bzrlib.merge3.Merge3 : API documentation

3-way merge of texts.

Given BASE, OTHER, THIS, tries to produce a combined text incorporating the changes from both BASE->OTHER and BASE->THIS. All three will typically be sequences of lines.

Method	__init__	Constructor.
Method	merge_lines	Return merge in cvs-like form.
Method	merge_annotated	Return merge with conflicts, showing origin of lines.
Method	merge_groups	Yield sequence of line groups. Each one is a tuple:
Method	merge_regions	Return sequences of matching and conflicting regions.
Method	reprocess_merge_regions	Where there are conflict regions, remove the agreed lines.
Static Method	mismatch_region	Undocumented
Method	find_sync_regions	Return a list of sync regions, where both descendents match the base.
Method	find_unconflicted	Return a list of ranges in base that are not conflicted.
Method	_refine_cherrypick_conflict	When cherrypicking b => a, ignore matches with b and base.

def __init__(self, base, a, b, is_cherrypick=False, allow_objects=False):

Constructor.

Parameters	base	lines in BASE
	a	lines in A
	b	lines in B
	is_cherrypick	flag indicating if this merge is a cherrypick. When cherrypicking b => a, matches with b and base do not conflict.
	allow_objects	if True, do not require that base, a and b are plain Python strs. Also prevents BinaryFile from being raised. Lines can be any sequence of comparable and hashable Python objects.

def merge_lines(self, name_a=None, name_b=None, name_base=None, start_marker='<<<<<<<', mid_marker='=======', end_marker='>>>>>>>', base_marker=None, reprocess=False):

Return merge in cvs-like form.

def merge_annotated(self):

Return merge with conflicts, showing origin of lines.

Most useful for debugging merge.

def merge_groups(self):

Yield sequence of line groups. Each one is a tuple:

'unchanged', lines: Lines unchanged from base
'a', lines: Lines taken from a
'same', lines: Lines taken from a (and equal to b)
'b', lines: Lines taken from b
'conflict', base_lines, a_lines, b_lines: Lines from base were changed to either a or b and conflict.

def merge_regions(self):

Return sequences of matching and conflicting regions.

This returns tuples, where the first value says what kind we have:

'unchanged', start, end: Take a region of base[start:end]
'same', astart, aend: b and a are different from base but give the same result
'a', start, end: Non-clashing insertion from a[start:end]

Method is as follows:

The two sequences align only on regions which match the base and both descendents. These are found by doing a two-way diff of each one against the base, and then finding the intersections between those regions. These "sync regions" are by definition unchanged in both and easily dealt with.

The regions in between can be in any of three cases: conflicted, or changed on only one side.

def _refine_cherrypick_conflict(self, zstart, zend, astart, aend, bstart, bend):

When cherrypicking b => a, ignore matches with b and base.

def reprocess_merge_regions(self, merge_regions):

Where there are conflict regions, remove the agreed lines.

Lines where both A and B have made the same changes are eliminated.

@staticmethod
def mismatch_region(next_a, region_ia, next_b, region_ib):

Undocumented

def find_sync_regions(self):

Return a list of sync regions, where both descendents match the base.

Generates a list of (base1, base2, a1, a2, b1, b2). There is always a zero-length sync region at the end of all the files.

b.m.Merge3(object) : class documentation