[PATCH] Add Blast+.

[Ricardo Wurmus]

>From 81cbb9bfa523d56c68d5f9f4feed3676edb5a414 Mon Sep 17 00:00:00 2001
From: Ricardo Wurmus <address@hidden>
Date: Tue, 16 Jun 2015 16:24:24 +0200
Subject: [PATCH] gnu: Add Blast+.

* gnu/packages/bioinformatics.scm (blast+): New variable.
---
 gnu/packages/bioinformatics.scm | 156 ++++++++++++++++++++++++++++++++++++++++
 1 file changed, 156 insertions(+)

diff --git a/gnu/packages/bioinformatics.scm b/gnu/packages/bioinformatics.scm
index ac4c50d..4a55040 100644
--- a/gnu/packages/bioinformatics.scm
+++ b/gnu/packages/bioinformatics.scm
@@ -31,6 +31,7 @@
   #:use-module (gnu packages base)
   #:use-module (gnu packages boost)
   #:use-module (gnu packages compression)
+  #:use-module (gnu packages cpio)
   #:use-module (gnu packages file)
   #:use-module (gnu packages java)
   #:use-module (gnu packages linux)
@@ -294,6 +295,161 @@ into separate processes; and more.")
     (inputs
      `(("python2-numpy" ,python2-numpy)))))
 
