Subject: CVS commit: pkgsrc/textproc/docbook-xsl
To: None <pkgsrc-changes@netbsd.org>
From: Thomas Klausner <wiz@netbsd.org>
List: pkgsrc-changes
Date: 05/30/2006 22:13:52
Module Name:	pkgsrc
Committed By:	wiz
Date:		Tue May 30 22:13:52 UTC 2006

Modified Files:
	pkgsrc/textproc/docbook-xsl: Makefile PLIST distinfo
Removed Files:
	pkgsrc/textproc/docbook-xsl/patches: patch-aa patch-ab patch-ac

Log Message:
Update to 1.70.1. Make pkglint happy.

Release: 1.70.1

This is a stable release of the 1.70 stylesheets. It includes only a few small
changes from 1.70.0.

The following is a list of changes that have been made since the 1.70.0
release.

FO

The following changes have been made to the fo code since the 1.70.0 release.

  * Added three new attribute sets (revhistory.title.properties,
    revhistory.table.properties and revhistory.table.cell.properties) for
    controlling appearance of revhistory in FO output.

  * Support DBv5 revisions with full author name (not only authorinitials)

HTML

The following changes have been made to the html code since the 1.70.0
release.

  * Support DBv5 revisions with full author name (not only authorinitials)

HTMLHelp

The following changes have been made to the htmlhelp code since the 1.70.0
release.

  * htmlhelp.generate.index is now param, not variable. This means that you
    can override its setting from outside. This is useful when you generate
    indexterms on the fly (see
    http://www.xml.com/pub/a/2004/07/14/dbndx.html?page=3).

  * Support chunk.tocs.and.lots in HTML Help

Params

The following changes have been made to the params code since the 1.70.0
release.

  * Added three new attribute sets (revhistory.title.properties,
    revhistory.table.properties and revhistory.table.cell.properties) for
    controlling appearance of revhistory in FO output.

Release: 1.70.0

As with all DocBook Project dot-zero releases, this is an experimental
release. It will be followed shortly by a stable release.

This release adds a number of new features, including:

  * support for selecting alternative index-collation methods (in particular,
    support for using a collation library developed by Eliot Kimber)

  * improved handling of DocBook 5 document instances (through a
    namespace-stripping mechanism)

  * full support for CALS and HTML tables in manpages output

  * a mechanism for preserving relative URIs in documents that make use of
    XInclude

  * support for the "new" .90 version of FOP

  * enhanced capabilities for controlling formatting of lists in HTML and FO
    output

  * autogeneration of AUTHOR and COPYRIGHT sections in manpages output

  * support for generating crop marks in FO/PDF output

  * support for qandaset as a root element in FO output

  * support for floatstyle and orient on all table types

  * support for floatstyle in figure, and example

  * pgwide.properties attribute-set supports extending figure, example and
    table into the left indent area instead of spanning multiple columns.

The following is a detailed list of enhancements and API changes that have
been made since the 1.69.1 release.

Common

The following changes have been made to the common code since the 1.69.1
release.

  * Add the xsl:key for the kimber indexing method.

  * Add support for qandaset.

  * Support dbhtml/dbfo start PI for orderedlist numbering in both HTML and FO

  * Added CVS header.

  * Changed content model of text element to ANY rather than #PCDATA because
    they could contain markup.

  * Added refentry.meta.get.quietly param.

    If zero (the default), notes and warnings about "missing" markup are
    generated during gathering of refentry metadata. If non-zero, the metadata
    is gathered "quietly" -- that is, the notes and warnings are suppressed.

    NOTE: If you are processing a large amount of refentry content, you may be
    able to speed up processing significantly by setting a non-zero value for
    refentry.meta.get.quietly.

  * After namespace stripping, the source document is the temporary tree
    created by the stripping process and it has the wrong base URI for
    relative references. Earlier versions of this code used to try to fix that
    by patching the elements with relative @fileref attributes. That was
    inadequate because it calculated an absolute base URI without considering
    that there might be xml:base attributes already in effect. It seems
    obvious now that the right thing to do is simply to put the xml:base on
    the root of the document. And that seems to work.

  * Added support for "software" and "sectdesc" class values on refmiscinfo;
    "software" is treated identically to "source", and "setdesc" is treated
    identically to "manual".

  * Added support for DocBook 5 namespace-stripping in manpages stylesheet.
    Closes request #1210692.

  * Added <xsl:template match="/"> to make stripns.xsl usable as a standalone
    stylesheet for stripping out DocBook 5/NG to DocBook 4. Note that DocBook
    XSLT drivers that include this stylesheet all override the match="/"
    template.

  * Number figures, examples, and tables from book if there is no prefix (i.e.
    if chapter.autolabel is set to 0). This avoids having the list of figures
    where the figures mysteriously restart their numeration periodically when
    chapter.autolabel is set to 0.

