Files with <a...></a> Tags Don't Beautify

[enuddleyarbl]

I run into this quite a bit. It might be with other tags as well, but it looks like files that include <a...></a> tags don't want to beautify (Tools > Beautify All Files) in the Editor. Nothing seems to happen when I try. Is that expected behavior?

The following is a Scrambled version of part of a book I'm messing with right now. Supposedly, it's been beautified. But, the heading and paragraph don't even get indented. Just removing the leading <a> tag pair lets the indentations happen.

<a id="GBS.0245.01"></a>

  <section class="aboutauthorpage">

<h2><span epub:type="pagebreak" id="pg_2" title="2"></span>Zqpsi zhs Cn*jmxp</h2>

<p class="center"><img src="../images/epub_Scalzi_2016_no_cr_fmt.jpeg" alt="Photo of John Scalzi" class="imgsetwidth"/><a id="GBS.0245.02"></a></p>

<p><a href="http://us.macmillan.com/author/johnscalzi?utm_source=ebook&amp;utm_medium=adcard&amp;utm_term=ebookreaders&amp;utm_content=johnscalzi_authorpage_macdotcom&amp;utm_campaign=9780765388926"><span class="smallcaps">Yxty Tojzak</span></a> pf evu xd okf moam nae*k*qhk vzs di*uaticmp BC ik*wwauq vo yywwat xh thk qgvi dncwuj. Vzs ilz*nbrajs xcl*efmk*jyg oy*wqa, <i>Bmc Aab’v Bgl</i>, qir rgx dco*ijqq sqf*wson’e Ndoo Z. Zdjg*hdun Elzvn cym Lnvn Ktj Ncsmze. Lxj <i>Hkx Zqfo Krbbk</i> jibv*bacwssn bs*dyzsn <i>Wjx Wzrf Ebpmjq; Qgxyr Pw*piwv; Ark*ffrsmi</i>, gkyxt bct 5372’y Ybsv Bvofl isp Sbsk Uaise; wnk <i>Ctkm Xo</i>. Ql*cy*gzqy nsiv lhb uoehbm wmnc qcap, <i>Izen*idyk</i> (<a href="http://whatever.scalzi.com">mpan*mctr.vdbmmh.rnn</a><a id="GBS.0245.03"></a>), pgi oqeb sdgcnr oyr hwi scvom Kvbh Icaphd. Bk taobr ps Wfbk mtnf qfy sqoy anj upaco*ywl. Ysw tda gcno mh ckz bzekb jn*saewk <a href="http://us.macmillan.com/authoralerts?authorName=johnscalzi&amp;authorRefId=1351591&amp;utm_source=ebook&amp;utm_medium=adcard&amp;utm_term=ebookreaders&amp;utm_content=johnscalzi_authoralertsignup_macdotcom&amp;utm_campaign=9780765388926">beci</a>.</p>

<a id="GBS.0246.01"></a><a id="GBS.0246.02"></a>
</section>

</body>

[jackie_w]

I think <a> tags are supposed to be enclosed in a block tag, e.g. <p> or <div>. Correcting them will probably make Beautify work properly. I've often seen "detached" <span> tags similarly misused in epubs.

In your particular example those Google Books epub <a> tags are usually worthless (IMO) so just removing them will also make the Beautify problem go away.

[enuddleyarbl]

I'd been wondering what those "GBS..." tags were. Yes. A Google Books book. I'll probably just delete the lot of them. But, this happens in quite a few books where they're just trying to put in dead-tree page numbers. I also see them with simple non-page number ids: <a id="frontmatter"> or something. Again, with these GBS ids, I'll probably just trash them. For the others, I try to preserve them. If I'm lucky, I can just do a regex search/replace and move the id into a <section> or <h#> tag. But, sometimes, those already contain another id and I don't know what to do with those.

I can't find anything about anchor elements (<a> tags) having to be enclosed in a block-type tag anywhere. AFAICT, the spec doesn't talk about it and the examples don't indicate one way or another. But, neither Calibre's Check Book tool or EPubCheck have any problem with them that way.

It's just odd that the Beautify tool just seems to do nothing with pages that start out like that if the checking tools don't see anything wrong.

[theducks]

Originally Posted by enuddleyarbl

I'd been wondering what those "GBS..." tags were. Yes. A Google Books book. I'll probably just delete the lot of them. But, this happens in quite a few books where they're just trying to put in dead-tree page numbers. I also see them with simple non-page number ids: <a id="frontmatter"> or something. Again, with these GBS ids, I'll probably just trash them. For the others, I try to preserve them. If I'm lucky, I can just do a regex search/replace and move the id into a <section> or <h#> tag. But, sometimes, those already contain another id and I don't know what to do with those.

