[phossler]

Thanks.

Small issue

If there's <style> in the html file, the [Replace All] will process that text and generate errors

For example, I had a jacket.xhtml file in an epub, and the RE split some things that generated errors (not typos or mis-spellings)

Spoiler:

<style type="text/css">
.cbj_banner {
background: #eee;
col or: black;
border: thin solid black;
margin: 1 em;
padding: 1 em;
}
table.cbj_header td.cbj_title {
font-size: 1.5 em;
font-style: italic;
text-align: c enter;
}
table.cbj_header td.cbj_series {
text-align: c enter;
}
table.cbj_header td.cbj_author {
text-align: c enter;
}
table.cbj_header td.cbj_pubdata {
text-align: c enter;
}
table.cbj_header {
width: 100%;
}
table.cbj_header td.cbj_label {
text-align: right;
width: 33%;
}
table.cbj_header td.cbj_content {
text-align: left;
width: 67%;
}
hr.metadata_divider {
width: 90%;
margin-left: 5%;
border-top: solid white 0;
border-right: solid white 0;
border-bottom: solid black 1px;
border-left: solid white 0;
}
hr {
border-top: 0 solid white;
border-right: 0 solid white;
border-bottom: 2px solid black;
border-left: 0 solid white;
margin-left: 10%;
width: 80%;
}
.cbj_footer {
font-size: 0.8 em;
margin-top: 8px;
text-align: c enter;
}
</style>

Any way to make the function a little smarter?

I can always regenerate the jacket.xhtml file, but any other files that have <style> in would probably be changed also