[Libreoffice-commits] .: 3 commits - src/formula.py src/xlsrecord.py

[Kohei Yoshida]

 src/formula.py   |   49 ++++++++++++++++++++++++++++++++++++-------------
 src/xlsrecord.py |    7 +++++--
 2 files changed, 41 insertions(+), 15 deletions(-)

New commits:
commit f7297f738e736c0ad0d76f48baa55aa35baedc7a
Author: Kohei Yoshida <kyoshida at novell.com>
Date:   Thu Apr 14 13:55:37 2011 -0400

    Improved parsing of function tokens.

diff --git a/src/formula.py b/src/formula.py
index 440aec0..f5572b3 100644
--- a/src/formula.py
+++ b/src/formula.py
@@ -209,7 +209,7 @@ class PtgRef3d(PtgBase):
     def getText (self):
         return "(xti=%d,%s)"%(self.ixti, self.cell.getName())
 
-class _FuncVar(PtgBase):
+class PtgFuncVar(PtgBase):
 
     funcTab = {
         0x0000: 'COUNT',
@@ -597,17 +597,13 @@ class _FuncVar(PtgBase):
     def getText (self):
         if self.isCeTab:
             # I'll support this later.
-            return ''
+            raise FormulaParserError("special built-in function not supported yet")
 
-        if not _FuncVar.funcTab.has_key(self.funcType):
+        if not PtgFuncVar.funcTab.has_key(self.funcType):
             # unknown function name
             return '#NAME!'
 
-        if self.argCount > 0:
-            # I'll support functions with arguments later.
-            return ''
-
-        return _FuncVar.funcTab[self.funcType] + "()"
+        return "(func: %s; arg: %d)"%(PtgFuncVar.funcTab[self.funcType], self.argCount)
 
 _tokenMap = {
     0x01: PtgExp,
@@ -621,7 +617,7 @@ _tokenMap = {
     0x7B: _Area3d,
 
     0x3A: PtgRef3d,
-    0x42: _FuncVar
+    0x42: PtgFuncVar
 }
 
 class FormulaParser(object):
commit 0f888e5c6a676e786c601603cf734afd214bd70f
Author: Kohei Yoshida <kyoshida at novell.com>
Date:   Thu Apr 14 13:48:33 2011 -0400

    Pick up more formula token types.

diff --git a/src/formula.py b/src/formula.py
index 1b18978..440aec0 100644
--- a/src/formula.py
+++ b/src/formula.py
@@ -151,6 +151,29 @@ class PtgExp(PtgBase):
     def getText (self):
         return "(ptgexp: row=%d, col=%d)"%(self.row, self.col)
 
+class PtgMissArg(PtgBase):
+    def parseBytes (self):
+        pass
+
+    def getText (self):
+        return '(arg missing)'
+
+class PtgRef(PtgBase):
+    def parseBytes (self):
+        self.row = self.strm.readUnsignedInt(2)
+        self.col = self.strm.readUnsignedInt(2)
+
+    def getText (self):
+        return "(ref: row=%d, col=%d)"%(self.row, self.col)
+
+class PtgStr(PtgBase):
+    def parseBytes (self):
+        length = self.strm.readUnsignedInt(1)
+        self.value = self.strm.readUnicodeString(length)
+
+    def getText (self):
+        return "(str: '%s')"%self.value
+
 class PtgNameX(PtgBase):
     def parseBytes (self):
         self.xti = self.strm.readUnsignedInt(2)
@@ -159,12 +182,12 @@ class PtgNameX(PtgBase):
     def getText (self):
         return "(name: xti=%d, name=%d)"%(self.xti, self.nameID)
 
-class _Int(PtgBase):
+class PtgInt(PtgBase):
     def parseBytes (self):
         self.value = self.strm.readUnsignedInt(2)
 
     def getText (self):
-        return "%d"%self.value
+        return "(int: %d)"%self.value
 
 class _Area3d(PtgBase):
     def parseBytes (self):
@@ -588,7 +611,10 @@ class _FuncVar(PtgBase):
 
 _tokenMap = {
     0x01: PtgExp,
-    0x1E: _Int,
+    0x16: PtgMissArg,
+    0x17: PtgStr,
+    0x1E: PtgInt,
+    0x24: PtgRef,
     0x3B: _Area3d,
     0x59: PtgNameX,
     0x5B: _Area3d,
commit 63edbb942b1678bbb08ebdf34cc20ff0b0205403
Author: Kohei Yoshida <kyoshida at novell.com>
Date:   Thu Apr 14 13:28:05 2011 -0400

    Catch unknown token error from the formula parser.

diff --git a/src/formula.py b/src/formula.py
index 76000ff..1b18978 100644
--- a/src/formula.py
+++ b/src/formula.py
@@ -29,6 +29,7 @@ import struct, sys
 import globals
 
 class InvalidCellAddress(Exception): pass
+class FormulaParserError(Exception): pass
 
 def toColName (colID):
     if colID > 255:
@@ -614,7 +615,7 @@ associated token classes will be without the leading underscore (_)."""
             b = self.strm.readUnsignedInt(1)
             if not _tokenMap.has_key(b):
                 # Unknown token.  Stop parsing.
-                return
+                raise FormulaParserError("unknown token 0x%2.2X"%b)
 
             token = _tokenMap[b](self.strm, b)
             token.parse()
diff --git a/src/xlsrecord.py b/src/xlsrecord.py
index a860af3..3704c8f 100644
--- a/src/xlsrecord.py
+++ b/src/xlsrecord.py
@@ -949,8 +949,11 @@ class Formula(BaseRecordHandler):
     def parseBytes (self):
         self.__parseBytes()
         fparser = formula.FormulaParser(self.header, self.tokens)
-        fparser.parse()
-        ftext = fparser.getText()
+        try:
+            fparser.parse()
+            ftext = fparser.getText()
+        except formula.FormulaParserError as e:
+            ftext = "(Error: %s)"%e.args[0]
 
         self.appendCellPosition(self.col, self.row)
         self.appendLine("XF record ID: %d"%self.xf)