  * Add task template in title.markup mode.

  * Add children (with ids) of formal objects to target data.

  * Added support for case when personname doesn't contain specific name
    markup (as allowed in DocBook 5.0)

Extensions

The following changes have been made to the extensions code since the 1.69.1
release.

  * Support Xalan 2.7

  * Handle the case where the imageFn is actually a URI. This still needs
    work.

FO

The following changes have been made to the fo code since the 1.69.1 release.

  * Adapted to the new indexing code. Now works just like a wrapper that calls
    kosek indexing method, originally implemented here.

  * Added parameters for header/footer table minimum height.

  * Add the index.method parameter.

  * Integrate support for three indexing methods: - the original English-only
    method. - Jirka Kosek's method using EXSLT extensions. - Eliot Kimber's
    method using Saxon extensions. Use the 'index.method' parameter to select.

  * Add support for TOC for qandaset in fo output.

  * Added parameter ulink.hyphenate.chars. Added parameter insert.link
    .page.number.

  * Implemented feature request #942524 to add insert.link.page.number to
    allow link element cross references to have a page number.

  * Add support for ulink.hyphenate.chars so more characters can be break
    points in urls.

  * Implemented patch #1075144 to make the url text in a ulink in FO output an
    active link as well.

  * table footnotes now have their own table.footnote.properties attribute
    set.

  * Add qandaset to root.elements.

  * Added mode="page.sequence" to make it easier to put content into a page
    sequence. First used for qandaset.

  * Implemented feature request #1434408 to support formatting of biblioentry.

  * Added biblioentry.properties.

  * Support PTC/Arbortext bookmarks

  * Added table.footnote.properties to permit table footnotes to format
    differently from regular footnotes.

  * Refactored table templates to unify their processing and support all
    options in all types. Now table and informaltable, in both Cals and Html
    markup, use the same templates where possible, and all support pgwide,
    rotation, and floats. There is also a placeholder table.container template
    to support wrapping a table in a layout table, so the XEP table title
    "continued" extension can be more easily implemented.

  * Added new attribute set toc.line.properties for controlling appearance of
    lines in ToC/LoT

  * Added support for float to example and equation. Added support for pgwide
    to figure, example, and equation (the latter two via a dbfo pgwide="1"
    processing instruction).

  * Add pgwide.properties attribute-set.

  * Added refclass.suppress param.

    If the value of refclass.suppress is non-zero, then display refclass
    contents is suppressed in output. Affects HTML and FO output only.

  * Improved support for task subelements

  * Adjusted spacing around K&R-formatted Funcdef and Paramdef output such
    that it can more easily be discerned where one ends and the other begins.
    Closes #1213264.

  * Made handling of paramdef/parameter in FO output consistent with that in
    HTML and manpages output. Closes #1213259.

  * Made handling of Refnamediv consistent with formatting in HTML and
    manpages output; specifically, changed so that Refname (comma-separated
    list of multiple instances found) is used (instead of Refentrytitle as
    previously), then em-dash, then the Refpurpose. Closes #1212562.

  * Added output of Releaseinfo to recto titlepage ("copyright" page) for Book
    in FO output. This makes it consistent with HTML output. Closes #1327034.
    Thanks to Paul DuBois for reporting.

  * Added condition for setting block-progression-dimension.minimum on
    table-row, instead of height, when fop1.extensions is non-zero. For an
    explanation of the reason for the change, see:
    http://wiki.apache.org/xmlgraphics-fop/Troubleshooting/CommonLogMessages

  * Added new refclass.suppress param for suppressing display of Refclass in
    HTML and FO output. Did not add it to manpages because manpages stylesheet
    is currently just silently ignoring Refclass anyway. Closes request
    #1461065. Thanks to Davor Ocelic (docelic) for reporting.

  * Add support for keep-together PI to informal objects.

  * Add support for fop1.extensions.

  * Add support for fop1 bookmarks.

  * Add fop1.extentions parameter to add support for fop development version.

  * Start supporting fop development version, which will become fop version 1.

  * Add template for task in mode="xref-to".

  * table footnotes now also get footnote.properties attribute-set.

  * Added index.separator named template to compute the separator punctuation
    based on locale.

  * Added support for link, olink, and xref within OO Classsynopsis and
    children. (Because DocBook NG/5 allows it).

  * Support date as an inline

  * Added new parameter keep.relative.image.uris

  * Map Unicode space characters U+2000-U+200A to fo:leaders.

  * Output a real em dash for em-dash dingbat (instead of two hypens).

  * Support default label width parameters for itemized and ordered lists

  * Generate localized title for Refsynopsisdiv if no appropriate Title
    descendant found in source. Closes #1212398. This change makes behavior
    for the Synopsis title consistent with the behavior of HTML and manpages
    output.

