Page MenuHomePhabricator
Feed All Stories

Wed, Sep 25

epriestley added a revision to T13395: Strip libphutil for parts: D20838: Add "PhutilProseDiff" classes to "phabricator/".
Wed, Sep 25, 9:40 PM · libphutil
epriestley requested review of D20837: Move PhutilProseDiff out of "libphutil/".
Wed, Sep 25, 9:37 PM
epriestley added a revision to T13395: Strip libphutil for parts: D20837: Move PhutilProseDiff out of "libphutil/".
Wed, Sep 25, 9:36 PM · libphutil
epriestley added a revision to T13414: Raise the effective corpus size limit in "PhutilProseDifferenceEngine" by using "diff" for coarse passes: D20837: Move PhutilProseDiff out of "libphutil/".
Wed, Sep 25, 9:36 PM · Infrastructure
epriestley added a revision to T13425: Support diffs between abstract block lists in the UI: D20837: Move PhutilProseDiff out of "libphutil/".
Wed, Sep 25, 9:36 PM · Differential
epriestley added a comment to T13425: Support diffs between abstract block lists in the UI.

Wed, Sep 25, 9:18 PM · Differential
epriestley requested review of D20836: Use a hash-and-diff strategy to produce a diff layout for block-based documents.
Wed, Sep 25, 9:07 PM
epriestley added a revision to T13414: Raise the effective corpus size limit in "PhutilProseDifferenceEngine" by using "diff" for coarse passes: D20836: Use a hash-and-diff strategy to produce a diff layout for block-based documents.
Wed, Sep 25, 9:05 PM · Infrastructure
epriestley added a revision to T13425: Support diffs between abstract block lists in the UI: D20836: Use a hash-and-diff strategy to produce a diff layout for block-based documents.
Wed, Sep 25, 9:05 PM · Differential
epriestley requested review of D20835: Improve behavior of inline comment highlight reticle for block diffs.
Wed, Sep 25, 8:27 PM
epriestley added a revision to T13425: Support diffs between abstract block lists in the UI: D20835: Improve behavior of inline comment highlight reticle for block diffs.
Wed, Sep 25, 8:25 PM · Differential
epriestley added a comment to T13425: Support diffs between abstract block lists in the UI.

??? it just works ???

Wed, Sep 25, 7:50 PM · Differential
epriestley requested review of D20834: Correct DOM node metadata to let inline comments work against block-based diffs.
Wed, Sep 25, 7:49 PM
epriestley added a revision to T13425: Support diffs between abstract block lists in the UI: D20834: Correct DOM node metadata to let inline comments work against block-based diffs.
Wed, Sep 25, 7:48 PM · Differential
epriestley requested review of D20833: Allow the Jupyter engine to elect to emit diffs, and emit Jupyter documents as blocks.
Wed, Sep 25, 7:06 PM
epriestley added a revision to T13425: Support diffs between abstract block lists in the UI: D20833: Allow the Jupyter engine to elect to emit diffs, and emit Jupyter documents as blocks.
Wed, Sep 25, 7:04 PM · Differential
epriestley added a comment to T13425: Support diffs between abstract block lists in the UI.

Jupyter as blocks, no diffing or inlines yet:

Wed, Sep 25, 7:03 PM · Differential
epriestley added a comment to T13425: Support diffs between abstract block lists in the UI.
  • Differential shows a "this file is big, so syntax highlighting is disabled by default" warning even when a document engine which does not use syntax highlighting renders the document.
Wed, Sep 25, 6:46 PM · Differential
epriestley added a revision to T13425: Support diffs between abstract block lists in the UI: D20832: Add a "View as..." option to diff dropdowns for selecting between document engines.
Wed, Sep 25, 6:21 PM · Differential
epriestley updated the summary of D20832: Add a "View as..." option to diff dropdowns for selecting between document engines.
Wed, Sep 25, 6:21 PM
epriestley requested review of D20832: Add a "View as..." option to diff dropdowns for selecting between document engines.
Wed, Sep 25, 5:51 PM
epriestley added a comment to T13425: Support diffs between abstract block lists in the UI.
  • There's also no "render as native source" option, but there is a "View as Source" option, which doesn't work. Gotcha!
Wed, Sep 25, 5:47 PM · Differential
epriestley added a comment to T13425: Support diffs between abstract block lists in the UI.
  • The options in the "View As..." dropdown are exhaustive, and most do not work, because they aren't based on the changeset being rendered (so we'll give you an option to render a Jupyter notebook as audio, for example). This isn't trivial to fix and it isn't terribly important for this to function as an escape hatch back to old behavior.
  • Since we expect most documents to have a relatively small number of options here, a list of clickable options might be better than a <select /> dropdown.
Wed, Sep 25, 5:46 PM · Differential
epriestley added a comment to T13425: Support diffs between abstract block lists in the UI.

