LaTeX_funs.hs revision 026af5e4803af16acb4fe738725eb0c096499cbe
{- |
Module : $Header$
Description : auxiliary functions for LaTeX printing
Copyright : (c) Klaus Luettich, Uni Bremen 2002-2006
License : GPLv2 or higher, see LICENSE.txt
Maintainer : Christian.Maeder@dfki.de
Stability : provisional
Portability : portable
Auxiliary functions for LaTeX printing
Functions to calculate the length of a given word as it would be
printed with LaTeX according to one of four categories of words
useful for CASL:
* keywords -- all the things that were printed in boldface
* structid -- all the names used in the structured context of CASL
* annotation -- all the comments and annotations of CASL in a smaller font
* axiom -- identifiers in math mode for CASL Basic specs
-}
module Common.LaTeX_funs
( calcLineLen
, axiom_width
, latex_macro
, flushright
, casl_comment_latex
, casl_normal_latex
, hc_sty_small_keyword
, hc_sty_plain_keyword
, hc_sty_casl_keyword
, hc_sty_axiom
, hc_sty_structid
, hc_sty_structid_indexed
, hc_sty_id
, startTab, endTab, setTab
, setTabWSp
, startAnno
, endAnno
, escapeSpecial
, escapeLatex
) where
import qualified Data.Map as Map
import Data.Char
import Data.List (isPrefixOf)
import Common.LaTeX_maps
import Common.Lib.Pretty as Pretty
import Common.Parsec
import Text.ParserCombinators.Parsec as Parsec
-- | a constant String for starting a LaTeX indentation with tab stop
startTab :: String
startTab = "\\@begT@"
-- | a constant String for releasing a LaTeX indentation with tab stop
endTab :: String
endTab = "\\@endT@"
-- | a constant String to set a tab stop and enable it
setTab :: String
setTab = "\\="
-- | a constant String indicating the start of a space based indentation
setTabWSp :: String
setTabWSp = "\\@setTS@{"
{- | functions for calculating an integer value according to a given
length in LaTeX points.
-}
calcLineLen :: Int -> Int
calcLineLen len = scaleDown $ len * 351
-- Units per mm found in: Karsten Guenther, "Einfuehrung in LaTeX2e" (p.376)
scaleDown :: Int -> Int
scaleDown = (`div` 44) . (+ 15)
{- functions to calculate a word-width in integer with a given word
type or purpose
-}
data Word_type =
Keyword | StructId | Normal | Comment | Annotation | AnnotationBold | Axiom
deriving (Show, Eq)
calc_word_width :: Word_type -> String -> Int
calc_word_width wt s = scaleDown $ calc_word_widthAux wt s
calc_word_widthAux :: Word_type -> String -> Int
calc_word_widthAux wt s = Map.findWithDefault
(sum_char_width_deb (showString "In map \"" . shows wt . showString "\" \'")
wFM k_wFM s - correction) s wFM
where (wFM, k_wFM) = case wt of
Keyword -> (keyword_map, key_keyword_map)
StructId -> (structid_map, key_structid_map)
Comment -> (comment_map, key_comment_map)
Annotation -> (annotation_map, key_annotation_map)
AnnotationBold -> (annotationbf_map,
key_annotationbf_map)
Axiom -> (axiom_map, key_axiom_map)
Normal -> (normal_map, key_normal_map)
correction = case wt of
Axiom -> itCorrection s
_ -> 0
itCorrection :: String -> Int
itCorrection [] = 0
itCorrection s
| length s < 2 || head s == '\\' = 0
| otherwise = itCorrection' 0 s
where itCorrection' :: Int -> String -> Int
itCorrection' _ [] = error "itCorrection' applied to empty List"
itCorrection' r ys@[y1, y2]
| not (isAlphaNum y1) = r
| not (isAlphaNum y2) = r
| otherwise = r + lookupCorrection ys
itCorrection' r (y1 : ys@(y2 : _))
| not (isAlphaNum y1) = itCorrection' r ys
| otherwise =