    Also, added xsl:use-attribute-sets="normal.para.spacing" to block
    generated for Cmdsynopsis output. Previously, that block had no spacing at
    all specified, which resulted it being crammed up to closely to the
    Synopsis head.

  * Added parameters to support localization of index item punctuation.

  * Added index.number.separator, index.range.separator, and
    index.term.separator parameters to support localization of punctuation in
    index entries.

  * Added "Cross References" section in HTML doc (for consistency with the FO
    doc). Also, moved the existing FO "Cross References" section to follow the
    "Linking" section.

  * Added ID attribues to all Reference elements (e.g., id="tables" for the
    doc for section on Table params). So pages for all subsections of ref docs
    now have stable filenames instead of arbitrary generated filenames.

  * Added two new parameters for handling of multi-term varlistentry elements:

    variablelist.term.break.after: When the variablelist.term.break.after is
    non-zero, it will generate a line break after each term multi-term
    varlistentry.

    variablelist.term.separator: When a varlistentry contains multiple term
    elements, the string specified in the value of the
    variablelist.term.separator parameter is placed after each term except the
    last. The default is ", " (a comma followed by a space). To suppress
    rendering of the separator, set the value of variablelist.term.separator
    to the empty string ("").

    These parameters are primarily intended to be useful if you have
    multi-term varlistentries that have long terms.

    Closes #1306676. Thanks to Sam Steingold for providing an example "lots of
    long terms" doc that demonstrated the value of having these options.

    Also, added normalize-space() call to processing of each term.

    This change affects all output formats (HTML, PDF, manpages). The default
    behavior should pretty much remain the same as before, but it is possible
    (as always) that the change may introduce some new bugginess.

  * Add sidebar titlepage placeholder attset for styles.

  * Add titlepage for sidebar.

  * Implemented RFE #1292615.

    Added bunch of new parameters (attribute sets) that affect list
    presentation: list.block.properties, itemizedlist.properties, orderedlist
    .properties, itemizedlist.label.properties and orderedlist.label
    .properties. Default behaviour of stylesheets has not been changed but
    further customizations will be much more easier.

  * Implemented RFE #1242092.

    You can enable crop marks in your document by setting crop.marks=1 and
    xep.extensions=1. Appearance of crop marks can be controlled by parameters
    crop.mark.bleed (6pt), crop.mark.offset (24pt) and crop.mark.width
    (0.5pt).

    Also there is new named template called user-xep-pis. You can overwrite it
    in order to produce some PIs that can control XEP as described in
    http://www.renderx.com/reference.html#Output_Formats

HTML

The following changes have been made to the html code since the 1.69.1
release.

  * implemented index.method parameter and three methods.

  * added index.method parameter to support 3 indexing methods.

  * Implemented feature request #1072510 as a processing instruction to permit
    including external HTML content into HTML output.

  * Added new parameter chunk.tocs.and.lots.has.title which controls presence
    of title in a separate chunk with ToC/LoT. Disabling title can be very
    useful if you are generating frameset output (well, yes those frames, but
    some customers really want them ;-).

  * Support dbhtml/dbfo start PI for orderedlist numbering in both HTML and FO

  * Allow ToC without title also for set and book.

  * Implemented floats uniformly for figure, example, equation and
    informalfigure, informalexample, and informalequation.

  * Added the autotoc.label.in.hyperlink param.

    If the value of autotoc.label.in.hyperlink is non-zero, labels are
    included in hyperlinked titles in the TOC. If it is instead zero, labels
    are still displayed prior to the hyperlinked titles, but are not
    hyperlinked along with the titles.

    Closes patch #1065868. Thanks to anatoly techtonik for the patch.

  * Added two new params: html.head.legalnotice.link.types and html.head.
    legalnotice.link.multiple.

    If the value of the generate.legalnotice.link is non-zero, then the
    stylesheet generates (in the head section of the HTML source) either a
    single HTML link element or, if the value of the html.head.legalnotice.
    link.multiple is non-zero, one link element for each link type specified.
    Each link has the following attributes:

    - a rel attribute whose value is derived from the value of html.head.
    legalnotice.link.types

    - an href attribute whose value is set to the URL of the file containing
    the legalnotice

    - a title attribute whose value is set to the title of the corresponding
    legalnotice (or a title programatically determined by the stylesheet)

    For example:

    <link rel="copyright" href="ln-id2524073.html" title="Legal Notice">

    Closes #1476450. Thanks to Sam Steingold.

  * Added refclass.suppress param.

    If the value of refclass.suppress is non-zero, then display refclass
    contents is suppressed in output. Affects HTML and FO output only.

