URL
https://opencores.org/ocsvn/openrisc_me/openrisc_me/trunk
Subversion Repositories openrisc_me
[/] [openrisc/] [trunk/] [gnu-src/] [gcc-4.5.1/] [gcc/] [ada/] [g-spipat.adb] - Rev 307
Go to most recent revision | Compare with Previous | Blame | View Log
------------------------------------------------------------------------------ -- -- -- GNAT LIBRARY COMPONENTS -- -- -- -- G N A T . S P I T B O L . P A T T E R N S -- -- -- -- B o d y -- -- -- -- Copyright (C) 1998-2008, AdaCore -- -- -- -- GNAT is free software; you can redistribute it and/or modify it under -- -- terms of the GNU General Public License as published by the Free Soft- -- -- ware Foundation; either version 2, or (at your option) any later ver- -- -- sion. GNAT is distributed in the hope that it will be useful, but WITH- -- -- OUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY -- -- or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License -- -- for more details. You should have received a copy of the GNU General -- -- Public License distributed with GNAT; see file COPYING. If not, write -- -- to the Free Software Foundation, 51 Franklin Street, Fifth Floor, -- -- Boston, MA 02110-1301, USA. -- -- -- -- As a special exception, if other files instantiate generics from this -- -- unit, or you link this unit with other files to produce an executable, -- -- this unit does not by itself cause the resulting executable to be -- -- covered by the GNU General Public License. This exception does not -- -- however invalidate any other reasons why the executable file might be -- -- covered by the GNU Public License. -- -- -- -- GNAT was originally developed by the GNAT team at New York University. -- -- Extensive contributions were provided by Ada Core Technologies Inc. -- -- -- ------------------------------------------------------------------------------ -- Note: the data structures and general approach used in this implementation -- are derived from the original MINIMAL sources for SPITBOL. The code is not -- a direct translation, but the approach is followed closely. In particular, -- we use the one stack approach developed in the SPITBOL implementation. with Ada.Strings.Unbounded.Aux; use Ada.Strings.Unbounded.Aux; with GNAT.Debug_Utilities; use GNAT.Debug_Utilities; with System; use System; with Ada.Unchecked_Conversion; with Ada.Unchecked_Deallocation; package body GNAT.Spitbol.Patterns is ------------------------ -- Internal Debugging -- ------------------------ Internal_Debug : constant Boolean := False; -- Set this flag to True to activate some built-in debugging traceback -- These are all lines output with PutD and Put_LineD. procedure New_LineD; pragma Inline (New_LineD); -- Output new blank line with New_Line if Internal_Debug is True procedure PutD (Str : String); pragma Inline (PutD); -- Output string with Put if Internal_Debug is True procedure Put_LineD (Str : String); pragma Inline (Put_LineD); -- Output string with Put_Line if Internal_Debug is True ----------------------------- -- Local Type Declarations -- ----------------------------- subtype String_Ptr is Ada.Strings.Unbounded.String_Access; subtype File_Ptr is Ada.Text_IO.File_Access; function To_Address is new Ada.Unchecked_Conversion (PE_Ptr, Address); -- Used only for debugging output purposes subtype AFC is Ada.Finalization.Controlled; N : constant PE_Ptr := null; -- Shorthand used to initialize Copy fields to null type Natural_Ptr is access all Natural; type Pattern_Ptr is access all Pattern; -------------------------------------------------- -- Description of Algorithm and Data Structures -- -------------------------------------------------- -- A pattern structure is represented as a linked graph of nodes -- with the following structure: -- +------------------------------------+ -- I Pcode I -- +------------------------------------+ -- I Index I -- +------------------------------------+ -- I Pthen I -- +------------------------------------+ -- I parameter(s) I -- +------------------------------------+ -- Pcode is a code value indicating the type of the pattern node. This -- code is used both as the discriminant value for the record, and as -- the case index in the main match routine that branches to the proper -- match code for the given element. -- Index is a serial index number. The use of these serial index -- numbers is described in a separate section. -- Pthen is a pointer to the successor node, i.e the node to be matched -- if the attempt to match the node succeeds. If this is the last node -- of the pattern to be matched, then Pthen points to a dummy node -- of kind PC_EOP (end of pattern), which initializes pattern exit. -- The parameter or parameters are present for certain node types, -- and the type varies with the pattern code. type Pattern_Code is ( PC_Arb_Y, PC_Assign, PC_Bal, PC_BreakX_X, PC_Cancel, PC_EOP, PC_Fail, PC_Fence, PC_Fence_X, PC_Fence_Y, PC_R_Enter, PC_R_Remove, PC_R_Restore, PC_Rest, PC_Succeed, PC_Unanchored, PC_Alt, PC_Arb_X, PC_Arbno_S, PC_Arbno_X, PC_Rpat, PC_Pred_Func, PC_Assign_Imm, PC_Assign_OnM, PC_Any_VP, PC_Break_VP, PC_BreakX_VP, PC_NotAny_VP, PC_NSpan_VP, PC_Span_VP, PC_String_VP, PC_Write_Imm, PC_Write_OnM, PC_Null, PC_String, PC_String_2, PC_String_3, PC_String_4, PC_String_5, PC_String_6, PC_Setcur, PC_Any_CH, PC_Break_CH, PC_BreakX_CH, PC_Char, PC_NotAny_CH, PC_NSpan_CH, PC_Span_CH, PC_Any_CS, PC_Break_CS, PC_BreakX_CS, PC_NotAny_CS, PC_NSpan_CS, PC_Span_CS, PC_Arbno_Y, PC_Len_Nat, PC_Pos_Nat, PC_RPos_Nat, PC_RTab_Nat, PC_Tab_Nat, PC_Pos_NF, PC_Len_NF, PC_RPos_NF, PC_RTab_NF, PC_Tab_NF, PC_Pos_NP, PC_Len_NP, PC_RPos_NP, PC_RTab_NP, PC_Tab_NP, PC_Any_VF, PC_Break_VF, PC_BreakX_VF, PC_NotAny_VF, PC_NSpan_VF, PC_Span_VF, PC_String_VF); type IndexT is range 0 .. +(2 **15 - 1); type PE (Pcode : Pattern_Code) is record Index : IndexT; -- Serial index number of pattern element within pattern Pthen : PE_Ptr; -- Successor element, to be matched after this one case Pcode is when PC_Arb_Y | PC_Assign | PC_Bal | PC_BreakX_X | PC_Cancel | PC_EOP | PC_Fail | PC_Fence | PC_Fence_X | PC_Fence_Y | PC_Null | PC_R_Enter | PC_R_Remove | PC_R_Restore | PC_Rest | PC_Succeed | PC_Unanchored => null; when PC_Alt | PC_Arb_X | PC_Arbno_S | PC_Arbno_X => Alt : PE_Ptr; when PC_Rpat => PP : Pattern_Ptr; when PC_Pred_Func => BF : Boolean_Func; when PC_Assign_Imm | PC_Assign_OnM | PC_Any_VP | PC_Break_VP | PC_BreakX_VP | PC_NotAny_VP | PC_NSpan_VP | PC_Span_VP | PC_String_VP => VP : VString_Ptr; when PC_Write_Imm | PC_Write_OnM => FP : File_Ptr; when PC_String => Str : String_Ptr; when PC_String_2 => Str2 : String (1 .. 2); when PC_String_3 => Str3 : String (1 .. 3); when PC_String_4 => Str4 : String (1 .. 4); when PC_String_5 => Str5 : String (1 .. 5); when PC_String_6 => Str6 : String (1 .. 6); when PC_Setcur => Var : Natural_Ptr; when PC_Any_CH | PC_Break_CH | PC_BreakX_CH | PC_Char | PC_NotAny_CH | PC_NSpan_CH | PC_Span_CH => Char : Character; when PC_Any_CS | PC_Break_CS | PC_BreakX_CS | PC_NotAny_CS | PC_NSpan_CS | PC_Span_CS => CS : Character_Set; when PC_Arbno_Y | PC_Len_Nat | PC_Pos_Nat | PC_RPos_Nat | PC_RTab_Nat | PC_Tab_Nat => Nat : Natural; when PC_Pos_NF | PC_Len_NF | PC_RPos_NF | PC_RTab_NF | PC_Tab_NF => NF : Natural_Func; when PC_Pos_NP | PC_Len_NP | PC_RPos_NP | PC_RTab_NP | PC_Tab_NP => NP : Natural_Ptr; when PC_Any_VF | PC_Break_VF | PC_BreakX_VF | PC_NotAny_VF | PC_NSpan_VF | PC_Span_VF | PC_String_VF => VF : VString_Func; end case; end record; subtype PC_Has_Alt is Pattern_Code range PC_Alt .. PC_Arbno_X; -- Range of pattern codes that has an Alt field. This is used in the -- recursive traversals, since these links must be followed. EOP_Element : aliased constant PE := (PC_EOP, 0, N); -- This is the end of pattern element, and is thus the representation of -- a null pattern. It has a zero index element since it is never placed -- inside a pattern. Furthermore it does not need a successor, since it -- marks the end of the pattern, so that no more successors are needed. EOP : constant PE_Ptr := EOP_Element'Unrestricted_Access; -- This is the end of pattern pointer, that is used in the Pthen pointer -- of other nodes to signal end of pattern. -- The following array is used to determine if a pattern used as an -- argument for Arbno is eligible for treatment using the simple Arbno -- structure (i.e. it is a pattern that is guaranteed to match at least -- one character on success, and not to make any entries on the stack. OK_For_Simple_Arbno : constant array (Pattern_Code) of Boolean := (PC_Any_CS | PC_Any_CH | PC_Any_VF | PC_Any_VP | PC_Char | PC_Len_Nat | PC_NotAny_CS | PC_NotAny_CH | PC_NotAny_VF | PC_NotAny_VP | PC_Span_CS | PC_Span_CH | PC_Span_VF | PC_Span_VP | PC_String | PC_String_2 | PC_String_3 | PC_String_4 | PC_String_5 | PC_String_6 => True, others => False); ------------------------------- -- The Pattern History Stack -- ------------------------------- -- The pattern history stack is used for controlling backtracking when -- a match fails. The idea is to stack entries that give a cursor value -- to be restored, and a node to be reestablished as the current node to -- attempt an appropriate rematch operation. The processing for a pattern -- element that has rematch alternatives pushes an appropriate entry or -- entry on to the stack, and the proceeds. If a match fails at any point, -- the top element of the stack is popped off, resetting the cursor and -- the match continues by accessing the node stored with this entry. type Stack_Entry is record Cursor : Integer; -- Saved cursor value that is restored when this entry is popped -- from the stack if a match attempt fails. Occasionally, this -- field is used to store a history stack pointer instead of a -- cursor. Such cases are noted in the documentation and the value -- stored is negative since stack pointer values are always negative. Node : PE_Ptr; -- This pattern element reference is reestablished as the current -- Node to be matched (which will attempt an appropriate rematch). end record; subtype Stack_Range is Integer range -Stack_Size .. -1; type Stack_Type is array (Stack_Range) of Stack_Entry; -- The type used for a history stack. The actual instance of the stack -- is declared as a local variable in the Match routine, to properly -- handle recursive calls to Match. All stack pointer values are negative -- to distinguish them from normal cursor values. -- Note: the pattern matching stack is used only to handle backtracking. -- If no backtracking occurs, its entries are never accessed, and never -- popped off, and in particular it is normal for a successful match -- to terminate with entries on the stack that are simply discarded. -- Note: in subsequent diagrams of the stack, we always place element -- zero (the deepest element) at the top of the page, then build the -- stack down on the page with the most recent (top of stack) element -- being the bottom-most entry on the page. -- Stack checking is handled by labeling every pattern with the maximum -- number of stack entries that are required, so a single check at the -- start of matching the pattern suffices. There are two exceptions. -- First, the count does not include entries for recursive pattern -- references. Such recursions must therefore perform a specific -- stack check with respect to the number of stack entries required -- by the recursive pattern that is accessed and the amount of stack -- that remains unused. -- Second, the count includes only one iteration of an Arbno pattern, -- so a specific check must be made on subsequent iterations that there -- is still enough stack space left. The Arbno node has a field that -- records the number of stack entries required by its argument for -- this purpose. --------------------------------------------------- -- Use of Serial Index Field in Pattern Elements -- --------------------------------------------------- -- The serial index numbers for the pattern elements are assigned as -- a pattern is constructed from its constituent elements. Note that there -- is never any sharing of pattern elements between patterns (copies are -- always made), so the serial index numbers are unique to a particular -- pattern as referenced from the P field of a value of type Pattern. -- The index numbers meet three separate invariants, which are used for -- various purposes as described in this section. -- First, the numbers uniquely identify the pattern elements within a -- pattern. If Num is the number of elements in a given pattern, then -- the serial index numbers for the elements of this pattern will range -- from 1 .. Num, so that each element has a separate value. -- The purpose of this assignment is to provide a convenient auxiliary -- data structure mechanism during operations which must traverse a -- pattern (e.g. copy and finalization processing). Once constructed -- patterns are strictly read only. This is necessary to allow sharing -- of patterns between tasks. This means that we cannot go marking the -- pattern (e.g. with a visited bit). Instead we construct a separate -- vector that contains the necessary information indexed by the Index -- values in the pattern elements. For this purpose the only requirement -- is that they be uniquely assigned. -- Second, the pattern element referenced directly, i.e. the leading -- pattern element, is always the maximum numbered element and therefore -- indicates the total number of elements in the pattern. More precisely, -- the element referenced by the P field of a pattern value, or the -- element returned by any of the internal pattern construction routines -- in the body (that return a value of type PE_Ptr) always is this -- maximum element, -- The purpose of this requirement is to allow an immediate determination -- of the number of pattern elements within a pattern. This is used to -- properly size the vectors used to contain auxiliary information for -- traversal as described above. -- Third, as compound pattern structures are constructed, the way in which -- constituent parts of the pattern are constructed is stylized. This is -- an automatic consequence of the way that these compound structures -- are constructed, and basically what we are doing is simply documenting -- and specifying the natural result of the pattern construction. The -- section describing compound pattern structures gives details of the -- numbering of each compound pattern structure. -- The purpose of specifying the stylized numbering structures for the -- compound patterns is to help simplify the processing in the Image -- function, since it eases the task of retrieving the original recursive -- structure of the pattern from the flat graph structure of elements. -- This use in the Image function is the only point at which the code -- makes use of the stylized structures. type Ref_Array is array (IndexT range <>) of PE_Ptr; -- This type is used to build an array whose N'th entry references the -- element in a pattern whose Index value is N. See Build_Ref_Array. procedure Build_Ref_Array (E : PE_Ptr; RA : out Ref_Array); -- Given a pattern element which is the leading element of a pattern -- structure, and a Ref_Array with bounds 1 .. E.Index, fills in the -- Ref_Array so that its N'th entry references the element of the -- referenced pattern whose Index value is N. ------------------------------- -- Recursive Pattern Matches -- ------------------------------- -- The pattern primitive (+P) where P is a Pattern_Ptr or Pattern_Func -- causes a recursive pattern match. This cannot be handled by an actual -- recursive call to the outer level Match routine, since this would not -- allow for possible backtracking into the region matched by the inner -- pattern. Indeed this is the classical clash between recursion and -- backtracking, and a simple recursive stack structure does not suffice. -- This section describes how this recursion and the possible associated -- backtracking is handled. We still use a single stack, but we establish -- the concept of nested regions on this stack, each of which has a stack -- base value pointing to the deepest stack entry of the region. The base -- value for the outer level is zero. -- When a recursive match is established, two special stack entries are -- made. The first entry is used to save the original node that starts -- the recursive match. This is saved so that the successor field of -- this node is accessible at the end of the match, but it is never -- popped and executed. -- The second entry corresponds to a standard new region action. A -- PC_R_Remove node is stacked, whose cursor field is used to store -- the outer stack base, and the stack base is reset to point to -- this PC_R_Remove node. Then the recursive pattern is matched and -- it can make history stack entries in the normal matter, so now -- the stack looks like: -- (stack entries made by outer level) -- (Special entry, node is (+P) successor -- cursor entry is not used) -- (PC_R_Remove entry, "cursor" value is (negative) <-- Stack base -- saved base value for the enclosing region) -- (stack entries made by inner level) -- If a subsequent failure occurs and pops the PC_R_Remove node, it -- removes itself and the special entry immediately underneath it, -- restores the stack base value for the enclosing region, and then -- again signals failure to look for alternatives that were stacked -- before the recursion was initiated. -- Now we need to consider what happens if the inner pattern succeeds, as -- signalled by accessing the special PC_EOP pattern primitive. First we -- recognize the nested case by looking at the Base value. If this Base -- value is Stack'First, then the entire match has succeeded, but if the -- base value is greater than Stack'First, then we have successfully -- matched an inner pattern, and processing continues at the outer level. -- There are two cases. The simple case is when the inner pattern has made -- no stack entries, as recognized by the fact that the current stack -- pointer is equal to the current base value. In this case it is fine to -- remove all trace of the recursion by restoring the outer base value and -- using the special entry to find the appropriate successor node. -- The more complex case arises when the inner match does make stack -- entries. In this case, the PC_EOP processing stacks