Page MenuHomePhabricator

Alphare (Raphaël Gomès)
User

Projects

User does not belong to any projects.

User Details

User Since
Apr 18 2019, 7:59 AM (98 w, 4 d)

Recent Activity

Fri, Mar 5

Alphare closed D10109: requirements: also add a fncache constant.
Fri, Mar 5, 4:34 AM
Alphare closed D10108: requirements: also add a store constant.
Fri, Mar 5, 4:34 AM
Alphare closed D10107: requirements: also add a dotencode constant.
Fri, Mar 5, 4:34 AM
Alphare committed rHGb4c2a2af25e2: requirements: also add a fncache constant.
requirements: also add a fncache constant
Fri, Mar 5, 4:34 AM
Alphare committed rHGab58098bebed: requirements: also add a store constant.
requirements: also add a store constant
Fri, Mar 5, 4:34 AM
Alphare committed rHGc3773636ddbb: requirements: also add a dotencode constant.
requirements: also add a dotencode constant
Fri, Mar 5, 4:34 AM
Alphare committed rHGe83327af26f1: pure-parsers: document index class constants.
pure-parsers: document index class constants
Fri, Mar 5, 4:34 AM
Alphare closed D10104: pure-parsers: document index class constants.
Fri, Mar 5, 4:34 AM
Alphare accepted D10112: rhg: Fall back to Python if unsupported extensions are enabled.
Fri, Mar 5, 4:05 AM
Alphare accepted D10111: rhg: Sort config files when adding a directory.
Fri, Mar 5, 4:04 AM
Alphare accepted D10110: rhg: Align config file parse error formatting with Python.
Fri, Mar 5, 4:03 AM
Alphare accepted D10102: rhg: Fall back to Python for bundle repositories.
Fri, Mar 5, 4:02 AM
Alphare added a comment to D10101: rhg: Fall back to Python on --repository with an URL.

Maybe I'm trying too hard, but is looking for : in the repo bytes too wide to detect an URL? I know actual URL parsing is out of the question right now, but maybe we can do a little bit more since : is allowed in filenames on many platforms. What do you think?

Fri, Mar 5, 4:00 AM
Alphare accepted D10100: rhg: Print non-absolutized path in "repository {} not found" errors.
Fri, Mar 5, 3:58 AM
Alphare accepted D10099: rhg: Align "malformed --config" error message with Python.
Fri, Mar 5, 3:48 AM
Alphare accepted D10098: rhg: Fall back to Python on unsupported `rhg config <section>`.
Fri, Mar 5, 3:48 AM
Alphare accepted D10097: rhg: Fall back to Python for unsupported revset syntax.
Fri, Mar 5, 3:48 AM
Alphare accepted D10096: tests: Add `rhg` and `no-rhg` for #require and #if in .t files.
Fri, Mar 5, 3:46 AM
Alphare accepted D10095: tests: Add `--rhg` and `--with-rhg=<path>` options for `run-tests.py`.
Fri, Mar 5, 3:44 AM
Alphare accepted D10094: rhg: Make fallback to Python the default behavior.
Fri, Mar 5, 3:42 AM
D10093: rhg: Add support for automatic fallback to Python now requires changes to proceed.
Fri, Mar 5, 3:36 AM
Alphare added a comment to D10093: rhg: Add support for automatic fallback to Python.

Please add a small description to your changeset that explains the new config value and why we would want rhg to host a fallback system itself when it already has a specific error code for this.

Fri, Mar 5, 3:35 AM
Alphare accepted D10091: rhg: Add a `rhg.on-unsupported` configuration key.
Fri, Mar 5, 3:30 AM
Alphare added a comment to D10091: rhg: Add a `rhg.on-unsupported` configuration key.

Your changeset description uses abort twice, FYI, otherwise this looks good.

Fri, Mar 5, 3:27 AM
Alphare accepted D10090: rhg: Make configuration available as early as possible in main().
Fri, Mar 5, 3:21 AM