  * Improved support for task subelements

  * Added new refclass.suppress param for suppressing display of Refclass in
    HTML and FO output. Did not add it to manpages because manpages stylesheet
    is currently just silently ignoring Refclass anyway. Closes request
    #1461065. Thanks to Davor Ocelic (docelic) for reporting.

  * Process alt text with normalize-space(). Replace tab indents with spaces.

  * Content of citation element is automatically linked to the bibliographic
    entry with the corresponding abbrev.

  * Add template for task in mode="xref-to".

  * Suppress ID warnings if the .warnings parameter is 0

  * Add support for floatstyle to figure.

  * Handling of xref to area/areaset need support in extensions code also. I
    currently have no time to touch extensions code, so code is here to be
    enabled when extension is fixed also.

  * Added 3 parameters for overriding gentext for index punctuation.

  * Added parameters to support localization of index item punctuation. Added
    index.separator named template to compute the separator punctuation based
    on locale.

  * Added a <div class="{$class}-contents"> wrapper around output of contents
    of all formal objects. Also, added an optional <br class="{class}-break"/>
    linebreak after all formal objects.

    WARNING: Because this change places an additional DIV between the DIV
    wrapper for the equation and the equation contents, it may break some
    existing CSS stylesheets that have been created with the assumption that
    there would never be an intervening DIV there.

    The following is an example of what Equation output looks like as a result
    of the changes described above.

    <div class="equation"> <a name="three" id="three"></a>

    <p class="title"><b>(1.3)</b></p>

    <div class="equation-contents"> <span class="mathphrase">1+1=3</span>
    </div> </div><br class="equation-break">

    Rationale: These changes allow CSS control of the placement of the
    formal-object title relative to the formal-object contents. For example,
    using the CSS "float" property enables the title and contents to be
    rendered on the same line. Example stylesheet:

    .equation { margin-top: 20px; margin-bottom: 20px; } .equation-contents {
    float: left; }

    .equation .title { margin-top: 0; float: right; margin-right: 200px; }

    .equation .title b { font-weight: normal; }

    .equation-break { clear: both; }

    Note that the purpose of the ".equation-break" class is to provide a way
    to clear off the floats.

    If you want to instead have the equation title rendered to the left of the
    equation contents, you can do something like this:

    .equation { margin-top: 20px; width: 300px; margin-bottom: 20px; }
    .equation-contents { float: right; }

    .equation .title { margin-top: 0; float: left; margin-right: 200px; }

    .equation .title b { font-weight: normal; }

    .equation-break { clear: both; }

  * Added a chunker.output.quiet top-level parameter so that the chunker can
    be made quiet by default

  * Added support for link, olink, and xref within OO Classsynopsis and
    children. (Because DocBook NG/5 allows it).

  * New parameter: id.warnings. If non-zero, warnings are generated for titled
    objects that don't have titles. True by default; I wonder if this will be
    too aggressive?

  * If the keep.relative.image.uris parameter is true, don't use the absolute
    URI (as calculated from xml:base) in the img src attribute, us the value
    the author specified. Note that we still have to calculate the absolute
    filename for use in the image intrinsics extension.

  * Support date as an inline

  * Added new parameter keep.relative.image.uris

  * Added two new parameters for handling of multi-term varlistentry elements:

    variablelist.term.break.after: When the variablelist.term.break.after is
    non-zero, it will generate a line break after each term multi-term
    varlistentry.

    variablelist.term.separator: When a varlistentry contains multiple term
    elements, the string specified in the value of the
    variablelist.term.separator parameter is placed after each term except the
    last. The default is ", " (a comma followed by a space). To suppress
    rendering of the separator, set the value of variablelist.term.separator
    to the empty string ("").

    These parameters are primarily intended to be useful if you have
    multi-term varlistentries that have long terms.

    Closes #1306676. Thanks to Sam Steingold for providing an example "lots of
    long terms" doc that demonstrated the value of having these options.

    Also, added normalize-space() call to processing of each term.

    This change affects all output formats (HTML, PDF, manpages). The default
    behavior should pretty much remain the same as before, but it is possible
    (as always) that the change may introduce some new bugginess.

  * Added "wrapper-name" param to inline.charseq named template, enabling it
    to output inlines other than just "span". Acronym and Abbrev templates now
    use inline.charseq to output HTML "acronym" and "abbr" elements (instead
    of "span"). Closes #1305468. Thanks to Sam Steingold for suggesting the
    change.

Manpages

The following changes have been made to the manpages code since the 1.69.1
release.

  * Added the following params:

    - man.indent.width (string-valued) - man.indent.refsect (boolean) -
    man.indent.blurbs (boolean) - man.indent.lists (boolean) -
    man.indent.verbatims (boolean)

    Note that in earlier snapshots, man.indent.width was named
    man.indentation.default.value and the boolean params had names like
    man.indentation.*.adjust. Also the man.indent.blurbs param was called
    man.indentation.authors.adjust (or something).

