gnu: ccextractor: Update to 0.94 and fix build.

* gnu/packages/gstreamer.scm (ccextractor): Update to 0.94.
[source]: Apply patches.  Add FIXME comment.
[build-system]: Switch to gnu-build-system.
[arguments]: Delete #tests? argument.  Delete trailing #t and use gexps.
Adjust configure flags accordingly, removing WITH_SHARING (doesn't exist for
Autoconf) and adding '--without-rust'.  Override check phase.
[native-inputs]: Remove labels.  Remove python-wrapper and perl.
[propagated-inputs]: Remove labels.  Remove nanomsg and zlib.  Replace
leptonica with leptonica-1.80.  replace tesseract-ocr with tesseract-ocr-4.
* gnu/packages/patches/ccextractor-add-missing-header.patch: New file.
* gnu/packages/patches/ccextractor-autoconf-tesseract.patch: Likewise.
* gnu/packages/patches/ccextractor-fix-ocr.patch: Likewise.
* gnu/local.mk (dist_patch_DATA): Register them.

5 files changed, 139 insertions, 38 deletions

diff --git a/gnu/local.mk b/gnu/local.mk
index 0a1a5d077f..2944211e12 100644
--- a/gnu/local.mk
+++ b/gnu/local.mk
@@ -966,6 +966,9 @@ dist_patch_DATA =						\
   %D%/packages/patches/calibre-remove-test-sqlite.patch		\
   %D%/packages/patches/calibre-remove-test-unrar.patch		\
   %D%/packages/patches/catdoc-CVE-2017-11110.patch		\
+  %D%/packages/patches/ccextractor-add-missing-header.patch	\
+  %D%/packages/patches/ccextractor-autoconf-tesseract.patch	\
+  %D%/packages/patches/ccextractor-fix-ocr.patch		\
   %D%/packages/patches/chez-scheme-bin-sh.patch			\
   %D%/packages/patches/circos-remove-findbin.patch		\
   %D%/packages/patches/cdparanoia-fpic.patch			\
diff --git a/gnu/packages/gstreamer.scm b/gnu/packages/gstreamer.scm
index 2bb4d48991..f33a30a02b 100644
--- a/gnu/packages/gstreamer.scm
+++ b/gnu/packages/gstreamer.scm
@@ -10,6 +10,7 @@
 ;;; Copyright © 2019, 2020 Marius Bakke <mbakke@fastmail.com>
 ;;; Copyright © 2020 Liliana Marie Prikler <liliana.prikler@gmail.com>
 ;;; Copyright © 2020 Michael Rohleder <mike@rohleder.de>
+;;; Copyright © 2023 Maxim Cournoyer <maxim.cournoyer@gmail.com>
 ;;;
 ;;; This file is part of GNU Guix.
 ;;;
@@ -32,7 +33,6 @@
   #:use-module (guix gexp)
   #:use-module (guix download)
   #:use-module (guix git-download)
-  #:use-module (guix build-system cmake)
   #:use-module (guix build-system gnu)
   #:use-module (guix build-system meson)
   #:use-module (guix build-system trivial)
@@ -171,49 +171,48 @@ module for the DMA capture of the video flow.")
 (define-public ccextractor
   (package
     (name "ccextractor")
-    (version "0.88")
+    (version "0.94")
     (source
      (origin
        (method git-fetch)
-       (uri
-        (git-reference
-         (url "https://github.com/CCExtractor/ccextractor")
-         (commit (string-append "v" version))))
+       (uri (git-reference
+             (url "https://github.com/CCExtractor/ccextractor")
+             (commit (string-append "v" version))))
        (file-name (git-file-name name version))
+       ;; FIXME: Delete the 'src/thirdparty directory and unbundle the
+       ;; libraries it contains, such as freetype, libpng, zlib, and others.
+       (patches (search-patches "ccextractor-add-missing-header.patch"
+                                "ccextractor-autoconf-tesseract.patch"
+                                "ccextractor-fix-ocr.patch"))
        (sha256
-        (base32 "1sya45hvv4d46bk7541yimmafgvgyhkpsvwfz9kv6pm4yi1lz6nb"))))
-    (build-system cmake-build-system)
+        (base32 "1hrk4xlzkvk9pnv0yr4whcsh8h4fzk42mrf30dsr3xzh1lgpfslg"))))
+    (build-system gnu-build-system)
     (arguments
-     `(#:tests? #f                      ; No target
-       #:configure-flags
-       (list
-        "-DWITH_FFMPEG=ON"
-        "-DWITH_OCR=ON"
-        "-DWITH_SHARING=ON"
-        "-DWITH_HARDSUBX=ON")
-       #:phases
-       (modify-phases %standard-phases
-         ;; The package is in a sub-dir of this repo.
