diff --git a/README.md b/README.md new file mode 100644 --- /dev/null +++ b/README.md @@ -0,0 +1,11 @@ +Automatic hyphenation for Hakyll +================================ + +This library provides [Hakyll][1] compiler to automatically hyphenate each word +in HTML output with Unicode [SOFT-HYPHEN][2] characters. It understands HTML +markup and doesn't try, for example, to hyphenate tag names and attributes. It +uses [hyphenation][3] package to do actual work of splitting words. + +[1]: http://jaspervdj.be/hakyll +[2]: http://en.wikipedia.org/wiki/Soft_hyphen +[3]: http://hackage.haskell.org/package/hyphenation diff --git a/hakyll-contrib-hyphenation.cabal b/hakyll-contrib-hyphenation.cabal --- a/hakyll-contrib-hyphenation.cabal +++ b/hakyll-contrib-hyphenation.cabal @@ -1,7 +1,12 @@ name: hakyll-contrib-hyphenation -version: 0.1.0.2 -synopsis: Automatic hyphenation for Hakyll --- description: +version: 0.1.0.3 +synopsis: automatic hyphenation for Hakyll +description: This library provides Hakyll compiler to automatically + hyphenate each word in HTML output with Unicode SOFT-HYPHEN + characters. It understands HTML markup and doesn't try, for + example, to hyphenate tag names and attributes. It uses + + package to do actual work of splitting words. category: Web stability: experimental author: Pavel Kretov @@ -13,6 +18,7 @@ build-type: Simple cabal-version: >= 1.6 tested-with: GHC == 7.6.3 +extra-source-files: README.md library hs-source-dirs: src diff --git a/src/Hakyll/Contrib/Hyphenation.hs b/src/Hakyll/Contrib/Hyphenation.hs --- a/src/Hakyll/Contrib/Hyphenation.hs +++ b/src/Hakyll/Contrib/Hyphenation.hs @@ -8,36 +8,50 @@ -- they'll be shown only in case when line break actually occured, otherwise -- leaving them invisible. -- --- Of course, this approach has its downsides you must be aware of, so use it --- with caution. Some of these downsides: increased size of generated HTML; --- more difficult text copying and pasing; the text may not even be displayd --- at all in some rare situations. --- -- Here is simple usage pattern for producing hyphenated texts, for example, -- English and Russian: -- --- > import Hakyll.Contrib.Hyphenation (hyphenateHtml, english_US, russian) --- > --- > match "posts/*" $ do --- > route $ idRoute --- > compile $ pandocCompiler --- > >>= hyphenateHtml english_US --- > >>= hyphenateHtml russian --- > >>= loadAndApplyTemplate "..." ctx --- > >>= relativizeUrls +-- > import Hakyll.Contrib.Hyphenation (hyphenateHtml, english_US, russian) +-- > +-- > match "posts/*" $ do +-- > route $ setExtension "html" +-- > compile $ pandocCompiler +-- > >>= hyphenateHtml english_US +-- > >>= hyphenateHtml russian +-- > >>= loadAndApplyTemplate "..." ctx +-- > >>= relativizeUrls +-- +-- Of course, this approach has its downsides you must be aware of, so use it +-- with caution. Some of these downsides: +-- +-- * generated HTML code will slightly increase in size; +-- +-- * soft hyphens will remain in text copied and pasted; +-- +-- * the text may be displayed in a wrong way in some rare situations +-- (for example, in old Unicode-unaware browsers). +-- +-- Taking these into account, you may better prefer to try +-- instead, or +-- not to use text hyphenation at all. module Hakyll.Contrib.Hyphenation ( + module Text.Hyphenation.Language, hyphen, hyphenateText, - hyphenateHtml, - module Text.Hyphenation) where + hyphenateHtmlText, + hyphenateHtml) where -import Data.Char (isSpace) -import Data.List (intercalate) -import Data.List.Split (split, condense, whenElt) -import Text.Hyphenation -import Text.HTML.TagSoup -import Hakyll (Item, Compiler, itemBody, itemSetBody) +import Data.Char (isSpace) +import Data.List (intercalate) +import Data.List.Split (split, condense, whenElt) +import Text.Hyphenation (Hyphenator, hyphenate) +import Text.HTML.TagSoup (Tag(TagText), isTagText, fromTagText, + parseTags, renderTags) +import Hakyll (Item, Compiler, itemBody, itemSetBody) +-- Module from 'hyphenate' package, re-exported for convenience of end-user. +import Text.Hyphenation.Language + -- | Soft-hyphen character. -- This character indicates positions where word line can be safely broken and -- normally displayed only if word break occured otherwise being invisible. @@ -46,15 +60,15 @@ -- | Hyphenate every word of text. -- This function takes a hyphenator (usually a language definition from package --- 'Text.Hyphenation', for example 'english_US') and input string, then returns --- a new string with soft-hyphens inserted into each word. It does not preserve --- whitespaces and instead merges any number of consequent whitespaces into a --- single whitespace. +-- , for example +-- 'english_US') and input string, then returns a new string with soft-hyphens +-- inserted into each word. It does not preserve whitespaces and instead merges +-- any number of consequent whitespaces into a single whitespace. hyphenateText :: Hyphenator -> String -> String -hyphenateText lang = id - . concat - . map (\s -> intercalate hyphen $ hyphenate lang s) - . (split . condense . whenElt) isSpace +hyphenateText lang str = id + $ concat + $ map (\s -> intercalate hyphen $ hyphenate lang s) + $ (split . condense . whenElt) isSpace str -- | Hyphenate every word of HTML-formatted text. -- This function takes a hyphenator and input HTML-formatted string, then @@ -63,19 +77,18 @@ -- any other special markup, instead only text content is affected. It does -- not also preserve original HTML code formatting. hyphenateHtmlText :: Hyphenator -> String -> String -hyphenateHtmlText lang = id - . renderTags - . map (\t -> if not $ isTagText t +hyphenateHtmlText lang str = id + $ renderTags + $ map (\t -> if not $ isTagText t then t else TagText $ hyphenateText lang $ fromTagText t ) - . parseTags + $ parseTags str -- | Hyphenate HTML body of Hakyll item. -- This is the convenience function for using in Hakyll rules as a part of -- compilers chain. Takes a hyphenator, then hyphenates item's body and returns -- a new item with updated body wrapped into 'Compiler' monad. hyphenateHtml :: Hyphenator -> Item String -> Compiler (Item String) -hyphenateHtml lang item = id - $ return +hyphenateHtml lang item = return $ (flip itemSetBody) item $ hyphenateHtmlText lang (itemBody item)