e-Book Tools - a Word add-in

[Toxaris]

Originally Posted by Hitch

Hi:

You mean...what you mean is, it's not that easy to make the program do it? You don't mean, it's not that easy for people, right? (Seriously...just clarifying what you said.)

Hitch

I mean making the program do it. Getting the fonts is easy, that is just one call. Getting the correct filenames of these fonts... not that easy at all. Getting the Bold/Italic/Bold Italic as well as the regular, even harder. There could be cases with localized versions that the current version does not even work. When the names are known, it is easier, as long as you take into account that not everybody uses c:\Windows. Oh, and the methods of getting the filenames also change depending on which version of Windows you use.
For people it is easy, go to the font directory and just copy it...

[Hitch]

Originally Posted by Toxaris

I mean making the program do it. Getting the fonts is easy, that is just one call. Getting the correct filenames of these fonts... not that easy at all. Getting the Bold/Italic/Bold Italic as well as the regular, even harder. There could be cases with localized versions that the current version does not even work. When the names are known, it is easier, as long as you take into account that not everybody uses c:\Windows. Oh, and the methods of getting the filenames also change depending on which version of Windows you use.
For people it is easy, go to the font directory and just copy it...

Right. That's what I thought you meant. :-)

I wish it were not so easy. I think that between Atlantis and whats-its, um, Jutoh, there are likely thousands, if not tens of thousands, (especially good old TNR) unlicensed fonts in eBooks around the globe. I'm stupid. I license fonts all the time. I even licensed an extortionate (to my way of thinking) font, just because I HAD to have it. It was a sans, with an absolutely lovely G. (I know. I'm odd, at best.)

And have I used it? HARDLY. Planned to use it for my (then) new website. Ended up using something normal from the Goog.

Hitch

[BetterRed]

Tools Version 1.18, Word 2010, Windows 10

When I try to update the S&R 'temporary' xml file nothing happens, and the Exit button does nothing, I have to click red X to dismiss the dialogue



Reverted to Version 1.16 - same behaviour

This looks like the checksum not changing problem we had before. I want to use my latest C:\_Miscellaneous\myReplaceRules.xml but once again I can't - even if I delete %AppData%\Toxaris\ePUBToolsSRRules.xml I can't

Maybe it would it be better if I maintained %AppData%\Toxaris\ePUBToolsSRRules.xml directly? I might give that a go tomorrow.

BR

[Toxaris]

Hmm. Are you sure the files are not in use? What you can try is to delete the option CRC32hash from the configuation file. That will cause a new hash to be created. I will look into the code to see what could cause this. Usually it is caused by open files.

The behaviour should not change in version 1.16 since nothing has been changed there.

[BetterRed]

@Toxaris - files are not in use (other than by epubtools), to prove that I tried it immediately after a Windows restart.

I removed the CRC32HASH and tried again, same result.

I'm attaching the config and log in a zip. The log is short because I tried this yesterday :-

• uninstalled epubtools,
• deleted %appdata%toxaris,
• did fresh install, restored epubtools.cfg and ePUBToolsSRRules.xml from backup into the new %appdata%toxaris.

However, I tried editing ePUBToolsSRRules.xml directly with Notepad++. This seems to work fine. Of course I did it when Word wasn't running. I also edit Word's exclusion lists directly with Notepad++. I plan on editing ePUBToolsSRRules directly from now on.

I've had this problem several times - hopefully direct editing means I'll not get it again. Am I the only person who prefers to edit XML with a text editor, rather than fiddle around with Word tables? Wouldn't be surpised if I am.

I probably edit my S&R rules more frequently than those who only use it to fix scanning glitches. I also (mainly) use it to de-jargonise.

BR

[Toxaris]

It remains very strange. You are the only one, that I know off, that has this problem. I personally usually use the internal editor. I will look at the files and the code. If I remember correctly, that button just makes a copy and recreates the hash. So, no real rocket science. Oh, and of course it wants to validate the XML file.

[BetterRed]

Originally Posted by Toxaris

If I remember correctly, that button just makes a copy and recreates the hash. So, no real rocket science.

Apart from the fact that it uses invisible data (the hashes) to determine if the file has changed, rather than the visible modify timestamp of the designated file being later than the visible modify timestamp of the copy in %appdata%/toxaris.

I can see that the hash can be used avoid superfluous copying, thereby saving a tiny amount of disk wear-and-tear and a blink of an eye in time when the Recreate temporary S/R XML each use is checked. But if the options are as I have them. then surely the hash is irrelevant. I thought the whole point of that configuration was to give the user absolute control over the copying.

Originally Posted by Toxaris

Oh, and of course it wants to validate the XML file.

And if it doesn't validate - then does (should) it show an error message?

