some fixes for new corpora

Makefile.def

@@ -164,15 +164,18 @@ GZCAT    := ${GZIP} -cd
 ZCAT     := gzip -cd
 
 SORT     := sort -T ${TMPDIR} -S1G --parallel=${THREADS}
-# SORT   := sort -T ${TMPDIR} --parallel=${THREADS} -S1G --compress-program=${GZIP}
-# SORT   := sort -T ${TMPDIR} --parallel=${THREADS} -S1G
+# SORT   := sort -T ${TMPDIR} -S1G --parallel=${THREADS} --compress-program=${GZIP}
 
 UNIQ     := ${SORT} -u
 MERGE    := ${SORT} -m -u
 
-ifneq (${shell which parsort 2>/dev/null},)
-  SORT   := parsort -T ${TMPDIR} -S1G
-endif
+
+## I don't really trust parsort ...
+
+# ifneq (${shell which parsort 2>/dev/null},)
+#   SORT   := parsort -T ${TMPDIR} -S1G
+# endif
+
 
 
 ## seems to be necessary to run with threads on HPC nodes

corpus/HPLT/Makefile

@@ -30,14 +30,6 @@ all-job:
 # 	${MAKE} HPC_CORES=1 HPC_MEM=4g HPC_TIME=72:00 fetch-job.submit
 
 
-new-sr:
-	${MAKE} -C src all
-	${MAKE} LANGUAGE=sr annotate_files
-	${MAKE} LANGUAGE=en annotate_files
-	${MAKE} moses
-	${MAKE} copy-tmx-files
-	${MAKE} publish
-
 
 fetch-job:
 	${MAKE} -C src fetch
@@ -68,6 +60,7 @@ publish-job:
 	${MAKE} moses
 	${MAKE} copy-tmx-files
 	${MAKE} publish
+	${MAKE} release
 #	${MAKE} wordalign/Makefile
 #	${MAKE} -C wordalign submit-all
 #	${MAKE} HPC_CORES=8 HPC_MEM=8g HPC_TIME=72:00 udparse.submit

corpus/HPLT/Makefile.def

@@ -4,13 +4,13 @@
 
 CORPUS  = HPLT
 
-RELEASE = 1
+RELEASE = 1.1
 VERSION = v${RELEASE}
 
 # with SRCHTML you can define some extra HTML code that will be placed just
 # after the name of the sub-corpus
 # EXTRAHTML is code which will be put just after the 'Download' header
 
 SRCHTML = Parallel corpora from Web Crawls collected in the <a href="https://hplt-project.org/">HPLT project</a>
-EXTRAHTML = The data is released under the <a href="https://creativecommons.org/share-your-work/public-domain/cc0/">Creative Commons CC0 license</a> ("no rights reserved"), see <a href="http://paracrawl.eu/download.html">http://paracrawl.eu/download.html</a> for more details
+EXTRAHTML = The data packaging is released under the <a href="https://creativecommons.org/share-your-work/public-domain/cc0/">Creative Commons CC0 license</a> ("no rights reserved"), see <a href="https://hplt-project.org/">https://hplt-project.org/</a> for more details
 CITENOTE = Please, acknowledge the HPLT project at <a href="https://hplt-project.org/">https://hplt-project.org/</a>. This version is derived from the original release at their website adjusted for redistribution via the OPUS corpus collection. Please, acknowledge OPUS as well for this service.