Thu, Mar 4

Alphare updated the summary of D10026: changegroup: add v4 changegroup for revlog v2 exchange.
Thu, Mar 4, 10:21 AM
Alphare updated the diff for D10032: sidedata-exchange: rewrite sidedata on-the-fly whenever possible.
Thu, Mar 4, 10:21 AM
Alphare updated the summary of D10031: revlog-index: add `replace` method.
Thu, Mar 4, 10:20 AM
Alphare updated the diff for D10029: changegroupv4: add sidedata helpers.
Thu, Mar 4, 10:20 AM
Alphare updated the diff for D10030: revlogv2: temporarily forbid inline revlogs.
Thu, Mar 4, 10:20 AM
Alphare updated the diff for D10028: sidedata-exchange: add `wanted_sidedata` and `sidedata_computers` to repos.
Thu, Mar 4, 10:20 AM
Alphare updated the summary of D10027: delta: add sidedata field to revision delta.
Thu, Mar 4, 10:19 AM
Alphare updated the diff for D9993: sidedata: move to new sidedata storage in revlogv2.
Thu, Mar 4, 10:18 AM
Alphare updated the diff for D10025: revlogv2: don't assume that the sidedata of the last rev is right after data.
Thu, Mar 4, 10:18 AM
Alphare added inline comments to D9844: revlogv2: allow upgrading to v2.
Thu, Mar 4, 10:13 AM
Alphare updated the diff for D9844: revlogv2: allow upgrading to v2.
Thu, Mar 4, 10:12 AM
Alphare created D10113: format: remove sidedata format variant.
Thu, Mar 4, 10:12 AM
Alphare updated the diff for D9843: revlog: introduce v2 format.
Thu, Mar 4, 10:12 AM
Alphare updated the diff for D10109: requirements: also add a fncache constant.
Thu, Mar 4, 10:11 AM
Alphare updated the diff for D10108: requirements: also add a store constant.
Thu, Mar 4, 10:11 AM
Alphare updated the diff for D10107: requirements: also add a dotencode constant.
Thu, Mar 4, 10:11 AM
Alphare updated the summary of D10104: pure-parsers: document index class constants.
Thu, Mar 4, 10:10 AM
Alphare added inline comments to D10031: revlog-index: add `replace` method.
Thu, Mar 4, 10:06 AM
Alphare added inline comments to D10104: pure-parsers: document index class constants.
Thu, Mar 4, 10:02 AM
Alphare closed D10106: requirements: also add a generaldelta constant.
Thu, Mar 4, 7:39 AM
Alphare closed D10105: requirements: add constant for revlog v1 requirement.
Thu, Mar 4, 7:39 AM
Alphare committed rHGf4c325bf80fc: requirements: also add a generaldelta constant.
requirements: also add a generaldelta constant
Thu, Mar 4, 7:38 AM
Alphare committed rHGee91966aec0f: requirements: add constant for revlog v1 requirement.
requirements: add constant for revlog v1 requirement
Thu, Mar 4, 7:38 AM
Alphare committed rHG3941fe53670d: error: add `hint` attribute to `SidedataHashError`.
error: add `hint` attribute to `SidedataHashError`
Thu, Mar 4, 7:38 AM
Alphare closed D10024: error: add `hint` attribute to `SidedataHashError`.
Thu, Mar 4, 7:38 AM
Alphare closed D10023: changegroup: use the local variable instead of reaching through self.
Thu, Mar 4, 7:38 AM
Alphare committed rHG357d2ea95ce9: changegroup: use the local variable instead of reaching through self.
changegroup: use the local variable instead of reaching through self
Thu, Mar 4, 7:38 AM
Alphare added inline comments to D10107: requirements: also add a dotencode constant.
Thu, Mar 4, 4:04 AM
Alphare planned changes to D9844: revlogv2: allow upgrading to v2.
Thu, Mar 4, 3:35 AM
Alphare added inline comments to D9843: revlog: introduce v2 format.
Thu, Mar 4, 3:34 AM
Alphare updated the diff for D9844: revlogv2: allow upgrading to v2.
Thu, Mar 4, 3:34 AM
Alphare updated the diff for D9843: revlog: introduce v2 format.
Thu, Mar 4, 3:33 AM
Alphare updated the summary of D9846: cext: add support for revlogv2.
Thu, Mar 4, 3:33 AM
Alphare created D10109: requirements: also add a fncache constant.
Thu, Mar 4, 3:33 AM
Alphare created D10106: requirements: also add a generaldelta constant.
Thu, Mar 4, 3:32 AM
Alphare created D10108: requirements: also add a store constant.
Thu, Mar 4, 3:32 AM
Alphare created D10105: requirements: add constant for revlog v1 requirement.
Thu, Mar 4, 3:32 AM
Alphare created D10107: requirements: also add a dotencode constant.
Thu, Mar 4, 3:32 AM
Alphare created D10104: pure-parsers: document index class constants.
Thu, Mar 4, 3:31 AM

Wed, Mar 3

Alphare added inline comments to D9993: sidedata: move to new sidedata storage in revlogv2.
Wed, Mar 3, 6:52 AM

Tue, Mar 2

Alphare added a comment to D10071: copies-rust: add a macro-based unit-testing framework.

The plan is that @marmoute can write test cases using this harness at some point. I’m much less familiar with the *intended* behavior of various algorithms, so while I could write test cases by reverse-engineering the code I risk testing bugs without necessarily realizing they’re bugs.

Tue, Mar 2, 9:33 AM
Alphare added a comment to D10071: copies-rust: add a macro-based unit-testing framework.

This was a harder one, macro_rules!, for all their usefulness are not the easiest thing to digest.

