Date   

Re: Where do YOU go for high-quality information about DITA? #resources

Rodolfo M. Raya
 

Hi Kris,

In my computer's desktop there is a folder containing the full 1.3 specification, another one with the errata text plus a folder with errata grammars, a folder with dita-lwdita-master and a big one with DITA samples from different clients collected through the years. 

At https://ditatranslation.com/articles/index.html I published a few articles that I wrote or co-authored and are relevant for DITA localization. Some of those articles were also published by DITA Adoption TC or CIDM.

Those are my preferred consulting sources.

Regards,
Rodolfo


Re: How can we make use of the 'wiki' feature of groups.io? #admin

Kristen James Eberlein
 

You are really missing the boat here. Please see my comments inline.

Best,
Kris

Kristen James Eberlein
Chair, OASIS DITA Technical Committee
Principal consultant, Eberlein Consulting
www.eberleinconsulting.com
+1 919 622-1501; kriseberlein (skype)

On 1/12/2020 7:58 PM, Ron Wheeler wrote:
Is there any way to fix  http://docs.oasis-open.org/dita/v1.2/os/spec/DITA1.2-spec.html ?
<kje>The OASIS DITA specification is a legal specification. It cannot be modified in any way after it is released; a specific version can be superceded by an errata or a DITA release. Is there a reason that you referenced version 1.2? DITA 1.2 was released over nine years ago, and it has been replaced by DITA 1.3, DITA 1.3 Errata One, and DITA 1.3 Errata Two.</kje>

The descriptions are very cryptic and often seem almost self-referential.
The examples are often edge cases and only show one possible use of a language feature that can appear in many other contexts and have many attributes applied.

<kje>The DITA specification is NOT end-user documentation. Its primary audience is implementors who build implementations of the DITA spec, such as DITA-OT and ditac. Given that audience, we certainly do our best to cover edge cases, as edge cases are where implementors might make different implementation choices.</kje>

It appears to be no longer maintained. Most pages have not been touched in many years.

At the very least the examples should have been augmented to cover more of the features described in the specification.

https://www.dita-ot.org/3.4/topics/troubleshooting-overview.html is also another place where some work could be done.
I believe some error codes are missing in https://www.dita-ot.org/3.4/topics/error-messages.html

<kje>Ironically, the topic that you mention in the DITA-OT documentation (https://www.dita-ot.org/3.4/topics/error-messages.html) cannot be out-of-date. The topic is programmatically generated from the code that stores the error messages. Thus, if DITA-OT has an error message, the error message is present in the DITA topic and thus the documentation. Now, I'm sure that some explanations are missing or can be improved; please see https://www.dita-ot.org/contributing for information about how to suggest edits to the documentation.</kje>

https://www.dita-ot.org/3.4/topics/dita-and-dita-ot-resources.html is not up to date. This group is not mentioned for starters.

<kje>Yes, the dita-users group is mentioned at its then current location at Yahoo! Groups. At the time DITA-OT 3.4 was released, we were awaiting the transfer of the groups messages from Yahoo! to Groups.io. Please note the test version of the DITA-OT documentation correctly places the group at Groups.io.</kje>

I am not sure that adding another wiki is the best answer. Fixing and expanding some of the existing resources might be more productive.

It is very odd that a community of technical writers has such poor and disorganized documentation.


Re: Odd results with glossary elements and PDF generation #glossary #PDF

Radu Coravu
 

Hi Dan,

You need to explicitly set the "print='yes'" attribute on the glossref:

