch) same_alias git://git.external2.org/project2 (push) project3: same_alias ssh://git.internal.com:29418/project3 (fetch) same_alias ssh://git.internal.com:29418/project3 (push) Change-Id: I2c48263097ff107f0c978f3e83966ae71d06cb90
2012-07-02 14:32:50 +00:00
alias = node.getAttribute('alias')
if alias == '':
alias = None
2008-10-21 14:00:00 +00:00
fetch = self._reqatt(node, 'fetch')
review = node.getAttribute('review')
if review == '':
review = None
manifestUrl = self.manifestProject.config.GetString('remote.origin.url')
Add remote alias support in manifest The `alias` is an optional attribute in element `remote`. It can be used to override attibute `name` to be set as the remote name in each project's .git/config. Its value can be duplicated while attribute `name` has to be unique across the manifest file. This helps each project to be able to have same remote name which actually points to different remote url. It eases some automation scripts to be able to checkout/push to same remote name but actually different remote url, like: repo forall -c "git checkout -b work same_remote/work" repo forall -c "git push same_remote work:work" for example: The manifest with 'alias' will look like: <?xml version='1.0' encoding='UTF-8'?> <manifest> <remote alias="same_alias" fetch="git://git.external1.org/" name="ext1" review="http://review.external1.org"/> <remote alias="same_alias" fetch="git://git.external2.org/" name="ext2" review="http://review.external2.org"/> <remote alias="same_alias" fetch="ssh://git.internal.com:29418" name="int" review="http://review.internal.com"/> <default remote="int" revision="int-branch" sync-j="2"/> <project name="path/to/project1" path="project1" remote="ext1"/> <project name="path/to/project2" path="project2" remote="ext2"/> <project name="path/to/project3" path="project3"/> ... </manifest> In each project, use command "git remote -v" project1: same_alias git://git.external1.org/project1 (fetch) same_alias git://git.external1.org/project1 (push) project2: same_alias git://git.external2.org/project2 (fetch) same_alias git://git.external2.org/project2 (push) project3: same_alias ssh://git.internal.com:29418/project3 (fetch) same_alias ssh://git.internal.com:29418/project3 (push) Change-Id: I2c48263097ff107f0c978f3e83966ae71d06cb90
2012-07-02 14:32:50 +00:00
return _XmlRemote(name, alias, fetch, manifestUrl, review)
2008-10-21 14:00:00 +00:00
def _ParseDefault(self, node):
"""
reads a <default> element from the manifest file
"""
d = _Default()
d.remote = self._get_remote(node)
d.revisionExpr = node.getAttribute('revision')
if d.revisionExpr == '':
d.revisionExpr = None
sync_j = node.getAttribute('sync-j')
if sync_j == '' or sync_j is None:
d.sync_j = 1
else:
d.sync_j = int(sync_j)
sync_c = node.getAttribute('sync-c')
if not sync_c:
d.sync_c = False
else:
d.sync_c = sync_c.lower() in ("yes", "true", "1")
Represent git-submodule as nested projects, take 2 (Previous submission of this change broke Android buildbot due to incorrect regular expression for parsing git-config output. During investigation, we also found that Android, which pulls Chromium, has a workaround for Chromium's submodules; its manifest includes Chromium's submodules. This new change, in addition to fixing the regex, also take this type of workarounds into consideration; it adds a new attribute that makes repo not fetch submodules unless submodules have a project element defined in the manifest, or this attribute is overridden by a parent project element or by the default element.) We need a representation of git-submodule in repo; otherwise repo will not sync submodules, and leave workspace in a broken state. Of course this will not be a problem if all projects are owned by the owner of the manifest file, who may simply choose not to use git-submodule in all projects. However, this is not possible in practice because manifest file owner is unlikely to own all upstream projects. As git submodules are simply git repositories, it is natural to treat them as plain repo projects that live inside a repo project. That is, we could use recursively declared projects to denote the is-submodule relation of git repositories. The behavior of repo remains the same to projects that do not have a sub-project within. As for parent projects, repo fetches them and their sub-projects as normal projects, and then checks out subprojects at the commit specified in parent's commit object. The sub-project is fetched at a path relative to parent project's working directory; so the path specified in manifest file should match that of .gitmodules file. If a submodule is not registered in repo manifest, repo will derive its properties from itself and its parent project, which might not always be correct. In such cases, the subproject is called a derived subproject. To a user, a sub-project is merely a git-submodule; so all tips of working with a git-submodule apply here, too. For example, you should not run `repo sync` in a parent repository if its submodule is dirty. Change-Id: I4b8344c1b9ccad2f58ad304573133e5d52e1faef
2012-01-11 03:28:42 +00:00
sync_s = node.getAttribute('sync-s')
if not sync_s:
d.sync_s = False
else:
d.sync_s = sync_s.lower() in ("yes", "true", "1")
2008-10-21 14:00:00 +00:00
return d
def _ParseNotice(self, node):
"""
reads a <notice> element from the manifest file
The <notice> element is distinct from other tags in the XML in that the
data is conveyed between the start and end tag (it's not an empty-element
tag).
