Summary: This is just a quick pass to fix a few bugs and spacing issues, Phortune itself could probably use some more custom UI, but that'll require some thought and abstraction. This also adds a new taller table CSS, which I mayyyy make automatic on tables with few rows, we'll see.
Test Plan: Browsed my Phortune account, tested new spacing on `admin` for 'full effect'
Reviewers: epriestley, btrahan
Reviewed By: btrahan
Subscribers: Korvin, epriestley
Differential Revision: https://secure.phabricator.com/D12115
Summary: Fixes T7665. Unfortunately when the user messes this up its not easy to partially recover so we just reset that time to the default.
Test Plan: set time to "00:00 AM" and got a sensible error.
Reviewers: chad, epriestley
Reviewed By: epriestley
Subscribers: Korvin, epriestley
Maniphest Tasks: T7665
Differential Revision: https://secure.phabricator.com/D12174
Summary: Fixes T7655. We'll set tighter spacing around edit clusters. Also darkened up the date marker and remove unused `phabricator-transaction-view` CSS that was still scattered around the site.
Test Plan: Test a full and column multi-edit spam. Visited Ponder and Diffusion, noticed no issues using those apps. Grepped for other users of `phabricator-transaction-view`
Reviewers: epriestley, btrahan
Reviewed By: btrahan
Subscribers: Korvin, epriestley
Maniphest Tasks: T7655
Differential Revision: https://secure.phabricator.com/D12148
Summary:
Fixes T7629 plus an un filed bug that's breaking creating new threads since we need to add participants EVEN EARLIER than we were doing it now that policy is actually enforced.
Back to the main thrust of this, there is one UI corner case - in the main view if you go from 1:1 to 1:1:1 (i.e. add a 3rd recipient, or Nth in a row) the icon only updates on page reload. I figure this will get sorted out at a later refactor as we make the client better / share more code with durable column.
One other small behavioral oddity is in the main view sometime we start loading with no conpherence. in that case, rather than show some incorrect icon, we show no icon (and "no title") and then things change at load. Seems okay-ish.
Finally, @chad - the CSS is a very work-man-like "use the built in stuff you can specify from PHP" so I'm sure it needs some love.
Test Plan: made all sorts of rooms and threads and liked the icons. noted smooth loading action as i switched around
Reviewers: chad, epriestley
Reviewed By: epriestley
Subscribers: Korvin, chad, epriestley
Maniphest Tasks: T7629
Differential Revision: https://secure.phabricator.com/D12163
Summary:
Ref T6755. In Git and Subversion, running `git clone http://google.com/` or `svn checkout http://google.com/` does not echo the response body.
In Mercurial, it does. Censor it from the output of `hg pull` and `hg clone`. This prevents an attacker from:
- Creating a Mercurial remote repository with URI `http://10.0.0.1/secrets/`; and
- reading the secrets out of the error message after the clone fails.
Test Plan: Set a Mercurial remote URI to a non-Mercurial repository, ran `repository update`, saw censored error message.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T6755
Differential Revision: https://secure.phabricator.com/D12170
Summary: Ref T6755. I'll add some notes there about specifics.
Test Plan:
- Made connections to HTTP and HTTPS URIs.
- Added some debugging code to verify that HTTP URIs were pre-resolved.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T6755
Differential Revision: https://secure.phabricator.com/D12169
Summary:
Ref T6755. Although we do not return response bodies, it is possible to perform crude portscanning if you can execute a DNS rebinding attack (which, for now, remains theoretical).
Limit users to 60 requests / hour to make it less feasible. This would require ~30 years to portscan all ports on a `/32` netblock.
Users who can guess that services may exist can confirm their existence more quickly than this, but if the attacker already had a very small set of candidate services it seems unlikely that portscanning would be of much use in executing the attack.
This protection should eventually be applied to T4190, too (that task also has other considerations).
Test Plan: Set rate limit very low, hit rate limit.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T6755
Differential Revision: https://secure.phabricator.com/D12168
Summary:
Fixes T7602. This is similar to the existing behavior for "changes planned" and "needs revision" revisions.