In an effort to "do no harm", I'm planning to add a "Render with Document Engine..." option to the View Options dropdown next. This will let you (for example) view a Jupyter notebook as a raw source diff if you want, if the "fancy" diff is broken or unhelpful for some reason, so you always have an escape hatch back to a lower level representation.

Wed, Sep 25, 5:23 PM · Differential
epriestley requested review of D20831: Render image diffs as abstract blocks diffs via DocumentEngine.
Wed, Sep 25, 5:16 PM
epriestley added a revision to T13425: Support diffs between abstract block lists in the UI: D20831: Render image diffs as abstract blocks diffs via DocumentEngine.
Wed, Sep 25, 5:15 PM · Differential
epriestley added a comment to T13425: Support diffs between abstract block lists in the UI.
  • TwoUp image comments aren't triggering (also in master).
  • OneUp image comments need some cell span adjustments.
Wed, Sep 25, 5:10 PM · Differential
epriestley requested review of D20830: Allow DocumentEngine to elect into diff construction.
Wed, Sep 25, 4:28 PM
epriestley added a revision to T13425: Support diffs between abstract block lists in the UI: D20830: Allow DocumentEngine to elect into diff construction.
Wed, Sep 25, 4:26 PM · Differential
epriestley triaged T13425: Support diffs between abstract block lists in the UI as Normal priority.
Wed, Sep 25, 4:19 PM · Differential
epriestley added a comment to T13299: Herald only triggers for commits once, during the initial publish phase.

I've moved the issue in PHI1264 (roughly, "commenting on an unpublished commit should still work normally") here from T13284, since it ended up not really fitting into that scope particularly well.

Wed, Sep 25, 4:01 PM · Diffusion, Herald
epriestley updated the task description for T13299: Herald only triggers for commits once, during the initial publish phase.
Wed, Sep 25, 4:00 PM · Diffusion, Herald
epriestley updated the task description for T13284: When a previously non-permanent ref is marked permanent, unpublished commits reachable from that ref are not published.
Wed, Sep 25, 4:00 PM · Diffusion
epriestley committed rP06edcf270921: Fix an issue where ancestors of permanent refs might not be published during… (authored by epriestley).
Fix an issue where ancestors of permanent refs might not be published during…
Wed, Sep 25, 3:55 PM
epriestley closed D20829: Fix an issue where ancestors of permanent refs might not be published during import or if a branch is later made permanent.
Wed, Sep 25, 3:55 PM
epriestley closed T13284: When a previously non-permanent ref is marked permanent, unpublished commits reachable from that ref are not published as Resolved by committing rP06edcf270921: Fix an issue where ancestors of permanent refs might not be published during….
Wed, Sep 25, 3:55 PM · Diffusion
epriestley added a comment to T13284: When a previously non-permanent ref is marked permanent, unpublished commits reachable from that ref are not published.

This fix is not retroactive (and, I think, isn't easy to make retroactive in an obviously safe way) so you'll need to manually repair affected repositories if you want to fix existing commits affected by this bug. This bug doesn't likely doesn't have any far-reaching implications so there's no need to go on a grand adventure to hunt these down, but if you've run into some you can fix them like this:

Wed, Sep 25, 3:53 PM · Diffusion
epriestley requested review of D20829: Fix an issue where ancestors of permanent refs might not be published during import or if a branch is later made permanent.
Wed, Sep 25, 3:49 PM
epriestley added a revision to T13284: When a previously non-permanent ref is marked permanent, unpublished commits reachable from that ref are not published: D20829: Fix an issue where ancestors of permanent refs might not be published during import or if a branch is later made permanent.
Wed, Sep 25, 3:48 PM · Diffusion
epriestley added a comment to T13284: When a previously non-permanent ref is marked permanent, unpublished commits reachable from that ref are not published.