https://github.com/dita-ot/dita-ot/issues/3328

Regards,
Radu

Radu Coravu
<oXygen/> XML Editor
http://www.oxygenxml.com

On 1/10/2020 10:03 PM, Dan Vint wrote:
I've been following the instructions for using the glossary elements
https://www.oxygenxml.com/doc/versions/21.1/ug-editor/topics/dita-creating-glossary.html
This all works perfectly for HTML output. When I produce the associated PDF file though it fails. In the OT log I'm getting
   [keyref] file:/opt/ixiasoft/outgen/temp/Dita2Pdf.danvint.15080.1578671565937/content/authoring/npf1578670232926.dita:11:29: [DOTJ047I][INFO] Unable to find key definition for key reference "gAdapter" in root scope. The href attribute may be used as fallback if it exists
   [keyref] file:/opt/ixiasoft/outgen/temp/Dita2Pdf.danvint.15080.1578671565937/content/authoring/npf1578670232926.dita:12:33: [DOTJ047I][INFO] Unable to find key definition for key reference "gCertificate" in root scope. The href attribute may be used as fallback if it exists
   [keyref] file:/opt/ixiasoft/outgen/temp/Dita2Pdf.danvint.15080.1578671565937/content/authoring/npf1578670232926.dita:13:32: [DOTJ047I][INFO] Unable to find key definition for key reference "gExtensible" in root scope. The href attribute may be used as fallback if it exists
My topic has this content as a test
<p>We have a glossary that can be linked to with &lt;term keyref="gloss_ddl" />. We have terms like
<termkeyref="gAdapter"/>and
<termkeyref="gCertificate"/>and
<termkeyref="gExtensible"/>and
<termkeyref="gOpenid"/>and
<termkeyref="gPKI"/> as well as
<termkeyref="gWebservice"/>.</p>
I also have a map for the formal glossary content that uses the same keys
<topicref keyref="gAdapter" />
<topicref keyref="gCertificate" />
<topicref keyref="gExtensible"/>
<topicref keyref="gOpenid" />
<topicref keyref="gPKI" />
<topicref keyref="gWebservice" />
These are not found either.
I have a separate map with the definitions
   <glossref keyref="vjp1572627554113" keys="gAdapter" />
<glossref keyref="pof1572627554169" keys="gCertificate" />
<glossref keyref="adv1572627554227" keys="gExtensible" />
<glossref keyref="sui1572627554285" keys="gOpenid" />
<glossref keyref="oom1572627554342" keys="gPKI" />
<glossref keyref="ikt1572627554399" keys="gWebservice" />
It seems like the key processing in PDF is not using the glossrefs properly.


Re: How can we make use of the 'wiki' feature of groups.io? #admin

Ron Wheeler
 

Thanks.


I am just a longtime user and use Eclipse (Spring Tool Suite) as my authoring environment.
I use standalone DITA-OT as well as the Maven plug-ins to publish my documents.

I do get by through Google which finds many useful pieces of documentation.

This discussion seems to have been prompted by a desire to create a new repository for DITA documentation oriented to actual use of DITA.

My point is that the documentation on the core organizations should be improved before we start a new wiki out of frustration.

Ron


On 2020-01-12 8:14 p.m., Tom Magliery wrote:

Hi Ron,

The first link you mentioned cannot be "fixed" per se. At least, that page can never be changed. It is a specification, not "documentation".

FWIW, there is a newer version of the DITA specification--version 1.3. Many changes and improvements were made to the specification from version 1.2 to 1.3. You can find version 1.3 here:

http://docs.oasis-open.org/dita/dita/v1.3/dita-v1.3-part0-overview.html

If you have suggestions for new DITA features or about the specification in general, they should be addressed to the DITA Technical Committee. In particular if they pertain directly to the specification, you should send comments to the "dita-comment" mailing list:
https://www.oasis-open.org/committees/tc_home.php?wg_abbrev=dita#feedback

But if you are a DITA user (as opposed to someone developing applications that support DITA in some way) you may benefit more from reading documentation about your own DITA tools than from reading the DITA specification itself. 

mag

-- 
Tom Magliery


On 2020-01-12 16:58, Ron Wheeler wrote:

Is there any way to fix  http://docs.oasis-open.org/dita/v1.2/os/spec/DITA1.2-spec.html ?

The descriptions are very cryptic and often seem almost self-referential.
The examples are often edge cases and only show one possible use of a language feature that can appear in many other contexts and have many attributes applied.

It appears to be no longer maintained. Most pages have not been touched in many years.

At the very least the examples should have been augmented to cover more of the features described in the specification.

https://www.dita-ot.org/3.4/topics/troubleshooting-overview.html is also another place where some work could be done.
I believe some error codes are missing in https://www.dita-ot.org/3.4/topics/error-messages.html

https://www.dita-ot.org/3.4/topics/dita-and-dita-ot-resources.html is not up to date. This group is not mentioned for starters.

I am not sure that adding another wiki is the best answer. Fixing and expanding some of the existing resources might be more productive.