The white space (carriage returns, indentation) for the notice element is
relevant and is parsed in a way that is based on how python docstrings work.
In fact, the code is remarkably similar to here:
http://www.python.org/dev/peps/pep-0257/
"""
# Get the data out of the node...
notice = node.childNodes[0].data
# Figure out minimum indentation, skipping the first line (the same line
# as the <notice> tag)...
minIndent = sys.maxint
lines = notice.splitlines()
for line in lines[1:]:
lstrippedLine = line.lstrip()
if lstrippedLine:
indent = len(line) - len(lstrippedLine)
minIndent = min(indent, minIndent)
# Strip leading / trailing blank lines and also indentation.
cleanLines = [lines[0].strip()]
for line in lines[1:]:
cleanLines.append(line[minIndent:].rstrip())
# Clear completely blank lines from front and back...
while cleanLines and not cleanLines[0]:
del cleanLines[0]
while cleanLines and not cleanLines[-1]:
del cleanLines[-1]
return '\n'.join(cleanLines)
Represent git-submodule as nested projects, take 2 (Previous submission of this change broke Android buildbot due to incorrect regular expression for parsing git-config output. During investigation, we also found that Android, which pulls Chromium, has a workaround for Chromium's submodules; its manifest includes Chromium's submodules. This new change, in addition to fixing the regex, also take this type of workarounds into consideration; it adds a new attribute that makes repo not fetch submodules unless submodules have a project element defined in the manifest, or this attribute is overridden by a parent project element or by the default element.) We need a representation of git-submodule in repo; otherwise repo will not sync submodules, and leave workspace in a broken state. Of course this will not be a problem if all projects are owned by the owner of the manifest file, who may simply choose not to use git-submodule in all projects. However, this is not possible in practice because manifest file owner is unlikely to own all upstream projects. As git submodules are simply git repositories, it is natural to treat them as plain repo projects that live inside a repo project. That is, we could use recursively declared projects to denote the is-submodule relation of git repositories. The behavior of repo remains the same to projects that do not have a sub-project within. As for parent projects, repo fetches them and their sub-projects as normal projects, and then checks out subprojects at the commit specified in parent's commit object. The sub-project is fetched at a path relative to parent project's working directory; so the path specified in manifest file should match that of .gitmodules file. If a submodule is not registered in repo manifest, repo will derive its properties from itself and its parent project, which might not always be correct. In such cases, the subproject is called a derived subproject. To a user, a sub-project is merely a git-submodule; so all tips of working with a git-submodule apply here, too. For example, you should not run `repo sync` in a parent repository if its submodule is dirty. Change-Id: I4b8344c1b9ccad2f58ad304573133e5d52e1faef
2012-01-11 03:28:42 +00:00
def _JoinName(self, parent_name, name):
return os.path.join(parent_name, name)
def _UnjoinName(self, parent_name, name):
return os.path.relpath(name, parent_name)
def _ParseProject(self, node, parent = None):
2008-10-21 14:00:00 +00:00
"""
reads a <project> element from the manifest file
"""
2008-10-21 14:00:00 +00:00
name = self._reqatt(node, 'name')
Represent git-submodule as nested projects, take 2 (Previous submission of this change broke Android buildbot due to incorrect regular expression for parsing git-config output. During investigation, we also found that Android, which pulls Chromium, has a workaround for Chromium's submodules; its manifest includes Chromium's submodules. This new change, in addition to fixing the regex, also take this type of workarounds into consideration; it adds a new attribute that makes repo not fetch submodules unless submodules have a project element defined in the manifest, or this attribute is overridden by a parent project element or by the default element.) We need a representation of git-submodule in repo; otherwise repo will not sync submodules, and leave workspace in a broken state. Of course this will not be a problem if all projects are owned by the owner of the manifest file, who may simply choose not to use git-submodule in all projects. However, this is not possible in practice because manifest file owner is unlikely to own all upstream projects. As git submodules are simply git repositories, it is natural to treat them as plain repo projects that live inside a repo project. That is, we could use recursively declared projects to denote the is-submodule relation of git repositories. The behavior of repo remains the same to projects that do not have a sub-project within. As for parent projects, repo fetches them and their sub-projects as normal projects, and then checks out subprojects at the commit specified in parent's commit object. The sub-project is fetched at a path relative to parent project's working directory; so the path specified in manifest file should match that of .gitmodules file. If a submodule is not registered in repo manifest, repo will derive its properties from itself and its parent project, which might not always be correct. In such cases, the subproject is called a derived subproject. To a user, a sub-project is merely a git-submodule; so all tips of working with a git-submodule apply here, too. For example, you should not run `repo sync` in a parent repository if its submodule is dirty. Change-Id: I4b8344c1b9ccad2f58ad304573133e5d52e1faef
