Date   

Each (un-nested) topic of my map outputs as a chapter

angelinab416
 

Why does my stylesheet make each topic in a map appear on its own page as a chapter?? Grrr.

<bookmap>
  <chapter href="map1"/>
  <chapter href="map2"/>
</bookmap>

<map title="map1">
  <topicref href="topic1"/>
  <topicref href="topic2"/>
</map>


<map title="map2">
  <topicref href="topic3"/>
  <topicref href="topic4"/>
</map>

In the output, topics 1-4 render as chapters. :(

In my mind, there should be two chapters as evidenced by my two chapter elements. I understand that the map and chapter kinda go away during processing, but certainly there's a way to get what I want without having to nest all my chapter topics under one topic, isn't there?

Kind regards,
Angelina

(Using v 1.6.3 with FOP)


Re: Task: same image for multiple steps

John Glass
 

borealforest,

I thought about using <stepsection>; the problem is that a <stepsection> actually relates to a single step, and, besides that, there can only be a single <stepsection> before each <step>, and I already intend to use <stepsection> to create notes about a step, when I need them: it would become tricky to handle multiple tags within <stepsection> each with different styling requirements: the image must be styled to be side by side with the steps, while a note must stay with the steps.


It is a common requirement to represent tasks side by side with images.

An it seems that the task content model does not cleanly allow for that.


My end goal is to express the relationship of an image with a set of steps, not with a step in particular (which is what <stepsection> suggests). The clearer the model, the easier the styling downstream.


How does the DITA community represent tasks steps and images side-by-side?

An image usually relates to multiple steps, and that should emerge from the model.


Thanks,

John


Help needed with DITA, CHM, and context-sensitive help

MIRIAM GOULD (TC) /6229 <miriam.gould@...>
 

I’ve run into a dead end with my attempt to finalize a conversion of content to DITA. I need the converted content to be context sensitive because it will be used for both standalone online help (as CHM) and a user guide. This is my first time working with context sensitivity, so everything about it is new for me. After doing a bunch of reading online, I’ve gotten a workflow to work involving using oXygen to transform my DITA content to CHM, importing the .hhp into RoboHelp, and then using RoboHelp to recompile my CHM with the .h file.

 

I would like to take the RoboHelp step out of the workflow because it’s an extra tool and feels like an extra complication for the authors, who are already having to deal with all the headaches of moving into DITA. I’m happy to use RoboHelp for the initial creation of the .h file because it’s easy to understand what I’m doing, but I feel like once I have a .h, the authors should be able to include it as part of the DITA-to-CHM transform whenever they’re doing a rev. I thought I could edit the args.htmlhelp.includefile parameter to point to the .h and have it all work, but it didn’t. I’ve seen references to using the parameter to point to an include.txt file instead that would contain the information about the .h and the .ali, but I couldn’t find any examples of what this include.txt file should look like.

 

Am I just wrong that this is something I should be able to do? Do we need to stick with the RoboHelp step until DITA 1.3 is released? Or is the include.txt file the missing step, and I just need to know how to create that?

 

Miriam

 

Miriam Robinson Gould

DITA Conversion Lead

Omnicell, Inc.

 


Re: Task: same image for multiple steps

peter.fournier@...
 

Does this help? It's valid.  


<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE dita PUBLIC "-//OASIS//DTD DITA Composite//EN" "ditabase.dtd">
<dita>
  <task id="task_snk_wyh_nm">
    <title>Task title</title>
    <taskbody>
      <steps>
        <stepsection>Descriptive paragraph <fig>
            <title>Figure inside stepsection</title>
          </fig>
        </stepsection>
        <step>
          <cmd>A step</cmd>
          <info>Information about the step<fig>
              <title>Figure inside info</title>
            </fig>
          </info>
          <substeps>
            <substep>
              <cmd>A SUB step </cmd>
              <info>Information about the substep<fig>
                  <title>Figure inside info in a substep</title>
                </fig>
              </info>
            </substep>
          </substeps>
        </step>
      </steps>
    </taskbody>
  </task>
</dita>


Re: Looking for a translation forum/mailing list

Frank.Ralf@...
 