Tue, Mar 2, 9:18 AM
Alphare accepted D10068: copies-rust: rewrite ChangedFiles binary parsing.
Tue, Mar 2, 8:36 AM
Alphare added a comment to D10089: rust: Upgrade Cargo.lock to the newer format.

This should probably go to the stable branch though

Tue, Mar 2, 3:59 AM
Alphare accepted D10089: rust: Upgrade Cargo.lock to the newer format.
Tue, Mar 2, 3:59 AM
Alphare added a comment to D10085: copies-rust: remove an unnecessary format!() inside assert!().

But looking up this blog post I see that support for this new format was introduced in 1.40.0. Since the oldest version we support is 1.41.1 (for Debian) maybe we should just do the upgrade.

Tue, Mar 2, 3:39 AM

Thu, Feb 25

Alphare added inline comments to D9843: revlog: introduce v2 format.
Thu, Feb 25, 11:46 AM

Wed, Feb 24

Alphare closed D10034: narrow: fix flaky behavior described in issue6150.
Wed, Feb 24, 11:07 AM
Alphare committed rHGb994db7c4d1e: narrow: fix flaky behavior described in issue6150.
narrow: fix flaky behavior described in issue6150
Wed, Feb 24, 11:07 AM
Alphare accepted D10064: re2: byteify some regex used to get check re2 availability.
Wed, Feb 24, 9:34 AM

Tue, Feb 23

D10018: tags: update the hgtagsfnodes cache if invalid file nodes are found now requires changes to proceed.

I'm waiting for the "next patch" mentioned in the commit message before this can be accepted, since the behavior changed from an abort to a silent omission of valid data, which is a regression.

Tue, Feb 23, 4:47 AM
Alphare accepted D10017: tags: return set of invalid nodes from _tagsfromfnodes().
Tue, Feb 23, 4:42 AM
Alphare accepted D10016: tags: return tag cache source from _readtagcache().
Tue, Feb 23, 4:41 AM
Alphare accepted D10015: debugtagscache: verify that filenode is correct.
Tue, Feb 23, 4:33 AM
Alphare accepted D9812: tests: demonstrate a case where a corrupt tag cache causes an abort.
Tue, Feb 23, 4:31 AM
Alphare added a comment to D10014: debugcommands: prevent using `is False`.

Is if foo == False valid in this case? I probably got my wires crossed with is None. (I'm fine with the code change here, just wondering for future reference.)

Tue, Feb 23, 4:20 AM
Alphare accepted D10013: hgtagsfnodes: refactor code to compute fnode into separate fn.
Tue, Feb 23, 4:16 AM
Alphare accepted D9994: error: remove shortening of node in error message.
Tue, Feb 23, 4:14 AM
Alphare added a comment to D9992: upgrade: speed up when we have only nodemap to downgrade.

I think I'd rather we also delete the docket file(s) and transaction files if we can, but this would not be a blocker for this patch in particular to land.

Tue, Feb 23, 4:11 AM
Alphare accepted D9991: upgrade: write nodemap for manifests too.
Tue, Feb 23, 4:02 AM
Alphare accepted D10033: helptext: fix a recent typo.
Tue, Feb 23, 4:00 AM

Mon, Feb 22

Alphare accepted D10047: test-copies: add subcase titles for various "conflicting" information variant.
Mon, Feb 22, 10:59 AM
Alphare accepted D10046: test-copies: improve description of the B+F case.
Mon, Feb 22, 10:52 AM
Alphare accepted D10045: test-copies: improve description of the C+H case.
Mon, Feb 22, 10:52 AM
Alphare accepted D10044: test-copies: improve description of the B+C "revert/restore" case.
Mon, Feb 22, 10:47 AM
Alphare accepted D10043: test-copies: improve description of the G+C case.
Mon, Feb 22, 10:43 AM
Alphare accepted D10042: test-copies: improve description of the G+F case.
Mon, Feb 22, 10:42 AM
Alphare accepted D10041: test-copies: improve description of the D+G case.
Mon, Feb 22, 10:40 AM
Alphare accepted D10040: test-copies: improve description of the A+E case.
Mon, Feb 22, 10:34 AM
Alphare accepted D10039: test-copies: improve description of the B+D case.
Mon, Feb 22, 10:33 AM
Alphare accepted D10038: test-copies: improve description of the B+C case.
Mon, Feb 22, 10:32 AM
Alphare accepted D10037: test-copies: improve description of the A+B case.
Mon, Feb 22, 10:30 AM
Alphare accepted D10036: test-copies: use intermediate variable some commit descriptions.
Mon, Feb 22, 10:30 AM
Alphare added inline comments to D10035: test-copies: don't use empty file for "same content" cases.
Mon, Feb 22, 10:28 AM

Fri, Feb 19

Alphare added a comment to D10034: narrow: fix flaky behavior described in issue6150.

/!\ This change is destined for the stable branch

Fri, Feb 19, 1:40 PM