It is very odd that a community of technical writers has such poor and disorganized documentation.



Where do YOU go for high-quality information about DITA? #resources

Kristen James Eberlein
 

I wanted to take some of the content from a recent post and turn it in a more positive direction.

Question: Where do you go for high-quality information about DITA?

My personal response:

I use the DITA specification (as published at oasis-open.org) and the DITA Open Toolkit (DITA-OT) documentation heavily. I also use whitepapers published by the DITA Adoption Technical Committee, especially the one about troubleshooting authored by Bob Thomas (https://www.oasis-open.org/committees/download.php/53516/DITA13TroubleshootingArticle_FINAL_03jul14.pdf).

You? Curious minds want to know.

--
Best,
Kris

Kristen James Eberlein
Chair, OASIS DITA Technical Committee
Principal consultant, Eberlein Consulting
www.eberleinconsulting.com
+1 919 622-1501; kriseberlein (skype)


Re: How can we make use of the 'wiki' feature of groups.io? #admin

Tom Magliery
 

Hi Ron,

The first link you mentioned cannot be "fixed" per se. At least, that page can never be changed. It is a specification, not "documentation".

FWIW, there is a newer version of the DITA specification--version 1.3. Many changes and improvements were made to the specification from version 1.2 to 1.3. You can find version 1.3 here:

http://docs.oasis-open.org/dita/dita/v1.3/dita-v1.3-part0-overview.html

If you have suggestions for new DITA features or about the specification in general, they should be addressed to the DITA Technical Committee. In particular if they pertain directly to the specification, you should send comments to the "dita-comment" mailing list:
https://www.oasis-open.org/committees/tc_home.php?wg_abbrev=dita#feedback

But if you are a DITA user (as opposed to someone developing applications that support DITA in some way) you may benefit more from reading documentation about your own DITA tools than from reading the DITA specification itself. 

mag

-- 
Tom Magliery


On 2020-01-12 16:58, Ron Wheeler wrote:

Is there any way to fix  http://docs.oasis-open.org/dita/v1.2/os/spec/DITA1.2-spec.html ?

The descriptions are very cryptic and often seem almost self-referential.
The examples are often edge cases and only show one possible use of a language feature that can appear in many other contexts and have many attributes applied.

It appears to be no longer maintained. Most pages have not been touched in many years.

At the very least the examples should have been augmented to cover more of the features described in the specification.

https://www.dita-ot.org/3.4/topics/troubleshooting-overview.html is also another place where some work could be done.
I believe some error codes are missing in https://www.dita-ot.org/3.4/topics/error-messages.html

https://www.dita-ot.org/3.4/topics/dita-and-dita-ot-resources.html is not up to date. This group is not mentioned for starters.

I am not sure that adding another wiki is the best answer. Fixing and expanding some of the existing resources might be more productive.

It is very odd that a community of technical writers has such poor and disorganized documentation.



Re: How can we make use of the 'wiki' feature of groups.io? #admin

Ron Wheeler
 

Is there any way to fix  http://docs.oasis-open.org/dita/v1.2/os/spec/DITA1.2-spec.html ?

The descriptions are very cryptic and often seem almost self-referential.
The examples are often edge cases and only show one possible use of a language feature that can appear in many other contexts and have many attributes applied.

It appears to be no longer maintained. Most pages have not been touched in many years.

At the very least the examples should have been augmented to cover more of the features described in the specification.

https://www.dita-ot.org/3.4/topics/troubleshooting-overview.html is also another place where some work could be done.
I believe some error codes are missing in https://www.dita-ot.org/3.4/topics/error-messages.html

https://www.dita-ot.org/3.4/topics/dita-and-dita-ot-resources.html is not up to date. This group is not mentioned for starters.

I am not sure that adding another wiki is the best answer. Fixing and expanding some of the existing resources might be more productive.

It is very odd that a community of technical writers has such poor and disorganized documentation.


reltable loses navtitle when collection-type="family" is added #reltable

Ron Wheeler
 

I am using DITA to document the plans for condominiums in a condo association.

Many units have the same plan. The relatble links the plan to the condos bult with the plan and links each condo unit to the appropriate plan.