itCorrection'
(r + lookupCorrection [y1, y2])
ys
itCorrection' _ _ = error ("itCorrection' doesn't work with " ++ s)
lookupCorrection str = Map.findWithDefault def_cor str
italiccorrection_map
def_cor = 610
sum_char_width_deb :: (String -> String) -- only used for an hackie debug thing
-> Map.Map String Int
-> Map.Map Char [String] -> String -> Int
sum_char_width_deb _pref_fun cFM key_cFM s = sum_char_width' s 0
where sum_char_width' [] r = r
sum_char_width' [c] r = r + case c of
'}' -> 0
'{' -> 0
' ' -> lookupWithDefault_cFM "~"
_ -> lookupWithDefault_cFM [c]
sum_char_width' full@(c1 : rest@(c2 : cs)) r
| isLigature [c1, c2] = case Map.lookup [c1, c2] cFM of
Just l -> sum_char_width' cs (r + l)
Nothing -> sum_char_width' rest nl
| [c1, c2] == "\\ " =
sum_char_width' cs (r + lookupWithDefault_cFM "~")
| c1 == ' ' =
sum_char_width' rest (r + lookupWithDefault_cFM "~")
| otherwise = case prefixIsKey full key_cFM of
Just key -> sum_char_width'
(drop (length key) full)
$ r + (cFM Map.! key)
Nothing -> if c1 == '\\' then
sum_char_width'
(dropWhile isAlpha rest)
$ r + lookupWithDefault_cFM "~"
else sum_char_width' rest nl
where nl = r + lookupWithDefault_cFM [c1]
lookupWithDefault_cFM s' = Map.findWithDefault 2200 s' cFM
-- 2200 may not be optimal
prefixIsKey :: String -> Map.Map Char [String] -> Maybe String
prefixIsKey [] _ = Nothing
prefixIsKey ls@(c : _) key_cFM = case filter (`isPrefixOf` ls)
$ Map.findWithDefault [] c key_cFM of
[] -> Nothing
s : _ -> Just s
isLigature :: String -> Bool
isLigature s = case s of
[_, _] -> Map.findWithDefault False s ligatures
_ -> False
keyword_width, structid_width, axiom_width, annotationbf_width,
comment_width, normal_width :: String -> Int
annotationbf_width = calc_word_width AnnotationBold
keyword_width = calc_word_width Keyword
structid_width = calc_word_width StructId
comment_width = calc_word_width Comment
normal_width = calc_word_width Normal
axiom_width = sum . map (calc_word_width Axiom) . parseAxiomString
{- |
latex_macro creates a document ('Doc') containing String
that has a zero width.
So it can be used for LaTeX-macros not needing any space, i.e.
@\textit{@ or @}@ -}
latex_macro :: String -> Doc
latex_macro = sizedText 0
casl_keyword_latex, casl_annotationbf_latex,
casl_axiom_latex,
casl_comment_latex, casl_structid_latex,
casl_normal_latex :: String -> Doc
casl_annotationbf_latex s = sizedText (annotationbf_width s) s
casl_structid_latex s = sizedText (structid_width s) s
casl_comment_latex s = sizedText (comment_width s) s
casl_keyword_latex s = sizedText (keyword_width s) s
casl_normal_latex s = sizedText (normal_width s) s
casl_axiom_latex s = sizedText (axiom_width s) s
-- | sort, op, pred, type and its plurals
hc_sty_casl_keyword :: String -> Doc
hc_sty_casl_keyword str =
sizedText (keyword_width "preds") $ '\\' : map toUpper str
hc_sty_plain_keyword :: String -> Doc
hc_sty_plain_keyword kw =
latex_macro "\\KW{" <> casl_keyword_latex (escapeSpecial kw)
<> latex_macro "}"
hc_sty_small_keyword :: String -> Doc
hc_sty_small_keyword kw =
latex_macro "\\KW{" <> casl_annotationbf_latex (escapeSpecial kw)
<> latex_macro "}"
hc_sty_axiom, hc_sty_structid, hc_sty_id, hc_sty_structid_indexed
:: String -> Doc
hc_sty_structid sid = latex_macro "\\SId{" <> sid_doc <> latex_macro "}"
where sid_doc = casl_structid_latex (escapeSpecial sid)
hc_sty_structid_indexed sid =
latex_macro "\\SIdIndex{" <> sid_doc <> latex_macro "}"
