textwrap.wrap: new argument for more pleasing output

[82d5d3e7-0df6-496f-8325-d2e5a662bf01]

BPO	12485
Nosy	@birkenfeld, @merwok, @bitdancer, @wiggin15
Files	textwrap.py-beautiful-2011-07-11_22-01-31_r71296+.diff: Addition of beautiful algorithm to textwrap.py Issue12485.v2.patch: Addition of beautiful algorithm to textwrap.py with tests

^{Note: these values reflect the state of the issue at the time it was migrated and might not reflect the current state.}

Show more details

GitHub fields: ```python assignee = None closed_at = None created_at = labels = ['type-feature', 'library'] title = 'textwrap.wrap: new argument for more pleasing output' updated_at = user = 'https://bugs.python.org/parent5446' ``` bugs.python.org fields: ```python activity = actor = 'wiggin15' assignee = 'none' closed = False closed_date = None closer = None components = ['Library (Lib)'] creation = creator = 'parent5446' dependencies = [] files = ['22626', '38978'] hgrepos = [] issue_num = 12485 keywords = ['patch', 'needs review'] message_count = 7.0 messages = ['139731', '139785', '139825', '139827', '139880', '140170', '240827'] nosy_count = 6.0 nosy_names = ['georg.brandl', 'eric.araujo', 'r.david.murray', 'wiggin15', 'parent5446', 'gavanderlinden'] pr_nums = [] priority = 'normal' resolution = None stage = 'test needed' status = 'open' superseder = None type = 'enhancement' url = 'https://bugs.python.org/issue12485' versions = ['Python 3.5'] ```

[82d5d3e7-0df6-496f-8325-d2e5a662bf01]

Python's textwrap module can be helpful at times, but personally I think there are a couple of things that could be added.

First, when it comes to text wrapping, usually you're not dealing with a monospace font where each letter is the same size. If you're working with the Python Imaging Library for example, there is a function that you pass the text to in order to determine how wide (or tall) a font is. Therefore, it would be useful to have a parameter where the user can pass a function that gives a custom width for a set of text. The default for this parameter, of course, would be len.

Also, this module uses a rough and efficient algorithm for wrapping text, but the results are not always aesthetically pleasing (one word hanging off on a line). Sometimes the user may want something that is wrapped more beautifully, so to say, such as is found in TeX. So there should also be a beautiful option that goes back and redistributes the text so that it is more aesthetically pleasing.

This isn't exactly that important (minor improvements to a module that is probably not used much), but I figured I'd get it out there as I run into the problem all the time when trying to wrap text to be put in images of a set size.

[merwok]

Hi! Thanks for the report and patch. Are your two requests related? If not, it would be best to open two reports.

[82d5d3e7-0df6-496f-8325-d2e5a662bf01]

Nah, they're both unrelated. I'll separate the changes and remake the patches. (I'll keep this entry for the beautification part.)

[82d5d3e7-0df6-496f-8325-d2e5a662bf01]

OK, so here is the patch for just the new algorithm.

[merwok]

xrange does not exist in Python 3, it’s called range. You should have seen yesterday that I changed the versions: as a new feature, this cannot go into stable releases, only into the next one.

I’m adding Georg to nosy per http://docs.python.org/devguide/experts

[82d5d3e7-0df6-496f-8325-d2e5a662bf01]

OK, sorry to get back so late, but here's the updated patch without xrange. I saw the version change but forgot that I used xrange in the function (old habits I guess).

[c01133da-2897-42c8-8a27-dc35d72938a9]

Updated the patch and added tests. Fixed a problem with the previous patch: result of map function was assumed to be list, however map in Python3 returns an interator. So I replaced it with a list comprehension.