a0da281 · typeable/pandoc

git.iko.soy/typeable/pandoc

Man writer: break lines at end of each sentence.

groff expects this and treats . and ? differently when followed by line ending as opposed to ordinary space. Also, don't escape periods. Instead, use zero-width character \& to avoid unwanted interpretation of periods at start of line. Resolves Issue #148. git-svn-id: https://pandoc.googlecode.com/svn/trunk@1580 788f1e2b-df1e-0410-8736-df70ead52e1b

commita0da2816bda209dec2d701f90abcda57944d7ad3 authorfiddlosopher <fiddlosopher@788f1e2b-df1e-0410-8736-df70ead52e1b> dateSat 6 Jun 2009, 21:27 (17 years ago) parent314c332

3 files changed+153−107

compare with master

unifiedsplit

src/Text/Pandoc/Writers/Man.hs+36−3 tests/tables.man+13−11 tests/writer.man+104−93

src/Text/Pandoc/Writers/Man.hs+36−3

@@ -35,6 +35,7 @@ import Text.Printf ( printf )

3535import Data.List ( isPrefixOf, drop, nub, intersperse, intercalate )

3636import Text.PrettyPrint.HughesPJ hiding ( Str )

3737import Control.Monad.State

38+import Control.Monad ( liftM )

3839

3940type Notes = [[Block]]

4041type Preprocessors = [String] -- e.g. "t" for tbl

@@ -100,7 +101,7 @@ noteToMan opts num note = do

100101

101102-- | Association list of characters to escape.

102103manEscapes :: [(Char, String)]

103−manEscapes = [('\160', "\\ "), ('\'', "\\[aq]")] ++ backslashEscapes ".@\\"

104+manEscapes = [('\160', "\\ "), ('\'', "\\[aq]")] ++ backslashEscapes "@\\"

104105

105106-- | Escape special characters for Man.

106107escapeString :: String -> String

@@ -110,15 +111,43 @@ escapeString = escapeStringUsing manEscapes

110111escapeCode :: String -> String

111112escapeCode = escapeStringUsing (manEscapes ++ backslashEscapes "\t ")

112113

114+-- We split inline lists into sentences, and print one sentence per

115+-- line. groff/troff treats the line-ending period differently.

116+-- See http://code.google.com/p/pandoc/issues/detail?id=148.

117+

118+-- | Returns the first sentence in a list of inlines, and the rest.

119+breakSentence :: [Inline] -> ([Inline], [Inline])

120+breakSentence [] = ([],[])

121+breakSentence xs =

122+ let isSentenceEndInline (Str ".") = True

123+ isSentenceEndInline (Str "?") = True

124+ isSentenceEndInline _ = False

125+ (as, bs) = break isSentenceEndInline xs

126+ in case bs of

127+ [] -> (as, [])

128+ [c] -> (as ++ [c], [])

129+ (c:Space:cs) -> (as ++ [c], cs)

130+ (Str ".":Str ")":cs) -> (as ++ [Str ".", Str ")"], cs)

131+ (c:cs) -> (as ++ [c] ++ ds, es)

132+ where (ds, es) = breakSentence cs

133+

134+-- | Split a list of inlines into sentences.

135+splitSentences :: [Inline] -> [[Inline]]

136+splitSentences xs =

137+ let (sent, rest) = breakSentence xs

138+ in if null rest then [sent] else sent : splitSentences rest

139+

113140-- | Convert Pandoc block element to man.

114141blockToMan :: WriterOptions -- ^ Options

115142 -> Block -- ^ Block element

116143 -> State WriterState Doc

117144blockToMan _ Null = return empty

118145blockToMan opts (Plain inlines) =

119− wrapIfNeeded opts (inlineListToMan opts) inlines

146+ liftM vcat $ mapM (wrapIfNeeded opts (inlineListToMan opts)) $

147+ splitSentences inlines

120148blockToMan opts (Para inlines) = do

121− contents <- wrapIfNeeded opts (inlineListToMan opts) inlines

149+ contents <- liftM vcat $ mapM (wrapIfNeeded opts (inlineListToMan opts)) $

150+ splitSentences inlines

122151 return $ text ".PP" $$ contents

123152blockToMan _ (RawHtml str) = return $ text str

124153blockToMan _ HorizontalRule = return $ text $ ".PP\n * * * * *"

@@ -237,6 +266,10 @@ blockListToMan opts blocks =

237266

238267-- | Convert list of Pandoc inline elements to man.

