MobileRead Forums - View Single Post - Search & Replace problems with pdf conversion

Longmatys · 01-25-2012, 03:52 AM

Hi All,

I have some difficulty with search & replace. I'm quite familiar with regex, but I'm not sure if I understand correctly, how does it work in calibre.

My problem is, that I'm trying to convert pdf to mobi. Everything is quite ok, and the last thing is, that there are headings and page numbers to be removed. Page numbers were quite easy, but I have problems with page headings. I tried to use the debug mode, but it does not help me either.

My question is, from debugging process I got three directories:

input
parsed
processed
structure

From what I read, the regex processing should be run against parsed files. But I really doubt it.

I use two regex patterns:

\n[0-9]+ <br>
0[^\n]*?0\s*<br>

The first one removes page numbers and works fine. The second one is meant to remove lines, which begins and ends with 0 with any text between them (it contains the name of chapters - so it keeps changing)

I'm confused, that the first one works - in xhtml files in parsed directory, there are no

Code:

<br>

tags??

Is there a way, where I can step in the process - do some manual search/replace and then continue? Maybe via html export/import?

Input exhibit:

Code:

<hr>
<A name=8></a>0&nbsp;&nbsp;Doba&nbsp;jedová&nbsp;0&nbsp;<br>
dostávají&nbsp;velmi&nbsp;obtížně&nbsp;a&nbsp;pomalu,&nbsp;protože&nbsp;znamenají&nbsp;velký&nbsp;zásah&nbsp;<br>

parsed exhibit:

Code:

na  veřejnost 0  Doba jedová 0 </p>
<p>dostávají velmi obtížně a pomal

processed exhibit:

Code:

veřejnost 0  Doba jedová 0 </p>
<p class="calibre1">dostávají velmi obtížně a pomalu

structure exhibit:

Code:

veřejnost 0  Doba jedová 0 </p>
<p>dostávají velmi

regex builder:

Code:

<hr>
<A name=8></a><IMG src="index-8_1.jpg"><br>
0  Doba jedová 0 <br>
dostávají velmi obtížně a pomalu

Does anyone know how calibre behaves to endlines? Can I use them in regexes? (I did, but is it wise?)

01-25-2012, 03:52 AM	#1
Longmatys Junior Member Posts: 1 Karma: 10 Join Date: Jan 2012 Device: Kindle 3	Search & Replace problems with pdf conversion Hi All, I have some difficulty with search & replace. I'm quite familiar with regex, but I'm not sure if I understand correctly, how does it work in calibre. My problem is, that I'm trying to convert pdf to mobi. Everything is quite ok, and the last thing is, that there are headings and page numbers to be removed. Page numbers were quite easy, but I have problems with page headings. I tried to use the debug mode, but it does not help me either. My question is, from debugging process I got three directories: input parsed processed structure From what I read, the regex processing should be run against parsed files. But I really doubt it. I use two regex patterns: \n[0-9]+ <br> 0[^\n]?0\s<br> The first one removes page numbers and works fine. The second one is meant to remove lines, which begins and ends with 0 with any text between them (it contains the name of chapters - so it keeps changing) I'm confused, that the first one works - in xhtml files in parsed directory, there are no Code: <br> tags?? Is there a way, where I can step in the process - do some manual search/replace and then continue? Maybe via html export/import? Input exhibit: Code: <hr> <A name=8></a>0  Doba jedová 0 <br> dostávají velmi obtížně a pomalu, protože znamenají velký zásah <br> parsed exhibit: Code: na veřejnost 0 Doba jedová 0 </p> <p>dostávají velmi obtížně a pomal processed exhibit: Code: veřejnost 0 Doba jedová 0 </p> <p class="calibre1">dostávají velmi obtížně a pomalu structure exhibit: Code: veřejnost 0 Doba jedová 0 </p> <p>dostávají velmi regex builder: Code: <hr> <A name=8></a><IMG src="index-8_1.jpg"><br> 0 Doba jedová 0 <br> dostávají velmi obtížně a pomalu Does anyone know how calibre behaves to endlines? Can I use them in regexes? (I did, but is it wise?)