XML characters in metadata

[Milica Who]

Hi, this is the first time I'm posting here...sorry if I'm in the wrong place, but I'm not very familiar with how mobile read works (i.e. threads and such).

Anyway I have a problem when editing metadata with Calibre. Actually with showing said metadata in my ebook reader (I use Bookari, a.k.a. Mantano). You see, when I download metadata from, let's say, Amazon, and modify epub, comment (book info) should show in ebook reader, but it doesn't. Eventually, I discovered that the problem is in these xml tags: < and > in content.opf. Mantano won't read description in content.opf that starts with these tags. I have no idea how can I configure calibre not to put these.

Books the comments of which Mantano reads have only <dc:description> and </dc:description> at the beginning and the end , respectively, in content.opf. HTML source (in edit metadata window) of books that have their own description (which Mantano reads) and the books the metadata of which Calibre downloaded shows that tags used are the same (<p class="description"> , <div>, etc.) in both cases. When I erase those xml tags from description section in content.opf and leave only bare text and save changes, Calibre just puts them back. When I erase them in Sigil, they stay erased. So, the only thing that's left is that Calibre is forcing these tags in content.opf.

Is there anyway around this? Without manually editing content.opf of every book. Something in Calibre configuration?

Thanks in advance! And thanks for all the work!

[eschwartz]

calibre auto-HTMLizes all comments, but you can "cheat" and clean out the tags using bulk search & replace.

See this thread for more details: https://www.mobileread.com/forums/sho...d.php?t=268119

Edit: And this thread has a plugin which should also do the job I think: https://www.mobileread.com/forums/sho...d.php?t=271734

[Milica Who]

Thank you so, so much for this!!! I've been searching for a solution on the net for the past three days, but nowhere in the results were those threads...Plug-in you pointed me to worked perfectly!

[eschwartz]

To be fair, I found it myself because I remembered the author of the plugin and did an attachment search.

[Terisa de morgan]

Well, I've removed, at least, the faulty releases

[Milica Who]

Originally Posted by Terisa de morgan

Well, I've removed, at least, the faulty releases

Thank you for this plug-in! I was really considering editing books one by one with Sigil, so to say you saved me a loooot of time would be an understatement really!

[Terisa de morgan]

Originally Posted by Milica Who

Thank you for this plug-in! I was really considering editing books one by one with Sigil, so to say you saved me a loooot of time would be an understatement really!

You're welcome. I didn't think it could be useful really, that's the reason it's only at that thread.

[BetterRed]

They say prevention is better than cure, see ==>> Preferences->Download Metadata->Convert all downloaded comments to plain text.

Originally Posted by Terisa de morgan

You're welcome. I didn't think it could be useful really, that's the reason it's only at that thread.

@Terisa de morgan - any reason why your plugin can't be made official? The above won't fix what's been already downloaded, and some folks might prefer to normalise selectively.

BR

[eschwartz]

Do plaintext downloaded comments lose the

<p class="description">

???

[BetterRed]

@eschwartz - seems so - on the test I did before prior post I didn't see any mark up between <dc:description> and <\dc:description> in content.opf, and I chose a book riddled with bold and italics in its comments.

I only noticed the setting the other day when responding to the issue regarding downloading Authors from B&N.

BR

[eschwartz]

Nice find, then!

[Milica Who]

Originally Posted by BetterRed

They say prevention is better than cure, see ==>> Preferences->Download Metadata->Convert all downloaded comments to plain text.


@Terisa de morgan - any reason why your plugin can't be made official? The above won't fix what's been already downloaded, and some folks might prefer to normalise selectively.

BR

I checked that box before I even attempted to download metadata, and the problem I had with xml tags was still present. Believe me I tried everything before I posted the question, and so far only the plug-in worked...Just now I tried it on a book that already came with its own metadata, but also with those problematic xml tags in content.opf. And the plug-in worked on that book, too.

[Milica Who]

Originally Posted by eschwartz

Do plaintext downloaded comments lose the

<p class="description">

???

In HTML source of metadata they don't...aside from <p class="description">, they still have for example <div> for paragraphs and <br> for line breaks. The problem was that Calibre translated < and > into &lt; and &gt; respectively in the content.opf which should have no tags at all. With plug-in that doesn't happen - there's no <p class="description">, <div> and <br> tags in the description section of content.opf, you just have bare text but separated in paragraphs following the <div> tag pattern of the HTML source. The same goes for line breaks.

[BetterRed]

Originally Posted by Milica Who

I checked that box before I even attempted to download metadata, and the problem I had with xml tags was still present. Believe me I tried everything before I posted the question, and so far only the plug-in worked...Just now I tried it on a book that already came with its own metadata, but also with those problematic xml tags in content.opf. And the plug-in worked on that book, too.

Originally Posted by Milica Who

In HTML source of metadata they don't...aside from <p class="description">, they still have for example <div> for paragraphs and <br> for line breaks. The problem was that Calibre translated < and > into &lt; and &gt; respectively in the content.opf which should have no tags at all. With plug-in that doesn't happen - there's no <p class="description">, <div> and <br> tags in the description section of content.opf, you just have bare text but separated in paragraphs following the <div> tag pattern of the HTML source. The same goes for line breaks.

@Milica - if calibre isn't removing all the markup in your downloaded Comments when the setting is checked then IMO that's a defect that should be rectified. Perhaps you could report the issue at Bugs : calibre with details of a book and metadata source that has the problem.

I did a spot check on one book and the result was fine - the dc:description was free of any HTML markup. I don't use the metadata download feature much, what's more I normally have Comments unchecked.

BTW - I am not suggesting that the setting obviates the need for Terisa's plugin, in fact I've suggested it be made a registered plugin.

BR

[Milica Who]

Originally Posted by BetterRed

@Milica - if calibre isn't removing all the markup in your downloaded Comments when the setting is checked then IMO that's a defect that should be rectified. Perhaps you could report the issue at Bugs : calibre with details of a book and metadata source that has the problem.

I did a spot check on one book and the result was fine - the dc:description was free of any HTML markup. I don't use the metadata download feature much, what's more I normally have Comments unchecked.

BTW - I am not suggesting that the setting obviates the need for Terisa's plugin, in fact I've suggested it be made a registered plugin.

BR

I don't know if those tags, or HTML markup as you called it (I'm not very familiar with these stuff ), are problematic for Mantano e-reader only or for other ebook readers (that have the ability to read description) too. If this problem appears in other ebook readers, too, then I also think it would be good to report a bug. If this issue is only Mantano related, I don't know if that's enough to report a bug...?

I agree, it would still be good to make this official plug-in, regardless of this being a bug or not .