dd: update from Plan 9

R=rsc
CC=plan9port.codebot
http://codereview.appspot.com/4850052

man/man1/dd.1

@@ -0,0 +1,198 @@
+.TH DD 1
+.SH NAME
+dd \- convert and copy a file
+.SH SYNOPSIS
+.B dd
+[
+.I option value
+]
+\&...
+.SH DESCRIPTION
+.I Dd\^
+copies the specified input file
+to the specified output with
+possible conversions.
+The standard input and output are used by default.
+The input and output block size may be
+specified to take advantage of raw physical I/O.
+The options are
+.TF "quiet\ \ \fIn
+.PD
+.TP
+.BI -if\  f
+Open file
+.I f
+for input.
+.TP
+.BI -of\  f
+Open file
+.I f
+for output.
+.TP
+.BI -ibs\  n\^
+Set input block size to
+.I n\^
+bytes (default 512).
+.TP
+.BI -obs\  n\^
+Set output block size (default 512).
+.TP
+.BI -bs\  n\^
+Set both input and output block size,
+superseding
+.I ibs\^
+and
+.IR obs .
+If no conversion is specified,
+preserve the input block size instead of packing short blocks
+into the output buffer.
+This is particularly efficient since no in-core copy need be done.
+.TP
+.BI -cbs\  n\^
+Set conversion buffer size.
+.TP
+.BI -skip\  n\^
+Skip
+.I n
+input records before copying.
+.TP
+.BI -iseek\  n\^
+Seek
+.I n
+records forward on input file
+before copying.
+.TP
+.BI -files\  n\^
+Catenate
+.I n 
+input files (useful only for magnetic tape or similar input device).
+.TP
+.BI -oseek\  n\^
+Seek
+.I n\^
+records from beginning of output file before copying.
+.TP
+.BI -count\  n\^
+Copy only
+.I n
+input records.
+.TP
+.BI -trunc\  n\^
+By default,
+.I dd
+truncates the output file when it opens it;
+.B -trunc
+.B 0
+opens it without truncation.
+.TP
+.BI -quiet\  n\^
+By default,
+.I dd
+prints the number of blocks read and written
+once it is finished.
+.B -quiet
+.B 1
+silences this summary.
+.HP
+\fL-conv\ ascii\ \ \ \ \fRConvert
+.SM EBCDIC
+to
+.SM ASCII.
+.PD0
+.RS "\w'\fLconv\ \fP'u"
+.TP "\w'\fLunblock\ \ \fP'u"
+.B ebcdic
+Convert
+.SM ASCII
+to
+.SM EBCDIC.
+.TP
+.B ibm
+Like
+.B ebcdic
+but with a slightly different character map.
+.TP
+.B block
+Convert variable length
+.SM ASCII
+records to fixed length.
+.TP
+.B unblock
+Convert fixed length
+.SM ASCII
+records to variable length.
+.TP
+.B lcase
+Map alphabetics to lower case.
+.TP
+.B ucase
+Map alphabetics to upper case.
+.TP
+.B swab
+Swap every pair of bytes.
+.TP
+.B noerror
+Do not stop processing on an error.
+.TP
+.B sync
+Pad every input record to
+.I  ibs\^
+bytes.
+.RE
+.PD
+.PP
+.fi
+Where sizes are specified,
+a number of bytes is expected.
+A number may end with
+.L k
+or
+.LR b
+to specify multiplication by
+1024 or 512 respectively;
+a pair of numbers may be separated by
+.L x
+to indicate a product.
+Multiple conversions may be specified in the style:
+.LR "-conv ebcdic,ucase" .
+.PP
+.L Cbs\^
+is used only if
+.LR ascii\^ ,
+.LR unblock\^ ,
+.LR ebcdic\^ ,
+.LR ibm\^ ,
+or
+.L block\^
+conversion is specified.
+In the first two cases,
+.I n
+characters are copied into the conversion buffer, any specified
+character mapping is done,
+trailing blanks are trimmed and new-line is added
+before sending the line to the output.
+In the latter three cases, characters are read into the
+conversion buffer and blanks are added to make up an
+output record of size
+.IR n .
+If
+.L cbs\^
+is unspecified or zero, the
+.LR ascii\^ ,
+.LR ebcdic\^ ,
+and
+.L ibm\^
+options convert the character set without changing the block
+structure of the input file; the
+.L unblock\^
+and
+.L block\^
+options become a simple file copy.
+.SH SOURCE
+.B \*9/src/cmd/dd.c
+.SH "SEE ALSO"
+.IR cp (1)
+.SH DIAGNOSTICS
+.I Dd
+reports the number of full + partial input and output
+blocks handled.