My original reltable without collection-type="family" worked in the sense that plans had a list of condos built using that plan and each condo had a link to the plan.
The headings came out on the PDF and were in the correct language.
The problem that I was trying to address was that each condo had its own navtitle on the plan which looked a bit silly and took a lot of space.

Adding collection-type="family" did get rid of the extra headings but now I have a "Related Concepts" title on the condo page that lists similar condos using the same plan.
The "Condominiums built with this plan" or "Copropriétés construite avec ce plan" on the condo page is replaced by English text "Related Concepts" which is a NO-NO in Quebec.

The "Condominiums built with this plan" or "Copropriétés construite avec ce plan" does appear on the Plan page once per plan rather than once per address which is great.

It looks like a bug where the navtitle is replaced on one side of the relation.

Can I add something to get my navtitle back or eliminate the list of similar condos on the condo unit page(suppress one side of the relationship)?


<reltable>
        <relheader>
             <relcolspec type="topic">
              <topichead>
                <topicmeta>
                  <navtitle audience="en">Condominiums built with this plan</navtitle>
                  <navtitle audience="fr">Copropriétés construite avec ce plan</navtitle>
                </topicmeta>
              </topichead>
            </relcolspec>
                <relcolspec type="topic">
              <topichead>
                <topicmeta>
                  <navtitle audience="en">Plan for this condominium</navtitle>
                  <navtitle audience="fr">Plan pour cette copropriété</navtitle>
                </topicmeta>
              </topichead>
            </relcolspec>
        </relheader>
        <relrow>
            <relcell collection-type="family">
            <topicref keyref="1_14588Aumais"/><topicref keyref="3_14588Aumais"/><topicref keyref="1_14628Aumais"/><topicref keyref="3_14628Aumais" />
            </relcell>
            <relcell><topicref keyref="leftodd12"/></relcell>
        </relrow>


Re: How To Increase DITA OT Performance #DITA-OT

Ron Wheeler
 

It may be that DITA is I/O bound rather than CPU bound. Adding more processors thrashing the same disk may not help.

I had very good results by creating a RAM drive and using that for document generation of a 100+ page manual. 

I currently use an SSD drive which seems to be very fast with a project of 150 dita files

It is easy to create a RAM drive and test this hypothesis.


Re: Odd results with glossary elements and PDF generation #glossary #PDF

Dan Vint
 

Thanks, I’ll take a look at when I’m back at work

 

From: dita-users@groups.io <dita-users@groups.io> On Behalf Of Wayne Brissette
Sent: Friday, January 10, 2020 12:56 PM
To: dita-users@groups.io
Subject: Re: [dita-users] Odd results with glossary elements and PDF generation

 

Dan:

 

I sent you a sample project that should get you started. I put it together for last year’s DITA North America presentation. 

 

-Wayne



On Jan 10, 2020, at 2:03 PM, Dan Vint <dvint@...> wrote:

 

I've been following the instructions for using the glossary elements

 

This all works perfectly for HTML output. When I produce the associated PDF file though it fails. In the OT log I'm getting 

 

   [keyref] file:/opt/ixiasoft/outgen/temp/Dita2Pdf.danvint.15080.1578671565937/content/authoring/npf1578670232926.dita:11:29: [DOTJ047I][INFO] Unable to find key definition for key reference "gAdapter" in root scope. The href attribute may be used as fallback if it exists

   [keyref] file:/opt/ixiasoft/outgen/temp/Dita2Pdf.danvint.15080.1578671565937/content/authoring/npf1578670232926.dita:12:33: [DOTJ047I][INFO] Unable to find key definition for key reference "gCertificate" in root scope. The href attribute may be used as fallback if it exists

   [keyref] file:/opt/ixiasoft/outgen/temp/Dita2Pdf.danvint.15080.1578671565937/content/authoring/npf1578670232926.dita:13:32: [DOTJ047I][INFO] Unable to find key definition for key reference "gExtensible" in root scope. The href attribute may be used as fallback if it exists

 

 


Re: Odd results with glossary elements and PDF generation #glossary #PDF

Wayne Brissette
 

Dan:

I sent you a sample project that should get you started. I put it together for last year’s DITA North America presentation. 

