[Libreoffice-bugs] [Bug 121449] New: 6.2 export problems found (bit map import from pdf alters objects order and epub export quality)

bugzilla-daemon at bugs.documentfoundation.org bugzilla-daemon at bugs.documentfoundation.org
Thu Nov 15 19:34:19 UTC 2018


https://bugs.documentfoundation.org/show_bug.cgi?id=121449

            Bug ID: 121449
           Summary: 6.2 export problems found (bit map import from pdf
                    alters objects order and epub export quality)
           Product: LibreOffice
           Version: unspecified
          Hardware: All
                OS: All
            Status: UNCONFIRMED
          Severity: normal
          Priority: medium
         Component: Writer
          Assignee: libreoffice-bugs at lists.freedesktop.org
          Reporter: lbartolome at gmail.com

Description:
I use Writer to create both pdfs for imprints and epubs. My current LO is 5.3.7
and I'm testing 6.2.0.0.alpha1 (x64). 
Since there's no option to export to pdf with trim marks and color bars, I
created a pdf with them on another program from a page with a logo on it. I
covered the logo with a white square before saving this and importing as bit
map on page style.
LO 5 correctly imported it without the logo showing, but it wasn't so with LO
6.2, which also made the lines fuzzy.

Regarding the epub export, I am using Writer2Latex 1.6 extension (writer2html)
in LO 5 and the LO6 default epub export filter. 

Neither of them created a cover for the book even though I selected an image in
the LO 6.2 filter menu. 

LO 6.2 epub export filter only allows to select either new file at header or
page break when it should be able to let you chose whether one or both and at
which header level even if it's not adding additional options as W2L does.

LO 6.2 epub filter didn't export chapter numbers, which W2L handled correctly.

LO 6.2 epub filter added and F in front of the footnote number and added a line
between notes, which make it look pretty weird on ADE. 

And LO 6.2 is not exporting style formating correctly, creating a lot of
garbage code in the epub instead of a clean export. 
I'm copying the output from LO 6.2 epub export filter
<p class="para0"><span class="span0">Rom</span><span
class="span0">a</span><span class="span0">. </span><span class="span0">(41158
palabras)</span></p>
<p class="para1"> </p>
<p class="para1"><span class="span1">Nuestros lectores </span><span
class="span1">desearán</span><span class="span1"> —al menos esa es nuestra
esperanza</span><span class="span1">—</span><span class="span1"> </span><span
class="span1">posponer</span><span class="span1"> por unos instantes la
explicación que va a tener lugar entre P</span><span
class="span1">è</span><span class="span1">trus y Regin</span><span
class="span1">e</span><span class="span1">, a fin de seguir </span><span
class="span1">en su</span><span class="span1"> peregrina</span><span
class="span1">ción a</span><span class="span1"> uno de los héroes de esta
historia, héroe abandonado </span><span class="span1">durante mucho</span><span
class="span1"> tiempo y </span><span class="span1">en el</span><span
class="span1"> que nos parece que </span><span
class="span1">querían</span><span class="span1"> inter</span><span
class="span1">e</span><span class="span1">s</span><span
class="span1">arse</span><span class="span1">. </span></p>


Compare the same lines from W2L export
    <h1 id="toc0"><span class="SectionNumber">CAPÍTULO I  </span>Roma. (41158
palabras)</h1>
    <p class="Standard" lang="fr-FR" xml:lang="fr-FR"> </p>
    <p class="Standard">Nuestros lectores desearán —al menos esa es nuestra
esperanza— posponer por unos instantes la explicación que va a tener lugar
entre Pètrus y Regine, a fin de seguir en su peregrinación a uno de los héroes
de esta historia, héroe abandonado durante mucho tiempo y en el que nos parece
que querían interesarse. </p>

You can see that LO epub export filter is not exporting language information,
which is used by epub readers to show the right dictionary for the words.


Actual Results:
Messy epub
Fuzzy lines in bitmap

Expected Results:
Output should have been at least as good as in previous LO for both things:
1. Handling of pdf imports as bitmaps keeping quality
2. epub filter should have been worked from the best existing filter around
(Writer2Latex's writer2html) to add up to it.
3. If the software had done trim/crop marks and color bars, I shouldn't have
needed another program to create the bitmap.


Reproducible: Always


User Profile Reset: No



Additional Info:

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/libreoffice-bugs/attachments/20181115/3ded05c6/attachment-0001.html>


More information about the Libreoffice-bugs mailing list