Markdown filter

.circleci/config.yml

@@ -94,10 +94,29 @@ jobs:
     - run: CXX=clang++-8 CC=clang-8 ./sanity-check.sh
     - run: ./config-opt && make -C build -j2
 
+  tests:
+    working_directory: ~/GNUAspell/aspell
+    parallelism: 1
+    docker:
+    - image: circleci/buildpack-deps:19.04
+    steps:
+    # Machine Setup
+    - checkout
+    #
+    - run: sudo apt-get -y update
+    - run: sudo apt-get -y install make autopoint texinfo libtool intltool bzip2 gettext libipc-system-simple-perl
+    - run: sudo apt-get -y purge aspell
+    #
+    - run: ./autogen
+    - run: make -j2 -C test
+
 workflows:
   version: 2
   sanity_check:
     jobs:
       - build_jessie
       - build_18_04
       - build_19_04
+  tests:
+    jobs:
+      - tests

Makefile.am

@@ -166,6 +166,7 @@ optfiles = \
   modules/filter/email-filter.info\
   modules/filter/tex-filter.info\
   modules/filter/sgml-filter.info\
+  modules/filter/markdown-filter.info\
   modules/filter/html-filter.info\
   modules/filter/context-filter.info\
   modules/filter/nroff-filter.info\
@@ -175,6 +176,7 @@ optfiles = \
 fltfiles = \
   modules/filter/modes/html.amf \
   modules/filter/modes/sgml.amf \
+  modules/filter/modes/markdown.amf \
   modules/filter/modes/tex.amf \
   modules/filter/modes/email.amf \
   modules/filter/modes/ccpp.amf \
@@ -196,6 +198,7 @@ libaspell_la_SOURCES +=\
   modules/filter/email.cpp\
   modules/filter/tex.cpp\
   modules/filter/sgml.cpp\
+  modules/filter/markdown.cpp\
   modules/filter/context.cpp\
   modules/filter/nroff.cpp\
   modules/filter/texinfo.cpp
@@ -208,7 +211,7 @@ filter_ldflags = -module -avoid-version
 ### Add name of filter library containing your filter. Name always
 ### must look like lib<filtername>-filter.la see development manual
 filter_LTLIBRARIES = email-filter.la tex-filter.la\
-	             sgml-filter.la context-filter.la\
+	             sgml-filter.la markdown-filter.la context-filter.la\
                      nroff-filter.la texinfo-filter.la
 
 email_filter_la_SOURCES = modules/filter/email.cpp
@@ -223,6 +226,10 @@ sgml_filter_la_SOURCES  = modules/filter/sgml.cpp
 sgml_filter_la_LIBADD = libaspell.la
 sgml_filter_la_LDFLAGS  = ${filter_ldflags}
 
+markdown_filter_la_SOURCES  = modules/filter/markdown.cpp
+markdown_filter_la_LIBADD = libaspell.la
+markdown_filter_la_LDFLAGS  = ${filter_ldflags}
+
 context_filter_la_SOURCES  = modules/filter/context.cpp
 context_filter_la_LIBADD = libaspell.la
 context_filter_la_LDFLAGS  = ${filter_ldflags}

auto/mk-src.in

@@ -842,8 +842,11 @@ class: document checker
 	method: process
 
 		desc => Process a string.
-			The string passed in should only be split on
-			white space characters.  Furthermore, between
+			The document is expected to be passed in one or more
+			lines at a time.  Splitting the document on
+			white space characters instead of new lines is
+			permissible but some filters which are line based
+			may give incorrect results.  Furthermore, between
 			calls to reset, each string should be passed
 			in exactly once and in the order they appeared
 			in the document.  Passing in strings out of

config-debug

@@ -4,4 +4,4 @@ mkdir -p build
 cd build
 ../configure --enable-maintainer-mode \
   --disable-shared --disable-pspell-compatibility\
-  --prefix="`pwd`/../inst" CFLAGS='-g -Wall' CXXFLAGS='-g -Wall' "$@"
+  --prefix="`pwd`/../inst" CFLAGS='-g -Wall -Wno-unused' CXXFLAGS='-g -Wall -Wno-unused' "$@"

manual/aspell.texi

@@ -1355,6 +1355,55 @@ The @option{html} filter is like the SGML Filter Mode but specialized for
 HTML.  By default, 'script' and 'style' are members of the skip list in
 HTML mode.
 