    The behavior now is: If the value of a particular man.indent.* boolean
    param is non-zero, the corresponding contents (refsect*, list items,
    authorblurb/personblurb, vervatims) are displayed with a left margin
    indented by a width equal to the value of man.indent.width.

  * Added man.table.footnotes.divider param.

    In each table that contains footenotes, the string specified by the man.
    table.footnotes.divider parameter is output before the list of footnotes
    for the table.

  * Added the man.output.in.separate.dir, man.output.base.dir, and
    man.output.subdirs.enabled parameters.

    The man.output.base.dir parameter specifies the base directory into which
    man-page files are output. The man.output.subdirs.enabled parameter
    controls whether the files are output in subdirectories within the base
    directory.

    The values of the man.output.base.dir and man.output.subdirs.enabled
    parameters are used only if the value of man.output.in.separate.dir
    parameter is non-zero. If the value of man.output.in.separate.dir is zero,
    man-page files are not output in a separate directory.

  * Added man.font.table.headings and man.font.table.title params, for
    controlling font in table headings and titles.

  * Added man.font.funcsynopsisinfo and man.font.funcprototype params, for
    specifying the roff font (for example, BI, B, I) for funcsynopsisinfo and
    funcprototype output.

  * Added man.segtitle.suppress param.

    If the value of man.segtitle.suppress is non-zero, then display of
    segtitle contents is suppressed in output.

  * Added man.output.manifest.enabled and man.output.manifest.filename params.

    If man.output.manifest.enabled is non-zero, a list of filenames for man
    pages generated by the stylesheet transformation is written to the file
    named by man.output.manifest.filename

  * Added refentry.meta.get.quietly param.

    If zero (the default), notes and warnings about "missing" markup are
    generated during gathering of refentry metadata. If non-zero, the metadata
    is gathered "quietly" -- that is, the notes and warnings are suppressed.

    NOTE: If you are processing a large amount of refentry content, you may be
    able to speed up processing significantly by setting a non-zero value for
    refentry.meta.get.quietly.

  * Changed names of all boolean indentation params to man.indent.* Also
    discarded individual man.indent.*.value params and switched to just using
    a common man.indent.width param (3n by default).

  * Added boolean man.output.in.separate.dir param, to control whether or not
    man files are output in separate directory.

  * Added options for controlling indentation of verbatim output. Controlled
    through the man.indentation.verbatims.adjust and
    man.indentation.verbatims.value params. Closes #1242997

  * Added options for controlling indentation in lists and in *blurb output in
    the AUTHORS section. Controlled through the man.indentation.lists.adjust,
    man.indentation.lists.value, man.indentation.authors.adjust, and
    man.indentation.authors.value parameters. Default is 3 characters (instead
    of the roff default of 8 characters). Closes #1449369.

    Also, removed the indent that was being set on informalexample outuput. I
    will instead add an option for indenting verbatims, which I think is what
    the informalexample indent was intended for originally.

  * Changed line-spacing call before synopfragment to use ".sp -1n" ("n" units
    specified) instead of plain ".sp -1"

  * Added support for writing man files into a specific output directory and
    into appropriate subdirectories within that output directory. Controlled
    through the man.base.dir parameter (similar to the base.dir support in the
    HTML stylesheet) and the man.subdirs.enabled parameter, which
    automatically determines the name of an appropriate subdir (for example,
    man/man7, man/man1, etc.) based on the section number/manvolnum of the
    source Refentry.

    Closes #1255036 and #1170317. Thanks to Denis Bradford for the original
    feature request, and to Costin Stroie for submitting a patch that was very
    helpful in implementing the support.

  * Refined XPath statements and notification messages for refentry metadata
    handling.

  * Added support for copyright and legalnotice. The manpages stylesheets now
    output a COPYRIGHT section, after the AUTHORS section, if a copyright or
    legalnotice is found in the source. The section contains the copyright
    contents followed by the legalnotice contents. Closes #1450209.

  * Drastically reworked all of the XPath expressions used in refentry
    metadata gathering -- completely removed $parentinfo and turned $info into
    a set of nodes that includes the *info contents of the Refentry plus the
    *info contents all all of its ancestor elements. The basic XPath
    expression now used throughout is (using the example of checking for a
    date):

    (($info[//date])[last()]/date)[1].

    That selects the "last" *info/date date in document order -- that is, the
    one eitther on the Refentry itself or on the closest ancestor to the
    Refentry.

    It's likely this change may break some things; may need to pick up some
    pieces later.

    Also, changed the default value for the man.th.extra2.max.length from 40
    to 30.

  * Added support for DocBook 5 namespace-stripping in manpages stylesheet.
    Closes request #1210692.

