added improvements from diff to diff1; increased number of digits printed in diff, diff1, lin_reg, lin_reg1, and lin_reg2;

5 files changed, 75 insertions, 57 deletions

diff --git a/diff.awk b/diff.awk
index fb93b57..03bc398 100644
--- a/diff.awk
+++ b/diff.awk
@@ -13,13 +13,18 @@ BEGIN {
 }
 
 NR == 1 {
-    ### orig columns
+    ### orig data columns
     for (n=1; n<=NF; n++) {
-        ($n !~ number) ? header[n] = $n : header[n] = "col" n
+        if ($n !~ number) {
+            header[n] = $n
+            dheader[n] = "d" $n
+        }
+        else {
+            header[n] = "col" n
+            dheader[n] = "dcol" n
+        }
         printf(header[n])
-        ($n !~ number) ? dheader[n] = "d" $n : dheader[n] = "dcol" n
-        if (n <= NF)
-            printf(OFS)
+        printf(OFS)
     }
 
     ### diff columns
@@ -34,12 +39,12 @@ NF {
     if (NF > nf_max)
         nf_max = NF
 
-    ### orig columns
+    ### data columns
     for (y=1; y<=nf_max; y++) {
         if ($y == header[y] || $y == dheader[y])
             continue
         if ($y ~ number)
-            printf($y)
+            printf("%.12g", $y)
         else
             printf("")
         printf(OFS)
@@ -47,15 +52,18 @@ NF {
 
     ### diff columns
     for (y=1; y<=nf_max; y++) {
+        if ($y == header[y] || $y == dheader[y])
+            continue
         if ($y ~ number) {
             data[y] = $y
             (data_prev[y] ~ number) ? diff[y] = data[y] - data_prev[y] : diff[y] = ""
             data_prev[y] = data[y]
-            printf(diff[y])
-            if (y < nf_max)
-                printf(OFS)
+            printf("%.12g", diff[y])
         }
-        diff[y] = ""
+        else
+            diff[y] = ""
+        if (y < nf_max)
+            printf(OFS)
     }
-    printf("\n")
+    printf(ORS)
 }
diff --git a/diff1.awk b/diff1.awk
index 42cc5a8..e2c4a99 100644
--- a/diff1.awk
+++ b/diff1.awk
@@ -1,4 +1,4 @@
-#!/usr/bin/awk -f
+#!usr/bin/awk -f
 
 ### diff.awk
 # numerical diff along columns
@@ -13,30 +13,40 @@ BEGIN {
 }
 
 NR == 1 {
-    header_nf = NF
+    ### orig data columns
     for (n=1; n<=NF; n++) {
-        if ($n !~ number)
-            header[n] = "d" $n
-        else
-            header[n] = "col_" n
-        printf("%s,", header[n])
+        if ($n !~ number) {
+            header[n] = $n
+            dheader[n] = "d" $n
+        }
+        else {
+            header[n] = "col" n
+            dheader[n] = "dcol" n
+        }
+        printf(dheader[n])
+        if (n < NF)
+            printf(OFS)
     }
 }
 
-NF != 0 {
-    if (NF > max_nf)
-        max_nf = NF
+NF {
+    if (NF > nf_max)
+        nf_max = NF
 
-    ### iterate over columns
-    for (y=1; y<=max_nf; y++) {
+    ### data columns
+    for (y=1; y<=nf_max; y++) {
+        if ($y == header[y] || $y == dheader[y])
+            continue
         if ($y ~ number) {
-
             data[y] = $y
-            diff[y] = data[y] - data_prev[y]
-            printf("%g,", diff[y])
+            (data_prev[y] ~ number) ? diff[y] = data[y] - data_prev[y] : diff[y] = ""
             data_prev[y] = data[y]
-
+            printf("%.12g", diff[y])
         }
+        else
+            diff[y] = ""
+        if (y < nf_max)
+            printf(OFS)
     }
-    printf("\n")
+    printf(ORS)
 }
diff --git a/lin_reg.awk b/lin_reg.awk
index 51f2e87..f82e2e8 100644
--- a/lin_reg.awk
+++ b/lin_reg.awk
@@ -4,6 +4,7 @@
 # simple linear regression between columns
 
 BEGIN {
+    OFS = "%.9g"
     sign = "[+-]?"
     decimal = "[0-9]+[.]?[0-9]*"
     fraction = "[.][0-9]*"
@@ -14,17 +15,16 @@ BEGIN {
 NR == 1 {
     header_nf = NF
     for (n=1; n<=NF; n++) {
-        ($n !~ number) ? header[n] = $n : header[n] = "col_" n
+        ($n !~ number) ? header[n] = $n : header[n] = "col" n
     }
 }
 
 NF != 0 {
-    #printf("\n%s: %s", NR, $0)
-    if (NF > max_nf)
-        max_nf = NF
+    if (NF > nf_max)
+        nf_max = NF
 
     ### iterate over columns
-    for (y=1; y<=max_nf; y++) {
+    for (y=1; y<=nf_max; y++) {
         if ($y ~ number) {
 
             ### mean
@@ -42,7 +42,7 @@ NF != 0 {
             (count[y] > 1) ? var[y] = sum_delta2[y]/(count[y] - 1) : var[y] = 0
 
             # x = row, y = col
-            for (x=1; x<=max_nf; x++) {
+            for (x=1; x<=nf_max; x++) {
                 count[x,y] += 1
                 sum_xy[x,y] += $x*$y
                 sum_delta_xy[x,y] += delta[x]*delta[y]
@@ -81,10 +81,10 @@ NF != 0 {
 }
 