Hi Matt,

For the technical side of translations, especially XML based, the best starting point is the open source Okapi Framework http://okapi.opentag.com

They also have a very responsive user group at http://groups.yahoo.com/neo/groups/okapitools/info

hth
Frank


Re: Colaboration and feedback #content-delivery

George Cristian Bina
 

One option is, like you mention, to review on the DITA output, and we provide a webhelp output with feedback support, this can be used by users to submit feedback on a specific topic.

However, I think an even better approach for review is to revire directly on the DITA document, thus you remove any additional work needed to synchronize the comments from the SMEs on the output and the DITA source. More, you can allow the SMEs to perform also simple changes to the document, not only comments, and then just accept or reject those propose changes. Some of our partners integrated the oXygen Author Component and they can offer such direct DITA review support, working inside the browse:
- Bluestream
- IXIASOFT
- Componize
- RSI
- Vasont
and others.

Best Regards,
George
--
George Cristian Bina
<oXygen/> XML Editor, Schema Editor and XSLT Editor/Debugger
http://www.oxygenxml.com

On 12/18/13, 3:26 PM, nmcdonagh wrote:
Hi all

Not sure this is the best place or not, but I am looking for ideas on
the best way to get feedback on published DITA content, so that we can
collaborate with SME's collect their comments, and then pull their
comments back into our DITA content. I know one option could be using
LiveContent with Reach etc but are there other options out there, right
now we are publishing from our CMS into Eclipse InfoCenters.

Thanks


Re: Looking for a translation forum/mailing list

Joe Pairman
 

Hi Matt,

Try proz.com. People there were very helpful when I had a question about Arabic nunation marks.

Keep your spirits up — the initial translation issues can feel a bit nightmarish, but things quickly settle down (at least if you have good translators who work with you to come up with solutions).

Cheers,
Joe


On Wed, Dec 18, 2013 at 11:49 PM, <mkaatman@...> wrote:
 

Does anyone know of a forum/mailing list or other resources that is active and dedicated to discussing translations and managing translation memory?


Every day translation introduces me to new nightmares.



Re: Colaboration and feedback #content-delivery

Joe Gelb <idg@...>
 

Hi Noel,

SuiteShare from Suite Solutions can also be used for SME reviewing.

Joe

 

From: dita-users@... [mailto:dita-users@...] On Behalf Of nmcdonagh
Sent: Wednesday, December 18, 2013 3:26 PM
To: dita-users@...
Subject: [dita-users] Colaboration and feedback

 

 

Hi all

Not sure this is the best place or not, but I am looking for ideas on the best way to get feedback on published DITA content, so that we can collaborate with SME's collect their comments, and then pull their comments back into our DITA content. I know one option could be using LiveContent with Reach etc but are there other options out there, right now we are publishing from our CMS into Eclipse InfoCenters.

Thanks

No virus found in this message.
Checked by AVG - www.avg.com
Version: 2014.0.4259 / Virus Database: 3658/6931 - Release Date: 12/18/13


Re: generating PDF

Don R. Day <donday@...>
 

Hi, Balaji. The simplest way to get started producing PDF using DITA-OT is to use the command-line dialog.

1. Assuming that you are using Windows, open your file explorer to the root folder of your DITA-OT installation.

2. Run the startcmd.bat file by doubleclicking on it.

3. Start the dialog by entering: ant -f build_demo.xml
A sample interaction might look like this (where the bold highlighting represents what you would enter):

Buildfile: C:\DITA-OT1.8.M2\build_demo.xml

prompt.init:


prompt:

     [echo] Please enter the filename for the DITA map that you
     [echo] want to build including the directory path (if any).
     [echo] The filename must have the .ditamap extension.
     [echo] Note that relative paths that climb (..) are not supported yet.
     [echo] To build the sample, press return without entering anything.
    [input] The DITA map filename:  [C:\DITA-OT1.8.M2\samples\hierarchy.ditamap]

samples\sequence.ditamap

     [echo]
     [echo] Please enter the name of the output directory or press return
     [echo] to accept the default.
    [input] The output directory (out):  [out]
out\samples\pdf
     [echo]
     [echo] Please enter the type of output to generate.
     [echo] Options include: eclipse, tocjs, htmlhelp, javahelp, pdf, or web
     [echo] Use lowercase letters.
     [echo]
    [input] The output type:  (eclipse, tocjs, htmlhelp, javahelp, pdf, [web], d
ocbook)
pdf
     [echo]
     [echo] Ready to build samples\sequence.ditamap
     [echo] for pdf in out
     [echo]
    [input] Continue?  (Y, [y], N, n)
y

These are the basic inputs for any PDF conversion, but you don't have to use just this method. As others have hinted, you can encode those input choices into an Ant build file that you can run at any time. For an example, see samples\ant_sample.xml which sets up exactly the same kind of build request. Run this build file by entering:

C:\DITA-OT1.8.M2>ant -f samples\ant_sample\sample_pdf.xml

The process and outcome should be identical to that of the interactive dialog.

