PBB Copying Error

The PBB compiler is still buggy. It has just been released for beta testing for the first time (Beta 6) about a week ago, and we've been hammering on it since and reporting bugs. They haven't released an update with bug fixes yet, but I expect it to take several releases before it's quite stable and usable.

If you'd like to join in with our beta testing efforts, you'll need to be able to document how to reproduce the problem you've encountered. See if you can narrow down the DOCX file to the bare minimum that will demonstrate this problem. Then post the document here if it's something you can legally share.

Another thing to note: we have discovered that Word generates pretty yucky XML code behind the scenes, and this currently confuses the PBB compiler. If you're just wanting to get some work done with this document, you can clean up the XML by saving the DOCX file as RTF, then opening it back up again in Word and saving it out as DOCX again. Then rebuild the book in Logos from that DOCX file source. If that doesn't clean up the blue text problem and split Scripture references that you're seeing, then you've found a genuine bug, which will be helpful for Logos Development to know about.

https://community.logos.com/discussion/comment/256279#Comment_256279

Hi Rosie thanks for your reply. I was going to try and insert a screen shot with the forum post. What format do I need to save a screen shot in to upload it into the post? I am running windows 7 with word 2007.

Steve Caswell

https://community.logos.com/discussion/comment/256283#Comment_256283

You can post a screenshot in any graphic image format (.jpg, .png, etc.). Just click the paperclip icon to attach it. See here for more info: http://wiki.logos.com/Screenshot

As for posting a Word document, you might want to zip it first. In Win 7, right click on the document name in Windows Explorer and choose "Send to > Compressed (zipped) folder" - then attach that zipped folder to a post here. Since it isn't an image file, delete the Max width and Max height values before clicking Insert:

https://community.logos.com/discussion/comment/256284#Comment_256284

Here goes I will try and attach the sreen shot of the word document and underneath it I have the Logos PBB that has not coppied accurately.

Steve Caswell

https://community.logos.com/discussion/comment/256289#Comment_256289

Unfortunately the screenshot of your document shows the part above where the problem occurs. You would have needed to scroll down a page more before taking the screenshot.

But really, there could be all kinds of stuff going on internally in the DOCX file that we cannot see from a screenshot. We'd need to see the XML code inside it to know what's going on. Would you be able to post the actual document itself (or better yet, a shortened version of it that contains just this problematic section, as it's easier to troubleshoot with a smaller document)?

EDIT: Also, as PBB is a beta feature this thread could be moved by Logos to the PC Beta forum.

https://community.logos.com/discussion/comment/256293#Comment_256293

Here goes again, I have not posted screen shots before.

Steve Caswell

https://community.logos.com/discussion/comment/256294#Comment_256294

Steve

From the screenshot it looks as though instances of "1 Timothy" are being inserted into the PBB document at various points which seems to be causing some confusion to the compiler. But, also from the screenshot, these are not visible in the docx file.

As Rosie suggested it would be useful to post the document itself - or an extract from it - to look as what is actually within it.

Graham

https://community.logos.com/discussion/comment/256295#Comment_256295

Thanks Graham, The document shown in the top screen shot is the DOCX File and has none of the errors that show in the PBB copy. I wonder why a lot of the text has copied across in the blue color like it is a link to a Scripture reference. That is quite strange.

Steve Caswell

https://community.logos.com/discussion/comment/256295#Comment_256295

Hi Steve

What I think is going on (and bear in mind I knew nothing about PBBs until a week ago!) is that there is something in the Docx file which isn't visible when viewed in Word which is generating the additional 1 Timothy instances.

As these are not properly terminated references, the PBB compiler treats all of the subsequent characters as part of the reference until it finds something that it recognises as a terminator - hence the long sections of blue text.

To diagnose this further we would really need a copy of the document (or at least a subset of it that demonstrates the problem) to examine as opposed to a screenshot.

