##// END OF EJS Templates
inotify: server: new data structure to keep track of changes....
inotify: server: new data structure to keep track of changes. == Rationale for the new structure == Current structure was a dictionary tree. One directory was tracked as a dictionary: - keys: file/subdir name - values: - for a file, the status (a/r/m/...) - for a subdir, the directory representing the subdir It allowed efficient lookups, no matter of the type of the terminal leaf: for part in path.split('/'): tree = tree[part] However, there is no way to represent a directory and a file with the same name because keys are conflicting in the dictionary. Concrete example: Initial state: root dir |- foo (file) |- bar (file) # data state is: {'foo': 'n', 'bar': 'n'} Remove foo: root dir |- bar (file) # Data becomes {'foo': 'r'} until next commit. Add foo, as a directory, and foo/barbar file: root dir |- bar (file) |-> foo (dir) |- barbar (file) # New state should be represented as: {'foo': {'barbar': 'a'}, 'bar': 'n'} however, the key "foo" is already used and represents the old file. The dirstate: D foo A foo/barbar cannot be represented, hence the need for a new structure. == The new structure == 'directory' class. Represents one directory level. * Notable attributes: Two dictionaries: - 'files' Maps filename -> status for the current dir. - 'dirs' Maps subdir's name -> directory object representing the subdir * methods - walk(), formerly server.walk - lookup(), old server.lookup - dir(), old server.dir This new class allows embedding all the tree walks/lookups in its own class, instead of having everything mixed together in server. Incidently, since files and directories are not stored in the same dictionaries, we are solving the previous key conflict problem. The small drawback is that lookup operation is a bit more complex: for a path a/b/c/d/e we have to check twice the leaf, if e is a directory or a file.

File last commit:

r9040:4743d1a6 default
r9115:b55d4471 default
Show More
test-convert-cvs-branch
113 lines | 2.0 KiB | text/plain | TextLexer
/ tests / test-convert-cvs-branch
Frank Kingswood
convert: cvs.py - Allow user to use built-in CVS changeset code....
r6690 #!/bin/sh
Dirkjan Ochtman
change wiki/bts URLs to point to new hostname
r8936 # This is http://mercurial.selenic.com/bts/issue1148
# and http://mercurial.selenic.com/bts/issue1447
Frank Kingswood
convert: cvs.py - Allow user to use built-in CVS changeset code....
r6690
"$TESTDIR/hghave" cvs || exit 80
cvscall()
{
cvs -f "$@"
}
echo "[extensions]" >> $HGRCPATH
echo "convert = " >> $HGRCPATH
echo "graphlog = " >> $HGRCPATH
echo "[convert]" >> $HGRCPATH
echo "cvsps=builtin" >> $HGRCPATH
Patrick Mezard
convert/cvs: add an option to disable remote log caching...
r8169 echo "cvsps.cache=0" >> $HGRCPATH
Frank Kingswood
convert: cvs.py - Allow user to use built-in CVS changeset code....
r6690
echo % create cvs repository
mkdir cvsrepo
cd cvsrepo
Martin Geisler
tests: avoid export FOO=bar bashism
r8350 CVSROOT=`pwd`
export CVSROOT
CVS_OPTIONS=-f
export CVS_OPTIONS
Frank Kingswood
convert: cvs.py - Allow user to use built-in CVS changeset code....
r6690 cd ..
cvscall -q -d "$CVSROOT" init
echo % Create a new project
mkdir src
cd src
Frank Kingswood
tests: allow cvs import to reorder its filename list...
r7465 echo "1" > a
echo "1" > b
cvscall import -m "init" src v0 r0 | sort
Frank Kingswood
convert: cvs.py - Allow user to use built-in CVS changeset code....
r6690 cd ..
cvscall co src
cd src
echo % Branch the project
cvscall tag -b BRANCH
Mads Kiilerich
test-convert-cvs*: mute output from "cvs up"...
r7812 cvscall up -r BRANCH > /dev/null
Frank Kingswood
convert: cvs.py - Allow user to use built-in CVS changeset code....
r6690
echo % Modify file a, then b, then a
echo "2" > a
cvscall ci -m "mod a" | grep '<--' | sed -e 's:.*src/\(.*\),v.*:checking in src/\1,v:g'
echo "2" > b
cvscall ci -m "mod b" | grep '<--' | sed -e 's:.*src/\(.*\),v.*:checking in src/\1,v:g'
echo "3" > a
cvscall ci -m "mod a again" | grep '<--' | sed -e 's:.*src/\(.*\),v.*:checking in src/\1,v:g'
echo % Convert
cd ..
hg convert src | sed -e 's/connecting to.*cvsrepo/connecting to cvsrepo/g'
echo % Check the result
Martin Geisler
tests: replace #...# syntax with {...}
r8523 hg -R src-hg glog --template '{rev} ({branches}) {desc} files: {files}\n'
Henrik Stuart
convert: better support for CVS branchpoints (issue1447)...
r8756
echo ""
echo % issue 1447
cvscall()
{
echo cvs -f "$@"
cvs -f "$@"
sleep 1
}
cvsci()
{
echo cvs -f ci "$@"
cvs -f ci "$@" >/dev/null 2>&1
sleep 1
}
cvscall -Q -d `pwd`/cvsmaster2 init >/dev/null 2>&1
cd cvsmaster2
David Champion
test: fix for compatibilty with true Bourne /bin/sh...
r9040 CVSROOT=`pwd`
export CVSROOT
Henrik Stuart
convert: better support for CVS branchpoints (issue1447)...
r8756 mkdir foo
cd ..
cvscall -Q co -d cvswork2 foo
cd cvswork2
echo foo > a.txt
echo bar > b.txt
cvscall -Q add a.txt b.txt
cvsci -m "Initial commit"
echo foo > b.txt
cvsci -m "Fix b on HEAD"
echo bar > a.txt
cvsci -m "Small fix in a on HEAD"
cvscall -Q tag -b BRANCH
cvscall -Q up -P -rBRANCH
echo baz > b.txt
cvsci -m "Change on BRANCH in b"
hg debugcvsps -x --parents foo | sed -e 's/Author:.*/Author:/' -e 's/Date:.*/Date:/'
cd ..