239268inlineListToMan :: WriterOptions -> [Inline] -> State WriterState Doc

269+-- if list starts with ., insert a zero-width character \& so it

270+-- won't be interpreted as markup if it falls at the beginning of a line.

271+inlineListToMan opts lst@(Str "." : _) = mapM (inlineToMan opts) lst >>=

272+ (return . (text "\\&" <>) . hcat)

240273inlineListToMan opts lst = mapM (inlineToMan opts) lst >>= (return . hcat)

241274

242275-- | Convert Pandoc inline element to man.

tests/tables.man+13−11

@@ -1,7 +1,7 @@

11.PP

22Simple table with caption:

33.PP

4−Demonstration of simple table syntax\.

4+Demonstration of simple table syntax.

55.TS

66tab(@);

77rw(10.50n) lw(6.13n) cw(11.38n) lw(8.75n).

@@ -90,7 +90,7 @@ T}

9090.PP

9191Simple table indented two spaces:

9292.PP

93−Demonstration of simple table syntax\.

93+Demonstration of simple table syntax.

9494.TS

9595tab(@);

9696rw(10.50n) lw(6.13n) cw(11.38n) lw(8.75n).

@@ -135,7 +135,7 @@ T}

135135.PP

136136Multiline table with caption:

137137.PP

138−Here\[aq]s the caption\. It may span multiple lines\.

138+Here\[aq]s the caption. It may span multiple lines.

139139.TS

140140tab(@);

141141cw(10.50n) lw(9.63n) rw(11.38n) lw(23.63n).

@@ -154,18 +154,19 @@ First

154154T}@T{

155155row

156156T}@T{

157−12\.0

157+12.0

158158T}@T{

159−Example of a row that spans multiple lines\.

159+Example of a row that spans multiple lines.

160160T}

161161T{

162162Second

163163T}@T{

164164row

165165T}@T{

166−5\.0

166+5.0

167167T}@T{

168−Here\[aq]s another one\. Note the blank line between rows\.

168+Here\[aq]s another one.

169+Note the blank line between rows.

169170T}

170171.TE

171172.PP

@@ -189,17 +190,18 @@ First

189190T}@T{

190191row

191192T}@T{

192−12\.0

193+12.0

193194T}@T{

194−Example of a row that spans multiple lines\.

195+Example of a row that spans multiple lines.

195196T}

196197T{

197198Second

198199T}@T{

199200row

200201T}@T{

201−5\.0

202+5.0

202203T}@T{

203−Here\[aq]s another one\. Note the blank line between rows\.

204+Here\[aq]s another one.

205+Note the blank line between rows.

204206T}

205207.TE

tests/writer.man+104−93

@@ -1,7 +1,7 @@

11.TH Pandoc "" "July 17, 2006" "Test Suite"

22.PP

3−This is a set of tests for pandoc\. Most of them are adapted from

4−John Gruber's markdown test suite\.

3+This is a set of tests for pandoc.

4+Most of them are adapted from John Gruber's markdown test suite.

55.PP

66 * * * * *

77.SH Headers

@@ -21,19 +21,22 @@ with no blank line

2121 * * * * *

2222.SH Paragraphs

2323.PP

24−Here's a regular paragraph\.

24+Here's a regular paragraph.

2525.PP

26−In Markdown 1\.0\.0 and earlier\. Version 8\. This line turns into

27−a list item\. Because a hard-wrapped line in the middle of a

28−paragraph looked like a list item\.

26+In Markdown 1.0.0 and earlier.

27+Version 8.

28+This line turns into a list item.

29+Because a hard-wrapped line in the middle of a paragraph looked

30+like a list item.

2931.PP

30−Here's one with a bullet\. * criminey\.

32+Here's one with a bullet.

33+* criminey.

3134.PP

3235There should be a hard line break

3336.PD 0

3437.P

3538.PD

36−here\.

39+here.

3740.PP

3841 * * * * *

3942.SH Block Quotes

@@ -41,7 +44,8 @@ here\.

4144E-mail style:

4245.RS

4346.PP

44−This is a block quote\. It is pretty short\.

47+This is a block quote.

48+It is pretty short.

4549.RE

4650.RS

4751.PP

@@ -70,9 +74,9 @@ nested

7074.RE

7175.RE

7276.PP

73−This should not be a block quote: 2 > 1\.

77+This should not be a block quote: 2 > 1.

7478.PP

75−And a following paragraph\.

79+And a following paragraph.

