chapter alone of the Handbook. Add an entity here, used with ⋼,
so that we can mark up each of those with <acronym>NIS</acronym>
without bloating the source SGML files too much. We shouldn't add
every such acronym here, but in the case of NIS it really helps the
readability.
callouts with consistent number icons. Callouts above number 15 would
still be displayed in bracketed text.
Obtained from: docbook-xsl distribution
Created by: Dave Pawson
is not defined.
- URL_RELPREFIX must always point the top page's URL
(http://www.FreeBSD.org) in relative form, defined in Makefile.
- &url.base; can be used in doc/<langcode>/*. It will be
replaced with the real URL which points the top page.
For translators: Please make sure to define URL_RELPREFIX to
point the top page. The localized docs have different directory
hierarchy in its URL from the English version.
Tested by: den
Replace &url.main; -> http://www.FreeBSD.org
when URLS_ABSOLUTE is set (for release building),
and -> ../../../.. by default (for mirror sites).
Patch was slightly modified by: hrs
Discussed with: des, ceri, hrs, trhodes, simon
- Add some new CVSUP sites in China
- Merge the two South Africa sections into one.
New China CVSUP sites provided by Xin Li <delphij (at) frontfree.net>
on behalf of The FreeBSD Simplified Chinese Project.
scaling model "got confused" and most of messages become unreadable.
Resolve this problem, by spliting of branches diagram into three
diagrams: head, releng3 and releng4. It reduces width usage and gives
us more control over scalling control.
The previous fix only worked in some cases, so back it out and add a
more clean fix. This fix also makes "make clean" work as expected again
for the shared PNG images.
The real problem is the assumption in doc.images.mk that the
${IMAGES_PDF} variable only need to contain images converted to PDF
format and not images which are already in PNG format. We need to list
the PNG images in ${IMAGES_PDF} since they might be in a shared image
directory, and if we don't list them in ${IMAGES_PDF} they will not be
copied to the working directory and will therefor not be found during
build.
Prompted by: phantom
After introduction of shared images concept it was not possible to realize
how images should be put into packages, so built images references were
used. It result up tgz file with lots filename which contained of '..'.
In order to workaround this issue I used temporary directory to install
complete packaging document, then generate PLIST based on content of
temporary directory and create actual package then.
Prodded by: The FreeBSD Russian Documentation Project
doc.docbook.mk includes bsd.subdir.mk via bsd.obj.mk which is responsible
for processing of 'SUBDIR' make variable. But since we are handling
'SUBDIR' here explicitly, doing same thing (second time!) via
bsd.subdir.mk's rules makes no sense and only adds disk IO overhead.
It reduces install recursive calls by factor 4.
for "BODY P". This seems to have no effect within other browsers, but
stops Netscape from displaying images on top of the document text.
PR: docs/48472
Submitted by: Alan Larson <larson@eng.paix.net>
in the docproj port. TEX, LATEX, PDFTEX, JADETEX, and PDFJADETEX
has been renamed to TEX_CMD, LATEX_CMD, PDFTEX_CMD, JADETEX_CMD, and
PDFJADETEX_CMD for consistency. And TEXCMDS has been renamed
TEX_CMDSEQ because TEXCMDS and TEX_CMD are misleading names.
Discussed on: -doc
- Use trademark entities.
- Add trademark attributions.
- Don't join trademarks with other words, e.g. using hyphens.
trademark.ent:
- Add trademark entity for Sun.
- The structure of transtable.xml is revised. <word> should be
bracketed with <group>.
- A sorting order of the FreeBSD mirror sites rendered using
mirrors.xml has been determined by sort(1).
- A template "transtable-lookup" has been added for localization on
word-by-word basis.
- Replace English month names in news.xml, press.xml, and
advisories.xml with numbers that correspond to the names.
The number->name translation is performed on the fly.
- Since information in mirrors.xml is used in www/ tree, it depends on
doc/ tree now. When WITHOUT_DOC is defined it can be built without
doc/ tree, but some information becomes unavailable. For example,
a list of the mirror sites generated in index.html becomes a dummy
one, and calling "transtable-lookup" with a word returns the word
itself.
Neither www/ nor doc/ build should be broken due to this commit, but
until the necessary changes are applied in the localized directories,
the transtable does not work; it simply generates non-localized contents
even if transtable.xsl is used.
- Add several templates for index.xsl into includes.misc.xsl.
- The structure of mirrors.xml has been changed; <host> is
a container of <name> and <url> now.
- add doc.common.mk, which defines variables, targets, and
dependencies commonly used in www/ and doc/.