-Wayne

On Jan 10, 2020, at 2:03 PM, Dan Vint <dvint@...> wrote:

I've been following the instructions for using the glossary elements

This all works perfectly for HTML output. When I produce the associated PDF file though it fails. In the OT log I'm getting 

   [keyref] file:/opt/ixiasoft/outgen/temp/Dita2Pdf.danvint.15080.1578671565937/content/authoring/npf1578670232926.dita:11:29: [DOTJ047I][INFO] Unable to find key definition for key reference "gAdapter" in root scope. The href attribute may be used as fallback if it exists
   [keyref] file:/opt/ixiasoft/outgen/temp/Dita2Pdf.danvint.15080.1578671565937/content/authoring/npf1578670232926.dita:12:33: [DOTJ047I][INFO] Unable to find key definition for key reference "gCertificate" in root scope. The href attribute may be used as fallback if it exists
   [keyref] file:/opt/ixiasoft/outgen/temp/Dita2Pdf.danvint.15080.1578671565937/content/authoring/npf1578670232926.dita:13:32: [DOTJ047I][INFO] Unable to find key definition for key reference "gExtensible" in root scope. The href attribute may be used as fallback if it exists



Odd results with glossary elements and PDF generation #glossary #PDF

Dan Vint
 

I've been following the instructions for using the glossary elements

This all works perfectly for HTML output. When I produce the associated PDF file though it fails. In the OT log I'm getting 

   [keyref] file:/opt/ixiasoft/outgen/temp/Dita2Pdf.danvint.15080.1578671565937/content/authoring/npf1578670232926.dita:11:29: [DOTJ047I][INFO] Unable to find key definition for key reference "gAdapter" in root scope. The href attribute may be used as fallback if it exists
   [keyref] file:/opt/ixiasoft/outgen/temp/Dita2Pdf.danvint.15080.1578671565937/content/authoring/npf1578670232926.dita:12:33: [DOTJ047I][INFO] Unable to find key definition for key reference "gCertificate" in root scope. The href attribute may be used as fallback if it exists
   [keyref] file:/opt/ixiasoft/outgen/temp/Dita2Pdf.danvint.15080.1578671565937/content/authoring/npf1578670232926.dita:13:32: [DOTJ047I][INFO] Unable to find key definition for key reference "gExtensible" in root scope. The href attribute may be used as fallback if it exists


My topic has this content as a test

<p>We have a glossary that can be linked to with &lt;term keyref="gloss_ddl" />. We have terms like 

<term keyref="gAdapter" /> and 

<term keyref="gCertificate" /> and

<term keyref="gExtensible" /> and

<term keyref="gOpenid" /> and

<term keyref="gPKI" /> as well as

<term keyref="gWebservice"  />.</p>


I also have a map for the formal glossary content that uses the same keys


   <topicref keyref="gAdapter" />

   <topicref keyref="gCertificate" />

   <topicref keyref="gExtensible"  />

   <topicref keyref="gOpenid" />

   <topicref keyref="gPKI" />

   <topicref keyref="gWebservice" />


These are not found either. 


I have a separate map with the definitions


   <glossref keyref="vjp1572627554113" keys="gAdapter" />

   <glossref keyref="pof1572627554169" keys="gCertificate" />

   <glossref keyref="adv1572627554227" keys="gExtensible" />

   <glossref keyref="sui1572627554285" keys="gOpenid" />

   <glossref keyref="oom1572627554342" keys="gPKI" />

   <glossref keyref="ikt1572627554399" keys="gWebservice" />


It seems like the key processing in PDF is not using the glossrefs properly.




Re: Question about DITA launcher inside a Cygwin environment #Oxygen

Mica Semrick
 

I'd expect that since you are using windows, git, when run from oxygen, has no knowledge of your cygwin environment and thus cannot launch your script with bash.

I'd guess that you'd need to give the file an extension, like .sh, then associate the .sh mimetype with bash.

