Chromium Code Reviews
|
|
Chromium Code Reviews
Issue
2875173002:
Add hook to download Fuchsia SDK (Closed)
DescriptionAdd hook to download Fuchsia SDK
Used in (new) hooks_os section of DEPS to only be run when target_os includes
'fuchsia'. Writes the current version into third_party/fuchsia-sdk/.hash to
avoid updating when no change has been made to the version.
BUG=706592
Review-Url: https://codereview.chromium.org/2875173002
Cr-Commit-Position: refs/heads/master@{#472008}
Committed: https://chromium.googlesource.com/chromium/src/+/4cd62becec2902f99720eb0c5b4602eecfbd6ef1
Patch Set 1 #Patch Set 2 : . #
Total comments: 2
Patch Set 3 : . #
Total comments: 4
Patch Set 4 : . #Patch Set 5 : use hooks_os #Patch Set 6 : . #
Total comments: 2
Patch Set 7 : note that we can move android hooks now #Messages
Total messages: 44 (20 generated)
The CQ bit was checked by scottmg@chromium.org to run a CQ dry run
Dry run: CQ is trying da patch. Follow status at: https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...
Description was changed from ========== Add hook to download Fuchsia SDK BUG=706592 ========== to ========== Add hook to download Fuchsia SDK Checks that .gclient target_os contains 'fuchsia' before taking any action. BUG=706592 ==========
Description was changed from ========== Add hook to download Fuchsia SDK Checks that .gclient target_os contains 'fuchsia' before taking any action. BUG=706592 ========== to ========== Add hook to download Fuchsia SDK Checks that .gclient target_os contains 'fuchsia' before taking any action. Writes the current version into $root/.hash to avoid updating when no change has been made to the version. BUG=706592 ==========
thakis@chromium.org changed reviewers: + thakis@chromium.org
Thanks! nit: I thought with $root you meant src/, but you mean the sdk root (which makes a lot more sense) – maybe you could tweak the cl description a bit. https://codereview.chromium.org/2875173002/diff/40001/build/fuchsia/update_sd... File build/fuchsia/update_sdk.py (right): https://codereview.chromium.org/2875173002/diff/40001/build/fuchsia/update_sd... build/fuchsia/update_sdk.py:20: import cr.base Since you're suing cr.auto and cr.loader below, shouldn't this be "import cr"? Also, isn't cr this mostly not officially supported gclient wrapper that someone wrote 'cause they didn't like using depot_tools directly? I'm not sure we should use that from hooks. Do other hooks use this? https://codereview.chromium.org/2875173002/diff/40001/build/fuchsia/update_sd... build/fuchsia/update_sdk.py:44: target_os = cr.context.gclient.get('target_os', []) How hard is it to add hooks_os to gclient directly, like deps_os? It looks like several of our hooks could be per-os hooks – we could move all of those android third-party hooks out of the run-everywhere hooks for example.
dpranke@chromium.org changed reviewers: + dpranke@chromium.org - thakis@chromium.org
https://codereview.chromium.org/2875173002/diff/20001/build/fuchsia/update_sd... File build/fuchsia/update_sdk.py (right): https://codereview.chromium.org/2875173002/diff/20001/build/fuchsia/update_sd... build/fuchsia/update_sdk.py:47: return 0 Why use CR for this, rather than just loading .gclient directly?