Add a bin/repository refs --rebuild or similar flag to repair repositories affected by this bug.

Wed, Sep 25, 3:43 PM · Diffusion
epriestley added a comment to T13284: When a previously non-permanent ref is marked permanent, unpublished commits reachable from that ref are not published.

If this theory holds up, the "swap a branch" case can be reproduced like this:

Wed, Sep 25, 3:39 PM · Diffusion
epriestley added a comment to T13284: When a previously non-permanent ref is marked permanent, unpublished commits reachable from that ref are not published.

Retrying this on the patch:

Wed, Sep 25, 3:19 PM · Diffusion
epriestley added a comment to T13424: Support builds with TeamCity in Harbormaster.

The problem isn't that it's in rough shape (I'm fine with bringing rough stuff upstream), but that it's something I may eventually want to license as a paid extension. I generally want to stop bringing "free glue for paid systems" upstream (T13229).

Wed, Sep 25, 2:22 PM · Harbormaster
swisspol added a comment to T13424: Support builds with TeamCity in Harbormaster.

How about landing this as a prototype?

Wed, Sep 25, 4:46 AM · Harbormaster
epriestley added a comment to T13284: When a previously non-permanent ref is marked permanent, unpublished commits reachable from that ref are not published.

The "import a repo" case can be reproduced like this:

Wed, Sep 25, 1:50 AM · Diffusion

Tue, Sep 24

epriestley added a comment to T13284: When a previously non-permanent ref is marked permanent, unpublished commits reachable from that ref are not published.

If this theory holds up, the "swap a branch" case can be reproduced like this:

Tue, Sep 24, 11:57 PM · Diffusion
epriestley added a comment to T13284: When a previously non-permanent ref is marked permanent, unpublished commits reachable from that ref are not published.

If this theory holds up, the "swap a branch" case can be reproduced like this:

Tue, Sep 24, 11:17 PM · Diffusion
epriestley added a comment to T13284: When a previously non-permanent ref is marked permanent, unpublished commits reachable from that ref are not published.

The algorithm in PhabricatorRepositoryRefEngine->updateRefs() is approximately:

Tue, Sep 24, 10:49 PM · Diffusion
epriestley added a comment to T13284: When a previously non-permanent ref is marked permanent, unpublished commits reachable from that ref are not published.

See PHI1447. In this case, the install reports a sequence of events roughly like this:

Tue, Sep 24, 10:09 PM · Diffusion
epriestley lowered the priority of T13413: Chrome 77 rendering engine freezes when elements are nested with particular word-wrap / white-space CSS behavior from Normal to Wishlist.

This now appears to be fixed in the release version of Chrome. We can remove the workaround in some future change.

