From b31e025f25e4dfe20e6d4d0ee88a267cf2167264 Mon Sep 17 00:00:00 2001
From: bill-auger <mr.j.spam.me@gmail.com>
Date: Sun, 10 Feb 2019 22:37:12 -0500
Subject: add 'sort-entries' script and non-uniform sort warning to 'check.sh'

---
 .gitignore   |  1 +
 check.sh     | 13 +++++++++++++
 sort-entries | 22 ++++++++++++++++++++++
 3 files changed, 36 insertions(+)
 create mode 100755 sort-entries

diff --git a/.gitignore b/.gitignore
index 1acfe1f..83a5258 100644
--- a/.gitignore
+++ b/.gitignore
@@ -7,6 +7,7 @@
 !find-replacements
 !find-deprecated-pkgs
 !README
+!sort-entries
 !SYNTAX
 !your-freedom_emu-blacklist.txt
 !your-privacy-blacklist.txt
diff --git a/check.sh b/check.sh
index fe96b1c..9e84cc7 100755
--- a/check.sh
+++ b/check.sh
@@ -6,6 +6,10 @@
 # notice and this notice are preserved.  This file is offered as-is,
 # without any warranty.
 
+readonly CSV_CHAR=':'
+readonly SEP_CHAR='!'
+
+
 # Verify the blacklist entries are correctly formatted.
 
 bad_entries="$(egrep -v '^[^:]*:[^:]*:(sv|debian|parabola|fsf|fedora)?:[^:]*:.*$' *.txt)"
@@ -21,3 +25,12 @@ if [[ ! -z "$unsourced" ]]; then
     printf "[citation needed]:\n\n%s\n" "$unsourced" >&2
     exit 1
 fi
+
+
+# TODO: this could be removed someday - see note in 'sort-entries' script
+unsortable="$(grep ${SEP_CHAR} *.txt)"
+
+if [[ ! -z "$unsortable" ]]; then
+    printf "[non-uniform sorting] entry contains '$s' char" ${SEP_CHAR} >&2
+    exit 1
+fi
diff --git a/sort-entries b/sort-entries
new file mode 100755
index 0000000..f688029
--- /dev/null
+++ b/sort-entries
@@ -0,0 +1,22 @@
+#!/bin/bash
+
+# TODO: the best sorting results are acheived when the field separator
+#       precedes any valid package name character in ASCII order -
+#       the lowest of which is ASCII 43 '+'; so
+#       ASCII 33 ('!') serves this purpose quite well -
+#       someday, we should re-write the tools to use '!' instead of ':' -
+#       then the sort command alone would yeild the same results as this script
+#       and the warning could be removed from 'check.sh'
+
+readonly CSV_CHAR=':'
+readonly SEP_CHAR='!'
+
+
+for blacklist in *.txt
+do  if   grep ${SEP_CHAR} ${blacklist}
+    then echo "can not sort: '${blacklist}' - contains '${SEP_CHAR}' char"
+    else echo "sorting: '${blacklist}'"
+         cat ${blacklist} | tr "${CSV_CHAR}" "${SEP_CHAR}" | sort | uniq | tr "${SEP_CHAR}" "${CSV_CHAR}" > ${blacklist}.temp
+         mv ${blacklist}.temp ${blacklist}
+    fi
+done
-- 
cgit v1.2.2