View Single Post
Old 05-31-2013, 12:26 PM   #24
Toxaris
Wizard
Toxaris ought to be getting tired of karma fortunes by now.Toxaris ought to be getting tired of karma fortunes by now.Toxaris ought to be getting tired of karma fortunes by now.Toxaris ought to be getting tired of karma fortunes by now.Toxaris ought to be getting tired of karma fortunes by now.Toxaris ought to be getting tired of karma fortunes by now.Toxaris ought to be getting tired of karma fortunes by now.Toxaris ought to be getting tired of karma fortunes by now.Toxaris ought to be getting tired of karma fortunes by now.Toxaris ought to be getting tired of karma fortunes by now.Toxaris ought to be getting tired of karma fortunes by now.
 
Toxaris's Avatar
 
Posts: 4,520
Karma: 121692313
Join Date: Oct 2009
Location: Heemskerk, NL
Device: PRS-T1, Kobo Touch, Kobo Aura
Quote:
Originally Posted by abeonis View Post
If I understand well, there are 2 open issues:
1. We have to check the images afterwards, “sometimes something goes wrong with the numbering, probably something typical Word”. You personally place the pictures in Sigil.
2. Unordered lists are converted to ordered lists

Are these issues still there? I write Technical/Business eBooks and use intensively both elements and I use images for tables.
The first one does not exist with the add-in, only with the macro. The second one can occur. It depends on the source. If I create a document from scratch in Word, it works fine. If the source is ABBYY it will not always go right. I have not found the reason yet.

Quote:
Originally Posted by abeonis View Post
I write myself in MS-Word, so I am the only responsible to do it correctly (I am my unique client). However, at the end I run manually some S&R, directly in MS-Word, one after one … so I will definitely have a look at your “Search & Replace” module. By the way can we save the modified MS-Word file?
Of course, I could not see a reason why not. Just keep in mind that the source containing the S&R rules must be in the format as described for the procedure to workd.

Quote:
Originally Posted by abeonis View Post
Arrrgh! Could you please explain what it means?
Try it, it is easy enough to see what it does. Just keep in mind that in languages that use a lot of accents, there will be a lot of hits.

Quote:
Originally Posted by abeonis View Post
Regarding the Create ePUB module, you limited the matadatas to the IDPF requirement. That’s OK. I will have to check if Sigil generates a uuid in case we do not include an ISBN in your form. I’d like to suggest you add 2 metadatas, the publisher and the pub date. Some platforms require them (Barnes&Noble). It should not be a big deal for you.
I know I can add this metadata in Sigil. By the way, is there any INI file where we could save all the options we commonly use in the tool?
There will always be an uuid. There must be, even if there is not an ISBN. At the moment I will not add the publisher and publication date. I might do it later though. I would advise to create the enhancement request at the issue list (link at the site). In that way I will not forget about it.
There is no INI file. All the settings can be set and saved within the add-in. The settings will be retained if the current version is replaced by another version.

Quote:
Originally Posted by abeonis View Post
Finally, let’s have a dream. Is it planned to be able to run the tool from a command file?
No. There will always be manual effort required. Not all checks can be done automatic.

Quote:
Originally Posted by abeonis View Post
Again, many thanks for your efforts to solve this bottleneck. If at some point in your development you need to translate your GUI to spanish or french, just let me know how I can contribute to your effort.
If there is desire to have other languages available, let me know. I will see if I can come up with something then. The list is not that extensive.

Quote:
Originally Posted by abeonis View Post
Both intermal and external links are not rendered in HTML. They are recognized as underlined text (<u>text</u>.
Interesting. I would be interested in the Word file. Links should be retained. This is not the behavior I would like.

Quote:
Originally Posted by abeonis View Post
It supports a very simple table like the one in the docx file. If you merge cells or change the border, the Convert to HTML is not able to achieve its job.
That is according to design. It is also stated in the manual. Only simple tables will be converted at the moment.

Quote:
Originally Posted by abeonis View Post
The code generated is
<img src="../Images/image001.jpg" alt=""/>

I am not sure the alt attribute with a blank value will pass ePub validation. I need to check. Adding the name of the file to the alt attribute should help.

Is it possible that the tool adds the height and width attributes? (I use them to avoid kindle resizing).
It will validate. Not a problem at all actually. Adding the filename to the alt attribute will not be done. It should describe the picture and a filename will create a strange thing for the hearing impaired where the text is read.
I can add the height and width probably. Again, please enter an enhancement request. That will help me to organize.

Quote:
Originally Posted by abeonis View Post
If I want to check the HTML in a browser before creating the ePub, it will help if the tool creates the ../Images/ dir and put there the images it extracts from the compiled .docx
Enter an enhancement request. No garantuee though, I don't really see added value for that, since it is an intermediate state. I do not extract the images from the compiled docx. That is the case for the macro, but not for the add-in. For the add-in I use a totally different method that works a whole lot better. The result from the macro could not be used for the add-in. They are temporary saved in the TEMP directory though.

Quote:
Originally Posted by abeonis View Post
Regarding the HTML generated:

The line:
<link rel="stylesheet" type="text/css" href="..\Styles\stylesheet.css">
Should be:
<link rel="stylesheet" type="text/css" href="../Styles/stylesheet.css">

The following line is added at the end of the file, I don't know why
<p>&amp;nbsp;</p>
It is not a big deal, but you are right. They are corrected automatically. I will take a small look at it.

The last line informs me that you probably do not use the latest version. The ampersand should not be converted to the HTML entity in the latest version. That being said, I am not quite sure I uploaded the latest version, but pretty sure.

Quote:
Originally Posted by abeonis View Post
I always use xhtml files. I need to check as well if ePub vaIidation supports html files. Maybe the experts here can help.
An ePUB always contains XHTML files. So, it is actually not an HTML file, but an XHTML file. The file generated is HTML, but the one used in the ePUB is XHTML. Some changes are done.

If you want to know if there are new versions, you can use RSS (I think) or the newsletter on my site. If I upload a new version, I will sent out a newsletter. A big update I will also post here.

Issues and enhancements can be posted and requested here
Toxaris is offline   Reply With Quote