Hope this makes sense

Graham

Mark Barnes

https://community.logos.com/discussion/comment/256321#Comment_256321

It's very likely that Graham is right. If you want to try and fix it yourself, you could try and clean the .docx file yourself, try the steps under: Cleaning your docx file

Rich DeRuiter

https://community.logos.com/discussion/comment/256294#Comment_256294

I haven't had much time to play with this until today again, and saw some of these problems too.

One case of yours parallels one of mine: abbreviating a Bible book without using a period (1 Tim 5:22) in your case).

That doesn't parallel all your occurrences, but might solve some issues, if you put the period in there.

For the others, follow the advice to strip the extra code through the work around described in the Wiki. (BTW, don't we also need to turn off/disable smart tags for this to work? I wasn't able to follow all the discussion, so correct me if I'm wrong.)

Rich DeRuiter

https://community.logos.com/discussion/comment/256465#Comment_256465

I know, I'm replying to myself. But this seemed cleaner than editing myself.

I just tried again, after saving to docx via WordPad. The results were
worse than before. It looks like nearly every instance of a Bible book
beginning with a number is affected. In one place I appear to have no
less than three passages overlapping the hyperlinks.

Here's what I just tried: opening and saving a document in WordPad from/to docx, then importing the file directly into the PBB. Note that none of the text has any mark-up or PBB codes. It's simply text.

I'm posting two screenshots in the first notice the hover is over Mark3:23. Preceding that reference is 211. Those numbers come from 2Corinthians 11:14, 1Peter 3:22 and 1Corinthians 7:5 (see second screenshot for full text).

Second Screen Shot:

EDIT: after opening in MSWord and resaving, the problems go away. This suggests a problem with the WordPad rendering of docx documents.

https://community.logos.com/discussion/comment/256465#Comment_256465

(BTW, don't we also need to turn off/disable smart tags for this to work? I wasn't able to follow all the discussion, so correct me if I'm wrong.)

Hi Richard

Yes, that is correct.

With Smart Tags enabled, Bible references get stripped out

Graham

https://community.logos.com/discussion/comment/256476#Comment_256476

Richard's reply made me go back and look at Steve's screen shots again and - while it still looks like some strange formatting - the culprits seem visible!

For example:

"who had the necessary spiritual qualifications. Godly character was the main quality required. Acts 6:1-5; 1 Timothy 3:1-13; Titus 1:5-9 in Acts 6 we find"

is rendered as:

"who had the necessary1 Timothy spiritual qualifications. Godly character was the main quality required. Acts 6:1-5; 3:1-13; Titus 1:5-9 in Acts 6 we find"

What has happened is that the words "1 Timothy" have been moved to earlier in the paragraph resulting in the effects seen.

Similarly in the next occurence, the "1 Tim" of "1 Timothy 5:22" is taken to before Acts 13:1-3.

So the question comes down to what is there in the actual XML making up the file which causes the PBB compiler to treat it in this way?

Hope this helps track it down.

Graham

https://community.logos.com/discussion/comment/256479#Comment_256479

I have discovered that the program has trouble distinguishing between people's names and Bible Book names. If the name of a person is a book with a number in front of it it seems to be a problem, ie 1 & 2 Timothy and 1 & 2 Peter. Once I abbreviated these scripture references it solved 90% of the problems.

Steve Caswell

https://community.logos.com/discussion/comment/256516#Comment_256516

I have discovered that the program has trouble distinguishing between people's names and Bible Book names. If the name of a person is a book with a number in front of it it seems to be a problem

That would surprise me. I doubt that the PBB compiler would have any smarts about what constitutes a person's name. I'm guessing that it's rather that editing that part of the document caused some invisible changes in the XML code behind the visible text.

