alexmv (Alex Vandiver)
User

Projects

Today

  • Clear sailing ahead.

Tomorrow

  • Clear sailing ahead.

Sunday

  • Clear sailing ahead.

Badges

User Details

User Since
Aug 3 2015, 8:30 AM (81 w, 4 d)
Availability
Available

Recent Activity

Thu, Feb 16

alexmv committed rPHU863226f4e56c: Prevent backtracking on long JSON strings with escape codes (authored by alexmv).
Prevent backtracking on long JSON strings with escape codes
Thu, Feb 16, 10:43 PM
alexmv closed D17363: Prevent backtracking on long JSON strings with escape codes by committing rPHU863226f4e56c: Prevent backtracking on long JSON strings with escape codes.
Thu, Feb 16, 10:43 PM
alexmv added a comment to D17363: Prevent backtracking on long JSON strings with escape codes.

Yup, will try to get this upstreamed.

Thu, Feb 16, 10:23 PM
alexmv updated the diff for D17363: Prevent backtracking on long JSON strings with escape codes.

Non-backtracking forms of the two repetition characters.

Thu, Feb 16, 10:22 PM
alexmv updated the summary of D17363: Prevent backtracking on long JSON strings with escape codes.
Thu, Feb 16, 10:21 PM
epriestley awarded D17363: Prevent backtracking on long JSON strings with escape codes a Evil Spooky Haunted Tree token.
Thu, Feb 16, 2:30 AM
alexmv created D17363: Prevent backtracking on long JSON strings with escape codes.
Thu, Feb 16, 2:15 AM

Tue, Feb 14

alexmv closed T10881: git diff parsing fails with --submodule=log as "Resolved".
Tue, Feb 14, 1:41 AM · Contributor Onboarding, Arcanist, Bug Report
alexmv committed rARCf3037bf216e5: [git] Override `diff.submodule` so `git diff` output is always parseable (authored by alexmv).
[git] Override `diff.submodule` so `git diff` output is always parseable
Tue, Feb 14, 1:40 AM
alexmv closed D17327: [git] Override `diff.submodule` so `git diff` output is always parseable.
Tue, Feb 14, 1:40 AM

Thu, Feb 9

alexmv updated the diff for D17327: [git] Override `diff.submodule` so `git diff` output is always parseable.

Fix typo in comment

Thu, Feb 9, 12:32 AM
alexmv updated the test plan for D17327: [git] Override `diff.submodule` so `git diff` output is always parseable.
Thu, Feb 9, 12:26 AM
alexmv created D17327: [git] Override `diff.submodule` so `git diff` output is always parseable.
Thu, Feb 9, 12:25 AM

Jan 17 2017

alexmv closed T11713: `utf8_decode` is provided by XML extension, which may not be installed as "Resolved".

Fixed by D17188.

Jan 17 2017, 5:54 PM · libphutil, Bug Report

Jan 12 2017

alexmv committed rPHU6cd38ce75fea: Fall back to from utf8_decode to phutil_utf8v for codepoint counting (authored by alexmv).
Fall back to from utf8_decode to phutil_utf8v for codepoint counting
Jan 12 2017, 3:32 PM
alexmv closed D17188: Fall back to from utf8_decode to phutil_utf8v for codepoint counting by committing rPHU6cd38ce75fea: Fall back to from utf8_decode to phutil_utf8v for codepoint counting.
Jan 12 2017, 3:32 PM
alexmv closed D17188: Fall back to from utf8_decode to phutil_utf8v for codepoint counting.
Jan 12 2017, 3:32 PM
alexmv updated the diff for D17188: Fall back to from utf8_decode to phutil_utf8v for codepoint counting.

Rebase and re-run unit tests.

Jan 12 2017, 3:31 PM
alexmv committed rPHUe4b06e06ee78: Stop hardcoding the YACC error message (authored by alexmv).
Stop hardcoding the YACC error message
Jan 12 2017, 3:30 PM
alexmv closed D17189: Stop hardcoding the YACC error message.
Jan 12 2017, 3:30 PM
alexmv created D17189: Stop hardcoding the YACC error message.
Jan 12 2017, 3:26 PM
alexmv added a comment to D17188: Fall back to from utf8_decode to phutil_utf8v for codepoint counting.

