Remove Lynx.

Use tools/html2text.py to convert HTML to text.

Remove some now-obsolete documentation.

Change-Id: Ib21a1ab10c789182da5fcc68e98917a00f2fa650
Reviewed-on: https://code.wireshark.org/review/25733
Petri-Dish: Gerald Combs <gerald@wireshark.org>
Tested-by: Petri Dish Buildbot
Reviewed-by: Gerald Combs <gerald@wireshark.org>

3 files changed, 3 insertions, 137 deletions

diff --git a/docbook/CMakeLists.txt b/docbook/CMakeLists.txt
index 259c4d3231..56d0fa0268 100644
--- a/docbook/CMakeLists.txt
+++ b/docbook/CMakeLists.txt
@@ -12,7 +12,6 @@
 # - Make the build targets top-level on Windows, similar to the NSIS,
 #   WiX, and PortableApps targets?
 
-find_package( LYNX )
 find_package( Asciidoctor 1.5 )
 find_package( XSLTPROC )
 
diff --git a/docbook/Makefile.am b/docbook/Makefile.am
index 3c6855e911..c190d59817 100644
--- a/docbook/Makefile.am
+++ b/docbook/Makefile.am
@@ -458,10 +458,8 @@ wsdg_html_chunked/index.html: developer-guide.xml
 # Release notes
 if HAVE_ASCIIDOCTOR
 release_note_deps = release-notes.html
-if HAVE_LYNX
 release_note_deps += release-notes.txt
 endif
-endif
 release_notes: $(release_note_deps)
 
 # create html single page file
@@ -475,8 +473,7 @@ release-notes.html: $(RELEASE_NOTES_SOURCE)
 # create txt single page file (through HTML)
 release-notes.txt: release-notes.html
 	@ echo --- RELEASE NOTES - TXT ---
-	$(LYNX) \
-		-dump \
+	$(PYTHON) $(srcdir)/tools/html2text.py \
 		$(builddir)/release-notes.html \
 		> $(builddir)/release-notes.txt
 
diff --git a/docbook/README.adoc b/docbook/README.adoc
index aa63ae0124..bfb9a3d0d1 100644
--- a/docbook/README.adoc
+++ b/docbook/README.adoc
@@ -28,7 +28,8 @@ formats is done via the following steps:
 
 Ultimately we'd like to reduce the toolchain requirements to AsciidoctorJ
 alone, but that's not yet possible. Additional tooling is required for
-the HTML,
+the HTML and HTMLHelp targets. See the Developer's Guide for instructions
+on installing required packages for your platform.
 
 == AsciidoctorJ (recommended) or Asciidoctor
 
@@ -56,12 +57,6 @@ formats based on XSL stylesheets. It either ships as part of the
 operating system or is available via an installable package on
 most Linux distributions.
 
-== Lynx
-
-https://invisible-island.net/lynx/[Lynx] is a text based web browser which can
-convert HTML to plain text. We may add w3m, elinks, or other alternatives
-in the future.
-
 == HTML Help Workshop (Windows only)
 
