From bc5aa386675cb39aaf034d73af406c3ddffb4cb0 Mon Sep 17 00:00:00 2001 From: Vagrant Cascadian Date: Thu, 5 Sep 2019 16:12:08 -0700 Subject: [PATCH] gnu: Add odt2txt. * gnu/packages/textutils (odt2txt): New variable. * gnu/packages/package-management (diffoscope)[native-inputs]: Add odt2txt. [use-modules]: Remove accidentally added libreoffice package module. --- gnu/packages/package-management.scm | 2 +- gnu/packages/textutils.scm | 38 +++++++++++++++++++++++++++++ 2 files changed, 39 insertions(+), 1 deletion(-) diff --git a/gnu/packages/package-management.scm b/gnu/packages/package-management.scm index b9eabfb55c..b712bd7ec9 100644 --- a/gnu/packages/package-management.scm +++ b/gnu/packages/package-management.scm @@ -62,7 +62,6 @@ #:use-module (gnu packages image) #:use-module (gnu packages imagemagick) #:use-module (gnu packages java) - #:use-module (gnu packages libreoffice) #:use-module (gnu packages linux) #:use-module (gnu packages lisp) #:use-module (gnu packages llvm) @@ -655,6 +654,7 @@ transactions from C or Python.") ("llvm" ,llvm) ("lz4" ,lz4) ("mono" ,mono) + ("odt2txt" ,odt2txt) ;; no unversioned openjdk available ("openjdk:jdk" ,openjdk12 "jdk") ("openssh" ,openssh) diff --git a/gnu/packages/textutils.scm b/gnu/packages/textutils.scm index 388f164428..2da31973f4 100644 --- a/gnu/packages/textutils.scm +++ b/gnu/packages/textutils.scm @@ -879,6 +879,44 @@ indentation. (home-page "http://docx2txt.sourceforge.net") (license license:gpl3+))) +(define-public odt2txt + (package + (name "odt2txt") + (version "0.5") + (source + (origin + (method git-fetch) + (uri (git-reference + (url "https://github.com/dstosberg/odt2txt/") + (commit (string-append "v" version)))) + (file-name (git-file-name name version)) + (sha256 + (base32 + "0im3kzvhxkjlx57w6h13mc9584c74ma1dyymgvpq2y61av3gc35v")))) + (build-system gnu-build-system) + (arguments + `(#:tests? #f ; no make check + #:make-flags (list "CC=gcc" + (string-append "DESTDIR=" (assoc-ref %outputs "out"))) + #:phases + (modify-phases %standard-phases + ;; no configure script + (delete 'configure)))) + (inputs + `(("zlib" ,zlib))) + (home-page "https://github.com/dstosberg/odt2txt/") + (synopsis "Converter from OpenDocument Text to plain text") + (description "odt2txt is a command-line tool which extracts the text out +of OpenDocument Texts, as produced by OpenOffice.org, KOffice, StarOffice and +others. + +odt2txt can also extract text from some file formats similar to OpenDocument +Text, such as OpenOffice.org XML (*.sxw), which was used by OpenOffice.org +version 1.x and older StarOffice versions. To a lesser extent, odt2txt may be +useful to extract content from OpenDocument spreadsheets (*.ods) and +OpenDocument presentations (*.odp).") + (license license:gpl2))) + (define-public opencc (package (name "opencc")