2012-01-11 03:28:42 +00:00
if parent:
name = self._JoinName(parent.name, name)
2008-10-21 14:00:00 +00:00
remote = self._get_remote(node)
if remote is None:
remote = self._default.remote
if remote is None:
raise ManifestParseError, \
"no remote for project %s within %s" % \
(name, self.manifestFile)
revisionExpr = node.getAttribute('revision')
if not revisionExpr:
revisionExpr = self._default.revisionExpr
if not revisionExpr:
2008-10-21 14:00:00 +00:00
raise ManifestParseError, \
"no revision for project %s within %s" % \
(name, self.manifestFile)
path = node.getAttribute('path')
if not path:
path = name
if path.startswith('/'):
raise ManifestParseError, \
"project %s path cannot be absolute in %s" % \
(name, self.manifestFile)
rebase = node.getAttribute('rebase')
if not rebase:
rebase = True
else:
rebase = rebase.lower() in ("yes", "true", "1")
sync_c = node.getAttribute('sync-c')
if not sync_c:
sync_c = False
else:
sync_c = sync_c.lower() in ("yes", "true", "1")
Represent git-submodule as nested projects, take 2 (Previous submission of this change broke Android buildbot due to incorrect regular expression for parsing git-config output. During investigation, we also found that Android, which pulls Chromium, has a workaround for Chromium's submodules; its manifest includes Chromium's submodules. This new change, in addition to fixing the regex, also take this type of workarounds into consideration; it adds a new attribute that makes repo not fetch submodules unless submodules have a project element defined in the manifest, or this attribute is overridden by a parent project element or by the default element.) We need a representation of git-submodule in repo; otherwise repo will not sync submodules, and leave workspace in a broken state. Of course this will not be a problem if all projects are owned by the owner of the manifest file, who may simply choose not to use git-submodule in all projects. However, this is not possible in practice because manifest file owner is unlikely to own all upstream projects. As git submodules are simply git repositories, it is natural to treat them as plain repo projects that live inside a repo project. That is, we could use recursively declared projects to denote the is-submodule relation of git repositories. The behavior of repo remains the same to projects that do not have a sub-project within. As for parent projects, repo fetches them and their sub-projects as normal projects, and then checks out subprojects at the commit specified in parent's commit object. The sub-project is fetched at a path relative to parent project's working directory; so the path specified in manifest file should match that of .gitmodules file. If a submodule is not registered in repo manifest, repo will derive its properties from itself and its parent project, which might not always be correct. In such cases, the subproject is called a derived subproject. To a user, a sub-project is merely a git-submodule; so all tips of working with a git-submodule apply here, too. For example, you should not run `repo sync` in a parent repository if its submodule is dirty. Change-Id: I4b8344c1b9ccad2f58ad304573133e5d52e1faef
2012-01-11 03:28:42 +00:00
sync_s = node.getAttribute('sync-s')
if not sync_s:
sync_s = self._default.sync_s
else:
sync_s = sync_s.lower() in ("yes", "true", "1")
upstream = node.getAttribute('upstream')
groups = ''
if node.hasAttribute('groups'):
groups = node.getAttribute('groups')
groups = [x for x in re.split(r'[,\s]+', groups) if x]
Represent git-submodule as nested projects, take 2 (Previous submission of this change broke Android buildbot due to incorrect regular expression for parsing git-config output. During investigation, we also found that Android, which pulls Chromium, has a workaround for Chromium's submodules; its manifest includes Chromium's submodules. This new change, in addition to fixing the regex, also take this type of workarounds into consideration; it adds a new attribute that makes repo not fetch submodules unless submodules have a project element defined in the manifest, or this attribute is overridden by a parent project element or by the default element.) We need a representation of git-submodule in repo; otherwise repo will not sync submodules, and leave workspace in a broken state. Of course this will not be a problem if all projects are owned by the owner of the manifest file, who may simply choose not to use git-submodule in all projects. However, this is not possible in practice