|
|
Chromium Code Reviews
DescriptionWhitelist fileSystemProvider API to Secure Shell app
Secure Shell is a legacy packaged app, and as such cannot declare
fileSystemProvider permission. Secure Shell will require access to
fileSystemProvider API - to enable Secure Shell to access the API,
expose fileSystemProvider permission to the app via whitelist.
BUG=673004
Review-Url: https://codereview.chromium.org/2613823002
Cr-Commit-Position: refs/heads/master@{#444498}
Committed: https://chromium.googlesource.com/chromium/src/+/635e633fc4db2ac47139a493fa36646735627464
Patch Set 1 #Patch Set 2 : manifest feature whitelis #
Messages
Total messages: 26 (13 generated)
The CQ bit was checked by tbarzic@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...
tbarzic@chromium.org changed reviewers: + rdevlin.cronin@chromium.org, rginda@chromium.org
rginda: fyi
The CQ bit was unchecked by commit-bot@chromium.org
Dry run: This issue passed the CQ dry run.
lgtm Thanks!
Hi tbarzic, Does _manifest_features.json require changing as well? Or is it intelligent enough to carry over? (i.e. if the app has permission to use the API, it has the manifest permissions as well) Reference: https://chromium.googlesource.com/chromium/src/+/master/chrome/common/extensi...
On 2017/01/05 23:33:52, mcdermottm wrote: > Hi tbarzic, > > Does _manifest_features.json require changing as well? Or is it intelligent > enough to carry over? (i.e. if the app has permission to use the API, it has the > manifest permissions as well) > > Reference: > https://chromium.googlesource.com/chromium/src/+/master/chrome/common/extensi... oh, I missed that one - I think I have to update both
The CQ bit was checked by tbarzic@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...
The CQ bit was unchecked by commit-bot@chromium.org
Dry run: This issue passed the CQ dry run.
On 2017/01/05 23:54:20, tbarzic wrote: > On 2017/01/05 23:33:52, mcdermottm wrote: > > Hi tbarzic, > > > > Does _manifest_features.json require changing as well? Or is it intelligent > > enough to carry over? (i.e. if the app has permission to use the API, it has > the > > manifest permissions as well) > > > > Reference: > > > https://chromium.googlesource.com/chromium/src/+/master/chrome/common/extensi... > > oh, I missed that one - I think I have to update both Awesome-- thank you! (I am the intern requesting this whitelisting)
legacy chrome apps have been deprecated for years. It makes me a little sad to see us adding yet-more-functionality to this rather than transitioning away... can you outline on the bug a bit more of the reasons that are difficult to address?
On 2017/01/10 16:00:04, Devlin (US OOO on Jan 16) wrote: > legacy chrome apps have been deprecated for years. It makes me a little sad to > see us adding yet-more-functionality to this rather than transitioning away... > can you outline on the bug a bit more of the reasons that are difficult to > address? We'd love to get off of the deprecated platform, and put in some considerable effort two years ago. In the end we stopped because neither of the options meets the needs. Extensions: 1. Aren't allowed to capture Ctrl-N/T/W. 2. Will likely never have access to raw sockets. 3. Have no equivalent of "open in window". 4. Have no app icon in the ChromeOS launcher. Of these, #1 is the deal breaker (except on Mac, where Cmd-N/T/W are distinct). #2 we could work around with the same whitelist we have today. 3 and 4 would cause problems for some users. V2 Apps: 1. Inability to load window content from the web breaks our corp OAuth to the ssh relays. 2. Inability to send postMessage to extensions breaks our integration with gnubby. 3. Deprecated everywhere except Chrome OS. These are all deal breakers for corp users. The first issue we might be able to work around with a rewrite of portions of Secure Shell and the relay servers. The second is hard to imagine a fix for, other than a change to Chrome's policy on app->extension messaging. The third I think is a deal breaker. I've met with the Owen from the web platform team about this in the past. We're hoping that the problems with extensions can be addressed by the web platform over time and Secure Shell can move into extension or open web space at that time.
On 2017/01/17 18:11:29, rginda wrote: > On 2017/01/10 16:00:04, Devlin (US OOO on Jan 16) wrote: > > legacy chrome apps have been deprecated for years. It makes me a little sad > to > > see us adding yet-more-functionality to this rather than transitioning away... > > can you outline on the bug a bit more of the reasons that are difficult to > > address? > > We'd love to get off of the deprecated platform, and put in some considerable > effort two years ago. In the end we stopped because neither of the options > meets the needs. > > Extensions: > 1. Aren't allowed to capture Ctrl-N/T/W. > 2. Will likely never have access to raw sockets. > 3. Have no equivalent of "open in window". > 4. Have no app icon in the ChromeOS launcher. > > Of these, #1 is the deal breaker (except on Mac, where Cmd-N/T/W are distinct). > #2 we could work around with the same whitelist we have today. 3 and 4 would > cause problems for some users. Extensions can capture Ctrl+N/T/W using the commands API, can't they?
