September 2017 Update

Task/Text-processing-2/C/text-processing-2-1.c

@@ -1,84 +0,0 @@
-#include <stdio.h>
-#include <string.h>
-#include <stdlib.h>
-#include <unistd.h>
-#include <sys/types.h>
-#include <sys/stat.h>
-#include <fcntl.h>
-
-typedef struct { const char *s; int ln, bad; } rec_t;
-int cmp_rec(const void *aa, const void *bb)
-{
-	const rec_t *a = aa, *b = bb;
-	return a->s == b->s ? 0 : !a->s ? 1 : !b->s ? -1 : strncmp(a->s, b->s, 10);
-}
-
-int read_file(const char *fn)
-{
-	int fd = open(fn, O_RDONLY);
-	if (fd == -1) return 0;
-
-	struct stat s;
-	fstat(fd, &s);
-
-	char *txt = malloc(s.st_size);
-	read(fd, txt, s.st_size);
-	close(fd);
-
-	int i, j, lines = 0, k, di, bad;
-	for (i = lines = 0; i < s.st_size; i++)
-		if (txt[i] == '\n') {
-			txt[i] = '\0';
-			lines++;
-		}
-
-	rec_t *rec = calloc(sizeof(rec_t), lines);
-	const char *ptr, *end;
-	rec[0].s = txt;
-	rec[0].ln = 1;
-	for (i = 0; i < lines; i++) {
-		if (i + 1 < lines) {
-			rec[i + 1].s = rec[i].s + strlen(rec[i].s) + 1;
-			rec[i + 1].ln = i + 2;
-		}
-		if (sscanf(rec[i].s, "%4d-%2d-%2d", &di, &di, &di) != 3) {
-			printf("bad line %d: %s\n", i, rec[i].s);
-			rec[i].s = 0;
-			continue;
-		}
-		ptr = rec[i].s + 10;
-
-		for (j = k = 0; j < 25; j++) {
-			if (!strtod(ptr, (char**)&end) && end == ptr) break;
-			k++, ptr = end;
-			if (!(di = strtol(ptr, (char**)&end, 10)) && end == ptr) break;
-			k++, ptr = end;
-			if (di < 1) rec[i].bad = 1;
-		}
-
-		if (k != 48) {
-			printf("bad format at line %d: %s\n", i, rec[i].s);
-			rec[i].s = 0;
-		}
-	}
-
-	qsort(rec, lines, sizeof(rec_t), cmp_rec);
-	for (i = 1, bad = rec[0].bad, j = 0; i < lines && rec[i].s; i++) {
-		if (rec[i].bad) bad++;
-		if (strncmp(rec[i].s, rec[j].s, 10)) {
-			j = i;
-		} else
-			printf("dup line %d: %.10s\n", rec[i].ln, rec[i].s);
-	}
-
-	free(rec);
-	free(txt);
-	printf("\n%d out %d lines good\n", lines - bad, lines);
-	return 0;
-}
-
-int main()
-{
-	read_file("readings.txt");
-	return 0;
-}

Task/Text-processing-2/C/text-processing-2-2.c

@@ -1,7 +0,0 @@
-dup line 85: 1990-03-25
-dup line 456: 1991-03-31
-dup line 820: 1992-03-29
-dup line 1184: 1993-03-28
-dup line 1911: 1995-03-26
-
-5017 out 5471 lines good

Task/Text-processing-2/Zkl/text-processing-2.zkl

@@ -0,0 +1,19 @@
+   // the RegExp engine has a low limit on groups so
+   // I can't use it to select all fields, only verify them
+re:=RegExp(0'|^(\d+-\d+-\d+)| + 0'|\s+\d+\.\d+\s+-*\d+| * 24 + ".+$");
+w:=[1..].zip(File("readings.txt"));  //-->lazy (line #,line)
+reg datep,N, good=0, dd=0;
+foreach n,line in (w){
+   N=n;		// since n is local to this scope
+   if (not re.search(line)){ println("Line %d: malformed".fmt(n)); continue; }
+   date:=line[re.matchedNs[1].xplode()];  // I can group the date field
+   if (datep==date){ dd+=1; println("Line %4d: dup date: %s".fmt(n,date)); }
+   datep=date;
+   if (line.replace("\t"," ").split(" ").filter()[1,*]  // blow fields apart, drop date
+         .pump(Void,Void.Read, // get (reading,status)
+            fcn(_,s){  // stop on first problem status and return True
+               if(s.strip().toInt()<1) T(Void.Stop,True) else False
+       })) continue;
+   good+=1;
+}
+println("%d records read, %d duplicate dates, %d valid".fmt(N,dd,good));