Formatting Epub Output - Page 5

kazbates · 07-19-2011, 11:37 AM

Quote:

Originally Posted by snarkophilus

In an Explorer window on Windows 7, click the "Organize" button on the top left, then the "Folder and search options" button. In the "View" table, uncheck the "Hide extensions for known file types" option. This then should include the file extension in the filenames in the Explorer windows, and allow you to change the .epub to .zip (with some warning about making the file potentially unusable). You should also be able to rename the epub file from a DOS command prompt as well.
<SNIP>

Bless you, Simon, that worked! I tried using the DOS command line before but couldn't remember the actual command and Rename didn't work. I have now renamed the epub to zip and unzipped it.

Interestingly enough, the ASCII characters are in the html files that I viewed using the MS IE9 browser. It looks more and more like I'm just going to have to do a global replace. If that's the case, I will just merge (append, actually) the files together that I want and then do Find/Replace. There has to be an easier way using html, but it's beyond me at this point.

I wonder what would happen if I loaded the html file into Book Designer. . .

kazbates · 07-19-2011, 09:38 PM

Okay, I'm totally frustrated. I've tried to do a Find/Replace of the characters but the file has the same character (â) to represent multiple symbols (', ", —, etc.). Consequently, I have to go through each html file line-by-line.

I can use the encoding feature in Notepad++ and encode in UTF-8 but I can't seem to save the encoded file.

I don't understand why Sigil alters the initial html code so that the file is no longer encoding in UTF-8.

I know I'm being a pain, but does anyone have any other ideas?. . .Help. . .

DaleDe · 07-19-2011, 10:03 PM

the same character should not represent multiple characters. It is likely that a pair of characters is representing each character and the â is common in all of them. Select the multiple characters and then substitute a single character. It is possible that the â is beside an invisible non printing character I suppose. Select a couple of characters with the â in the mix and then do a find and next find and see if it can be searched independently and if you then you can safely do the replace. This thing is a mess, Can you send me one of items as a sample? PM me.

Dale

JSWolf · 07-19-2011, 10:05 PM

OK, start off getting a copy of notepad++ 5.9.2. It's what I use. I'll go on from there once I find a free eBook we can all download and strip the DRM so we can be on the same page with the same code. I'm thinking one of the Star Wars serial ePub because it's short and will do fine.

kazbates · 07-19-2011, 10:09 PM

Quote:

Originally Posted by DaleDe

the same character should not represent multiple characters. It is likely that a pair of characters is representing each character and the â is common in all of them. Select the multiple characters and then substitute a single character. It is possible that the â is beside an invisible non printing character I suppose. Select a couple of characters with the â in the mix and then do a find and next find and see if it can be searched independently and if you then you can safely do the replace. This thing is a mess, Can you send me one of items as a sample? PM me.

Dale

Thanks, Dale. I've tried all kinds of combinations. I'll PM you.

Quote:

Originally Posted by JSWolf

OK, start off getting a copy of notepad++ 5.9.2. It's what I use. I'll go on from there once I find a free eBook we can all download and strip the DRM so we can be on the same page with the same code. I'm thinking one of the Star Wars serial ePub because it's short and will do fine.

Thanks, Jon. I downloaded Notepad++ and tried to use it to edit out the special characters that are giving me such a headache. I will PM you, as well.

DaleDe · 07-19-2011, 10:09 PM

One last stab in the dark. Do an ePub to ePub conversion using Calibre, just in case it can fix it.

Dale

kazbates · 07-19-2011, 10:21 PM

Great minds, and all that. That was one of my first ideas. No luck. The special characters are still not right.

Faster · 07-22-2011, 02:28 PM

Fixing those b..... nasty characters:

Open a blank document in Web View in MS Word.

SIGIL
Open the problem epub in Sigil.
Go into Code View.
Do a Find/Replace as follows:

(Look in: All HTML Files)
(Search Mode: Regular expression)
(Check: All)

Code:

Find:	(</body>)
Replace:	<hr class="sigilChapterBreak" />\1

The number of replacements should agree with the number of hml files in the Text folder. If not, discard and start again.

Next go to the left panel and RIGHT CLICK on the second HTML file. From the context menu that appears select 'Merge With Previous'. Repeat this until you have only one large concatenated file.
Go back to the Code View window. Select All CTRL A and Copy CTRL C.

WORD
Go to the blank MS Word document and paste CTRL V.
Select All and change the font to Arial size 14. (No effect on your epub but easier to select characters.)
Go to Tools > AutoCorrect Options > AutoFormat As You Type and ensure "Straight quote" with "smart quotes" is checked.