7680.PP

7781 * * * * *

7882.SH Code Blocks

@@ -184,16 +188,17 @@ Three

184188.PP

185189Multiple paragraphs:

186190.IP "1." 3

187−Item 1, graf one\.

191+Item 1, graf one.

188192.RS 4

189193.PP

190−Item 1\. graf two\. The quick brown fox jumped over the lazy dog's

191−back\.

194+Item 1.

195+graf two.

196+The quick brown fox jumped over the lazy dog's back.

192197.RE

193198.IP "2." 3

194−Item 2\.

199+Item 2.

195200.IP "3." 3

196−Item 3\.

201+Item 3.

197202.SS Nested

198203.IP \[bu] 2

199204Tab

@@ -273,7 +278,7 @@ Nesting:

273278Upper Alpha

274279.RS 4

275280.IP "I." 3

276−Upper Roman\.

281+Upper Roman.

277282.RS 4

278283.IP "(6)" 4

279284Decimal start with 6

@@ -286,19 +291,20 @@ Lower alpha with paren

286291.PP

287292Autonumbering:

288293.IP "1." 3

289−Autonumber\.

294+Autonumber.

290295.IP "2." 3

291−More\.

296+More.

292297.RS 4

293298.IP "1." 3

294−Nested\.

299+Nested.

295300.RE

296301.PP

297302Should not be a list item:

298303.PP

299−M\.A\.\ 2007

304+M.A.\ 2007

300305.PP

301−B\. Williams

306+B.

307+Williams

302308.PP

303309 * * * * *

304310.SH Definition Lists

@@ -497,51 +503,53 @@ Hr's:

497503 * * * * *

498504.SH Inline Markup

499505.PP

500−This is \f[I]emphasized\f[], and so \f[I]is this\f[]\.

506+This is \f[I]emphasized\f[], and so \f[I]is this\f[].

501507.PP

502−This is \f[B]strong\f[], and so \f[B]is this\f[]\.

508+This is \f[B]strong\f[], and so \f[B]is this\f[].

503509.PP

504−An \f[I]emphasized link (/url)\f[]\.

510+An \f[I]emphasized link (/url)\f[].

505511.PP

506−\f[B]\f[I]This is strong and em\.\f[]\f[]

512+\f[B]\f[I]This is strong and em.\f[]\f[]

507513.PP

508−So is \f[B]\f[I]this\f[]\f[] word\.

514+So is \f[B]\f[I]this\f[]\f[] word.

509515.PP

510−\f[B]\f[I]This is strong and em\.\f[]\f[]

516+\f[B]\f[I]This is strong and em.\f[]\f[]

511517.PP

512−So is \f[B]\f[I]this\f[]\f[] word\.

518+So is \f[B]\f[I]this\f[]\f[] word.

513519.PP

514520This is code: \f[B]>\f[], \f[B]$\f[], \f[B]\\\f[], \f[B]\\$\f[],

515−\f[B]<html>\f[]\.

521+\f[B]<html>\f[].

516522.PP

517−[STRIKEOUT:This is \f[I]strikeout\f[]\.]

523+[STRIKEOUT:This is \f[I]strikeout\f[].]

518524.PP

519−Superscripts: a^bc^d a^\f[I]hello\f[]^ a^hello\ there^\.

525+Superscripts: a^bc^d a^\f[I]hello\f[]^ a^hello\ there^.

520526.PP

521−Subscripts: H~2~O, H~23~O, H~many\ of\ them~O\.

527+Subscripts: H~2~O, H~23~O, H~many\ of\ them~O.

522528.PP

523529These should not be superscripts or subscripts, because of the

524−unescaped spaces: a^b c^d, a~b c~d\.

530+unescaped spaces: a^b c^d, a~b c~d.

525531.PP

526532 * * * * *

527533.SH Smart quotes, ellipses, dashes

528534.PP

529−\[lq]Hello,\[rq] said the spider\. \[lq]`Shelob' is my name\.\[rq]

535+\[lq]Hello,\[rq] said the spider.

536+\[lq]`Shelob' is my name.\[rq]

530537.PP

531−`A', `B', and `C' are letters\.

538+`A', `B', and `C' are letters.

532539.PP

533−`Oak,' `elm,' and `beech' are names of trees\. So is `pine\.'

540+`Oak,' `elm,' and `beech' are names of trees.

541+So is `pine.'

534542.PP

535−`He said, \[lq]I want to go\.\[rq]' Were you alive in the 70's?