On 2017/01/17 20:57:49, Devlin (possibly slow) wrote: > On 2017/01/17 18:11:29, rginda wrote: > > On 2017/01/10 16:00:04, Devlin (US OOO on Jan 16) wrote: > > > legacy chrome apps have been deprecated for years. It makes me a little sad > > to > > > see us adding yet-more-functionality to this rather than transitioning > away... > > > can you outline on the bug a bit more of the reasons that are difficult to > > > address? > > > > We'd love to get off of the deprecated platform, and put in some considerable > > effort two years ago. In the end we stopped because neither of the options > > meets the needs. > > > > Extensions: > > 1. Aren't allowed to capture Ctrl-N/T/W. > > 2. Will likely never have access to raw sockets. > > 3. Have no equivalent of "open in window". > > 4. Have no app icon in the ChromeOS launcher. > > > > Of these, #1 is the deal breaker (except on Mac, where Cmd-N/T/W are > distinct). > > #2 we could work around with the same whitelist we have today. 3 and 4 would > > cause problems for some users. > > Extensions can capture Ctrl+N/T/W using the commands API, can't they? I'm not sure if that API allows an extension to override those keys. Bue even if it does, that API adds commands to chrome itself. We want to capture them only in the scope of an active Secure Shell session. We'd essentially have to override Ctrl-N/T/W for all of chrome, and then detect which, if any Secure Shell window were active and route the command to it. If not, somehow defer processing of the command to chrome.
On 2017/01/17 20:57:49, Devlin (possibly slow) wrote: > On 2017/01/17 18:11:29, rginda wrote: > > On 2017/01/10 16:00:04, Devlin (US OOO on Jan 16) wrote: > > > legacy chrome apps have been deprecated for years. It makes me a little sad > > to > > > see us adding yet-more-functionality to this rather than transitioning > away... > > > can you outline on the bug a bit more of the reasons that are difficult to > > > address? > > > > We'd love to get off of the deprecated platform, and put in some considerable > > effort two years ago. In the end we stopped because neither of the options > > meets the needs. > > > > Extensions: > > 1. Aren't allowed to capture Ctrl-N/T/W. > > 2. Will likely never have access to raw sockets. > > 3. Have no equivalent of "open in window". > > 4. Have no app icon in the ChromeOS launcher. > > > > Of these, #1 is the deal breaker (except on Mac, where Cmd-N/T/W are > distinct). > > #2 we could work around with the same whitelist we have today. 3 and 4 would > > cause problems for some users. > > Extensions can capture Ctrl+N/T/W using the commands API, can't they? I'm not sure if that API allows an extension to override those keys. Bue even if it does, that API adds commands to chrome itself. We want to capture them only in the scope of an active Secure Shell session. We'd essentially have to override Ctrl-N/T/W for all of chrome, and then detect which, if any Secure Shell window were active and route the command to it. If not, somehow defer processing of the command to chrome.
I think this probably warrants more discussion, because I think we can probably solve some of these cases. But we don't need to block this cl on that discussion. :) lgtm
The CQ bit was checked by tbarzic@chromium.org
The patchset sent to the CQ was uploaded after l-g-t-m from rginda@chromium.org Link to the patchset: https://codereview.chromium.org/2613823002/#ps20001 (title: "manifest feature whitelis")
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": 20001, "attempt_start_ts": 1484773651275870,
"parent_rev": "0bc260f9e8b1531e8f02c0b88f918f17e9193bfc", "commit_rev":
"635e633fc4db2ac47139a493fa36646735627464"}
Message was sent while issue was closed.
Description was changed from ========== Whitelist fileSystemProvider API to Secure Shell app Secure Shell is a legacy packaged app, and as such cannot declare fileSystemProvider permission. Secure Shell will require access to fileSystemProvider API - to enable Secure Shell to access the API, expose fileSystemProvider permission to the app via whitelist. BUG=673004 ========== to ========== Whitelist fileSystemProvider API to Secure Shell app Secure Shell is a legacy packaged app, and as such cannot declare fileSystemProvider permission. Secure Shell will require access to fileSystemProvider API - to enable Secure Shell to access the API, expose fileSystemProvider permission to the app via whitelist. BUG=673004 Review-Url: https://codereview.chromium.org/2613823002 Cr-Commit-Position: refs/heads/master@{#444498} Committed: https://chromium.googlesource.com/chromium/src/+/635e633fc4db2ac47139a493fa36... ==========
Message was sent while issue was closed.
Committed patchset #2 (id:20001) as https://chromium.googlesource.com/chromium/src/+/635e633fc4db2ac47139a493fa36... |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