Tue, Sep 24, 6:10 PM · Google Chrome
epriestley committed rP6af776f84a66: Allow installs to provide "Request a Username Change" instructions (authored by epriestley).
Allow installs to provide "Request a Username Change" instructions
Tue, Sep 24, 6:09 PM
epriestley closed D20828: Allow installs to provide "Request a Username Change" instructions.
Tue, Sep 24, 6:09 PM
epriestley closed T13420: Update "Change Username" to make it more friendly for non-administrators as Resolved by committing rP6af776f84a66: Allow installs to provide "Request a Username Change" instructions.
Tue, Sep 24, 6:09 PM · People, Auth
epriestley committed rP0c331458a88b: When non-administrators click "Change Username", explain why they can't continue (authored by epriestley).
When non-administrators click "Change Username", explain why they can't continue
Tue, Sep 24, 6:09 PM
epriestley closed D20827: When non-administrators click "Change Username", explain why they can't continue.
Tue, Sep 24, 6:08 PM
epriestley committed rP058b7ae9450d: Update "Change Username" instructions to be less foreboding (authored by epriestley).
Update "Change Username" instructions to be less foreboding
Tue, Sep 24, 6:08 PM
epriestley closed D20826: Update "Change Username" instructions to be less foreboding.
Tue, Sep 24, 6:08 PM
epriestley committed rPb1d4d5c00ce1: Add an "{anchor #xyz}" rule to Remarkup (authored by epriestley).
Add an "{anchor #xyz}" rule to Remarkup
Tue, Sep 24, 6:04 PM
epriestley closed T4280: Embed anchor in remark up text and later link to it. as Resolved by committing rPb1d4d5c00ce1: Add an "{anchor #xyz}" rule to Remarkup.
Tue, Sep 24, 6:04 PM · Remarkup
epriestley closed T4280: Embed anchor in remark up text and later link to it. , a subtask of T13410: Update Phriction UI structural behaviors, as Resolved.
Tue, Sep 24, 6:04 PM · Phriction
epriestley closed D20825: Add an "{anchor #xyz}" rule to Remarkup.
Tue, Sep 24, 6:04 PM
epriestley committed rPbff72ce3b582: Generate more friendly anchor names for header sections in Remarkup (authored by epriestley).
Generate more friendly anchor names for header sections in Remarkup
Tue, Sep 24, 6:03 PM
epriestley closed D20821: Generate more friendly anchor names for header sections in Remarkup.
Tue, Sep 24, 6:03 PM
epriestley committed rP74d6bcbdceb9: Allow a user to target "#anchor" by navigating to any prefix (authored by epriestley).
Allow a user to target "#anchor" by navigating to any prefix
Tue, Sep 24, 5:56 PM
epriestley closed D20820: Allow a user to target "#anchor" by navigating to any prefix.
Tue, Sep 24, 5:56 PM
epriestley requested review of D20828: Allow installs to provide "Request a Username Change" instructions.
Tue, Sep 24, 5:53 PM
epriestley added a revision to T13420: Update "Change Username" to make it more friendly for non-administrators: D20828: Allow installs to provide "Request a Username Change" instructions.
Tue, Sep 24, 5:51 PM · People, Auth
epriestley updated the test plan for D20827: When non-administrators click "Change Username", explain why they can't continue.
Tue, Sep 24, 5:49 PM
epriestley requested review of D20827: When non-administrators click "Change Username", explain why they can't continue.
Tue, Sep 24, 5:43 PM
epriestley added a revision to T13420: Update "Change Username" to make it more friendly for non-administrators: D20827: When non-administrators click "Change Username", explain why they can't continue.
Tue, Sep 24, 5:41 PM · People, Auth
epriestley requested review of D20826: Update "Change Username" instructions to be less foreboding.
Tue, Sep 24, 5:38 PM
epriestley added a revision to T13420: Update "Change Username" to make it more friendly for non-administrators: D20826: Update "Change Username" instructions to be less foreboding.
Tue, Sep 24, 5:36 PM · People, Auth
epriestley requested review of D20825: Add an "{anchor #xyz}" rule to Remarkup.
Tue, Sep 24, 5:27 PM
epriestley added a revision to T4280: Embed anchor in remark up text and later link to it. : D20825: Add an "{anchor #xyz}" rule to Remarkup.
Tue, Sep 24, 5:26 PM · Remarkup
epriestley added a revision to T13410: Update Phriction UI structural behaviors: D20825: Add an "{anchor #xyz}" rule to Remarkup.
Tue, Sep 24, 5:26 PM · Phriction
epriestley committed rPHU3b15fd0a01d4: Ignore all files in "src/extensions/", not just immediate children (authored by epriestley).
Ignore all files in "src/extensions/", not just immediate children
Tue, Sep 24, 5:02 PM
epriestley closed D20824: Ignore all files in "src/extensions/", not just immediate children.
Tue, Sep 24, 5:02 PM
epriestley closed T13418: "arc liberate" incorrectly includes subdirectories of "src/extensions/" in library map as Resolved by committing rPHU3b15fd0a01d4: Ignore all files in "src/extensions/", not just immediate children.
Tue, Sep 24, 5:02 PM · libphutil
epriestley requested review of D20824: Ignore all files in "src/extensions/", not just immediate children.
Tue, Sep 24, 4:41 PM
epriestley added a revision to T13418: "arc liberate" incorrectly includes subdirectories of "src/extensions/" in library map: D20824: Ignore all files in "src/extensions/", not just immediate children.
Tue, Sep 24, 4:40 PM · libphutil
epriestley added a comment to T13418: "arc liberate" incorrectly includes subdirectories of "src/extensions/" in library map.