a special entry -- whose cursor value saves the saved inner base value (the one that -- references the corresponding PC_R_Remove value), and whose node -- pointer references a PC_R_Restore node, so the stack looks like: -- (stack entries made by outer level) -- (Special entry, node is (+P) successor, -- cursor entry is not used) -- (PC_R_Remove entry, "cursor" value is (negative) -- saved base value for the enclosing region) -- (stack entries made by inner level) -- (PC_Region_Replace entry, "cursor" value is (negative) -- stack pointer value referencing the PC_R_Remove entry). -- If the entire match succeeds, then these stack entries are, as usual, -- ignored and abandoned. If on the other hand a subsequent failure -- causes the PC_Region_Replace entry to be popped, it restores the -- inner base value from its saved "cursor" value and then fails again. -- Note that it is OK that the cursor is temporarily clobbered by this -- pop, since the second failure will reestablish a proper cursor value. --------------------------------- -- Compound Pattern Structures -- --------------------------------- -- This section discusses the compound structures used to represent -- constructed patterns. It shows the graph structures of pattern -- elements that are constructed, and in the case of patterns that -- provide backtracking possibilities, describes how the history -- stack is used to control the backtracking. Finally, it notes the -- way in which the Index numbers are assigned to the structure. -- In all diagrams, solid lines (built with minus signs or vertical -- bars, represent successor pointers (Pthen fields) with > or V used -- to indicate the direction of the pointer. The initial node of the -- structure is in the upper left of the diagram. A dotted line is an -- alternative pointer from the element above it to the element below -- it. See individual sections for details on how alternatives are used. ------------------- -- Concatenation -- ------------------- -- In the pattern structures listed in this section, a line that looks -- like ----> with nothing to the right indicates an end of pattern -- (EOP) pointer that represents the end of the match. -- When a pattern concatenation (L & R) occurs, the resulting structure -- is obtained by finding all such EOP pointers in L, and replacing -- them to point to R. This is the most important flattening that -- occurs in constructing a pattern, and it means that the pattern -- matching circuitry does not have to keep track of the structure -- of a pattern with respect to concatenation, since the appropriate -- successor is always at hand. -- Concatenation itself generates no additional possibilities for -- backtracking, but the constituent patterns of the concatenated -- structure will make stack entries as usual. The maximum amount -- of stack required by the structure is thus simply the sum of the -- maximums required by L and R. -- The index numbering of a concatenation structure works by leaving -- the numbering of the right hand pattern, R, unchanged and adjusting -- the numbers in the left hand pattern, L up by the count of elements -- in R. This ensures that the maximum numbered element is the leading -- element as required (given that it was the leading element in L). ----------------- -- Alternation -- ----------------- -- A pattern (L or R) constructs the structure: -- +---+ +---+ -- | A |---->| L |----> -- +---+ +---+ -- . -- . -- +---+ -- | R |----> -- +---+ -- The A element here is a PC_Alt node, and the dotted line represents -- the contents of the Alt field. When the PC_Alt element is matched, -- it stacks a pointer to the leading element of R on the history stack -- so that on subsequent failure, a match of R is attempted. -- The A node is the highest numbered element in the pattern. The -- original index numbers of R are unchanged, but the index numbers -- of the L pattern are adjusted up by the count of elements in R. -- Note that the difference between the index of the L leading element -- the index of the R leading element (after building the alt structure) -- indicates the number of nodes in L, and this is true even after the -- structure is incorporated into some larger structure. For example, -- if the A node has index 16, and L has index 15 and R has index -- 5, then we know that L has 10 (15-5) elements in it. -- Suppose that we now concatenate this structure to another pattern -- with 9 elements in it. We will now have the A node with an index -- of 25, L with an index of 24 and R with an index of 14. We still -- know that L has 10 (24-14) elements in it, numbered 15-24, and -- consequently the successor of the alternation structure has an -- index with a value less than 15. This is used in Image to figure -- out the original recursive structure of a pattern. -- To clarify the interaction of the alternation and concatenation -- structures, here is a more complex example of the structure built -- for the pattern: -- (V or W or X) (Y or Z) -- where A,B,C,D,E are all single element patterns: -- +---+ +---+ +---+ +---+ -- I A I---->I V I---+-->I A I---->I Y I----> -- +---+ +---+ I +---+ +---+ -- . I . -- . I . -- +---+ +---+ I +---+ -- I A I---->I W I-->I I Z I----> -- +---+ +---+ I +---+ -- . I -- . I -- +---+ I -- I X I------------>+ -- +---+ -- The numbering of the nodes would be as follows: -- +---+ +---+ +---+ +---+ -- I 8 I---->I 7 I---+-->I 3 I---->I 2 I----> -- +---+ +---+ I +---+ +---+ -- . I . -- . I . -- +---+ +---+ I +---+ -- I 6 I---->I 5 I-->I I 1 I----> -- +---+ +---+ I +---+ -- . I -- . I -- +---+ I -- I 4 I------------>+ -- +---+ -- Note: The above structure actually corresponds to -- (A or (B or C)) (D or E) -- rather than -- ((A or B) or C) (D or E) -- which is the more natural interpretation, but in fact alternation -- is associative, and the construction of an alternative changes the -- left grouped pattern to the right grouped pattern in any case, so -- that the Image function produces a more natural looking output. --------- -- Arb -- --------- -- An Arb pattern builds the structure -- +---+ -- | X |----> -- +---+ -- . -- . -- +---+ -- | Y |----> -- +---+ -- The X node is a PC_Arb_X node, which matches null, and stacks a -- pointer to Y node, which is the PC_Arb_Y node that matches one -- extra character and restacks itself. -- The PC_Arb_X node is numbered 2, and the PC_Arb_Y node is 1 ------------------------- -- Arbno (simple case) -- ------------------------- -- The simple form of Arbno can be used where the pattern always -- matches at least one character if it succeeds, and it is known -- not to make any history stack entries. In this case, Arbno (P) -- can construct the following structure: -- +-------------+ -- | ^ -- V | -- +---+ | -- | S |----> | -- +---+ | -- . | -- . | -- +---+ | -- | P |---------->+ -- +---+ -- The S (PC_Arbno_S) node matches null stacking a pointer to the -- pattern P. If a subsequent failure causes P to be matched and -- this match succeeds, then node A gets restacked to try another -- instance if needed by a subsequent failure. -- The node numbering of the constituent pattern P is not affected. -- The S node has a node number of P.Index + 1. -------------------------- -- Arbno (complex case) -- -------------------------- -- A call to Arbno (P), where P can match null (or at least is not -- known to require a non-null string) and/or P requires pattern stack -- entries, constructs the following structure: -- +--------------------------+ -- | ^ -- V | -- +---+ | -- | X |----> | -- +---+ | -- . | -- . | -- +---+ +---+ +---+ | -- | E |---->| P |---->| Y |--->+ -- +---+ +---+ +---+ -- The node X (PC_Arbno_X) matches null, stacking a pointer to the -- E-P-X structure used to match one Arbno instance. -- Here E is the PC_R_Enter node which matches null and creates two -- stack entries. The first is a special entry whose node field is -- not used at all, and whose cursor field has the initial cursor. -- The second entry corresponds to a standard new region action. A -- PC_R_Remove node is stacked, whose cursor field is used to store -- the outer stack base, and the stack base is reset to point to -- this PC_R_Remove node. Then the pattern P is matched, and it can -- make history stack entries in the normal manner, so now the stack -- looks like: -- (stack entries made before assign pattern) -- (Special entry, node field not used, -- used only to save initial cursor) -- (PC_R_Remove entry, "cursor" value is (negative) <-- Stack Base -- saved base value for the enclosing region) -- (stack entries made by matching P) -- If the match of P fails, then the PC_R_Remove entry is popped and -- it removes both itself and the special entry underneath it, -- restores the outer stack base, and signals failure. -- If the match of P succeeds, then node Y, the PC_Arbno_Y node, pops -- the inner region. There are two possibilities. If matching P left -- no stack entries, then all traces of the inner region can be removed. -- If there are stack entries, then we push an PC_Region_Replace stack -- entry whose "cursor" value is the inner stack base value, and then -- restore the outer stack base value, so the stack looks like: -- (stack entries made before assign pattern) -- (Special entry, node field not used, -- used only to save initial cursor) -- (PC_R_Remove entry, "cursor" value is (negative) -- saved base value for the enclosing region) -- (stack entries made by matching P) -- (PC_Region_Replace entry, "cursor" value is (negative) -- stack pointer value referencing the PC_R_Remove entry). -- Now that we have matched another instance of the Arbno pattern, -- we need to move to the successor. There are two cases. If the -- Arbno pattern matched null, then there is no point in seeking -- alternatives, since we would just match a whole bunch of nulls. -- In this case we look through the alternative node, and move -- directly to its successor (i.e. the successor of the Arbno -- pattern). If on the other hand a non-null string was matched, -- we simply follow the successor to the alternative node, which -- sets up for another possible match of the Arbno pattern. -- As noted in the section on stack checking, the stack count (and -- hence the stack check) for a pattern includes only one iteration -- of the Arbno pattern. To make sure that multiple iterations do not -- overflow the stack, the Arbno node saves the stack count required -- by a single iteration, and the Concat function increments this to -- include stack entries required by any successor. The PC_Arbno_Y -- node uses this count to ensure that sufficient stack remains -- before proceeding after matching each new instance. -- The node numbering of the constituent pattern P is not affected. -- Where N is the number of nodes in P, the Y node is numbered N + 1, -- the E node is N + 2, and the X node is N + 3. ---------------------- -- Assign Immediate -- ---------------------- -- Immediate assignment (P * V) constructs the following structure -- +---+ +---+ +---+ -- | E |---->| P |---->| A |----> -- +---+ +---+ +---+ -- Here E is the PC_R_Enter node which matches null and creates two -- stack entries. The first is a special entry whose node field is -- not used at all, and whose cursor field has the initial cursor. -- The second entry corresponds to a standard new region action. A -- PC_R_Remove node is stacked, whose cursor field is used to store -- the outer stack base, and the stack base is reset to point to -- this PC_R_Remove node. Then the pattern P is matched, and it can -- make history stack entries in the normal manner, so now the stack -- looks like: -- (stack entries made before assign pattern) -- (Special entry, node field not used, -- used only to save initial cursor) -- (PC_R_Remove entry, "cursor" value is (negative) <-- Stack Base -- saved base value for the enclosing region) -- (stack entries made by matching P) -- If the match of P fails, then the PC_R_Remove entry is popped -- and it removes both itself and the special entry underneath it, -- restores the outer stack base, and signals failure. -- If the match of P succeeds, then node A, which is the actual -- PC_Assign_Imm node, executes the assignment (using the stack -- base to locate the entry with the saved starting cursor value), -- and the pops the inner region. There are two possibilities, if -- matching P left no stack entries, then all traces of the inner -- region can be removed. If there are stack entries, then we push -- an PC_Region_Replace stack entry whose "cursor" value is the -- inner stack base value, and then restore the outer stack base -- value, so the stack looks like: -- (stack entries made before assign pattern) -- (Special entry, node field not used, -- used only to save initial cursor) -- (PC_R_Remove entry, "cursor" value is (negative) -- saved base value for the enclosing region) -- (stack entries made by matching P) -- (PC_Region_Replace entry, "cursor" value is the (negative) -- stack pointer value referencing the PC_R_Remove entry). -- If a subsequent failure occurs, the PC_Region_Replace node restores -- the inner stack base value and signals failure to explore rematches -- of the pattern P. -- The node numbering of the constituent pattern P is not affected. -- Where N is the number of nodes in P, the A node is numbered N + 1, -- and the E node is N + 2. --------------------- -- Assign On Match -- --------------------- -- The assign on match (**) pattern is quite similar to the assign -- immediate pattern, except that the actual assignment has to be -- delayed. The following structure is constructed: -- +---+ +---+ +---+ -- | E |---->| P |---->| A |----> -- +---+ +---+ +---+ -- The operation of this pattern is identical to that described above -- for deferred assignment, up to the point where P has been matched. -- The A node, which is the PC_Assign_OnM node first pushes a -- PC_Assign node onto the history stack. This node saves the ending -- cursor and acts as a flag for the final assignment, as further -- described below. -- It then stores a pointer to itself in the special entry node field. -- This was otherwise unused, and is now used to retrieve the address -- of the variable to be assigned at the end of the pattern. -- After that the inner region is terminated in the usual manner, -- by stacking a PC_R_Restore entry as described for the assign -- immediate case. Note that the optimization of completely -- removing the inner region does not happen in this case, since -- we have at least one stack entry (the PC_Assign one we just made). -- The stack now looks like: -- (stack entries made before assign pattern) -- (Special entry, node points to copy of -- the PC_Assign_OnM node, and the -- cursor field saves the initial cursor). -- (PC_R_Remove entry, "cursor" value is (negative) -- saved base value for the enclosing region) -- (stack entries made by matching P) -- (PC_Assign entry, saves final cursor) -- (PC_Region_Replace entry, "cursor" value is (negative) -- stack pointer value referencing the PC_R_Remove entry). -- If a subsequent failure causes the PC_Assign node to execute it -- simply removes itself and propagates the failure. -- If the match succeeds, then the history stack is scanned for -- PC_Assign nodes, and the assignments are executed (examination -- of the above diagram will show that all the necessary data is -- at hand for the assignment). -- To optimize the common case where no assign-on-match operations -- are present, a global flag Assign_OnM is maintained which is -- initialize to False, and gets set True as part of the execution -- of the PC_Assign_OnM node. The scan of the history stack for -- PC_Assign entries is done only if this flag is set. -- The node numbering of the constituent pattern P is not affected. -- Where N is the number of nodes in P, the A node is numbered N + 1, -- and the E node is N + 2. --------- -- Bal -- --------- -- Bal builds a single node: -- +---+ -- | B |----> -- +---+ -- The node B is the PC_Bal node which matches a parentheses balanced -- string, starting at the current cursor position. It then updates -- the cursor past this matched string, and stacks a pointer to itself -- with this updated cursor value on the history stack, to extend the -- matched string on a subsequent failure. -- Since this is a single node it is numbered 1 (the reason we include -- it in the compound patterns section is that it backtracks). ------------ -- BreakX -- ------------ -- BreakX builds the structure -- +---+ +---+ -- | B |---->| A |----> -- +---+ +---+ -- ^ . -- | . -- | +---+ -- +<------| X | -- +---+ -- Here the B node is the BreakX_xx node that performs a normal Break -- function. The A node is an alternative (PC_Alt) node that matches -- null, but stacks a pointer to node X (the PC_BreakX_X node) which -- extends the match one character (to eat up the previously detected -- break character), and then rematches the break. -- The B node is numbered 3, the alternative node is 1, and the X -- node is 2. ----------- -- Fence -- ----------- -- Fence builds a single node: -- +---+ -- | F |----> -- +---+ -- The element F, PC_Fence, matches null, and stacks a pointer to a -- PC_Cancel element which will abort the match on a subsequent failure. -- Since this is a single element it is numbered 1 (the reason we -- include it in the compound patterns section is that it backtracks). -------------------- -- Fence Function -- -------------------- -- A call to the Fence function builds the structure: -- +---+ +---+ +---+ -- | E |---->| P |---->| X |----> -- +---+ +---+ +---+ -- Here E is the PC_R_Enter node which matches null and creates two -- stack entries. The first is a special entry which is not used at -- all in the fence case (it is present merely for uniformity with -- other cases of region enter operations). -- The second entry corresponds to a standard new region action. A -- PC_R_Remove node is stacked, whose cursor field is used to store -- the outer stack base, and the stack base is reset to point to -- this PC_R_Remove node. Then the pattern P is matched, and it can -- make history stack entries in the normal manner, so now the stack -- looks like: -- (stack entries made before fence pattern) -- (Special entry, not used at all) -- (PC_R_Remove entry, "cursor" value is (negative) <-- Stack Base -- saved base value for the enclosing region) -- (stack entries made by matching P) -- If the match of P fails, then the PC_R_Remove entry is popped -- and it removes both itself and the special entry underneath it, -- restores the outer stack base, and signals failure. -- If the match of P succeeds, then node X, the PC_Fence_X node, gets -- control. One might be tempted to think that at this point, the -- history stack entries made by matching P can just be removed since -- they certainly are not going to be used for rematching (that is -- whole point of Fence after all!) However, this is wrong, because -- it would result in the loss of possible assign-on-match entries -- for deferred pattern assignments. -- Instead what we do is to make a special entry whose node references -- PC_Fence_Y, and whose cursor saves the inner stack base value, i.e. -- the pointer to the PC_R_Remove entry. Then the outer stack base -- pointer is restored, so the stack looks like: -- (stack entries made before assign pattern) -- (Special entry, not used at all) -- (PC_R_Remove entry, "cursor" value is (negative) -- saved base value for the enclosing region) -- (stack entries made by matching P) -- (PC_Fence_Y entry, "cursor" value is (negative) stack -- pointer value referencing the PC_R_Remove entry). -- If a subsequent failure occurs, then the PC_Fence_Y entry removes -- the entire inner region, including all entries made by matching P, -- and alternatives prior to the Fence pattern are sought. -- The node numbering of the constituent pattern P is not affected. -- Where N is the number of nodes in P, the X node is numbered N + 1, -- and the E node is N + 2. ------------- -- Succeed -- ------------- -- Succeed builds a single node: -- +---+ -- | S |----> -- +---+ -- The node S is the PC_Succeed node which matches null, and stacks -- a pointer to itself on the history stack, so that a subsequent -- failure repeats the same match. -- Since this is a single node it is numbered 1 (the reason we include -- it in the compound patterns section is that it backtracks). --------------------- -- Write Immediate -- --------------------- -- The structure built for a write immediate operation (P * F, where -- F is a file access value) is: -- +---+ +---+ +---+ -- | E |---->| P |---->| W |----> -- +---+ +---+ +---+ -- Here E is the PC_R_Enter node and W is the PC_Write_Imm node. The -- handling is identical to that described above for Assign Immediate, -- except that at the point where a successful match occurs, the matched -- substring is written to the referenced file. -- The node numbering of the constituent pattern P is not affected. -- Where N is the number of nodes in P, the W node is numbered N + 1, -- and the E node is N + 2. -------------------- -- Write On Match -- -------------------- -- The structure built for a write on match operation (P ** F, where -- F is a file access value) is: -- +---+ +---+ +---+ -- | E |---->| P |---->| W |----> -- +---+ +---+ +---+ -- Here E is the PC_R_Enter node and W is the PC_Write_OnM node. The -- handling is identical to that described above for Assign On Match, -- except that at the point where a successful match has completed, -- the matched substring is written to the referenced file. -- The node numbering of the constituent pattern P is not affected. -- Where N is the number of nodes in P, the W node is numbered N + 1, -- and the E node is N + 2. ----------------------- -- Constant Patterns -- ----------------------- -- The following pattern elements are referenced only from the pattern -- history stack. In each case the processing for the pattern element -- results in pattern match abort, or further failure, so there is no -- need for a successor and no need for a node number CP_Assign : aliased PE := (PC_Assign, 0, N); CP_Cancel : aliased PE := (PC_Cancel, 0, N); CP_Fence_Y : aliased PE := (PC_Fence_Y, 0, N); CP_R_Remove : aliased PE := (PC_R_Remove, 0, N); CP_R_Restore : aliased PE := (PC_R_Restore, 0, N); ----------------------- -- Local Subprograms -- ----------------------- function Alternate (L, R : PE_Ptr) return PE_Ptr; function "or" (L, R : PE_Ptr) return PE_Ptr renames Alternate; -- Build pattern structure corresponding to the alternation of L, R. -- (i.e. try to match L, and if that fails, try to match R). function Arbno_Simple (P : PE_Ptr) return PE_Ptr; -- Build simple Arbno pattern, P is a pattern that is guaranteed to -- match at least one character if it succeeds and to require no -- stack entries under all circumstances. The result returned is -- a simple Arbno structure as previously described. function Bracket (E, P, A : PE_Ptr) return PE_Ptr; -- Given two single node pattern elements E and A, and a (possible -- complex) pattern P, construct the concatenation E-->P-->A and -- return a pointer to E. The concatenation does not affect the -- node numbering in P. A has a number one higher than the maximum -- number in P, and E has a number two higher than the maximum -- number in P (see for example the Assign_Immediate structure to -- understand a typical use of this function). function BreakX_Make (B : PE_Ptr) return Pattern; -- Given a pattern element for a Break pattern, returns the -- corresponding BreakX compound pattern structure. function Concat (L, R : PE_Ptr; Incr : Natural) return PE_Ptr; -- Creates a pattern element that represents a concatenation of the -- two given pattern elements (i.e. the pattern L followed by R). -- The result returned is always the same as L, but the pattern -- referenced by L is modified to have R as a successor. This -- procedure does not copy L or R, so if a copy is required, it -- is the responsibility of the caller. The Incr parameter is an -- amount to be added to the Nat field of any P_Arbno_Y node that is -- in the left operand, it represents the additional stack space -- required by the right operand. function C_To_PE (C : PChar) return PE_Ptr; -- Given a character, constructs a pattern element that matches -- the single character. function Copy (P : PE_Ptr) return PE_Ptr; -- Creates a copy of the pattern element referenced by the given -- pattern element reference. This is a deep copy, which means that -- it follows the Next and Alt pointers. function Image (P : PE_Ptr) return String; -- Returns the image of the address of the referenced pattern element. -- This is equivalent to Image (To_Address (P)); function Is_In (C : Character; Str : String) return Boolean; pragma Inline (Is_In); -- Determines if the character C is in string Str procedure Logic_Error; -- Called to raise Program_Error with an appropriate message if an -- internal logic error is detected. function Str_BF (A : Boolean_Func) return String; function Str_FP (A : File_Ptr) return String; function Str_NF (A : Natural_Func) return String; function Str_NP (A : Natural_Ptr) return String; function Str_PP (A : Pattern_Ptr) return String; function Str_VF (A : VString_Func) return String; function Str_VP (A : VString_Ptr) return String; -- These are debugging routines, which return a representation of the -- given access value (they are called only by Image and Dump) procedure Set_Successor (Pat : PE_Ptr; Succ : PE_Ptr); -- Adjusts all EOP pointers in Pat to point to Succ. No other changes -- are made. In particular, Succ is unchanged, and no index numbers -- are modified. Note that Pat may not be equal to EOP on entry. function S_To_PE (Str : PString) return PE_Ptr; -- Given a string, constructs a pattern element that matches the string procedure Uninitialized_Pattern; pragma No_Return (Uninitialized_Pattern); -- Called to raise Program_Error with an appropriate error message if -- an uninitialized pattern is used in any pattern construction or -- pattern matching operation. procedure XMatch (Subject : String; Pat_P : PE_Ptr; Pat_S : Natural; Start : out Natural; Stop : out Natural); -- This is the common pattern match routine. It is passed a string and -- a pattern, and it indicates success or failure, and on success the -- section of the string matched. It does not perform any assignments -- to the subject string, so pattern replacement is for the caller. -- -- Subject The subject string. The lower bound is always one. In the -- Match procedures, it is fine to use strings whose lower bound -- is not one, but we perform a one time conversion before the -- call to XMatch, so that XMatch does not have to be bothered -- with strange lower bounds. -- -- Pat_P Points to initial pattern element of pattern to be matched -- -- Pat_S Maximum required stack entries for pattern to be matched -- -- Start If match is successful, starting index of matched section. -- This value is always non-zero. A value of zero is used to -- indicate a failed match. -- -- Stop If match is successful, ending index of matched section. -- This can be zero if we match the null string at the start, -- in which case Start is set to zero, and Stop to one. If the -- Match fails, then the contents of Stop is undefined. procedure XMatchD (Subject : String; Pat_P : PE_Ptr; Pat_S : Natural; Start : out Natural; Stop : out Natural); -- Identical in all respects to XMatch, except that trace information is -- output on Standard_Output during execution of the match. This is the -- version that is called if the original Match call has Debug => True. --------- -- "&" -- --------- function "&" (L : PString; R : Pattern) return Pattern is begin return (AFC with R.Stk, Concat (S_To_PE (L), Copy (R.P), R.Stk)); end "&"; function "&" (L : Pattern; R : PString) return Pattern is begin return (AFC with L.Stk, Concat (Copy (L.P), S_To_PE (R), 0)); end "&"; function "&" (L : PChar; R : Pattern) return Pattern is begin return (AFC with R.Stk, Concat (C_To_PE (L), Copy (R.P), R.Stk)); end "&"; function "&" (L : Pattern; R : PChar) return Pattern is begin return (AFC with L.Stk, Concat (Copy (L.P), C_To_PE (R), 0)); end "&"; function "&" (L : Pattern; R : Pattern) return Pattern is begin return (AFC with L.Stk + R.Stk, Concat (Copy (L.P), Copy (R.P), R.Stk)); end "&"; --------- -- "*" -- --------- -- Assign immediate -- +---+ +---+ +---+ -- | E |---->| P |---->| A |----> -- +---+ +---+ +---+ -- The node numbering of the constituent pattern P is not affected. -- Where N is the number of nodes in P, the A node is numbered N + 1, -- and the E node is N + 2. function "*" (P : Pattern; Var : VString_Var) return Pattern is Pat : constant PE_Ptr := Copy (P.P); E : constant PE_Ptr := new PE'(PC_R_Enter, 0, EOP); A : constant PE_Ptr := new PE'(PC_Assign_Imm, 0, EOP, Var'Unrestricted_Access); begin return (AFC with P.Stk + 3, Bracket (E, Pat, A)); end "*"; function "*" (P : PString; Var : VString_Var) return Pattern is Pat : constant PE_Ptr := S_To_PE (P); E : constant PE_Ptr := new PE'(PC_R_Enter, 0, EOP); A : constant PE_Ptr := new PE'(PC_Assign_Imm, 0, EOP, Var'Unrestricted_Access); begin return (AFC with 3, Bracket (E, Pat, A)); end "*"; function "*" (P : PChar; Var : VString_Var) return Pattern is Pat : constant PE_Ptr := C_To_PE (P); E : constant PE_Ptr := new PE'(PC_R_Enter, 0, EOP); A : constant PE_Ptr := new PE'(PC_Assign_Imm, 0, EOP, Var'Unrestricted_Access); begin return (AFC with 3, Bracket (E, Pat, A)); end "*"; -- Write immediate -- +---+ +---+ +---+ -- | E |---->| P |---->| W |----> -- +---+ +---+ +---+ -- The node numbering of the constituent pattern P is not affected. -- Where N is the number of nodes in P, the W node is numbered N + 1, -- and the E node is N + 2. function "*" (P : Pattern; Fil : File_Access) return Pattern is Pat : constant PE_Ptr := Copy (P.P); E : constant PE_Ptr := new PE'(PC_R_Enter, 0, EOP); W : constant PE_Ptr := new PE'(PC_Write_Imm, 0, EOP, Fil); begin return (AFC with 3, Bracket (E, Pat, W)); end "*"; function "*" (P : PString; Fil : File_Access) return Pattern is Pat : constant PE_Ptr := S_To_PE (P); E : constant PE_Ptr := new PE'(PC_R_Enter, 0, EOP); W : constant PE_Ptr := new PE'(PC_Write_Imm, 0, EOP, Fil); begin return (AFC with 3, Bracket (E, Pat, W)); end "*"; function "*" (P : PChar; Fil : File_Access) return Pattern is Pat : constant PE_Ptr := C_To_PE (P); E : constant PE_Ptr := new PE'(PC_R_Enter, 0, EOP); W : constant PE_Ptr := new PE'(PC_Write_Imm, 0, EOP, Fil); begin return (AFC with 3, Bracket (E, Pat, W)); end "*"; ---------- -- "**" -- ---------- -- Assign on match -- +---+ +---+ +---+ -- | E |---->| P |---->| A |----> -- +---+ +---+ +---+ -- The node numbering of the constituent pattern P is not affected. -- Where N is the number of nodes in P, the A node is numbered N + 1, -- and the E node is N + 2. function "**" (P : Pattern; Var : VString_Var) return Pattern is Pat : constant PE_Ptr := Copy (P.P); E : constant PE_Ptr := new PE'(PC_R_Enter, 0, EOP); A : constant PE_Ptr := new PE'(PC_Assign_OnM, 0, EOP, Var'Unrestricted_Access); begin return (AFC with P.Stk + 3, Bracket (E, Pat, A)); end "**"; function "**" (P : PString; Var : VString_Var) return Pattern is Pat : constant PE_Ptr := S_To_PE (P); E : constant PE_Ptr := new PE'(PC_R_Enter, 0, EOP); A : constant PE_Ptr := new PE'(PC_Assign_OnM, 0, EOP, Var'Unrestricted_Access); begin return (AFC with 3, Bracket (E, Pat, A)); end "**"; function "**" (P : PChar; Var : VString_Var) return Pattern is Pat : constant PE_Ptr := C_To_PE (P); E : constant PE_Ptr := new PE'(PC_R_Enter, 0, EOP); A : constant PE_Ptr := new PE'(PC_Assign_OnM, 0, EOP, Var'Unrestricted_Access); begin return (AFC with 3, Bracket (E, Pat, A)); end "**"; -- Write on match -- +---+ +---+ +---+ -- | E |---->| P |---->| W |----> -- +---+ +---+ +---+ -- The node numbering of the constituent pattern P is not affected. -- Where N is the number of nodes in P, the W node is numbered N + 1, -- and the E node is N + 2. function "**" (P : Pattern; Fil : File_Access) return Pattern is Pat : constant PE_Ptr := Copy (P.P); E : constant PE_Ptr := new PE'(PC_R_Enter, 0, EOP); W : constant PE_Ptr := new PE'(PC_Write_OnM, 0, EOP, Fil); begin return (AFC with P.Stk + 3, Bracket (E, Pat, W)); end "**"; function "**" (P : PString; Fil : File_Access) return Pattern is Pat : constant PE_Ptr := S_To_PE (P); E : constant PE_Ptr := new PE'(PC_R_Enter, 0, EOP); W : constant PE_Ptr := new PE'(PC_Write_OnM, 0, EOP, Fil); begin return (AFC with 3, Bracket (E, Pat, W)); end "**"; function "**" (P : PChar; Fil : File_Access) return Pattern is Pat : constant PE_Ptr := C_To_PE (P); E : constant PE_Ptr := new PE'(PC_R_Enter, 0, EOP); W : constant PE_Ptr := new PE'(PC_Write_OnM, 0, EOP, Fil); begin return (AFC with 3, Bracket (E, Pat, W)); end "**"; --------- -- "+" -- --------- function "+" (Str : VString_Var) return Pattern is begin return (AFC with 0, new PE'(PC_String_VP, 1, EOP, Str'Unrestricted_Access)); end "+"; function "+" (Str : VString_Func) return Pattern is begin return (AFC with 0, new PE'(PC_String_VF, 1, EOP, Str)); end "+"; function "+" (P : Pattern_Var) return Pattern is begin return (AFC with 3, new PE'(PC_Rpat, 1, EOP, P'Unrestricted_Access)); end "+"; function "+" (P : Boolean_Func) return Pattern is begin return (AFC with 3, new PE'(PC_Pred_Func, 1, EOP, P)); end "+"; ---------- -- "or" -- ---------- function "or" (L : PString; R : Pattern) return Pattern is begin return (AFC with R.Stk + 1, S_To_PE (L) or Copy (R.P)); end "or"; function "or" (L : Pattern; R : PString) return Pattern is begin return (AFC with L.Stk + 1, Copy (L.P) or S_To_PE (R)); end "or"; function "or" (L : PString; R : PString) return Pattern is begin return (AFC with 1, S_To_PE (L) or S_To_PE (R)); end "or"; function "or" (L : Pattern; R : Pattern) return Pattern is begin return (AFC with Natural'Max (L.Stk, R.Stk) + 1, Copy (L.P) or Copy (R.P)); end "or"; function "or" (L : PChar; R : Pattern) return Pattern is begin return (AFC with 1, C_To_PE (L) or Copy (R.P)); end "or"; function "or" (L : Pattern; R : PChar) return Pattern is begin return (AFC with 1, Copy (L.P) or C_To_PE (R)); end "or"; function "or" (L : PChar; R : PChar) return Pattern is begin return (AFC with 1, C_To_PE (L) or C_To_PE (R)); end "or"; function "or" (L : PString; R : PChar) return Pattern is begin return (AFC with 1, S_To_PE (L) or C_To_PE (R)); end "or"; function "or" (L : PChar; R : PString) return Pattern is begin return (AFC with 1, C_To_PE (L) or S_To_PE (R)); end "or"; ------------ -- Adjust -- ------------ -- No two patterns share the same pattern elements, so the adjust -- procedure for a Pattern assignment must do a deep copy of the -- pattern element structure. procedure Adjust (Object : in out Pattern) is begin Object.P := Copy (Object.P); end Adjust; --------------- -- Alternate -- --------------- function Alternate (L, R : PE_Ptr) return PE_Ptr is begin -- If the left pattern is null, then we just add the alternation -- node with an index one greater than the right hand pattern. if L = EOP then return new PE'(PC_Alt, R.Index + 1, EOP, R); -- If the left pattern is non-null, then build a reference vector -- for its elements, and adjust their index values to accommodate -- the right hand elements. Then add the alternation node. else declare Refs : Ref_Array (1 .. L.Index); begin Build_Ref_Array (L, Refs); for J in Refs'Range loop Refs (J).Index := Refs (J).Index + R.Index; end loop; end; return new PE'(PC_Alt, L.Index + 1, L, R); end if; end Alternate; --------- -- Any -- --------- function Any (Str : String) return Pattern is begin return (AFC with 0, new PE'(PC_Any_CS, 1, EOP, To_Set (Str))); end Any; function Any (Str : VString) return Pattern is begin return Any (S (Str)); end Any; function Any (Str : Character) return Pattern is begin return (AFC with 0, new PE'(PC_Any_CH, 1, EOP, Str)); end Any; function Any (Str : Character_Set) return Pattern is begin return (AFC with 0, new PE'(PC_Any_CS, 1, EOP, Str)); end Any; function Any (Str : not null access VString) return Pattern is begin return (AFC with 0, new PE'(PC_Any_VP, 1, EOP, VString_Ptr (Str))); end Any; function Any (Str : VString_Func) return Pattern is begin return (AFC with 0, new PE'(PC_Any_VF, 1, EOP, Str)); end Any; --------- -- Arb -- --------- -- +---+ -- | X |----> -- +---+ -- . -- . -- +---+ -- | Y |----> -- +---+ -- The PC_Arb_X element is numbered 2, and the PC_Arb_Y element is 1 function Arb return Pattern is Y : constant PE_Ptr := new PE'(PC_Arb_Y, 1, EOP); X : constant PE_Ptr := new PE'(PC_Arb_X, 2, EOP, Y); begin return (AFC with 1, X); end Arb; ----------- -- Arbno -- ----------- function Arbno (P : PString) return Pattern is begin if P'Length = 0 then return (AFC with 0, EOP); else return (AFC with 0, Arbno_Simple (S_To_PE (P))); end if; end Arbno; function Arbno (P : PChar) return Pattern is begin return (AFC with 0, Arbno_Simple (C_To_PE (P))); end Arbno; function Arbno (P : Pattern) return Pattern is Pat : constant PE_Ptr := Copy (P.P); begin if P.Stk = 0 and then OK_For_Simple_Arbno (Pat.Pcode) then return (AFC with 0, Arbno_Simple (Pat)); end if; -- This is the complex case, either the pattern makes stack entries -- or it is possible for the pattern to match the null string (more -- accurately, we don't know that this is not the case). -- +--------------------------+ -- | ^ -- V | -- +---+ | -- | X |----> | -- +---+ | -- . | -- . | -- +---+ +---+ +---+ | -- | E |---->| P |---->| Y |--->+ -- +---+ +---+ +---+ -- The node numbering of the constituent pattern P is not affected. -- Where N is the number of nodes in P, the Y node is numbered N + 1, -- the E node is N + 2, and the X node is N + 3. declare E : constant PE_Ptr := new PE'(PC_R_Enter, 0, EOP); X : constant PE_Ptr := new PE'(PC_Arbno_X, 0, EOP, E); Y : constant PE_Ptr := new PE'(PC_Arbno_Y, 0, X, P.Stk + 3); EPY : constant PE_Ptr := Bracket (E, Pat, Y); begin X.Alt := EPY; X.Index := EPY.Index + 1; return (AFC with P.Stk + 3, X); end; end Arbno; ------------------ -- Arbno_Simple -- ------------------ -- +-------------+ -- | ^ -- V | -- +---+ | -- | S |----> | -- +---+ | -- . | -- . | -- +---+ | -- | P |---------->+ -- +---+ -- The node numbering of the constituent pattern P is not affected. -- The S node has a node number of P.Index + 1. -- Note that we know that P cannot be EOP, because a null pattern -- does not meet the requirements for simple Arbno. function Arbno_Simple (P : PE_Ptr) return PE_Ptr is S : constant PE_Ptr := new PE'(PC_Arbno_S, P.Index + 1, EOP, P); begin Set_Successor (P, S); return S; end Arbno_Simple; --------- -- Bal -- --------- function Bal return Pattern is begin return (AFC with 1, new PE'(PC_Bal, 1, EOP)); end Bal; ------------- -- Bracket -- ------------- function Bracket (E, P, A : PE_Ptr) return PE_Ptr is begin if P = EOP then E.Pthen := A; E.Index := 2; A.Index := 1; else E.Pthen := P; Set_Successor (P, A); E.Index := P.Index + 2; A.Index := P.Index + 1; end if; return E; end Bracket; ----------- -- Break -- ----------- function Break (Str : String) return Pattern is begin return (AFC with 0, new PE'(PC_Break_CS, 1, EOP, To_Set (Str))); end Break; function Break (Str : VString) return Pattern is begin return Break (S (Str)); end Break; function Break (Str : Character) return Pattern is begin return (AFC with 0, new PE'(PC_Break_CH, 1, EOP, Str)); end Break; function Break (Str : Character_Set) return Pattern is begin return (AFC with 0, new PE'(PC_Break_CS, 1, EOP, Str)); end Break; function Break (Str : not null access VString) return Pattern is begin return (AFC with 0, new PE'(PC_Break_VP, 1, EOP, Str.all'Unchecked_Access)); end Break; function Break (Str : VString_Func) return Pattern is begin return (AFC with 0, new PE'(PC_Break_VF, 1, EOP, Str)); end Break; ------------ -- BreakX -- ------------ function BreakX (Str : String) return Pattern is begin return BreakX_Make (new PE'(PC_BreakX_CS, 3, N, To_Set (Str))); end BreakX; function BreakX (Str : VString) return Pattern is begin return BreakX (S (Str)); end BreakX; function BreakX (Str : Character) return Pattern is begin return BreakX_Make (new PE'(PC_BreakX_CH, 3, N, Str)); end BreakX; function BreakX (Str : Character_Set) return Pattern is begin return BreakX_Make (new PE'(PC_BreakX_CS, 3, N, Str)); end BreakX; function BreakX (Str : not null access VString) return Pattern is begin return BreakX_Make (new PE'(PC_BreakX_VP, 3, N, VString_Ptr (Str))); end BreakX; function BreakX (Str : VString_Func) return Pattern is begin return BreakX_Make (new PE'(PC_BreakX_VF, 3, N, Str)); end BreakX; ----------------- -- BreakX_Make -- ----------------- -- +---+ +---+ -- | B |---->| A |----> -- +---+ +---+ -- ^ . -- | . -- | +---+ -- +<------| X | -- +---+ -- The B node is numbered 3, the alternative node is 1, and the X -- node is 2. function BreakX_Make (B : PE_Ptr) return Pattern is X : constant PE_Ptr := new PE'(PC_BreakX_X, 2, B); A : constant PE_Ptr := new PE'(PC_Alt, 1, EOP, X); begin B.Pthen := A; return (AFC with 2, B); end BreakX_Make; --------------------- -- Build_Ref_Array -- --------------------- procedure Build_Ref_Array (E : PE_Ptr; RA : out Ref_Array) is procedure Record_PE (E : PE_Ptr); -- Record given pattern element if not already recorded in RA, -- and also record any referenced pattern elements recursively. --------------- -- Record_PE -- --------------- procedure Record_PE (E : PE_Ptr) is begin PutD (" Record_PE called with PE_Ptr = " & Image (E)); if E = EOP or else RA (E.Index) /= null then Put_LineD (", nothing to do"); return; else Put_LineD (", recording" & IndexT'Image (E.Index)); RA (E.Index) := E; Record_PE (E.Pthen); if E.Pcode in PC_Has_Alt then Record_PE (E.Alt); end if; end if; end Record_PE; -- Start of processing for Build_Ref_Array begin New_LineD; Put_LineD ("Entering Build_Ref_Array"); Record_PE (E); New_LineD; end Build_Ref_Array; ------------- -- C_To_PE -- ------------- function C_To_PE (C : PChar) return PE_Ptr is begin return new PE'(PC_Char, 1, EOP, C); end C_To_PE; ------------ -- Cancel -- ------------ function Cancel return Pattern is begin return (AFC with 0, new PE'(PC_Cancel, 1, EOP)); end Cancel; ------------ -- Concat -- ------------ -- Concat needs to traverse the left operand performing the following -- set of fixups: -- a) Any successor pointers (Pthen fields) that are set to EOP are -- reset to point to the second operand. -- b) Any PC_Arbno_Y node has its stack count field incremented -- by the parameter Incr provided for this purpose. -- d) Num fields of all pattern elements in the left operand are -- adjusted to include the elements of the right operand. -- Note: we do not use Set_Successor in the processing for Concat, since -- there is no point in doing two traversals, we may as well do everything -- at the same time. function Concat (L, R : PE_Ptr; Incr : Natural) return PE_Ptr is begin if L = EOP then return R; elsif R = EOP then return L; else declare Refs : Ref_Array (1 .. L.Index); -- We build a reference array for L whose N'th element points to -- the pattern element of L whose original Index value is N. P : PE_Ptr; begin Build_Ref_Array (L, Refs); for J in Refs'Range loop P := Refs (J); P.Index := P.Index + R.Index; if P.Pcode = PC_Arbno_Y then P.Nat := P.Nat + Incr; end if; if P.Pthen = EOP then P.Pthen := R; end if; if P.Pcode in PC_Has_Alt and then P.Alt = EOP then P.Alt := R; end if; end loop; end; return L; end if; end Concat; ---------- -- Copy -- ---------- function Copy (P : PE_Ptr) return PE_Ptr is begin if P = null then Uninitialized_Pattern; else declare Refs : Ref_Array (1 .. P.Index); -- References to elements in P, indexed by Index field Copy : Ref_Array (1 .. P.Index); -- Holds copies of elements of P, indexed by Index field E : PE_Ptr; begin Build_Ref_Array (P, Refs); -- Now copy all nodes for J in Refs'Range loop Copy (J) := new PE'(Refs (J).all); end loop; -- Adjust all internal references for J in Copy'Range loop E := Copy (J); -- Adjust successor pointer to point to copy if E.Pthen /= EOP then E.Pthen := Copy (E.Pthen.Index); end if; -- Adjust Alt pointer if there is one to point to copy if E.Pcode in PC_Has_Alt and then E.Alt /= EOP then E.Alt := Copy (E.Alt.Index); end if; -- Copy referenced string if E.Pcode = PC_String then E.Str := new String'(E.Str.all); end if; end loop; return Copy (P.Index); end; end if; end Copy; ---------- -- Dump -- ---------- procedure Dump (P : Pattern) is subtype Count is Ada.Text_IO.Count; Scol : Count; -- Used to keep track of column in dump output Refs : Ref_Array (1 .. P.P.Index); -- We build a reference array whose N'th element points to the -- pattern element whose Index value is N. Cols : Natural := 2; -- Number of columns used for pattern numbers, minimum is 2 E : PE_Ptr; procedure Write_Node_Id (E : PE_Ptr); -- Writes out a string identifying the given pattern element ------------------- -- Write_Node_Id -- ------------------- procedure Write_Node_Id (E : PE_Ptr) is begin if E = EOP then Put ("EOP"); for J in 4 .. Cols loop Put (' '); end loop; else declare Str : String (1 .. Cols); N : Natural := Natural (E.Index); begin Put ("#"); for J in reverse Str'Range loop Str (J) := Character'Val (48 + N mod 10); N := N / 10; end loop; Put (Str); end; end if; end Write_Node_Id; -- Start of processing for Dump begin New_Line; Put ("Pattern Dump Output (pattern at " & Image (P'Address) & ", S = " & Natural'Image (P.Stk) & ')'); Scol := Col; New_Line; while Col < Scol loop Put ('-'); end loop; New_Line; -- If uninitialized pattern, dump line and we are done if P.P = null then Put_Line ("Uninitialized pattern value"); return; end if; -- If null pattern, just dump it and we are all done if P.P = EOP then Put_Line ("EOP (null pattern)"); return; end if; Build_Ref_Array (P.P, Refs); -- Set number of columns required for node numbers while 10 ** Cols - 1 < Integer (P.P.Index) loop Cols := Cols + 1; end loop; -- Now dump the nodes in reverse sequence. We output them in reverse -- sequence since this corresponds to the natural order used to -- construct the patterns. for J in reverse Refs'Range loop E := Refs (J); Write_Node_Id (E); Set_Col (Count (Cols) + 4); Put (Image (E)); Put (" "); Put (Pattern_Code'Image (E.Pcode)); Put (" "); Set_Col (21 + Count (Cols) + Address_Image_Length); Write_Node_Id (E.Pthen); Set_Col (24 + 2 * Count (Cols) + Address_Image_Length); case E.Pcode is when PC_Alt | PC_Arb_X | PC_Arbno_S | PC_Arbno_X => Write_Node_Id (E.Alt); when PC_Rpat => Put (Str_PP (E.PP)); when PC_Pred_Func => Put (Str_BF (E.BF)); when PC_Assign_Imm | PC_Assign_OnM | PC_Any_VP | PC_Break_VP | PC_BreakX_VP | PC_NotAny_VP | PC_NSpan_VP | PC_Span_VP | PC_String_VP => Put (Str_VP (E.VP)); when PC_Write_Imm | PC_Write_OnM => Put (Str_FP (E.FP)); when PC_String => Put (Image (E.Str.all)); when PC_String_2 => Put (Image (E.Str2)); when PC_String_3 => Put (Image (E.Str3)); when PC_String_4 => Put (Image (E.Str4)); when PC_String_5 => Put (Image (E.Str5)); when PC_String_6 => Put (Image (E.Str6)); when PC_Setcur => Put (Str_NP (E.Var)); when PC_Any_CH | PC_Break_CH | PC_BreakX_CH | PC_Char | PC_NotAny_CH | PC_NSpan_CH | PC_Span_CH => Put (''' & E.Char & '''); when PC_Any_CS | PC_Break_CS | PC_BreakX_CS | PC_NotAny_CS | PC_NSpan_CS | PC_Span_CS => Put ('"' & To_Sequence (E.CS) & '"'); when PC_Arbno_Y | PC_Len_Nat | PC_Pos_Nat | PC_RPos_Nat | PC_RTab_Nat | PC_Tab_Nat => Put (S (E.Nat)); when PC_Pos_NF | PC_Len_NF | PC_RPos_NF | PC_RTab_NF | PC_Tab_NF => Put (Str_NF (E.NF)); when PC_Pos_NP | PC_Len_NP | PC_RPos_NP | PC_RTab_NP | PC_Tab_NP => Put (Str_NP (E.NP)); when PC_Any_VF | PC_Break_VF | PC_BreakX_VF | PC_NotAny_VF | PC_NSpan_VF | PC_Span_VF | PC_String_VF => Put (Str_VF (E.VF)); when others => null; end case; New_Line; end loop; New_Line; end Dump; ---------- -- Fail -- ---------- function Fail return Pattern is begin return (AFC with 0, new PE'(PC_Fail, 1, EOP)); end Fail; ----------- -- Fence -- ----------- -- Simple case function Fence return Pattern is begin return (AFC with 1, new PE'(PC_Fence, 1, EOP)); end Fence; -- Function case -- +---+ +---+ +---+ -- | E |---->| P |---->| X |----> -- +---+ +---+ +---+ -- The node numbering of the constituent pattern P is not affected. -- Where N is the number of nodes in P, the X node is numbered N + 1, -- and the E node is N + 2. function Fence (P : Pattern) return Pattern is Pat : constant PE_Ptr := Copy (P.P); E : constant PE_Ptr := new PE'(PC_R_Enter, 0, EOP); X : constant PE_Ptr := new PE'(PC_Fence_X, 0, EOP); begin return (AFC with P.Stk + 1, Bracket (E, Pat, X)); end Fence; -------------- -- Finalize -- -------------- procedure Finalize (Object : in out Pattern) is procedure Free is new Ada.Unchecked_Deallocation (PE, PE_Ptr); procedure Free is new Ada.Unchecked_Deallocation (String, String_Ptr); begin -- Nothing to do if already freed if Object.P = null then return; -- Otherwise we must free all elements else declare Refs : Ref_Array (1 .. Object.P.Index); -- References to elements in pattern to be finalized begin Build_Ref_Array (Object.P, Refs); for J in Refs'Range loop if Refs (J).Pcode = PC_String then Free (Refs (J).Str); end if; Free (Refs (J)); end loop; Object.P := null; end; end if; end Finalize; ----------- -- Image -- ----------- function Image (P : PE_Ptr) return String is begin return Image (To_Address (P)); end Image; function Image (P : Pattern) return String is begin return S (Image (P)); end Image; function Image (P : Pattern) return VString is Kill_Ampersand : Boolean := False; -- Set True to delete next & to be output to Result Result : VString := Nul; -- The result is accumulated here, using Append Refs : Ref_Array (1 .. P.P.Index); -- We build a reference array whose N'th element points to the -- pattern element whose Index value is N. procedure Delete_Ampersand; -- Deletes the ampersand at the end of Result procedure Image_Seq (E : PE_Ptr; Succ : PE_Ptr; Paren : Boolean); -- E refers to a pattern structure whose successor is given by Succ. -- This procedure appends to Result a representation of this pattern. -- The Paren parameter indicates whether parentheses are required if -- the output is more than one element. procedure Image_One (E : in out PE_Ptr); -- E refers to a pattern structure. This procedure appends to Result -- a representation of the single simple or compound pattern structure -- at the start of E and updates E to point to its successor. ---------------------- -- Delete_Ampersand -- ---------------------- procedure Delete_Ampersand is L : constant Natural := Length (Result); begin if L > 2 then Delete (Result, L - 1, L); end if; end Delete_Ampersand; --------------- -- Image_One -- --------------- procedure Image_One (E : in out PE_Ptr) is ER : PE_Ptr := E.Pthen; -- Successor set as result in E unless reset begin case E.Pcode is when PC_Cancel => Append (Result, "Cancel"); when PC_Alt => Alt : declare Elmts_In_L : constant IndexT := E.Pthen.Index - E.Alt.Index; -- Number of elements in left pattern of alternation Lowest_In_L : constant IndexT := E.Index - Elmts_In_L; -- Number of lowest index in elements of left pattern E1 : PE_Ptr; begin -- The successor of the alternation node must have a lower -- index than any node that is in the left pattern or a -- higher index than the alternation node itself. while ER /= EOP and then ER.Index >= Lowest_In_L and then ER.Index < E.Index loop ER := ER.Pthen; end loop; Append (Result, '('); E1 := E; loop Image_Seq (E1.Pthen, ER, False); Append (Result, " or "); E1 := E1.Alt; exit when E1.Pcode /= PC_Alt; end loop; Image_Seq (E1, ER, False); Append (Result, ')'); end Alt; when PC_Any_CS => Append (Result, "Any (" & Image (To_Sequence (E.CS)) & ')'); when PC_Any_VF => Append (Result, "Any (" & Str_VF (E.VF) & ')'); when PC_Any_VP => Append (Result, "Any (" & Str_VP (E.VP) & ')'); when PC_Arb_X => Append (Result, "Arb"); when PC_Arbno_S => Append (Result, "Arbno ("); Image_Seq (E.Alt, E, False); Append (Result, ')'); when PC_Arbno_X => Append (Result, "Arbno ("); Image_Seq (E.Alt.Pthen, Refs (E.Index - 2), False); Append (Result, ')'); when PC_Assign_Imm => Delete_Ampersand; Append (Result, "* " & Str_VP (Refs (E.Index).VP)); when PC_Assign_OnM => Delete_Ampersand; Append (Result, "** " & Str_VP (Refs (E.Index).VP)); when PC_Any_CH => Append (Result, "Any ('" & E.Char & "')"); when PC_Bal => Append (Result, "Bal"); when PC_Break_CH => Append (Result, "Break ('" & E.Char & "')"); when PC_Break_CS => Append (Result, "Break (" & Image (To_Sequence (E.CS)) & ')'); when PC_Break_VF => Append (Result, "Break (" & Str_VF (E.VF) & ')'); when PC_Break_VP => Append (Result, "Break (" & Str_VP (E.VP) & ')'); when PC_BreakX_CH => Append (Result, "BreakX ('" & E.Char & "')"); ER := ER.Pthen; when PC_BreakX_CS => Append (Result, "BreakX (" & Image (To_Sequence (E.CS)) & ')'); ER := ER.Pthen; when PC_BreakX_VF => Append (Result, "BreakX (" & Str_VF (E.VF) & ')'); ER := ER.Pthen; when PC_BreakX_VP => Append (Result, "BreakX (" & Str_VP (E.VP) & ')'); ER := ER.Pthen; when PC_Char => Append (Result, ''' & E.Char & '''); when PC_Fail => Append (Result, "Fail"); when PC_Fence => Append (Result, "Fence"); when PC_Fence_X => Append (Result, "Fence ("); Image_Seq (E.Pthen, Refs (E.Index - 1), False); Append (Result, ")"); ER := Refs (E.Index - 1).Pthen; when PC_Len_Nat => Append (Result, "Len (" & E.Nat & ')'); when PC_Len_NF => Append (Result, "Len (" & Str_NF (E.NF) & ')'); when PC_Len_NP => Append (Result, "Len (" & Str_NP (E.NP) & ')'); when PC_NotAny_CH => Append (Result, "NotAny ('" & E.Char & "')"); when PC_NotAny_CS => Append (Result, "NotAny (" & Image (To_Sequence (E.CS)) & ')'); when PC_NotAny_VF => Append (Result, "NotAny (" & Str_VF (E.VF) & ')'); when PC_NotAny_VP => Append (Result, "NotAny (" & Str_VP (E.VP) & ')'); when PC_NSpan_CH => Append (Result, "NSpan ('" & E.Char & "')"); when PC_NSpan_CS => Append (Result, "NSpan (" & Image (To_Sequence (E.CS)) & ')'); when PC_NSpan_VF => Append (Result, "NSpan (" & Str_VF (E.VF) & ')'); when PC_NSpan_VP => Append (Result, "NSpan (" & Str_VP (E.VP) & ')'); when PC_Null => Append (Result, """"""); when PC_Pos_Nat => Append (Result, "Pos (" & E.Nat & ')'); when PC_Pos_NF => Append (Result, "Pos (" & Str_NF (E.NF) & ')'); when PC_Pos_NP => Append (Result, "Pos (" & Str_NP (E.NP) & ')'); when PC_R_Enter => Kill_Ampersand := True; when PC_Rest => Append (Result, "Rest"); when PC_Rpat => Append (Result, "(+ " & Str_PP (E.PP) & ')'); when PC_Pred_Func => Append (Result, "(+ " & Str_BF (E.BF) & ')'); when PC_RPos_Nat => Append (Result, "RPos (" & E.Nat & ')'); when PC_RPos_NF => Append (Result, "RPos (" & Str_NF (E.NF) & ')'); when PC_RPos_NP => Append (Result, "RPos (" & Str_NP (E.NP) & ')'); when PC_RTab_Nat => Append (Result, "RTab (" & E.Nat & ')'); when PC_RTab_NF => Append (Result, "RTab (" & Str_NF (E.NF) & ')'); when PC_RTab_NP => Append (Result, "RTab (" & Str_NP (E.NP) & ')'); when PC_Setcur => Append (Result, "Setcur (" & Str_NP (E.Var) & ')'); when PC_Span_CH => Append (Result, "Span ('" & E.Char & "')"); when PC_Span_CS => Append (Result, "Span (" & Image (To_Sequence (E.CS)) & ')'); when PC_Span_VF => Append (Result, "Span (" & Str_VF (E.VF) & ')'); when PC_Span_VP => Append (Result, "Span (" & Str_VP (E.VP) & ')'); when PC_String => Append (Result, Image (E.Str.all)); when PC_String_2 => Append (Result, Image (E.Str2)); when PC_String_3 => Append (Result, Image (E.Str3)); when PC_String_4 => Append (Result, Image (E.Str4)); when PC_String_5 => Append (Result, Image (E.Str5)); when PC_String_6 => Append (Result, Image (E.Str6)); when PC_String_VF => Append (Result, "(+" & Str_VF (E.VF) & ')'); when PC_String_VP => Append (Result, "(+" & Str_VP (E.VP) & ')'); when PC_Succeed => Append (Result, "Succeed"); when PC_Tab_Nat => Append (Result, "Tab (" & E.Nat & ')'); when PC_Tab_NF => Append (Result, "Tab (" & Str_NF (E.NF) & ')'); when PC_Tab_NP => Append (Result, "Tab (" & Str_NP (E.NP) & ')'); when PC_Write_Imm => Append (Result, '('); Image_Seq (E, Refs (E.Index - 1), True); Append (Result, " * " & Str_FP (Refs (E.Index - 1).FP)); ER := Refs (E.Index - 1).Pthen; when PC_Write_OnM => Append (Result, '('); Image_Seq (E.Pthen, Refs (E.Index - 1), True); Append (Result, " ** " & Str_FP (Refs (E.Index - 1).FP)); ER := Refs (E.Index - 1).Pthen; -- Other pattern codes should not appear as leading elements when PC_Arb_Y | PC_Arbno_Y | PC_Assign | PC_BreakX_X | PC_EOP | PC_Fence_Y | PC_R_Remove | PC_R_Restore | PC_Unanchored => Append (Result, "???"); end case; E := ER; end Image_One; --------------- -- Image_Seq -- --------------- procedure Image_Seq (E : PE_Ptr; Succ : PE_Ptr; Paren : Boolean) is Indx : constant Natural := Length (Result); E1 : PE_Ptr := E; Mult : Boolean := False; begin -- The image of EOP is "" (the null string) if E = EOP then Append (Result, """"""); -- Else generate appropriate concatenation sequence else loop Image_One (E1); exit when E1 = Succ; exit when E1 = EOP; Mult := True; if Kill_Ampersand then Kill_Ampersand := False; else Append (Result, " & "); end if; end loop; end if; if Mult and Paren then Insert (Result, Indx + 1, "("); Append (Result, ")"); end if; end Image_Seq; -- Start of processing for Image begin Build_Ref_Array (P.P, Refs); Image_Seq (P.P, EOP, False); return Result; end Image; ----------- -- Is_In -- ----------- function Is_In (C : Character; Str : String) return Boolean is begin for J in Str'Range loop if Str (J) = C then return True; end if; end loop; return False; end Is_In; --------- -- Len -- --------- function Len (Count : Natural) return Pattern is begin -- Note, the following is not just an optimization, it is needed -- to ensure that Arbno (Len (0)) does not generate an infinite -- matching loop (since PC_Len_Nat is OK_For_Simple_Arbno). if Count = 0 then return (AFC with 0, new PE'(PC_Null, 1, EOP)); else return (AFC with 0, new PE'(PC_Len_Nat, 1, EOP, Count)); end if; end Len; function Len (Count : Natural_Func) return Pattern is begin return (AFC with 0, new PE'(PC_Len_NF, 1, EOP, Count)); end Len; function Len (Count : not null access Natural) return Pattern is begin return (AFC with 0, new PE'(PC_Len_NP, 1, EOP, Natural_Ptr (Count))); end Len; ----------------- -- Logic_Error -- ----------------- procedure Logic_Error is begin raise Program_Error with "Internal logic error in GNAT.Spitbol.Patterns"; end Logic_Error; ----------- -- Match -- ----------- function Match (Subject : VString; Pat : Pattern) return Boolean is S : String_Access; L : Natural; Start : Natural; Stop : Natural; pragma Unreferenced (Stop); begin Get_String (Subject, S, L); if Debug_Mode then XMatchD (S (1 .. L), Pat.P, Pat.Stk, Start, Stop); else XMatch (S (1 .. L), Pat.P, Pat.Stk, Start, Stop); end if; return Start /= 0; end Match; function Match (Subject : String; Pat : Pattern) return Boolean is Start, Stop : Natural; pragma Unreferenced (Stop); subtype String1 is String (1 .. Subject'Length); begin if Debug_Mode then XMatchD (String1 (Subject), Pat.P, Pat.Stk, Start, Stop); else XMatch (String1 (Subject), Pat.P, Pat.Stk, Start, Stop); end if; return Start /= 0; end Match; function Match (Subject : VString_Var; Pat : Pattern; Replace : VString) return Boolean is Start : Natural; Stop : Natural; S : String_Access; L : Natural; begin Get_String (Subject, S, L); if Debug_Mode then XMatchD (S (1 .. L), Pat.P, Pat.Stk, Start, Stop); else XMatch (S (1 .. L), Pat.P, Pat.Stk, Start, Stop); end if; if Start = 0 then return False; else Get_String (Replace, S, L); Replace_Slice (Subject'Unrestricted_Access.all, Start, Stop, S (1 .. L)); return True; end if; end Match; function Match (Subject : VString_Var; Pat : Pattern; Replace : String) return Boolean is Start : Natural; Stop : Natural; S : String_Access; L : Natural; begin Get_String (Subject, S, L); if Debug_Mode then XMatchD (S (1 .. L), Pat.P, Pat.Stk, Start, Stop); else XMatch (S (1 .. L), Pat.P, Pat.Stk, Start, Stop); end if; if Start = 0 then return False; else Replace_Slice (Subject'Unrestricted_Access.all, Start, Stop, Replace); return True; end if; end Match; procedure Match (Subject : VString; Pat : Pattern) is S : String_Access; L : Natural; Start : Natural; Stop : Natural; pragma Unreferenced (Start, Stop); begin Get_String (Subject, S, L); if Debug_Mode then XMatchD (S (1 .. L), Pat.P, Pat.Stk, Start, Stop); else XMatch (S (1 .. L), Pat.P, Pat.Stk, Start, Stop); end if; end Match; procedure Match (Subject : String; Pat : Pattern) is Start, Stop : Natural; pragma Unreferenced (Start, Stop); subtype String1 is String (1 .. Subject'Length); begin if Debug_Mode then XMatchD (String1 (Subject), Pat.P, Pat.Stk, Start, Stop); else XMatch (String1 (Subject), Pat.P, Pat.Stk, Start, Stop); end if; end Match; procedure Match (Subject : in out VString; Pat : Pattern; Replace : VString) is Start : Natural; Stop : Natural; S : String_Access; L : Natural; begin Get_String (Subject, S, L); if Debug_Mode then XMatchD (S (1 .. L), Pat.P, Pat.Stk, Start, Stop); else XMatch (S (1 .. L), Pat.P, Pat.Stk, Start, Stop); end if; if Start /= 0 then Get_String (Replace, S, L); Replace_Slice (Subject, Start, Stop, S (1 .. L)); end if; end Match; procedure Match (Subject : in out VString; Pat : Pattern; Replace : String) is Start : Natural; Stop : Natural; S : String_Access; L : Natural; begin Get_String (Subject, S, L); if Debug_Mode then XMatchD (S (1 .. L), Pat.P, Pat.Stk, Start, Stop); else XMatch (S (1 .. L), Pat.P, Pat.Stk, Start, Stop); end if; if Start /= 0 then Replace_Slice (Subject, Start, Stop, Replace); end if; end Match; function Match (Subject : VString; Pat : PString) return Boolean is Pat_Len : constant Natural := Pat'Length; S : String_Access; L : Natural; begin Get_String (Subject, S, L); if Anchored_Mode then if Pat_Len > L then return False; else return Pat = S (1 .. Pat_Len); end if; else for J in 1 .. L - Pat_Len + 1 loop if Pat = S (J .. J + (Pat_Len - 1)) then return True; end if; end loop; return False; end if; end Match; function Match (Subject : String; Pat : PString) return Boolean is Pat_Len : constant Natural := Pat'Length; Sub_Len : constant Natural := Subject'Length; SFirst : constant Natural := Subject'First; begin if Anchored_Mode then if Pat_Len > Sub_Len then return False; else return Pat = Subject (SFirst .. SFirst + Pat_Len - 1); end if; else for J in SFirst .. SFirst + Sub_Len - Pat_Len loop if Pat = Subject (J .. J + (Pat_Len - 1)) then return True; end if; end loop; return False; end if; end Match; function Match (Subject : VString_Var; Pat : PString; Replace : VString) return Boolean is Start : Natural; Stop : Natural; S : String_Access; L : Natural; begin Get_String (Subject, S, L); if Debug_Mode then XMatchD (S (1 .. L), S_To_PE (Pat), 0, Start, Stop); else XMatch (S (1 .. L), S_To_PE (Pat), 0, Start, Stop); end if; if Start = 0 then return False; else Get_String (Replace, S, L); Replace_Slice (Subject'Unrestricted_Access.all, Start, Stop, S (1 .. L)); return True; end if; end Match; function Match (Subject : VString_Var; Pat : PString; Replace : String) return Boolean is Start : Natural; Stop : Natural; S : String_Access; L : Natural; begin Get_String (Subject, S, L); if Debug_Mode then XMatchD (S (1 .. L), S_To_PE (Pat), 0, Start, Stop); else XMatch (S (1 .. L), S_To_PE (Pat), 0, Start, Stop); end if; if Start = 0 then return False; else Replace_Slice (Subject'Unrestricted_Access.all, Start, Stop, Replace); return True; end if; end Match; procedure Match (Subject : VString; Pat : PString) is S : String_Access; L : Natural; Start : Natural; Stop : Natural; pragma Unreferenced (Start, Stop); begin Get_String (Subject, S, L); if Debug_Mode then XMatchD (S (1 .. L), S_To_PE (Pat), 0, Start, Stop); else XMatch (S (1 .. L), S_To_PE (Pat), 0, Start, Stop); end if; end Match; procedure Match (Subject : String; Pat : PString) is Start, Stop : Natural; pragma Unreferenced (Start, Stop); subtype String1 is String (1 .. Subject'Length); begin if Debug_Mode then XMatchD (String1 (Subject), S_To_PE (Pat), 0, Start, Stop); else XMatch (String1 (Subject), S_To_PE (Pat), 0, Start, Stop); end if; end Match; procedure Match (Subject : in out VString; Pat : PString; Replace : VString) is Start : Natural; Stop : Natural; S : String_Access; L : Natural; begin Get_String (Subject, S, L); if Debug_Mode then XMatchD (S (1 .. L), S_To_PE (Pat), 0, Start, Stop); else XMatch (S (1 .. L), S_To_PE (Pat), 0, Start, Stop); end if; if Start /= 0 then Get_String (Replace, S, L); Replace_Slice (Subject, Start, Stop, S (1 .. L)); end if; end Match; procedure Match (Subject : in out VString; Pat : PString; Replace : String) is Start : Natural; Stop : Natural; S : String_Access; L : Natural; begin Get_String (Subject, S, L); if Debug_Mode then XMatchD (S (1 .. L), S_To_PE (Pat), 0, Start, Stop); else XMatch (S (1 .. L), S_To_PE (Pat), 0, Start, Stop); end if; if Start /= 0 then Replace_Slice (Subject, Start, Stop, Replace); end if; end Match; function Match (Subject : VString_Var; Pat : Pattern; Result : Match_Result_Var) return Boolean is Start : Natural; Stop : Natural; S : String_Access; L : Natural; begin Get_String (Subject, S, L); if Debug_Mode then XMatchD (S (1 .. L), Pat.P, Pat.Stk, Start, Stop); else XMatch (S (1 .. L), Pat.P, Pat.Stk, Start, Stop); end if; if Start = 0 then Result'Unrestricted_Access.all.Var := null; return False; else Result'Unrestricted_Access.all.Var := Subject'Unrestricted_Access; Result'Unrestricted_Access.all.Start := Start; Result'Unrestricted_Access.all.Stop := Stop; return True; end if; end Match; procedure Match (Subject : in out VString; Pat : Pattern; Result : out Match_Result) is Start : Natural; Stop : Natural; S : String_Access; L : Natural; begin Get_String (Subject, S, L); if Debug_Mode then XMatchD (S (1 .. L), Pat.P, Pat.Stk, Start, Stop); else XMatch (S (1 .. L), Pat.P, Pat.Stk, Start, Stop); end if; if Start = 0 then Result.Var := null; else Result.Var := Subject'Unrestricted_Access; Result.Start := Start; Result.Stop := Stop; end if; end Match; --------------- -- New_LineD -- --------------- procedure New_LineD is begin if Internal_Debug then New_Line; end if; end New_LineD; ------------ -- NotAny -- ------------ function NotAny (Str : String) return Pattern is begin return (AFC with 0, new PE'(PC_NotAny_CS, 1, EOP, To_Set (Str))); end NotAny; function NotAny (Str : VString) return Pattern is begin return NotAny (S (Str)); end NotAny; function NotAny (Str : Character) return Pattern is begin return (AFC with 0, new PE'(PC_NotAny_CH, 1, EOP, Str)); end NotAny; function NotAny (Str : Character_Set) return Pattern is begin return (AFC with 0, new PE'(PC_NotAny_CS, 1, EOP, Str)); end NotAny; function NotAny (Str : not null access VString) return Pattern is begin return (AFC with 0, new PE'(PC_NotAny_VP, 1, EOP, VString_Ptr (Str))); end NotAny; function NotAny (Str : VString_Func) return Pattern is begin return (AFC with 0, new PE'(PC_NotAny_VF, 1, EOP, Str)); end NotAny; ----------- -- NSpan -- ----------- function NSpan (Str : String) return Pattern is begin return (AFC with 0, new PE'(PC_NSpan_CS, 1, EOP, To_Set (Str))); end NSpan; function NSpan (Str : VString) return Pattern is begin return NSpan (S (Str)); end NSpan; function NSpan (Str : Character) return Pattern is begin return (AFC with 0, new PE'(PC_NSpan_CH, 1, EOP, Str)); end NSpan; function NSpan (Str : Character_Set) return Pattern is begin return (AFC with 0, new PE'(PC_NSpan_CS, 1, EOP, Str)); end NSpan; function NSpan (Str : not null access VString) return Pattern is begin return (AFC with 0, new PE'(PC_NSpan_VP, 1, EOP, VString_Ptr (Str))); end NSpan; function NSpan (Str : VString_Func) return Pattern is begin return (AFC with 0, new PE'(PC_NSpan_VF, 1, EOP, Str)); end NSpan; --------- -- Pos -- --------- function Pos (Count : Natural) return Pattern is begin return (AFC with 0, new PE'(PC_Pos_Nat, 1, EOP, Count)); end Pos; function Pos (Count : Natural_Func) return Pattern is begin return (AFC with 0, new PE'(PC_Pos_NF, 1, EOP, Count)); end Pos; function Pos (Count : not null access Natural) return Pattern is begin return (AFC with 0, new PE'(PC_Pos_NP, 1, EOP, Natural_Ptr (Count))); end Pos; ---------- -- PutD -- ---------- procedure PutD (Str : String) is begin if Internal_Debug then Put (Str); end if; end PutD; --------------- -- Put_LineD -- --------------- procedure Put_LineD (Str : String) is begin if Internal_Debug then Put_Line (Str); end if; end Put_LineD; ------------- -- Replace -- ------------- procedure Replace (Result : in out Match_Result; Replace : VString) is S : String_Access; L : Natural; begin Get_String (Replace, S, L); if Result.Var /= null then Replace_Slice (Result.Var.all, Result.Start, Result.Stop, S (1 .. L)); Result.Var := null; end if; end Replace; ---------- -- Rest -- ---------- function Rest return Pattern is begin return (AFC with 0, new PE'(PC_Rest, 1, EOP)); end Rest; ---------- -- Rpos -- ---------- function Rpos (Count : Natural) return Pattern is begin return (AFC with 0, new PE'(PC_RPos_Nat, 1, EOP, Count)); end Rpos; function Rpos (Count : Natural_Func) return Pattern is begin return (AFC with 0, new PE'(PC_RPos_NF, 1, EOP, Count)); end Rpos; function Rpos (Count : not null access Natural) return Pattern is begin return (AFC with 0, new PE'(PC_RPos_NP, 1, EOP, Natural_Ptr (Count))); end Rpos; ---------- -- Rtab -- ---------- function Rtab (Count : Natural) return Pattern is begin return (AFC with 0, new PE'(PC_RTab_Nat, 1, EOP, Count)); end Rtab; function Rtab (Count : Natural_Func) return Pattern is begin return (AFC with 0, new PE'(PC_RTab_NF, 1, EOP, Count)); end Rtab; function Rtab (Count : not null access Natural) return Pattern is begin return (AFC with 0, new PE'(PC_RTab_NP, 1, EOP, Natural_Ptr (Count))); end Rtab; ------------- -- S_To_PE -- ------------- function S_To_PE (Str : PString) return PE_Ptr is Len : constant Natural := Str'Length; begin case Len is when 0 => return new PE'(PC_Null, 1, EOP); when 1 => return new PE'(PC_Char, 1, EOP, Str (Str'First)); when 2 => return new PE'(PC_String_2, 1, EOP, Str); when 3 => return new PE'(PC_String_3, 1, EOP, Str); when 4 => return new PE'(PC_String_4, 1, EOP, Str); when 5 => return new PE'(PC_String_5, 1, EOP, Str); when 6 => return new PE'(PC_String_6, 1, EOP, Str); when others => return new PE'(PC_String, 1, EOP, new String'(Str)); end case; end S_To_PE; ------------------- -- Set_Successor -- ------------------- -- Note: this procedure is not used by the normal concatenation circuit, -- since other fixups are required on the left operand in this case, and -- they might as well be done all together. procedure Set_Successor (Pat : PE_Ptr; Succ : PE_Ptr) is begin if Pat = null then Uninitialized_Pattern; elsif Pat = EOP then Logic_Error; else declare Refs : Ref_Array (1 .. Pat.Index); -- We build a reference array for L whose N'th element points to -- the pattern element of L whose original Index value is N. P : PE_Ptr; begin Build_Ref_Array (Pat, Refs); for J in Refs'Range loop P := Refs (J); if P.Pthen = EOP then P.Pthen := Succ; end if; if P.Pcode in PC_Has_Alt and then P.Alt = EOP then P.Alt := Succ; end if; end loop; end; end if; end Set_Successor; ------------ -- Setcur -- ------------ function Setcur (Var : not null access Natural) return Pattern is begin return (AFC with 0, new PE'(PC_Setcur, 1, EOP, Natural_Ptr (Var))); end Setcur; ---------- -- Span -- ---------- function Span (Str : String) return Pattern is begin return (AFC with 0, new PE'(PC_Span_CS, 1, EOP, To_Set (Str))); end Span; function Span (Str : VString) return Pattern is begin return Span (S (Str)); end Span; function Span (Str : Character) return Pattern is begin return (AFC with 0, new PE'(PC_Span_CH, 1, EOP, Str)); end Span; function Span (Str : Character_Set) return Pattern is begin return (AFC with 0, new PE'(PC_Span_CS, 1, EOP, Str)); end Span; function Span (Str : not null access VString) return Pattern is begin return (AFC with 0, new PE'(PC_Span_VP, 1, EOP, VString_Ptr (Str))); end Span; function Span (Str : VString_Func) return Pattern is begin return (AFC with 0, new PE'(PC_Span_VF, 1, EOP, Str)); end Span; ------------ -- Str_BF -- ------------ function Str_BF (A : Boolean_Func) return String is function To_A is new Ada.Unchecked_Conversion (Boolean_Func, Address); begin return "BF(" & Image (To_A (A)) & ')'; end Str_BF; ------------ -- Str_FP -- ------------ function Str_FP (A : File_Ptr) return String is begin return "FP(" & Image (A.all'Address) & ')'; end Str_FP; ------------ -- Str_NF -- ------------ function Str_NF (A : Natural_Func) return String is function To_A is new Ada.Unchecked_Conversion (Natural_Func, Address); begin return "NF(" & Image (To_A (A)) & ')'; end Str_NF; ------------ -- Str_NP -- ------------ function Str_NP (A : Natural_Ptr) return String is begin return "NP(" & Image (A.all'Address) & ')'; end Str_NP; ------------ -- Str_PP -- ------------ function Str_PP (A : Pattern_Ptr) return String is begin return "PP(" & Image (A.all'Address) & ')'; end Str_PP; ------------ -- Str_VF -- ------------ function Str_VF (A : VString_Func) return String is function To_A is new Ada.Unchecked_Conversion (VString_Func, Address); begin return "VF(" & Image (To_A (A)) & ')'; end Str_VF; ------------ -- Str_VP -- ------------ function Str_VP (A : VString_Ptr) return String is begin return "VP(" & Image (A.all'Address) & ')'; end Str_VP; ------------- -- Succeed -- ------------- function Succeed return Pattern is begin return (AFC with 1, new PE'(PC_Succeed, 1, EOP)); end Succeed; --------- -- Tab -- --------- function Tab (Count : Natural) return Pattern is begin return (AFC with 0, new PE'(PC_Tab_Nat, 1, EOP, Count)); end Tab; function Tab (Count : Natural_Func) return Pattern is begin return (AFC with 0, new PE'(PC_Tab_NF, 1, EOP, Count)); end Tab; function Tab (Count : not null access Natural) return Pattern is begin return (AFC with 0, new PE'(PC_Tab_NP, 1, EOP, Natural_Ptr (Count))); end Tab; --------------------------- -- Uninitialized_Pattern -- --------------------------- procedure Uninitialized_Pattern is begin raise Program_Error with "uninitialized value of type GNAT.Spitbol.Patterns.Pattern"; end Uninitialized_Pattern; ------------ -- XMatch -- ------------ procedure XMatch (Subject : String; Pat_P : PE_Ptr; Pat_S : Natural; Start : out Natural; Stop : out Natural) is Node : PE_Ptr; -- Pointer to current pattern node. Initialized from Pat_P, and then -- updated as the match proceeds through its constituent elements. Length : constant Natural := Subject'Length; -- Length of string (= Subject'Last, since Subject'First is always 1) Cursor : Integer := 0; -- If the value is non-negative, then this value is the index showing -- the current position of the match in the subject string. The next -- character to be matched is at Subject (Cursor + 1). Note that since -- our view of the subject string in XMatch always has a lower bound -- of one, regardless of original bounds, that this definition exactly -- corresponds to the cursor value as referenced by functions like Pos. -- -- If the value is negative, then this is a saved stack pointer, -- typically a base pointer of an inner or outer region. Cursor -- temporarily holds such a value when it is popped from the stack -- by Fail. In all cases, Cursor is reset to a proper non-negative -- cursor value before the match proceeds (e.g. by propagating the -- failure and popping a "real" cursor value from the stack. PE_Unanchored : aliased PE := (PC_Unanchored, 0, Pat_P); -- Dummy pattern element used in the unanchored case Stack : Stack_Type; -- The pattern matching failure stack for this call to Match Stack_Ptr : Stack_Range; -- Current stack pointer. This points to the top element of the stack -- that is currently in use. At the outer level this is the special -- entry placed on the stack according to the anchor mode. Stack_Init : constant Stack_Range := Stack'First + 1; -- This is the initial value of the Stack_Ptr and Stack_Base. The -- initial (Stack'First) element of the stack is not used so that -- when we pop the last element off, Stack_Ptr is still in range. Stack_Base : Stack_Range; -- This value is the stack base value, i.e. the stack pointer for the -- first history stack entry in the current stack region. See separate -- section on handling of recursive pattern matches. Assign_OnM : Boolean := False; -- Set True if assign-on-match or write-on-match operations may be -- present in the history stack, which must then be scanned on a -- successful match. procedure Pop_Region; pragma Inline (Pop_Region); -- Used at the end of processing of an inner region. If the inner -- region left no stack entries, then all trace of it is removed. -- Otherwise a PC_Restore_Region entry is pushed to ensure proper -- handling of alternatives in the inner region. procedure Push (Node : PE_Ptr); pragma Inline (Push); -- Make entry in pattern matching stack with current cursor value procedure Push_Region; pragma Inline (Push_Region); -- This procedure makes a new region on the history stack. The -- caller first establishes the special entry on the stack, but -- does not push the stack pointer. Then this call stacks a -- PC_Remove_Region node, on top of this entry, using the cursor -- field of the PC_Remove_Region entry to save the outer level -- stack base value, and resets the stack base to point to this -- PC_Remove_Region node. ---------------- -- Pop_Region -- ---------------- procedure Pop_Region is begin -- If nothing was pushed in the inner region, we can just get -- rid of it entirely, leaving no traces that it was ever there if Stack_Ptr = Stack_Base then Stack_Ptr := Stack_Base - 2; Stack_Base := Stack (Stack_Ptr + 2).Cursor; -- If stuff was pushed in the inner region, then we have to -- push a PC_R_Restore node so that we properly handle possible -- rematches within the region. else Stack_Ptr := Stack_Ptr + 1; Stack (Stack_Ptr).Cursor := Stack_Base; Stack (Stack_Ptr).Node := CP_R_Restore'Access; Stack_Base := Stack (Stack_Base).Cursor; end if; end Pop_Region; ---------- -- Push -- ---------- procedure Push (Node : PE_Ptr) is begin Stack_Ptr := Stack_Ptr + 1; Stack (Stack_Ptr).Cursor := Cursor; Stack (Stack_Ptr).Node := Node; end Push; ----------------- -- Push_Region -- ----------------- procedure Push_Region is begin Stack_Ptr := Stack_Ptr + 2; Stack (Stack_Ptr).Cursor := Stack_Base; Stack (Stack_Ptr).Node := CP_R_Remove'Access; Stack_Base := Stack_Ptr; end Push_Region; -- Start of processing for XMatch begin if Pat_P = null then Uninitialized_Pattern; end if; -- Check we have enough stack for this pattern. This check deals with -- every possibility except a match of a recursive pattern, where we -- make a check at each recursion level. if Pat_S >= Stack_Size - 1 then raise Pattern_Stack_Overflow; end if; -- In anchored mode, the bottom entry on the stack is an abort entry if Anchored_Mode then Stack (Stack_Init).Node := CP_Cancel'Access; Stack (Stack_Init).Cursor := 0; -- In unanchored more, the bottom entry on the stack references -- the special pattern element PE_Unanchored, whose Pthen field -- points to the initial pattern element. The cursor value in this -- entry is the number of anchor moves so far. else Stack (Stack_Init).Node := PE_Unanchored'Unchecked_Access; Stack (Stack_Init).Cursor := 0; end if; Stack_Ptr := Stack_Init; Stack_Base := Stack_Ptr; Cursor := 0; Node := Pat_P; goto Match; ----------------------------------------- -- Main Pattern Matching State Control -- ----------------------------------------- -- This is a state machine which uses gotos to change state. The -- initial state is Match, to initiate the matching of the first -- element, so the goto Match above starts the match. In the -- following descriptions, we indicate the global values that -- are relevant for the state transition. -- Come here if entire match fails <<Match_Fail>> Start := 0; Stop := 0; return; -- Come here if entire match succeeds -- Cursor current position in subject string <<Match_Succeed>> Start := Stack (Stack_Init).Cursor + 1; Stop := Cursor; -- Scan history stack for deferred assignments or writes if Assign_OnM then for S in Stack_Init .. Stack_Ptr loop if Stack (S).Node = CP_Assign'Access then declare Inner_Base : constant Stack_Range := Stack (S + 1).Cursor; Special_Entry : constant Stack_Range := Inner_Base - 1; Node_OnM : constant PE_Ptr := Stack (Special_Entry).Node; Start : constant Natural := Stack (Special_Entry).Cursor + 1; Stop : constant Natural := Stack (S).Cursor; begin if Node_OnM.Pcode = PC_Assign_OnM then Set_String (Node_OnM.VP.all, Subject (Start .. Stop)); elsif Node_OnM.Pcode = PC_Write_OnM then Put_Line (Node_OnM.FP.all, Subject (Start .. Stop)); else Logic_Error; end if; end; end if; end loop; end if; return; -- Come here if attempt to match current element fails -- Stack_Base current stack base -- Stack_Ptr current stack pointer <<Fail>> Cursor := Stack (Stack_Ptr).Cursor; Node := Stack (Stack_Ptr).Node; Stack_Ptr := Stack_Ptr - 1; goto Match; -- Come here if attempt to match current element succeeds -- Cursor current position in subject string -- Node pointer to node successfully matched -- Stack_Base current stack base -- Stack_Ptr current stack pointer <<Succeed>> Node := Node.Pthen; -- Come here to match the next pattern element -- Cursor current position in subject string -- Node pointer to node to be matched -- Stack_Base current stack base -- Stack_Ptr current stack pointer <<Match>> -------------------------------------------------- -- Main Pattern Match Element Matching Routines -- -------------------------------------------------- -- Here is the case statement that processes the current node. The -- processing for each element does one of five things: -- goto Succeed to move to the successor -- goto Match_Succeed if the entire match succeeds -- goto Match_Fail if the entire match fails -- goto Fail to signal failure of current match -- Processing is NOT allowed to fall through case Node.Pcode is -- Cancel when PC_Cancel => goto Match_Fail; -- Alternation when PC_Alt => Push (Node.Alt); Node := Node.Pthen; goto Match; -- Any (one character case) when PC_Any_CH => if Cursor < Length and then Subject (Cursor + 1) = Node.Char then Cursor := Cursor + 1; goto Succeed; else goto Fail; end if; -- Any (character set case) when PC_Any_CS => if Cursor < Length and then Is_In (Subject (Cursor + 1), Node.CS) then Cursor := Cursor + 1; goto Succeed; else goto Fail; end if; -- Any (string function case) when PC_Any_VF => declare U : constant VString := Node.VF.all; S : String_Access; L : Natural; begin Get_String (U, S, L); if Cursor < Length and then Is_In (Subject (Cursor + 1), S (1 .. L)) then Cursor := Cursor + 1; goto Succeed; else goto Fail; end if; end; -- Any (string pointer case) when PC_Any_VP => declare U : constant VString := Node.VP.all; S : String_Access; L : Natural; begin Get_String (U, S, L); if Cursor < Length and then Is_In (Subject (Cursor + 1), S (1 .. L)) then Cursor := Cursor + 1; goto Succeed; else goto Fail; end if; end; -- Arb (initial match) when PC_Arb_X => Push (Node.Alt); Node := Node.Pthen; goto Match; -- Arb (extension) when PC_Arb_Y => if Cursor < Length then Cursor := Cursor + 1; Push (Node); goto Succeed; else goto Fail; end if; -- Arbno_S (simple Arbno initialize). This is the node that -- initiates the match of a simple Arbno structure. when PC_Arbno_S => Push (Node.Alt); Node := Node.Pthen; goto Match; -- Arbno_X (Arbno initialize). This is the node that initiates -- the match of a complex Arbno structure. when PC_Arbno_X => Push (Node.Alt); Node := Node.Pthen; goto Match; -- Arbno_Y (Arbno rematch). This is the node that is executed -- following successful matching of one instance of a complex -- Arbno pattern. when PC_Arbno_Y => declare Null_Match : constant Boolean := Cursor = Stack (Stack_Base - 1).Cursor; begin Pop_Region; -- If arbno extension matched null, then immediately fail if Null_Match then goto Fail; end if; -- Here we must do a stack check to make sure enough stack -- is left. This check will happen once for each instance of -- the Arbno pattern that is matched. The Nat field of a -- PC_Arbno pattern contains the maximum stack entries needed -- for the Arbno with one instance and the successor pattern if Stack_Ptr + Node.Nat >= Stack'Last then raise Pattern_Stack_Overflow; end if; goto Succeed; end; -- Assign. If this node is executed, it means the assign-on-match -- or write-on-match operation will not happen after all, so we -- is propagate the failure, removing the PC_Assign node. when PC_Assign => goto Fail; -- Assign immediate. This node performs the actual assignment when PC_Assign_Imm => Set_String (Node.VP.all, Subject (Stack (Stack_Base - 1).Cursor + 1 .. Cursor)); Pop_Region; goto Succeed; -- Assign on match. This node sets up for the eventual assignment when PC_Assign_OnM => Stack (Stack_Base - 1).Node := Node; Push (CP_Assign'Access); Pop_Region; Assign_OnM := True; goto Succeed; -- Bal when PC_Bal => if Cursor >= Length or else Subject (Cursor + 1) = ')' then goto Fail; elsif Subject (Cursor + 1) = '(' then declare Paren_Count : Natural := 1; begin loop Cursor := Cursor + 1; if Cursor >= Length then goto Fail; elsif Subject (Cursor + 1) = '(' then Paren_Count := Paren_Count + 1; elsif Subject (Cursor + 1) = ')' then Paren_Count := Paren_Count - 1; exit when Paren_Count = 0; end if; end loop; end; end if; Cursor := Cursor + 1; Push (Node); goto Succeed; -- Break (one character case) when PC_Break_CH => while Cursor < Length loop if Subject (Cursor + 1) = Node.Char then goto Succeed; else Cursor := Cursor + 1; end if; end loop; goto Fail; -- Break (character set case) when PC_Break_CS => while Cursor < Length loop if Is_In (Subject (Cursor + 1), Node.CS) then goto Succeed; else Cursor := Cursor + 1; end if; end loop; goto Fail; -- Break (string function case) when PC_Break_VF => declare U : constant VString := Node.VF.all; S : String_Access; L : Natural; begin Get_String (U, S, L); while Cursor < Length loop if Is_In (Subject (Cursor + 1), S (1 .. L)) then goto Succeed; else Cursor := Cursor + 1; end if; end loop; goto Fail; end; -- Break (string pointer case) when PC_Break_VP => declare U : constant VString := Node.VP.all; S : String_Access; L : Natural; begin Get_String (U, S, L); while Cursor < Length loop if Is_In (Subject (Cursor + 1), S (1 .. L)) then goto Succeed; else Cursor := Cursor + 1; end if; end loop; goto Fail; end; -- BreakX (one character case) when PC_BreakX_CH => while Cursor < Length loop if Subject (Cursor + 1) = Node.Char then goto Succeed; else Cursor := Cursor + 1; end if; end loop; goto Fail; -- BreakX (character set case) when PC_BreakX_CS => while Cursor < Length loop if Is_In (Subject (Cursor + 1), Node.CS) then goto Succeed; else Cursor := Cursor + 1; end if; end loop; goto Fail; -- BreakX (string function case) when PC_BreakX_VF => declare U : constant VString := Node.VF.all; S : String_Access; L : Natural; begin Get_String (U, S, L); while Cursor < Length loop if Is_In (Subject (Cursor + 1), S (1 .. L)) then goto Succeed; else Cursor := Cursor + 1; end if; end loop; goto Fail; end; -- BreakX (string pointer case) when PC_BreakX_VP => declare U : constant VString := Node.VP.all; S : String_Access; L : Natural; begin Get_String (U, S, L); while Cursor < Length loop if Is_In (Subject (Cursor + 1), S (1 .. L)) then goto Succeed; else Cursor := Cursor + 1; end if; end loop; goto Fail; end; -- BreakX_X (BreakX extension). See section on "Compound Pattern -- Structures". This node is the alternative that is stacked to -- skip past the break character and extend the break. when PC_BreakX_X => Cursor := Cursor + 1; goto Succeed; -- Character (one character string) when PC_Char => if Cursor < Length and then Subject (Cursor + 1) = Node.Char then Cursor := Cursor + 1; goto Succeed; else goto Fail; end if; -- End of Pattern when PC_EOP => if Stack_Base = Stack_Init then goto Match_Succeed; -- End of recursive inner match. See separate section on -- handing of recursive pattern matches for details. else Node := Stack (Stack_Base - 1).Node; Pop_Region; goto Match; end if; -- Fail when PC_Fail => goto Fail; -- Fence (built in pattern) when PC_Fence => Push (CP_Cancel'Access); goto Succeed; -- Fence function node X. This is the node that gets control -- after a successful match of the fenced pattern. when PC_Fence_X => Stack_Ptr := Stack_Ptr + 1; Stack (Stack_Ptr).Cursor := Stack_Base; Stack (Stack_Ptr).Node := CP_Fence_Y'Access; Stack_Base := Stack (Stack_Base).Cursor; goto Succeed; -- Fence function node Y. This is the node that gets control on -- a failure that occurs after the fenced pattern has matched. -- Note: the Cursor at this stage is actually the inner stack -- base value. We don't reset this, but we do use it to strip -- off all the entries made by the fenced pattern. when PC_Fence_Y => Stack_Ptr := Cursor - 2; goto Fail; -- Len (integer case) when PC_Len_Nat => if Cursor + Node.Nat > Length then goto Fail; else Cursor := Cursor + Node.Nat; goto Succeed; end if; -- Len (Integer function case) when PC_Len_NF => declare N : constant Natural := Node.NF.all; begin if Cursor + N > Length then goto Fail; else Cursor := Cursor + N; goto Succeed; end if; end; -- Len (integer pointer case) when PC_Len_NP => if Cursor + Node.NP.all > Length then goto Fail; else Cursor := Cursor + Node.NP.all; goto Succeed; end if; -- NotAny (one character case) when PC_NotAny_CH => if Cursor < Length and then Subject (Cursor + 1) /= Node.Char then Cursor := Cursor + 1; goto Succeed; else goto Fail; end if; -- NotAny (character set case) when PC_NotAny_CS => if Cursor < Length and then not Is_In (Subject (Cursor + 1), Node.CS) then Cursor := Cursor + 1; goto Succeed; else goto Fail; end if; -- NotAny (string function case) when PC_NotAny_VF => declare U : constant VString := Node.VF.all; S : String_Access; L : Natural; begin Get_String (U, S, L); if Cursor < Length and then not Is_In (Subject (Cursor + 1), S (1 .. L)) then Cursor := Cursor + 1; goto Succeed; else goto Fail; end if; end; -- NotAny (string pointer case) when PC_NotAny_VP => declare U : constant VString := Node.VP.all; S : String_Access; L : Natural; begin Get_String (U, S, L); if Cursor < Length and then not Is_In (Subject (Cursor + 1), S (1 .. L)) then Cursor := Cursor + 1; goto Succeed; else goto Fail; end if; end; -- NSpan (one character case) when PC_NSpan_CH => while Cursor < Length and then Subject (Cursor + 1) = Node.Char loop Cursor := Cursor + 1; end loop; goto Succeed; -- NSpan (character set case) when PC_NSpan_CS => while Cursor < Length and then Is_In (Subject (Cursor + 1), Node.CS) loop Cursor := Cursor + 1; end loop; goto Succeed; -- NSpan (string function case) when PC_NSpan_VF => declare U : constant VString := Node.VF.all; S : String_Access; L : Natural; begin Get_String (U, S, L); while Cursor < Length and then Is_In (Subject (Cursor + 1), S (1 .. L)) loop Cursor := Cursor + 1; end loop; goto Succeed; end; -- NSpan (string pointer case) when PC_NSpan_VP => declare U : constant VString := Node.VP.all; S : String_Access; L : Natural; begin Get_String (U, S, L); while Cursor < Length and then Is_In (Subject (Cursor + 1), S (1 .. L)) loop Cursor := Cursor + 1; end loop; goto Succeed; end; -- Null string when PC_Null => goto Succeed; -- Pos (integer case) when PC_Pos_Nat => if Cursor = Node.Nat then goto Succeed; else goto Fail; end if; -- Pos (Integer function case) when PC_Pos_NF => declare N : constant Natural := Node.NF.all; begin if Cursor = N then goto Succeed; else