I think that's unrelated to this change, and purely based on the yacc version in the environment. Mind if I split that out into a separate diff?

Jan 12 2017, 3:12 PM
alexmv created D17188: Fall back to from utf8_decode to phutil_utf8v for codepoint counting.
Jan 12 2017, 3:02 PM

Dec 21 2016

alexmv committed rP972604e0e59c: Set `TERM` to prevent `No entry for terminal type "unknown"` messages during… (authored by alexmv).
Set `TERM` to prevent `No entry for terminal type "unknown"` messages during…
Dec 21 2016, 11:17 PM
alexmv closed D17100: Set `TERM` to prevent `No entry for terminal type "unknown"` messages during fetch.
Dec 21 2016, 11:17 PM
alexmv created D17100: Set `TERM` to prevent `No entry for terminal type "unknown"` messages during fetch.
Dec 21 2016, 10:58 PM
alexmv added a comment to T12033: Large diffs can still repeatedly fail to insert.

metamta.email-body-limit is unset.

Dec 21 2016, 10:17 PM · Diffusion (v3), Restricted Project, Mail, Differential, Bug Report
alexmv added a comment to T12033: Large diffs can still repeatedly fail to insert.

Aha -- the diff as posted for review was mostly-reasonably sized. The diff as committed was the huge one:

mysql> select differential_diff.id as diff_id, creationMethod, SUM(addLines), SUM(delLines)
    ->   from differential_changeset
    ->   join differential_diff on differential_changeset.diffID = differential_diff.id
    ->  where differential_diff.revisionID = 253041
    ->  group by differential_diff.id;
+---------+----------------+---------------+---------------+
| diff_id | creationMethod | SUM(addLines) | SUM(delLines) |
+---------+----------------+---------------+---------------+
|  837437 | arc            |         11845 |          8567 |
|  837558 | commit         |       1752986 |        396108 |
+---------+----------------+---------------+---------------+
Dec 21 2016, 3:46 AM · Diffusion (v3), Restricted Project, Mail, Differential, Bug Report
alexmv added a comment to T12033: Large diffs can still repeatedly fail to insert.

bin/worker execute --id 35362582 did not work because the task was archived. Instead I ran ./bin/phd stop; ./bin/worker retry --id 35362582; ./bin/phd debug PhabricatorTaskmasterDaemon. The key seems to be the production of the .patch file itself, not the number of files. Somehow these users managed to cram a >100M patchfile into Differential:

1​<VERB> PhabricatorTaskmasterDaemon Working on task 35362582 (PhabricatorApplicationTransactionPublishWorker)...
2​object(PhabricatorMetaMTAMail)#421 (14) {
3​ ["actorPHID":protected"]=>
4​ string(30) "..."
5​ ["parameters":protected]=>
6​ array(18) {
7​ ["sensitive"]=>
8​ bool(false)
9​ ["subject"]=>
10​ string(83) "..."
11​ ["headers"]=>
12​ array(6) {
13​ [0]=>
14​ array(2) {
15​ [0]=>
16​ string(12) "Thread-Topic"
17​ [1]=>
18​ string(83) "..."
19​ }
20​ [1]=>
21​ array(2) {
22​ [0]=>
23​ string(14) "X-Herald-Rules"
24​ [1]=>
25​ string(26) "..."
26​ }
27​ [2]=>
28​ array(2) {
29​ [0]=>
30​ string(16) "X-Phabricator-To"
31​ [1]=>
32​ string(32) "..."
33​ }
34​ [3]=>
35​ array(2) {
36​ [0]=>
37​ string(16) "X-Phabricator-To"
38​ [1]=>
39​ string(32) "..."
40​ }
41​ [4]=>
42​ array(2) {
43​ [0]=>
44​ string(16) "X-Phabricator-To"
45​ [1]=>
46​ string(32) "..."
47​ }
48​ [5]=>
49​ array(2) {
50​ [0]=>
51​ string(16) "X-Phabricator-Cc"
52​ [1]=>
53​ string(32) "..."
54​ }
55​ }
56​ ["from"]=>
57​ string(30) "..."
58​ ["subject-prefix"]=>
59​ string(14) "[Differential]"
60​ ["vary-subject-prefix"]=>
61​ string(8) "[Closed]"
62​ ["thread-id"]=>
63​ string(51) "..."
64​ ["is-first-message"]=>
65​ bool(false)
66​ ["exclude"]=>
67​ array(0) {
68​ }
69​ ["herald-force-recipients"]=>
70​ array(0) {
71​ }
72​ ["mailtags"]=>
73​ array(2) {
74​ [0]=>
75​ string(20) "differential-updated"
76​ [1]=>
77​ string(22) "differential-committed"
78​ }
79​ ["is-bulk"]=>
80​ bool(true)
81​ ["body"]=>
82​ string(16820) "..."
83​ ["html-body"]=>
84​ string(18920) "..."
85​ ["attachments"]=>
86​ array(1) {
87​ [0]=>
88​ array(3) {
89​ ["filename"]=>
90​ string(20) "D253041.837558.patch"
91​ ["mimetype"]=>
92​ string(27) "text/x-patch; charset=utf-8"
93​ ["data"]=>
94​ string(112531411) "
95​Segmentation fault (core dumped)
96​Freeing active task leases...