I can't find anything about anchor elements (<a> tags) having to be enclosed in a block-type tag anywhere. AFAICT, the spec doesn't talk about it and the examples don't indicate one way or another. But, neither Calibre's Check Book tool or EPubCheck have any problem with them that way.

It's just odd that the Beautify tool just seems to do nothing with pages that start out like that if the checking tools don't see anything wrong.

I think reverse of your logic is the norm.
If it ain't a block tag, then it needs to be in one.
(and yes, you are not the only one seein' Naked anchors. There is a ton of bad/messy code and it ain't all from indies.)

[gbm]

Use the modify epub plugin to remove the google play page maps. see screenshot.

bernie

Originally Posted by enuddleyarbl

I'd been wondering what those "GBS..." tags were. Yes. A Google Books book. I'll probably just delete the lot of them. But, this happens in quite a few books where they're just trying to put in dead-tree page numbers. I also see them with simple non-page number ids: <a id="frontmatter"> or something. Again, with these GBS ids, I'll probably just trash them. For the others, I try to preserve them. If I'm lucky, I can just do a regex search/replace and move the id into a <section> or <h#> tag. But, sometimes, those already contain another id and I don't know what to do with those.

I can't find anything about anchor elements (<a> tags) having to be enclosed in a block-type tag anywhere. AFAICT, the spec doesn't talk about it and the examples don't indicate one way or another. But, neither Calibre's Check Book tool or EPubCheck have any problem with them that way.

It's just odd that the Beautify tool just seems to do nothing with pages that start out like that if the checking tools don't see anything wrong.

[enuddleyarbl]

I've removed those GBS anchors without problem.

I'll also note that the <div> and <figure> tag also suffer the same lack of Beautification if there's an <a></a> tag in there with an image (I usually see it between the <div> or <figure> start tag and the <img> start tag it contains. For instance:

<div id="fig_1"><a></a><img src="../images/00005.jpeg" alt="Thror's Map" class="imgfull"/></div>

will not beautify (ditto for the <figure> version). But, delete the <a> tag pair and it beautifies just fine.

[kovidgoyal]

Beautify is very conservative in what it will beautify because in general changing white-space in an HTML file is not safe, it can change rendered appearance.

[Karellen]

Originally Posted by jackie_w

I think <a> tags are supposed to be enclosed in a block tag, e.g. <p> or <div>.

Hate to disagree, but I've seen <a> tags used at block level, even encompassing <div> and <section> tags.
It turns the whole block into a hyperlink (if a href is added)

[JSWolf]

Originally Posted by gbm

Use the modify epub plugin to remove the google play page maps. see screenshot.

bernie

It doesn't work. I tried it yesterday with an eBook from Google Play Books. I've reported it this morning. I hope it can be fixed easily.

[jackie_w]

Originally Posted by Karellen

Hate to disagree, but I've seen <a> tags used at block level, even encompassing <div> and <section> tags.
It turns the whole block into a hyperlink (if a href is added)

Perhaps I could have worded it better. My use of the phrase "supposed to be" was purely pragmatic in the context of the OP's wish to get a better result from calibre's Beautify. It was not a comment about valid HTML.

I tend to stay out of conversations about HTML/CSS "best practice" which seem to quickly escalate into a thread more suited to MR's "Politics & Religion".

[gbm]

I have no idea what you did wrong , but it works for me.

bernie

Originally Posted by JSWolf

It doesn't work. I tried it yesterday with an eBook from Google Play Books. I've reported it this morning. I hope it can be fixed easily.

[JSWolf]

Originally Posted by gbm

I have no idea what you did wrong , but it works for me.

bernie

It doesn't remove the <a GB stuff.

[gbm]

It works. I do not know what you are doing that is incorrect.


bernie

Originally Posted by JSWolf

It doesn't remove the <a GB stuff.

[jackie_w]

@gbm and @JSWolf,

I don't use Modify Epub anymore but I took a quick look at the plugin code. I think the problem may be that the plugin is currently only looking for GBS anchors in their self-closing form

<a id="GBS..."/>

i.e. not

<a id="GBS..."></a>

ETA: If Beautify has been run all anchors will look like the latter.

[JSWolf]

Originally Posted by jackie_w

@gbm and @JSWolf,

I don't use Modify Epub anymore but I took a quick look at the plugin code. I think the problem may be that the plugin is currently only looking for GBS anchors in their self-closing form

<a id="GBS..."/>

i.e. not

<a id="GBS..."></a>

ETA: If Beautify has been run all anchors will look like the latter.

That would do it. Thank you. Do you think this is an easy fix?