Page MenuHomePhabricator

dsp (David Soria Parra)
User

Projects

User does not belong to any projects.

User Details

User Since
Jun 30 2017, 8:08 AM (102 w, 2 d)

Recent Activity

Jan 20 2018

dsp accepted D1912: p4fastimport: fix ignored executable bits.

that looks good to me. Thank you for the contribution!

Jan 20 2018, 1:20 AM

Jan 2 2018

dsp added inline comments to D1743: p4fastimport: allow sync commits only when imported repository is up-to-date.
Jan 2 2018, 2:42 PM
dsp requested review of D1743: p4fastimport: allow sync commits only when imported repository is up-to-date.
Jan 2 2018, 2:38 PM
dsp added inline comments to D1743: p4fastimport: allow sync commits only when imported repository is up-to-date.
Jan 2 2018, 2:37 PM

Dec 22 2017

dsp accepted D1314: p4fastimport: work around P4 symlink/directory bug.

looks good. Thank you!

Dec 22 2017, 4:51 PM

Dec 21 2017

dsp created D1743: p4fastimport: allow sync commits only when imported repository is up-to-date.
Dec 21 2017, 8:18 PM
dsp created D1742: p4fastimport: return the last changelist number that affects the client view.
Dec 21 2017, 8:18 PM
dsp created D1741: p4fastimport: fix symlink test on MacOS.
Dec 21 2017, 8:18 PM

Nov 29 2017

dsp requested changes to D1543: repack: move to flock based locking.

I am putting this back into the queue as it will break Windows builds, where hg repack currently works. fcntl does not exist on Windows. We can probably get a away with implementing a windows version that relies on the old repo._lock() mechanism.

Nov 29 2017, 9:24 PM

Nov 24 2017

dsp closed D1497: hbisect: pass repo into hbisect.bisect.
Nov 24 2017, 11:30 PM
dsp committed rHG8287df8b7be5: hbisect: use a defaultdict to avoid large allocations for a large changelogs.
hbisect: use a defaultdict to avoid large allocations for a large changelogs
Nov 24 2017, 11:30 PM
dsp closed D1499: hbisect: use a defaultdict to avoid large allocations for a large changelogs.
Nov 24 2017, 11:30 PM
dsp committed rHGec25c8275cfa: hbisect: use a revset for ancestor calculation.
hbisect: use a revset for ancestor calculation
Nov 24 2017, 11:30 PM
dsp closed D1498: hbisect: use a revset for ancestor calculation.
Nov 24 2017, 11:30 PM
dsp committed rHGfd8b6b183073: hbisect: pass repo into hbisect.bisect.
hbisect: pass repo into hbisect.bisect
Nov 24 2017, 11:30 PM

Nov 23 2017

dsp created D1497: hbisect: pass repo into hbisect.bisect.
Nov 23 2017, 5:24 PM
dsp created D1499: hbisect: use a defaultdict to avoid large allocations for a large changelogs.
Nov 23 2017, 5:24 PM
dsp created D1498: hbisect: use a revset for ancestor calculation.
Nov 23 2017, 5:24 PM

Nov 22 2017

dsp accepted D1491: phabricator: fix lint warning.
Nov 22 2017, 6:15 PM

Nov 21 2017

dsp added inline comments to D1314: p4fastimport: work around P4 symlink/directory bug.
Nov 21 2017, 3:56 PM

Nov 8 2017

dsp accepted D1315: p4fastimport: better handling of symlinks that later turn in to directories.
Nov 8 2017, 1:49 PM
dsp accepted D1306: tests: split p4fastimport-import-lfs.t into normal LFS vs SQLite.
Nov 8 2017, 1:41 PM

Oct 5 2017

dsp closed D920: p4fastimport: fix resetting file flags.
Oct 5 2017, 1:28 PM
dsp committed rFBHGXfad65366d820: p4fastimport: fix resetting file flags.
p4fastimport: fix resetting file flags
Oct 5 2017, 1:28 PM

