Hello Regina, Thank you for the clarification so far. Let's try to make this complex feature more obvious and easier understandable to ODF spec users. The better we explain it, the easier becomes ODF to be implemented. As a definition of our goal, I would like to quote my earlier mentor: "Whenever I have to slow down my reading-velocity the text is too complex!" ;) Now when I look up @draw:textarea-vertical-align , the specification reads: " The draw:textarea-vertical-align attribute specifies the vertical alignment of the text area inside a shape." There is yet no reference to any logical objects representing "text area" or "shape". Searching the string "text area" does not reveal a <text:p>, but lead me only to the <draw:text-areas> element. You mentioned other complex relations, that might need clarification, but how about starting with this one? How could we improve the description of the feature you presented? From my perspective: Adding references to some chapter within the spec to define "text area" and "shape" sounds like a good start, right? Kind regards, Svante ? 2017-11-08 1:46 GMT+01:00 Regina Henschel <
regina.henschel@libreoffice.org > : Hi Svante, Svante Schubert schrieb am 07.11.2017 um 23:24: Thank you for the example document, Regina. I have opened it in LibreOffice Impress and Powerpoint 2016 and looked into the XML. While PowerPoint does not know how to handle the text, LibreOffice is showing it but aligning it nicely. Regarding the alignment, I got a question. While I understand that the text is grouped easily within the images of slide 1, where does the XML state that the two paragraphs on slide 3 are within the lower part of the image. That is the attribute draw:textarea-vertical-align=" bottom" in the <style:graphic-properties> element of the style "gr6". I have unpacked the image to find out that the lower white part is part of it. LibreOffice would not need this extra part, because you can use a negative padding in LibreOffice. But negative values are not allowed for padding. Another way to get the text down is to use draw:textarea-vertical-align=" middle" and a value for fo:padding-top, that is larger as the image height. But that construct looks strange to me too. A valid way to get a text outside the area of the image is to use a negative value in the rect() value of the fo:clip attribute. The part which goes beyond the image is transparent and therefore the text might conflict with the background. For a solid colored slide, that is surely the correct method. Here I have enlarged my picture, because it was save and quick. It only puzzles me, that the first paragraph of slide 3 has the same style as all paragraphs within the image on slide 4, but they are totally different aligned within the image. The alignment is done with attribute fo:text-align. The paragraph with the URL on slide 3 has its own paragraph style P8, and there you find fo:text-align="end". The paragraphs on slide 4 have the paragraph style P7. The green color of the text on slide 3 belongs to the text-properties. Besides the alignment you need to look at the values for padding. The alignment on slide 4 is a combination of the draw:textarea-vertical-align=" top" and draw:textarea-horizontal-align ="left" in the graphic-properties of style gr7 with the paragraph alignment in style P7. You can think as if the entire set of paragraphs are inside a rectangular box and this box is attracted by the edges of the image, if a draw:textarea-xxx-align value is set. The value "justify" in style gr6, which belongs to slide 3 means, that the box is strechted to the right and to the left edge. That is the setting "Full width" in the UI of LibreOffice. Because of that there is a visible difference between the right aligned paragraph with the URL and the green left aligned paragraph. The values "left" and "top" in the style of slide 4 bind the box to the top-left of the image. The width of the box corresponds to the maximal width of the paragraphs. Likely I am missing the obvious, but perhaps you (or someone who might know) might speed up my search. It is not "obvious". The interaction between draw:textarea-xxx-align, fo:text-align, fo:padding and fo:clip is complex. Kind regards Regina ? 2017-11-07 19:50 GMT+01:00 Regina Henschel <
regina.henschel@libreoffice.o rg <mailto: regina.henschel@libreo ffice.org >>: Hi all, here are some examples for using text on images. The presentation advances with click. Kind regards Regina ------------------------------ ------------------------------ --------- To unsubscribe from this mail list, you must leave the OASIS TC that generates this mail. Follow this link to all your TCs in OASIS at:
https://www.oasis-open.org/app s/org/workgroup/portal/my_work groups.php <
https://www.oasis-open.org/ap ps/org/workgroup/portal/my_wor kgroups.php >