because manifest file owner is unlikely to own all upstream projects. As git submodules are simply git repositories, it is natural to treat them as plain repo projects that live inside a repo project. That is, we could use recursively declared projects to denote the is-submodule relation of git repositories. The behavior of repo remains the same to projects that do not have a sub-project within. As for parent projects, repo fetches them and their sub-projects as normal projects, and then checks out subprojects at the commit specified in parent's commit object. The sub-project is fetched at a path relative to parent project's working directory; so the path specified in manifest file should match that of .gitmodules file. If a submodule is not registered in repo manifest, repo will derive its properties from itself and its parent project, which might not always be correct. In such cases, the subproject is called a derived subproject. To a user, a sub-project is merely a git-submodule; so all tips of working with a git-submodule apply here, too. For example, you should not run `repo sync` in a parent repository if its submodule is dirty. Change-Id: I4b8344c1b9ccad2f58ad304573133e5d52e1faef
2012-01-11 03:28:42 +00:00
if parent is None:
relpath, worktree, gitdir = self.GetProjectPaths(name, path)
else:
Represent git-submodule as nested projects, take 2 (Previous submission of this change broke Android buildbot due to incorrect regular expression for parsing git-config output. During investigation, we also found that Android, which pulls Chromium, has a workaround for Chromium's submodules; its manifest includes Chromium's submodules. This new change, in addition to fixing the regex, also take this type of workarounds into consideration; it adds a new attribute that makes repo not fetch submodules unless submodules have a project element defined in the manifest, or this attribute is overridden by a parent project element or by the default element.) We need a representation of git-submodule in repo; otherwise repo will not sync submodules, and leave workspace in a broken state. Of course this will not be a problem if all projects are owned by the owner of the manifest file, who may simply choose not to use git-submodule in all projects. However, this is not possible in practice because manifest file owner is unlikely to own all upstream projects. As git submodules are simply git repositories, it is natural to treat them as plain repo projects that live inside a repo project. That is, we could use recursively declared projects to denote the is-submodule relation of git repositories. The behavior of repo remains the same to projects that do not have a sub-project within. As for parent projects, repo fetches them and their sub-projects as normal projects, and then checks out subprojects at the commit specified in parent's commit object. The sub-project is fetched at a path relative to parent project's working directory; so the path specified in manifest file should match that of .gitmodules file. If a submodule is not registered in repo manifest, repo will derive its properties from itself and its parent project, which might not always be correct. In such cases, the subproject is called a derived subproject. To a user, a sub-project is merely a git-submodule; so all tips of working with a git-submodule apply here, too. For example, you should not run `repo sync` in a parent repository if its submodule is dirty. Change-Id: I4b8344c1b9ccad2f58ad304573133e5d52e1faef
2012-01-11 03:28:42 +00:00
relpath, worktree, gitdir = self.GetSubprojectPaths(parent, path)
default_groups = ['all', 'name:%s' % name, 'path:%s' % relpath]
groups.extend(set(default_groups).difference(groups))
2008-10-21 14:00:00 +00:00
project = Project(manifest = self,
name = name,
remote = remote.ToRemoteSpec(name),
2008-10-21 14:00:00 +00:00
gitdir = gitdir,
worktree = worktree,
Represent git-submodule as nested projects, take 2 (Previous submission of this change broke Android buildbot due to incorrect regular expression for parsing git-config output. During investigation, we also found that Android, which pulls Chromium, has a workaround for Chromium's submodules; its manifest includes Chromium's submodules. This new change, in addition to fixing the regex, also take this type of workarounds into consideration; it adds a new attribute that makes repo not fetch submodules unless submodules have a project element defined in the manifest, or this attribute is overridden by a parent project element or by the default element.) We need a representation of git-submodule in repo; otherwise repo will not sync submodules, and leave workspace in a broken state. Of course this will not be a problem if all projects are owned by the owner of the manifest file, who may simply choose not to use git-submodule in all projects. However, this is not possible in practice because manifest file owner is unlikely to own all upstream projects. As git submodules are simply git repositories, it is natural to treat them as plain repo projects that live inside a repo project. That is, we could use recursively