In 882fd11 (merge-recursive: Delay content merging for renames 2010-09-20),
there was code that checked for whether we could skip updating a file in
the working directory, based on whether the merged version matched the
current working copy. Due to the desire to handle directory/file conflicts
that were resolvable, that commit deferred content merging by first
updating the index with the unmerged entries and then moving the actual
merging (along with the skip-the-content-update check) to another function
that ran later in the merge process. As part moving the content merging
code, a bug was introduced such that although the message about skipping
the update would be printed (whenever GIT_MERGE_VERBOSITY was sufficiently
high), the file would be unconditionally updated in the working copy
anyway.
When we detect that the file does not need to be updated in the working
copy, update the index appropriately and then return early before updating
the working copy.
Signed-off-by: Elijah Newren <newren@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Some vintage of gcc does not seem to notice last_len is only used when
last_file is already set to non-NULL at which point last_len is also
set.
Noticed on FreeBSD 8
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The code had several places where individual checks were done to remove
files that could be in the way of directories in D/F conflicts. Not all
D/F conflicts could have a path cleared for them in such a manner, however,
leading to the need to create make_room_for_directories_of_df_conflicts()
as done in the previous patch. That new function could not have been
incorporated into the code sooner, since not all relevant code paths had
been deferred to process_df_entry() yet, leading to the creation of even
more of these now-redundant path removals.
Clean out all of these extra D/F path clearing cases.
Signed-off-by: Elijah Newren <newren@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When there are unmerged entries present, make sure to check for D/F
conflicts first and remove any files present in HEAD that would be in the
way of creating files below the correspondingly named directory. Such
files will be processed again at the end of the merge in
process_df_entry(); at that time we will be able to tell if we need to
and can reinstate the file, whether we need to place its contents in a
different file due to the directory still being present, etc.
Signed-off-by: Elijah Newren <newren@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
If all the paths below some directory involved in a D/F conflict were not
removed during the rest of the merge, then the contents of the file whose
path conflicted needs to be recorded in file with an alternative filename.
Signed-off-by: Elijah Newren <newren@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
If all the paths below some directory involved in a D/F conflict were not
removed during the rest of the merge, then the contents of the file whose
path conflicted needs to be recorded in file with an alternative filename.
Signed-off-by: Elijah Newren <newren@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This function is called from process_df_entry(), near the end of the merge.
Rather than just checking whether one of the sides of the merge had a
directory at the same path as one of our files, check whether that
directory is still present by this point of our merge.
Signed-off-by: Elijah Newren <newren@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
If all the paths below some directory involved in a D/F conflict were not
removed during the rest of the merge, then the contents of the file whose
path conflicted needs to be recorded in file with an alternative filename.
Signed-off-by: Elijah Newren <newren@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When handling merges with modify/delete conflicts, if the modified path is
involved in a D/F conflict, handle the issue in process_df_entry() rather
than process_entry().
Signed-off-by: Elijah Newren <newren@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Move the handling of content merging for renames from process_renames() to
process_df_entry().
Signed-off-by: Elijah Newren <newren@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Move the handling of rename/delete conflicts from process_renames() to
process_df_entry().
Signed-off-by: Elijah Newren <newren@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Move the handling of rename/rename conflicts where one file is renamed on
both sides to the same file, from process_renames() to process_entry().
Here we avoid the three way merge logic by just using
update_stages_and_entry() to move the higher stage entries in the index
from the rename source to the rename destination, and then allow
process_entry() to do its magic.
Signed-off-by: Elijah Newren <newren@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Move the handling of rename/rename conflicts where one file is renamed to
two different files, from process_renames() to process_df_entry().
Signed-off-by: Elijah Newren <newren@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When a commit moves A to B while another commit created B (or moved C to
B), and these two different commits serve as different merge-bases for a
later merge, c94736a (merge-recursive: don't segfault while handling
rename clashes 2009-07-30) added some special code to avoid segfaults.
Since that commit, the two versions of B are merged in place (which could
be potentially conflicting) and the intermediate result is used as the
virtual ancestor.
However, right before this special merge, try_merge was turned on, meaning
that process_renames() would try an alternative merge that ignores the
'add' part of the conflict, and, if the merge is clean, store that as the
new virtual ancestor. This could cause incorrect merging of criss-cross
merges; it would typically result in just recording a slightly confusing
merge base, but in some cases it could cause silent acceptance of one side
of a merge as the final resolution when a conflict should have been
flagged.
When we do a special merge for such a rename/add conflict between
merge-bases, turn try_merge off to avoid an inappropriate second merge.
