<html> <head> <base href="https://bugs.documentfoundation.org/"> </head> <body> <div> <a class="bz_bug_link bz_status_NEW " title="NEW - Set language for multiple paragraphs only sets it for the last selected one" href="https://bugs.documentfoundation.org/show_bug.cgi?id=121431#c4">Comment # 4</a> on <a class="bz_bug_link bz_status_NEW " title="NEW - Set language for multiple paragraphs only sets it for the last selected one" href="https://bugs.documentfoundation.org/show_bug.cgi?id=121431">bug 121431</a> from <a class="email" href="mailto:adalbert.hanssen@gmx.de" title="Adalbert Hanßen <adalbert.hanssen@gmx.de>"> Adalbert Hanßen</a> <pre>Today I tested the bug again in LibreOfficeDev Version: 6.4.0.0.alpha0+ Build ID: 0328e0af9532d5ab26840cf58b9dcbb69bb8c33d CPU threads: 4; OS: Linux 4.4; UI render: default; VCL: gtk3; TinderBox: Linux-rpm_deb-x86_64@86-TDF, Branch:master, Time: 2019-06-28_14:36:08 Locale: de-DE (de_DE.UTF-8); UI-Language: en-US Calc: threaded I marked a text portion spanning several paragraphs. I used Tools>Language>for Paragraph. No language carried a checkmark at that state of my testing. Then I continued such that finally I checked Tools>Language>for Paragraph>English (USA). But this did not change the language: Everything was still underlined with red curly lines indicating typing errors. I checked, if the checkmark for the paragraphs being English (USA) was set: No, it was not, no checkmark on any language in Tools>Language>for Paragraph. Then I repeated my trial with the selection only spanning one paragraph or less and that worked such that the content of the whole paragraph was turned to English (USA). After this, I checked if the checkmark for English (USA) was set fpr that paragraph: No, it was not set. No checkmark at any language in Tools>Language>for Paragraph. Trying to find more about this glitch, I checked several parts of the paragraph to which I had applied Tools>Language>for Paragraph>English (USA), but I checked the properties of the marked test portions with Tools>Language>for Selection. This showed me that the whole content of the paragraph had been turned to English (USA), but as characters, not as a paragraph. Preliminary Conclusion: a) EITHER the language flag is a property of characters rather than a property of paragraphs. b) OR there is a language flag which can be associated to paragraphs, but it is handled by setting this flag to the characters in the selection and the whole surrounding paragraph in which it is located. I don't know enough about the underlying data structures. At least, the function is not properly implemented in the case for a selection spanning more than one paragraph if a) is true and the option to apply a language to a paragraph is meant as "extending the selection to the involved paragraphs first, if necessary". My proposal aa) for case a): In this case the option to assign a language for a paragraph should EITHER be removed, since there is nothing to associate a language to a paragraph OR ab) (still for for case a): it should be implemented to also work, if the selection spans more than one paragraph: extending to all associated paragraphs. bb) for case b): If the data structures are such that they apply a language to characters and to paragraphs, the hierarchy of those settings should be explained in the help function and the function should be applied accordingly. The current user interface at this point is just anti-logical (since it works for less than one paragraph extending but does not work at all, if the selection extends across paragraph bounds). It just surprises the user. Sometimes he is happy, sometimes he is unhappy. But the user does not know about it ahead (unless he has spent as much time as I did to find out when it works and when it doesn't). Probably removing the option to set the selection for paragraphs is the quickest one to implement. It has the advantage to make the program more logical to use and less "mystic exceptions" to learn for a beginner.</pre> </div> <hr> You are receiving this mail because: <ul> <li>You are the assignee for the bug.</li> </ul> </body> </html>