Amended the following in flight:

test-check-code.t says hi.

I like what this patch does. I am not sure why this needs to be experimental since hg revert -i is not. Also, let's document this config option in hg help revert.

@taapas1128 when you get time, can you see whether reading the branchmap can slow down things or not. It other words, can you test some performance numbers with it?

In D6082#88981, @Kwan wrote:

In D6082#88852, @mharbison72 wrote:

In D6082#88851, @pulkit wrote:

@mharbison72 thanks for tips on adding test. Will add tests in next iteration.

I found that Differentials do have a branch field, maybe we can use that? https://secure.phabricator.com/source/phabricator/browse/master/src/applications/differential/customfield/DifferentialBranchField.php

Seems worth a shot. I don’t know anything about it, but presumably this would be rendered specially in the web UI, like the test plan, etc. That sounds better than as a follow up comment. It also seems natural enough that maybe it can be done unconditionally, instead of needing the argument.

Yeah, it shows in the Diff Detail pane, like here. Unfortunately I think it can be only set when using the creatediff endpoint, which is what I had to change my fork to do (though I could be wrong, the conduit docs are too sparse to be sure).

@lothiraldan There is no significant performance benefit which I am able to notice. There is minor improvement of around 0.2 sec on our large repo but that is not stable.

In D6147#89602, @lothiraldan wrote:

The series looks good to me after a very quick review. I think a more in-deep review will be necessary.

@pulkit Do you have measured some performance improvement with this series?

In D6098#89437, @indygreg wrote:

In D6098#89413, @pulkit wrote:

I applied this locally and did hg diff -c . --git --binary locally and the output is empty. Not sure what's broken here.

Maybe it is a Phabricator/integration bug? Try this:

# HG changeset patch
# User Gregory Szorc <gregory.szorc@gmail.com>
# Date 1552069233 28800
#      Fri Mar 08 10:20:33 2019 -0800
# Node ID 577eab065a686f838bf13def7810d8d6f5f7f7ec
# Parent  7e95ade0f369d7509d04d6c0eefc06ca3d26c6e7
wix: restore COPYING.rtf

8427fea04017 accidentally blew away the content of this file.
As part of restoring the content, I updated the copyright year
to 2019.

Differential Revision: https://phab.mercurial-scm.org/D6098

diff --git a/contrib/packaging/wix/COPYING.rtf b/contrib/packaging/wix/COPYING.rtf
index e69de29bb2d1d6434b8b29ae775ad8c2e48c5391..320dd93cce160f2950b9e47d0144efecfe40a1d6
GIT binary patch
literal 1687
zc$|$@-*4MC5Z-fu{11n|tSvG}Qsalr!vL>H8l#D0*bWK<e4!=Ex|JdolCohK{@-^;
z$%~5t+kk+vtmAz@?jwB=E|jT|#$#OaW|L;aOcbgt6JE76n>5L84x+Rsgtev*i6{+f
zQRbC&UWEjfSSt@xwS&2>r5PQMmlp?-TWuZffRS*jNW{h|WfHO0t;&YOQLOnWHp(a`
zgLUxa%b*GjFdW>ZgN*)$$?fPUwy1XB(G)kL%~R1xR|t*6G!rhaUS~H6t#zwIDqjB-
z9pz7-B2k|u@T6ScI+Pm3&+-<sBpLpaCFw0Z$q;DCze}w_ngVR7l=F;ndGWS(cxl8-
zg_qReAE;*_=%FO3;B6Up(y0;bZ40?HKsKNc4%cl^Ktc)fnuOQ3z;e%Dw5CuFhy(=X
zY7e%I4_~G)@Fn`BVb-oS=J1HQGG2l7(C|0DQU$E{vHYIV7d<-MTk0wUS~CKLED*LN
zlAa+u0rcIjfY`j3feAaX4R88DEcc*gc-uGVtXa~=#@UUN#FUZV#1()hHm+<#S}JNu
zPkY!Q8NF7}dR4ai3aA#oO_mSOFH0CtSMY5-pO2@@)rcxXW$CTjs;(2(jYeurnw&IY
z&(xpp=AR#^e*DMeak5-7`}@gqdbe1>{qr2gFdNU8lg}@Y<2lS;=CkL;UC(`Dz}VQF
zP1nkM8qRBru5(c~1eVLSR-}B#eI=*2fM8kK&S5~a>Ey6Ixo!3iH&r0PnnoCn<QFke
zK#)XuS6Qg#{xt4>_&;PKoLPFsMrn*B2$nNcQhgz7xl!~MO|IL5X)sWCjL1^os2uqN
zy~l#|QD$gcZfU-+Ej@EY`zACs^xT1+?xR4X;d(T3jT(tIVe@zA)ZqOaIFHcSKul{4
zokQCC7+p%8gh-kGuN7B2#R-r3kU5<fiWOJz4x-~*A0(P?1}xa}VL(?z2azh+RwyOv
zmh#p!{+ks+Q`}-tj?ei3`jQV<U7)Zeti0-{#a+(_RnehZ(_IC!*rF#kCm>>py(_lw
z&^y`3YN6DB9oRA+cs?zPwCdB&6{+=<cwG?uR*_M;!YyIT7_R?w6KESFNy5o9f6_c+
zu{0n`29f)%3ZeP+`g*t94K}77Shu;>T?2ppd(VZgcvQOJ+qu*<WFHO>tJ&TBaWef1
x@A?k)>Ao9yrIo>aklXs*@p#}p&g*axtrmHi!oO`e9*?JMy2GYlg`X}j{sP5+QltO?

In D5296#88016, @yuja wrote:

• self.entries = set(decodedir(fp.read()).splitlines()) + + self.entries = [] + totalsize = self.vfs.stat('fncache').st_size

I don't think totalsize has to be stat()ed. We can just loop over until
fp.read() reaches to EOF. It's unreliable to assume that the stat('fncache')
see the identical file to the fp open()ed before.

+ chunksize = (10 ** 6) # 1 MB
+ chunk = b''
+ chunksize = min(totalsize, chunksize)

Do we have any test covering totalsize > chunksize case?

Nice work, I have left some inline comments. Can you also add a test where we pushing multiple branches and not every branch is a closed branch?

In D2010#89433, @martinvonz wrote:

In D2010#89337, @indygreg wrote:

Should we queue this patch or abandon it?

I'm for it, even though it leads to inconsistency. However, we may want to discuss ahead of time what our long-term plan for existing symbols is. Do we eventually want to remove that inconsistency? I took a quick look for examples where it seemed obviously not worth it to rename and it was harder to find good examples than I had expected. Perhaps bail_if_changed and extensions.wrap_function are some of the more frequently used. But most very commonly used functions seem to have short names already. So maybe even if we wanted to eventually make it consistent, it won't be as bad as people have feared? I still don't feel very strongly, but I wanted to highlight what it would mean in practice.

I applied this locally and did hg diff -c . --git --binary locally and the output is empty. Not sure what's broken here.

Can you try to add a test for this?

In D2010#89337, @indygreg wrote:

Should we queue this patch or abandon it?

In D6005#89182, @martinvonz wrote:

Can you update the commit message to explain how this works? I think I saw somewhere else that it first uncommits everything and then does an interactive amend. Is that correct?

@mharbison72 thanks for tips on adding test. Will add tests in next iteration.

This review is intended for default branch.

This review is intended for default branch.

testing comments, please ignore

I am not sure the new message is much helpful. As JordiGH and others pointed out on IRC, it will definitely be helpful if we can show which operation has the lock right now.

Amended the following in flight and resolved conflicts while rebasing the next one.