BR

[BetterRed]

Originally Posted by BetterRed

And if it doesn't validate - then does (should) it show an error message?

Apparently not.

I just ran my xml file through this XML Validation and discovered I had lost a '/' on a closing tag.

After fixing that error, the EpubTools S&R->Create temporary file features works.

I wonder if there's an xml validator PI for Notepad++?

Added : of course there is, I already have it - XML Tools

BR

[Toxaris]

Originally Posted by BetterRed

Apparently not.

I just ran my xml file through this XML Validation and discovered I had lost a '/' on a closing tag.

After fixing that error, the EpubTools S&R->Create temporary file features works.

I wonder if there's an xml validator PI for Notepad++?

Added : of course there is, I already have it - XML Tools

BR

It does not? That is strange. In the code it is specifically specified that a message must be given in case the validation fails. Otherwise it would be useless of course... Can you sent me the xml file with the error? I would like to test this.
A message box should be shown in case the file does not validate. What sometimes happens, is that the message box is hidden behind other forms and therefore it seems that a program is stuck when it is actually waiting for answer. Such a messagebox can be hard to select. I have seen it in rare cases. I will see if I can add some code to be absolutely sure this cannot happen.

I use the CRC for multiple reasons. One is that the timestamp is not always enough. I have seen cases the timestamp is not changed or is changed with no changes to the file. In any case calculating a CRC32 hash is no rocket science in itself, it is a relative simple process. It is also calculated very quickly.

**EDIT** I have run a test case of what you described (forgetting a slash in the ending tag) and your are correct. The validation will crash instead of reporting that it does not validate (and why). I will fix this of course.

***EDIT 2*** I have fixed this... The program will now not only crash, but also gives the reason why (as it should and was designed).

[Toxaris]

Originally Posted by BetterRed

Apart from the fact that it uses invisible data (the hashes) to determine if the file has changed, rather than the visible modify timestamp of the designated file being later than the visible modify timestamp of the copy in %appdata%/toxaris.

Actually the button you use do not use the hashes or any check in that regard. Even if it is the same file, it will just create a new file. The button you pressed is pretty 'dumb' in that area, as is intended.

[BetterRed]

Originally Posted by Toxaris

***EDIT 2*** I have fixed this... The program will now not only crash, but also gives the reason why (as it should and was designed).

Do you really mean 'crash' - crash what - Windows, Word, Epub-Tools

From now on I'll be checking my XML in Notepad++, I've also flicked the auto-validate switch on. So hopefully Epub-tools will never see any errors.

Originally Posted by Toxaris

Actually the button you use do not use the hashes or any check in that regard. Even if it is the same file, it will just create a new file. The button you pressed is pretty 'dumb' in that area, as is intended.

That's what I understood too - that's why it irritated me so, the behaviour didn't match.

I am wondering if the ongoing frustrations I've had with the XML rules feature have been due unreported errors in the XML file ever since the getgo. I've never explicitly fixed any errors, instead what I've been doing is to restore my rules and the saved rules from backup and then re-enter the changes and add comments as padding to force a checksum difference Ψ²

BR

[Toxaris]

Originally Posted by BetterRed

Do you really mean 'crash' - crash what - Windows, Word, Epub-Tools

I did mean crash. When I ran the test case on my development machine, it went straight to the debugger and give the issue. It would not go forward nor backwards. That is why you had to end the process/form hard.

Anyway, it is fixed now. If you want, I can sent you the validation scheme for the XML.

I will wait a bit for the release of this, as I am testing the option to have equations/formulas exported as SVG in the ePUB instead of a bitmap image. That should give better images.

[BetterRed]

No hurry, now we've tracked down the cause and I've enabled appropriate settings in the XML-Tools PI for Notepad++ the add-in should never see any bad XML, from me at least.

And if it should happen again I know what to look for - there's no shortage of XML validators

BR

[Toxaris]

Originally Posted by BetterRed

No hurry, now we've tracked down the cause and I've enabled appropriate settings in the XML-Tools PI for Notepad++ the add-in should never see any bad XML, from me at least.

And if it should happen again I know what to look for - there's no shortage of XML validators

BR

Well, validation with regards to tags is one thing. The add-in also validates if the correct tags are used and if they are all there (there is one optional). So not only structure is checked.

[BetterRed]

@Toxaris - FYI, I bit the bullet! Converted my S&R rule tables to spreadsheets, which IMO is the best approach for reading and maintaining tabular data.

I'll have a script to save those as csv files, convert to xml, and then concatenate the xml 'header', the xml tables and the xml 'footer' to create a new MyReplaceRules.xml. That's the file I input into EPUB-Tools->Settings->S&R->Create temp S/R XML, as per my last screen shot.

This 'approach' has been rattling around in my head for some time.

BR