Description was changed from ========== Add hook to download Fuchsia SDK Checks that .gclient target_os contains 'fuchsia' before taking any action. Writes the current version into $root/.hash to avoid updating when no change has been made to the version. BUG=706592 ========== to ========== Add hook to download Fuchsia SDK Checks that .gclient target_os contains 'fuchsia' before taking any action. Writes the current version into third_party/fuchsia-sdk/.hash to avoid updating when no change has been made to the version. BUG=706592 ==========
scottmg@chromium.org changed reviewers: + thakis@chromium.org
Thanks https://codereview.chromium.org/2875173002/diff/20001/build/fuchsia/update_sd... File build/fuchsia/update_sdk.py (right): https://codereview.chromium.org/2875173002/diff/20001/build/fuchsia/update_sd... build/fuchsia/update_sdk.py:47: return 0 On 2017/05/11 22:31:30, Dirk Pranke wrote: > Why use CR for this, rather than just loading .gclient directly? No good reason. I'll do that if the hooks_os thing doesn't pan out. https://codereview.chromium.org/2875173002/diff/40001/build/fuchsia/update_sd... File build/fuchsia/update_sdk.py (right): https://codereview.chromium.org/2875173002/diff/40001/build/fuchsia/update_sd... build/fuchsia/update_sdk.py:20: import cr.base On 2017/05/11 22:31:19, Nico wrote: > Since you're suing cr.auto and cr.loader below, shouldn't this be "import cr"? > > Also, isn't cr this mostly not officially supported gclient wrapper that someone > wrote 'cause they didn't like using depot_tools directly? I'm not sure we should > use that from hooks. Do other hooks use this? I'd actually never heard of it before 20mins ago, I just cs'd for target_os and it seemed like a thing to use. Should we delete it or is it just sorta deprecated? https://codereview.chromium.org/2875173002/diff/40001/build/fuchsia/update_sd... build/fuchsia/update_sdk.py:44: target_os = cr.context.gclient.get('target_os', []) On 2017/05/11 22:31:19, Nico wrote: > How hard is it to add hooks_os to gclient directly, like deps_os? It looks like > several of our hooks could be per-os hooks – we could move all of those android > third-party hooks out of the run-everywhere hooks for example. I admit I just got a little avoid-y when I landed at https://cs.chromium.org/chromium/tools/depot_tools/gclient_eval.py?rcl=fc6905... . I can take a look to see how involved it'd be though.
On Thu, May 11, 2017 at 6:38 PM, <scottmg@chromium.org> wrote: > Reviewers: Dirk Pranke, Nico > CL: https://codereview.chromium.org/2875173002/ > > Message: > Thanks > > > https://codereview.chromium.org/2875173002/diff/20001/ > build/fuchsia/update_sdk.py > File build/fuchsia/update_sdk.py (right): > > https://codereview.chromium.org/2875173002/diff/20001/ > build/fuchsia/update_sdk.py#newcode47 > build/fuchsia/update_sdk.py:47: return 0 > On 2017/05/11 22:31:30, Dirk Pranke wrote: > > Why use CR for this, rather than just loading .gclient directly? > > No good reason. I'll do that if the hooks_os thing doesn't pan out. > > https://codereview.chromium.org/2875173002/diff/40001/ > build/fuchsia/update_sdk.py > File build/fuchsia/update_sdk.py (right): > > https://codereview.chromium.org/2875173002/diff/40001/ > build/fuchsia/update_sdk.py#newcode20 > build/fuchsia/update_sdk.py:20: import cr.base > On 2017/05/11 22:31:19, Nico wrote: > > Since you're suing cr.auto and cr.loader below, shouldn't this be > "import cr"? > > > > Also, isn't cr this mostly not officially supported gclient wrapper > that someone > > wrote 'cause they didn't like using depot_tools directly? I'm not sure > we should > > use that from hooks. Do other hooks use this? > > I'd actually never heard of it before 20mins ago, I just cs'd for > target_os and it seemed like a thing to use. Should we delete it or is > it just sorta deprecated? > > https://codereview.chromium.org/2875173002/diff/40001/ > build/fuchsia/update_sdk.py#newcode44 > build/fuchsia/update_sdk.py:44: target_os = > cr.context.gclient.get('target_os', []) > On 2017/05/11 22:31:19, Nico wrote: > > How hard is it to add hooks_os to gclient directly, like deps_os? It > looks like > > several of our hooks could be per-os hooks – we could move all of > those android > > third-party hooks out of the run-everywhere hooks for example. > > I admit I just got a little avoid-y when I landed at > https://cs.chromium.org/chromium/tools/depot_tools/gclient_eval.py?rcl= > fc6905908d38fd49928c15de786f7bd005d76b31&l=43 > . I can take a look to see how involved it'd be though. > Whoa yeah, I would've run too. Looks pretty new though: https://chromium.googlesource.com/chromium/tools/depot_tools/+/beec006697a116... So maybe there's a way from stopping this code from getting overly frameworky? > > Description: > Add hook to download Fuchsia SDK > > Checks that .gclient target_os contains 'fuchsia' before taking any action. > Writes the current version into third_party/fuchsia-sdk/.hash to avoid > updating > when no change has been made to the version. > > BUG=706592 > > Affected files (+88, -0 lines): > M DEPS > A build/fuchsia/update_sdk.py > > > Index: DEPS > diff --git a/DEPS b/DEPS > index 52a209de8fdf7dae7888243bd219335a12c1be1e.. > 252aa16fa564b61002fed7863595d24478b7f6e3 100644 > --- a/DEPS > +++ b/DEPS > @@ -599,6 +599,15 @@ hooks = [ > ], > }, > { > + 'name': 'fuchsia_sdk', > + 'pattern': '.', > + 'action': [ > + 'python', > + 'src/build/fuchsia/update_sdk.py', > + '3886a632c4a472fa7ef9eab35b45b63359e387fa', > + ], > + }, > + { > 'name': 'intellij', > 'pattern': '.', > 'action': ['python', > Index: build/fuchsia/update_sdk.py > diff --git a/build/fuchsia/update_sdk.py b/build/fuchsia/update_sdk.py > new file mode 100755 > index 0000000000000000000000000000000000000000.. > 0d2603d57a63f97991e3e64c2f5434f535072d8f > --- /dev/null > +++ b/build/fuchsia/update_sdk.py > @@ -0,0 +1,79 @@ > +#!/usr/bin/env python > +# Copyright 2017 The Chromium Authors. All rights reserved. > +# Use of this source code is governed by a BSD-style license that can be > +# found in the LICENSE file. > + > +"""Updates the Fuchsia SDK to the given revision. Checks .gclient > target_os > +so that it doesn't download unnecessarily.""" > + > +import os > +import shutil > +import subprocess > +import sys > +import tarfile > +import tempfile > + > +REPOSITORY_ROOT = os.path.abspath(os.path.join( > + os.path.dirname(__file__), '..', '..')) > +sys.path.append(os.path.join(REPOSITORY_ROOT, 'build')) > +sys.path.append(os.path.join(REPOSITORY_ROOT, 'tools', 'cr')) > + > +import cr.base > +import find_depot_tools > + > + > +def EnsureDirExists(path): > + if not os.path.exists(path): > + print 'Creating directory %s' % path > + os.makedirs(path) > + > + > +def main(): > + if len(sys.argv) != 2: > + print >>sys.stderr, 'usage: %s <sdk_hash>' % sys.argv[0] > + return 1 > + > + # Check whether we're targeting Fuchsia. > + user_path = os.path.expanduser(os.path.join('~', '.config', 'cr')) > + cr.auto.user.__path__.append(user_path) > + cr.loader.Scan() > + with cr.base.context.Create( > + description='Fuchsia SDK update tool', > + epilog='') as context: > + > + cr.base.client.DetectClient() > + target_os = cr.context.gclient.get('target_os', []) > + if 'fuchsia' not in target_os: > + # Nothing to do. > + return 0 > + > + sdk_hash = sys.argv[1] > + output_dir = os.path.join(REPOSITORY_ROOT, 'third_party', 'fuchsia-sdk') > + > + hash_filename = os.path.join(output_dir, '.hash') > + if os.path.exists(hash_filename): > + with open(hash_filename, 'r') as f: > + if f.read().strip() == sdk_hash: > + # Nothing to do. > + return 0 > + > + print 'Downloading SDK %s...' % sdk_hash > + > + if os.path.isdir(output_dir): > + shutil.rmtree(output_dir) > + > + bucket = 'gs://fuchsia-build/fuchsia/sdk/linux64/' > + with tempfile.NamedTemporaryFile() as f: > + cmd = [os.path.join(find_depot_tools.DEPOT_TOOLS_PATH, 'gsutil.py'), > + 'cp', bucket + sdk_hash, f.name] > + subprocess.check_call(cmd) > + f.seek(0) > + EnsureDirExists(output_dir) > + tarfile.open(mode='r:gz', fileobj=f).extractall(path=output_dir) > + > + with open(hash_filename, 'w') as f: > + f.write(sdk_hash) > + > + > +if __name__ == '__main__': > + sys.exit(main()) > > > -- You received this message because you are subscribed to the Google Groups "Chromium-reviews" group. To unsubscribe from this group and stop receiving emails from it, send an email to chromium-reviews+unsubscribe@chromium.org.
