Canary Tests

Tsaukpaetra

@Tsaukpaetra said in Canary Tests:

@PJH said in Canary Tests:

I don't see the web littered with URLs containing :80 or :433, so I'm going to call "almost always" somewhat exaggerated.

Oh, they just moved the 80 to be http: and the 443 to https:. Just because a default has been communally adopted doesn't mean the underlying technology suddenly doesn't need it.

Not quite. http://www.example.com:443 and https://www.example.com:80 are both perfectly cromulent (if somewhat obfuscatory) URLs.

The bit after the last colon tells you the port, the bit before the first colon tell you what protocol you use over that port (said protocols having defaults of 80 and 443.)

.... How does this change or otherwise refute my statement in any way?

PJH

@Tsaukpaetra said in Canary Tests:

.... How does this change or otherwise refute my statement in any way?

Your implication that the bit before the first colon overrides whatever's after the second colon is wrong?

Tsaukpaetra

@PJH said in Canary Tests:

@Tsaukpaetra said in Canary Tests:

.... How does this change or otherwise refute my statement in any way?

Your implication that the bit before the first colon overrides whatever's after the second colon is wrong?

I don't recall implying that...

PJH

@Tsaukpaetra said in Canary Tests:

I don't recall implying that...

I must have imagined it..

Tsaukpaetra

@PJH said in Canary Tests:

@Tsaukpaetra said in Canary Tests:

I don't recall implying that...

I must have imagined it..

Keep that imagination, I hear having one is rather healthy!

dkf

@Tsaukpaetra said in Canary Tests:

I don't recall implying that...

Funnily enough, I was getting the same impression as @PJH; there must be a way to systematically misinterpret your communication that we were both doing.

kazitor

@dkf simple:

you and @PJH are both @boomzilla alts
you and @PJH share a shoulder alien
QED

dkf

@kazitor said in Canary Tests:

you and @PJH are both @boomzilla alts

This is a null statement. We are all @boomzilla alts. It is axiomatic.

kazitor

@dkf and yet you were confused by you sharing an opinion with another alt

dkf

@kazitor This surprises you how? Surely you've learned that this is how it is from your fellow alts by now?

Tsaukpaetra

@dkf said in Canary Tests:

@Tsaukpaetra said in Canary Tests:

I don't recall implying that...

Funnily enough, I was getting the same impression as @PJH; there must be a way to systematically misinterpret your communication that we were both doing.

I will attempt to deconstruct my statement then:

@Tsaukpaetra said in Canary Tests:

@PJH said in Canary Tests:

I don't see the web littered with URLs containing :80 or :433, so I'm going to call "almost always" somewhat exaggerated.

Oh, they just moved the 80 to be http: and the 443 to https:. Just because a default has been communally adopted doesn't mean the underlying technology suddenly doesn't need it.

Refer: @PJH stating that a lack of URLs littering the web containing port numbers.
Statement: http: means 80 and https: means 443 by default due to communally adopted standards.
Statement: Having a default, does not mean the requirement for having the data represented by the default is not needed for the underlying technology.

In other words, just because you say http://webserver doesn't mean the program making the connection due to interpretation of this string doesn't need to know that the port it needs to communicate on. It still needs to know it's 80 (unless otherwise specified, natch).

Zecc

@dkf said in Canary Tests:

@Tsaukpaetra said in Canary Tests:

I don't recall implying that...

Funnily enough, I was getting the same impression as @PJH; there must be a way to systematically misinterpret your communication that we were both doing.

I am going to side with @Tsaukpaetra on this one, specially given:

@Tsaukpaetra said in Canary Tests:

@PJH said in Canary Tests:

I don't see the web littered with URLs containing :80 or :433, so I'm going to call "almost always" somewhat exaggerated.

Oh, they just moved the 80 to be http: and the 443 to https:. Just because a default has been communally adopted doesn't mean the underlying technology suddenly doesn't need it.

Edit: d by only 32 minutes.

Tsaukpaetra

@Zecc said in Canary Tests:

only 32 minutes.

I gotta get them in sometimes. I knew I stayed up late for a reason this time!

dcon

@levicki said in Canary Tests:

@dkf said in Canary Tests:

That's where you should be thinking of hitting the previous developer with a rolled up newspaper, like a puppy that's chewed up your best chair.

