limit word lengths to 2-15

dict2dawg crashes when given a 1-letter word. Easier to fix in the
filtering that has to be there anyway.

xwords4/dawg/Catalan/info.txt

@@ -47,7 +47,7 @@ LANGFILTER: tr -d '\r'
 LANGFILTER: | tr 'a-zç' 'A-ZÇ'
 LANGFILTER: | sed -e 's/L·L/1/g' -e 's/L\.L/1/g' -e 's/L-L/1/g'
 LANGFILTER: | sed -e 's/NY/2/g' -e 's/QU/3/g'
-LANGFILTER: | grep '^[Ç1-3A-JL-VXZ\.]*$'
+LANGFILTER: | grep -x '[Ç1-3A-JL-VXZ\.]\{2,15\}'
 
 # substitute in the octal control character values 
 LANGFILTER: | tr '123' '\001\002\003'

xwords4/dawg/Danish/info.txt

@@ -24,7 +24,7 @@ LANGFILTER: | tr [a-zåæø] [A-ZÅÆØ]
 # no words not containing a vowel
 LANGFILTER: | grep '[AEIOUYÅÆØ]' 
 # none with illegal chars
-LANGFILTER: | grep '^[A-PR-VX-ZÅÆØ]\+$'
+LANGFILTER: | grep -x '[A-PR-VX-ZÅÆØ]\{2,15\}'
 # remove duplicates
 LANGFILTER: | sort -u
 

xwords4/dawg/Dutch/info.txt

@@ -23,7 +23,7 @@ LANGFILTER: tr -d '\r'
 # uppercase all
 LANGFILTER: | tr [a-z] [A-Z]
 # none with illegal chars
-LANGFILTER: | grep '^[A-Z]\+$'
+LANGFILTER: | grep -x '[A-Z]\{2,15\}'
 LANGFILTER: | sort -u
 
 # Until I can figure out how to force sort to use a locale's collation

xwords4/dawg/English/info.txt

@@ -20,7 +20,7 @@ LANGCODE:en_US
 # deal with DOS files
 LANGFILTER: tr -d '\r'
 LANGFILTER: | tr [a-z] [A-Z]
-LANGFILTER: | grep '^[A-Z]*$'
+LANGFILTER: | grep -x '[A-Z]\{2,15\}'
 LANGFILTER: | sort -u
 
 # We can trust sort (above) to do the right thing since there's no

xwords4/dawg/French/info.txt

@@ -20,7 +20,7 @@ LANGCODE:fr_FR
 LANGFILTER: tr -d '\r'
 
 LANGFILTER: | tr [a-z] [A-Z]
-LANGFILTER: | grep '^[A-Z]*$'
+LANGFILTER: | grep -x '[A-Z]\{2,15\}'
 LANGFILTER: | tr '\n' '\000'
 LANGFILTER: | sort -u -z
 

xwords4/dawg/German/info.txt

@@ -28,7 +28,7 @@ LANGFILTER: | tr [a-zäöü] [A-ZÄÖÜ]
 # no words not containing a vowel
 # LANGFILTER: | grep '[AEIOUÄÖÜ]' 
 # none with illegal chars
-LANGFILTER: | grep '^[A-ZÄÖÜ]\+$'
+LANGFILTER: | grep -x '[A-ZÄÖÜ]\{2,15\}'
 
 # Until I can figure out how to force sort to use a locale's collation
 # rules we can't trust sort in the filtering rules above and so must

xwords4/dawg/Greek/info.txt

@@ -21,7 +21,7 @@ CHARSET: utf-8
 
 LANGFILTER: tr -d '\r'
 LANGFILTER: | tr 'αβγδεζηθικλμνξοπρστυφχψω' 'ΑΒΓΔΕΖΗΘΙΚΛΜΝΞΟΠΡΣΤΥΦΧΨΩ'
-LANGFILTER: | grep '^[ΑΒΓΔΕΖΗΘΙΚΛΜΝΞΟΠΡΣΤΥΦΧΨΩ]*$'
+LANGFILTER: | grep -x '[ΑΒΓΔΕΖΗΘΙΚΛΜΝΞΟΠΡΣΤΥΦΧΨΩ]\{2,15\}'
 