On 2017/05/11 22:38:18, scottmg wrote: > Thanks > > https://codereview.chromium.org/2875173002/diff/20001/build/fuchsia/update_sd... > File build/fuchsia/update_sdk.py (right): > > https://codereview.chromium.org/2875173002/diff/20001/build/fuchsia/update_sd... > build/fuchsia/update_sdk.py:47: return 0 > On 2017/05/11 22:31:30, Dirk Pranke wrote: > > Why use CR for this, rather than just loading .gclient directly? > > No good reason. I'll do that if the hooks_os thing doesn't pan out. > > https://codereview.chromium.org/2875173002/diff/40001/build/fuchsia/update_sd... > File build/fuchsia/update_sdk.py (right): > > https://codereview.chromium.org/2875173002/diff/40001/build/fuchsia/update_sd... > build/fuchsia/update_sdk.py:20: import cr.base > On 2017/05/11 22:31:19, Nico wrote: > > Since you're suing cr.auto and cr.loader below, shouldn't this be "import cr"? > > > > Also, isn't cr this mostly not officially supported gclient wrapper that > someone > > wrote 'cause they didn't like using depot_tools directly? I'm not sure we > should > > use that from hooks. Do other hooks use this? > > I'd actually never heard of it before 20mins ago, I just cs'd for target_os and > it seemed like a thing to use. Should we delete it or is it just sorta > deprecated? > > https://codereview.chromium.org/2875173002/diff/40001/build/fuchsia/update_sd... > build/fuchsia/update_sdk.py:44: target_os = cr.context.gclient.get('target_os', > []) > On 2017/05/11 22:31:19, Nico wrote: > > How hard is it to add hooks_os to gclient directly, like deps_os? It looks > like > > several of our hooks could be per-os hooks – we could move all of those > android > > third-party hooks out of the run-everywhere hooks for example. > > I admit I just got a little avoid-y when I landed at > https://cs.chromium.org/chromium/tools/depot_tools/gclient_eval.py?rcl=fc6905... > . I can take a look to see how involved it'd be though. Actually, one thing that comes to mind is that we lose the defined ordering that a bunch of hooks rely on, if we have separate blocks. Hmm. It wouldn't matter for this one, but as a general thing, it looks like the Android ones have some of that.
On Thu, May 11, 2017 at 6:40 PM, <scottmg@chromium.org> wrote: > On 2017/05/11 22:38:18, scottmg wrote: > > Thanks > > > > > https://codereview.chromium.org/2875173002/diff/20001/ > build/fuchsia/update_sdk.py > > File build/fuchsia/update_sdk.py (right): > > > > > https://codereview.chromium.org/2875173002/diff/20001/ > build/fuchsia/update_sdk.py#newcode47 > > build/fuchsia/update_sdk.py:47: return 0 > > On 2017/05/11 22:31:30, Dirk Pranke wrote: > > > Why use CR for this, rather than just loading .gclient directly? > > > > No good reason. I'll do that if the hooks_os thing doesn't pan out. > > > > > https://codereview.chromium.org/2875173002/diff/40001/ > build/fuchsia/update_sdk.py > > File build/fuchsia/update_sdk.py (right): > > > > > https://codereview.chromium.org/2875173002/diff/40001/ > build/fuchsia/update_sdk.py#newcode20 > > build/fuchsia/update_sdk.py:20: import cr.base > > On 2017/05/11 22:31:19, Nico wrote: > > > Since you're suing cr.auto and cr.loader below, shouldn't this be > "import > cr"? > > > > > > Also, isn't cr this mostly not officially supported gclient wrapper > that > > someone > > > wrote 'cause they didn't like using depot_tools directly? I'm not sure > we > > should > > > use that from hooks. Do other hooks use this? > > > > I'd actually never heard of it before 20mins ago, I just cs'd for > target_os > and > > it seemed like a thing to use. Should we delete it or is it just sorta > > deprecated? > > > > > https://codereview.chromium.org/2875173002/diff/40001/ > build/fuchsia/update_sdk.py#newcode44 > > build/fuchsia/update_sdk.py:44: target_os = > cr.context.gclient.get('target_os', > > []) > > On 2017/05/11 22:31:19, Nico wrote: > > > How hard is it to add hooks_os to gclient directly, like deps_os? It > looks > > like > > > several of our hooks could be per-os hooks – we could move all of those > > android > > > third-party hooks out of the run-everywhere hooks for example. > > > > I admit I just got a little avoid-y when I landed at > > > https://cs.chromium.org/chromium/tools/depot_tools/gclient_eval.py?rcl= > fc6905908d38fd49928c15de786f7bd005d76b31&l=43 > > . I can take a look to see how involved it'd be though. > > Actually, one thing that comes to mind is that we lose the defined > ordering that > a bunch of hooks rely on, if we have separate blocks. Hmm. It wouldn't > matter > for this one, but as a general thing, it looks like the Android ones have > some > of that. > Hm, but "first deps, then deps_os in .gclient target_os order" would probably be good enough for most things. Most deps just download stuff. -- You received this message because you are subscribed to the Google Groups "Chromium-reviews" group. To unsubscribe from this group and stop receiving emails from it, send an email to chromium-reviews+unsubscribe@chromium.org.