Now you can copy this build file to another location and modify the properties (ditamap name, output directory, output process, etc.) and run it for any custom setup you wish.

I hope this was the kind of "getting started" help that you were looking for.

DITA and XML Consultant, Learning by Wrote
Co-Chair, OASIS DITA Technical Committee
"Where is the wisdom we have lost in knowledge?
Where is the knowledge we have lost in information?"
--T.S. Eliot

On 12/17/2013 12:01 AM, Balaji Kothandaraman wrote:
Hi,

I  would like to generate PDF using one of the ditamaps that I am using for doc purpose. Please advise me on the procedure to do the same. Also please let me know if I have to make changes to specific settings as well to generate the PDF.
Any help would be much appreciated.

Thanks,
Balaji K


--
  • "Where is the wisdom we have lost in knowledge?
  • Where is the knowledge we have lost in information?"
  • --T.S. Eliot


Looking for a translation forum/mailing list

mkaatman@...
 

Does anyone know of a forum/mailing list or other resources that is active and dedicated to discussing translations and managing translation memory?


Every day translation introduces me to new nightmares.


Re: Last page of PDF

Severin Foreman
 

Hi Anna,

I'm not sure how you would hide the back page. When I implemented the back-matter page sequence, we used the last page of the document to display the company logo and contact information. There is another technique to insert a marker into the last page of the document without adding a new page sequence, which is described in this post.


You might want to read Kevin's cautions on using the technique, but it might be your only option for FOP. 

Best,
Severin


From: "afloreanini@..."
To: dita-users@...
Sent: Wednesday, December 18, 2013 6:43 AM
Subject: [dita-users] RE: Last page of PDF



Hi Severin,
your first hint was useful, I realized that I was modifying the wrong root template. I'm using the FOP engine so the correct file is root-processing_fop.xsl.
Now I can get the total document pages, but I have a side effect: in the back of my document there is a new blank page that I'm not able to hide.

In other words, I need a "fake" back page: only for defining  the "endofdoc" block. Following is my template:


   
     
       
     
   
 
Many thanks,
Anna




Re: Colaboration and feedback #content-delivery

Jeremy H. Griffith <jeremy@...>
 

On Wed, 18 Dec 2013 13:26:06 -0000, "nmcdonagh"
<nmcdonagh@yahoo.com> wrote:

Not sure this is the best place or not, but I am
looking for ideas on the best way to get feedback
on published DITA content, so that we can collaborate
with SME's collect their comments,
Most SMEs on Windows seem to prefer Word for review.
DIYA2Go (free) can produce Word files with Track Changes
locked on, so it is easy for them to comment and for you
to find the comments.

and then pull their comments back into our DITA content.
That you'd need to do by hand, with copy/paste. But
SME comments are rarely appropriate for insertion
as-is anyway. ;-)

I know one option could be using LiveContent with Reach
etc but are there other options out there, right now we
are publishing from our CMS into Eclipse InfoCenters.
You may still need to do that on non-Windows systems.


-- Jeremy H. Griffith <jeremy@omsys.com>
DITA2Go site: http://www.dita2go.com/


Re: Colaboration and feedback #content-delivery

mkaatman@...
 

Antidot's fluid topics solution has some pretty nice functionality similar to what you're looking for.


Re: Topicrefs pointing to non-DITA content fail: tocjs, d4p html2, Oxygen webhelp

Julio J Vazquez
 

Hi Mark,

As I understand it, @scope="local" generally means that the referenced topics are built at the same time as the current context. In general, topicrefs with this scope should be DITA files.

