LaTeX reader: citation handling changes.

Previously, a LaTeX citation would always be parsed as a Citation
element, with the raw LaTeX in the [Inline] part.

Now, the LaTeX citation is parsed as a Citation element only if
`--biblio` was specified (i.e. only if there is a nonempty set
of references in readerReferences).  Otherwise it is parsed as
raw LaTeX.

This will make it possible to simplify some things in the markdown
writer.  It also makes the LaTeX reader behave more like the Markdown
reader.

pandoc.cabal

@@ -109,6 +109,7 @@ Extra-Source-Files:
                  tests/lalune.jpg,
                  tests/movie.jpg,
                  tests/biblio.bib,
+                 tests/biblio.mods,
                  tests/chicago-author-date.csl,
                  tests/ieee.csl,
                  tests/mhra.csl,
@@ -383,6 +384,7 @@ Test-Suite test-pandoc
                   QuickCheck >= 2.4 && < 2.6,
                   HUnit >= 1.2 && < 1.3,
                   containers >= 0.1 && < 0.6,
+                   citeproc-hs >= 0.3 && < 0.4,
                   ansi-terminal >= 0.5 && < 0.7
   Other-Modules:  Tests.Old
                   Tests.Helpers

src/Text/Pandoc/Readers/LaTeX.hs

@@ -1021,7 +1021,10 @@ cites mode multi = try $ do
 citation :: String -> CitationMode -> Bool -> LP Inlines
 citation name mode multi = do
   (c,raw) <- withRaw $ cites mode multi
-  return $ cite c (rawInline "latex" $ "\\" ++ name ++ raw)
+  refs <- getOption readerReferences
+  return $ if null refs
+              then rawInline "latex" $ "\\" ++ name ++ raw
+              else cite c mempty
 
 complexNatbibCitation :: CitationMode -> LP Inlines
 complexNatbibCitation mode = try $ do
@@ -1040,8 +1043,10 @@ complexNatbibCitation mode = try $ do
                    optional $ char ';'
                    return $ addPrefix pref $ addSuffix suff $ cits'
   (c:cits, raw) <- withRaw $ grouped parseOne
-  return $ cite (c{ citationMode = mode }:cits)
-           (rawInline "latex" $ "\\citetext" ++ raw)
+  refs <- getOption readerReferences
+  return $ if null refs
+              then rawInline "latex" $ "\\citetext" ++ raw
+              else cite (c{ citationMode = mode}:cits) mempty
 
 -- tables
 

src/Text/Pandoc/Writers/Markdown.hs

@@ -644,7 +644,8 @@ inlineToMarkdown opts (LineBreak)
   | otherwise                              = return $ "  " <> cr
 inlineToMarkdown _ Space = return space
 inlineToMarkdown opts (Cite (c:cs) lst)
-  | not (isEnabled Ext_citations opts) = inlineListToMarkdown opts lst
+  | not (null lst) = inlineListToMarkdown opts lst
+  -- if lst is null, citeproc wasn't run; print a pandoc markdown citation
   | citationMode c == AuthorInText = do
     suffs <- inlineListToMarkdown opts $ citationSuffix c
     rest <- mapM convertOne cs

tests/Tests/Old.hs

@@ -186,7 +186,7 @@ markdownCitationTests
      ++ [test "natbib" wopts "markdown-citations.txt"
          "markdown-citations.txt"]
   where
-    ropts             = ["-r", "markdown", "-w", "markdown-citations", "--bibliography",
+    ropts             = ["-r", "markdown", "-w", "markdown", "--bibliography",
                          "biblio.bib", "--no-wrap"]
     wopts             = ["-r", "markdown", "-w", "markdown", "--no-wrap", "--natbib"]
     styleToTest style = test style (ropts ++ ["--csl", style ++ ".csl"])

tests/Tests/Readers/LaTeX.hs

@@ -2,14 +2,20 @@
 module Tests.Readers.LaTeX (tests) where
 
 import Text.Pandoc.Definition
+import Data.Monoid (mempty)
+import Text.CSL (Reference, readBiblioFile)
 import Test.Framework
 import Tests.Helpers
 import Tests.Arbitrary()
 import Text.Pandoc.Builder
 import Text.Pandoc
+import System.IO.Unsafe (unsafePerformIO)
+
+refs :: [Reference]
+refs = unsafePerformIO $ readBiblioFile "biblio.mods"
 
 latex :: String -> Pandoc
-latex = readLaTeX def
+latex = readLaTeX def{ readerReferences = refs }
 
 infix 4 =:
 (=:) :: ToString c
@@ -71,7 +77,7 @@ baseCitation = Citation{ citationId      = "item1"
                        }
 
