calc_blklen violates specification section "9.5 Segmented Streaming"

[Parakleta]

The relevant section reads:

9.5 Segmented Streaming

If the receiver cannot overlap serial and disk I/O, it uses the
ZRINIT frame to specify a buffer length which the sender will not
overflow. The sending program sends a ZCRCW data subpacket and waits
for a ZACK header before sending the next segment of the file.

If the sending program supports reverse data stream sampling or
interrupt, error recovery will be faster (on average) than a protocol
(such as YMODEM) that sends large blocks.

A sufficiently large receiving buffer allows throughput to closely
approach that of full streaming. For example, 16kb segmented
streaming adds about 3 per cent to full streaming ZMODEM file
transfer times when the round trip delay is five seconds.

If a maximum buffer length is provided in the ZRINIT frame this is read by lsz and stored in the Rxbuflen and blklen variables, but then later when calc_blklen is called the blklen value is overridden (set to 1024) and the Rxbuflen value is never again consulted.

This is problematic for my use case because I trying to use ZMODEM to program small embedded devices which have limited RAM and a specific flash block size which must be programmed. I am trying to set the buffer length to one flash page so that I can read it in, checksum it, and then write it to flash before accepting the next subpacket.

I think the solution to this is to set the start_blklen and max_blklen variables as well when Rxbuflen is set, if it is smaller than Tframelen, exactly the same way that these are set when it is.

[Parakleta]

As an addition to this, the following section of code makes it impossible to test this behaviour using pipes to a simulated receiver as it overrides the Rxbuflen value:

lrzsz/src/lsz.c

Lines 1260 to 1263 in 8cb2a6a

	fstat(0, &f);
	if (! (S_ISCHR(f.st_mode))) {
	Rxbuflen = MAX_BLOCK;
	}

[Parakleta]

Further, the commit fbdffdb breaks the required behaviour further by preventing a ZCRCW request from being made after each Rxbuflen sized block. The comment stating that the newcnt variable was never updated was incorrect. The following code

lrzsz/src/lsz.c

Lines 1593 to 1594 in 8cb2a6a

	if (e == ZCRCW)
	goto waitack;

jumps (I know, the goto's are atrocious) to above where newcnt was set, thus updating it again to allow another full block.