Also fix the "Update Diff" workflow so it correctly selects closed revisions as attachable.
Test Plan: Updated an abandoned revision, saw it change to "Needs Review".
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T7602
Differential Revision: https://secure.phabricator.com/D12167
Summary:
Fixes T7664. When there are a large number of tasks (400+) with the same subpriority (which can happen if the subpriority features are rarely used), it may take more than 30 seconds to rebalance them.
Make the algorithm more aggressive about rebalancing homogenous blocks of tasks.
This may need to get even fancier, but I'd guess it can process blocks 1-2 orders of magnitude larger, which should be ~all installs.
(If someone still hits issues with this, I'll make it fancier.)
Once a block is rebalanced, it doesn't need to be rebalanced again (at least, not as a whole block) so we basically just need to get over the initial hurdle here and then we're good.
In the worst case, we can provide `bin/maniphest rebalance` or similar and do the rebalance step offline.
And, in any case, we have more test coverage here now.
Test Plan:
- Existing tests.
- New tests.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T7664
Differential Revision: https://secure.phabricator.com/D12166
Summary: Enables a basic tooltip when using icon buttons and a convenience method for setting an icon.
Test Plan: Built a UIExample.
Reviewers: btrahan, epriestley
Reviewed By: epriestley
Subscribers: Korvin, epriestley
Differential Revision: https://secure.phabricator.com/D12172
Summary: Ref T1460, this adds additional buttons colors and styles for use in inline comments. Will also backport to Calendar and PHUIInfoView
Test Plan:
Review new buttons and hover states in UI Examples.
{F350549}
{F350550}
Reviewers: btrahan, epriestley
Reviewed By: epriestley
Subscribers: Korvin, epriestley
Maniphest Tasks: T1460
Differential Revision: https://secure.phabricator.com/D12162
Summary: Fixes T7586. If you can't edit a room, the pertinent UI is greyed out. One exception is the title of the room in the full viewer; this crumb is not disabled as it would be hard to read. Otherwise though, everything is disabled nicely.
Test Plan: tried to add participants when I wasn't allowed to and got an error. added participants otherwise okay. tried to edit title when i wasn't allowed and got an error. otherwise okay. left conpherence threads / rooms successfully.
Reviewers: epriestley, chad
Reviewed By: epriestley
Subscribers: Korvin, epriestley
Maniphest Tasks: T7586
Differential Revision: https://secure.phabricator.com/D12161
Summary: Ref T7660. I'm not toggling "inline-state-is-draft" correctly in JS yet since it's a little tricky (you can reload to see it) but the main state should work.
Test Plan:
- Clicked "done", saw comment opacity fade with placeholder style.
Reviewers: chad
Reviewed By: chad
Subscribers: epriestley
Maniphest Tasks: T7660
Differential Revision: https://secure.phabricator.com/D12160
Summary: Fixes T5658. Over a long period of time, some cruft can build up here. Only show revisions which have been updated in the last 30 days.
Test Plan:
- Viewed panel in Differential and Diffusion.
- Changed limit from 30 days to 30 seconds and saw no revisions.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T5658
Differential Revision: https://secure.phabricator.com/D12158
Summary:
Ref T6755. This mitigates an attack where you:
- compromise an administrative account;
- configure "text/plain" as an "image" MIME type; and
- create a new macro sourced from a sensitive resource which is locally accessible over HTTP GET, using DNS rebinding.
You can then view the content of the resource in Files. By preventing the compromised account from reconfiguring the MIME types, the server will instead destroy the response and prevent the attacker from seeing it.
In general, these options should change very rarely, and they often sit just beyond the edge of security vulnerabilities anyway.
For example, if you ignore the warnings about an alternate file domain and elect to serve content from the primary domain, it's still somewhat difficult for an attacker to exploit the vulnerability. If they can add "text/html" or "image/svg+xml" as image MIME types, it becomes trivial. In this case not having an alternate domain is the main issue, but easy modification of this config increases risk/exposure.