The @scope="peer" was designed specifically for the use case you describe, when content is built by different processes and you're pointing to results external to the current build. As I remember things, this was done specifically for Eclipse information centers which would frequently have this requirement.

I temper this with the fact that my memory is sometimes suspect but I think just changing @scope to external should resolve your problems as @scope="external" I think expects a URI..

HTH,

Julio J. Vazquez
Write Sprit
http://www.writespritservices.com


Re: Topicrefs pointing to non-DITA content fail: tocjs, d4p html2, Oxygen webhelp

ekimber@contrext.com
 

Can you open an issue at http://sourceforge.net/p/dita4publishers/bugs/
and provide a sample data set I can test with? Are the HTML files
parseable as XML? That is, can I apply an identity transform to them?

The other option would be to use the same technique I use for images,
which is to generate an Ant script that simply copies the files to the
output location.

Cheers,

E.
—————
Eliot Kimber, Owner
Contrext, LLC
http://contrext.com

On 12/18/13, 8:52 AM, "Mark Peters" <markpeters.work@gmail.com> wrote:







Hi Eliot,

Yes, your assumptions are correct on all counts. I'll admit you're right,
it's a strange process. We're selling a commercially packaged version of
open source software (which we sponsor). The commercial docs include all
of the open source docs, which are saved as a set of HTML pages. These
pages are self-contained. The TOC should reference them. And if the HTML2
plugin copies the HTML pages to the output directory, I don't see why the
processed DITA topics couldn't reference the precompiled HTML pages, too.
But these pages would have no "knowledge" of the DITA topics.


Thanks,
Mark




On Wed, Dec 18, 2013 at 9:26 AM, Eliot Kimber <ekimber@contrext.com>
wrote:








So the HTML pages are in with the other DITA source (that is, they
are
literally local files, not on some Web site somewhere?).

If so, then I think that with a scope of “local” your expectation is
reasonable, if a little strange, meaning that since the pages are produced
separately, there’s no obvious way to coordinate them with your DITA
pages, at least so that they could reliably link into your DITA source,
but maybe that’s not a requirement?

But if I understand, you essentially want the generated ToC to include
links to these separately-provided HTML pages, which need to be copied to
the same output location as the HTML generated from the DITA source?


Cheers,

E.
—————
Eliot Kimber, Owner
Contrext, LLC
http://contrext.com


On 12/18/13, 7:10 AM, "Mark Peters" <markpeters.work@gmail.com> wrote:







Hi Eliot,

Was your expectation that navigation (ToC) links would be to those
resources as they exist outside the scope of the publication?

If so they should have a @scope value of “external”.

My expectation (hope) was that the set of referenced HTML files would be
copied to the output directory and referenced in the TOC alongside the
processed DITA topics. To that end, @scope is set to the default value,
"local." If I set the value to "external," wouldn't the referenced
preformed HTML pages be opened in a separate browser window? I'd like the
complete set of content (DITA topics and preformed HTML pages) to appear
as seamless as possible.

Thanks,
Mark




On Tue, Dec 17, 2013 at 10:07 AM, Eliot Kimber <ekimber@contrext.com>
wrote:








How are you coding the topic refs? Do they have scope=“local” or
scope=“external”?

The meaning of navigation topic refs to non-DITA resources is necessarily
fuzzily defined in the spec. It’s allowed, but what the presentation
effect should be is very much processor dependent.

Was your expectation that navigation (ToC) links would be to those
resources as they exist outside the scope of the publication?

If so they should have a @scope value of “external”.

For D4P I’m sure I never considered this case. If you can specify clearly
what the behavior should be I can try to accommodate it.

Cheers,

E.
—————
Eliot Kimber, Owner
Contrext, LLC
http://contrext.com


On 12/17/13, 8:47 AM, "Mark Peters" <markpeters.work@gmail.com> wrote:







Hi,

The DITA 1.2 spec seems to allow writers to point topicref @href
attributes to HTML or XHTML files. The @format attribute accepts "HTML"
as a value. One of my company's products builds on an open source
product
that we sponsor. As a prerequisite step to building output for the
commercial product documentation, we pull in the HTML open source doc as
a set of pages. My ditamap points to these pages. The pages are all
valid
HTML.

When I build output using DITA-OT 1.7.5 and the xhtml transformation,
the
build is successful. The index.html file includes links to all of the
preformed HTML files. The links open the referenced files.


