From aca5dc9ed0acd80a8405c2985b17183d4ae82e9a Mon Sep 17 00:00:00 2001
From: Matt Turnbull <mcturnbull@fastmail.fm>
Date: Tue, 24 Mar 2009 14:58:14 +0000
Subject: [PATCH] Parse another PokerStars datetime.

Make tail work:
  -log instead of print inode # change detected,
  -make sure splitter regexes have groups around them
  -fix a bug
  -increase buffering size to something >> a single hand

Conflicts:

	pyfpdb/HandHistoryConverter.py
---
 pyfpdb/EverleafToFpdb.py       |  2 +-
 pyfpdb/FulltiltToFpdb.py       |  2 +-
 pyfpdb/HandHistoryConverter.py | 11 +++++++----
 pyfpdb/PokerStarsToFpdb.py     |  9 +++++++--
 4 files changed, 16 insertions(+), 8 deletions(-)

diff --git a/pyfpdb/EverleafToFpdb.py b/pyfpdb/EverleafToFpdb.py
index d5deeeb8..29808295 100755
--- a/pyfpdb/EverleafToFpdb.py
+++ b/pyfpdb/EverleafToFpdb.py
@@ -27,7 +27,7 @@ from HandHistoryConverter import *
 class Everleaf(HandHistoryConverter):
     
     # Static regexes
-    re_SplitHands  = re.compile(r"\n\n+")
+    re_SplitHands  = re.compile(r"(\n\n\n+)")
     re_GameInfo    = re.compile(ur"^(Blinds )?(?P<CURRENCY>\$| €|)(?P<SB>[.0-9]+)/(?:\$| €)?(?P<BB>[.0-9]+) (?P<LIMIT>NL|PL|) ?(?P<GAME>(Hold\'em|Omaha|7 Card Stud))", re.MULTILINE)
                      #re.compile(ur"^(Blinds )?(?P<CURRENCY>\$| €|)(?P<SB>[.0-9]+)/(?:\$| €)?(?P<BB>[.0-9]+) (?P<LIMIT>NL|PL|) (?P<GAME>(Hold\'em|Omaha|7 Card Stud))", re.MULTILINE)
     re_HandInfo    = re.compile(ur".*#(?P<HID>[0-9]+)\n.*\n(Blinds )?(?:\$| €|)(?P<SB>[.0-9]+)/(?:\$| €|)(?P<BB>[.0-9]+) (?P<GAMETYPE>.*) - (?P<DATETIME>\d\d\d\d/\d\d/\d\d - \d\d:\d\d:\d\d)\nTable (?P<TABLE>.+$)", re.MULTILINE)
diff --git a/pyfpdb/FulltiltToFpdb.py b/pyfpdb/FulltiltToFpdb.py
index 6a897bad..f3d30229 100755
--- a/pyfpdb/FulltiltToFpdb.py
+++ b/pyfpdb/FulltiltToFpdb.py
@@ -28,7 +28,7 @@ class Fulltilt(HandHistoryConverter):
     
     # Static regexes
     re_GameInfo     = re.compile('- (?P<CURRENCY>\$|)?(?P<SB>[.0-9]+)/\$?(?P<BB>[.0-9]+) (Ante \$(?P<ANTE>[.0-9]+) )?- (?P<LIMIT>(No Limit|Pot Limit|Limit))? (?P<GAME>(Hold\'em|Omaha Hi|Razz))')
-    re_SplitHands   = re.compile(r"\n\n+")
+    re_SplitHands   = re.compile(r"(\n\n+)")
     re_HandInfo     = re.compile('.*#(?P<HID>[0-9]+): Table (?P<TABLE>[- a-zA-Z]+) (\((?P<TABLEATTRIBUTES>.+)\) )?- \$?(?P<SB>[.0-9]+)/\$?(?P<BB>[.0-9]+) (Ante \$(?P<ANTE>[.0-9]+) )?- (?P<GAMETYPE>[a-zA-Z\' ]+) - (?P<DATETIME>.*)')
     re_Button       = re.compile('^The button is in seat #(?P<BUTTON>\d+)', re.MULTILINE)
     re_PlayerInfo   = re.compile('Seat (?P<SEAT>[0-9]+): (?P<PNAME>.*) \(\$(?P<CASH>[.0-9]+)\)\n')
diff --git a/pyfpdb/HandHistoryConverter.py b/pyfpdb/HandHistoryConverter.py
index 98a96638..8af8fbd1 100644
--- a/pyfpdb/HandHistoryConverter.py
+++ b/pyfpdb/HandHistoryConverter.py
@@ -72,7 +72,8 @@ import gettext
 gettext.install('myapplication')
 
 class HandHistoryConverter():
-    READ_CHUNK_SIZE = 1000 # bytes to read at a time from file
+
+    READ_CHUNK_SIZE = 10000 # bytes to read at a time from file (in tail mode)
     def __init__(self, in_path = '-', out_path = '-', sitename = None, follow=False):
         logging.info("HandHistory init called")
         