-m


On January 10, 2020 7:51:46 AM PST, Grant Hogarth <grant.hogarth@...> wrote:
Hello;
I'm trying to create a workflow that starts with multiple DITA files inside oXygen that end up as a single md file in GitHub.
I have installed the GitHub plugin (built on on the JGit library) in oXygen so that I can commit to a GitHub repository.
When I make the commit, I am trying to make a DITA-OT call using pre-commit githooks. The githooks script calls DITA-OT and uses it to merge the files before submitting them as a single file.
Unfortunately, this is not working.
The script (see below) works as expected when run from a GitBash or Windows command window. But running it from within oXygen appears to ignore the githooks.
Environment: Windows 10, Cygwin 3.0.7-1, DITA-OT 3.4, Java 11+28.
----
Githooks Script
#!/bin/sh set -x # In order to run this pre-commit hook the user will need the following setup # 1. The DITA-OT command utility needs to be added to the path # In Windows, typically this path is: "C:\DEV\DITA-OT\dita-ot-3.4\bin" # 2. Git needs to be installed on your computer. (https://git-scm.com/download/) # 3. Run 'git config core.hooksPath .githooks' in the repo folder to add the Git symlink. # There is no confirmation. # 3. Run 'chmod +x .githooks/pre-commit' in the repo folder to add the Git symlink. # There is no confirmation. # Populate all the .DITA* files for this commit ditafiles=$(git diff --cached --name-only --diff-filter=ACM | grep '.dita*') # Get the absolute path of the `.git/hooks` directory export GIT_HOOKS=`cd "\`dirname "\\\`readlink "$0" || echo $0\\\`"\`" && pwd` # Loop over the files, find their directory, and generate the README within that directory for arg in $ditafiles; do BASE_DIRECTORY=$(echo "$arg" | cut -d "/" -f1) # Set the absolute path of the build directory export BUILD_FILES="$GIT_HOOKS/../$BASE_DIRECTORY" # Now that we have the base path to the project we can run the DITA command to generate that specific README dita --input=$BUILD_FILES/README.ditamap --format=markdown -Dchunk=to-content --output=$BUILD_FILES/docs/ done # Exit with status of last command exit 0  
------
The JAVA_HOME variable user path (not system path, if that makes a difference) points to jdk-11.
Running the following in a Git Bash window works.
git add . 
git commit -m "bash commit"

As this is basically the commands generated from the other code sample above, I'm stumped, as is the support person at oXygen.
  The following is the response I got from the oXygen support person:  
I tried to use your script as a pre-commit hook, myself. I've started Oxygen from the command line launcher because I've noticed that if the pre-commit hook uses echo command then I will see its output inside Oxygen's console. When I committed from Oxygen, what appeared in the console was just this line:

CODE: SELECT ALL

Buildfile: \cygdrive\d\projects\oxygen\frameworks\dita\DITA-OT3.x\build.xml does not exist!
I've noticed that the dita script has some logic to identify when it's being executed in cygwin context, but I can't actually say what it's doing in there. I've made another small test:
- the dita script was added in my path, like we've previously discussed
- I've started a cygwin console
- I've execute the command:

CODE: SELECT ALL

dita --input=README.ditamap --format=markdown -Dchunk=to-content --output=docs/
I've received the same output:

CODE: SELECT ALL

Buildfile: \cygdrive\d\projects\eXml\frameworks\dita\DITA-OT3.x\build.xml does not exist!
Error: Error: Build failed
So it would appear that the dita launcher doesn't work inside a cygwin environment...    

Is this true, or are there other settings that need to be made?
Regards,
Grant Hogarth 
Technical Writer, SA Tools
Workiva Inc. 
1700 Platte St, Suite 200, Denver, Colorado 80202 
Mobile: 1-801-815-8353 


Re: Chapter and section numbering with Oxygen XML - DITA Map PDF based on HTML5 & CSS #CSS #Oxygen #PDF

Matt Lorenzi <mjlorenzi@...>
 

Hi Radu,
So originally my intent was to mirror the numbering outcome we achieved using FrameMaker. There we achieved numbering for all topics as well as the DITA <section>, so a topic might show like this:

3.0 Safety
3.1 Safety Wording/Symbols (DITA <section>)
3.2 General Warning and Cautions (DITA <section>)
3.3 Electrical Safety (DITA <section>)


So in the example above this is all in one topic, yet I achieve this numbering (both in the body and the TOC). I am curious how we achieved this numbering out of FrameMaker since this is not standard DITA-OT behaviour. Perhaps we had some custom scripting done at the time.

I guess to preserve this hierarchy in my document I will need to break out these <sections> into DITA topics. And if I just want to have non-hierarchical headings within a topic, I can continue to use <section><title></title></section>.