Dec 21 2016, 3:36 AM · Diffusion (v3), Restricted Project, Mail, Differential, Bug Report
alexmv edited P2027 Masterwork From Distant Lands.
Dec 21 2016, 3:27 AM

Dec 20 2016

alexmv added a comment to T12033: Large diffs can still repeatedly fail to insert.

It will likely be easiest to replicate again after upgrading to Week 52 stable this Friday, since I can't use precise same diff, and don't know the precise set of replication criteria yet.

Dec 20 2016, 1:52 AM · Diffusion (v3), Restricted Project, Mail, Differential, Bug Report

Dec 19 2016

alexmv added a comment to T12033: Large diffs can still repeatedly fail to insert.

Hm; we weren't running HEAD when we reported T11748. We have an upgrade planned to latest-stable in a week or so -- will that be sufficient to re-test with? Or is there a repository that we can apply a 4,800-file diff to on Phacility?

Dec 19 2016, 11:16 PM · Diffusion (v3), Restricted Project, Mail, Differential, Bug Report
alexmv created T12033: Large diffs can still repeatedly fail to insert.
Dec 19 2016, 10:54 PM · Diffusion (v3), Restricted Project, Mail, Differential, Bug Report

Nov 18 2016

alexmv added a comment to D16897: Update the "local:commits" property after amending.

This is a necessary, but not sufficient[1], step to be able to make the server better able to reason about stacked diffs. We use this as part of our CI integration, but it is currently used in the arc patch workflow as part of the message when the base commit is unavailable; it is used to suggest This diff is against commit D12345 rather than against commit 8badf00d.

Nov 18 2016, 8:50 PM
alexmv updated the test plan for D16897: Update the "local:commits" property after amending.
Nov 18 2016, 3:07 AM
alexmv retitled D16897: Update the "local:commits" property after amending from to Update the "local:commits" property after amending.
Nov 18 2016, 3:04 AM

Oct 14 2016

alexmv awarded T1022: arc patch fails to properly apply a patch which changes a symlink to a real file a Like token.
Oct 14 2016, 8:19 PM · Arcanist

Sep 29 2016

alexmv created T11714: Unified diffs don't allow selecting text in inline comments.
Sep 29 2016, 11:50 PM · Differential, Bug Report
alexmv created T11713: `utf8_decode` is provided by XML extension, which may not be installed.
Sep 29 2016, 11:30 PM · libphutil, Bug Report

Sep 21 2016

alexmv committed rARC483e985d08d2: Check both UNIX- and Windows-style paths from linter output (authored by alexmv).
Check both UNIX- and Windows-style paths from linter output
Sep 21 2016, 9:29 PM
alexmv closed T11248: ArcanistLinter's addLintMessage incorrectly normalizes path on Windows as "Resolved" by committing rARC483e985d08d2: Check both UNIX- and Windows-style paths from linter output.
Sep 21 2016, 9:29 PM · Windows, Bug Report
alexmv closed D16579: Check both UNIX- and Windows-style paths from linter output.
Sep 21 2016, 9:29 PM
alexmv retitled D16579: Check both UNIX- and Windows-style paths from linter output from Normalize paths stored in linters to UNIX-style to Check both UNIX- and Windows-style paths from linter output.
Sep 21 2016, 9:28 PM
alexmv updated the diff for D16579: Check both UNIX- and Windows-style paths from linter output.