 END {
-    for (y=1; y<=max_nf; y++) {
-        for (x=1; x<=max_nf; x++) {
+    for (y=1; y<=nf_max; y++) {
+        for (x=1; x<=nf_max; x++) {
             if (x != y && r[x,y]) {
-                printf("\n  %g \t (%s) \t = (%g +/- %g)(%s) \t + (%g +/- %g)",
+                printf("\n  %.9g \t (%s) \t = (%.9g +/- %.9g)(%s) \t + (%.9g +/- %.9g)",
                     10.0*log(r[x,y]*r[x,y])/log(10), header[y], b[x,y], b_err[x,y], header[x],
                     a[x,y], a_err[x,y])
             }
diff --git a/lin_reg1.awk b/lin_reg1.awk
index cd3db8e..2d21c29 100644..100755
--- a/lin_reg1.awk
+++ b/lin_reg1.awk
@@ -1,7 +1,7 @@
 #!/usr/bin/awk -f
 
 ### lin_reg1.awk
-# simple linear regression between individual text columns
+# simple linear regression between columns
 
 BEGIN {
     OFS = "%.9g"
@@ -15,16 +15,16 @@ BEGIN {
 NR == 1 {
     header_nf = NF
     for (n=1; n<=NF; n++) {
-        ($n !~ number) ? header[n] = $n : header[n] = sprintf("col_%g", n)
+        ($n !~ number) ? header[n] = $n : header[n] = "col" n
     }
 }
 
 NF != 0 {
-    if (NF > max_nf)
-        max_nf = NF
+    if (NF > nf_max)
+        nf_max = NF
 
     ### iterate over columns
-    for (y=1; y<= NF; y++) {
+    for (y=1; y<=nf_max; y++) {
         if ($y ~ number) {
 
             ### mean
@@ -39,7 +39,7 @@ NF != 0 {
             sum_delta2[y] += delta[y]*delta[y]
 
             # x = row, y = col
-            for (x=1; x<=max_nf; x++) {
+            for (x=1; x<=nf_max; x++) {
                 count[x,y] += 1
                 sum_xy[x,y] += $x*$y
                 sum_delta_xy[x,y] += delta[x]*delta[y]
@@ -64,11 +64,11 @@ NF != 0 {
 }
 
 END {
-    for (y=1; y<=max_nf; y++) {
-        for (x=1; x<=max_nf; x++) {
+    for (y=1; y<=nf_max; y++) {
+        for (x=1; x<=nf_max; x++) {
             if (x != y && r[x,y]) {
                 r2[x,y] = r[x,y]*r[x,y]
-                printf("\n  %f \t (%s) \t = %g(%s) \t + %g",
+                printf("\n  %.9g \t (%s) \t = %.9g(%s) \t + %.9g",
                     10.0*log(r2[x,y])/log(10), header[y], b[x,y], header[x], a[x,y])
             }
         }
diff --git a/lin_reg2.awk b/lin_reg2.awk
index 7553c4e..916b6cb 100644
--- a/lin_reg2.awk
+++ b/lin_reg2.awk
@@ -1,9 +1,10 @@
 #!/usr/bin/awk -f
 
-### lin_reg.awk
+### lin_reg2.awk
 # simple linear regression between columns
 
 BEGIN {
+    OFS = "%.9g"
     sign = "[+-]?"
     decimal = "[0-9]+[.]?[0-9]*"
     fraction = "[.][0-9]*"
@@ -14,17 +15,16 @@ BEGIN {
 NR == 1 {
     header_nf = NF
     for (n=1; n<=NF; n++) {
-        ($n !~ number) ? header[n] = $n : header[n] = n
+        ($n !~ number) ? header[n] = $n : header[n] = "col" n
     }
 }
 
 NF != 0 {
-    #printf("\n%s: %s", NR, $0)
-    if (NF > max_nf)
-        max_nf = NF
+    if (NF > nf_max)
+        nf_max = NF
 
     ### iterate over columns
-    for (y=1; y<=max_nf; y++) {
+    for (y=1; y<=nf_max; y++) {
         if ($y ~ number) {
 
             ### mean
@@ -42,7 +42,7 @@ NF != 0 {
             (count[y] - 1) ? var[y] = sum_delta2[y]/(count[y] - 1) : var[y] = 0
 
             # x = row, y = col
-            for (x=1; x<=max_nf; x++) {
+            for (x=1; x<=nf_max; x++) {
                 count[x,y] += 1
                 sum_xy[x,y] += $x*$y
                 sum_delta_xy[x,y] += delta[x]*delta[y]
@@ -81,10 +81,10 @@ NF != 0 {
 }
 
 END {
-    for (y=1; y<=max_nf; y++) {
-        for (x=1; x<=max_nf; x++) {
+    for (y=1; y<=nf_max; y++) {
+        for (x=1; x<=nf_max; x++) {
             if (x != y && r[x,y]) {
-                printf("\n  %g \t (%s) \t = (%g +/- %g)(%s) \t + (%g +/- %g)",
+                printf("\n  %.9g \t (%s) \t = (%.9g +/- %.9g)(%s) \t + (%.9g +/- %.9g)",
                     10.0*log(r[x,y]*r[x,y])/log(10), header[y], b[x,y], b_err[x,y], header[x],
                     a[x,y], a_err[x,y])
             }