Could you please post the actual document (attach the .docx file itself), instead of merely a screenshot image of it, and I'll see if cleaning the XML solves the rest of the problem.

https://community.logos.com/discussion/comment/256522#Comment_256522

Hi Rosie, I will try and attach the whole file. Here goes.0753.A Research Paper on Ordination - Folder Margins.docx

https://community.logos.com/discussion/comment/256526#Comment_256526

Hi, Steve. Something very strange is going on inside this document. I followed the procedure for "Cleaning your .docx file" and that seemed to clean up part of the problem, but even so there are a few places where the blue highlighted linking is messed up. I took a closer look at the first such location.

Here's what it looks like in Logos (notice the Scripture reference that got split up between two different places):

Here's what it looks like in the .docx file -- seems fine, at least superficially. The 1 Timothy 5:22 reference is all together, at the end of a paragraph:

But now let's look at the RTF version of this file and we can see that internally things are pretty chopped up:

I don't really know how to read RTF, but at least I could find where in the code the reference in question came, and I highlighted its pieces for you.

Finally, here's what is actually inside the .docx file (a docx file is really just a kind of zip file that contains other files, one of which is document.xml, which you can open if you rename the .docx file to a .zip extension and then unzip it and look in the word folder that inside that). Here is that section of the text in XML. It's even yuckier than the RTF, and this is after doing the XML cleanup routine:

I don't know how it got so gross in the first place (other than Word's XML-generation being buggy) or how to clean it up from this nasty state. It's not much different in the original .docx file you gave me (before I tried to clean up the XML on it):

I'm not sure what to say, other than to make sure Logos is aware of this problem. It's probably Word's fault to begin with, but perhaps there is something Logos can do to work around it. Or maybe there is some other XML cleanup process we can figure out among us that will fix this into readable XML that the Logos PBB compiler can handle. At the moment, I'm stumped, though.

https://community.logos.com/discussion/comment/256573#Comment_256573

Mark? Graham? Do you have any ideas on how to proceed from here?

MJ. Smith

https://community.logos.com/discussion/comment/256574#Comment_256574

The code Rosie highlighted in a more readable form:

.c9Y6TC { color:#0000ff; }
.cLUY2 { color:#ea8f0f; }
.c18YC0 { color:#d00020; }
.c5M9S0 { color:#000090; }

<w:r w:rsidRPr="00FC1D67">

                <w:rPr>

                    <w:rFonts w:ascii="Ebrima" w:hAnsi="Ebrima" w:cstheme="minorHAnsi"/>

                    <w:sz w:val="24"/>

                    <w:szCs w:val="24"/>

                </w:rPr>

                <w:t>1 Tim</w:t>

            </w:r>

            <w:r w:rsidR="00FC1D67" w:rsidRPr="00FC1D67">

                <w:rPr>

                    <w:rFonts w:ascii="Ebrima" w:hAnsi="Ebrima" w:cstheme="minorHAnsi"/>

                    <w:sz w:val="24"/>

                    <w:szCs w:val="24"/>

                </w:rPr>

                <w:t>othy</w:t>

            </w:r>

            <w:proofErr w:type="gramEnd"/>

            <w:r w:rsidRPr="00FC1D67">

                <w:rPr>

                    <w:rFonts w:ascii="Ebrima" w:hAnsi="Ebrima" w:cstheme="minorHAnsi"/>

                    <w:sz w:val="24"/>

                    <w:szCs w:val="24"/>

                </w:rPr>

                <w:t xml:space="preserve"> 5:22</w:t>

            </w:r>

            <w:r w:rsidR="00F25FC3" w:rsidRPr="00FC1D67">

                <w:rPr>

                    <w:rFonts w:ascii="Ebrima" w:hAnsi="Ebrima" w:cstheme="minorHAnsi"/>

                    <w:sz w:val="24"/>

                    <w:szCs w:val="24"/>

                </w:rPr>

                <w:t xml:space="preserve"> </w:t>

            </w:r>

        </w:p>

Two files I compiled tonight have similar unnecessary garbage in the .docx but they both processed correctly. I ran both your file and my Coptic Ruth commentary file through Stylus Studio which declared both of them to be well-formed XML. Notice that is "well-formed" not "elegant" and I didn't run a validation process