Somehow another change snuck in, though I don't see it locally?

Sep 21 2016, 9:09 PM
alexmv updated D16579: Check both UNIX- and Windows-style paths from linter output.
Sep 21 2016, 9:08 PM
alexmv updated the diff for D16579: Check both UNIX- and Windows-style paths from linter output.

Do normalization on checking in isRelevantMessage, not during
storage in paths.

Sep 21 2016, 9:05 PM
alexmv planned changes to D16579: Check both UNIX- and Windows-style paths from linter output.
Sep 21 2016, 7:19 PM
alexmv added a comment to D16579: Check both UNIX- and Windows-style paths from linter output.

That seems reasonable. I'll push an updated diff later today, which tries both forms of slashes in isRelemvantMessage() if running on Windows.

Sep 21 2016, 7:19 PM
alexmv added a comment to T11201: Exception when destroying an object with edges that no longer have corresponding edge definitions.

This is still logged as an EXCEPTION. Is DELETE FROM phabricator_repository.edge WHERE type = 39 a reasonable step to take, to prevent administrators unfamiliar with the non-fatal-ness of this from being worried?

Sep 21 2016, 7:07 PM · Bug Report

Sep 20 2016

alexmv retitled D16579: Check both UNIX- and Windows-style paths from linter output from to Normalize paths stored in linters to UNIX-style.
Sep 20 2016, 10:05 PM
alexmv added a revision to T11248: ArcanistLinter's addLintMessage incorrectly normalizes path on Windows: D16579: Check both UNIX- and Windows-style paths from linter output.
Sep 20 2016, 10:05 PM · Windows, Bug Report

Aug 27 2016

alexmv added a comment to T6878: Tagged commits which are not ancestors of any branch head don't get imported.

This can cause a storm of commit mail when upgrading. If there are old tags which weren't imported at the time, when Phabricator is upgraded to a version containing D16129 it can fire off thousands of commit emails, flooding inboxes and making some folks grumpy at their Phabricator administrators.

Aug 27 2016, 3:00 AM · Wikimedia, Restricted Project, Diffusion

Aug 19 2016

alexmv committed rARC89e8b4852384: Update documentation for changed splitGitDiffPaths function (authored by alexmv).
Update documentation for changed splitGitDiffPaths function
Aug 19 2016, 9:55 PM
alexmv closed D16425: Update documentation for changed splitGitDiffPaths function.
Aug 19 2016, 9:55 PM
alexmv retitled D16425: Update documentation for changed splitGitDiffPaths function from to Update documentation for changed splitGitDiffPaths function.
Aug 19 2016, 9:24 PM

Aug 17 2016

alexmv committed rARCd0957c344156: Parse git renames with inconsistent quoting or custom prefixes (authored by alexmv).
Parse git renames with inconsistent quoting or custom prefixes
Aug 17 2016, 12:45 AM
alexmv closed D16405: Parse git renames with inconsistent quoting or custom prefixes.
Aug 17 2016, 12:45 AM
alexmv accepted D16406: Fix a typo in "Internationalization" documentation.
Aug 17 2016, 12:43 AM
alexmv updated the diff for D16405: Parse git renames with inconsistent quoting or custom prefixes.

Address review comments

Aug 17 2016, 12:43 AM
alexmv added inline comments to D16405: Parse git renames with inconsistent quoting or custom prefixes.
Aug 17 2016, 12:43 AM
epriestley awarded D16405: Parse git renames with inconsistent quoting or custom prefixes a Piece of Eight token.
Aug 17 2016, 12:29 AM
alexmv committed rARCee6357386d0b: Correctly parse file renames and copies from `git diff --raw` (authored by alexmv).
Correctly parse file renames and copies from `git diff --raw`
Aug 17 2016, 12:27 AM
alexmv closed D16387: Correctly parse file renames and copies from `git diff --raw`.
Aug 17 2016, 12:27 AM
alexmv retitled D16405: Parse git renames with inconsistent quoting or custom prefixes from to Parse git renames with inconsistent quoting or custom prefixes.
Aug 17 2016, 12:05 AM

Aug 16 2016

alexmv updated D16387: Correctly parse file renames and copies from `git diff --raw`.
Aug 16 2016, 12:39 AM
alexmv added a comment to D16387: Correctly parse file renames and copies from `git diff --raw`.

