Hi,
I'm using lxml's html diff functionality in a project and it has been working
well so far. Sometime back though I
noticed that the generated diff changes the structure of the html in a manner
that's less than ideal for our use case.
An example:
>>> from lxml.html import diff
>>> a = "<div id='first'>some old text</div><div id='last'>more old text</div>"
>>> b = "<div id='first'>some old text</div><div id='middle'>and new
>>> text</div><div id='last'>more old text</div>"
>>> diff.htmldiff(a, b)
('<div id="middle"> <div id="first"><ins>some old text</ins></div><ins>and
new</ins> <del>some old</del> text</div><div
id="last">more old '
'text</div>')
>>>
As you can see, the div with id=middle has been inserted at the beginning of
the document and it encloses the div with
id=first. I believe this happens because lxml unconditionally inserts
'unbalanced tags' at the beginning of set of
'chunks' when it surrounds the chunks with the <ins> tags:
https://github.com/lxml/lxml/blob/master/src/lxml/html/diff.py#L241
Could we potentially be a bit smarter about this and insert the unbalanced
tags, as we encounter them instead ? For
instance, closing out the opened `<ins>` tag, inserting the unbalanced tag and
opening a new `<ins>` tag. Something
like:
Secondly, it would be great if someone could help me understand (or give me
pointers to) why I'm seeing this differences
of behaviour between what's being returned from the compiled version of the
code and executing the exact same set of
functions from the REPL !?!
(assuming the same context in the REPL as above)
>>> tokens_a = diff.tokenize(a)
>>> tokens_b = diff.tokenize(b)
>>> diff.htmldiff_tokens(tokens_a, tokens_b)
['<div id="middle"> ', '<ins>', '<div id="first">', 'some ', 'old ', 'text',
'</div>', 'and ', 'new', '</ins> ',
'<del>', 'some ', 'old', '</del> ', 'text', '</div>', '<div id="last">', 'more
', 'old ', 'text', '</div>']
>>> s = diff.InsensitiveSequenceMatcher(tokens_a, tokens_b)
>>> commands = s.get_opcodes()
>>> list(commands)
[('delete', 0, 9, 0, 0)]
The htmldiff_tokens() is obviously getting 2 opcodes, either an insert or
replace, followed by a delete, but if I
instantiate InsensitiveSequenceMatcher() from the REPL, it generates only the
delete ! This is driving me nuts ! Am I
doing something wrong ?
cheers,
Steve
_______________________________________________
lxml - The Python XML Toolkit mailing list -- lxml@python.org
To unsubscribe send an email to lxml-le...@python.org
https://mail.python.org/mailman3/lists/lxml.python.org/
Member address: arch...@mail-archive.com