- move www/<lang>/includes.xsl to the language independent
directory and split into several files.
- add transtable*.xsl and transtable.xml to support localized
mirror names.
- make doc/{en_US.ISO8859-1,ja_JP.eucJP}/books/handbook/ use
mirrors.xml (mirrors and eresources).
- make www/{en,ja}/index.xsl use mirrors.xml.
For details, please see doc/share/sgml/README.mirrors for the moment.
Reviewed by: simon and Alex Dupre <sysadmin@alexdupre.com>
of %show-all-trademark-symbols% from the HTML only section to the
general section.
Reported by: Jon Allen <jonallen@paunix.org>, bmah
Pointy hat to: simon
- Remove the text about keeping the trademark list in doc/ and the one
in www/ synchronized. The doc/ trademark list is too big, so it's
better just to add the needed trademark attributions in the www/
trademark file.
- Add a missing period in the VMware trademark attribution.
- Sort the entries in this file; somebody pass me an alphabet.
the lines at column 90, instead of the default 68. This makes the
resulting HTML considerably smaller.
The change also band-aid a problem where Tidy wrap lines which shouldn't
have been wrapped. This can cause extra spaces in the resulting HTML,
e.g. resulting in "( audio/lame)" instead of "(audio/lame)".
Discussed on: -doc
required symbols, by setting the make variable
WITH_ALL_TRADEMARK_SYMBOLS, to any non empty value.
This feature is mainly useful for document writers, to make sure all
trademarks are marked up correctly.
- Use trademark entities
- Add trademark attributions.
- Always use correct case for trademarks.
- Don't join trademarks with other words, e.g. using hyphens.
trademark.ent:
- Add entities for more trademarks / companies.
- Sort the Sun trademark attribution entity.
- Use trademark entities, including XFree86 now that the XFree86 Core
Team has responded with their trademark policy.
- Add trademark attributions.
- Always use correct case for trademarks.
- Don't join trademarks with other words, e.g. using hyphens.
- Don't use trademarks as nouns (e.g. don't use "Windows NT's").
trademark.ent:
- Add entities for more trademarks / companies.
- Fix the Xiph.Org entry.
- Use trademark entities.
- Add trademark attributions.
- Always use correct case for trademarks.
- Don't join trademarks with other words, e.g. using hyphens.
- Don't use trademarks as nouns (e.g. don't use "Windows NT's").
diskless-x/article.sgml:
- While I'm here anyway add trademark tags for all trademarks, now
that the stylesheet handles it right.
trademark.ent:
- Add entities for more trademarks / companies.
Also rename the 3Com trademark entity from t3com to tm.3com. This
should make the entity naming convention a bit more consitent. This is
done now since the entity is not used too many places yet.
empty and unclosed tags. I used the word "should" because sometimes it
does not work, however most of malformed tags are found now.
That idea came after a talk with Denis Peplin.
- Use trademark entities.
- Add trademark attributions.
- Always use correct case for trademarks.
- Don't join trademarks with other words, e.g. using hyphens.
- Use 'UNIX like systems' instead of 'UNIX systems'.
trademark.ent:
- Add entities for more trademarks / companies.
Each trademark will only have a trademark symbol appended the first
time in a document for print / html, and the first time on each HTML
page for html-split.
It is possible to always append the trademark symbol for a trademark,
by setting the role attribute to "force" in the trademark tag.
The code differentiates between trademarks in normal text and titles,
so if there is a trademark in a title, the symbol will be displayed
the first time in the title, and the first time in the normal text.
Basically this means that all trademarks can / should be marked up in
the documents, and the stylesheet will do "the right thing".
Looked at by: murray
Approved by: mwlucas (temp mentor)
- I added various trademarks needed in trademarks.ent but I did not
used them in the section. They should be added when the trademarks
tagging will be definitively defined.
PR: docs/55652 [1]
Submitted by: Dan Pelleg <daniel+handbook@pelleg.org> [1]
Add trademark symbols and attributions:
- Add trademark tags as needed, generally on first use in each
article.
- Add an attribution to the legal section mentioning all trademarks
referenced.
- Always use correct case for trademarks.
trademark.ent:
Add Macromedia, Inc trademark attribution.
share/sgml, since a lot of the entities are language independent, and
it makes work simpler for translators to have them in share/sgml.
The trademark.ent file contain:
- Commonly used trademarks defined as entities for ease of use.
- All trademark attributions defined as entities for ease of use.
Suggested by: hrs
Approved by: ceri (mentor)