After further testing, this is only relevant now because git < 2.9 did not do rename detection by default; it was enabled in 2.9.

Aug 16 2016, 12:36 AM

Aug 10 2016

alexmv retitled D16387: Correctly parse file renames and copies from `git diff --raw` from to Correctly parse file renames and copies from `git diff --raw`.
Aug 10 2016, 9:44 PM

Jul 12 2016

alexmv committed rARC8f69a5c378c3: Use phutil functions to copy/move files (authored by alexmv).
Use phutil functions to copy/move files
Jul 12 2016, 5:22 PM
alexmv closed D16273: Use phutil functions to copy/move files by committing rARC8f69a5c378c3: Use phutil functions to copy/move files.
Jul 12 2016, 5:22 PM
alexmv committed rPHU2d457275bea6: Add a method to copy files, preserving attributes (authored by alexmv).
Add a method to copy files, preserving attributes
Jul 12 2016, 5:20 PM
alexmv closed D16275: Add a method to copy files, preserving attributes.
Jul 12 2016, 5:19 PM
alexmv retitled D16273: Use phutil functions to copy/move files from Provide windows command-line equivalents for `cp` and `mv` commands to Use phutil functions to copy/move files.
Jul 12 2016, 12:20 AM
alexmv updated the diff for D16273: Use phutil functions to copy/move files.

Use new libphutil functions.

Jul 12 2016, 12:19 AM
alexmv retitled D16275: Add a method to copy files, preserving attributes from to Add a method to copy files, preserving attributes.
Jul 12 2016, 12:18 AM

Jul 11 2016

alexmv retitled D16273: Use phutil functions to copy/move files from to Provide windows command-line equivalents for `cp` and `mv` commands.
Jul 11 2016, 10:45 PM

Feb 22 2016

alexmv added a comment to T10416: 7b8d38cd causes hangs with SSH master connection.

Using SSH master connections has two advantages:

  1. Connections are indeed faster. In a minimal case (SSH'ing to an adjacent machine on the same network), it reduces the time for a no-op SSH connection from 0.14s to 0.01s. The utility of this grows as one needs to tunnel SSH connections in order to access a machine; for instance, SSH'ing to our Phabricator instance requires bouncing through a bastion host; using SSH master connections reduces the time to connect from ~0.7s to 0.01s. This is notable since arc may produce mulitiple git commands which attempt to fetch from upstream.
  2. Authentication information is preserved. If the connection does two-factor authentication, for instance, the user may be supplied with an authentication prompt when the master is set up. Re-using the master preserves this authenticated session, and reduces the amount of re-authorization that the user must do.
Feb 22 2016, 8:46 PM · Arcanist, Bug Report
alexmv created T10416: 7b8d38cd causes hangs with SSH master connection.
Feb 22 2016, 7:27 PM · Arcanist, Bug Report

Nov 23 2015

alexmv accepted D14554: Address feedback from D14530.

Ack -- sorry for having forgot that fixup.

Nov 23 2015, 6:31 PM
alexmv committed rARCe730ececbc96: Examine upstream path instead of assuming "origin" (authored by alexmv).
Examine upstream path instead of assuming "origin"
Nov 23 2015, 6:29 PM
alexmv closed D14530: Examine upstream path instead of assuming "origin".
Nov 23 2015, 6:29 PM

Nov 20 2015

alexmv committed rARCa77a77817a03: Try switching to the branch of the same name, instead of detached head (authored by alexmv).
Try switching to the branch of the same name, instead of detached head
Nov 20 2015, 6:58 PM
alexmv closed D14420: Try switching to the branch of the same name, instead of detached head.
Nov 20 2015, 6:58 PM
alexmv retitled D14530: Examine upstream path instead of assuming "origin" from to Examine upstream path instead of assuming "origin".
Nov 20 2015, 6:50 PM

Nov 6 2015

alexmv added a comment to D14420: Try switching to the branch of the same name, instead of detached head.