Now do a series of Find/Replaces CTRL H. (Click 'More' and make sure you've got Search 'All')
Find the first cluster of unwanted characters.
Put the cursor in front, then hold down <SHIFT> and use the RIGHT ARROW key to extend the selection over the cluster. Pay careful attention as to whether a space is or isn't a part of the unwanted cluster. COPY and Paste into the 'Find what' box.
In the 'Replace with' box enter the missing character. I'm afraid you'll have to resign yourself to single quotation marks rather than double otherwise you"ll be sorry!
It will take half a dozen F/Rs to correct the whole book. Replacements are typically leading and trailing quote marks, apostrophes and commas.

Note: if you generate a mistake, as I did, with spacing around commas, eg <space>comma<no space> then you can search for <space>comma and replace with comma<space>.

Finally in Word:
Select All CTRL A and Copy CTRL C.

SIGIL
Back in Code View and Select All CTRL A and Paste/Replace CTRL V with the text from Word.
Now press function key F6. This restores the text files sequence and any TOC will work as before.
Check it out in Book View and Save.

JSWolf · 07-23-2011, 09:46 PM

I am away for the weekend and will be home on Sunday. I'll have a look at the eBook on Sunday night via ADE and see if I can see anything funny. Can you point out what pages have these problems?

JSWolf · 07-23-2011, 09:49 PM

Since you have Sony Readers, I'm going to help you install fonts. That way we can actually make the book look better by changing from the default ADE font to a better font easily. Unless of course, the ePub already has a font family embedded.

kazbates · 07-25-2011, 09:59 AM

Okay! Here is what I had to do to correct the special character encoding issue:
Use 7zip and copy all html files to my desktop. Load all those files into Notepad++. Do a Replace in All Open files of "charset=iso-8859-1" with "charset=UTF-8". Save all the files and drap back to 7zip.

None of this is necessary, of course, unless I want to use the epubmerge.py script. I can easily edit the css file using 7zip and Notepad++ to change the margins and the justification. As Jon pointed out, this is a lot of work just to create an omnibus when it would make just as much sense to keep them all separate. :0

Thanks so much to everyone who has given me so much of their time and expertise to walk me through all of this mess. Hopefully, this thread will help others who are struggling with epub formatting issues.

07-22-2011, 02:28 PM	#68
Faster Connoisseur Posts: 61 Karma: 12096 Join Date: Sep 2010 Location: Tasmania Device: Sony PRS 650	Fixing those b..... nasty characters: Open a blank document in Web View in MS Word. SIGIL Open the problem epub in Sigil. Go into Code View. Do a Find/Replace as follows: (Look in: All HTML Files) (Search Mode: Regular expression) (Check: All) Code: Find: (</body>) Replace: <hr class="sigilChapterBreak" />\1 The number of replacements should agree with the number of hml files in the Text folder. If not, discard and start again. Next go to the left panel and RIGHT CLICK on the second HTML file. From the context menu that appears select 'Merge With Previous'. Repeat this until you have only one large concatenated file. Go back to the Code View window. Select All CTRL A and Copy CTRL C. WORD Go to the blank MS Word document and paste CTRL V. Select All and change the font to Arial size 14. (No effect on your epub but easier to select characters.) Go to Tools > AutoCorrect Options > AutoFormat As You Type and ensure "Straight quote" with "smart quotes" is checked. Now do a series of Find/Replaces CTRL H. (Click 'More' and make sure you've got Search 'All') Find the first cluster of unwanted characters. Put the cursor in front, then hold down <SHIFT> and use the RIGHT ARROW key to extend the selection over the cluster. Pay careful attention as to whether a space is or isn't a part of the unwanted cluster. COPY and Paste into the 'Find what' box. In the 'Replace with' box enter the missing character. I'm afraid you'll have to resign yourself to single quotation marks rather than double otherwise you"ll be sorry! It will take half a dozen F/Rs to correct the whole book. Replacements are typically leading and trailing quote marks, apostrophes and commas. Note: if you generate a mistake, as I did, with spacing around commas, eg <space>comma<no space> then you can search for <space>comma and replace with comma<space>. Finally in Word: Select All CTRL A and Copy CTRL C. SIGIL Back in Code View and Select All CTRL A and Paste/Replace CTRL V with the text from Word. Now press function key F6. This restores the text files sequence and any TOC will work as before. Check it out in Book View and Save.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
PDF to EPUB No output file	pearlheartgtr	Conversion	1	04-12-2011 01:03 AM
EPUB output	kovidgoyal	Calibre	920	02-05-2011 11:59 AM
EPUB output justification	toki08	Calibre	10	01-08-2011 04:14 PM
Formatting errors in Mobi output	agriainc	Calibre	3	08-22-2010 10:16 AM
epub output metadata	troymc	Calibre	5	05-22-2010 12:23 AM