+@subsubsection Markdown Filter
+
+The @option{markdown} filter/mode skips over code and links in Markdown
+documents.  The mode also handles HTML via the @option{html} filter
+which runs after the Markdown filer.
+
+Properly parsing Markdown requires reading the entire document in at
+once, which doesn't fit well into the existing filter framework so the
+filter parses the document a line at a time and makes a best guess
+when a line ends without enough information to correctly parse a a
+tag.  For example in the following:
+@example
+`This looks like inline code
+
+But it's not, becuase the of the blank line.
+@end example
+the first line will be incorrectly skipped.
+
+It supports the following two options that may be useful:
+
+@table @b
+
+@item markdown-multiline-tags
+Enables or disables support for HTML tags that might span multiple
+lines (outside of HTML blocks).  Enabled by default (to disable from
+the command line use @samp{--dont-markdown-multiline-tags}).
+
+@item markdown-skip-ref-labels
+Boolean option the controls if ref lables are skipped.  For when this
+option is enabled @samp{[ref]} will be skipped over in:
+@example
+[link][ref]
+
+[ref]: #c2 "Chapter 2"
+@end example
+Enabled by default (to disable from
+the command line use @samp{--dont-markdown-skip-ref-labels}).
+
+@end table
+
+The Markdown filter also supports the @option{markdown-raw-start-tags}
+and @option{markdown-block-start-tags} list options.  These options
+are a list of HTML tags that start an HTML block.  If the block starts
+with a tag listed in @option{markdown-raw-start-tags} then the block
+ends with the corresponding closing tag.  If the block starts with a
+tag listed in @option{markdown-block-start-tags} then the block ends
+with a blank line.  The defaults for these two options are taken from
+the CommonMark spec (version 0.29, 2019-04-06).
+
 @subsubsection @TeX{}/LaTeX Filter
 
 The @option{tex} (all lowercase) filter mode skips over @TeX{}
@@ -4272,7 +4321,17 @@ a C++ program.
 @heading Changes from 0.60.7 to 0.68.8 (???)
 @itemize @bullet
 @item
+Add Markdown filter.
+@item
+Implement @code{aspell filter} command.
+@item
 Remove unused @option{sug-edit-dist} option.
+@item
+In order to work with the new Markdown filter
+@code{AspellDocumentChecker} now expects the document a line at a
+time.  If the document is split on white space characters instead,
+nothing will break, but new filters such as the Markdown filter may
+give incorrect results.
 @end itemize
 
 @heading Changes from 0.60.6.1 to 0.60.7 (July 29, 2019)

modules/filter/markdown-filter.info

@@ -0,0 +1,97 @@
+# SGML filter option file
+
+LIB-FILE markdown-filter
+
+#This Filter is usable with the following version(s) of Aspell
+ASPELL >=0.60
+
+DESCRIPTION filter for Markdown/CommonMark documents
+
+STATIC filter
+
+OPTION skip-ref-labels
+TYPE bool
+DESCRIPTION skip link labels in link reference definitions
+DEFAULT true
+ENDOPTION
+
+OPTION multiline-tags
+TYPE bool
+DESCRIPTION support tags that span multiple lines outside of HTML blocks
+DEFAULT true
+ENDOPTION
+
+OPTION raw-start-tags
+TYPE list
+DESCRIPTION html tags that start a HTML block that allows blank lines
+DEFAULT script
+DEFAULT pre
+DEFAULT style
+ENDOPTION
+
+OPTION block-start-tags
+TYPE list
+DESCRIPTION html tags that start a HTML block that end with a blank line
+DEFAULT address
+DEFAULT article
+DEFAULT aside
+DEFAULT base
+DEFAULT basefont
+DEFAULT blockquote
+DEFAULT body
+DEFAULT caption
+DEFAULT center
+DEFAULT col
+DEFAULT colgroup
+DEFAULT dd
+DEFAULT details
+DEFAULT dialog
+DEFAULT dir
+DEFAULT div
+DEFAULT dl
+DEFAULT dt
+DEFAULT fieldset
+DEFAULT figcaption
+DEFAULT figure
+DEFAULT footer
+DEFAULT form
+DEFAULT frame
+DEFAULT frameset
+DEFAULT h1
+DEFAULT h2
+DEFAULT h3
+DEFAULT h4
+DEFAULT h5
+DEFAULT h6
+DEFAULT head
+DEFAULT header
+DEFAULT hr
+DEFAULT html
+DEFAULT iframe
+DEFAULT legend
+DEFAULT li
+DEFAULT link
+DEFAULT main
+DEFAULT menu
+DEFAULT menuitem
+DEFAULT nav
+DEFAULT noframes
+DEFAULT ol
+DEFAULT optgroup
+DEFAULT option
+DEFAULT p
+DEFAULT param
+DEFAULT section
+DEFAULT source
+DEFAULT summary
+DEFAULT table
+DEFAULT tbody
+DEFAULT td
+DEFAULT tfoot
+DEFAULT th
+DEFAULT thead
+DEFAULT title
+DEFAULT tr
+DEFAULT track
+DEFAULT ul
+ENDOPTION