Closed GoogleCodeExporter closed 9 years ago
Hi! Concerning the problem of "missing fonts", I think "bashley101" is wrong.
No software can include all the fonts existing on earth : this is a real
problem because a PDF file may reference any possible font without embedding
it. In this case, the book is not readable with the original fonts. This
problem is well known, but most ebook readers ignore it (even the well-noted
Moon+ ! ). As far as I know, Ebookdroid is the only software which warns you if
the original font is missing and which proposes TWO solutions: either copying
the missing font somewhere into your device, or (following my own suggestion a
long time ago) using by default a replacement font that belongs to the same
family. This leads to a not perfect but similar rendering! Very well done!
Original comment by philippe...@gmail.com
on 30 May 2014 at 6:43
Yes. you are right. That message indicates that rendering of the document will
differ from authored.
EBookDroid always uses some font substitution if there are no exact font found,
but sometimes it produces strange results. For example: font ArialMT uses
different character encoding than regular Arial (Helvetica), so document will
be gibberish. But if you put arialmt.ttf (from your Windows installation of
some font sites in internet) to /sdcard/fonts (or any other configured in
settings folder) or just in the same folder as PDF document, it will be
rendered correctly.
Now, concerning "decoder restarts". As you know, EBookDroid uses third party
code in the PDF rendering engine. We stated this in about and on main page of
this site. Unfortunately on some documents that code hangs. On older version of
EBookDroid when we used that code as library that cused whole EBookDroid
application to stop responding or crash. When we switched to using engines as
separate processes it not only allows us to separate EBookDroid to close source
part and GPL plugins, but allows us to kill and restart rendering engines even
if they totally hangs.
You can tune timeout in the performance settings after which EBookDroid decides
that decoder hangs. For some files (for example scanned PDFs with huge pages,
runned on non-top devices) this timeout should be set to higher value. If you
have a powerful device and do not want to wait for pages you can set it to
lower value.
Please note that "white page" means that request for rendering was sent to
engine and not yet received and you should wait some time. If engine three
times failed to render image that page will be shown as "gray page".
So, both of your reported issues not a bugs but a designed behaviour. That's
why I am marking this issue as "WontFix".
If you do not like current behaviour, you can try to describe how do you want
to change it and can investigate what can be done.
Original comment by Andrei.K...@gmail.com
on 30 May 2014 at 7:21
Original issue reported on code.google.com by
bashley101
on 30 May 2014 at 12:18