Menu
▾
▴
Re: [Betterform-users] problem saving &
|
From: daniele i. <dan...@gm...> - 2011-10-20 08:36:12
|
Hello,
I have tried with dojo.xsl but without success, because for me it's not the
right way to proceed, in fact I have tried also to modify the & in &
before to send the response in betterform servlet so at client side I
receive & but the the browser show &
I have tried it in IE8 and mozilla so it seems strange that is the browser
that has a bug on this otherwise also other website should have the same
problem.... I'm really confusing.
I have tried also another way: before I used a template with encoding UTF-8
and the default browser was ISO-8859-1 I changed it but the same problem.
The strange thing is that output and textarea component are working while
input no.(also with the previous difference of encoding UTF-8 for the xml
stored and ISO-8859-1 for the browser) Do you have any idea? I have
finished mines :(
Thank you
Daniele Ippoliti
2011/10/19 Lars Windauer <lar...@be...>
> Daniele,
>
> html-form-controls.xsl is only utilized in the non-scripted version, hence
> it is out of scope.
>
> To debug the xslt I would utilize xsl:message to cerify the template is
> called as expected.
>
> Further I would enable all logging in log4j.xml and copy the abstract
> XForms UI (you see in the console if you request form by a browser) of the
> console into an XML file and transform this directly using Oxygen.
>
> Hth
>
> Lars
>
> On 19.10.2011, at 17:45, daniele ippoliti <dan...@gm...> wrote:
>
> Hello Joern,
>
> definitely I tried to follow your suggestion regarding the dojo.xsl. But
> apparently if I modify the line 781 in the buildControl putting
>
> <xsl:when test="contains(@mediatype,'text/html') or
> $lname='input'">
> <xsl:value-of select="bf:data/text()"
> disable-output-escaping="yes"/>
> </xsl:when>
>
>
> instead of
>
> <xsl:when test="contains(@mediatype,'text/html')>
> <xsl:value-of select="bf:data/text()"
> disable-output-escaping="yes"/>
> </xsl:when>
>
> nothing happens to the input field.
>
> I have also restarted a server because I saw that you use a sort of cache
> for the xsl file transformers
>
> Do you have some idea? Which is the difference between dojo.xsl and
> html-form-controls.xsl ? it seems that they have the same scope to transform
> the xform in html and this make me confusing.... :-S
>
> Thank you for your help
>
> daniele ippoliti
>
> 2011/10/19 daniele ippoliti <dan...@gm...>
>
>> or maybe the quick fix could be transfor the outputStream received from
>> the generateUI method in the class WebProcessor.?
>>
>> thank you
>>
>> Daniele Ippoliti
>>
>>
>> 2011/10/19 Joern Turner <joe...@be...>
>>
>>> output behaves like this as its simply html output so the standard
>>> browser behavior applies. My idea is that it might have to do with
>>> 'disable-output-escaping' in dojo.xsl when the values are written to input
>>> or textarea - but that's just an idea. But i would first look into the xslt.
>>>
>>>
>>> Am Wed Oct 19 09:33:31 2011 schrieb daniele ippoliti:
>>>
>>>> ok Joern,
>>>> I can try to fix the problem, but I need some help to do this,
>>>> apparently seems that for xf:output the transformation works infact I'm able
>>>> to see the & and not the ASCII code. could you tell me where in betterform I
>>>> can start to see the trasnformationtion from xf:output to the real html, in
>>>> this way I'm going to compare that structure on that one that you have for
>>>> xf:input and xf:textarea.
>>>>
>>>> thank you for your help
>>>> Daniele Ippoliti
>>>>
>>>> I can try to fix it, I want just to
>>>>
>>>> 2011/10/18 Dave Finton <dav...@gm... <mailto:
>>>> dav...@gm...>**>
>>>>
>>>>
>>>> I just tried unencoding the & string via the eXist function
>>>> util:parse-html(), and while it did succeed in translating the
>>>> "&" to a "&", I ended up getting the following error:*
>>>>
>>>> XPST0003 : Ampersands (&) must be escaped.*
>>>>
>>>>
>>>> So, this means it is not possible to manually unencode the string
>>>> on the server side. I have not tried client-side.
>>>>
>>>>
>>>> On Tue, Oct 18, 2011 at 9:53 AM, Dave Finton
>>>> <dav...@gm... <mailto:dav...@gm...>**> wrote:
>>>>
>>>> So, if I understand correctly, if you load the text with the
>>>> & into a betterForm field (like xf:input or xf:textarea),
>>>> then the unencode does not happen? Hang on just one second...
>>>>
>>>> ... 2 minutes of testing later ...
>>>>
>>>> Ah HA! I've reproduced the problem! I my case, when I put "&"
>>>> in an xf:input form (in this example the search bar), the
>>>> model is used to automatically fill the same field when the
>>>> next page loads. Instead of seeing "&" as I would expect, I
>>>> get the full "&" sequence in my field. I am using the
>>>> latest development branch of betterForm.
