Fix long-standing excessive file merges Since switching to the multihead approach, we've been creating excessive file-level merges where files are marked as merged with their ancestors. This explicitly checks at commit time whether the two parent versions are linearly related, and if so, reduces the file check-in to a non-merge. Then the file is compared against the remaining parent, and, if equal, skips check-in of that file (as it's not changed). Since we're not checking in all files that were different between versions, we no longer need to mark so many files for merge. This removes most of the 'm' state marking as well. Finally, it is possible to do a tree-level merge with no file-level changes. This will happen if one user changes file A and another changes file B. Thus, if we have have two parents, we allow commit to proceed even if there are no file-level changes.

     1

#!/bin/sh

     2

     3

cat <<'EOF' > merge

     4

#!/bin/sh

     5

echo merging for `basename $1`

     6

EOF

     7

chmod +x merge

     8

HGMERGE=./merge; export HGMERGE

     9

    10

mkdir A1

    11

cd A1

    12

hg init

    13

echo This is file foo1 > foo

    14

echo This is file bar1 > bar

    15

hg add foo bar

    16

hg commit -m "commit text" -d "0 0"

    17

    18

cd ..

    19

hg clone A1 B1

    20

    21

cd A1

    22

rm bar

    23

hg remove bar

    24

hg commit -m "commit test" -d "0 0"

    25

    26

cd ../B1

    27

echo This is file foo22 > foo

    28

hg commit -m "commit test" -d "0 0"

    29

    30

cd ..

    31

hg clone A1 A2

    32

hg clone B1 B2

    33

    34

cd A1

    35

hg pull ../B1

    36

hg update -m

    37

hg commit -m "commit test" -d "0 0"

    38

echo bar should remain deleted.

    39

hg manifest

    40

    41

cd ../B2

    42

hg pull ../A2

    43

hg update -m

    44

hg commit -m "commit test" -d "0 0"

    45

echo bar should remain deleted.

    46

hg manifest