File problem in 0.4.0

[adv_dp_fan]

I noticed that 4.0 was posted for download today and so pulled it over to try out. Everything was fine until I had to create a chapter split in on file. When I did several of the xhtml files in the text folder in the book view suddenly jumped to different positions. I moved them back to the proper positions but the next chapter split moved things around again. Tried saving and reloading the book only to notice that when ever I saved and reloaded that some of the xhtml files where missing and I'd have to recreate them.

Was trying the Windows 64 bit version under Win 7 and tested the 32 bit version as well, same issues. Any ideas?

Thanks.

[ghostyjack]

This has been occuring since the first release of 0.4.0.

I was hoping this was fixed.

I'll stick with 0.3.4 for the time being then.

[weedfreak]

Yep, splitting chapters can cause all sorts of problems, so can merging chapters, often one of the merged chapters disappears and is not added to the other one. Html tidy is now next to useless whether it is enabled or not a syntax error will throw up a error page with a selection for automatic or manual fix, It does not seem to matter which you chose the document is returned to the state it was in before the edit.
It also seems that this version does not work with Calibre "open with .. " plugin.
Validating also seems to throw up more errors than before, some of which seem meaningless to me.

[charleski]

It always helps if you attach an example.

I'm guessing that the filenames in your ePub contain spaces. Luckily I have some time today and coded a fix that should cover the last of the problems with spaces.

And SERIOUSLY, people - if you have a problem ATTACH AN EXAMPLE THAT SHOWS IT HAPPENING. Vague reports of problems aren't helpful, since I have to spend more time trying to work out what the problem actually is that I do in fixing it.

[user_none]

Originally Posted by weedfreak

syntax error will throw up a error page with a selection for automatic or manual fix, It does not seem to matter which you chose the document is returned to the state it was in before the edit.

Before if you switched to the book view it would auto correct. If you auto correct there is the potential for data loss. This is to give you a chance to fix the problem before it turns into something bigger. Also, automatic correction cannot read your mind so it tries to figure out what you mean with minimal data loss.

[weedfreak]

Before it would auto correct if you had chosen to turn auto correct on, if you had turned off auto correct it would do nothing except add Style="" to the body tag unless there was some style already there. It may have done other things but I never really checked.

Now it does not matter if you have the auto correct on or off so you may as well delete that button. The page for selecting now whether to auto or manually correct seems to be doing the same thing whichever you select, that is reverting to the last version of the file, when you last changed view. Selecting manual correction does show the code for the 'error' but as soon as you try to edit it it reverts to the last version again. I have not yet seen any instance of auto correct doing anything like even attempting a fix, it only reverts the file.

I have lost more data today using 4.0 than in the last six months of daily use of previous versions. For my purposes 4.0 is of no use at all. I cannot send examples of the errors as the data I would want to show is deleted by the programme.

[charleski]

Originally Posted by weedfreak

Before it would auto correct if you had chosen to turn auto correct on, if you had turned off auto correct it would do nothing except add Style="" to the body tag unless there was some style already there. It may have done other things but I never really checked.

In previous versions HTMLTidy did a lot of auto-correction behind the scenes, whether you had it turned on or not. With version 0.4, that auto-correction has now been exposed to the user and you have a choice whether to use that or fix the problem yourself.

Now it does not matter if you have the auto correct on or off so you may as well delete that button. The page for selecting now whether to auto or manually correct seems to be doing the same thing whichever you select, that is reverting to the last version of the file, when you last changed view.

Sorry, but this just isn't happening. I've tested the current release on several epubs: turn Tidy off, insert some errors, get the error dialog, click Fix manually and the errors are there exactly as I typed them. I can step back using undo and get back to the original version, then step forward with redo and get back to the errors I inserted.

I cannot send examples of the errors as the data I would want to show is deleted by the programme.

You're clearly doing something strange that hasn't been anticipated. If you want this fixed then you need to post an example to the tracker along with the exact steps you took that produced the problem. Because as far as I can see it's working perfectly.

[adv_dp_fan]

Some of the files had spaces in them but the ones I was working with at the time didn't. In 3.4 the new chapter split off was placed right after the file it was split from, in 4.0 it seemed each split chapter was placed at the end of the book and some other random chapters were moved. I can deal with that, it was the save and then restore which had missing chapters which worried me the most. Rearranging data is one thing, lost data is something else.