Oct 3 2017

dsp updated the test plan for D920: p4fastimport: fix resetting file flags.
Oct 3 2017, 1:10 AM
dsp created D920: p4fastimport: fix resetting file flags.
Oct 3 2017, 1:09 AM

Sep 26 2017

dsp closed D766: remotefilelog: rename nodehash to node.
Sep 26 2017, 5:05 PM
dsp committed rFBHGX844b0f3ee953: remotefilelog: rename nodehash to node.
remotefilelog: rename nodehash to node
Sep 26 2017, 5:05 PM
dsp closed D767: p4fastimport: we must return the value from the function we are decorating.
Sep 26 2017, 5:05 PM
dsp committed rFBHGXb91ab9b55ccc: p4fastimport we must return the value from the function we are decorating.
p4fastimport we must return the value from the function we are decorating
Sep 26 2017, 5:05 PM

Sep 21 2017

dsp closed D732: lfs: implement remotefilelog prefetch.
Sep 21 2017, 5:54 PM
dsp closed D729: remotefilelog: remove unecessary return.
Sep 21 2017, 5:53 PM
dsp committed rFBHGX112cf5496b4b: remotefilelog: remove unecessary return.
remotefilelog: remove unecessary return
Sep 21 2017, 5:53 PM
dsp created D766: remotefilelog: rename nodehash to node.
Sep 21 2017, 12:30 PM
dsp created D767: p4fastimport: we must return the value from the function we are decorating.
Sep 21 2017, 12:30 PM

Sep 19 2017

dsp updated the diff for D732: lfs: implement remotefilelog prefetch.
Sep 19 2017, 11:13 PM

Sep 18 2017

dsp created D732: lfs: implement remotefilelog prefetch.
Sep 18 2017, 10:21 PM
dsp created D729: remotefilelog: remove unecessary return.
Sep 18 2017, 8:26 PM

Sep 12 2017

dsp committed rFBHGX7027f9ec4c1f: p4fastimport: add --base option to define branchpoint.
p4fastimport: add --base option to define branchpoint
Sep 12 2017, 11:08 AM

Sep 8 2017

dsp accepted D646: changegroup: fix to allow empty manifest parts.
Sep 8 2017, 2:54 PM
dsp added inline comments to D544: releasenotes: update docstrings with information on additional flags.
Sep 8 2017, 2:43 PM

Aug 30 2017

dsp committed rHGda07367d683b: mdiff: add a --ignore-space-at-eol option.
mdiff: add a --ignore-space-at-eol option
Aug 30 2017, 1:16 PM
dsp closed D422: mdiff: add a --ignore-space-at-eol option by committing rHGae7308d7604e: mdiff: add a --ignore-space-at-eol option.
Aug 30 2017, 9:54 AM
dsp committed rHGae7308d7604e: mdiff: add a --ignore-space-at-eol option.
mdiff: add a --ignore-space-at-eol option
Aug 30 2017, 9:53 AM

Aug 29 2017

dsp updated the diff for D422: mdiff: add a --ignore-space-at-eol option.
Aug 29 2017, 9:22 PM

Aug 24 2017

dsp updated the diff for D422: mdiff: add a --ignore-space-at-eol option.
Aug 24 2017, 6:39 PM
dsp added inline comments to D422: mdiff: add a --ignore-space-at-eol option.
Aug 24 2017, 1:55 PM
dsp accepted D493: dirstate: perform transactions with _copymap using single call, where possible.

I wonder if that is something that we can find with check-code to reduce the overall usage of these kind of patterns.

Aug 24 2017, 1:07 AM
dsp accepted D494: test-casecollision-merge: fix the test.

Thanks

Aug 24 2017, 1:05 AM
dsp requested changes to D492: context: remove unnecessary default values for matchers (API).

if match can't be None, then we should remove it as the default value. Otherwise I look at this in a year and think "why do we not handle a None matcher, when it clearly can be None if the value for match is not set".

Aug 24 2017, 1:05 AM
dsp accepted D491: context: always pass a matcher into _matchstatus() (API).

I guess either works. I personally prefer the inheritance here as it feels easier to get it right from an unsuspected caller, but i can see benefits on both sides.

