regex search/replace

[Sharlene]

Hey guys,

I couldn't find an answer to my question in the forums, so I figured i'd ask directly.

I have an epub that i need to reformat.

A like like this:

<p>&ldquo;Sorry.&rdquo;</p>

Needs to be changed to this:

<p>
&ldquo;Sorry.&rdquo;
</p>

I have to change all of the instances of this in my document, but i'm not sure how to add a new line break.

I typed in find >& and replace it with >\n& but that clearly doesn't work.

Any and all help is appreciated. Cheers.

[WS64]

Sigil will convert whatever you try back to <p>&ldquo;Sorry.&rdquo;</p>
If you really want it that way you can't use Sigil.

[Sharlene]

Ok. Thanks.

[theducks]

Originally Posted by Sharlene

Hey guys,

I couldn't find an answer to my question in the forums, so I figured i'd ask directly.

I have an epub that i need to reformat.

A like like this:

<p>&ldquo;Sorry.&rdquo;</p>

Needs to be changed to this:

<p>
&ldquo;Sorry.&rdquo;
</p>

I have to change all of the instances of this in my document, but i'm not sure how to add a new line break.

I typed in find >& and replace it with >\n& but that clearly doesn't work.

Any and all help is appreciated. Cheers.

There is no difference to the Rendering engine between the 2 versions.

If your intent was to add newlines to the ouput:

<p><br /> &ldquo;Sorry.&rdquo;<br /></p>

[huebi]

Originally Posted by Sharlene

A like like this:

<p>&ldquo;Sorry.&rdquo;</p>

Needs to be changed to this:

<p>
&ldquo;Sorry.&rdquo;
</p>

Why? The Book renderer will show the same result.

[Sharlene]

I know that the output will show the same, however, one of the companies that we are sending our books to says that one of our books didn't pass their system test.

please see below as to the wording that they sent to me (to me it doesn't make any sense, but i've got to fix the file and send it to them):

When content files in the epub contain a paragraph tag immediately
followed by the entity &ldquo; there will likely be an error with the
ACS server packaging. Example instance of this:

<p>&ldquo;Sorry.&rdquo;</p>

In many cases, simply following the paragraph tag with a newline will
avoid the problem:

<p>
&ldquo;Sorry.&rdquo;</p>

However, in other cases, both the beginning paragraph tag and the
ending paragraph tag must be on their own lines:

<p>
&ldquo;Sorry.&rdquo;
</p>

So the best practice for avoiding this problem is to always place
opening and closing paragraph tags on their own lines, in every
content document. For most developers, this is a simple find and
replace operation across all the content files. In cases where we have
these problems, the epub file will have to be edited in this way in
order to work with the ACS server and be successfully processed in our
system.

[DiapDealer]

That's just an insane requirement.

I'm afraid you'll have to accomplish it with some other regex-capable editor. And know that Sigil will just change it back again if you ever edit the epub with Sigil in the future.

[st_albert]

As others have noted, within Sigil, even if you make the changes your client suggests, Sigil will remove the extra linefeeds when you save the file and you'll be back where you started. I don't know how to turn off this behavior.

The hard way to get around this would be, after you are completely done with everything, outside of Sigil, unzip the epub and use some other text editor to make the suggested changes. Save and re-zip. Don't load into Sigil again.

But as kludge, you could avoid the occurrence of <p>&ldquo; by doing a global find & replace as follows (within Sigil):

find: <p>&ldquo;

replace: <p><span>&ldquo;</span>

or something to that effect.

[theducks]

Originally Posted by DiapDealer

That's just an insane requirement.

I'm afraid you'll have to accomplish it with some other regex-capable editor. And know that Sigil will just change it back again if you ever edit the epub with Sigil in the future.

nobody else write code like that. dump those dweebs

[Timur]

If you can not dump them, you can use these patterns in any editor that supports regexps(e.g. Notepad++ in Windows, jEdit in Win/Mac/Linux). Just extract all files and folders from epub, open all the html/xhtml files in the editor, and use replace operation on all opened files:

Find:

^(\s*)(<p[^>]*>)(.*)</p>\s*$

Replace:

\1\2\n\1\3\n\1</p>

[Hitch]

Originally Posted by theducks

nobody else write code like that. dump those dweebs

Ducky:

Some moron is attempting to use Adobe Content Server (to encrypt the file) without knowing HOW it actually works--that's why she's getting those idiotic messages. I have lost track of the countless hours of nonsense crapola we used to experience (back when I started up) from total jackasses at "retail" eBook companies who tried to use ACS, and, not having paid the thousands of dollars to actually learn how to use it, kept getting ingestion errors and blaming the ePUBs.

I've now made it our company policy that if B&N or Amazon has an intake issue, we'll address it. I won't jump through hoops for Apple's insanely bizarre error messages (oh, no! you have a MOBI ISBN typed in your file! You've MENTIONED Amazon! You set a book on the AMAZON River! We'll take you out and shoot you!) or any small-world retailer that doesn't know how to use Content Server.

THAT is why she's getting this amazingly ridiculous error message. I mean, hello? You can't have an OPENING quote following an opening p tag? Moses on a Pony! And people wonder this biz has such a fast burn-through rate...

Hitch