Signed-off-by: Elijah Newren <newren@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Enable calling merge_content() and providing more information about renames
and D/F conflicts (which we will want to do from process_df_entry()).
Signed-off-by: Elijah Newren <newren@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
To facilitate having this function called later using information stored
in a rename_df_conflict_info struct, accept a diff_filepair instead of a
rename.
Signed-off-by: Elijah Newren <newren@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Modify process_df_entry() (mostly just indentation level changes) to
get it ready for handling more D/F conflict type cases.
Signed-off-by: Elijah Newren <newren@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
If an entry has an associated rename_df_conflict_info, skip it and allow
it to be processed by process_df_entry().
Signed-off-by: Elijah Newren <newren@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
process_renames() and process_entry() have nearly identical code for
doing three-way file merging to resolve content changes. Since we are
already deferring some of the current rename handling in order to better
handle D/F conflicts, it seems to make sense to defer content merging as
well and remove the (nearly) duplicated code sections for handling this
merging.
To facilitate this process, add a new update_stages_and_entry() function
which will map the higher stage index entries from two files involved in a
rename into the resulting rename destination's index entries, and update
the associated stage_data structure.
Signed-off-by: Elijah Newren <newren@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Since we need to resolve paths (including renames) in-core first and defer
checking of D/F conflicts (namely waiting to see if directories are still
in the way after all paths are resolved) before updating files involved in
D/F conflicts, we will need to first process_renames, then record some
information about the rename needed at D/F resolution time, and then make
use of that information when resolving D/F conflicts at the end.
This commit adds some relevant data structures for storing the necessary
information.
Signed-off-by: Elijah Newren <newren@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This move is in preparation for merge_content growing and being called from
multiple places in order to handle D/F conflicts.
I also snuck in a small change to the output in the case that the merged
content for the file matches the current file contents, to make it better
match (and thus more able to take over) how other merge_file() calls in
process_renames() are handled.
Signed-off-by: Elijah Newren <newren@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This move is in preparation for the function being called from multiple
places in order to handle D/F conflicts.
Signed-off-by: Elijah Newren <newren@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This move is in preparation for the function growing and being called from
multiple places in order to handle D/F conflicts.
Signed-off-by: Elijah Newren <newren@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Since we want to resolve merges in-core and then detect at the end whether
D/F conflicts remain in the way, we should just apply renames in-core and
let logic elsewhere check for D/F conflicts.
Signed-off-by: Elijah Newren <newren@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The names conflict_rename_rename and conflict_rename_rename_2 did not make
it clear what they were handling. Since the first of these handles one
file being renamed in both branches to different files, while the latter
handles two different files being renamed to the same thing, add a little
'1to2' and '2to1' suffix on these and an explanatory comment to make their
intent clearer.
Signed-off-by: Elijah Newren <newren@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
process_renames() had a variable named "stage" and derived variables
src_other and dst_other whose purpose was not immediately obvious; also,
I want to extend the scope of this variable and use it later, so it should
have a more descriptive name. Do so, and add a brief comment explaining
how it is used and what it relates to.
Signed-off-by: Elijah Newren <newren@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
In 3734893 (merge-recursive: Fix D/F conflicts 2010-07-09),
process_df_entry() was added to process_renames() and process_entry() but
in a somewhat restrictive manner. Modify the code slightly to make it
clearer how we could chain more such functions if necessary, and alter
process_df_entry() to handle such chaining.
Signed-off-by: Elijah Newren <newren@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
In merge-recursive.c, whenever there was a rename where a file name on one
side of the rename matches a directory name on the other side of the merge,
then the very first check that
string_list_has_string(&o->current_directory_set, ren1_dst)
would trigger forcing it into marking it as a rename/directory conflict.
However, if the path is only renamed on one side and a simple three-way
merge between the separate files resolves cleanly, then we don't need to
mark it as a rename/directory conflict. So, we can simply move the check
for rename/directory conflicts after we've verified that there isn't a
rename/rename conflict and that a threeway content merge doesn't work.
This changes the particular error message one gets in the case where the
directory name that a file on one side of the rename matches is not also
part of the rename pair. For example, with commits containing the files:
COMMON -> (HEAD, MERGE )
--------- --------------- -------
sub/file1 -> (sub/file1, newsub)
<NULL> -> (newsub/newfile, <NULL>)
then previously when one tried to merge MERGE into HEAD, one would get
CONFLICT (rename/directory): Rename sub/file1->newsub in HEAD directory newsub added in merge
Renaming sub/file1 to newsub~HEAD instead
Adding newsub/newfile
Automatic merge failed; fix conflicts and then commit the result.