We'd like to display a left-nav bar with a collapsible TOC -- a standard
two- or three-pane frameset or HTML5 display. But I've tried three
separate transformations to produce this output, without success.

* The tocjs transformation produces a successful build. No errors or
warnings. The left-nav TOC displays the navtitles for the preformed HTML
files. But the links are static. The toctree.js file includes the
@navtitle values, but the @href values are empty.

* The DITA4Publishers HTML2 transformation produces a successful build.
No errors. The left-nav TOC does not list any of the mapped preformed
HTML files. Instead, there are many warnings such as:

[xslt] + [WARNING] generate-graphic-map, get-graphic-refs: Failed to
resolve topic reference to href
[xslt] "rabbit-web-docs/access-control.html"

* The OxygenXML Author 14 webhelp transformation would probably work for
our purposes. The left-nav TOC actually works correctly. But all
searches
from the Search tab in Firefox produce "No results for <x>" messages.
(The search tab is not clickable in Chrome or IE.) The issues don't
appear to be related to the mapped HTML content.


Have you successfully built two- or three-pane HTML output that includes
preformed HTML content? If so, how did you do it?


We haven't finalized on DITA as an authoring markup language and so
aren't ready to invest a lot in transformation software yet.

Thanks for any help.

Mark

--
Senior technical writer
Pivotal




































--
Senior technical writer
Pivotal


































--
Senior technical writer
Pivotal









Re: Topicrefs pointing to non-DITA content fail: tocjs, d4p html2, Oxygen webhelp

Mark Peters
 

Hi Eliot,

Yes, your assumptions are correct on all counts. I'll admit you're right, it's a strange process. We're selling a commercially packaged version of open source software (which we sponsor). The commercial docs include all of the open source docs, which are saved as a set of HTML pages. These pages are self-contained. The TOC should reference them. And if the HTML2 plugin copies the HTML pages to the output directory, I don't see why the processed DITA topics couldn't reference the precompiled HTML pages, too. But these pages would have no "knowledge" of the DITA topics.

Thanks,
Mark



On Wed, Dec 18, 2013 at 9:26 AM, Eliot Kimber <ekimber@...> wrote:
 

So the HTML pages are in with the other DITA source (that is, they are
literally local files, not on some Web site somewhere?).

If so, then I think that with a scope of “local” your expectation is
reasonable, if a little strange, meaning that since the pages are produced
separately, there’s no obvious way to coordinate them with your DITA
pages, at least so that they could reliably link into your DITA source,
but maybe that’s not a requirement?

But if I understand, you essentially want the generated ToC to include
links to these separately-provided HTML pages, which need to be copied to
the same output location as the HTML generated from the DITA source?



Cheers,

E.
—————
Eliot Kimber, Owner
Contrext, LLC
http://contrext.com

On 12/18/13, 7:10 AM, "Mark Peters" <markpeters.work@...> wrote:

>
>
>
>
>
>
> Hi Eliot,
>
>> Was your expectation that navigation (ToC) links would be to those
>resources as they exist outside the scope of the publication?
>
>> If so they should have a @scope value of “external”.
>
>
>My expectation (hope) was that the set of referenced HTML files would be
>copied to the output directory and referenced in the TOC alongside the
>processed DITA topics. To that end, @scope is set to the default value,
>"local." If I set the value to "external," wouldn't the referenced
>preformed HTML pages be opened in a separate browser window? I'd like the
>complete set of content (DITA topics and preformed HTML pages) to appear
>as seamless as possible.
>
>Thanks,
>Mark
>
>
>
>
>On Tue, Dec 17, 2013 at 10:07 AM, Eliot Kimber <ekimber@...>
>wrote:
>
>
>
>
>
>
>
>
> How are you coding the topic refs? Do they have scope=“local” or
>scope=“external”?
>
>The meaning of navigation topic refs to non-DITA resources is necessarily
>fuzzily defined in the spec. It’s allowed, but what the presentation
>effect should be is very much processor dependent.
>
>Was your expectation that navigation (ToC) links would be to those
>resources as they exist outside the scope of the publication?
>
>If so they should have a @scope value of “external”.
>
>For D4P I’m sure I never considered this case. If you can specify clearly
>what the behavior should be I can try to accommodate it.
>
>Cheers,
>
>E.
>—————
>Eliot Kimber, Owner
>Contrext, LLC
>http://contrext.com
>
>
>On 12/17/13, 8:47 AM, "Mark Peters" <markpeters.work@...> wrote:
>
>>
>>
>>
>>
>>
>>
>> Hi,
>>
>>The DITA 1.2 spec seems to allow writers to point topicref @href
>>attributes to HTML or XHTML files. The @format attribute accepts "HTML"
>>as a value. One of my company's products builds on an open source product
>>that we sponsor. As a prerequisite step to building output for the
>>commercial product documentation, we pull in the HTML open source doc as
>>a set of pages. My ditamap points to these pages. The pages are all valid
>>HTML.
>>
>>When I build output using DITA-OT 1.7.5 and the xhtml transformation, the
>>build is successful. The index.html file includes links to all of the
>>preformed HTML files. The links open the referenced files.
>>
>>
>>We'd like to display a left-nav bar with a collapsible TOC -- a standard
>>two- or three-pane frameset or HTML5 display. But I've tried three
>>separate transformations to produce this output, without success.
>>
>>
>
>>* The tocjs transformation produces a successful build. No errors or
>>warnings. The left-nav TOC displays the navtitles for the preformed HTML
>>files. But the links are static. The toctree.js file includes the
>>@navtitle values, but the @href values are empty.
>>
>>
>
>>* The DITA4Publishers HTML2 transformation produces a successful build.
>>No errors. The left-nav TOC does not list any of the mapped preformed
>>HTML files. Instead, there are many warnings such as:
>>
>>[xslt] + [WARNING] generate-graphic-map, get-graphic-refs: Failed to
>>resolve topic reference to href
>>[xslt] "rabbit-web-docs/access-control.html"
>>
>>
>
>>* The OxygenXML Author 14 webhelp transformation would probably work for
>>our purposes. The left-nav TOC actually works correctly. But all searches
>>from the Search tab in Firefox produce "No results for " messages.
>>(The search tab is not clickable in Chrome or IE.) The issues don't
>>appear to be related to the mapped HTML content.
>>
>>
>>Have you successfully built two- or three-pane HTML output that includes
>>preformed HTML content? If so, how did you do it?
>>
>>
>>We haven't finalized on DITA as an authoring markup language and so
>>aren't ready to invest a lot in transformation software yet.
>>
>>Thanks for any help.
>>
>>Mark
>>
>>--
>>Senior technical writer
>>Pivotal
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>--
>Senior technical writer
>Pivotal
>
>
>
>
>
>
>
>
>




--
Senior technical writer
Pivotal


Task: same image for multiple steps

John Glass
 

Hi,

I need to represent procedural content where a single task is split in subtasks sharing the same prerequisites.

I think have two options: either (a) creating multiple tasks, duplicating or conref'ing the prerequisites; or (b) configuring a new shell by allowing an unbounded number of <steps> elements within taskbody.

However, if I go for (b) I noticed that <steps> extends <ol>, so there will be no way to include a figure within <steps> because <ol> cannot accept anything else than <li> (<step>) in its content model.

I'm looking for suggestions.
Thanks.
-John


Re: Last page of PDF

afloreanini@...
 

Hi Severin,
your first hint was useful, I realized that I was modifying the wrong root template. I'm using the FOP engine so the correct file is root-processing_fop.xsl.
Now I can get the total document pages, but I have a side effect: in the back of my document there is a new blank page that I'm not able to hide.

In other words, I need a "fake" back page: only for defining  the "endofdoc" block. Following is my template:

<xsl:template name="createBackMatter">
    <fo:page-sequence master-reference="back-matter" format="1">
      <fo:flow flow-name="xsl-region-body">
        <fo:block id="endofdoc"/>
      </fo:flow>
    </fo:page-sequence>
  </xsl:template>

Many thanks,
Anna


Re: Topicrefs pointing to non-DITA content fail: tocjs, d4p html2, Oxygen webhelp

ekimber@contrext.com
 

So the HTML pages are in with the other DITA source (that is, they are
literally local files, not on some Web site somewhere?).