Oh, no, that's unrelated:

Tue, Sep 24, 4:37 PM · libphutil
epriestley added a comment to T13418: "arc liberate" incorrectly includes subdirectories of "src/extensions/" in library map.

Looking at this in slightly more detail, I think the immediate issue is just that .gitignore specifies /src/extensions/*, not /src/extensions/**.

Tue, Sep 24, 4:36 PM · libphutil
jmeador added a comment to T13418: "arc liberate" incorrectly includes subdirectories of "src/extensions/" in library map.

I actually tried to use the extensions directory to hide Composer's vendored files and ran into this same issue.

Tue, Sep 24, 4:33 PM · libphutil
epriestley requested review of D20821: Generate more friendly anchor names for header sections in Remarkup.
Tue, Sep 24, 4:09 PM
epriestley added a project to T13423: Conduit log paging is broken, fails with `Call to undefined method PhabricatorConduitLogQuery::withIDs` error: Conduit.
Tue, Sep 24, 3:37 PM · Conduit
epriestley committed rPbe2a0f873368: Fix a cursor paging issue in Conduit call logs (authored by epriestley).
Fix a cursor paging issue in Conduit call logs
Tue, Sep 24, 3:37 PM
epriestley closed D20823: Fix a cursor paging issue in Conduit call logs.
Tue, Sep 24, 3:36 PM
epriestley closed T13423: Conduit log paging is broken, fails with `Call to undefined method PhabricatorConduitLogQuery::withIDs` error as Resolved by committing rPbe2a0f873368: Fix a cursor paging issue in Conduit call logs.
Tue, Sep 24, 3:36 PM · Conduit
epriestley triaged T13423: Conduit log paging is broken, fails with `Call to undefined method PhabricatorConduitLogQuery::withIDs` error as Normal priority.
Tue, Sep 24, 3:36 PM · Conduit
epriestley requested review of D20823: Fix a cursor paging issue in Conduit call logs.
Tue, Sep 24, 3:36 PM
epriestley added a revision to T13423: Conduit log paging is broken, fails with `Call to undefined method PhabricatorConduitLogQuery::withIDs` error: D20823: Fix a cursor paging issue in Conduit call logs.
Tue, Sep 24, 3:34 PM · Conduit

Mon, Sep 23

epriestley edited the content of Changelog.
Mon, Sep 23, 9:09 PM
epriestley created an object: 2019 Week 38 (Late September).
Mon, Sep 23, 9:08 PM
epriestley committed rPf0f352ffd083: (stable) Promote 2019 Week 38 (authored by epriestley).
(stable) Promote 2019 Week 38
Mon, Sep 23, 9:08 PM
epriestley committed rP09d86c2d2031: Unprototype "Facts" to clear Maniphest/chart fatals (authored by epriestley).
Unprototype "Facts" to clear Maniphest/chart fatals
Mon, Sep 23, 8:01 PM
epriestley closed D20822: Unprototype "Facts" to clear Maniphest/chart fatals.
Mon, Sep 23, 8:01 PM
epriestley requested review of D20822: Unprototype "Facts" to clear Maniphest/chart fatals.
Mon, Sep 23, 7:44 PM
epriestley added a revision to T13279: Build Charting for Facts: D20822: Unprototype "Facts" to clear Maniphest/chart fatals.
Mon, Sep 23, 7:43 PM · Plans, Facts
epriestley added a comment to T13424: Support builds with TeamCity in Harbormaster.

(This has been made to exist, at least roughly; see PHI1448.)

Mon, Sep 23, 4:34 PM · Harbormaster
epriestley closed T12011: Support builds with Travis CI, a subtask of T9456: Evaluate upstream support for third-party build systems, as Wontfix.
Mon, Sep 23, 4:33 PM · Restricted Project, Harbormaster