I couldn't attach the example as it was a personal scan of copyrighted material but I can see if I can duplicate it with some random text. It happens every time with the book I was working with but I haven't tried 4.0 on anything else yet.

I'll dig deeper and see if I can get more details to help track it down.

[user_none]

If you open a bug report we can mark it private so the file is not publicly available. Once open just ask and once we've marked it as private you can upload the file.

[theducks]

Originally Posted by adv_dp_fan

Some of the files had spaces in them but the ones I was working with at the time didn't. In 3.4 the new chapter split off was placed right after the file it was split from, in 4.0 it seemed each split chapter was placed at the end of the book and some other random chapters were moved. I can deal with that, it was the save and then restore which had missing chapters which worried me the most. Rearranging data is one thing, lost data is something else.

I couldn't attach the example as it was a personal scan of copyrighted material but I can see if I can duplicate it with some random text. It happens every time with the book I was working with but I haven't tried 4.0 on anything else yet.

I'll dig deeper and see if I can get more details to help track it down.

I saw both of those issues (and reported them) in RC1... Could not be found- closed
(I did not realize spaces might be a triggering condition and was not watching for them)
Maybe Sigil can 'tidy-up' file names while shuffling the source structure to it's standards on first process.

[charleski]

I've fixed a string of completely different problems with the code for splitting chapters. There've been at least four distinct problems involving completely different parts of the code. This last issue with spaces was actually a problem with renaming files and had nothing to do with the chapter-splitting routines, but I believe I've now fixed all the places where a space in the filename could cause problems.

Since many production workflows involve generating filenames with specific identifiers embedded in them, I didn't want to change the filenames themselves (as calibre does). But in general it's a good idea to avoid using spaces.

[adv_dp_fan]

Okay did some brief testing with 32 bit 4.0 on Win 7.

It does appear to be an issue with spaces in the filenames. I first started an empty book and added some random text to the initial file. Added a new item and added more random text to the new file. Then I picked a random spot in the first file and did a chapter split. Everything split okay and the new file appeared right after its parent as expected. Saved and reloaded and everything looked fine.

Then I changed the file names of the second and third files to include spaces, went back to a random spot in the first file and did a chapter split and the files were no longer in expected order. I then saved and restarted and reloaded and the new file and all the text in it were missing.

Tried a few other things and each time, seemed fine if the files names didn't have spaces and started acting strange and losing data if they had spaces.

Of course a LOT of my books have spaces in the filenames...

[JSWolf]

Originally Posted by adv_dp_fan

Okay did some brief testing with 32 bit 4.0 on Win 7.

It does appear to be an issue with spaces in the filenames. I first started an empty book and added some random text to the initial file. Added a new item and added more random text to the new file. Then I picked a random spot in the first file and did a chapter split. Everything split okay and the new file appeared right after its parent as expected. Saved and reloaded and everything looked fine.

Then I changed the file names of the second and third files to include spaces, went back to a random spot in the first file and did a chapter split and the files were no longer in expected order. I then saved and restarted and reloaded and the new file and all the text in it were missing.

Tried a few other things and each time, seemed fine if the files names didn't have spaces and started acting strange and losing data if they had spaces.

Of course a LOT of my books have spaces in the filenames...

The best way to deal is to remove the spaces in the filenames. They actually shouldn't be there.

[alansplace]

Originally Posted by adv_dp_fan

Okay did some brief testing with 32 bit 4.0 on Win 7.....

and i on 0.4.0 64bit on win 7. i have a file with spaces in the .html filenames and any split i create gets moved to above the cover graphics file till i rename the files w/o any spaces, then the splits occur normally.

[crutledge]

I just experimented with a book (A Young Man in a Hurry.htm) loaded it to sigil and split and saved. Then I split a file at random (A Young Man in a Hurry_0010.xhtml).

A Young Man in a Hurry_0010.xhtml popped to the top of Book Browser and the split off file (Section0001.xthtml) at the bottom.

I then renamed the book with no spaces (A_Young_Man_in_a_Hurry.htm) and retested. Every thing works as advertised.

It's strange to see Section0001.xthtm appear in the middle of Book Browser.

For two other files I get the same-same.

If the file (book) name is to be used as section names, a major problem with final book names is to be solved. The final epub file name will be the book name with underscores. If we go back to the old Section0002, Section0003,, etc. this problem goes away.