+(define-public blast+
+  (package
+    (name "blast+")
+    (version "2.2.30")
+    (source (origin
+              (method url-fetch)
+              (uri (string-append
+                    "ftp://ftp.ncbi.nlm.nih.gov/blast/executables/blast+/";
+                    version "/ncbi-blast-" version "+-src.tar.gz"))
+              (sha256
+               (base32
+                "0h0fj5cpx6zpfwixgx5f5xbr4rn3cnai0x3j7grrg50vr18jvxr6"))))
+    (build-system gnu-build-system)
+    (arguments
+     `(;; There are three(!) tests for this massive library, and all fail with
+       ;; "unparsable timing stats".
+       ;; ERR [127] --  [util/regexp] test_pcre.sh     (unparsable timing 
stats)
+       ;; ERR [127] --  [serial/datatool] datatool.sh     (unparsable timing 
stats)
+       ;; ERR [127] --  [serial/datatool] datatool_xml.sh     (unparsable 
timing stats)
+       #:tests? #f
+       #:out-of-source? #t
+       #:parallel-build? #f ; not supported
+       #:phases
+       (modify-phases %standard-phases
+         (add-before
+          'configure 'set-HOME
+          ;; $HOME needs to be set at some point during the configure phase
+          (lambda _ (setenv "HOME" "/tmp") #t))
+         (add-after
+          'unpack 'enter-dir
+          (lambda _ (chdir "c++") #t))
+         (add-after
+          'enter-dir 'fix-build-system
+          (lambda _
+            ;; Proceed even though the weird build system says that generated
+            ;; files are out of date
+            (setenv "NCBICXX_RECONF_POLICY" "warn")
+
+            ;; Remove bundled bzip2 and zlib
+            (delete-file-recursively "src/util/compress/bzip2")
+            (delete-file-recursively "src/util/compress/zlib")
+            (substitute* "src/util/compress/Makefile.in"
+              (("bzip2 zlib api") "api"))
+
+            ;; Remove useless msbuild directory
+            (delete-file-recursively 
"src/build-system/project_tree_builder/msbuild")
+
+            ;; Some of the files we're patching are
+            ;; ISO-8859-1-encoded, so choose it as the default
+            ;; encoding so the byte encoding is preserved.
+            (with-fluids ((%default-port-encoding #f))
+              (substitute* (find-files "src/build-system" "config.*")
+                (("LN_S=/bin/\\$LN_S") (string-append "LN_S=" (which "ln")))
+                (("/bin/sh") (which "bash"))
+                (("^PATH=.*") "")))
+
+            ;; fix static and generated shebangs
+            (substitute* (find-files "scripts/common/check" "\\.sh")
+              (("/bin/sh") (which "bash")))
+
+            ;; rewrite "/var/tmp" in check script
+            (substitute* "scripts/common/check/check_make_unix.sh"
+              (("/var/tmp") (string-append (getcwd) "/build/build")))
+
+            ;; fix path to "echo"
+            (substitute* '("src/build-system/Makefile.rules_with_autodep.in"
+                           "src/build-system/Makefile.meta.gmake=no"
+                           "src/build-system/Makefile.meta_r"
+                           "src/build-system/Makefile.requirements")
+              (("/bin/echo") (which "echo")))
+
+            ;; fix path to "basename"
+            (substitute* '("src/build-system/Makefile.in.top")
+              (("/usr/bin/basename") (which "basename")))
+
+            ;; fix path to "mv"
+            (substitute* '("src/build-system/Makefile.rules_with_autodep.in"
+                           "src/build-system/Makefile.meta_p")
+              (("/bin/mv") (which "mv")))
+
+            ;; fix path to "rm"
+            (substitute* '("src/build-system/Makefile.mk.in"
+                           "src/build-system/Makefile.meta.in"
+                           "scripts/common/impl/run_with_lock.sh")
+              (("/bin/rm") (which "rm")))
+
+            ;; fix path to "cp"
+            (substitute* '("src/build-system/Makefile.configurables.real"
+                           "src/build-system/Makefile.mk.in"
+                           "src/build-system/configure"
+                           "src/build-system/configure.ac"
+                           "scripts/common/impl/if_diff.sh")
+              (("/bin/cp") (which "cp")))
+
+            ;; fix path to "mkdir"
+            (substitute* '("src/build-system/Makefile.mk.in"
+                           "src/build-system/Makefile.meta.in")
+              (("/bin/mkdir") (which "mkdir")))
+
+            ;; fix path to "dirname"
+            (substitute* '("src/build-system/Makefile.configurables.real"
+                           "src/build-system/Makefile.meta_p")
+              (("/usr/bin/dirname") (which "dirname")))
+
+            ;; make call to "date" deterministic
+            (substitute* "src/build-system/Makefile.meta_l"
+              (("/bin/date") "echo -n 0"))
+
+            ;; do not reset PATH
+            (substitute* (find-files "scripts/common/impl/" "\\.sh")
+              (("^ *PATH=.*") "")
+              (("action=/bin/") "action=")
+              (("export PATH") "echo -n 0"))
+            #t))
+         (replace
+          'configure
+          (lambda* (#:key inputs outputs #:allow-other-keys)
+            (let ((out (assoc-ref outputs "out"))
+                  (lib (string-append (assoc-ref outputs "lib") "/lib"))
+                  (include (string-append (assoc-ref outputs "include")
+                                          "/include/ncbi-tools++")))
+              ;; The 'configure' script doesn't recognize things like
+              ;; '--enable-fast-install'.
+              (zero? (system* "./configure.orig"
+                              (string-append "--with-build-root=" (getcwd) 
"/build")
+                              (string-append "--prefix=" out)
+                              (string-append "--libdir=" lib)
+                              (string-append "--includedir=" include)
+                              (string-append "--with-bz2="
+                                             (assoc-ref inputs "bzip2"))
+                              (string-append "--with-z="
+                                             (assoc-ref inputs "zlib"))
+                              ;; Each library is built twice by default, once
+                              ;; with "-static" in its name, and again
+                              ;; without.
+                              "--without-static"
+                              "--with-dll"))))))))
+    (outputs '("out"       ; 19 MB
+               "lib"       ; 203MB
+               "include")) ; 32MB
+    (inputs
+     `(("bzip2" ,bzip2)
+       ("zlib" ,zlib)))
+    (native-inputs
+     `(("cpio" ,cpio)))
+    (home-page "http://blast.ncbi.nlm.nih.gov";)
+    (synopsis "Basic local alignment search tool")
+    (description
+     "BLAST is a popular method of performing a DNA or protein sequence
+similarity search, using heuristics to produce results quickly.  It also
+calculates an “expect value” that estimates how many matches would have
+occurred at a given score by chance, which can aid a user in judging how much
+confidence to have in an alignment.")
+    (license license:public-domain)))
+
 (define-public bowtie
   (package
     (name "bowtie")
-- 
2.1.0