 The HTML Help compiler is part of the
@@ -147,128 +142,3 @@ which isn't currently supported by Asciidoctor or via DocBook XSL.
 The default help application (Qt Assistant) is ugly. We'd probably want
 to write our own help viewer app or integrate help directly via
 QHelpEngine.
-
-= Packages For Windows
-
-Installing the asciidoc package will pull in almost all the other required Cygwin packages.
-You may need to run "build-docbook-catalog" from a Cygwin bash prompt in order to register your catalog properly.
-
-Tool/File           Cygwin Package          Opt./Mand.  Comments
----------           --------------          ----------  --------
-asciidoc            Doc/asciidoc            M           cygwin python is a dependency and will also be installed (if not installed)
-xsltproc:           Libs/libxslt            M
-xsl stylesheets:    Text/docbook-xsl        M           docbook.xsl, chunk.xsl and htmlhelp.xsl
-docbookx.dtd:       Text/docbook-xml42      M           a later version may be required (e.g. Doc/docbook-xml45), depending on your asciidoc installation
-docbookx.dtd:       Text/docbook-xml45      M           current asciidoc installations require this
-lynx:               Web/lynx                M
-dblatex             Text/dblatex            O           A number of dependencies will also be installed
-fop:                -                       O           URL: http://xml.apache.org/fop/ - install it into docbok\fop-1.x or wireshark_lib_dir\fop-1.x
-hhc:                -                       O           URL: http://msdn.microsoft.com/library/default.asp?url=/library/en-us/htmlhelp/html/hwMicrosoftHTMLHelpDownloads.asp
-zip:                Archive/zip             O
-getopt:             Base/util-linux         O           Required to run "build-docbook-catalog"
-
-
-Packages for Suse 9.3
----------------------
-Tool/File           Package                 Opt./Mand.  Comments
----------           -------                 ----------  --------
-xsltproc:           libxslt                 M
-xsl stylesheets:    docbook-xsl-stylesheets M           docbook.xsl and chunk.xsl
-docbookx.dtd:       docbook_4               M
-fop:                fop                     O
-
-
-Packages for Gentoo
--------------------
-Like with all packages do ...
-Check dependencies: emerge -p <package>
-Install it:         emerge <package>
-
-Tool/File           Package                  Opt./Mand.   Comments
----------           -------                  ----------   --------
-xsltproc:           libxslt                  M
-xsl stylesheets:    docbook-xsl-stylesheets  M            docbook.xsl and chunk.xsl
-                                                          Necessary docbook catalogs are built automatically by portage in /etc/xml and /etc/sgml
-                                                            docbook.xsl and chunk.xsl using "/usr/bin/build-docbook-catalog".
-                                                            So docbook runs out of the box on Gentoo.
-docbookx.dtd:       docbook-xml-dtd          M
-fop:                fop                      O            Has a lot of JAVA dependencies.
-Quanta+             quanta or kdewebdev      O            Nice HTML/XML/SGML and Docbook editor with Syntaxhighlighting, Autocompletion, etc.
-
-Tip: The actual DTD version of Gentoo is 4.4, but wireshark docs still use 4.2.
-     To be able to generate the docs, change the version in the second line of
-     developer-guide.xml or install an older version of the DTD.
-     See into the Gentoo handbook howto unmask old versions.
-
-
-Packages for Fedora
--------------------
-Tool/File           Package                 Opt./Mand.  Comments
----------           -------                 ----------  --------
-xsltproc:           libxslt                 M
-xsl stylesheets:    docbook-style-xsl       M           docbook.xsl and chunk.xsl
-docbookx.dtd:       docbook-dtds            M           provides v4.1, v4.2, v4.3, v4.4 DTDs
-asciidoc:           ascidoc                 M
-
-fop:                fop                     O           See above
-
-Note: There are required dependencies (such as xml-common and sgml-common);
-      yum is your friend for doing package installs including required
-      dependencies.
-
-
-Packages for Debian
--------------------
-Tool/File           Package                 Opt./Mand.  Comments
----------           -------                 ----------  --------
-xsltproc:           libxslt                 M
-xsl stylesheets:    docbook-xsl             M
-chunk.xsl:          docbook-xsl             M
-htmlhelp.xsl:       docbook-xsl             M
-docbookx.dtd:       docbook-xml             M
-fop:                fop                     O           See above
-
-
-
-Makefile:
---------------------------
-There are several ways and tools to do these conversion, following is a short
-description of the way the makefile targets are doing things and which output
-files required for a release in that format.
-
-all
-Will generate both guides in all available output formats (see below).
-
-make wsug
-Will generate Wireshark User's Guide in all available output formats.
-
-make wsug_html
-The HTML file is generated using xsltproc and the XSL stylesheets from
-Norman Walsh. This is a conversion into a single HTML page.
-output: wsug_html
-
-make wsug_html_chunked
-The HTML files are generated using xsltproc and the XSL stylesheets from
-Norman Walsh. This is a conversion into chunked (multiple) HTML pages.
-output: wsug_html_chunked
-
-make wsug_pdf_us
-make wsug_pdf_a4
-The PDF is generated using an intermediate format named XSL-FO (XSL
-formatting objects). xsltproc converts the XML to a FO file, and then FOP
-(Apache's formatting object processor) is used to generate the PDF document,
-in US letter or A4 paper format.
-Tip: You will get lot's of INFO/WARNING/ERROR messages when generating PDF,
-but the conversion works just fine.
-output: user-guide-us.pdf user-guide-a4.pdf
-
-make wsug_chm
-On Win32 platforms, the "famous" HTML help format can be generated by using a
-special HTML chunked conversion and then use the htmlhelp compiler from
-Microsoft.
-output: htmlhelp.chm
-
-Using the prefix wsdg_ instead of wsug_ will build the same targets but for the
-Wireshark Developer's Guide.
-
-The makefile is written to be run with make on UNIX/Linux platforms.