`difflib`

Compute deltas between sequences. The core algorithm is Hunt-McIlroy on hashable elements (lines, characters, tokens).

Source-of-record: Lib/difflib.py, difflib docs.

Free functions

Function	Returns
`get_close_matches(word, possibilities, n=3, cutoff=0.6)`	Best matches.
`ndiff(a, b, linejunk=None, charjunk=IS_CHARACTER_JUNK)`	Pretty line-by-line diff.
`restore(sequence, which)`	Recover side 1 or 2 from `ndiff`.
`unified_diff(a, b, fromfile='', tofile='', fromfiledate='', tofiledate='', n=3, lineterm='\\n')`	Unified diff.
`context_diff(...)`	Context diff (same kwargs).

SequenceMatcher(isjunk=None, a='', b='', autojunk=True)

Method	Returns
`ratio()`	Similarity 0.0 to 1.0.
`quick_ratio()` / `real_quick_ratio()`	Upper bounds.
`get_matching_blocks()`	List of `(i, j, n)` triples.
`get_opcodes()`	List of `(tag, i1, i2, j1, j2)`.
`get_grouped_opcodes(n=3)`	Grouped for diff display.
`find_longest_match(alo, ahi, blo, bhi)`	LCS-like longest matching block.
`set_seqs(a, b)` / `set_seq1(a)` / `set_seq2(b)`	Set sequences (caches).

Opcode tags: 'equal', 'replace', 'delete', 'insert'.

Differ(linejunk=None, charjunk=None) produces multi-line diffs with markers: ' ' equal, '-' left only, '+' right only, '?' hint.

HtmlDiff(tabsize=8, wrapcolumn=None, linejunk=None, charjunk=IS_CHARACTER_JUNK).

Method	Returns
`make_table(fromlines, tolines, fromdesc='', todesc='', context=False, numlines=5)`	HTML table.
`make_file(...)`	Full HTML page.

difflib.IS_LINE_JUNK(line) returns True for blank or '#'-only lines. difflib.IS_CHARACTER_JUNK(ch) returns True for space or tab.

Area	State
`SequenceMatcher`, `Differ`	Complete.
`unified_diff` / `context_diff`	Complete; byte-identical output.
`HtmlDiff`	Complete.
Junk filters	Complete.