On 04/02/12 11:37, lkcl luke wrote: > On Sat, Feb 4, 2012 at 11:02 AM, Phil Charlesworth > <[email protected]> wrote: > >> On 04/02/12 01:34, C Anthony Risinger wrote: >> >>> On Fri, Feb 3, 2012 at 4:39 PM, Stef Mientki<[email protected]> >>> wrote: >>> >>> >>>> On 03-02-2012 19:45, Phil Charlesworth wrote: >>>> >>>> >>>>> I have posted Issue 681 about this problem: >>>>> >>>>> http://code.google.com/p/pyjamas/issues/detail?id=681 >>>>> >>>>> Please see the issue for details. >>>>> >>>>> >>>> Somewhat releated problem, >>>> diacritic characters are ok in PYJS, but wrong in PYJD ((IE). >>>> >>>> I had the same effect in some other situation, >>>> and found by making a type, >>>> that IE handles the character encoding wrong. >>>> >>>> The correct way to specify encoding in the web is: >>>> <meta http-equiv="content-type" content="text/html; charset=UTF-8"> >>>> which doesn't work correctly in IE >>>> >>>> by making the following type >>>> <meta http-equiv="content-type" content="text/html;" charset=UTF-8> >>>> the encoding works in Mozilla, Chrome and IE >>>> >>>> >>> html5 boilerplate: >>> >>> https://github.com/h5bp/html5-boilerplate/blob/master/index.html >>> >>> ... uses: >>> >>> <meta charset="utf-8"> >>> >>> ... and i read in other h5bp discussions that IE needs that statement >>> within the first 512 bytes. >>> >>> per: >>> >>> http://www.w3schools.com/html5/att_meta_charset.asp >>> >>> ... and elsewhere, the `charset` attribute is proper html5-way of >>> doing it, and should be used moving forward -- the http-equiv-ness is >>> legacy. i'd opt for the html5, since that's in front of us (and >>> probably the reason IE works in that case). >>> >>> this is one of at least 5-10 items we should blatantly copy from the >>> h5bp template. >>> >>> >>> >> I don't think this is anything to do with what encoding the browser is >> using - it's to do with the fact that Python is using the ascii codec by >> default and there isn't an easy way to change that. >> I found lots of stuff on the internet about the problem and the link >> below struck me as particularly apposite. >> >> http://blog.webforefront.com/archives/2011/02/python_ascii_co.html >> >> I have >> # -*- coding: utf-8 -*- >> at the head of the python module, so I can be sure what I'm getting in >> unicode but I stll need to circumvent the str() calls in setText() and >> getText(), which I am doing (as a workaround) by >> content = getattr(self.txa.getElement(),'value') #for getText() >> DOM.setAttribute(self.txa.getElement(), "value", content) # >> for setText() >> where self.txa is a TextArea widget >> > the thing is that i'm reluctant to change something as fundamental as > DOM.setAttribute without analysing its full impact across *all* the > pyjd engines. > > anything like this is going to require comprehensive and thorough testing. > > l. Fair enough P.