You never, ever hit puppies. You use positive reinforcement.

However, the previous developer does get the newspaper.

dcon

@levicki said in Canary Tests:

@dcon said in Canary Tests:

However, the previous developer does get the newspaper.

I prefer hitting them with a shielded SCSI cable until the code improves.

Or the Cat5-o-nine tails?

TwelveBaud

@dkf On Windows, you specify a two-byte address family when creating a socket, and then an sockaddr when connecting or binding containing the same two-byte family code and a 14-to-126-byte-or-more blob of arbitrary data. For IPv4, this is the port and then (binary) IP address, little endian. For IPv6 it also has spots reserved for flow control information and a scope ID. For IrDA it's a four-byte device ID assigned by IrDA and a 25-byte "well-known service" name; for traditional Unix sockets (which are now, finally, supported on Windows) it's their canonical full on-disk path in a PATH_MAX-sized buffer. And those are just the common ones. It gets icky. Oh, and GetAddrInfoW returns a completely different structure (ADDR_INFO_EX4W) that you need to be able to copy into the appropriately-sized sockaddr structure. Yaaay.

Click here if you want to be disappointed in your life choices!

#define AF_UNSPEC       0               // unspecified
#define AF_UNIX         1               // local to host (pipes, portals)
#define AF_INET         2               // internetwork: UDP, TCP, etc.
#define AF_IMPLINK      3               // arpanet imp addresses
#define AF_PUP          4               // pup protocols: e.g. BSP
#define AF_CHAOS        5               // mit CHAOS protocols
#define AF_NS           6               // XEROX NS protocols
#define AF_IPX          AF_NS           // IPX protocols: IPX, SPX, etc.
#define AF_ISO          7               // ISO protocols
#define AF_OSI          AF_ISO          // OSI is ISO
#define AF_ECMA         8               // european computer manufacturers
#define AF_DATAKIT      9               // datakit protocols
#define AF_CCITT        10              // CCITT protocols, X.25 etc
#define AF_SNA          11              // IBM SNA
#define AF_DECnet       12              // DECnet
#define AF_DLI          13              // Direct data link interface
#define AF_LAT          14              // LAT
#define AF_HYLINK       15              // NSC Hyperchannel
#define AF_APPLETALK    16              // AppleTalk
#define AF_NETBIOS      17              // NetBios-style addresses
#define AF_VOICEVIEW    18              // VoiceView
#define AF_FIREFOX      19              // Protocols from Firefox
#define AF_UNKNOWN1     20              // Somebody is using this!
#define AF_BAN          21              // Banyan
#define AF_ATM          22              // Native ATM Services
#define AF_INET6        23              // Internetwork Version 6
#define AF_CLUSTER      24              // Microsoft Wolfpack
#define AF_12844        25              // IEEE 1284.4 WG AF
#define AF_IRDA         26              // IrDA
#define AF_NETDES       28              // Network Designers OSI & gateway
#define AF_TCNPROCESS   29
#define AF_TCNMESSAGE   30
#define AF_ICLFXBM      31
#define AF_BTH          32              // Bluetooth RFCOMM/L2CAP protocols
#define AF_LINK         33
#define AF_HYPERV       34

dkf

@TwelveBaud said in Canary Tests:

And those are just the common ones. It gets icky.

It does indeed.

Zenith

Alright, here's one:

CREATE PROCEDURE messegeroutr12 (@ BOtype VARCHAR(5000), OUT x VARCHAR(5000), OUT x2 XML)
BEGIN
IF @ BOtype = 'A'
  code that returns an INT
IF @ BOtype = 'B'
  code that selects a TABLE or 10
IF @ BOtype = 'CorespondanceIsCorraspondense'
  code that sets an OUT parameter
IF @ BOtype = 'D'
  code that returns "success" or "succeeded" or "" or "there was an error, try again" or FALSE
IF @ BOtype = 'Z'
  down the rabbit hole of procedure chaining we go...
IF @ BOtype = 'B1'
  code that alters global state and parameters (like @ BOtype)
IF @ BOtype = 'TicketINC09345274563'
  nightmares won't stop
...89 IFs and 0 ELSEs later...
END

Lorne Kates

@Zenith said in Canary Tests:

BO

The code is right: it stinks.