 rt :: String -> Inlines
-rt = rawInline "latex"
+rt = const mempty -- rawInline "latex"
 
 natbibCitations :: Test
 natbibCitations = testGroup "natbib"

tests/biblio.mods

@@ -0,0 +1,106 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<modsCollection xmlns="http://www.loc.gov/mods/v3">
+<mods ID="item1">
+    <titleInfo>
+        <title>First Book</title>
+    </titleInfo>
+    <name type="personal">
+        <namePart type="given">John</namePart>
+        <namePart type="family">Doe</namePart>
+        <role>
+            <roleTerm authority="marcrelator" type="text">author</roleTerm>
+        </role>
+    </name>
+    <originInfo>
+        <issuance>monographic</issuance>
+        <dateIssued>2005</dateIssued>
+        <publisher>Cambridge University Press</publisher>
+        <place>
+            <placeTerm type="text">Cambridge</placeTerm>
+        </place>
+    </originInfo>
+    <typeOfResource>text</typeOfResource>
+    <genre authority="marcgt">book</genre>
+    <identifier type="citekey">item1</identifier>
+</mods>
+<mods ID="item2">
+    <titleInfo>
+        <title>Article</title>
+    </titleInfo>
+    <name type="personal">
+        <namePart type="given">John</namePart>
+        <namePart type="family">Doe</namePart>
+        <role>
+            <roleTerm authority="marcrelator" type="text">author</roleTerm>
+        </role>
+    </name>
+    <originInfo>
+        <dateIssued>2006</dateIssued>
+    </originInfo>
+    <typeOfResource>text</typeOfResource>
+    <relatedItem type="host">
+        <titleInfo>
+            <title>Journal of Generic Studies</title>
+        </titleInfo>
+        <originInfo>
+            <issuance>continuing</issuance>
+        </originInfo>
+        <genre authority="marcgt">periodical</genre>
+        <genre>academic journal</genre>
+    </relatedItem>
+    <identifier type="citekey">item2</identifier>
+    <part>
+        <date>2006</date>
+        <detail type="volume"><number>6</number></detail>
+        <extent unit="page">
+            <start>33</start>
+            <end>34</end>
+        </extent>
+    </part>
+</mods>
+<mods ID="пункт3">
+    <titleInfo>
+        <title>Why Water Is Wet</title>
+    </titleInfo>
+    <name type="personal">
+        <namePart type="given">John</namePart>
+        <namePart type="family">Doe</namePart>
+        <role>
+            <roleTerm authority="marcrelator" type="text">author</roleTerm>
+        </role>
+    </name>
+    <name type="personal">
+        <namePart type="given">Jenny</namePart>
+        <namePart type="family">Roe</namePart>
+        <role>
+            <roleTerm authority="marcrelator" type="text">author</roleTerm>
+        </role>
+    </name>
+    <originInfo>
+        <issuance>monographic</issuance>
+        <dateIssued>2007</dateIssued>
+    </originInfo>
+    <typeOfResource>text</typeOfResource>
+    <relatedItem type="host">
+        <titleInfo>
+            <title>Third Book</title>
+        </titleInfo>
+        <name type="personal">
+            <namePart type="given">Sam</namePart>
+            <namePart type="family">Smith</namePart>
+            <role>
+                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
+            </role>
+        </name>
+        <originInfo>
+            <dateIssued>2007</dateIssued>
+            <publisher>Oxford University Press</publisher>
+            <place>
+                <placeTerm type="text">Oxford</placeTerm>
+            </place>
+        </originInfo>
+        <genre>collection</genre>
+    </relatedItem>
+    <identifier type="citekey">пункт3</identifier>
+</mods>
+</modsCollection>

tests/latex-reader.native

@@ -260,7 +260,7 @@ Pandoc (Meta {docTitle = [Str "Pandoc",Space,Str "Test",Space,Str "Suite"], docA
 ,HorizontalRule
 ,Header 1 ("",[],[]) [Str "LaTeX"]
 ,BulletList
- [[Para [Cite [Citation {citationId = "smith.1899", citationPrefix = [], citationSuffix = [Str ",",Space,Str "22-23"], citationMode = AuthorInText, citationNoteNum = 0, citationHash = 0}] [RawInline "latex" "\\cite[22-23]{smith.1899}"]]]
+ [[Para [RawInline "latex" "\\cite[22-23]{smith.1899}"]]
  ,[Para [RawInline "latex" "\\doublespacing"]]
  ,[Para [Math InlineMath "2+2=4"]]
  ,[Para [Math InlineMath "x \\in y"]]