redo epub to epub - don't use original-epub

[cybmole]

i want to re-convert some epub books in order to use the CSS filter options.
I don't however want to convert from the original-epub files because if I do that I will lose recent edits made to the epub versions. i want to convert from epub to epub, as if the original-epub format was not present, and not to overwrite that original-epub format file at all

so is there any way to run a bulk epub to epub conversion with a setting that says don't use the original-epub files

otherwise its a complicated fudge. I'd need to go into DOS, rename all instances of that extension to fool calibre, then rename it back again afterwards ?

[BetterRed]

AFAIK - once created... by an EPUB to EPUB conversion or Polish Book... an ORIGINAL_EPUB will not be used in conversion

In any subsequent EPUB to EPUB conversion the existing EPUB is used as input and it will be overwritten by the output EPUB.

If there is only an ORIGINAL_EPUB then you can't even get a conversion - see attachment

BR

[theducks]

Also note: ANY (single) book conversion allows you to pick the source from the input side pulldown, you don't have to use the default/last used if you don't want to.
This is NOT true for bulk mode converts

[cybmole]

Originally Posted by BetterRed

AFAIK - once created... by an EPUB to EPUB conversion or Polish Book... an ORIGINAL_EPUB will not be used in conversion

In any subsequent EPUB to EPUB conversion the existing EPUB is used as input and it will be overwritten by the output EPUB.

If there is only an ORIGINAL_EPUB then you can't even get a conversion - see attachment

BR

I agree with your last point.

I need to retest the other points. I did one re-convert & noted that front & back matter, copyright etc. pages were present & I usually edit those out after the initial convert. Maybe I picked a bad test case where I'd not done that....

RETEST RESULT - Sorry but you were wrong:

TEST process
1. pick any book where there is just the one epub format
2. run epub to epub - this creates the original-epub format
3. now edit that book - type in THIS IS A TEST at start of chapter 1
4 redo point 2.
5. open book in editor - the text added in 3. is not there

so step 4 DID use original-epub as source

[DoctorOhh]

For conversions the Original_epub file, if present, will always be used over any existing epub as the source for the conversion.

[cybmole]

Originally Posted by DoctorOhh

For conversions the Original_epub file, if present, will always be used over any existing epub as the source for the conversion.

what i thought - so there's absolutely no way around that except to remove or rename the original files, then restore them after re-converting

or do a double bulk run 1- epub _> azw3 followed by 2. azw3-> epub.
That double step shoudl leave the origanl fiels untouched and should make mimimal unwanted code chages ( given that books have ben thoguh calibre convert once already ) ?

[BetterRed]

Originally Posted by cybmole

I agree with your last point.

I need to retest the other points. I did one re-convert & noted that front & back matter, copyright etc. pages were present & I usually edit those out after the initial convert. Maybe I picked a bad test case where I'd not done that....

RETEST RESULT - Sorry but you were wrong:

TEST process
1. pick any book where there is just the one epub format
2. run epub to epub - this creates the original-epub format
3. now edit that book - type in THIS IS A TEST at start of chapter 1
4 redo point 2.
5. open book in editor - the text added in 3. is not there

so step 4 DID use original-epub as source

My bad, I sort of misread you post -

I've always been confused over this, its why I don't retain ORIGINAL_xxxx in the calibre book folders. As soon as I have a ORIGINAL_xxxx I put it in a 7z archive, delete the ORIGINAL_xxxx and add the 7z to the book (if its not there already).

I don't understand why the conversion dialogue can't be explicit. If there's an EPUB, an ORIGINAL_EPUB, an RTF and an ORIGINAL_RTF then why not display the one that's going to be used, and offer all 4 in the Input drop down for me to explicitly select the one that I want... assuming the default isn't the one I want.

BR

[cybmole]

Originally Posted by BetterRed

My bad -

I've always been confused over this
BR

me too, but the whole point of having a backup up calibre library should be that I don't also have to create other files/folders for original, un-tweaked documents

though in my paranoia I have taken to doing that for all non amazon purchases. get them de-drm'd then stash a safe copy. in case other stores do a "sony"

so I want the original-epub fiels in calibre, but it's surely not unreasonable to also want to reconvert an edited epub ; my need for example is to apply some CSS filters not previously applied and thus avoid manual labour.

& I hate trying to remember dos syntax for bulk file renaming
its something like
rename *.original_epub *.ebak,
but it needs to happen within the calibre blackbox folder structure

[BetterRed]

Originally Posted by cybmole

me too...

I hate trying to remember dos syntax for bulk file renaming

me too - at my time of life I can't be bothered trying to remember all that tedious dross.

Do yourself a favour get a copy of xplorer2 lite (it's free) it has a browse flat feature that's gives a flat view of files in folder hierarchy. Filter by 'original_epub', rename them to say original_save. Do your conversions then browse flat again and rename them back to original_epub.

In screen shot you see I am ready to rename all the chm files in my Program Files Portable folder (I forgot the . in front of the chm in the filter box, hence that stray xml)

BR