  * Fixed handling of table footnotes. With this checkin, the table support in
    the manpages stylesheet is now basically feature complete. So this change
    closes request #619532, "No support for tables" -- the oldest currently
    open manpages feature request, submitted by Ben Secrest (blsecres) on
    2002-10-07. Congratulations to me [patting myself on the back].

  * Added handling for table titles. Also fixed handling of nested tables;
    nest tables are now "extracted" and displayed just after their parent
    tables.

  * Added option for turning off bold formatting in Funcsynopsis. Boldface
    formatting in function synopsis is mandated in the man(7) man page and is
    used almost universally in existing man pages. Despite that, it really
    does look like crap to have an entire Funcsynopsis output in bold, so I
    added params for turning off the bold formatting and/or replacing it with
    a different roff special font (e.g., "RI" for alternating roman/italic
    instead of the default "BI" for alternating bold/italic). The new params
    are "man.funcprototype.font" and "man.funcsynopsisinfo.font". To be
    documented later.

    Closes #1452247. Thanks to Joe Orton for the feature request.

  * Use AUTHORS instead of AUTHOR if we have multiple people to attribute.
    Also, fixed checking such that we generate author section even if we don't
    have an author (as long as there is at least one other person/entity we
    can put in the section). Also adjusted assembly of content for Author
    metainfo field such that we now not only use author, but try to find a
    "best match" if we can't find an author name to put there.

    Closes #1233592. Thanks to Sam Steingold for the request.

  * Changes for request #1243027, "Impove handling of AUTHOR section." This
    adds support for Collab, Corpauthor, Corpcredt, Orgname, Publishername,
    and Publisher. Also adds support for output of Affiliation and its
    children, and support for using gentext strings for auto-attributing roles
    (Author, Editor, Publisher, Translator, etc.). Also did a lot of code
    cleanup and modularization of all the AUTHOR handling code. And fixed a
    bug that was causing Author info to not be picked up correctly for
    metainfo comment we embed in man-page source.

  * Support bold output for "emphasis remap='B'". (because Eric Raymond's
    doclifter(1) tool converts groff source marked up with ".B" request or
    "\fB" escapes to DocBook "emphasis remap='B'".)

  * Added support for Segmentedlist. Details: Output is tabular, with no
    option for "list" type output. Output for Segtitle elements can be
    supressed by setting man.segtitle.suppress. If Segtitle content is output,
    it is rendered in italic type (not bold because not all terminals support
    bold and so italic ensures the stand out on those terminals). Extra space
    (.sp line) at end of table code ensures that it gets handled correctly in
    the case where its source is the child of a Para. Closes feature-request
    #1400097. Thanks to Daniel Leidert for the patch and push, and to Alastair
    Rankine for filing the original feature request.

  * Improved handling or Author/Editor/Othercredit.

    Reworked content of (non-visible) comment added at top of each page
    (metadata stuff).

    Added support for generating a manifest file (useful for cleaning up after
    builds, etc.)

  * Added two new parameters for handling of multi-term varlistentry elements:

    variablelist.term.break.after: When the variablelist.term.break.after is
    non-zero, it will generate a line break after each term multi-term
    varlistentry.

    variablelist.term.separator: When a varlistentry contains multiple term
    elements, the string specified in the value of the
    variablelist.term.separator parameter is placed after each term except the
    last. The default is ", " (a comma followed by a space). To suppress
    rendering of the separator, set the value of variablelist.term.separator
    to the empty string ("").

    These parameters are primarily intended to be useful if you have
    multi-term varlistentries that have long terms.

    Closes #1306676. Thanks to Sam Steingold for providing an example "lots of
    long terms" doc that demonstrated the value of having these options.

    Also, added normalize-space() call to processing of each term.

    This change affects all output formats (HTML, PDF, manpages). The default
    behavior should pretty much remain the same as before, but it is possible
    (as always) that the change may introduce some new bugginess.

Params

The following changes have been made to the params code since the 1.69.1
release.

  * New parameters to set header/footer table minimum height.

  * Support multiple indexing methods for different languages.

  * Remove qandaset and qandadiv from generate.toc for fo output because
    formerly it wasn't working, but now it is and the default behavior should
    stay the same.

  * add support for page number references to link element too.

  * Add support for more characters to hyphen on when ulink.hyphenate is
    turned on.

  * New attribute-set to format biblioentry and bibliomixed.

  * Added new parameter chunk.tocs.and.lots.has.title which controls presence
    of title in a separate chunk with ToC/LoT. Disabling title can be very
    useful if you are generating frameset output (well, yes those frames, but
    some customers really want them ;-).

  * Added new attribute set toc.line.properties for controlling appearance of
    lines in ToC/LoT

  * Allow table footnotes to have different properties from regular footnotes.