After this patch, the error message will instead become:
Removing newsub
Adding newsub/newfile
CONFLICT (file/directory): There is a directory with name newsub in merge. Adding newsub as newsub~HEAD
Automatic merge failed; fix conflicts and then commit the result.
That makes more sense to me, because git can't know that there's a conflict
until after it's tried resolving paths involving newsub/newfile to see if
they are still in the way at the end (and if newsub/newfile is not in the
way at the end, there should be no conflict at all, which did not hold with
git previously).
Signed-off-by: Elijah Newren <newren@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
>Due to this this (and maybe all the tests) need to depend on the
>SYMLINKS prereq.
Here's a third attempt with no use of symlinks in the test:
Skip the entire rename/add conflict case if the file added on the
other branch has the same contents as the file being renamed. This
avoids giving the user an extra copy of the same file and presenting a
conflict that is confusing and pointless.
A simple test of this case has been added in
t/t3030-merge-recursive.sh.
Signed-off-by: Ken Schalk <ken.schalk@intel.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This is a preparation patch to let setup_unpack_trees_porcelain set
show_all_errors itself.
Signed-off-by: Matthieu Moy <Matthieu.Moy@imag.fr>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The function is currently dealing only with error messages, but the
intent of calling it is really to notify the unpack-tree mechanics that
it is running in porcelain mode.
Signed-off-by: Matthieu Moy <Matthieu.Moy@imag.fr>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
In 5a2580d (merge_recursive: Fix renames across paths below D/F conflicts
2010-07-09) and ae74548 (merge-recursive: Fix multiple file rename across
D/F conflict 2010-08-17), renames across D/F conflicts were fixed by
making process_renames() consider as unprocessed renames whose dst_entry
"still" had higher stage entries. The assumption was that those higher
stage entries would have been cleared out of dst_entry by that point in
cases where the conflict could be resolved (normal renames with no D/F
conflicts). That is not the case -- higher stage entries will remain in
all cases.
Fix this by checking for higher stage entries corresponding to D/F
conflicts, namely that stages 2 and 3 have exactly one nonzero mode between
them. The nonzero mode stage corresponds to a file at the path, while the
stage with a zero mode will correspond to a directory at that path (since
rename/delete conflicts will have already been handled before this codepath
is reached.)
Signed-off-by: Elijah Newren <newren@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
In 5a2580d (merge_recursive: Fix renames across paths below D/F conflicts
2010-07-09), detection was added for renames across paths involved in a
directory<->file conflict. However, the change accidentally involved
reusing an outer loop index ('i') in an inner loop, changing its values
and causing a slightly different type of breakage for cases where there are
multiple renames across the D/F conflict. Fix by creating a new temporary
variable 'i'.
Signed-off-by: Elijah Newren <newren@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When an error is encountered, it calls add_rejected_file() which either
- directly displays the error message and stops if in plumbing mode
(i.e. if show_all_errors is not initialized at 1)
- or stores it so that it will be displayed at the end with display_error_msgs(),
Storing the files by error type permits to have a list of files for
which there is the same error instead of having a serie of almost
identical errors.
As each bind_overlap error combines a file and an old file, a list cannot be
done, therefore, theses errors are not stored but directly displayed.
Signed-off-by: Matthieu Moy <Matthieu.Moy@imag.fr>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
To limit the number of possible error messages, the error messages for
the case would_lose_untracked_file and would_lose_orphaned in
unpack_trees_options.msgs were handled with a single string,
parameterized by an action string ("overwritten" or "removed").
Instead, we consider them as two different cases, with unparameterized
string. This will make it easier to make separate lists sorted by error
types later.
Only the bind_overlap case still takes two %s parameters, but that's
unavoidable.
Signed-off-by: Matthieu Moy <Matthieu.Moy@imag.fr>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
A porcelain message was first added in checkout.c in the commit
8ccba008 (Junio C Hamano, Sat May 17 21:03:49 2008, unpack-trees:
allow Porcelain to give different error messages) to give better feedback
in the case of merge errors.
This patch adapts the porcelain messages for the case of checkout
instead. This way, when having a checkout error, "merge" no longer
appears in the error message.
While we're there, we add an advice in the case of
would_lose_untracked_file.
Signed-off-by: Matthieu Moy <Matthieu.Moy@imag.fr>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The list of error messages was introduced as a structure, but an array
indexed over an enum is more flexible, since it allows one to store a
type of error message (index in the array) in a variable.
This change needs to rename would_lose_untracked ->
would_lose_untracked_file to avoid a clash with the function
would_lose_untracked in merge-recursive.c.