Aug 24 2017, 1:03 AM
dsp accepted D487: test-extension: enable demandimport explicitly.
Aug 24 2017, 12:53 AM

Aug 23 2017

dsp updated the diff for D422: mdiff: add a --ignore-space-at-eol option.
Aug 23 2017, 10:40 PM
dsp updated the diff for D422: mdiff: add a --ignore-space-at-eol option.
Aug 23 2017, 10:39 PM

Aug 22 2017

dsp accepted D478: run-tests: remove unused pygments token type.
Aug 22 2017, 2:31 PM

Aug 21 2017

dsp accepted D461: morestatus: check whether the conflict message is None before printing.
Aug 21 2017, 11:17 AM

Aug 16 2017

dsp accepted D423: pushvars: do not mangle repo state.
Aug 16 2017, 11:28 PM
dsp added a comment to D423: pushvars: do not mangle repo state.

this looks good to me. One consideration that users would get warnings about pushvars parsing failing fairly late now, compared to directly before we attempt a push. This can be annoying if searching for changes takes a long time. We could attempt to test if they are parsable in commands.py to avoid this.

Aug 16 2017, 11:28 PM
dsp updated the diff for D422: mdiff: add a --ignore-space-at-eol option.
Aug 16 2017, 9:17 PM
dsp added a comment to D422: mdiff: add a --ignore-space-at-eol option.

Updated the diff with your suggested changed

Aug 16 2017, 9:17 PM
dsp updated the summary of D422: mdiff: add a --ignore-space-at-eol option.
Aug 16 2017, 9:15 PM
dsp created D422: mdiff: add a --ignore-space-at-eol option.
Aug 16 2017, 6:13 PM

Aug 11 2017

dsp committed rFBHGXfa5b954dd692: p4fastimport: decode special perforce characters in filenames.
p4fastimport: decode special perforce characters in filenames
Aug 11 2017, 3:28 PM
dsp committed rFBHGX08f1f14f5520: p4fastimport: remove stale comment.
p4fastimport: remove stale comment
Aug 11 2017, 3:28 PM

Aug 8 2017

dsp added inline comments to D245: obsstore: rename precursors into predecessors.
Aug 8 2017, 11:54 AM
dsp accepted D244: obsolete: rename _addprecursors into _addpredecessors.
Aug 8 2017, 11:52 AM
dsp accepted D243: obsmarker: rename precnode into prednode.
Aug 8 2017, 11:52 AM

Aug 3 2017

dsp committed rFBHGXb8bb6d8167e6: p4fastimport: fix symlink handling.
p4fastimport: fix symlink handling
Aug 3 2017, 5:45 AM
dsp committed rFBHGX64fb9152ca66: p4fastimport: remove documentation of non-existing option.
p4fastimport: remove documentation of non-existing option
Aug 3 2017, 5:45 AM
dsp committed rFBHGX8dd6879c1acb: p4fastimport: don't inline data as a workaround for an inline separation issue.
p4fastimport: don't inline data as a workaround for an inline separation issue
Aug 3 2017, 5:45 AM
dsp committed rFBHGX0a82ba9095a2: p4fastimport: make tests not depending on environment.
p4fastimport: make tests not depending on environment
Aug 3 2017, 5:45 AM
dsp committed rFBHGX3d1bfe28be54: p4fastimport: append transactions instead of overwriting.
p4fastimport: append transactions instead of overwriting
Aug 3 2017, 5:45 AM

Aug 1 2017

dsp requested changes to D202: tests: fix test-flagprocessor.t flakiness.
Aug 1 2017, 11:39 AM
dsp updated the summary of D202: tests: fix test-flagprocessor.t flakiness.
Aug 1 2017, 11:34 AM
dsp updated the summary of D202: tests: fix test-flagprocessor.t flakiness.
Aug 1 2017, 11:33 AM

Jul 31 2017

dsp committed rFBHGXa62c8a2e903a: fbarcconfig: support arc feature.
fbarcconfig: support arc feature
Jul 31 2017, 1:00 PM

Jul 28 2017

dsp committed rFBHGX43a6b7f586bb: p4fastimport: ignore keyword expansion when using cvs tools.
p4fastimport: ignore keyword expansion when using cvs tools
Jul 28 2017, 9:29 AM
dsp committed rFBHGXf8ab9fab783b: p4fastimport: use new bookmarks.applychanges API.
p4fastimport: use new bookmarks.applychanges API
Jul 28 2017, 9:29 AM

Jul 16 2017

dsp accepted D96: sparse: require [section] in sparse config files (BC).
Jul 16 2017, 3:49 PM
dsp updated subscribers of D96: sparse: require [section] in sparse config files (BC).