  * Set properties for pgwide="1" objects.

  * Added the autotoc.label.in.hyperlink param.

    If the value of autotoc.label.in.hyperlink is non-zero, labels are
    included in hyperlinked titles in the TOC. If it is instead zero, labels
    are still displayed prior to the hyperlinked titles, but are not
    hyperlinked along with the titles.

    Closes patch #1065868. Thanks to anatoly techtonik for the patch.

  * Added two new params: html.head.legalnotice.link.types and html.head.
    legalnotice.link.multiple.

    If the value of the generate.legalnotice.link is non-zero, then the
    stylesheet generates (in the head section of the HTML source) either a
    single HTML link element or, if the value of the html.head.legalnotice.
    link.multiple is non-zero, one link element for each link type specified.
    Each link has the following attributes:

    - a rel attribute whose value is derived from the value of html.head.
    legalnotice.link.types

    - an href attribute whose value is set to the URL of the file containing
    the legalnotice

    - a title attribute whose value is set to the title of the corresponding
    legalnotice (or a title programatically determined by the stylesheet)

    For example:

    <link rel="copyright" href="ln-id2524073.html" title="Legal Notice">

    Closes #1476450. Thanks to Sam Steingold.

  * Added the following params:

    - man.indent.width (string-valued) - man.indent.refsect (boolean) -
    man.indent.blurbs (boolean) - man.indent.lists (boolean) -
    man.indent.verbatims (boolean)

    Note that in earlier snapshots, man.indent.width was named
    man.indentation.default.value and the boolean params had names like
    man.indentation.*.adjust. Also the man.indent.blurbs param was called
    man.indentation.authors.adjust (or something).

    The behavior now is: If the value of a particular man.indent.* boolean
    param is non-zero, the corresponding contents (refsect*, list items,
    authorblurb/personblurb, vervatims) are displayed with a left margin
    indented by a width equal to the value of man.indent.width.

  * Added man.table.footnotes.divider param.

    In each table that contains footenotes, the string specified by the man.
    table.footnotes.divider parameter is output before the list of footnotes
    for the table.

  * Added the man.output.in.separate.dir, man.output.base.dir, and
    man.output.subdirs.enabled parameters.

    The man.output.base.dir parameter specifies the base directory into which
    man-page files are output. The man.output.subdirs.enabled parameter
    controls whether the files are output in subdirectories within the base
    directory.

    The values of the man.output.base.dir and man.output.subdirs.enabled
    parameters are used only if the value of man.output.in.separate.dir
    parameter is non-zero. If the value of man.output.in.separate.dir is zero,
    man-page files are not output in a separate directory.

  * Added man.font.table.headings and man.font.table.title params, for
    controlling font in table headings and titles.

  * Added man.font.funcsynopsisinfo and man.font.funcprototype params, for
    specifying the roff font (for example, BI, B, I) for funcsynopsisinfo and
    funcprototype output.

  * Changed to select="0" in refclass.suppress (instead of ..>0</..)

  * Added man.segtitle.suppress param.

    If the value of man.segtitle.suppress is non-zero, then display of
    segtitle contents is suppressed in output.

  * Added man.output.manifest.enabled and man.output.manifest.filename params.

    If man.output.manifest.enabled is non-zero, a list of filenames for man
    pages generated by the stylesheet transformation is written to the file
    named by man.output.manifest.filename

  * Added refclass.suppress param.

    If the value of refclass.suppress is non-zero, then display refclass
    contents is suppressed in output. Affects HTML and FO output only.

  * Added refentry.meta.get.quietly param.

    If zero (the default), notes and warnings about "missing" markup are
    generated during gathering of refentry metadata. If non-zero, the metadata
    is gathered "quietly" -- that is, the notes and warnings are suppressed.

    NOTE: If you are processing a large amount of refentry content, you may be
    able to speed up processing significantly by setting a non-zero value for
    refentry.meta.get.quietly.

  * Added support for "software" and "sectdesc" class values on refmiscinfo;
    "software" is treated identically to "source", and "setdesc" is treated
    identically to "manual".

  * Drastically reworked all of the XPath expressions used in refentry
    metadata gathering -- completely removed $parentinfo and turned $info into
    a set of nodes that includes the *info contents of the Refentry plus the
    *info contents all all of its ancestor elements. The basic XPath
    expression now used throughout is (using the example of checking for a
    date):

    (($info[//date])[last()]/date)[1].

    That selects the "last" *info/date date in document order -- that is, the
    one eitther on the Refentry itself or on the closest ancestor to the
    Refentry.

    It's likely this change may break some things; may need to pick up some
    pieces later.

    Also, changed the default value for the man.th.extra2.max.length from 40
    to 30.

