Eclipse Community Forums: Gendoc » Gendoc standard library for parsing and generating documents

Help

Home

Home » Eclipse Projects » Gendoc » Gendoc standard library for parsing and generating documents

Show: Today's Messages :: Show Polls :: Message Navigator

Gendoc standard library for parsing and generating documents [message #1701135]

Thu, 09 July 2015 13:35

Simon Delisle

Messages: 10
Registered: June 2015

Junior Member

Hi,

I have played a bit with Gendoc and I have encountered some problems on the parsing and generating part with LibreOffice.

Have you ever check for a third party standard library to parse the template and generate documents? If yes what are the possible solutions?

Thank you

Report message to a moderator

Re: Gendoc standard library for parsing and generating documents [message #1701701 is a reply to message #1701135]

Wed, 15 July 2015 06:46

Tristan Faure

Messages: 460
Registered: July 2009

Senior Member

Hello we have some users that generate doc for open office.
Maybe there are some cases that are not managed can you share publicly or privately your templates ?

About "Have you ever check for a third party standard library to parse the template and generate documents? If yes what are the possible solutions?" :
> We chose from the first time to not rely on standard library because of the specific process we perform on the document, if we decide to use POI or Tika or other technology we should make a global architecture redesign of gendoc.

Why do you think it would fix your problem in your case ?

Report message to a moderator

Re: Gendoc standard library for parsing and generating documents [message #1701930 is a reply to message #1701701]

Thu, 16 July 2015 19:21

Simon Delisle

Messages: 10
Registered: June 2015

Junior Member

The problem was in the context tag. There was a formatting problem that cause Gendoc to be unable to parse the model path correctly. I have solve this problem by modifying the content.xml by removing some formatting that LibreOffice had placed there.
I think that if GenDoc used a standard library, it could have solved this kind of problem.

Report message to a moderator

Re: Gendoc standard library for parsing and generating documents [message #1702285 is a reply to message #1701930]

Tue, 21 July 2015 10:33

Tristan Faure

Messages: 460
Registered: July 2009

Senior Member

Ok I understand your concern, and sorry you should not have to look inside XMLs.

Actually current added value of gendoc is to remove invalid characters in text. So the problem you had is obviously a bug.
Generally to bypass this problem you have to "remove formatting" on the text of your tag.

To improve gendoc and make it more reliable I see these actions :
- improve the way we recognize tags so if you can reproduce your problem we could add it to our tests
- improve the way errors are reported to users

Nevertheless switching to API like POI would change a lot in the core architecture of ReqCycle if you are interested in the full explanation we can discuss it directly by phone call.

Report message to a moderator

Re: Gendoc standard library for parsing and generating documents [message #1704117 is a reply to message #1702285]

Fri, 07 August 2015 12:20

Yves BERNARD

Messages: 152
Registered: July 2014

Senior Member

Cut then paste (special) as "unformatted text" might solve your problem. I used this workaround successfully with LibreOffice which seems to insert special control sequence when you modify text with some character styles applied (e.g. font, color, ...) that GenDoc cannot handle properly. These sequence are cleared by the cut/paste operation (assuming you paste as "unformated text")

Hope that helps.

Yves

Report message to a moderator

Previous Topic:	How using a Sirius Diagram with GenDoc
Next Topic:	Unexpected paragraph marks

Goto Forum:

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

]

Current Time: Fri Apr 19 16:39:36 GMT 2024

.:: Contact :: Home ::.

Breadcrumbs

Sign up to our Newsletter