543+`He said, \[lq]I want to go.\[rq]' Were you alive in the 70's?

536544.PP

537545Here is some quoted `\f[B]code\f[]' and a

538−\[lq]quoted link (http://example.com/?foo=1&bar=2)\[rq]\.

546+\[lq]quoted link (http://example.com/?foo=1&bar=2)\[rq].

539547.PP

540−Some dashes: one\[em]two \[em] three\[em]four \[em] five\.

548+Some dashes: one\[em]two \[em] three\[em]four \[em] five.

541549.PP

542−Dashes between numbers: 5\[en]7, 255\[en]66, 1987\[en]1999\.

550+Dashes between numbers: 5\[en]7, 255\[en]66, 1987\[en]1999.

543551.PP

544−Ellipses\&...and\&...and\&...\.

552+Ellipses\&...and\&...and\&....

545553.PP

546554 * * * * *

547555.SH LaTeX

@@ -563,18 +571,19 @@ Here's some display math:

563571.RE

564572.IP \[bu] 2

565573Here's one that has a line break in it:

566−\f[B]\\alpha\ +\ \\omega\ \\times\ x^2\f[]\.

574+\f[B]\\alpha\ +\ \\omega\ \\times\ x^2\f[].

567575.PP

568576These shouldn't be math:

569577.IP \[bu] 2

570−To get the famous equation, write \f[B]$e\ =\ mc^2$\f[]\.

578+To get the famous equation, write \f[B]$e\ =\ mc^2$\f[].

571579.IP \[bu] 2

572−$22,000 is a \f[I]lot\f[] of money\. So is $34,000\. (It worked if

573−\[lq]lot\[rq] is emphasized\.)

580+$22,000 is a \f[I]lot\f[] of money.

581+So is $34,000.

582+(It worked if \[lq]lot\[rq] is emphasized.)

574583.IP \[bu] 2

575−Shoes ($20) and socks ($5)\.

584+Shoes ($20) and socks ($5).

576585.IP \[bu] 2

577−Escaped \f[B]$\f[]: $73 \f[I]this should be emphasized\f[] 23$\.

586+Escaped \f[B]$\f[]: $73 \f[I]this should be emphasized\f[] 23$.

578587.PP

579588Here's a LaTeX table:

580589.PP

@@ -594,15 +603,15 @@ set membership: ∈

594603.IP \[bu] 2

595604copyright: ©

596605.PP

597−AT&T has an ampersand in their name\.

606+AT&T has an ampersand in their name.

598607.PP

599−AT&T is another way to write it\.

608+AT&T is another way to write it.

600609.PP

601−This & that\.

610+This & that.

602611.PP

603−4 < 5\.

612+4 < 5.

604613.PP

605−6 > 5\.

614+6 > 5.

606615.PP

607616Backslash: \\

608617.PP

@@ -628,7 +637,7 @@ Greater-than: >

628637.PP

629638Hash: #

630639.PP

631−Period: \.

640+Period: \&.

632641.PP

633642Bang: !

634643.PP

@@ -640,13 +649,13 @@ Minus: -

640649.SH Links

641650.SS Explicit

642651.PP

643−Just a URL (/url/)\.

652+Just a URL (/url/).

644653.PP

645−URL and title (/url/)\.

654+URL and title (/url/).

646655.PP

647−URL and title (/url/)\.

656+URL and title (/url/).

648657.PP

649−URL and title (/url/)\.

658+URL and title (/url/).

650659.PP

651660URL and title (/url/)

652661.PP

@@ -656,45 +665,45 @@ with_underscore (/url/with_underscore)

656665.PP

657666Email link (mailto:nobody@nowhere.net)

658667.PP

659−Empty ()\.

668+Empty ().

660669.SS Reference

661670.PP

662−Foo bar (/url/)\.

671+Foo bar (/url/).

663672.PP

664−Foo bar (/url/)\.

673+Foo bar (/url/).

665674.PP

666−Foo bar (/url/)\.

675+Foo bar (/url/).

667676.PP

668−With embedded [brackets] (/url/)\.

677+With embedded [brackets] (/url/).

669678.PP

670−b (/url/) by itself should be a link\.

679+b (/url/) by itself should be a link.

671680.PP

672−Indented once (/url)\.

681+Indented once (/url).

673682.PP

674−Indented twice (/url)\.

683+Indented twice (/url).

675684.PP

676−Indented thrice (/url)\.

685+Indented thrice (/url).