where sid_doc = casl_structid_latex (escapeSpecial sid)
hc_sty_id i = latex_macro "\\Id{" <> id_doc <> latex_macro "}"
where id_doc = casl_axiom_latex i
hc_sty_axiom ax = latex_macro "\\Ax{" <> ax_doc <> latex_macro "}"
where ax_doc = casl_axiom_latex ax
-- | flush argument doc to the right
flushright :: Doc -> Doc
flushright = (latex_macro "\\`" <>)
-- | a constant String for the start of annotations
startAnno :: String
startAnno = "{\\small{}"
-- | a constant string ending an annotation
endAnno :: String
endAnno = "%@%small@}"
escapeSpecial :: String -> String
escapeSpecial = concatMap $ \ c -> if elem c "_%$&{}#" then '\\' : [c] else
Map.findWithDefault [c] c escapeMap
{- http://dhelta.net/hprojects/HaTeX/code/HaTeX-3.1.0/Text/LaTeX/Base/Syntax.hs
changes _ to \_{} -}
escapeLatex :: String -> String
escapeLatex = concatMap $ \ c -> case () of
()
| elem c "_%$&{}#" -> "\\Ax{\\" ++ c : "}"
| elem c "<|>=-!()[]?:;,./*+@" -> "\\Ax{" ++ c : "}"
| otherwise -> Map.findWithDefault [c] c escapeMap
parseAxiomString :: String -> [String]
parseAxiomString s = case parse axiomString "" s of
Left _ -> [s]
Right l -> l
axiomString :: CharParser st [String]
axiomString = do
l <- many parseAtom
eof
return $ concat l
parseAtom :: CharParser st [String]
parseAtom = fmap concat
((tryString "\\Ax{" <|> tryString "\\Id{" <|> string "{")
>> many parseAtom << Parsec.char '}')
<|> do
b <- Parsec.char '\\'
s <- fmap (: []) (satisfy (\ c -> isSpace c
|| elem c "_~^|\'\",;:.`\\{}[]%$&#()"))
<|> many1 letter
return [b : s]
<|> do
s <- many1 letter
return [s]
<|> do
c <- satisfy (/= '}')
return [[c]]
-- | a character map for special latex characters
escapeMap :: Map.Map Char String
escapeMap = Map.fromList
[('\\', "\\Ax{\\setminus}"),
('^', "\\Ax{\\hat{\\ }}"),
('"', "''"),
('~', "\\Ax{\\sim}"),
('\160', "\\ "),
('\162', "\\Id{\\textcent}"),
('\164', "\\Id{\\textcurrency}"),
('\165', "\\Id{\\textyen}"),
('\166', "\\Id{\\textbrokenbar}"),
('\170', "\\Id{\\textordfeminine}"),
('\171', "\\Id{\\guillemotleft}"),
('\172', "\\Ax{\\neg}"),
('\173', "-"),
('\174', "\\Id{\\textregistered}"),
('\175', "\\Ax{\\bar{\\ }}"),
('\176', "\\Id{\\textdegree}"),
('\177', "\\Ax{\\pm}"),
('\178', "\\Ax{^2}"),
('\179', "\\Ax{^3}"),
('\180', "\\Ax{\\acute{\\ }}"),
('\181', "\\Ax{\\mu}"),
('\185', "\\Ax{^1}"),
('\186', "\\Id{\\textordmasculine}"),
('\187', "\\Id{\\guillemotright}"),
('\192', "\\Ax{\\grave{A}}"),
('\193', "\\Ax{\\acute{A}}"),
('\200', "\\Ax{\\grave{E}}"),
('\201', "\\Ax{\\acute{E}}"),
('\204', "\\Ax{\\grave{I}}"),
('\205', "\\Ax{\\acute{I}}"),
('\208', "\\Id{\\DH}"),
('\210', "\\Ax{\\grave{O}}"),
('\211', "\\Ax{\\acute{O}}"),
('\215', "\\Ax{\\times}"),
('\217', "\\Ax{\\grave{U}}"),
('\218', "\\Ax{\\acute{U}}"),
('\221', "\\Ax{\\acute{Y}}"),
('\222', "\\Id{\\TH}"),
('\224', "\\Ax{\\grave{a}}"),
('\225', "\\Ax{\\acute{a}}"),
('\232', "\\Ax{\\grave{e}}"),
('\233', "\\Ax{\\acute{e}}"),
('\236', "\\Ax{\\grave{\\Id{\\i}}}"),
('\237', "\\Ax{\\acute{\\Id{\\i}}}"),
('\240', "\\Id{\\dh}"),
('\242', "\\Ax{\\grave{o}}"),
('\243', "\\Ax{\\acute{o}}"),
('\247', "\\Ax{\\div}"),
('\249', "\\Ax{\\grave{u}}"),
('\250', "\\Ax{\\acute{u}}"),
('\253', "\\Ax{\\acute{y}}"),
('\254', "\\Id{\\th}")]
{- acute and grave characters don't work in a tabbing environment
\textcent upto textbrokenbar requires \usepackage{textcomp}
whereas \guillemot, eth, and thorn \usepackage[T1]{fontenc}
-}