Test Plan: Viewed affected config and saw that it is locked.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T6755
Differential Revision: https://secure.phabricator.com/D12154
Summary: Fixes T6378.
Test Plan: Set config to `/.*/`, created a new diff, everything was collapsed as generated.
Reviewers: btrahan, chad
Reviewed By: chad
Subscribers: epriestley
Maniphest Tasks: T6378
Differential Revision: https://secure.phabricator.com/D12159
Summary:
Fixes T7658. Currently, we remove the "undo" before placing the comment, but that causes us to lose track of which row we should be examining.
Instead, place the comment first, then remove the "undo".
Test Plan: This stuff is hard to test comprehensively, but the original report reproduced easily and is now fixed. I wasn't able to break anything by adding/editing/deleting comments.
Reviewers: btrahan, chad
Reviewed By: chad
Subscribers: epriestley
Maniphest Tasks: T7658
Differential Revision: https://secure.phabricator.com/D12157
Summary:
Ref T6755. This improves our resistance to SSRF attacks:
- Follow redirects manually and verify each component of the redirect chain.
- Handle authentication provider profile picture fetches more strictly.
Test Plan:
- Tried to download macros from various URIs which issued redirects, etc.
- Downloaded an actual macro.
- Went through external account workflow.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T6755
Differential Revision: https://secure.phabricator.com/D12151
Summary: This signature changed at some point after I tested things and I didn't catch it.
Test Plan: Destroyed a chunked large file with `bin/remove`.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Differential Revision: https://secure.phabricator.com/D12152
Summary:
Ref T7585. This implements everything specified, with a few caveats
- since rooms you have yet to join can't be viewed in the column yet, the column view has some bugs and isn't expected to work.
- the room you're looking at is just pre-pending to the top of the "recent" list
Test Plan: made a room that no one could join. verified when viewing that there was no comment ui. made a room that others could join. verified folks who had yet to join had a "join" button with an area for text. tried joining with / without message text and it worked in both cases
Reviewers: chad, epriestley
Reviewed By: epriestley
Subscribers: Korvin, epriestley
Maniphest Tasks: T7585
Differential Revision: https://secure.phabricator.com/D12149
Summary:
Ref T1266. We won't detect a move/copy if fewer than 3 lines are changed.
However, you may move a block like:
Complicated Line A
Trivial Line B
Complicated Line C
...where "Trivial Line B" is something like a curly brace. If you move this block somewhere that happened to previously have a similar trivial curly brace line, we won't be able to find 3 contiguous added lines in order to detect the copy/move.
Instead, consider both changed and unchanged lines when trying to find contiguous blocks. This allows us to detect across gaps where lines were not actually changed.
This new algorithm may be too liberal (for example, we may end up incorrectly identifying moved/copied code before or after changed lines, not just between changed lines), but we can keep an eye on it and tweak it. The algorithm is better factored and better covered, now.
Test Plan:
- Added a unit test for this case.
- Spot-checked a handful of diffs and generally saw behavior that made sense and looked better than before.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T1266
Differential Revision: https://secure.phabricator.com/D12146
Summary:
Ref T1266. This doesn't change any behaviors, but some of this code has a lot of really complicated conditionals and I tried to break that up a bit.
Also, reexpress this stuff in terms of the "structured" parser in D12144.
Test Plan: Unit tests still pass. They aren't hugely comprehensive but did reliably fail when I screwed stuff up.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T1266
Differential Revision: https://secure.phabricator.com/D12145
Summary:
Ref T1266. This prepares to fix case (2) on T1266 by improving the robustness of hunk parsing.
In particular, the copy detection code abuses this API because it isn't currently expressive or flexible enough.
Make it more flexible and cover it exhaustively.
I'll move callsites to the new stuff in upcoming revisions.