declared projects to denote the is-submodule relation of git repositories. The behavior of repo remains the same to projects that do not have a sub-project within. As for parent projects, repo fetches them and their sub-projects as normal projects, and then checks out subprojects at the commit specified in parent's commit object. The sub-project is fetched at a path relative to parent project's working directory; so the path specified in manifest file should match that of .gitmodules file. If a submodule is not registered in repo manifest, repo will derive its properties from itself and its parent project, which might not always be correct. In such cases, the subproject is called a derived subproject. To a user, a sub-project is merely a git-submodule; so all tips of working with a git-submodule apply here, too. For example, you should not run `repo sync` in a parent repository if its submodule is dirty. Change-Id: I4b8344c1b9ccad2f58ad304573133e5d52e1faef
2012-01-11 03:28:42 +00:00
relpath = relpath,
revisionExpr = revisionExpr,
revisionId = None,
rebase = rebase,
groups = groups,
sync_c = sync_c,
Represent git-submodule as nested projects, take 2 (Previous submission of this change broke Android buildbot due to incorrect regular expression for parsing git-config output. During investigation, we also found that Android, which pulls Chromium, has a workaround for Chromium's submodules; its manifest includes Chromium's submodules. This new change, in addition to fixing the regex, also take this type of workarounds into consideration; it adds a new attribute that makes repo not fetch submodules unless submodules have a project element defined in the manifest, or this attribute is overridden by a parent project element or by the default element.) We need a representation of git-submodule in repo; otherwise repo will not sync submodules, and leave workspace in a broken state. Of course this will not be a problem if all projects are owned by the owner of the manifest file, who may simply choose not to use git-submodule in all projects. However, this is not possible in practice because manifest file owner is unlikely to own all upstream projects. As git submodules are simply git repositories, it is natural to treat them as plain repo projects that live inside a repo project. That is, we could use recursively declared projects to denote the is-submodule relation of git repositories. The behavior of repo remains the same to projects that do not have a sub-project within. As for parent projects, repo fetches them and their sub-projects as normal projects, and then checks out subprojects at the commit specified in parent's commit object. The sub-project is fetched at a path relative to parent project's working directory; so the path specified in manifest file should match that of .gitmodules file. If a submodule is not registered in repo manifest, repo will derive its properties from itself and its parent project, which might not always be correct. In such cases, the subproject is called a derived subproject. To a user, a sub-project is merely a git-submodule; so all tips of working with a git-submodule apply here, too. For example, you should not run `repo sync` in a parent repository if its submodule is dirty. Change-Id: I4b8344c1b9ccad2f58ad304573133e5d52e1faef
2012-01-11 03:28:42 +00:00
sync_s = sync_s,
upstream = upstream,
parent = parent)
2008-10-21 14:00:00 +00:00
for n in node.childNodes:
if n.nodeName == 'copyfile':
2008-10-21 14:00:00 +00:00
self._ParseCopyFile(project, n)
if n.nodeName == 'annotation':
self._ParseAnnotation(project, n)
Represent git-submodule as nested projects, take 2 (Previous submission of this change broke Android buildbot due to incorrect regular expression for parsing git-config output. During investigation, we also found that Android, which pulls Chromium, has a workaround for Chromium's submodules; its manifest includes Chromium's submodules. This new change, in addition to fixing the regex, also take this type of workarounds into consideration; it adds a new attribute that makes repo not fetch submodules unless submodules have a project element defined in the manifest, or this attribute is overridden by a parent project element or by the default element.) We need a representation of git-submodule in repo; otherwise repo will not sync submodules, and leave workspace in a broken state. Of course this will not be a problem if all projects are owned by the owner of the manifest file, who may simply choose not to use git-submodule in all projects. However, this is not possible in practice because manifest file owner is unlikely to own all upstream projects. As git submodules are simply git repositories, it is natural to treat them as plain repo projects that live inside a repo project. That is, we could use recursively declared projects to denote the is-submodule relation of git repositories. The behavior of repo remains the same to projects that do not have a sub-project within. As for parent projects, repo fetches them and their sub-projects as normal projects, and then checks out subprojects at the commit specified in parent's commit object. The sub-project is fetched at a path relative to parent project's working directory; so the path specified in manifest file should match that of .gitmodules file. If a submodule is not registered in repo manifest, repo will derive its properties from itself and its parent project, which might not always be correct. In such cases, the subproject is called a derived subproject. To a user, a sub-project is merely a git-submodule; so all tips of working with a git-submodule apply here, too. For example, you should not run `repo sync` in a parent repository if its submodule is dirty. Change-Id: I4b8344c1b9ccad2f58ad304573133e5d52e1faef
