fix pull racing with push/commit (issue1320) changegroup() has a problem when nodes which does not descend from a node in <bases> are added to remote after the discovery phase. If that happens, changegroup() won't send the correct set of nodes, ie. some nodes will be missing. To correct it we have to find the set of nodes that both remote and self have (called <common>), and send all the nodes not in <common>. This fix has some overhead, in the worst case it will re-send a whole branch. A proper fix to avoid this overhead might be to change the protocol so that the <common> nodes are sent (instead of the <bases> of the missing nodes).

     1

#!/bin/sh

     2

     3

hg init repo

     4

cd repo

     5

echo foo > foo

     6

hg ci -qAm 'add foo' -d '0 0'

     7

echo >> foo

     8

hg ci -m 'change foo' -d '0 0'

     9

hg up -qC 0

    10

echo bar > bar

    11

hg ci -qAm 'add bar' -d '0 0'

    12

hg log

    13

cd ..

    14

hg init copy

    15

cd copy

    16

    17

echo '% pull a missing revision'

    18

hg pull -qr missing ../repo

    19

    20

echo '% pull -r 0'

    21

hg pull -qr 0 ../repo

    22

hg log

    23

    24

echo '% pull -r 1'

    25

hg pull -qr 1 ../repo

    26

hg log

    27

    28

# this used to abort: received changelog group is empty

    29

echo '% pull -r 1 again'

    30

hg pull -qr 1 ../repo