Test Plan: Unit tests.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T1266
Differential Revision: https://secure.phabricator.com/D12144
Summary: Fixes T7578. This was pretty easy because conpherence funnels all transacton stuff through this spot
Test Plan: made a new room so only my user was a participant. wrote "@myself will work and @anotherguy will be greyed out" and so it was as expected
Reviewers: chad, epriestley
Reviewed By: epriestley
Subscribers: Korvin, epriestley
Maniphest Tasks: T7578
Differential Revision: https://secure.phabricator.com/D12114
Summary: Ref T7584. This hits all the major bullets there. Next step on T7584 is figuring out how it integrates into the full UI and column UI. That said, this is a bit buggy feeling right now since Conpherence as is assumes you are a participant all over the place and rooms make no such assumption. I'll probably this bit up next.
Test Plan:
viewed /conpherence/room/ and saw stuff. viewed the "participant" query as two different users and saw different correct result sets. made a room via the button and it worked. tried to view a room I wasn't a participant in and it failed horribly, which is something to fix in a future diff
created a thread via "send message" on a user profile and it worked
Reviewers: chad, epriestley
Reviewed By: epriestley
Subscribers: Korvin, epriestley
Maniphest Tasks: T7584
Differential Revision: https://secure.phabricator.com/D12113
Summary: Fixes T7377. We don't expand projects into members when sending notifications right now. Instead, expand them.
Test Plan:
- Added a project as a reviewer to a revision, made a comment, saw project members receive a read notification + email (with appropriate preferences).
- There's meaningful test coverage on the core mail stuff.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T7377
Differential Revision: https://secure.phabricator.com/D12142
Summary:
Ref T5644. Ref T7472. Currently, we highlight each line of pattern search results in Diffusion.
- This is incredibly slow for non-PHP languages which need to shell out to Pygments.
- A lot of this highlighting isn't very useful anyway, because it doesn't have any context.
Instead, try to highlight pattern matches but don't highlight the source itself.
Test Plan: {F349637}
Reviewers: chad, btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T7472, T5644
Differential Revision: https://secure.phabricator.com/D12141
Summary:
Ref T5644. See some discussion in D8040.
When a file is very large (more than 64KB of text), don't activate syntax highlighting by default. This should prevent us from wasting resources running `pygmentize` on enormous files.
Users who want the file highlighted can still select "Highlight As...".
The tricky part of this diff is separating the headers into "changeset" headers and "undershield" (rendering) headers. Specifically, a file might have these headers/shields:
- "This file is newly added."
- "This file is generated. Show Changes"
- "Highlighting is disabled for this large file."
In this case, I want the user to see "added" and "generated" when they load the page, and only see "highlighting disabled" after they click "Show Changes". So there are several categories:
- "Changeset" headers, which discuss the changeset as a whole (binary file, image file, moved, added, deleted, etc.)
- "Property" headers, which describe metadata changes (not relevant here).
- "Shields", which hide files from view by default.
- "Undershield" headers, which provide rendering information that is only relevant if there is no shield on the file.
Test Plan:
- Viewed a diff with the library map, clicked "show changes", got a "highlighting disabled" header back with highlighting disabled.
- Enabled highlighting explicitly (this currently restores the shield, which it probably shouldn't, but that feels out of scope for this change). The deshielded file is highlighted per the user's request.
- Loaded context on normal files.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: joshuaspence, epriestley
Maniphest Tasks: T5644
Differential Revision: https://secure.phabricator.com/D12132
Summary:
Ref T1460. Ref T6403. Replace `Diffusion::INLINEDONE` with `Transactions::INLINESTATE` and generalize things enough that we can lift it into core.
The next change will lift Differential's similar implementation into the core.
Also start implementing a fix for T6403, providing an alternate hook for optional builtin transactions.
Test Plan: Changed inline state in Diffusion.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T6403, T1460
Differential Revision: https://secure.phabricator.com/D12129
Summary:
Ref T1460. See D12126. This is essentially the same change, but for Diffusion.
This is a bit copy/pastey. I'm going to make an effort to lift inline handling into the core before pushing this in, so hopefully that will clean things up a bit.