-         (add-after 'unpack 'chdir
-           (lambda _
-             (chdir "src")
-             #t))
-         (add-after 'chdir 'fix-build-errors
-           (lambda _
-             (substitute* "CMakeLists.txt"
-               (("libnanomsg")
-                "nanomsg"))
-             #t)))))
-    (native-inputs
-     `(("perl" ,perl)
-       ("pkg-config" ,pkg-config)
-       ("python" ,python-wrapper)))
-    (inputs
-     `(("ffmeg" ,ffmpeg-3.4)
-       ("nanomsg" ,nanomsg)
-       ("leptonica" ,leptonica)
-       ("ocr" ,tesseract-ocr)
-       ("zlib" ,zlib)))
+     (list #:configure-flags
+           #~(list "--enable-ffmpeg"
+                   "--enable-ocr"
+                   "--enable-hardsubx"
+                   ;; Disable Rust support, as there's no rust source included
+                   ;; and cargo wants to fetch the crates from the network
+                   ;; (see:
+                   ;; https://github.com/CCExtractor/ccextractor/issues/1502).
+                   "--without-rust")
+           #:phases #~(modify-phases %standard-phases
+                        (add-after 'unpack 'chdir
+                          (lambda _
+                            (chdir "linux")))
+                        (add-after 'chdir 'patch-pre-build.sh
+                          (lambda _
+                            (substitute* "pre-build.sh"
+                              (("/usr/bin/env") (which "env")))))
+                        (replace 'check
+                          (lambda* (#:key tests? #:allow-other-keys)
+                            (when tests?
+                              ;; There is no test suite; simply run the binary
+                              ;; to validate there are no obvious problems.
+                              (invoke "./ccextractor" "--help")))))))
+    (native-inputs (list autoconf automake pkg-config))
+    (inputs (list ffmpeg-3.4 leptonica-1.80 tesseract-ocr-4))
     (synopsis "Closed Caption Extractor")
     (description "CCExtractor is a tool that analyzes video files and produces
 independent subtitle files from the closed captions data.  It is portable, small,
diff --git a/gnu/packages/patches/ccextractor-add-missing-header.patch b/gnu/packages/patches/ccextractor-add-missing-header.patch
new file mode 100644
index 0000000000..6c6c02be51
--- /dev/null
+++ b/gnu/packages/patches/ccextractor-add-missing-header.patch
@@ -0,0 +1,33 @@
+Upstream status: https://github.com/CCExtractor/ccextractor/pull/1505
+
+From 272e0e1410135c3a5de105cecce400c7b2a9ed5b Mon Sep 17 00:00:00 2001
+From: Maxim Cournoyer <maxim.cournoyer@gmail.com>
+Date: Fri, 17 Mar 2023 09:49:29 -0400
+Subject: [PATCH] linux/Makefile.am: Add missing generated header.
+
+This header is generated by the pre-build.sh script.  The compilation
+fails if it is missing.
+
+* linux/Makefile.am (ccextractor_SOURCES): Add
+../src/lib_ccx/compile_info_real.h.
+---
+ linux/Makefile.am | 1 +
+ 1 file changed, 1 insertion(+)
+
+diff --git a/linux/Makefile.am b/linux/Makefile.am
+index fddc26a5..f83f1eff 100644
+--- a/linux/Makefile.am
++++ b/linux/Makefile.am
+@@ -118,6 +118,7 @@ ccextractor_SOURCES = \
+ 				../src/lib_ccx/bitstream.h \
+ 				../src/lib_ccx/ccx_common_option.c \
+ 				../src/lib_ccx/ccx_common_common.c \
++				../src/lib_ccx/compile_info_real.h \
+ 				../src/lib_ccx/utility.c \
+ 				../src/lib_ccx/activity.c \
+ 				../src/lib_ccx/asf_functions.c \
+
+base-commit: cb496a711923c984251483776e652ca9c027513c
+-- 
+2.39.1
+
diff --git a/gnu/packages/patches/ccextractor-autoconf-tesseract.patch b/gnu/packages/patches/ccextractor-autoconf-tesseract.patch
new file mode 100644
index 0000000000..3ce2925acf
--- /dev/null
+++ b/gnu/packages/patches/ccextractor-autoconf-tesseract.patch
@@ -0,0 +1,37 @@
+Upstream status: https://github.com/CCExtractor/ccextractor/pull/1504
+
+From f587050c2b5805ff5feb667736381dcc9991a5d3 Mon Sep 17 00:00:00 2001
+From: Maxim Cournoyer <maxim.cournoyer@gmail.com>
+Date: Fri, 17 Mar 2023 00:57:13 -0400
+Subject: [PATCH] linux/configure.ac: Fix tesseract conditional problem.
