1 files changed, 10 insertions, 4 deletions
diff --git a/de-sgm b/de-sgm
index 664c18c..fd4546e 100755
--- a/de-sgm
+++ b/de-sgm
@@ -1,7 +1,13 @@
 #!/bin/sh
 
-
-egrep -v "^[[:space:]]*(<\?xml.*\?>|</?(mteval|doc|srcset|refset|translator|reviewer)[^>]*>)[[:space:]]*$" \
-  | egrep -v "^[[:space:]]*<(url|description|keywords|talkid|title|translator|reviewer)[^>]*>.*</(url|description|keywords|talkid|title|translator|reviewer)>[[:space:]]*$" \
-  | sed "s|<seg[^>]*>\s*||" | sed "s|\s*</seg>$||"
+egrep -v -i "^[[:space:]]*(<\?xml.*\?>|</?(mteval|doc|srcset|refset|translator|reviewer)[^>]*>)[[:space:]]*$" \
+  | egrep -v -i "^[[:space:]]*<(url|description|keywords|talkid|title|translator|reviewer)[^>]*>.*</(url|description|keywords|talkid|title|translator|reviewer)>[[:space:]]*$" \
+  | sed "s|<seg[^>]*>\s*||" \
+  | sed "s|\s*</seg>\s*$||" \
+  | egrep -v -i "^[[:space:]]*<p>[[:space:]]*$|^[[:space:]]*</p>[[:space:]]*$" \
+  | sed "s|<speaker>\s*||" \
+  | sed "s|\s*</speaker>\s*$||" \
+  | sed "s|\s*<hl>\s*$||" \
+  | sed "s|\s*</hl>\s*$||" \
+  | grep -v -P "^\s*$"