I think we can safely assume that PT3 will read a HH if it's *identical* to the format of an accepted casino.
However, as developers, we would like to know which parts of the HH are used, and how how stringent the parsing is. I wouldn't hurt to know which casino had the simplest histories either.
Hand histories are design to be human readable, and contain a lot of redundant information and pleasantries that we don't want to have to recreate in order to have PT3 parse a data file and put it into a database.
Here's how simple it could be (and might be - we don't know unless we ask):
Here's what I would like to write in my ideal HH format:
Game: Flinstones Family, Hardrock Table, 2500 BC.
Seat 1: Fred
Seat 2: Jane
Seat 3: Wilma
Seat 4: Barney
Button: Seat 4
Dealt: [Cards] to 1
3 bets $2
4 bet $4
1 folds
2 calls
Flop: [Cards]
2 bets $2
3 bets $4
4 calls
1 calls
[..]
3 shows [CARDS]
3 wins $20
<END>
Here is an made-up example of pokerstars hand - I'd like to know how much of it I need. This will affect my decision to buy your product.
Table 'Carlova' 9-max Seat #6 is the button
Seat 1: hotMccoffee ($51.20 in chips)
...
Seat 9: dudetristen ($50 in chips)
Quad2x2: is sitting out
Quad2x2 leaves the table
Tobsen: is sitting out
Tobsen leaves the table
loxster: posts small blind $0.25
Foldarz: posts big blind $0.50
Do I need to include the worth of every player?
Do I need to include players that are sitting out.
Do I need to re-impress upon PT3 the cost of the small and big blind?
...
pepero64: folds
samsemilia: bets $1
samsemilia: raises $1 to $2
Do I really have to differentiate between a bet and a raise, and then tell PT3 that 1 + 1 = 2?
*** FLOP *** [7c 6c 3c]
biff4968: calls $3
samsemilia: raises $20 to $23 and is all-in
biff4968: folds
Uncalled bet ($22) returned to samsemilia
Again, why am I doing math here... if PT3 is using the initial worth of each player, then is it really taking any notice of the "all-in"?
Must I explain to it that uncalled bets are returned to players?
*** SHOW DOWN ***
samsemilia: shows [Qc Qs] (a flush, Queen high)
biff4968: mucks hand
samsemilia collected $14.05 from pot
*** SUMMARY ***
I'm fairly sure that PT3 isn't actually processing the description "a flush, Queen high"... Maybe I could write "A brown flushed turd" instead.
I don't really want to have to write translations for every type of hand, to perfectly match the HH format.
- a full house, Kings over Queens
a full house, Queens full of Kings
Kings full over Queens
a set
a set of 3's
a set of 3's with an Ace kicker
Aces and Kings
2 pair, Aces and Kings
Straight, ace high.
Straight Flush, ace high
Royal Flush
5 Kings
Full house, Kings over kings
Quads, kings with a king kicker.
two pair, KK and KK with a green leprechaun kicker
Total pot $14.75 | Rake $0.70
I like this line. Definately good to make sure that the bets as written = the pot.
Board [7c 6c 3c 9c Qd]
Seat 1: hotMccoffee folded before Flop (didn't bet)
Seat 2: canaans folded before Flop (didn't bet)
Seat 3: pepero64 folded before Flop (didn't bet)
Seat 4: samsemilia showed [Qc Qs] and won ($14.05) with a flush, Queen high
Seat 5: biff4968 mucked
Seat 6: lydiasboy84 (button) folded before Flop (didn't bet)
Seat 7: loxster (small blind) folded before Flop
Seat 8: Foldarz (big blind) folded before Flop
Seat 9: dudetristen folded before Flop (didn't bet)
10 lines of useless, painful to recreate, duplicate data. How many times does PT3 really need to see the common hands. Does it need to be reminded who folded, and what the winning hand was? Does it have Alzheimers?
I could probably live with the stars format, if I could just skip the summary line. However I fear that PT3 actually reads in some of those fields for cross checking.
I would like to know which ones - or be pointed to an easier history file.