Signed-off-by: Matthieu Moy <Matthieu.Moy@imag.fr>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Teach "git merge-recursive" a --renormalize option to enable the
merge.renormalize configuration. The --no-renormalize option can
be used to override it in the negative.
So in the future, you might be able to, e.g.:
git checkout -m -Xrenormalize otherbranch
or
git revert -Xrenormalize otherpatch
or
git pull --rebase -Xrenormalize
The bad part: merge.renormalize is still not honored for most
commands. And it reveals lots of places that -X has not been plumbed
in (so we get "git merge -Xrenormalize" but not much else).
NEEDSWORK: tests
Cc: Eyvind Bernhardsen <eyvind.bernhardsen@gmail.com>
Signed-off-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Add a “renormalize” bit to the ll-merge options word so callers can
decide on a case-by-case basis whether the merge is likely to have
overlapped with a change in smudge/clean rules.
This reveals a few commands that have not been taking that situation
into account, though it does not fix them.
No functional change intended.
Cc: Eyvind Bernhardsen <eyvind.bernhardsen@gmail.com>
Improved-by: Junio C Hamano <gitster@pobox.com>
Signed-off-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
ll_merge() takes its options in a flag word, which has a few
advantages:
- options flags can be cheaply passed around in registers, while
an option struct passed by pointer cannot;
- callers can easily pass 0 without trouble for no options,
while an option struct passed by value would not allow that.
The downside is that code to populate and access the flag word can be
somewhat opaque. Mitigate that with a few macros.
Cc: Avery Pennarun <apenwarr@gmail.com>
Cc: Bert Wesarg <bert.wesarg@googlemail.com>
Signed-off-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Add a "renormalize" option to struct merge_options so callers can
decide on a case-by-case basis whether the merge is likely to have
overlapped with a change in smudge/clean rules. The option defaults
to the global merge_renormalize setting for now.
No change in behavior intended.
Cc: Eyvind Bernhardsen <eyvind.bernhardsen@gmail.com>
Signed-off-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The merge machinery decides whether to resmudge and clean relevant
entries based on the global merge_renormalize setting, which is set by
"git merge" based on its configuration (and left alone by other
commands).
A nicer interface would make that decision a parameter to merge_trees
so callers would pass in a choice made on a call-by-call basis.
Start by making blob_unchanged stop examining the merge_renormalize
global.
In other words, this change is a trivial no-op, but it brings us
closer to something good.
Cc: Eyvind Bernhardsen <eyvind.bernhardsen@gmail.com>
Signed-off-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The rename logic in process_renames() handles renames and merging of file
contents and then marks files as processed. However, there may be higher
stage entries left in the index for other reasons (e.g., due to D/F
conflicts). By checking for such cases and marking the entry as not
processed, it allows process_entry() later to look at it and handle those
higher stages.
Signed-off-by: Elijah Newren <newren@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The D/F conflicts that can be automatically resolved (file or directory
unmodified on one side of history), have the nice property that
process_entry() can correctly handle all subpaths of the D/F conflict. In
the case of D->F conversions, it will correctly delete all non-conflicting
files below the relevant directory and the directory itself (note that both
untracked and conflicting files below the directory will prevent its
removal). So if we handle D/F conflicts after all other conflicts, they
become fairly simple to handle -- we just need to check for whether or not
a path (file/directory) is in the way of creating the new content. We do
this by having process_entry() defer handling such entries to a subsequent
process_df_entry() step.
Signed-off-by: Elijah Newren <newren@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This implements a simple merge strategy for submodule hashes. We check
whether one side of the merge candidates is already contained in the
other and then merge automatically.
If both sides contain changes we search for a merge in the submodule.
In case a single one exists we check that out and suggest it as the
merge resolution. A list of candidates is returned when we find multiple
merges that contain both sides of the changes.
This is useful for a workflow in which the developers can publish topic
branches in submodules and a separate maintainer merges them. In case
the developers always wait until their branch gets merged before tracking
them in the superproject all merges of branches that contain submodule
changes will be resolved automatically. If developers choose to track
their feature branch the maintainer might get a conflict but git will
search the submodule for a merge and suggest it/them as a resolution.
Signed-off-by: Heiko Voigt <hvoigt@hvoigt.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
If a file is modified due to normalization on one branch, and deleted on
another, a merge of the two branches will result in a delete/modify
conflict for that file even if it is otherwise unchanged.
Try to avoid the conflict by normalizing and comparing the "base" file
and the modified file when their sha1s differ. If they compare equal,
the file is considered unmodified and is deleted.
Signed-off-by: Eyvind Bernhardsen <eyvind.bernhardsen@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Elijah Newren