2012-01-11 03:28:42 +00:00
if n.nodeName == 'project':
project.subprojects.append(self._ParseProject(n, parent = project))
2008-10-21 14:00:00 +00:00
return project
Represent git-submodule as nested projects, take 2 (Previous submission of this change broke Android buildbot due to incorrect regular expression for parsing git-config output. During investigation, we also found that Android, which pulls Chromium, has a workaround for Chromium's submodules; its manifest includes Chromium's submodules. This new change, in addition to fixing the regex, also take this type of workarounds into consideration; it adds a new attribute that makes repo not fetch submodules unless submodules have a project element defined in the manifest, or this attribute is overridden by a parent project element or by the default element.) We need a representation of git-submodule in repo; otherwise repo will not sync submodules, and leave workspace in a broken state. Of course this will not be a problem if all projects are owned by the owner of the manifest file, who may simply choose not to use git-submodule in all projects. However, this is not possible in practice because manifest file owner is unlikely to own all upstream projects. As git submodules are simply git repositories, it is natural to treat them as plain repo projects that live inside a repo project. That is, we could use recursively declared projects to denote the is-submodule relation of git repositories. The behavior of repo remains the same to projects that do not have a sub-project within. As for parent projects, repo fetches them and their sub-projects as normal projects, and then checks out subprojects at the commit specified in parent's commit object. The sub-project is fetched at a path relative to parent project's working directory; so the path specified in manifest file should match that of .gitmodules file. If a submodule is not registered in repo manifest, repo will derive its properties from itself and its parent project, which might not always be correct. In such cases, the subproject is called a derived subproject. To a user, a sub-project is merely a git-submodule; so all tips of working with a git-submodule apply here, too. For example, you should not run `repo sync` in a parent repository if its submodule is dirty. Change-Id: I4b8344c1b9ccad2f58ad304573133e5d52e1faef
2012-01-11 03:28:42 +00:00
def GetProjectPaths(self, name, path):
relpath = path
if self.IsMirror:
worktree = None
gitdir = os.path.join(self.topdir, '%s.git' % name)
else:
worktree = os.path.join(self.topdir, path).replace('\\', '/')
gitdir = os.path.join(self.repodir, 'projects', '%s.git' % path)
return relpath, worktree, gitdir
def GetSubprojectName(self, parent, submodule_path):
return os.path.join(parent.name, submodule_path)
def _JoinRelpath(self, parent_relpath, relpath):
return os.path.join(parent_relpath, relpath)
def _UnjoinRelpath(self, parent_relpath, relpath):
return os.path.relpath(relpath, parent_relpath)
def GetSubprojectPaths(self, parent, path):
relpath = self._JoinRelpath(parent.relpath, path)
gitdir = os.path.join(parent.gitdir, 'subprojects', '%s.git' % path)
if self.IsMirror:
worktree = None
else:
worktree = os.path.join(parent.worktree, path).replace('\\', '/')
return relpath, worktree, gitdir
2008-10-21 14:00:00 +00:00
def _ParseCopyFile(self, project, node):
src = self._reqatt(node, 'src')
dest = self._reqatt(node, 'dest')
if not self.IsMirror:
# src is project relative;
# dest is relative to the top of the tree
project.AddCopyFile(src, dest, os.path.join(self.topdir, dest))
2008-10-21 14:00:00 +00:00
def _ParseAnnotation(self, project, node):
name = self._reqatt(node, 'name')
value = self._reqatt(node, 'value')
try:
keep = self._reqatt(node, 'keep').lower()
except ManifestParseError:
keep = "true"
if keep != "true" and keep != "false":
raise ManifestParseError, "optional \"keep\" attribute must be \"true\" or \"false\""
project.AddAnnotation(name, value, keep)
2008-10-21 14:00:00 +00:00
def _get_remote(self, node):
name = node.getAttribute('remote')
if not name:
return None
v = self._remotes.get(name)
if not v:
raise ManifestParseError, \
"remote %s not defined in %s" % \
(name, self.manifestFile)
return v
def _reqatt(self, node, attname):
"""
reads a required attribute from the node.
"""
v = node.getAttribute(attname)
if not v:
raise ManifestParseError, \
"no %s in <%s> within %s" % \
(attname, node.nodeName, self.manifestFile)
return v