677686.PP

678−This should [not][] be a link\.

687+This should [not][] be a link.

679688.PP

680689\f[CR]

681690 [not]:\ /url

682691\f[]

683692.PP

684−Foo bar (/url/)\.

693+Foo bar (/url/).

685694.PP

686−Foo biz (/url/)\.

695+Foo biz (/url/).

687696.SS With ampersands

688697.PP

689698Here's a

690−link with an ampersand in the URL (http://example.com/?foo=1&bar=2)\.

699+link with an ampersand in the URL (http://example.com/?foo=1&bar=2).

691700.PP

692701Here's a link with an amersand in the link text:

693−AT&T (http://att.com/)\.

702+AT&T (http://att.com/).

694703.PP

695−Here's an inline link (/script?foo=1&bar=2)\.

704+Here's an inline link (/script?foo=1&bar=2).

696705.PP

697−Here's an inline link in pointy braces (/script?foo=1&bar=2)\.

706+Here's an inline link in pointy braces (/script?foo=1&bar=2).

698707.SS Autolinks

699708.PP

700709With an ampersand: <http://example.com/?foo=1&bar=2>

@@ -703,7 +712,7 @@ In a list?

703712.IP \[bu] 2

704713<http://example.com/>

705714.IP \[bu] 2

706−It should\.

715+It should.

707716.PP

708717An e-mail address: <nobody@nowhere.net>

709718.RS

@@ -711,10 +720,10 @@ An e-mail address: <nobody@nowhere.net>

711720Blockquoted: <http://example.com/>

712721.RE

713722.PP

714−Auto-links should not occur here: \f[B]<http://example\.com/>\f[]

723+Auto-links should not occur here: \f[B]<http://example.com/>\f[]

715724.PP

716725\f[CR]

717− or\ here:\ <http://example\.com/>

726+ or\ here:\ <http://example.com/>

718727\f[]

719728.PP

720729 * * * * *

@@ -724,56 +733,58 @@ From \[lq]Voyage dans la Lune\[rq] by Georges Melies (1902):

724733.PP

725734[IMAGE: lalune (lalune.jpg)]

726735.PP

727−Here is a movie [IMAGE: movie (movie.jpg)] icon\.

736+Here is a movie [IMAGE: movie (movie.jpg)] icon.

728737.PP

729738 * * * * *

730739.SH Footnotes

731740.PP

732−Here is a footnote reference,[1] and another\.[2] This should

741+Here is a footnote reference,[1] and another.[2] This should

733742\f[I]not\f[] be a footnote reference, because it contains a

734−space\.[^my note] Here is an inline note\.[3]

743+space.[^my note] Here is an inline note.[3]

735744.RS

736745.PP

737−Notes can go in quotes\.[4]

746+Notes can go in quotes.[4]

738747.RE

739748.IP "1." 3

740−And in list items\.[5]

749+And in list items.[5]

741750.PP

742751This paragraph should not be part of the note, as it is not

743−indented\.

752+indented.

744753.SH NOTES

745754

746755.SS [1]

747756.PP

748−Here is the footnote\. It can go anywhere after the footnote

749−reference\. It need not be placed at the end of the document\.

757+Here is the footnote.

758+It can go anywhere after the footnote reference.

759+It need not be placed at the end of the document.

750760

751761.SS [2]

752762.PP

753−Here's the long note\. This one contains multiple blocks\.

763+Here's the long note.

764+This one contains multiple blocks.

754765.PP

755766Subsequent blocks are indented to show that they belong to the

756−footnote (as with list items)\.

767+footnote (as with list items).

757768.PP

758769\f[CR]

759770 \ \ {\ <code>\ }

760771\f[]

761772.PP

762773If you want, you can indent every line, but you can also be lazy

763−and just indent the first line of each block\.

774+and just indent the first line of each block.

764775

765776.SS [3]

766777.PP

767−This is \f[I]easier\f[] to type\. Inline notes may contain

768−links (http://google.com) and \f[B]]\f[] verbatim characters, as

769−well as [bracketed text]\.

778+This is \f[I]easier\f[] to type.

779+Inline notes may contain links (http://google.com) and \f[B]]\f[]

780+verbatim characters, as well as [bracketed text].

770781

771782.SS [4]

772783.PP

773−In quote\.

784+In quote.

774785

775786.SS [5]

776787.PP

777−In list\.

788+In list.

778789.SH AUTHORS

779790John MacFarlane, Anonymous