Boost logo

Boost :

Subject: Re: [boost] [release] 1.69.0 deadline for new libraries approaching
From: Steven Watanabe (watanabesj_at_[hidden])
Date: 2018-10-23 15:16:00


AMDG

On 10/23/2018 08:54 AM, Robert Ramey via Boost wrote:
> On 10/22/18 9:46 PM, Edward Diener via Boost wrote:
>> On 10/23/2018 12:33 AM, Robert Ramey via Boost wrote:
>>> On 10/22/18 8:02 PM, Edward Diener via Boost wrote:
>>>> 5. Building a library's documentation with CMake. Quite a number of
>>>> libraries currently use the Quickbook -> Boostbook -> html/pdf cycle
>>>> with built-in support by Boost Build for this with a jamfile.
>>>
>>> Maybe - but personally I just build my docs pdf and html with a
>>> simple shell script - it's about 5 lines long - easy as pie.
>>
>> 1) Shell scripts are different for Windows, Linux, and Mac OSs.
>> 2) Please provide your "simple" shell script for any OS, much less all
>> of them, that encapsulates the quickbook -> boostbook -> html/pdf
>> process in jam files.
>>
>> Try to understand that even if you do not use the quickbook ->
>> boostbook -> html/pdf build process a great number of libraries do use
>> it, for good reason, and those libraries are not going to re-invent
>> the wheel just because you have an alternate method for creating
>> documentation.
>
> Here is the the script I use to build html documentation for the safe
> numerics library:
>
> if test x = x$BOOST_ROOT
> then
>     echo BOOST_ROOT not set
>     exit 1
> fi
> xsltproc --nonet --xinclude bb2db.xsl safe_numerics.xml | xsltproc
> --nonet db2html.xsl -

You need a catalog file for --nonet to work. This script
is not self-contained, because it assumes that the
catalogs for boostbook and docbook are present in the
environment.

> cp pre-boost.jpg ../html
> cp pre-boost.jpg ../html/eliminate_runtime_penalty
> cp pre-boost.jpg ../html/promotion_policies
> cp pre-boost.jpg ../html/tutorial
> cp StepperMotor.gif ../html/
> cp stepper_profile.png ../html/
> cp $BOOST_ROOT/doc/src/boostbook.css ../html
> cp -R $BOOST_ROOT/doc/html/images ../html
>
> If you want to use it feel free.
>
> Note that it transforms boostbook => html.  How your boostbook file(s)
> are created is not addressed here. if you want to include quickbook in
> the pipeline or use something like it windows, You'll just have to alter
> the script to taste.  Sorry I can't do this for you.
>

And that's the real issue. Scripts like this are
work great as long as you're only running them in
a specific environment. If you try to make them
portable, they'll quickly become at least as complex
as the b2 implementation.

> Note that it's actual function is to cut B2 out of the mix. This makes
> things much, much simpler.
>
> BTW there are other versions to produce pdf and ebook from the boostbook
> files.
>

In Christ,
Steven Watanabe


Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk