ods2csv 0.1 → 0.1.0.1

raw patch · 2 files changed

+37/−14 lines, 2 filesdep ~optparse-applicative

Dependency ranges changed: optparse-applicative

Files

• ods2csv.cabal +6/−3
• src/Main.hs +31/−11

@@ -1,5 +1,5 @@ Name:                ods2csv-Version:             0.1+Version:             0.1.0.1 Synopsis:            Convert Open Document Spreadsheet ODS to CSV Description:   Usually you would convert ODS to CSV via @libreoffice --headless@@@ -16,7 +16,7 @@   but might not be true if generated or manipulated by other tools.   .   You can feed generated CSV files to LaTeX @datatool.sty@, @csvmidi@,-  or @csvreplace@ from Haskell package @spreadsheet@.+  Hledger's CSV import or @csvreplace@ from Haskell package @spreadsheet@.   .   Example runs of the program:   .@@ -33,8 +33,11 @@   The program supports UTF-8 encoding for FODS/XML input,   and applies Latin1 decoding otherwise.   For writing CSV it uses the system's default encoding.+  You may only feed FODS, not ODS via pipe/stdin to the program.   .   ToDo: Provide FODS parser as library function, maybe in @spreadsheet@.+  .+  Alternatives: <https://www.visidata.org/>, <https://csvkit.rtfd.org/> Homepage:            https://hub.darcs.net/thielema/ods2csv License:             BSD3 License-File:        LICENSE@@ -45,7 +48,7 @@ Cabal-Version:       >=1.10  Source-Repository this-  Tag:         0.1+  Tag:         0.1.0.1   Type:        darcs   Location:    https://hub.darcs.net/thielema/ods2csv 

@@ -5,12 +5,13 @@  import qualified Options.Applicative as OP +import qualified Text.HTML.Tagchup.Format as TagFormat import qualified Text.HTML.Tagchup.Parser as TagParser import qualified Text.HTML.Tagchup.Process as TagProc import qualified Text.HTML.Tagchup.Tag.Match as TagMatch import qualified Text.HTML.Tagchup.Tag as Tag import qualified Text.XML.Basic.Attribute as Attr-import qualified Text.XML.Basic.Name.MixedCase as Name -- Qualified would also be appropriate+import qualified Text.XML.Basic.Name.MixedCase as Name import qualified Text.XML.Basic.Name as NameC  import qualified Codec.Archive.Zip as Zip@@ -25,7 +26,8 @@ import qualified Data.Char as Char import Data.Function.HT (Id) import Data.Tuple.HT (mapSnd)-import Data.Maybe (mapMaybe)+import Data.Maybe.HT (toMaybe)+import Data.Maybe (mapMaybe, fromMaybe) import Data.Monoid ((<>))  import Control.Monad (join, guard)@@ -91,21 +93,39 @@       =<< readInput  +{-+LibreOffice escapes double spaces as in @Foo  Bar@ like so:++> <text:p>Foo <text:s/>Baar</text:p>+-}+flattenText :: [Tag.T Name.T String] -> String+flattenText =+   concatMap+      (\tag ->+         fromMaybe+            (printf "*** unexpected tag %s in cell text ***"+               (TagFormat.xml [tag] "")) $+         Tag.maybeText tag+         <|>+         toMaybe (TagMatch.openNameLit "text:s" tag) " "+         <|>+         toMaybe (TagMatch.closeNameLit "text:s" tag) "")+ extractTablesContents :: [Tag.T Name.T String] -> [(String, [[String]])] extractTablesContents =    map (mapSnd (          map             (concatMap                (uncurry $ \n ->-                  (\texts ->-                     replicate n $-                     case texts of-                        text:_ -> text-                        [] -> "")-                   .-                   mapMaybe Tag.maybeText-                   .-                   dropWhile (not . TagMatch.openNameLit "text:p"))+                  replicate n+                  .+                  flattenText+                  .+                  takeWhile (not . TagMatch.closeNameLit "text:p")+                  .+                  drop 1+                  .+                  dropWhile (not . TagMatch.openNameLit "text:p"))              .              snd              .