@@ -154,7 +155,7 @@ Tail the in_path file and yield handTexts separated by re_SplitHands"""
                     time.sleep(interval)
                     fd.seek(where)
                 else:
-                    print "%s changed inode numbers from %d to %d" % (self.in_path, fd_results[1], st_results[1])
+                    logging.debug("%s changed inode numbers from %d to %d" % (self.in_path, fd_results[1], st_results[1]))
                     fd = codecs.open(self.in_path, 'r', self.codepage)
                     fd.seek(where)
             else:
@@ -162,6 +163,7 @@ Tail the in_path file and yield handTexts separated by re_SplitHands"""
                 data = data + newdata
                 result = self.re_SplitHands.split(data)
                 result = iter(result)
+                data = ''
                 # --x       data (- is bit of splitter, x is paragraph)     yield,...,keep
                 # [,--,x]    result of re.split (with group around splitter)
                 # ,x        our output: yield nothing, keep x
@@ -181,9 +183,10 @@ Tail the in_path file and yield handTexts separated by re_SplitHands"""
                 # We want to yield all paragraphs followed by a splitter, i.e. all even indices except the last.
                 for para in result:
                     try:
-                        splitter = result.next()
+                        result.next()
+                        splitter = True
                     except StopIteration:
-                        splitter = None
+                        splitter = False
                     if splitter: # para is followed by a splitter
                         if para: yield para # para not ''
                     else:
diff --git a/pyfpdb/PokerStarsToFpdb.py b/pyfpdb/PokerStarsToFpdb.py
index 8b905005..4f150ead 100755
--- a/pyfpdb/PokerStarsToFpdb.py
+++ b/pyfpdb/PokerStarsToFpdb.py
@@ -27,7 +27,7 @@ class PokerStars(HandHistoryConverter):
 
     # Static regexes
     re_GameInfo     = re.compile("PokerStars Game #(?P<HID>[0-9]+):\s+(HORSE)? \(?(?P<GAME>Hold\'em|Razz|7 Card Stud|Omaha|Omaha Hi/Lo|Badugi) (?P<LIMIT>No Limit|Limit|Pot Limit),? \(?(?P<CURRENCY>\$|)?(?P<SB>[.0-9]+)/\$?(?P<BB>[.0-9]+)\) - (?P<DATETIME>.*$)", re.MULTILINE)
-    re_SplitHands   = re.compile('\n\n+')
+    re_SplitHands   = re.compile('(\n\n+)')
     re_HandInfo     = re.compile("^Table \'(?P<TABLE>[- a-zA-Z]+)\'(?P<TABLEATTRIBUTES>.+?$)?", re.MULTILINE)
     re_Button       = re.compile('Seat #(?P<BUTTON>\d+) is the button', re.MULTILINE)
     re_PlayerInfo   = re.compile('^Seat (?P<SEAT>[0-9]+): (?P<PNAME>.*) \(\$?(?P<CASH>[.0-9]+) in chips\)', re.MULTILINE)
@@ -128,7 +128,12 @@ follow :  whether to tail -f the input"""
         logging.debug("readHandInfo: %s" % info)
         for key in info:
             if key == 'DATETIME':
-                datetime = info[key].replace(" - "," ") # some are like "2009/02/26 - 15:22:55 ET"
+                datetime = info[key]                
+                #2008/11/16 1:22:47 CET [2008/11/15 19:22:47 ET]                
+                m2 = re.search(r".+\[(.+) ET\]", datetime)
+                if m2: datetime = m2.group(1)
+                #2009/02/26 - 15:22:55 ET
+                datetime = datetime.replace(" - "," ") # some are like "2009/02/26 - 15:22:55 ET"
                 datetime = datetime.replace(" (ET)","") # kludge for now.
                 datetime = datetime.replace(" ET","") # kludge for now.
                 hand.starttime = time.strptime(datetime, "%Y/%m/%d %H:%M:%S")