 LANGINFO: <p>Greek.  Uploaded wordlist must be in utf-8 format.
 LANGINFO: </p>

xwords4/dawg/Hex/info.txt

@@ -20,7 +20,7 @@ LANGCODE:HEX
 
 # uppercase all
 LANGFILTER: tr [a-f] [A-F]
-LANGFILTER: | grep '^[A-F]*$'
+LANGFILTER: | grep -x '[A-F]\{2,15\}'
 LANGFILTER: | sed 's/A/Ä/'
 LANGFILTER: | sed 's/E/Ë/'
 LANGFILTER: | sort -u

xwords4/dawg/Hungarian/info.txt

@@ -14,7 +14,7 @@ LANGFILTER: | sed -e 's,NY,4,g'
 LANGFILTER: | sed -e 's,^\(.*\)SZ\(.*\)$,\15\2\n\1SZ\2,g'
 LANGFILTER: | sed -e 's,TY,6,g'
 LANGFILTER: | sed -e 's,^\(.*\)ZS\(.*\)$,\17\2\n\1ZS\2,g'
-LANGFILTER: | grep '^[1-7AÁBCDEÉFGHIÍJKLMNOÓÖŐPRSTUÚÜŰVZ]\{2,15\}$'
+LANGFILTER: | grep -x '[1-7AÁBCDEÉFGHIÍJKLMNOÓÖŐPRSTUÚÜŰVZ]\{2,15\}'
 LANGFILTER: | tr '1234567' '\001\002\003\004\005\006\007'
 
 # High bit means "official".  Next 7 bits are an enum where

xwords4/dawg/Hëx/info.txt

@@ -20,7 +20,7 @@ LANGCODE:HEX
 
 # uppercase all
 LANGFILTER: tr [a-f] [A-F]
-LANGFILTER: | grep '^[A-F]*$'
+LANGFILTER: | grep -x '[A-F]\{2,15\}'
 LANGFILTER: | sed 's/A/Ä/'
 LANGFILTER: | sed 's/E/Ë/'
 LANGFILTER: | sort -u

xwords4/dawg/Italian/info.txt

@@ -20,7 +20,7 @@ LANGCODE:it_IT
 # deal with DOS files
 LANGFILTER: tr -d '\r'
 LANGFILTER: | tr [a-z] [A-Z]
-LANGFILTER: | grep '^[A-IL-VZ]*$'
+LANGFILTER: | grep -x '[A-IL-VZ]\{2,15\}'
 LANGFILTER: | sort -u
 
 D2DARGS: -r -term 10 -nosort

xwords4/dawg/Polish/info.txt

@@ -22,7 +22,7 @@ CHARSET:utf-8
 # deal with DOS files
 LANGFILTER: tr -d '\r'
 LANGFILTER: | tr [a-pr-uwyząćęłńóśźż] [A-PR-UWYZĄĆĘŁŃÓŚŹŻ]
-LANGFILTER: | grep '^[A-PR-UWYZĄĆĘŁŃÓŚŹŻ]*$'
+LANGFILTER: | grep -x '[A-PR-UWYZĄĆĘŁŃÓŚŹŻ]\{2,15\}'
 LANGFILTER: | tr '\n' '\000'
 
 D2DARGS: -r -term 0

xwords4/dawg/Portuguese/info.txt

@@ -24,7 +24,7 @@ LANGFILTER: | tr [a-zç] [A-ZÇ]
 # no words not containing a vowel
 LANGFILTER: | grep '[AEIOU]' 
 # none with illegal chars
-LANGFILTER: | grep '^[A-JL-VXZÇ]\+$'
+LANGFILTER: | grep -x '[A-JL-VXZÇ]\{2,15\}'
 
 # Until I can figure out how to force sort to use a locale's collation
 # rules we can't trust sort in the filtering rules above and so must

xwords4/dawg/Romanian/info.txt

@@ -23,7 +23,7 @@ LANGCODE:ro
 LANGFILTER: tr -d '\r'
 LANGFILTER: | tr [:lower:] [:upper:]
 LANGFILTER: | tr 'ĂÂÎȘȚ' 'AAIST'
-LANGFILTER: | grep '^[A-JL-PR-VXZ]*$'
+LANGFILTER: | grep -x '[A-JL-PR-VXZ]\{2,15\}'
 LANGFILTER: | sort -u
 
 # We can trust sort (above) to do the right thing since there's no

xwords4/dawg/Spanish/info.txt

@@ -30,7 +30,7 @@ LANGFILTER: | tr '\207\216\222\227\234\237\226' 'aeiouu\321'
 # uppercase
 LANGFILTER: | tr [a-zñ] [A-ZÑ]
 # remove words with illegal letters
-LANGFILTER: | grep '^[[A-JL-VX-ZÑ]*$'
+LANGFILTER: | grep -x '[[A-JL-VX-ZÑ]\{2,15\}'
 # substitute pairs (can't figure out how to use octal values)
 LANGFILTER: | sed 's/CH/1/g'
 LANGFILTER: | sed 's/LL/2/g'

xwords4/dawg/Swedish/info.txt

@@ -20,7 +20,7 @@ LANGCODE:sv_SE
 
 LANGFILTER: tr -d '\r'
 LANGFILTER: | tr [a-zäåæöü] [A-ZÄÅÆÖÜ]
-LANGFILTER: | grep '^[A-ZÄÅÆÖÜ]*$'
+LANGFILTER: | grep -x '[A-ZÄÅÆÖÜ]\{2,15\}'
 
 D2DARGS: -r -term 10