Change looks good to me. I am in favor of that BC change. /cc @durham for an opinion tho.

Jul 16 2017, 3:49 PM
dsp accepted D95: sparse: use set for capturing profiles.

Looks good! Does it make sense to have a test that codifies that sorting doesn't matter? e.g. testing that two sparse profiles

Jul 16 2017, 3:40 PM

Jul 14 2017

dsp accepted D78: codemod: rewrite nested with.

_like_

Jul 14 2017, 10:07 AM · Restricted Project

Jul 13 2017

dsp accepted D58: match: make unionmatcher a proper matcher.
Jul 13 2017, 2:06 PM
dsp accepted D57: match: write forceincludematcher using unionmatcher.

accepting this for visibility to hg-reviewers.

Jul 13 2017, 2:05 PM
dsp accepted D65: histedit: create transaction outside of try.
Jul 13 2017, 11:17 AM
dsp accepted D64: histedit: remove transaction from state object.
Jul 13 2017, 11:16 AM

Jul 12 2017

dsp accepted D64: histedit: remove transaction from state object.
Jul 12 2017, 7:56 PM
dsp added a comment to D57: match: write forceincludematcher using unionmatcher.
In D57#735, @martinvonz wrote:

What arguments do you pass to "hg debugwalk" to get a forceincludematcher?

Jul 12 2017, 7:28 PM
dsp added a comment to D59: sparse: override dirstate.walk() instead of dirstate._ignore.

-1 on this.

Jul 12 2017, 11:06 AM
dsp added a comment to D57: match: write forceincludematcher using unionmatcher.

I am slightly -1 on this, as it removes __repr__ for forceincludematcher. This would lead to remove information from hg debugwalk.

Jul 12 2017, 10:41 AM
dsp added a comment to D58: match: make unionmatcher a proper matcher.

LGTM

Jul 12 2017, 10:37 AM
dsp accepted D17: infinitepush: handle lfs correctly.

@stash thank you for taking care of this!

Jul 12 2017, 10:13 AM

Jul 11 2017

dsp added a comment to D56: py3: make localrepo filtered repo cache work on py3.

LGTM

Jul 11 2017, 9:15 PM
dsp accepted D48: smartlog: remove temporary hack for old revset compatibility.
Jul 11 2017, 7:53 PM
dsp added a comment to D32: revlog: use struct.Struct instances for slight performance wins.

Thanks for the explanation. Can we add this reasoning to the commit message maybe and clearly describe what the microoptimization is?

Jul 11 2017, 11:33 AM
dsp added inline comments to D32: revlog: use struct.Struct instances for slight performance wins.
Jul 11 2017, 11:27 AM
dsp added a comment to D31: revlog: micro-optimize the computation of hashes.

Do we have any data if this speeds up hash computation? The patch seems small enough to justify either way.

Jul 11 2017, 11:24 AM
dsp requested changes to D30: merge: Removed sorting in casefolding detection, for a slight performance win.

I think for the correctness of the algorithm sorted() is not needed. However it is there to ensure that the error is deterministic, as iterating over set is in arbitrary order. So running the same merge with multiple case conflicts can potentially lead to two different errors. sorted makes the iteration deterministic. Fujiwara might have more insights.

Jul 11 2017, 11:19 AM
dsp added inline comments to D17: infinitepush: handle lfs correctly.
Jul 11 2017, 11:10 AM

Jul 6 2017

dsp changed the visibility for D15: contrib: widen "direct use of `python`" net again.
Jul 6 2017, 5:40 PM