Test Plan: Submitted stuff in Diffusion and got checkmarks to publish.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T1460
Differential Revision: https://secure.phabricator.com/D12128
Summary:
Ref T1460. When a revision author updates/comments/etc on a revision, publish all their checkmarks.
This doesn't handle Diffusion/audits yet.
Test Plan: {F346870}
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: yelirekim, epriestley
Maniphest Tasks: T1460
Differential Revision: https://secure.phabricator.com/D12126
Summary:
Ref T1460. This just barely works, but throwing it up in case any of it sounds mechanically crazy before we build integrations/UI/etc.
Specifically, these are the behaviors:
- You can mark your own draft comments as "done" before you submit them. The intent is to let reviewers mark their stuff advisory/minor/not-important before they submit it, to hint to authors that they don't expect the feedback to necessarily be addressed (maybe it's a joke, maybe it's just discussion, maybe it's "consider..").
- You can mark others' published comments as "done" if you're the revision/commit author. The intent is to keep this lightweight by not requiring an audit trail of who marked what done when. If anyone could mark anything done, we'd have to have some way to show who marked stuff.
- When you mark stuff done (or unmark it), it goes into a "draft" state, where you see the change but others don't see it yet. The intent is twofold:
- Be consistent with how inlines work.
- Allow us to publish a "epriestley updated this revision + epriestley marked 15 inlines as done" story later if we want. This seems more useful than publishing 15 "epriestley marked one thing as done" stories.
- The actual bit where done-ness publishes isn't implemented.
- UI is bare bones.
- No integration with the rest of the UI yet.
Test Plan: Clicked some checkboxes.
Reviewers: chad, btrahan
Reviewed By: btrahan
Subscribers: paulshen, chasemp, epriestley
Maniphest Tasks: T1460
Differential Revision: https://secure.phabricator.com/D12033
Summary:
Fixes T1102. If you don't use `arc`, the web workflow requires some extra needless steps when updating diffs.
Provide a more streamlined "Update Diff" workflow.
Test Plan: {F347750}
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T1102
Differential Revision: https://secure.phabricator.com/D12131
Summary:
Ref T6755. This is a partial fix, but:
- Allow netblocks to be blacklisted instead of making the feature all-or-nothing.
- Default to disallow requests to all reserved private/local/special IP blocks. This should generally be a "safe" setting.
- Explain the risks better.
- Improve the errors rasied by Macro when failing.
- Removed `security.allow-outbound-http`, as it is superseded by this setting and is somewhat misleading.
- We still make outbound HTTP requests to OAuth.
- We still make outbound HTTP requests for repositories.
From a technical perspective:
- Separate URIs that are safe to link to or redirect to (basically, not "javascript://") from URIs that are safe to fetch (nothing in a private block).
- Add the default blacklist.
- Be more careful with response data in Macro fetching, and don't let the user see it if it isn't ultimately valid.
Additionally:
- I want to do this check before pulling repositories, but that's enough of a mess that it should go in a separate diff.
- The future implementation of T4190 needs to perform the fetch check.
Test Plan:
- Fetched a valid macro.
- Fetched a non-image, verified it didn't result in a viewable file.
- Fetched a private-ip-space image, got an error.
- Fetched a 404, got a useful-enough error without additional revealing response content (which is usually HTML anyway and not useful).
- Fetched a bad protocol, got an error.
- Linked to a local resource, a phriction page, a valid remote site, all worked.
- Linked to private IP space, which worked fine (we want to let you link and redierect to other private services, just not fetch them).
- Added and executed unit tests.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T6755
Differential Revision: https://secure.phabricator.com/D12136
Summary: Fixes T7609. When we moved the sidenav background to z-index -1, the footer then reappeared. Make the hiding more specific in CSS.
Test Plan: test mobile layout of homepage, check footer still exists on all other pages.
Reviewers: btrahan, epriestley
Reviewed By: epriestley
Subscribers: Korvin, epriestley
Maniphest Tasks: T7609
Differential Revision: https://secure.phabricator.com/D12138
Summary: Ref T7627. This centralizes this transaction construction code so the unit tests and Instances can both use it.