src/cmd/dd.c

@@ -1,23 +1,28 @@
 #include <u.h>
 #include <libc.h>
 
-#define	BIG	2147483647
+#define	BIG	((1UL<<31)-1)
+#define VBIG	((1ULL<<63)-1)
 #define	LCASE	(1<<0)
 #define	UCASE	(1<<1)
 #define	SWAB	(1<<2)
 #define NERR	(1<<3)
 #define SYNC	(1<<4)
+
 int	cflag;
 int	fflag;
+
 char	*string;
 char	*ifile;
 char	*ofile;
 char	*ibuf;
 char	*obuf;
+
 vlong	skip;
 vlong	oseekn;
 vlong	iseekn;
 vlong	count;
+
 long	files	= 1;
 long	ibs	= 512;
 long	obs	= 512;
@@ -31,18 +36,23 @@ long	nipr;
 long	nofr;
 long	nopr;
 long	ntrunc;
+
 int dotrunc = 1;
 int	ibf;
 int	obf;
+
 char	*op;
 int	nspace;
+
 uchar	etoa[256];
 uchar	atoe[256];
 uchar	atoibm[256];
 
+int	quiet;
+
 void	flsh(void);
 int	match(char *s);
-vlong	number(long big);
+vlong	number(vlong big);
 void	cnull(int cc);
 void	null(int c);
 void	ascii(int cc);
@@ -50,12 +60,12 @@ void	unblock(int cc);
 void	ebcdic(int cc);
 void	ibm(int cc);
 void	block(int cc);
-void	term(void);
+void	term(char*);
 void	stats(void);
 
 #define	iskey(s)	((key[0] == '-') && (strcmp(key+1, s) == 0))
 
-void
+int
 main(int argc, char *argv[])
 {
 	void (*conv)(int);
@@ -99,20 +109,24 @@ main(int argc, char *argv[])
 			dotrunc = number(BIG);
 			continue;
 		}
+		if(iskey("quiet")) {
+			quiet = number(BIG);
+			continue;
+		}
 		if(iskey("skip")) {
-			skip = number(BIG);
+			skip = number(VBIG);
 			continue;
 		}
 		if(iskey("seek") || iskey("oseek")) {
-			oseekn = number(BIG);
+			oseekn = number(VBIG);
 			continue;
 		}
 		if(iskey("iseek")) {
-			iseekn = number(BIG);
+			iseekn = number(VBIG);
 			continue;
 		}
 		if(iskey("count")) {
-			count = number(BIG);
+			count = number(VBIG);
 			continue;
 		}
 		if(iskey("files")) {
@@ -165,6 +179,8 @@ main(int argc, char *argv[])
 				cflag |= SYNC;
 				goto cloop;
 			}
+			fprint(2, "dd: bad conv %s\n", argv[c]);
+			exits("arg");
 		}
 		fprint(2, "dd: bad arg: %s\n", key);
 		exits("arg");
@@ -243,17 +259,17 @@ loop:
 			perror("read");
 			if((cflag&NERR) == 0) {
 				flsh();
-				term();
+				term("errors");
 			}
 			ibc = 0;
 			for(c=0; c<ibs; c++)
 				if(ibuf[c] != 0)
-					ibc = c;
+					ibc = c+1;
+			seek(ibf, ibs, 1);
 			stats();
-		}
+		}else if(ibc == 0 && --files<=0) {
-		if(ibc == 0 && --files<=0) {
 			flsh();
-			term();
+			term(nil);
 		}
 		if(ibc != ibs) {
 			nipr++;
@@ -290,12 +306,14 @@ flsh(void)
 	int c;
 
 	if(obc) {
+		/* don't perror dregs of previous errors on a short write */
+		werrstr("");
 		c = write(obf, obuf, obc);
 		if(c != obc) {
 			if(c > 0)
 				++nopr;
 			perror("write");
-			term();
+			term("errors");
 		}
 		if(obc == obs)
 			nofr++;
@@ -324,10 +342,10 @@ true:
 }
 
 vlong
-number(long big)
+number(vlong big)
 {
 	char *cs;
-	vlong n;
+	uvlong n;
 
 	cs = string;
 	n = 0;
@@ -340,11 +358,6 @@ number(long big)
 		n *= 1024;
 		continue;
 
-/*	case 'w':
-		n *= sizeof(int);
-		continue;
-*/
-
 	case 'b':
 		n *= 512;
 		continue;
@@ -352,11 +365,11 @@ number(long big)
 /*	case '*':*/
 	case 'x':
 		string = cs;
-		n *= number(BIG);
+		n *= number(VBIG);
 
 	case '\0':
-		if(n>=big || n<0) {
+		if(n > big) {
-			fprint(2, "dd: argument %lld out of range\n", n);
+			fprint(2, "dd: argument %llud out of range\n", n);
 			exits("range");
 		}
 		return n;
@@ -536,17 +549,17 @@ block(int cc)
 }
 
 void
-term(void)
+term(char *status)
 {
-
 	stats();
-	exits(0);
+	exits(status);
 }
 
 void
 stats(void)
 {
-
+	if(quiet)
+		return;
 	fprint(2, "%lud+%lud records in\n", nifr, nipr);
 	fprint(2, "%lud+%lud records out\n", nofr, nopr);
 	if(ntrunc)