07-19-2011, 09:38 PM	#62
kazbates Wizard Posts: 2,627 Karma: 406616 Join Date: Dec 2008 Location: Northern Virginia Device: SurfacePro, SurfaceBook 2	Okay, I'm totally frustrated. I've tried to do a Find/Replace of the characters but the file has the same character (â) to represent multiple symbols (', ", —, etc.). Consequently, I have to go through each html file line-by-line. I can use the encoding feature in Notepad++ and encode in UTF-8 but I can't seem to save the encoded file. I don't understand why Sigil alters the initial html code so that the file is no longer encoding in UTF-8. I know I'm being a pain, but does anyone have any other ideas?. . .Help. . .

07-19-2011, 10:03 PM	#63
DaleDe Grand Sorcerer Posts: 11,470 Karma: 13095790 Join Date: Aug 2007 Location: Grass Valley, CA Device: EB 1150, EZ Reader, Literati, iPad 2 & Air 2, iPhone 7	the same character should not represent multiple characters. It is likely that a pair of characters is representing each character and the â is common in all of them. Select the multiple characters and then substitute a single character. It is possible that the â is beside an invisible non printing character I suppose. Select a couple of characters with the â in the mix and then do a find and next find and see if it can be searched independently and if you then you can safely do the replace. This thing is a mess, Can you send me one of items as a sample? PM me. Dale

07-19-2011, 10:05 PM	#64
JSWolf Resident Curmudgeon Posts: 79,745 Karma: 145864619 Join Date: Nov 2006 Location: Roslindale, Massachusetts Device: Kobo Libra 2, Kobo Aura H2O, PRS-650, PRS-T1, nook STR, PW3	OK, start off getting a copy of notepad++ 5.9.2. It's what I use. I'll go on from there once I find a free eBook we can all download and strip the DRM so we can be on the same page with the same code. I'm thinking one of the Star Wars serial ePub because it's short and will do fine.

07-19-2011, 10:09 PM	#66
DaleDe Grand Sorcerer Posts: 11,470 Karma: 13095790 Join Date: Aug 2007 Location: Grass Valley, CA Device: EB 1150, EZ Reader, Literati, iPad 2 & Air 2, iPhone 7	One last stab in the dark. Do an ePub to ePub conversion using Calibre, just in case it can fix it. Dale

07-19-2011, 10:21 PM	#67
kazbates Wizard Posts: 2,627 Karma: 406616 Join Date: Dec 2008 Location: Northern Virginia Device: SurfacePro, SurfaceBook 2	Great minds, and all that. That was one of my first ideas. No luck. The special characters are still not right.

07-23-2011, 09:46 PM	#69
JSWolf Resident Curmudgeon Posts: 79,745 Karma: 145864619 Join Date: Nov 2006 Location: Roslindale, Massachusetts Device: Kobo Libra 2, Kobo Aura H2O, PRS-650, PRS-T1, nook STR, PW3	I am away for the weekend and will be home on Sunday. I'll have a look at the eBook on Sunday night via ADE and see if I can see anything funny. Can you point out what pages have these problems?

07-23-2011, 09:49 PM	#70
JSWolf Resident Curmudgeon Posts: 79,745 Karma: 145864619 Join Date: Nov 2006 Location: Roslindale, Massachusetts Device: Kobo Libra 2, Kobo Aura H2O, PRS-650, PRS-T1, nook STR, PW3	Since you have Sony Readers, I'm going to help you install fonts. That way we can actually make the book look better by changing from the default ADE font to a better font easily. Unless of course, the ePub already has a font family embedded.

07-25-2011, 09:59 AM	#71
kazbates Wizard Posts: 2,627 Karma: 406616 Join Date: Dec 2008 Location: Northern Virginia Device: SurfacePro, SurfaceBook 2	Okay! Here is what I had to do to correct the special character encoding issue: Use 7zip and copy all html files to my desktop. Load all those files into Notepad++. Do a Replace in All Open files of "charset=iso-8859-1" with "charset=UTF-8". Save all the files and drap back to 7zip. None of this is necessary, of course, unless I want to use the epubmerge.py script. I can easily edit the css file using 7zip and Notepad++ to change the margins and the justification. As Jon pointed out, this is a lot of work just to create an omnibus when it would make just as much sense to keep them all separate. :0 Thanks so much to everyone who has given me so much of their time and expertise to walk me through all of this mess. Hopefully, this thread will help others who are struggling with epub formatting issues.

Advert

Advert