@matheusbsbs / matheus-bagliano / commits / 35e47e3514

merge-ort: add implementation of rename collisions

Implement rename/rename(2to1) and rename/add handling, i.e. a file is renamed into a location where another file is added (with that other file either being a plain add or itself coming from a rename). Note that rename collisions can also have a special case stacked on top: the file being renamed on one side of history is deleted on the other (yielding either a rename/add/delete conflict or perhaps a rename/rename(2to1)/delete[/delete]) conflict. One thing to note here is that when there is a double rename, the code in question only handles one of them at a time; a later iteration through the loop will handle the other. After they've both been handled, process_entry()'s normal add/add code can handle the collision. This code replaces the following from merge-recurisve.c: * all the 2to1 code in process_renames() * the RENAME_TWO_FILES_TO_ONE case of process_entry() * handle_rename_rename_2to1() * handle_rename_add() Also, there is some shared code from merge-recursive.c for multiple different rename cases which we will no longer need for this case (or other rename cases): * handle_file_collision() * setup_rename_conflict_info() The consolidation of six separate codepaths into one is made possible by a change in design: process_renames() tweaks the conflict_info entries within opt->priv->paths such that process_entry() can then handle all the non-rename conflict types (directory/file, modify/delete, etc.) orthogonally. This means we're much less likely to miss special implementation of some kind of combination of conflict types (see commits brought in by 66c62eaec6 ("Merge branch 'en/merge-tests'", 2020-11-18), especially commit ef52778708 ("merge tests: expect improved directory/file conflict handling in ort", 2020-10-26) for more details). That, together with letting worktree/index updating be handled orthogonally in the merge_switch_to_result() function, dramatically simplifies the code for various special rename cases. Signed-off-by: Elijah Newren <newren@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>

Elijah Newren committed Dec 15, 2020 at 18:28 UTC 35e47e3514b5129e6075a67fa9f4659da2f1a1a1

1 file changed +51 -3

merge-ort.c

+51 -3

@@ -795,10 +795,58 @@ static int process_renames(struct merge_options *opt,

795 /* Need to check for special types of rename conflicts... */

796 if (collision && !source_deleted) {

797 /* collision: rename/add or rename/rename(2to1) */

798 - die("Not yet implemented");

798 + const char *pathnames[3];

799 + struct version_info merged;

800 +

801 + struct conflict_info *base, *side1, *side2;

802 + unsigned clean;

803 +

804 + pathnames[0] = oldpath;

805 + pathnames[other_source_index] = oldpath;

806 + pathnames[target_index] = newpath;

807 +

808 + base = strmap_get(&opt->priv->paths, pathnames[0]);

809 + side1 = strmap_get(&opt->priv->paths, pathnames[1]);

810 + side2 = strmap_get(&opt->priv->paths, pathnames[2]);

811 +

812 + VERIFY_CI(base);

813 + VERIFY_CI(side1);

814 + VERIFY_CI(side2);

815 +

816 + clean = handle_content_merge(opt, pair->one->path,

817 + &base->stages[0],

818 + &side1->stages[1],

819 + &side2->stages[2],

820 + pathnames,

821 + 1 + 2 * opt->priv->call_depth,

822 + &merged);

823 +

824 + memcpy(&newinfo->stages[target_index], &merged,

825 + sizeof(merged));

826 + if (!clean) {

827 + path_msg(opt, newpath, 0,

828 + _("CONFLICT (rename involved in "

829 + "collision): rename of %s -> %s has "

830 + "content conflicts AND collides "

831 + "with another path; this may result "

832 + "in nested conflict markers."),

833 + oldpath, newpath);

834 + }

835 } else if (collision && source_deleted) {

800 - /* rename/add/delete or rename/rename(2to1)/delete */

801 - die("Not yet implemented");

836 + /*

837 + * rename/add/delete or rename/rename(2to1)/delete:

838 + * since oldpath was deleted on the side that didn't

839 + * do the rename, there's not much of a content merge

840 + * we can do for the rename. oldinfo->merged.is_null

841 + * was already set, so we just leave things as-is so

842 + * they look like an add/add conflict.

843 + */

844 +

845 + newinfo->path_conflict = 1;

846 + path_msg(opt, newpath, 0,

847 + _("CONFLICT (rename/delete): %s renamed "

848 + "to %s in %s, but deleted in %s."),

849 + oldpath, newpath, rename_branch, delete_branch);

850 } else {

851 /*

852 * a few different cases...start by copying the