  * Added option for turning off bold formatting in Funcsynopsis. Boldface
    formatting in function synopsis is mandated in the man(7) man page and is
    used almost universally in existing man pages. Despite that, it really
    does look like crap to have an entire Funcsynopsis output in bold, so I
    added params for turning off the bold formatting and/or replacing it with
    a different roff special font (e.g., "RI" for alternating roman/italic
    instead of the default "BI" for alternating bold/italic). The new params
    are "man.funcprototype.font" and "man.funcsynopsisinfo.font". To be
    documented later.

    Closes #1452247. Thanks to Joe Orton for the feature request.

  * fop.extensions now only for FOP version 0.20.5 and earlier.

  * Support for fop1 different from fop 0.20.5 and earlier.

  * Reset default value to empty string so template uses gentext first, then
    the parameter value if not empty.

  * New parameter: id.warnings. If non-zero, warnings are generated for titled
    objects that don't have titles. True by default; I wonder if this will be
    too aggressive?

  * Added new parameter keep.relative.image.uris

  * Support default label width parameters for itemized and ordered lists

  * Added parameters to localize punctuation in indexes.

  * Added two new parameters for handling of multi-term varlistentry elements:

    variablelist.term.break.after: When the variablelist.term.break.after is
    non-zero, it will generate a line break after each term multi-term
    varlistentry.

    variablelist.term.separator: When a varlistentry contains multiple term
    elements, the string specified in the value of the
    variablelist.term.separator parameter is placed after each term except the
    last. The default is ", " (a comma followed by a space). To suppress
    rendering of the separator, set the value of variablelist.term.separator
    to the empty string ("").

    These parameters are primarily intended to be useful if you have
    multi-term varlistentries that have long terms.

    Closes #1306676. Thanks to Sam Steingold for providing an example "lots of
    long terms" doc that demonstrated the value of having these options.

    Also, added normalize-space() call to processing of each term.

    This change affects all output formats (HTML, PDF, manpages). The default
    behavior should pretty much remain the same as before, but it is possible
    (as always) that the change may introduce some new bugginess.

  * Convert 'no' to string in default value.

  * Implemented RFE #1292615.

    Added bunch of new parameters (attribute sets) that affect list
    presentation: list.block.properties, itemizedlist.properties, orderedlist
    .properties, itemizedlist.label.properties and orderedlist.label
    .properties. Default behaviour of stylesheets has not been changed but
    further customizations will be much more easier.

  * Implemented RFE #1242092.

    You can enable crop marks in your document by setting crop.marks=1 and
    xep.extensions=1. Appearance of crop marks can be controlled by parameters
    crop.mark.bleed (6pt), crop.mark.offset (24pt) and crop.mark.width
    (0.5pt).

    Also there is new named template called user-xep-pis. You can overwrite it
    in order to produce some PIs that can control XEP as described in
    http://www.renderx.com/reference.html#Output_Formats

  * Changed short descriptions in doc for *autolabel* params to match new
    autolabel behavior.

Profiling

The following changes have been made to the profiling code since the 1.69.1
release.

  * Profiling now works together with namespace stripping (V5 documents).
    Namespace striping should work with all stylesheets named profile-, even
    if they are not supporting namespace stripping in a non-profiling variant.

  * Moved profiling stage out of templates. This make possible to reuse
    profiled content by several templates and still maintaing node indentity
    (needed for example for HTML Help where content is processed multiple
    times).

    I don't know why this was not on the top level before. Maybe some XSLT
    processors choked on it. I hope this will be OK now.

Tools

The following changes have been made to the tools code since the 1.69.1
release.

  * Moved Makefile.DocBook from contrib module to xsl module.

WordML

The following changes have been made to the wordml code since the 1.69.1
release.

  * added contrib element, better handling of default paragraph style

  * added bridgehead

  * added blocks stylesheet to support bibliographies, glossaries and
    qandasets

  * added mediaobject caption

  * added callouts

  * added Word template file

  * added abstract, fixed itemizedlist, ulink

  * fixed Makefile added many features to Pages support added revhistory,
    inlines, highlights, abstract

  * fixed handling linebreaks when generating WordML added Apple Pages support


To generate a diff of this commit:
cvs rdiff -r1.39 -r1.40 pkgsrc/textproc/docbook-xsl/Makefile
cvs rdiff -r1.16 -r1.17 pkgsrc/textproc/docbook-xsl/PLIST \
    pkgsrc/textproc/docbook-xsl/distinfo
cvs rdiff -r1.1 -r0 pkgsrc/textproc/docbook-xsl/patches/patch-aa \
    pkgsrc/textproc/docbook-xsl/patches/patch-ab \
    pkgsrc/textproc/docbook-xsl/patches/patch-ac

Please note that diffs are not public domain; they are subject to the
copyright notices on the relevant files.