>>>>
>>>> So, at this point, I'd need to defer to the actual developers
>>>> of betterForms because I would not know how to fix this
>>>> problem. :D
>>>>
>>>> Joern and Lars, is there by chance a quick and easy way to
>>>> correct this behavior? I figure we're either looking at an
>>>> unimplemented feature ;-) or maybe a missing flag that we
>>>> aren't utilizing correctly (or at all). If you folks have any
>>>> guidance, I would be interested in a solution as well.
>>>>
>>>>
>>>> On Tue, Oct 18, 2011 at 9:30 AM, daniele ippoliti
>>>> <dan...@gm... <mailto:dan...@gm...>> wrote:
>>>>
>>>> hello,
>>>> thank you for your support :). let's say that a user is
>>>> filling a forum that use xform he fills it and he saved a
>>>> content of one field with an '&' character (this character
>>>> thanks to the transformation has been stored like &).
>>>>
>>>> The same user want to modify the file that he has defined
>>>> previously,but at this time, he will see instead of &
>>>> ---> & in the UI.
>>>>
>>>> For me is in betterform that is missing the second way
>>>> transformation. I want to say:
>>>> 1) One side & ---> & (perfect I agree is totally
>>>> necessary).
>>>> 2) Second side & ---> & is missing.
>>>>
>>>> for the point 2) on my side (client of betterfom) I should
>>>> write a javascript (with the DOM) that field by field goes
>>>> to parse the content in order to have & shown instead of
>>>> &, this is very unefficient, because the
>>>> transformation should be done at the moment of the
>>>> trasnformation from xf:input or xf:textarea or... (all the
>>>> field that could contain & ) to HTML, because you will
>>>> have one entry point, I don't know the internal structure
>>>> of betterform but you should have a transformation from
>>>> xf:input declared by the developer to HTML (div, span...)
>>>> and if you apply here the function to remove all the ASCII
>>>> codes with the characters you will invoke just one method
>>>> of conversion to have the job done for all the xf:input
>>>>
>>>> something like this (pseudocode)
>>>>
>>>> xfInputTransformation(){
>>>>
>>>> ...
>>>>
>>>> out.write("<div id.....");
>>>> out.write("<input id.....>");
>>>> out.write(**convertSpecialCharacter(**contenInput));
>>>> out.write("</input .>");
>>>>
>>>> ...
>>>> }
>>>>
>>>> Am I totally wrong ?
>>>>
>>>> Thank you
>>>>
>>>> Daniele
>>>>
>>>>
>>>>
>>>>
>>>> 2011/10/18 Dave Finton <dav...@gm...
>>>> <mailto:dav...@gm...>**>
>>>>
>>>>
>>>>
>>>> I agree with you that is necessary the
>>>> transformation on one side, when you submit in a
>>>> resource your instance to avoid XML error and this
>>>> has been done by betterform, but I thought that if
>>>> the user want fill in the form and insert & he
>>>> wants to see & in the next consultation of the
>>>> file, so should be under the responsability of
>>>> betterform/xform to do the oposite translation at
>>>> reading time and not only at writing time or not?
>>>>
>>>>
>>>> I think I may be confused on this bit. When you say
>>>> "next consultation of the file". In what kind of
>>>> context do you mean? If you're storing the input data
>>>> in HTML or XML format, then the escape sequences very
>>>> much need to stay intact. A stray "&" or ">" character
>>>> will destroy the structure of the markup, and cause
>>>> errors both for web browsers or XML parsers on the
>>>> next read.
>>>>
>>>> I'm probably mixing up the meaning of your last reply,
>>>> so a clarification would be most welcome. Thanks!
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> -- David Finton
>>>>
>>>>
>>>>
>>>>
>>>> -- David Finton
>>>>
>>>>
>>>>
>>>>
>>>> ------------------------------**------------------------------**
>>>> ------------------
>>>> All the data continuously generated in your IT infrastructure contains a
>>>> definitive record of customers, application performance, security
>>>> threats, fraudulent activity and more. Splunk takes this data and makes
>>>> sense of it. Business sense. IT sense. Common sense.
>>>> http://p.sf.net/sfu/splunk-**d2d-oct<http://p.sf.net/sfu/splunk-d2d-oct>
>>>>
>>>>
>>>> ______________________________**_________________
>>>> Betterform-users mailing list
>>>> Betterform-users@lists.**sourceforge.net<Bet...@li...>
>>>> https://lists.sourceforge.net/**lists/listinfo/betterform-**users<https://lists.sourceforge.net/lists/listinfo/betterform-users>
>>>>
>>>
>>>
>>>
>>
>
> ------------------------------------------------------------------------------
> All the data continuously generated in your IT infrastructure contains a
> definitive record of customers, application performance, security
> threats, fraudulent activity and more. Splunk takes this data and makes
> sense of it. Business sense. IT sense. Common sense.
> http://p.sf.net/sfu/splunk-d2d-oct
>
> _______________________________________________
> Betterform-users mailing list
> Bet...@li...
> https://lists.sourceforge.net/lists/listinfo/betterform-users
>
>
|