Added the missing build files for last

bin/last/build/gc-inc.sh

@@ -0,0 +1,20 @@
+#! /bin/sh
+
+# This generates source code from genetic codes.
+
+cat <<EOF
+const struct {
+  const char *name;
+  const char *text;
+} geneticCodes[] = {
+EOF
+
+cat "$@" | tr -d '",' |
+awk '
+$1 == "id" {print "{\"" $2 "\", \"\\"}
+$1 == "ncbieaa" {print "  AAs = " $2 "\\n\\"}
+/-- Base/ {print $2 " = " $3 "\\n\\"}
+/-- Base3/ {print "\"},"}
+'
+
+echo "};"

bin/last/build/mat-doc.sh

@@ -0,0 +1,29 @@
+#! /bin/sh
+
+# This generates text documentation from substitution score matrices.
+
+cat <<EOF
+LAST built-in scoring schemes
+=============================
+
+EOF
+
+for i in "$@"
+do
+    name=$(basename $i .mat)
+    nick=$(grep '^#nickname' $i | cut -d' ' -f2)
+    test "$nick" && name="$name or $nick"
+    echo $name
+    echo $name | sed 's/./-/g'  # underline
+    echo
+    grep '^# ' $i | cut -d' ' -f2-
+    echo It uses this matrix::
+    echo
+    grep -v '^#' $i | awk NF | sed 's/^/  /'
+    echo
+    grep -q '^#last' $i && {
+	echo It sets these default lastal parameter values:
+	grep '^#last' $i | cut -d' ' -f2-
+	echo
+    }
+done

bin/last/build/mat-inc.sh

@@ -0,0 +1,33 @@
+#! /bin/sh
+
+# This generates source code from substitution score matrices.
+
+cat <<EOF
+const struct {
+  const char *nickname;
+  const char *realname;
+} scoreMatrixNicknames [] = {
+EOF
+for i in "$@"
+do
+    name=$(basename $i .mat)
+    grep '^#nickname' $i | cut -d' ' -f2 | sed 's/.*/{"&", "'$name'"},/'
+done
+echo "};"
+
+echo
+
+cat <<EOF
+const struct {
+  const char *name;
+  const char *text;
+} scoreMatrices[] = {
+EOF
+for i in "$@"
+do
+    basename $i .mat | sed 's/.*/{"&", "\\/'
+    grep -v '^#[n ]' $i | awk NF | sed 's/$/\\n\\/'
+    echo '"},'
+    echo
+done
+echo "};"

bin/last/build/seed-doc.sh

@@ -0,0 +1,80 @@
+#! /bin/sh
+
+# This generates text documentation from subset seeds.
+
+cat <<'EOF'
+LAST seeding schemes
+====================
+
+LAST's critical first step is to find *seeds*, i.e. initial matches
+between query and reference sequences.  It can use various seeding
+schemes, which allow different kinds of mismatches at different seed
+positions.
+
+   A seeding scheme consists of a seed alphabet, such as::
+
+     1  A C G T
+     0  ACGT
+     T  AG CT
+
+   and one or more patterns, such as this one::
+
+     1T1T10T1101101
+
+   Each symbol in a pattern represents a grouping of sequence letters:
+   in this example, ``T`` represents the grouping ``AG CT``.  At each
+   position in an initial match, mismatches are allowed between
+   letters that are grouped at that position in the pattern.
+
+   Although the patterns have fixed lengths, LAST's initial matches do
+   not.  LAST finds shorter matches by using a prefix of the pattern,
+   and longer matches by cyclically repeating the pattern.
+
+   A *restricted* symbol omits letters of the main sequence alphabet,
+   which are then forbidden at those positions::
+
+     r  AG
+
+   An *exact* symbol groups no letters::
+
+     Y  C T
+
+   In 2nd and subsequent cycles, restricted symbols are made
+   unrestricted: if it is exact then the omitted letters are added as
+   separate groups, else they are added as one group.
+
+EOF
+
+for i in "$@"
+do
+    name=$(basename $i .seed)
+    abbr=$(grep '^#abbreviation' $i | cut -d' ' -f2)
+    test "$abbr" && name="$name (abbreviation: $abbr)"
+    echo $name
+    echo $name | sed 's/./-/g'  # underline
+    echo
+    grep '^# ' $i | cut -d' ' -f2-
+    echo It uses this seed alphabet::
+    echo
+    awk '!/^#/ && NF > 1 && length($1) == 1' $i | sed 's/^/  /'
+    echo
+    if [ $(awk '!/^#/ && length($1) > 1 || NF == 1' $i | wc -w) = 1 ]
+	then echo And this pattern::
+	else echo And these patterns::
+    fi
+    echo
+    awk '!/^#/ && length($1) > 1 || NF == 1' $i | sed 's/^/  /'
+    echo
+    grep -q '^#lastdb' $i && {
+	echo It sets this lastdb default:
+	grep '^#lastdb' $i | cut -d' ' -f2-
+	echo
+    }
+    grep -q '^#lastal' $i && {
+	echo It sets this lastal default:
+	grep '^#lastal' $i | cut -d' ' -f2-
+	echo
+    }
+done
+
+exit 0

bin/last/build/seed-inc.sh

@@ -0,0 +1,33 @@
+#! /bin/sh
+
+# This generates source code from subset seeds.
+
+cat <<EOF
+const struct {
+  const char *nickname;
+  const char *realname;
+} subsetSeedNicknames [] = {
+EOF
+for i in "$@"
+do
+    name=$(basename $i .seed)
+    grep '^#abbreviation' $i | cut -d' ' -f2 | sed 's/.*/{"&", "'$name'"},/'
+done
+echo "};"
+
+echo
+
+cat <<EOF
+const struct {
+  const char *name;
+  const char *text;
+} subsetSeeds[] = {
+EOF
+for i in "$@"
+do
+    basename $i .seed | sed 's/.*/{"&", "\\/'
+    grep -v '^#[a ]' $i | awk NF | sed 's/$/\\n\\/'
+    echo '"},'
+    echo
+done
+echo "};"