Closed GoogleCodeExporter closed 8 years ago
Hi,
> 2. We want to style a text and also to support carriage return (new lines)
from a HTML textarea input field.
I have decided to remove '\n' and '\r' because it's, because it's HTML ignore
those caracters. Ex: if you create an simple HTML file :
---------------------
<html>
<body>
a
b
</body>
</html>
---------------------
Browser (FF, Chrome, etc...) displays :
---------------------
a b
---------------------
If you see the text syling demo webapp
(http://xdocreport.opensagres.cloudbees.net/textStyling.jsp), it uses CKEditor
which generates '\n', '\r' and it must be ignored.
> a) <p>any text</p> causes a leading blank new line, that could not removed?!
When you set a field with html text styling in docx or odt, you have (for odt)
<p:text>$field</p:text>
So if $field="a b", it will generate :
---------------------
<p:text>a b</p:text>
---------------------
if $field="<p>a b</p>", it will generate
---------------------
<p:text></p:text>
<p:text>a b</p:text>
---------------------
It's the case you describe. So why do you not use "a b" instead of <p>a b</p>?
>b) In the Word document the paragraph style will not used from the MergeField
style because of this spare paragraph
Could you tell me a sample (docx+java code) please?
>c) A <br/> tag is not supported without a paragraph change
Could you tell me a sample (docx+java code) please?
> We are interested to use this tool. In witch classes and jars is the text
styling with HTML implemented?
1)
http://code.google.com/p/xdocreport/source/browse/document/fr.opensagres.xdocrep
ort.document/src/main/java/fr/opensagres/xdocreport/document/textstyling/html/HT
MLTextStylingContentHandler.java it a SAX Handler which parse HTML and call the
well methods of generic documentHandler
2)
http://code.google.com/p/xdocreport/source/browse/document/fr.opensagres.xdocrep
ort.document/src/main/java/fr/opensagres/xdocreport/document/textstyling/html/HT
MLTextStylingTransformer.java HTML text styling which ignore the '\r, '\n' and
call the SAX Handler HTMLTextStylingContentHandler
>There is also a issues with processing (German) umlauts like ä,ö,ü used via
ä >ü etc.. I think the French accent aigu will cause the same defective XML
and >DOCX files?!?
Yes you are right, the demo webapp failed with 'é'. Please create a separate
issue for that. Many thank's
Regards Angelo
Original comment by angelo.z...@gmail.com
on 4 Jul 2012 at 1:00
Hi Angelo, thanks for your quick response!
We also use CKEditor and I think other RichText2HTML Editors will generate HTML
Code like this:
<p>First Line and Return</p>
<p>Second Line and Shift+Return<br />
Thirst Line and Return</p>
<p>Last Line and nothing else</p>
XDocReport generates (in ODT and DOCX):
An additional Blank Line
First Line and Return
Second Line and Shift+ReturnThirst Line and Return
Last Line and nothing else
CKEditor (and others) cannot omit only the first <p> element, so everyone who
uses Text Styling have to code a workaround.
Would it make sense to detect leading <p> element and to omit the blank new
line inside XDocReport?
The same behaviour is if you are using the Wiki syntaxes or the <h1> element,
see here
http://xdocreport.opensagres.cloudbees.net/processReport?dispatch=load&reportId=
ODTTextStylingWithFreemarker.odt
I think b) will be caused by a). I'm not sure, but actually that is not so
important.
For c) That the <br> element will be ignored is shown above too.
Thanks for your additional information. I will have a look and add an new issue
for umlauts.
Best Regards,
Alexander
Original comment by googlec...@norz.de
on 4 Jul 2012 at 1:51
Hi Alexander,
I think the problem that you have is
1) <br> is ignored.
2) extra blank for new line
So if <br> is managed I think you will have not problem and \n must be ignored
as today.
> For c) That the <br> element will be ignored is shown above too.
I have created a new isse to manage <br /> =>
http://code.google.com/p/xdocreport/issues/detail?id=130
Omit the blank new line is very difficult (I could explain more if you wish),
but I think we could omit it for simple cas like this :
------------------------------------
<w:p>
<w:t>$html</w:t>
<w:p>
------------------------------------
but not with $html a
> I will have a look and add an new issue for umlauts.
I have seen that (http://code.google.com/p/xdocreport/issues/detail?id=129) and
I have dixed the problem for the 0.9.8 which will come soon.
Original comment by angelo.z...@gmail.com
on 8 Jul 2012 at 8:37
Original comment by angelo.z...@gmail.com
on 28 Sep 2012 at 9:34
Original issue reported on code.google.com by
googlec...@norz.de
on 4 Jul 2012 at 12:10