I believe the fuchsia sdk tarball is compatible with https://chromium.googlesource.com/chromium/tools/depot_tools/+/master/downloa... which may simplify the downloading process - if it isn't please let me know.
depot_tools change here: https://chromium-review.googlesource.com/503534
On 2017/05/11 23:21:18, jamesr1 wrote: > I believe the fuchsia sdk tarball is compatible with > https://chromium.googlesource.com/chromium/tools/depot_tools/+/master/downloa... > which may simplify the downloading process - if it isn't please let me know. Yeah, I think I could do that instead with the addition of hooks_os to gclient.
On 2017/05/11 23:21:18, jamesr1 wrote: > I believe the fuchsia sdk tarball is compatible with > https://chromium.googlesource.com/chromium/tools/depot_tools/+/master/downloa... > which may simplify the downloading process - if it isn't please let me know. It's whining about something. ________ running 'download_from_google_storage --no_resume --no_auth --extract --bucket fuchsia-build/fuchsia/sdk/linux64 -s src/third_party/fuchsia-sdk/fuchsia-sdk.tar.gz.sha1' in '/work/cr' src/third_party/fuchsia-sdk/fuchsia-sdk.tar.gz contains invalid entries. 0> Downloading src/third_party/fuchsia-sdk/fuchsia-sdk.tar.gz... 0> Error: src/third_party/fuchsia-sdk/fuchsia-sdk.tar.gz contains files outside src/third_party/fuchsia-sdk/fuchsia-sdk. Downloading 1 files took 7.279358 second(s) Error: Command 'download_from_google_storage --no_resume --no_auth --extract --bucket fuchsia-build/fuchsia/sdk/linux64 -s src/third_party/fuchsia-sdk/fuchsia-sdk.tar.gz.sha1' returned non-zero exit status 1 in /work/cr Maybe because there's a '.' in the root? I think d_f_g_s wants a subdir in the archive or something? Maybe there's some straightforward way to outwit it.
Ah, dfgs as always is my favourite thing. There might be more reasons, but at least these being symlinks it doesn't like: issym True ./sysroot/x86_64-fuchsia/debug-info/ld.so.1 issym True ./sysroot/x86_64-fuchsia/lib/libunwind.so issym True ./sysroot/x86_64-fuchsia/lib/libc++abi.so.1 issym True ./sysroot/x86_64-fuchsia/lib/libc++.so.2 issym True ./sysroot/x86_64-fuchsia/lib/libunwind.so.1 issym True ./sysroot/x86_64-fuchsia/lib/libc++abi.so It's also complaining about files outside the root, but that might be an incorrect error.
jamesr@google.com changed reviewers: + jamesr@google.com
I believe the symlinks are expected to be there by the toolchain (and possibly other tools, not sure), so I'm not sure we should try to remove them.
On 2017/05/11 23:41:16, scottmg wrote: > Ah, dfgs as always is my favourite thing. There might be more reasons, but at > least these being symlinks it doesn't like: > > issym True ./sysroot/x86_64-fuchsia/debug-info/ld.so.1 > issym True ./sysroot/x86_64-fuchsia/lib/libunwind.so > issym True ./sysroot/x86_64-fuchsia/lib/libc++abi.so.1 > issym True ./sysroot/x86_64-fuchsia/lib/libc++.so.2 > issym True ./sysroot/x86_64-fuchsia/lib/libunwind.so.1 > issym True ./sysroot/x86_64-fuchsia/lib/libc++abi.so > > It's also complaining about files outside the root, but that might be an > incorrect error. '.' being the root of the directory tree also confuses it. If I put the .sha1 file in third_party/fuchsia-sdk/ then it deletes that file when it extracts the archive (works, but confuses itself for next time, of course). I guess you'd probably have to repackage with something like 'sdk' or 'fuchsia-sdk' as the root, and then we'd have to stutter that in the build files too. Bah.
jamesr@google.com changed reviewers: - jamesr@google.com
I think we just shouldn't use download_from_google_storage for this - I'd rather not pick a directory structure when packaging up the sdk.