Thanks again!


Question about DITA launcher inside a Cygwin environment #Oxygen

 

Hello;
I'm trying to create a workflow that starts with multiple DITA files inside oXygen that end up as a single md file in GitHub.
I have installed the GitHub plugin (built on on the JGit library) in oXygen so that I can commit to a GitHub repository.
When I make the commit, I am trying to make a DITA-OT call using pre-commit githooks. The githooks script calls DITA-OT and uses it to merge the files before submitting them as a single file.
Unfortunately, this is not working.
The script (see below) works as expected when run from a GitBash or Windows command window. But running it from within oXygen appears to ignore the githooks.
Environment: Windows 10, Cygwin 3.0.7-1, DITA-OT 3.4, Java 11+28.
----
Githooks Script
#!/bin/sh set -x # In order to run this pre-commit hook the user will need the following setup # 1. The DITA-OT command utility needs to be added to the path # In Windows, typically this path is: "C:\DEV\DITA-OT\dita-ot-3.4\bin" # 2. Git needs to be installed on your computer. (https://git-scm.com/download/) # 3. Run 'git config core.hooksPath .githooks' in the repo folder to add the Git symlink. # There is no confirmation. # 3. Run 'chmod +x .githooks/pre-commit' in the repo folder to add the Git symlink. # There is no confirmation. # Populate all the .DITA* files for this commit ditafiles=$(git diff --cached --name-only --diff-filter=ACM | grep '.dita*') # Get the absolute path of the `.git/hooks` directory export GIT_HOOKS=`cd "\`dirname "\\\`readlink "$0" || echo $0\\\`"\`" && pwd` # Loop over the files, find their directory, and generate the README within that directory for arg in $ditafiles; do BASE_DIRECTORY=$(echo "$arg" | cut -d "/" -f1) # Set the absolute path of the build directory export BUILD_FILES="$GIT_HOOKS/../$BASE_DIRECTORY" # Now that we have the base path to the project we can run the DITA command to generate that specific README dita --input=$BUILD_FILES/README.ditamap --format=markdown -Dchunk=to-content --output=$BUILD_FILES/docs/ done # Exit with status of last command exit 0  
------
The JAVA_HOME variable user path (not system path, if that makes a difference) points to jdk-11.
Running the following in a Git Bash window works.
git add . 
git commit -m "bash commit"

As this is basically the commands generated from the other code sample above, I'm stumped, as is the support person at oXygen.
  The following is the response I got from the oXygen support person:  
I tried to use your script as a pre-commit hook, myself. I've started Oxygen from the command line launcher because I've noticed that if the pre-commit hook uses echo command then I will see its output inside Oxygen's console. When I committed from Oxygen, what appeared in the console was just this line:

CODE: SELECT ALL

Buildfile: \cygdrive\d\projects\oxygen\frameworks\dita\DITA-OT3.x\build.xml does not exist!
I've noticed that the dita script has some logic to identify when it's being executed in cygwin context, but I can't actually say what it's doing in there. I've made another small test:
- the dita script was added in my path, like we've previously discussed
- I've started a cygwin console
- I've execute the command:

CODE: SELECT ALL

dita --input=README.ditamap --format=markdown -Dchunk=to-content --output=docs/
I've received the same output:

CODE: SELECT ALL

Buildfile: \cygdrive\d\projects\eXml\frameworks\dita\DITA-OT3.x\build.xml does not exist!
Error: Error: Build failed
So it would appear that the dita launcher doesn't work inside a cygwin environment...    

Is this true, or are there other settings that need to be made?
Regards,
Grant Hogarth 
Technical Writer, SA Tools
Workiva Inc. 
1700 Platte St, Suite 200, Denver, Colorado 80202 
Mobile: 1-801-815-8353 


Re: Remove a Blank Preface Page #PDF

@ditaUserNYC
 

Yes, I am building a custom DITA-OT plug-in. I'm not modifying the master files. I'm just pointing out where to find it.


Re: Remove a Blank Preface Page #PDF

Kristen James Eberlein
 

Have you built a DITA-OT plug-in that you are using to override the DITA-OT defaults? Or are you modifying the DITA-OT files directly. The latter is a bad practice ...