Test Plan: See D12116.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T7627
Differential Revision: https://secure.phabricator.com/D12118
Summary: Fixes T7625. The way the regexp worked, "unusual" terminal characters required at least one character as a prefix in order to match. Allow 0 instead, so `#a1` matches.
Test Plan: Added and executed unit test.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T7625
Differential Revision: https://secure.phabricator.com/D12123
Summary: Ref T7611. This should let us figure out the root cause, hopefully.
Test Plan: iiam
Reviewers: joshuaspence, btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T7611
Differential Revision: https://secure.phabricator.com/D12124
Summary:
Fixes T7485. Before applications had proper policies, we gated access by requiring the viewer be an administrator.
This is now redundant (CAN_EDIT on applications has the same effect, and performs the same check), and may some day be wrong (we might let administrators configure a different policy to control who can configure applications). Today, it gets the policy dialog wrong.
Test Plan:
Clicked "Edit Policies" as a non-administrator, was unable to, got nice error:
{F346598}
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T7485
Differential Revision: https://secure.phabricator.com/D12125
Summary: Fixes T7620. I don't fully understand exactly what's going on here, but we don't actually need to call `flush()`.
Test Plan:
- Put timing code around the `echo`.
- Made a fake page that emitted a lot of data.
- Saw the `echo` block proportionate to data size under `curl --limit-rate ...`.
- See T7620.
- Downloaded a large file, got a reasonable progress bar and no obvious memory use issues.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: jlarouche, rbalik, epriestley
Maniphest Tasks: T7620
Differential Revision: https://secure.phabricator.com/D12127
Summary: Fixes T7647. We float both the time and description here and want the fluidity.
Test Plan: Review a new Conpherence in FF, Safari, Chrome.
Reviewers: btrahan, epriestley
Reviewed By: epriestley
Subscribers: Korvin, epriestley
Maniphest Tasks: T7647
Differential Revision: https://secure.phabricator.com/D12137
Summary:
Fixes T7563. Fixes T5201. Reframe this as two separate operations:
- Move before or after a task.
- Move to the beginning or end of a priority.
Then:
- Make all the order queries unambiguous and properly reversible, with an explicit `id` order.
- Just reuse `ManiphestTask` to get results in the correct order.
- Simplify the actual transaction apply logic.
- Detect and recover from cases where tasks have identical or similar subpriorities.
Test Plan:
- Wrote and executed unit tests.
- Dragged and dropped tasks within priorities and between priorities in the main Maniphest view.
- Dragged and dropped tasks within priorities in the workboard view, when ordered by priority.
- Also poked at the "natural" order, but that shouldn't be affected.
Reviewers: btrahan, chad
Reviewed By: chad
Subscribers: chad, epriestley
Maniphest Tasks: T5201, T7563
Differential Revision: https://secure.phabricator.com/D12121
Summary:
Via HackerOne. We aren't correctly escaping the date, so a user can XSS themselves by setting their date format creatively.
This construction is very unusual and I don't think we do anything similar elsewhere, so I can't come up with a systematic change which would prevent this in the general case.
Test Plan: Set date format to tag junk, got self-XSS before patch and proper escaping after the patch.
Reviewers: btrahan, chad
Reviewed By: chad
Subscribers: epriestley
Differential Revision: https://secure.phabricator.com/D12117
Summary:
Fixes T7621. The engine selection code started out making sense, but didn't make as much sense by the time I was done with it.
Specifically, from the vanilla file upload, we may incorrectly try to write directly to the chunk storage engine. This is incorrect, and produces a confusing/bad error.
Make chunk storage engines explicit and don't try to do single-file one-shot writes to them.
Test Plan:
- Tried to upload a large file with vanilla uploader, got better error message.
- Uploaded small and large files with drag and drop.
- Viewed {nav Files > Help/Options}.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T7621
Differential Revision: https://secure.phabricator.com/D12110