On 2017/05/11 23:49:38, jamesr1 wrote: > I believe the symlinks are expected to be there by the toolchain (and possibly > other tools, not sure), so I'm not sure we should try to remove them. Yeah. I added a symlink-allowing-flag to dfgs, but given the rooting problem too, I think I'll just stick with a gsutil wrapper.
Patchset #5 (id:80001) has been deleted
Description was changed from ========== Add hook to download Fuchsia SDK Checks that .gclient target_os contains 'fuchsia' before taking any action. Writes the current version into third_party/fuchsia-sdk/.hash to avoid updating when no change has been made to the version. BUG=706592 ========== to ========== Add hook to download Fuchsia SDK Used in (new) hooks_os section of DEPS to only be run when target_os includes 'fuchsia'. Writes the current version into third_party/fuchsia-sdk/.hash to avoid updating when no change has been made to the version. BUG=706592 ==========
hooks_os is in depot_tools now, anyone want to review what's left here now?
lgtm https://codereview.chromium.org/2875173002/diff/120001/DEPS File DEPS (right): https://codereview.chromium.org/2875173002/diff/120001/DEPS#newcode1174 DEPS:1174: ], Can you add a TODO here (on me, if you want) to move some of the android hooks to down here?
Thanks https://codereview.chromium.org/2875173002/diff/120001/DEPS File DEPS (right): https://codereview.chromium.org/2875173002/diff/120001/DEPS#newcode1174 DEPS:1174: ], On 2017/05/15 20:20:02, Nico wrote: > Can you add a TODO here (on me, if you want) to move some of the android hooks > to down here? Done.
The CQ bit was checked by scottmg@chromium.org
The patchset sent to the CQ was uploaded after l-g-t-m from thakis@chromium.org Link to the patchset: https://codereview.chromium.org/2875173002/#ps140001 (title: "note that we can move android hooks now")
CQ is trying da patch. Follow status at: https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...
The CQ bit was unchecked by commit-bot@chromium.org
Try jobs failed on following builders: win_clang on master.tryserver.chromium.win (JOB_FAILED, http://build.chromium.org/p/tryserver.chromium.win/builders/win_clang/builds/...)
The CQ bit was checked by scottmg@chromium.org
CQ is trying da patch. Follow status at: https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...
The CQ bit was unchecked by commit-bot@chromium.org
Try jobs failed on following builders: win_chromium_rel_ng on master.tryserver.chromium.win (JOB_FAILED, http://build.chromium.org/p/tryserver.chromium.win/builders/win_chromium_rel_...)
The CQ bit was checked by scottmg@chromium.org
CQ is trying da patch. Follow status at: https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...
CQ is committing da patch.
Bot data: {"patchset_id": 140001, "attempt_start_ts": 1494902814954850,
"parent_rev": "7e7ed9921f134e1e48aec31509e7195f7150deb8", "commit_rev":
"4cd62becec2902f99720eb0c5b4602eecfbd6ef1"}
Message was sent while issue was closed.
Description was changed from ========== Add hook to download Fuchsia SDK Used in (new) hooks_os section of DEPS to only be run when target_os includes 'fuchsia'. Writes the current version into third_party/fuchsia-sdk/.hash to avoid updating when no change has been made to the version. BUG=706592 ========== to ========== Add hook to download Fuchsia SDK Used in (new) hooks_os section of DEPS to only be run when target_os includes 'fuchsia'. Writes the current version into third_party/fuchsia-sdk/.hash to avoid updating when no change has been made to the version. BUG=706592 Review-Url: https://codereview.chromium.org/2875173002 Cr-Commit-Position: refs/heads/master@{#472008} Committed: https://chromium.googlesource.com/chromium/src/+/4cd62becec2902f99720eb0c5b46... ==========
Message was sent while issue was closed.
Committed patchset #7 (id:140001) as https://chromium.googlesource.com/chromium/src/+/4cd62becec2902f99720eb0c5b46... |