Best,
Kris

Kristen James Eberlein
Chair, OASIS DITA Technical Committee
Principal consultant, Eberlein Consulting
www.eberleinconsulting.com
+1 919 622-1501; kriseberlein (skype)

On 1/10/2020 8:24 AM, cbgarcia@... wrote:
Found a solution finally.

Edit the front-matter-attr.xsl file to:
  <xsl:attribute-set name="back-cover">
    <xsl:attribute name="force-page-count">auto</xsl:attribute>
  </xsl:attribute-set>

<xsl:attribute-set name="__back-cover">
    <xsl:attribute name="break-before">auto</xsl:attribute>
  </xsl:attribute-set>


Re: Remove a Blank Preface Page #PDF

@ditaUserNYC
 

Found a solution finally.

Edit the front-matter-attr.xsl file to:
  <xsl:attribute-set name="back-cover">
    <xsl:attribute name="force-page-count">auto</xsl:attribute>
  </xsl:attribute-set>

<xsl:attribute-set name="__back-cover">
    <xsl:attribute name="break-before">auto</xsl:attribute>
  </xsl:attribute-set>


Re: Chapter and section numbering with Oxygen XML - DITA Map PDF based on HTML5 & CSS #CSS #Oxygen #PDF

Radu Coravu
 

Hi Matt,

I stumbled on this thread, as the thread is long it was an easy stumble :)

So our DITA to PDF using CSS documentation mentions:

https://www.oxygenxml.com/doc/versions/21.1/ug-editor/topics/dcpp_numbering_types.html
Deep
Each chapter and section is numbered (sections are prefixed with the chapter number). Figures, tables, and pages are numbered sequentially from the start of the publication and they do not reset.
but I think the "section" wording there was originally indented to mean "topic" located on any level and not DITA <section>. I will add an internal issue to clarify this in our documentation.

Indeed one possibility to move forward is to convert the sections in your topic to inner topics, we have an "Convert Nested Topics to New Topics" action available in the Oxygen DITA Maps Manager view.

https://www.oxygenxml.com/doc/versions/21.1/ug-editor/topics/dita-maps-manager.html

Do you want as an outcome the numbered sections to just appear numbered in the PDF content or also to appear in the table of contents and the bookmarks area?
Because if you do not want the sections to appear also in the TOC and the bookmarks area I think this numbering could be achieved using CSS.

I see you also started a thread on the Oxygen XML Forum, I will probably reply also on that one:

https://www.oxygenxml.com/forum/viewtopic.php?p=56674#p56615

Regards,
Radu

Radu Coravu
<oXygen/> XML Editor
http://www.oxygenxml.com

On 1/10/2020 12:21 AM, Matt Lorenzi via Groups.Io wrote:
I hope someone at Oxygen stumbles on this thread - regardless I think the definition for *<section>* from the Oasis website clears things up and affirms we may have been using it wrong:
3.1.1.2.31 section
The <section> element represents an organizational division in a topic. Sections are used to organize subsets of information that are directly related to the topic. For example, the titles Reference Syntax, Example and Properties might represent section-level discourse within a topic about a command-line process—the content in each section relates uniquely to the subject of that topic. Multiple sections within a single topic do not represent a hierarchy, but rather peer divisions of that topic. Sections cannot be nested. A section may have an optional title.
The "multiple sections within a single topic do not represent a hierarchy" bit definitely negates the need for numbering.


Re: Chapter and section numbering with Oxygen XML - DITA Map PDF based on HTML5 & CSS #CSS #Oxygen #PDF

Matt Lorenzi <mjlorenzi@...>
 

I hope someone at Oxygen stumbles on this thread - regardless I think the definition for <section> from the Oasis website clears things up and affirms we may have been using it wrong:
3.1.1.2.31 section

The <section> element represents an organizational division in a topic. Sections are used to organize subsets of information that are directly related to the topic. For example, the titles Reference SyntaxExample and Properties might represent section-level discourse within a topic about a command-line process—the content in each section relates uniquely to the subject of that topic. Multiple sections within a single topic do not represent a hierarchy, but rather peer divisions of that topic. Sections cannot be nested. A section may have an optional title.

The "multiple sections within a single topic do not represent a hierarchy" bit definitely negates the need for numbering. 

921 - 940 of 45963