The problem was even though the upstream remote of the branch I was working on was in this arcanist, the remote named origin still pointed to our internal fork. Temporarily swapping the names of the upstreams around made it work.

Nov 6 2015, 7:09 PM
alexmv added a comment to D14420: Try switching to the branch of the same name, instead of detached head.

Nope, no staging area:

$ arc diff HEAD~
Linting...
 LINT OKAY  No lint problems.
Running unit tests...
   PASS   16ms★  ArcanistLibraryTestCase::testLibraryMap
   PASS    9ms★  ArcanistLibraryTestCase::testEverythingImplemented
   PASS   36ms★  ArcanistLibraryTestCase::testMethodVisibility
 UNIT OKAY  No unit test failures.
 SKIP STAGING  Unable to determine repository for this change.
Updated an existing Differential revision:
        Revision URI: https://secure.phabricator.com/D14420
Nov 6 2015, 2:06 AM
alexmv updated the diff for D14420: Try switching to the branch of the same name, instead of detached head.

Actual no-op diff

Nov 6 2015, 1:00 AM
alexmv updated the diff for D14420: Try switching to the branch of the same name, instead of detached head.

No-op diff

Nov 6 2015, 1:00 AM
alexmv added a comment to D14420: Try switching to the branch of the same name, instead of detached head.

Just uploaded. Should I try again?

Nov 6 2015, 12:44 AM
alexmv added a comment to D14420: Try switching to the branch of the same name, instead of detached head.

Landing via the UI failed with https://secure.phabricator.com/drydock/operation/58/ ; I'll use the CLI.

Nov 6 2015, 12:42 AM
alexmv added a task to D14420: Try switching to the branch of the same name, instead of detached head: T9723: When landing a branch tracking upstream, try switching to local tracking branch of the same name.
Nov 6 2015, 12:20 AM
alexmv added a revision to T9723: When landing a branch tracking upstream, try switching to local tracking branch of the same name: D14420: Try switching to the branch of the same name, instead of detached head.
Nov 6 2015, 12:20 AM · Arcanist
alexmv added inline comments to D14420: Try switching to the branch of the same name, instead of detached head.
Nov 6 2015, 12:17 AM

Nov 5 2015

alexmv added a comment to T9723: When landing a branch tracking upstream, try switching to local tracking branch of the same name.

Sent D14420.

Nov 5 2015, 11:44 PM · Arcanist
alexmv retitled D14420: Try switching to the branch of the same name, instead of detached head from to Try switching to the branch of the same name, instead of detached head.
Nov 5 2015, 11:43 PM
alexmv created T9723: When landing a branch tracking upstream, try switching to local tracking branch of the same name.
Nov 5 2015, 9:56 PM · Arcanist

Oct 31 2015

alexmv added a comment to T9676: Triangle workflows (pushing to a merge manager).

The URL to push to is constant. The ref to push to, however, would depend on what one was merging into -- hence what I mean by needing to be computed. Implementing this would also require replacing all of git-receive-pack on the server side; pushing into refs named by the server allows the server to not worry about magically hiding that overlap. As is, the server can manage the refs as normal refs, which works effectively.

Oct 31 2015, 1:55 AM · Restricted Project
alexmv added a comment to T9676: Triangle workflows (pushing to a merge manager).

How does removing the API call simplify matters? One still needs to have code to construct the name of the ref to push to (even if that's just a string concatenation). Doing that computation on a remote machine, rather than locally, makes little difference.

Oct 31 2015, 1:43 AM · Restricted Project
alexmv added a comment to T9676: Triangle workflows (pushing to a merge manager).

Hm, interesting. I see a few problems with that technique, though:

Oct 31 2015, 1:22 AM · Restricted Project

Oct 30 2015

alexmv created T9676: Triangle workflows (pushing to a merge manager).
Oct 30 2015, 9:33 PM · Restricted Project

Oct 28 2015

alexmv added a comment to T3855: `arc land` should work when the current branch/bookmark and the target branch/bookmark are the same.

D14365, applied locally, does fix both issues.

Oct 28 2015, 11:58 PM · Restricted Project, Arcanist