If so, then I think that with a scope of “local” your expectation is
reasonable, if a little strange, meaning that since the pages are produced
separately, there’s no obvious way to coordinate them with your DITA
pages, at least so that they could reliably link into your DITA source,
but maybe that’s not a requirement?

But if I understand, you essentially want the generated ToC to include
links to these separately-provided HTML pages, which need to be copied to
the same output location as the HTML generated from the DITA source?

Cheers,

E.
—————
Eliot Kimber, Owner
Contrext, LLC
http://contrext.com

On 12/18/13, 7:10 AM, "Mark Peters" <markpeters.work@gmail.com> wrote:







Hi Eliot,

Was your expectation that navigation (ToC) links would be to those
resources as they exist outside the scope of the publication?

If so they should have a @scope value of “external”.

My expectation (hope) was that the set of referenced HTML files would be
copied to the output directory and referenced in the TOC alongside the
processed DITA topics. To that end, @scope is set to the default value,
"local." If I set the value to "external," wouldn't the referenced
preformed HTML pages be opened in a separate browser window? I'd like the
complete set of content (DITA topics and preformed HTML pages) to appear
as seamless as possible.

Thanks,
Mark




On Tue, Dec 17, 2013 at 10:07 AM, Eliot Kimber <ekimber@contrext.com>
wrote:








How are you coding the topic refs? Do they have scope=“local” or
scope=“external”?

The meaning of navigation topic refs to non-DITA resources is necessarily
fuzzily defined in the spec. It’s allowed, but what the presentation
effect should be is very much processor dependent.

Was your expectation that navigation (ToC) links would be to those
resources as they exist outside the scope of the publication?

If so they should have a @scope value of “external”.

For D4P I’m sure I never considered this case. If you can specify clearly
what the behavior should be I can try to accommodate it.

Cheers,

E.
—————
Eliot Kimber, Owner
Contrext, LLC
http://contrext.com


On 12/17/13, 8:47 AM, "Mark Peters" <markpeters.work@gmail.com> wrote:







Hi,

The DITA 1.2 spec seems to allow writers to point topicref @href
attributes to HTML or XHTML files. The @format attribute accepts "HTML"
as a value. One of my company's products builds on an open source product
that we sponsor. As a prerequisite step to building output for the
commercial product documentation, we pull in the HTML open source doc as
a set of pages. My ditamap points to these pages. The pages are all valid
HTML.

When I build output using DITA-OT 1.7.5 and the xhtml transformation, the
build is successful. The index.html file includes links to all of the
preformed HTML files. The links open the referenced files.


We'd like to display a left-nav bar with a collapsible TOC -- a standard
two- or three-pane frameset or HTML5 display. But I've tried three
separate transformations to produce this output, without success.

* The tocjs transformation produces a successful build. No errors or
warnings. The left-nav TOC displays the navtitles for the preformed HTML
files. But the links are static. The toctree.js file includes the
@navtitle values, but the @href values are empty.

* The DITA4Publishers HTML2 transformation produces a successful build.
No errors. The left-nav TOC does not list any of the mapped preformed
HTML files. Instead, there are many warnings such as:

[xslt] + [WARNING] generate-graphic-map, get-graphic-refs: Failed to
resolve topic reference to href
[xslt] "rabbit-web-docs/access-control.html"

* The OxygenXML Author 14 webhelp transformation would probably work for
our purposes. The left-nav TOC actually works correctly. But all searches
from the Search tab in Firefox produce "No results for <x>" messages.
(The search tab is not clickable in Chrome or IE.) The issues don't
appear to be related to the mapped HTML content.


Have you successfully built two- or three-pane HTML output that includes
preformed HTML content? If so, how did you do it?


We haven't finalized on DITA as an authoring markup language and so
aren't ready to invest a lot in transformation software yet.

Thanks for any help.

Mark

--
Senior technical writer
Pivotal




































--
Senior technical writer
Pivotal









Colaboration and feedback #content-delivery

nmcdonagh
 

Hi all

Not sure this is the best place or not, but I am looking for ideas on the best way to get feedback on published DITA content, so that we can collaborate with SME's collect their comments, and then pull their comments back into our DITA content. I know one option could be using LiveContent with Reach etc but are there other options out there, right now we are publishing from our CMS into Eclipse InfoCenters.

Thanks

12761 - 12780 of 46276