+
+For tesseract-ocr's stock pkg-config, it would produce an error due to
+unquoted whitespace:
+
+  $ test ! -z `pkg-config --libs-only-l --silence-errors tesseract`
+  bash: test: syntax error: `-larchive' unexpected
+
+* linux/configure.ac: Use a positive test, and double-quote the $() command
+substitution.
+---
+ linux/configure.ac | 2 +-
+ 1 file changed, 1 insertion(+), 1 deletion(-)
+
+diff --git a/linux/configure.ac b/linux/configure.ac
+index 45fc11f3..118ee7fa 100644
+--- a/linux/configure.ac
++++ b/linux/configure.ac
+@@ -149,7 +149,7 @@ AS_IF([ (test x$ocr = xtrue || test x$hardsubx = xtrue) && test ! $HAS_LEPT -gt
+ AM_CONDITIONAL(HARDSUBX_IS_ENABLED, [ test x$hardsubx = xtrue ])
+ AM_CONDITIONAL(OCR_IS_ENABLED, [ test x$ocr = xtrue || test x$hardsubx = xtrue ])
+ AM_CONDITIONAL(FFMPEG_IS_ENABLED, [ test x$ffmpeg = xtrue ])
+-AM_CONDITIONAL(TESSERACT_PRESENT, [ test ! -z  `pkg-config --libs-only-l --silence-errors tesseract` ])
++AM_CONDITIONAL(TESSERACT_PRESENT, [ test -n  "$(pkg-config --libs-only-l --silence-errors tesseract)" ])
+ AM_CONDITIONAL(TESSERACT_PRESENT_RPI, [ test -d "/usr/include/tesseract" && test `ls -A /usr/include/tesseract | wc -l` -gt 0 ])
+ AM_CONDITIONAL(SYS_IS_LINUX, [ test `uname -s` = "Linux"])
+ AM_CONDITIONAL(SYS_IS_MAC, [ test `uname -s` = "Darwin"])
+
+base-commit: cb496a711923c984251483776e652ca9c027513c
+-- 
+2.39.1
+
diff --git a/gnu/packages/patches/ccextractor-fix-ocr.patch b/gnu/packages/patches/ccextractor-fix-ocr.patch
new file mode 100644
index 0000000000..9a06026724
--- /dev/null
+++ b/gnu/packages/patches/ccextractor-fix-ocr.patch
@@ -0,0 +1,29 @@
+Extracted from upstream commit 0264e7da2be67182deb031228eb07e6ed4943c81.
+
+diff --git a/src/lib_ccx/ocr.c b/src/lib_ccx/ocr.c
+index 31cc1c5c..73bd4818 100644
+--- a/src/lib_ccx/ocr.c
++++ b/src/lib_ccx/ocr.c
+@@ -331,6 +331,11 @@ char *ocr_bitmap(void *arg, png_color *palette, png_byte *alpha, unsigned char *
+ 	}
+ 
+ 	BOX *crop_points = ignore_alpha_at_edge(copy->alpha, copy->data, w, h, color_pix, &color_pix_out);
++
++	l_int32 x, y, _w, _h;
++
++	boxGetGeometry(crop_points, &x, &y, &_w, &_h);
++
+ 	// Converting image to grayscale for OCR to avoid issues with transparency
+ 	cpix_gs = pixConvertRGBToGray(cpix, 0.0, 0.0, 0.0);
+ 
+@@ -426,8 +431,8 @@ char *ocr_bitmap(void *arg, png_color *palette, png_byte *alpha, unsigned char *
+ 				{
+ 					for (int j = x1; j <= x2; j++)
+ 					{
+-						if (copy->data[(crop_points->y + i) * w + (crop_points->x + j)] != firstpixel)
+-							histogram[copy->data[(crop_points->y + i) * w + (crop_points->x + j)]]++;
++						if (copy->data[(y + i) * w + (x + j)] != firstpixel)
++							histogram[copy->data[(y + i) * w + (x + j)]]++;
+ 					}
+ 				}
+ 				/* sorted in increasing order of intensity */