Apache OpenOffice Community Forum

If a docx file contains a textbox, neither the textbox, nor the text within it, is displayed when opening the docx file in OOo3.4.1.

Observed behaviour
1 Neither text box, nor text within it, is displayed
2 File > Properties > Statistics > Word count = 0
3 Navigator shows the document is empty (eg Text Frames = 0 etc)

Expected behaviour
1 Both text box, and text within it, should be displayed
2 File > Properties > Statistics > Word count should count the words in the textbox
3 Presumably Navigator should show the document is not empty (eg Text Frames = 1?)

A workaround to display the textbox and text is to open the docx in W7 Microsoft Works wordprocessor, or Microsoft Office Word Viewer, proving the textbox and text are present in the docx file.

Several bug reports seem to have been raised about this - see https://issues.apache.org/ooo/show_bug.cgi?id=117910, https://issues.apache.org/ooo/show_bug.cgi?id=93842 and https://issues.apache.org/ooo/show_bug.cgi?id=117980 but no progress seems to have been made resolving the bug.

It seems to me to be a pretty serious compatibility issue that all text in textboxes in docx files is ignored by OOo.

Can anyone suggest how I can try to get this resolved?

Thanks

You can try transferring the file in .doc format. Remember that the vaious MS formats are proprietary, so their filters have been reverse engineered. Despite popular belief, OpenOffice is not an MS Office clone; if support for such formats is of major importance then consideration shoud be given to acquisition and use of MS Office

OOXML file format has not been designed to be compatible with anything else than MS Office. See: MS Office 2007 OOXML file format (docx, xslx, pptx, ppsx).

LibreOffice tries to implement better OOXML filters, you can try it.

RoryOF wrote:You can try transferring the file in .doc format. Remember that the vaious MS formats are proprietary, so their filters have been reverse engineered. Despite popular belief, OpenOffice is not an MS Office clone; if support for such formats is of major importance then consideration shoud be given to acquisition and use of MS Office

Rory,

Seriously? If Apache Open Office is going to survive, it must have MS Word compatibility. Word DOC is dead, dead, dead, just like XP PRO. DOCX is the standard of the world right now and if, for example, I wanted to share a file for editing and commenting with people at work, they are going to expect, no demand, DOCX.

So this is the most serious bug fix that Apache needs to address IMO and if they are not inclined to do so, then OO will slide into the dustbin of almost good enough software.

I had another thought about this.

First, I believe an MS Word Text Box is the same thing as an AOO graphics "Rectangle with text inside".

Second, perhaps the most serious aspect of this is that there is no indication that data is missing from the file.

XML is an eXtensible Markup Language, where the application defines its own tags. If I understand correctly, I can create my own tag <fred> ... </fred> in a file and when my application sees <fred>some data to be formatted</fred>, it formats the data according to what I have defined. Your application does not recognise the tag <fred> ... </fred> so just ignores it, and moves on to the next tag. It is for this reason that trying to keep up with Microsoft continuing changing of its markup is so difficult.

Mind you, even Microsoft seems to find it impossible to read its own XML - I have docx files with text boxes which the Microsoft Word Viewer corrupts! MS WordPad is supposed to handle docx files, but it gives an error on opening docx files I have saying it contains elements that (my) WordPad cannot handle and shows the images but not the text boxes.

It would be extremely useful for AOO to extract all XML statements in a docx file which AOO doesn't recognise, and output them in a separate file so that users knew that data was missing.

Edit: I wonder how difficult it would be to write an add-on which parsed an XML file and separated the "tags recognised by AOO content" and the "tags not recognised by AOO content". It would presumably be fairly easy to keep it up to date as it would just need a "look-up list" of AOO tags

There is ongoing work in progress to improve OO's XML handling; if anyone wishes to comment/be involved in this, please subscribe to the OpenOffice dev list, linked off
http://openoffice.apache.org/mailing-lists.html

Be aware that some of these lists are high traffic lists - perhaps as many as a hundred posts a day - so be wary of subscribing if you do not wish to receive a large volume of email.