goto Fail; end if; end; -- Pos (integer pointer case) when PC_Pos_NP => if Cursor = Node.NP.all then goto Succeed; else goto Fail; end if; -- Predicate function when PC_Pred_Func => if Node.BF.all then goto Succeed; else goto Fail; end if; -- Region Enter. Initiate new pattern history stack region when PC_R_Enter => Stack (Stack_Ptr + 1).Cursor := Cursor; Push_Region; goto Succeed; -- Region Remove node. This is the node stacked by an R_Enter. -- It removes the special format stack entry right underneath, and -- then restores the outer level stack base and signals failure. -- Note: the cursor value at this stage is actually the (negative) -- stack base value for the outer level. when PC_R_Remove => Stack_Base := Cursor; Stack_Ptr := Stack_Ptr - 1; goto Fail; -- Region restore node. This is the node stacked at the end of an -- inner level match. Its function is to restore the inner level -- region, so that alternatives in this region can be sought. -- Note: the Cursor at this stage is actually the negative of the -- inner stack base value, which we use to restore the inner region. when PC_R_Restore => Stack_Base := Cursor; goto Fail; -- Rest when PC_Rest => Cursor := Length; goto Succeed; -- Initiate recursive match (pattern pointer case) when PC_Rpat => Stack (Stack_Ptr + 1).Node := Node.Pthen; Push_Region; if Stack_Ptr + Node.PP.all.Stk >= Stack_Size then raise Pattern_Stack_Overflow; else Node := Node.PP.all.P; goto Match; end if; -- RPos (integer case) when PC_RPos_Nat => if Cursor = (Length - Node.Nat) then goto Succeed; else goto Fail; end if; -- RPos (integer function case) when PC_RPos_NF => declare N : constant Natural := Node.NF.all; begin if Length - Cursor = N then goto Succeed; else goto Fail; end if; end; -- RPos (integer pointer case) when PC_RPos_NP => if Cursor = (Length - Node.NP.all) then goto Succeed; else goto Fail; end if; -- RTab (integer case) when PC_RTab_Nat => if Cursor <= (Length - Node.Nat) then Cursor := Length - Node.Nat; goto Succeed; else goto Fail; end if; -- RTab (integer function case) when PC_RTab_NF => declare N : constant Natural := Node.NF.all; begin if Length - Cursor >= N then Cursor := Length - N; goto Succeed; else goto Fail; end if; end; -- RTab (integer pointer case) when PC_RTab_NP => if Cursor <= (Length - Node.NP.all) then Cursor := Length - Node.NP.all; goto Succeed; else goto Fail; end if; -- Cursor assignment when PC_Setcur => Node.Var.all := Cursor; goto Succeed; -- Span (one character case) when PC_Span_CH => declare P : Natural; begin P := Cursor; while P < Length and then Subject (P + 1) = Node.Char loop P := P + 1; end loop; if P /= Cursor then Cursor := P; goto Succeed; else goto Fail; end if; end; -- Span (character set case) when PC_Span_CS => declare P : Natural; begin P := Cursor; while P < Length and then Is_In (Subject (P + 1), Node.CS) loop P := P + 1; end loop; if P /= Cursor then Cursor := P; goto Succeed; else goto Fail; end if; end; -- Span (string function case) when PC_Span_VF => declare U : constant VString := Node.VF.all; S : String_Access; L : Natural; P : Natural; begin Get_String (U, S, L); P := Cursor; while P < Length and then Is_In (Subject (P + 1), S (1 .. L)) loop P := P + 1; end loop; if P /= Cursor then Cursor := P; goto Succeed; else goto Fail; end if; end; -- Span (string pointer case) when PC_Span_VP => declare U : constant VString := Node.VP.all; S : String_Access; L : Natural; P : Natural; begin Get_String (U, S, L); P := Cursor; while P < Length and then Is_In (Subject (P + 1), S (1 .. L)) loop P := P + 1; end loop; if P /= Cursor then Cursor := P; goto Succeed; else goto Fail; end if; end; -- String (two character case) when PC_String_2 => if (Length - Cursor) >= 2 and then Subject (Cursor + 1 .. Cursor + 2) = Node.Str2 then Cursor := Cursor + 2; goto Succeed; else goto Fail; end if; -- String (three character case) when PC_String_3 => if (Length - Cursor) >= 3 and then Subject (Cursor + 1 .. Cursor + 3) = Node.Str3 then Cursor := Cursor + 3; goto Succeed; else goto Fail; end if; -- String (four character case) when PC_String_4 => if (Length - Cursor) >= 4 and then Subject (Cursor + 1 .. Cursor + 4) = Node.Str4 then Cursor := Cursor + 4; goto Succeed; else goto Fail; end if; -- String (five character case) when PC_String_5 => if (Length - Cursor) >= 5 and then Subject (Cursor + 1 .. Cursor + 5) = Node.Str5 then Cursor := Cursor + 5; goto Succeed; else goto Fail; end if; -- String (six character case) when PC_String_6 => if (Length - Cursor) >= 6 and then Subject (Cursor + 1 .. Cursor + 6) = Node.Str6 then Cursor := Cursor + 6; goto Succeed; else goto Fail; end if; -- String (case of more than six characters) when PC_String => declare Len : constant Natural := Node.Str'Length; begin if (Length - Cursor) >= Len and then Node.Str.all = Subject (Cursor + 1 .. Cursor + Len) then Cursor := Cursor + Len; goto Succeed; else goto Fail; end if; end; -- String (function case) when PC_String_VF => declare U : constant VString := Node.VF.all; S : String_Access; L : Natural; begin Get_String (U, S, L); if (Length - Cursor) >= L and then S (1 .. L) = Subject (Cursor + 1 .. Cursor + L) then Cursor := Cursor + L; goto Succeed; else goto Fail; end if; end; -- String (pointer case) when PC_String_VP => declare U : constant VString := Node.VP.all; S : String_Access; L : Natural; begin Get_String (U, S, L); if (Length - Cursor) >= L and then S (1 .. L) = Subject (Cursor + 1 .. Cursor + L) then Cursor := Cursor + L; goto Succeed; else goto Fail; end if; end; -- Succeed when PC_Succeed => Push (Node); goto Succeed; -- Tab (integer case) when PC_Tab_Nat => if Cursor <= Node.Nat then Cursor := Node.Nat; goto Succeed; else goto Fail; end if; -- Tab (integer function case) when PC_Tab_NF => declare N : constant Natural := Node.NF.all; begin if Cursor <= N then Cursor := N; goto Succeed; else goto Fail; end if; end; -- Tab (integer pointer case) when PC_Tab_NP => if Cursor <= Node.NP.all then Cursor := Node.NP.all; goto Succeed; else goto Fail; end if; -- Unanchored movement when PC_Unanchored => -- All done if we tried every position if Cursor > Length then goto Match_Fail; -- Otherwise extend the anchor point, and restack ourself else Cursor := Cursor + 1; Push (Node); goto Succeed; end if; -- Write immediate. This node performs the actual write when PC_Write_Imm => Put_Line (Node.FP.all, Subject (Stack (Stack_Base - 1).Cursor + 1 .. Cursor)); Pop_Region; goto Succeed; -- Write on match. This node sets up for the eventual write when PC_Write_OnM => Stack (Stack_Base - 1).Node := Node; Push (CP_Assign'Access); Pop_Region; Assign_OnM := True; goto Succeed; end case; -- We are NOT allowed to fall though this case statement, since every -- match routine must end by executing a goto to the appropriate point -- in the finite state machine model. pragma Warnings (Off); Logic_Error; pragma Warnings (On); end XMatch; ------------- -- XMatchD -- ------------- -- Maintenance note: There is a LOT of code duplication between XMatch -- and XMatchD. This is quite intentional, the point is to avoid any -- unnecessary debugging overhead in the XMatch case, but this does mean -- that any changes to XMatchD must be mirrored in XMatch. In case of -- any major changes, the proper approach is to delete XMatch, make the -- changes to XMatchD, and then make a copy of XMatchD, removing all -- calls to Dout, and all Put and Put_Line operations. This copy becomes -- the new XMatch. procedure XMatchD (Subject : String; Pat_P : PE_Ptr; Pat_S : Natural; Start : out Natural; Stop : out Natural) is Node : PE_Ptr; -- Pointer to current pattern node. Initialized from Pat_P, and then -- updated as the match proceeds through its constituent elements. Length : constant Natural := Subject'Length; -- Length of string (= Subject'Last, since Subject'First is always 1) Cursor : Integer := 0; -- If the value is non-negative, then this value is the index showing -- the current position of the match in the subject string. The next -- character to be matched is at Subject (Cursor + 1). Note that since -- our view of the subject string in XMatch always has a lower bound -- of one, regardless of original bounds, that this definition exactly -- corresponds to the cursor value as referenced by functions like Pos. -- -- If the value is negative, then this is a saved stack pointer, -- typically a base pointer of an inner or outer region. Cursor -- temporarily holds such a value when it is popped from the stack -- by Fail. In all cases, Cursor is reset to a proper non-negative -- cursor value before the match proceeds (e.g. by propagating the -- failure and popping a "real" cursor value from the stack. PE_Unanchored : aliased PE := (PC_Unanchored, 0, Pat_P); -- Dummy pattern element used in the unanchored case Region_Level : Natural := 0; -- Keeps track of recursive region level. This is used only for -- debugging, it is the number of saved history stack base values. Stack : Stack_Type; -- The pattern matching failure stack for this call to Match Stack_Ptr : Stack_Range; -- Current stack pointer. This points to the top element of the stack -- that is currently in use. At the outer level this is the special -- entry placed on the stack according to the anchor mode. Stack_Init : constant Stack_Range := Stack'First + 1; -- This is the initial value of the Stack_Ptr and Stack_Base. The -- initial (Stack'First) element of the stack is not used so that -- when we pop the last element off, Stack_Ptr is still in range. Stack_Base : Stack_Range; -- This value is the stack base value, i.e. the stack pointer for the -- first history stack entry in the current stack region. See separate -- section on handling of recursive pattern matches. Assign_OnM : Boolean := False; -- Set True if assign-on-match or write-on-match operations may be -- present in the history stack, which must then be scanned on a -- successful match. procedure Dout (Str : String); -- Output string to standard error with bars indicating region level procedure Dout (Str : String; A : Character); -- Calls Dout with the string S ('A') procedure Dout (Str : String; A : Character_Set); -- Calls Dout with the string S ("A") procedure Dout (Str : String; A : Natural); -- Calls Dout with the string S (A) procedure Dout (Str : String; A : String); -- Calls Dout with the string S ("A") function Img (P : PE_Ptr) return String; -- Returns a string of the form #nnn where nnn is P.Index procedure Pop_Region; pragma Inline (Pop_Region); -- Used at the end of processing of an inner region. If the inner -- region left no stack entries, then all trace of it is removed. -- Otherwise a PC_Restore_Region entry is pushed to ensure proper -- handling of alternatives in the inner region. procedure Push (Node : PE_Ptr); pragma Inline (Push); -- Make entry in pattern matching stack with current cursor value procedure Push_Region; pragma Inline (Push_Region); -- This procedure makes a new region on the history stack. The -- caller first establishes the special entry on the stack, but -- does not push the stack pointer. Then this call stacks a -- PC_Remove_Region node, on top of this entry, using the cursor -- field of the PC_Remove_Region entry to save the outer level -- stack base value, and resets the stack base to point to this -- PC_Remove_Region node. ---------- -- Dout -- ---------- procedure Dout (Str : String) is begin for J in 1 .. Region_Level loop Put ("| "); end loop; Put_Line (Str); end Dout; procedure Dout (Str : String; A : Character) is begin Dout (Str & " ('" & A & "')"); end Dout; procedure Dout (Str : String; A : Character_Set) is begin Dout (Str & " (" & Image (To_Sequence (A)) & ')'); end Dout; procedure Dout (Str : String; A : Natural) is begin Dout (Str & " (" & A & ')'); end Dout; procedure Dout (Str : String; A : String) is begin Dout (Str & " (" & Image (A) & ')'); end Dout; --------- -- Img -- --------- function Img (P : PE_Ptr) return String is begin return "#" & Integer (P.Index) & " "; end Img; ---------------- -- Pop_Region -- ---------------- procedure Pop_Region is begin Region_Level := Region_Level - 1; -- If nothing was pushed in the inner region, we can just get -- rid of it entirely, leaving no traces that it was ever there if Stack_Ptr = Stack_Base then Stack_Ptr := Stack_Base - 2; Stack_Base := Stack (Stack_Ptr + 2).Cursor; -- If stuff was pushed in the inner region, then we have to -- push a PC_R_Restore node so that we properly handle possible -- rematches within the region. else Stack_Ptr := Stack_Ptr + 1; Stack (Stack_Ptr).Cursor := Stack_Base; Stack (Stack_Ptr).Node := CP_R_Restore'Access; Stack_Base := Stack (Stack_Base).Cursor; end if; end Pop_Region; ---------- -- Push -- ---------- procedure Push (Node : PE_Ptr) is begin Stack_Ptr := Stack_Ptr + 1; Stack (Stack_Ptr).Cursor := Cursor; Stack (Stack_Ptr).Node := Node; end Push; ----------------- -- Push_Region -- ----------------- procedure Push_Region is begin Region_Level := Region_Level + 1; Stack_Ptr := Stack_Ptr + 2; Stack (Stack_Ptr).Cursor := Stack_Base; Stack (Stack_Ptr).Node := CP_R_Remove'Access; Stack_Base := Stack_Ptr; end Push_Region; -- Start of processing for XMatchD begin New_Line; Put_Line ("Initiating pattern match, subject = " & Image (Subject)); Put ("--------------------------------------"); for J in 1 .. Length loop Put ('-'); end loop; New_Line; Put_Line ("subject length = " & Length); if Pat_P = null then Uninitialized_Pattern; end if; -- Check we have enough stack for this pattern. This check deals with -- every possibility except a match of a recursive pattern, where we -- make a check at each recursion level. if Pat_S >= Stack_Size - 1 then raise Pattern_Stack_Overflow; end if; -- In anchored mode, the bottom entry on the stack is an abort entry if Anchored_Mode then Stack (Stack_Init).Node := CP_Cancel'Access; Stack (Stack_Init).Cursor := 0; -- In unanchored more, the bottom entry on the stack references -- the special pattern element PE_Unanchored, whose Pthen field -- points to the initial pattern element. The cursor value in this -- entry is the number of anchor moves so far. else Stack (Stack_Init).Node := PE_Unanchored'Unchecked_Access; Stack (Stack_Init).Cursor := 0; end if; Stack_Ptr := Stack_Init; Stack_Base := Stack_Ptr; Cursor := 0; Node := Pat_P; goto Match; ----------------------------------------- -- Main Pattern Matching State Control -- ----------------------------------------- -- This is a state machine which uses gotos to change state. The -- initial state is Match, to initiate the matching of the first -- element, so the goto Match above starts the match. In the -- following descriptions, we indicate the global values that -- are relevant for the state transition. -- Come here if entire match fails <<Match_Fail>> Dout ("match fails"); New_Line; Start := 0; Stop := 0; return; -- Come here if entire match succeeds -- Cursor current position in subject string <<Match_Succeed>> Dout ("match succeeds"); Start := Stack (Stack_Init).Cursor + 1; Stop := Cursor; Dout ("first matched character index = " & Start); Dout ("last matched character index = " & Stop); Dout ("matched substring = " & Image (Subject (Start .. Stop))); -- Scan history stack for deferred assignments or writes if Assign_OnM then for S in Stack'First .. Stack_Ptr loop if Stack (S).Node = CP_Assign'Access then declare Inner_Base : constant Stack_Range := Stack (S + 1).Cursor; Special_Entry : constant Stack_Range := Inner_Base - 1; Node_OnM : constant PE_Ptr := Stack (Special_Entry).Node; Start : constant Natural := Stack (Special_Entry).Cursor + 1; Stop : constant Natural := Stack (S).Cursor; begin if Node_OnM.Pcode = PC_Assign_OnM then Set_String (Node_OnM.VP.all, Subject (Start .. Stop)); Dout (Img (Stack (S).Node) & "deferred assignment of " & Image (Subject (Start .. Stop))); elsif Node_OnM.Pcode = PC_Write_OnM then Put_Line (Node_OnM.FP.all, Subject (Start .. Stop)); Dout (Img (Stack (S).Node) & "deferred write of " & Image (Subject (Start .. Stop))); else Logic_Error; end if; end; end if; end loop; end if; New_Line; return; -- Come here if attempt to match current element fails -- Stack_Base current stack base -- Stack_Ptr current stack pointer <<Fail>> Cursor := Stack (Stack_Ptr).Cursor; Node := Stack (Stack_Ptr).Node; Stack_Ptr := Stack_Ptr - 1; if Cursor >= 0 then Dout ("failure, cursor reset to " & Cursor); end if; goto Match; -- Come here if attempt to match current element succeeds -- Cursor current position in subject string -- Node pointer to node successfully matched -- Stack_Base current stack base -- Stack_Ptr current stack pointer <<Succeed>> Dout ("success, cursor = " & Cursor); Node := Node.Pthen; -- Come here to match the next pattern element -- Cursor current position in subject string -- Node pointer to node to be matched -- Stack_Base current stack base -- Stack_Ptr current stack pointer <<Match>> -------------------------------------------------- -- Main Pattern Match Element Matching Routines -- -------------------------------------------------- -- Here is the case statement that processes the current node. The -- processing for each element does one of five things: -- goto Succeed to move to the successor -- goto Match_Succeed if the entire match succeeds -- goto Match_Fail if the entire match fails -- goto Fail to signal failure of current match -- Processing is NOT allowed to fall through case Node.Pcode is -- Cancel when PC_Cancel => Dout (Img (Node) & "matching Cancel"); goto Match_Fail; -- Alternation when PC_Alt => Dout (Img (Node) & "setting up alternative " & Img (Node.Alt)); Push (Node.Alt); Node := Node.Pthen; goto Match; -- Any (one character case) when PC_Any_CH => Dout (Img (Node) & "matching Any", Node.Char); if Cursor < Length and then Subject (Cursor + 1) = Node.Char then Cursor := Cursor + 1; goto Succeed; else goto Fail; end if; -- Any (character set case) when PC_Any_CS => Dout (Img (Node) & "matching Any", Node.CS); if Cursor < Length and then Is_In (Subject (Cursor + 1), Node.CS) then Cursor := Cursor + 1; goto Succeed; else goto Fail; end if; -- Any (string function case) when PC_Any_VF => declare U : constant VString := Node.VF.all; S : String_Access; L : Natural; begin Get_String (U, S, L); Dout (Img (Node) & "matching Any", S (1 .. L)); if Cursor < Length and then Is_In (Subject (Cursor + 1), S (1 .. L)) then Cursor := Cursor + 1; goto Succeed; else goto Fail; end if; end; -- Any (string pointer case) when PC_Any_VP => declare U : constant VString := Node.VP.all; S : String_Access; L : Natural; begin Get_String (U, S, L); Dout (Img (Node) & "matching Any", S (1 .. L)); if Cursor < Length and then Is_In (Subject (Cursor + 1), S (1 .. L)) then Cursor := Cursor + 1; goto Succeed; else goto Fail; end if; end; -- Arb (initial match) when PC_Arb_X => Dout (Img (Node) & "matching Arb"); Push (Node.Alt); Node := Node.Pthen; goto Match; -- Arb (extension) when PC_Arb_Y => Dout (Img (Node) & "extending Arb"); if Cursor < Length then Cursor := Cursor + 1; Push (Node); goto Succeed; else goto Fail; end if; -- Arbno_S (simple Arbno initialize). This is the node that -- initiates the match of a simple Arbno structure. when PC_Arbno_S => Dout (Img (Node) & "setting up Arbno alternative " & Img (Node.Alt)); Push (Node.Alt); Node := Node.Pthen; goto Match; -- Arbno_X (Arbno initialize). This is the node that initiates -- the match of a complex Arbno structure. when PC_Arbno_X => Dout (Img (Node) & "setting up Arbno alternative " & Img (Node.Alt)); Push (Node.Alt); Node := Node.Pthen; goto Match; -- Arbno_Y (Arbno rematch). This is the node that is executed -- following successful matching of one instance of a complex -- Arbno pattern. when PC_Arbno_Y => declare Null_Match : constant Boolean := Cursor = Stack (Stack_Base - 1).Cursor; begin Dout (Img (Node) & "extending Arbno"); Pop_Region; -- If arbno extension matched null, then immediately fail if Null_Match then Dout ("Arbno extension matched null, so fails"); goto Fail; end if; -- Here we must do a stack check to make sure enough stack -- is left. This check will happen once for each instance of -- the Arbno pattern that is matched. The Nat field of a -- PC_Arbno pattern contains the maximum stack entries needed -- for the Arbno with one instance and the successor pattern if Stack_Ptr + Node.Nat >= Stack'Last then raise Pattern_Stack_Overflow; end if; goto Succeed; end; -- Assign. If this node is executed, it means the assign-on-match -- or write-on-match operation will not happen after all, so we -- is propagate the failure, removing the PC_Assign node. when PC_Assign => Dout (Img (Node) & "deferred assign/write cancelled"); goto Fail; -- Assign immediate. This node performs the actual assignment when PC_Assign_Imm => Dout (Img (Node) & "executing immediate assignment of " & Image (Subject (Stack (Stack_Base - 1).Cursor + 1 .. Cursor))); Set_String (Node.VP.all, Subject (Stack (Stack_Base - 1).Cursor + 1 .. Cursor)); Pop_Region; goto Succeed; -- Assign on match. This node sets up for the eventual assignment when PC_Assign_OnM => Dout (Img (Node) & "registering deferred assignment"); Stack (Stack_Base - 1).Node := Node; Push (CP_Assign'Access); Pop_Region; Assign_OnM := True; goto Succeed; -- Bal when PC_Bal => Dout (Img (Node) & "matching or extending Bal"); if Cursor >= Length or else Subject (Cursor + 1) = ')' then goto Fail; elsif Subject (Cursor + 1) = '(' then declare Paren_Count : Natural := 1; begin loop Cursor := Cursor + 1; if Cursor >= Length then goto Fail; elsif Subject (Cursor + 1) = '(' then Paren_Count := Paren_Count + 1; elsif Subject (Cursor + 1) = ')' then Paren_Count := Paren_Count - 1; exit when Paren_Count = 0; end if; end loop; end; end if; Cursor := Cursor + 1; Push (Node); goto Succeed; -- Break (one character case) when PC_Break_CH => Dout (Img (Node) & "matching Break", Node.Char); while Cursor < Length loop if Subject (Cursor + 1) = Node.Char then goto Succeed; else Cursor := Cursor + 1; end if; end loop; goto Fail; -- Break (character set case) when PC_Break_CS => Dout (Img (Node) & "matching Break", Node.CS); while Cursor < Length loop if Is_In (Subject (Cursor + 1), Node.CS) then goto Succeed; else Cursor := Cursor + 1; end if; end loop; goto Fail; -- Break (string function case) when PC_Break_VF => declare U : constant VString := Node.VF.all; S : String_Access; L : Natural; begin Get_String (U, S, L); Dout (Img (Node) & "matching Break", S (1 .. L)); while Cursor < Length loop if Is_In (Subject (Cursor + 1), S (1 .. L)) then goto Succeed; else Cursor := Cursor + 1; end if; end loop; goto Fail; end; -- Break (string pointer case) when PC_Break_VP => declare U : constant VString := Node.VP.all; S : String_Access; L : Natural; begin Get_String (U, S, L); Dout (Img (Node) & "matching Break", S (1 .. L)); while Cursor < Length loop if Is_In (Subject (Cursor + 1), S (1 .. L)) then goto Succeed; else Cursor := Cursor + 1; end if; end loop; goto Fail; end; -- BreakX (one character case) when PC_BreakX_CH => Dout (Img (Node) & "matching BreakX", Node.Char); while Cursor < Length loop if Subject (Cursor + 1) = Node.Char then goto Succeed; else Cursor := Cursor + 1; end if; end loop; goto Fail; -- BreakX (character set case) when PC_BreakX_CS => Dout (Img (Node) & "matching BreakX", Node.CS); while Cursor < Length loop if Is_In (Subject (Cursor + 1), Node.CS) then goto Succeed; else Cursor := Cursor + 1; end if; end loop; goto Fail; -- BreakX (string function case) when PC_BreakX_VF => declare U : constant VString := Node.VF.all; S : String_Access; L : Natural; begin Get_String (U, S, L); Dout (Img (Node) & "matching BreakX", S (1 .. L)); while Cursor < Length loop if Is_In (Subject (Cursor + 1), S (1 .. L)) then goto Succeed; else Cursor := Cursor + 1; end if; end loop; goto Fail; end; -- BreakX (string pointer case) when PC_BreakX_VP => declare U : constant VString := Node.VP.all; S : String_Access; L : Natural; begin Get_String (U, S, L); Dout (Img (Node) & "matching BreakX", S (1 .. L)); while Cursor < Length loop if Is_In (Subject (Cursor + 1), S (1 .. L)) then goto Succeed; else Cursor := Cursor + 1; end if; end loop; goto Fail; end; -- BreakX_X (BreakX extension). See section on "Compound Pattern -- Structures". This node is the alternative that is stacked -- to skip past the break character and extend the break. when PC_BreakX_X => Dout (Img (Node) & "extending BreakX"); Cursor := Cursor + 1; goto Succeed; -- Character (one character string) when PC_Char => Dout (Img (Node) & "matching '" & Node.Char & '''); if Cursor < Length and then Subject (Cursor + 1) = Node.Char then Cursor := Cursor + 1; goto Succeed; else goto Fail; end if; -- End of Pattern when PC_EOP => if Stack_Base = Stack_Init then Dout ("end of pattern"); goto Match_Succeed; -- End of recursive inner match. See separate section on -- handing of recursive pattern matches for details. else Dout ("terminating recursive match"); Node := Stack (Stack_Base - 1).Node; Pop_Region; goto Match; end if; -- Fail when PC_Fail => Dout (Img (Node) & "matching Fail"); goto Fail; -- Fence (built in pattern) when PC_Fence => Dout (Img (Node) & "matching Fence"); Push (CP_Cancel'Access); goto Succeed; -- Fence function node X. This is the node that gets control -- after a successful match of the fenced pattern. when PC_Fence_X => Dout (Img (Node) & "matching Fence function"); Stack_Ptr := Stack_Ptr + 1; Stack (Stack_Ptr).Cursor := Stack_Base; Stack (Stack_Ptr).Node := CP_Fence_Y'Access; Stack_Base := Stack (Stack_Base).Cursor; Region_Level := Region_Level - 1; goto Succeed; -- Fence function node Y. This is the node that gets control on -- a failure that occurs after the fenced pattern has matched. -- Note: the Cursor at this stage is actually the inner stack -- base value. We don't reset this, but we do use it to strip -- off all the entries made by the fenced pattern. when PC_Fence_Y => Dout (Img (Node) & "pattern matched by Fence caused failure"); Stack_Ptr := Cursor - 2; goto Fail; -- Len (integer case) when PC_Len_Nat => Dout (Img (Node) & "matching Len", Node.Nat); if Cursor + Node.Nat > Length then goto Fail; else Cursor := Cursor + Node.Nat; goto Succeed; end if; -- Len (Integer function case) when PC_Len_NF => declare N : constant Natural := Node.NF.all; begin Dout (Img (Node) & "matching Len", N); if Cursor + N > Length then goto Fail; else Cursor := Cursor + N; goto Succeed; end if; end; -- Len (integer pointer case) when PC_Len_NP => Dout (Img (Node) & "matching Len", Node.NP.all); if Cursor + Node.NP.all > Length then goto Fail; else Cursor := Cursor + Node.NP.all; goto Succeed; end if; -- NotAny (one character case) when PC_NotAny_CH => Dout (Img (Node) & "matching NotAny", Node.Char); if Cursor < Length and then Subject (Cursor + 1) /= Node.Char then Cursor := Cursor + 1; goto Succeed; else goto Fail; end if; -- NotAny (character set case) when PC_NotAny_CS => Dout (Img (Node) & "matching NotAny", Node.CS); if Cursor < Length and then not Is_In (Subject (Cursor + 1), Node.CS) then Cursor := Cursor + 1; goto Succeed; else goto Fail; end if; -- NotAny (string function case) when PC_NotAny_VF => declare U : constant VString := Node.VF.all; S : String_Access; L : Natural; begin Get_String (U, S, L); Dout (Img (Node) & "matching NotAny", S (1 .. L)); if Cursor < Length and then not Is_In (Subject (Cursor + 1), S (1 .. L)) then Cursor := Cursor + 1; goto Succeed; else goto Fail; end if; end; -- NotAny (string pointer case) when PC_NotAny_VP => declare U : constant VString := Node.VP.all; S : String_Access; L : Natural; begin Get_String (U, S, L); Dout (Img (Node) & "matching NotAny", S (1 .. L)); if Cursor < Length and then not Is_In (Subject (Cursor + 1), S (1 .. L)) then Cursor := Cursor + 1; goto Succeed; else goto Fail; end if; end; -- NSpan (one character case) when PC_NSpan_CH => Dout (Img (Node) & "matching NSpan", Node.Char); while Cursor < Length and then Subject (Cursor + 1) = Node.Char loop Cursor := Cursor + 1; end loop; goto Succeed; -- NSpan (character set case) when PC_NSpan_CS => Dout (Img (Node) & "matching NSpan", Node.CS); while Cursor < Length and then Is_In (Subject (Cursor + 1), Node.CS) loop Cursor := Cursor + 1; end loop; goto Succeed; -- NSpan (string function case) when PC_NSpan_VF => declare U : constant VString := Node.VF.all; S : String_Access; L : Natural; begin Get_String (U, S, L); Dout (Img (Node) & "matching NSpan", S (1 .. L)); while Cursor < Length and then Is_In (Subject (Cursor + 1), S (1 .. L)) loop Cursor := Cursor + 1; end loop; goto Succeed; end; -- NSpan (string pointer case) when PC_NSpan_VP => declare U : constant VString := Node.VP.all; S : String_Access; L : Natural; begin Get_String (U, S, L); Dout (Img (Node) & "matching NSpan", S (1 .. L)); while Cursor < Length and then Is_In (Subject (Cursor + 1), S (1 .. L)) loop Cursor := Cursor + 1; end loop; goto Succeed; end; when PC_Null => Dout (Img (Node) & "matching null"); goto Succeed; -- Pos (integer case) when PC_Pos_Nat => Dout (Img (Node) & "matching Pos", Node.Nat); if Cursor = Node.Nat then goto Succeed; else goto Fail; end if; -- Pos (Integer function case) when PC_Pos_NF => declare N : constant Natural := Node.NF.all; begin Dout (Img (Node) & "matching Pos", N); if Cursor = N then goto Succeed; else goto Fail; end if; end; -- Pos (integer pointer case) when PC_Pos_NP => Dout (Img (Node) & "matching Pos", Node.NP.all); if Cursor = Node.NP.all then goto Succeed; else goto Fail; end if; -- Predicate function when PC_Pred_Func => Dout (Img (Node) & "matching predicate function"); if Node.BF.all then goto Succeed; else goto Fail; end if; -- Region Enter. Initiate new pattern history stack region when PC_R_Enter => Dout (Img (Node) & "starting match of nested pattern"); Stack (Stack_Ptr + 1).Cursor := Cursor; Push_Region; goto Succeed; -- Region Remove node. This is the node stacked by an R_Enter. -- It removes the special format stack entry right underneath, and -- then restores the outer level stack base and signals failure. -- Note: the cursor value at this stage is actually the (negative) -- stack base value for the outer level. when PC_R_Remove => Dout ("failure, match of nested pattern terminated"); Stack_Base := Cursor; Region_Level := Region_Level - 1; Stack_Ptr := Stack_Ptr - 1; goto Fail; -- Region restore node. This is the node stacked at the end of an -- inner level match. Its function is to restore the inner level -- region, so that alternatives in this region can be sought. -- Note: the Cursor at this stage is actually the negative of the -- inner stack base value, which we use to restore the inner region. when PC_R_Restore => Dout ("failure, search for alternatives in nested pattern"); Region_Level := Region_Level + 1; Stack_Base := Cursor; goto Fail; -- Rest when PC_Rest => Dout (Img (Node) & "matching Rest"); Cursor := Length; goto Succeed; -- Initiate recursive match (pattern pointer case) when PC_Rpat => Stack (Stack_Ptr + 1).Node := Node.Pthen; Push_Region; Dout (Img (Node) & "initiating recursive match"); if Stack_Ptr + Node.PP.all.Stk >= Stack_Size then raise Pattern_Stack_Overflow; else Node := Node.PP.all.P; goto Match; end if; -- RPos (integer case) when PC_RPos_Nat => Dout (Img (Node) & "matching RPos", Node.Nat); if Cursor = (Length - Node.Nat) then goto Succeed; else goto Fail; end if; -- RPos (integer function case) when PC_RPos_NF => declare N : constant Natural := Node.NF.all; begin Dout (Img (Node) & "matching RPos", N); if Length - Cursor = N then goto Succeed; else goto Fail; end if; end; -- RPos (integer pointer case) when PC_RPos_NP => Dout (Img (Node) & "matching RPos", Node.NP.all); if Cursor = (Length - Node.NP.all) then goto Succeed; else goto Fail; end if; -- RTab (integer case) when PC_RTab_Nat => Dout (Img (Node) & "matching RTab", Node.Nat); if Cursor <= (Length - Node.Nat) then Cursor := Length - Node.Nat; goto Succeed; else goto Fail; end if; -- RTab (integer function case) when PC_RTab_NF => declare N : constant Natural := Node.NF.all; begin Dout (Img (Node) & "matching RPos", N); if Length - Cursor >= N then Cursor := Length - N; goto Succeed; else goto Fail; end if; end; -- RTab (integer pointer case) when PC_RTab_NP => Dout (Img (Node) & "matching RPos", Node.NP.all); if Cursor <= (Length - Node.NP.all) then Cursor := Length - Node.NP.all; goto Succeed; else goto Fail; end if; -- Cursor assignment when PC_Setcur => Dout (Img (Node) & "matching Setcur"); Node.Var.all := Cursor; goto Succeed; -- Span (one character case) when PC_Span_CH => declare P : Natural := Cursor; begin Dout (Img (Node) & "matching Span", Node.Char); while P < Length and then Subject (P + 1) = Node.Char loop P := P + 1; end loop; if P /= Cursor then Cursor := P; goto Succeed; else goto Fail; end if; end; -- Span (character set case) when PC_Span_CS => declare P : Natural := Cursor; begin Dout (Img (Node) & "matching Span", Node.CS); while P < Length and then Is_In (Subject (P + 1), Node.CS) loop P := P + 1; end loop; if P /= Cursor then Cursor := P; goto Succeed; else goto Fail; end if; end; -- Span (string function case) when PC_Span_VF => declare U : constant VString := Node.VF.all; S : String_Access; L : Natural; P : Natural; begin Get_String (U, S, L); Dout (Img (Node) & "matching Span", S (1 .. L)); P := Cursor; while P < Length and then Is_In (Subject (P + 1), S (1 .. L)) loop P := P + 1; end loop; if P /= Cursor then Cursor := P; goto Succeed; else goto Fail; end if; end; -- Span (string pointer case) when PC_Span_VP => declare U : constant VString := Node.VP.all; S : String_Access; L : Natural; P : Natural; begin Get_String (U, S, L); Dout (Img (Node) & "matching Span", S (1 .. L)); P := Cursor; while P < Length and then Is_In (Subject (P + 1), S (1 .. L)) loop P := P + 1; end loop; if P /= Cursor then Cursor := P; goto Succeed; else goto Fail; end if; end; -- String (two character case) when PC_String_2 => Dout (Img (Node) & "matching " & Image (Node.Str2)); if (Length - Cursor) >= 2 and then Subject (Cursor + 1 .. Cursor + 2) = Node.Str2 then Cursor := Cursor + 2; goto Succeed; else goto Fail; end if; -- String (three character case) when PC_String_3 => Dout (Img (Node) & "matching " & Image (Node.Str3)); if (Length - Cursor) >= 3 and then Subject (Cursor + 1 .. Cursor + 3) = Node.Str3 then Cursor := Cursor + 3; goto Succeed; else goto Fail; end if; -- String (four character case) when PC_String_4 => Dout (Img (Node) & "matching " & Image (Node.Str4)); if (Length - Cursor) >= 4 and then Subject (Cursor + 1 .. Cursor + 4) = Node.Str4 then Cursor := Cursor + 4; goto Succeed; else goto Fail; end if; -- String (five character case) when PC_String_5 => Dout (Img (Node) & "matching " & Image (Node.Str5)); if (Length - Cursor) >= 5 and then Subject (Cursor + 1 .. Cursor + 5) = Node.Str5 then Cursor := Cursor + 5; goto Succeed; else goto Fail; end if; -- String (six character case) when PC_String_6 => Dout (Img (Node) & "matching " & Image (Node.Str6)); if (Length - Cursor) >= 6 and then Subject (Cursor + 1 .. Cursor + 6) = Node.Str6 then Cursor := Cursor + 6; goto Succeed; else goto Fail; end if; -- String (case of more than six characters) when PC_String => declare Len : constant Natural := Node.Str'Length; begin Dout (Img (Node) & "matching " & Image (Node.Str.all)); if (Length - Cursor) >= Len and then Node.Str.all = Subject (Cursor + 1 .. Cursor + Len) then Cursor := Cursor + Len; goto Succeed; else goto Fail; end if; end; -- String (function case) when PC_String_VF => declare U : constant VString := Node.VF.all; S : String_Access; L : Natural; begin Get_String (U, S, L); Dout (Img (Node) & "matching " & Image (S (1 .. L))); if (Length - Cursor) >= L and then S (1 .. L) = Subject (Cursor + 1 .. Cursor + L) then Cursor := Cursor + L; goto Succeed; else goto Fail; end if; end; -- String (vstring pointer case) when PC_String_VP => declare U : constant VString := Node.VP.all; S : String_Access; L : Natural; begin Get_String (U, S, L); Dout (Img (Node) & "matching " & Image (S (1 .. L))); if (Length - Cursor) >= L and then S (1 .. L) = Subject (Cursor + 1 .. Cursor + L) then Cursor := Cursor + L; goto Succeed; else goto Fail; end if; end; -- Succeed when PC_Succeed => Dout (Img (Node) & "matching Succeed"); Push (Node); goto Succeed; -- Tab (integer case) when PC_Tab_Nat => Dout (Img (Node) & "matching Tab", Node.Nat); if Cursor <= Node.Nat then Cursor := Node.Nat; goto Succeed; else goto Fail; end if; -- Tab (integer function case) when PC_Tab_NF => declare N : constant Natural := Node.NF.all; begin Dout (Img (Node) & "matching Tab ", N); if Cursor <= N then Cursor := N; goto Succeed; else goto Fail; end if; end; -- Tab (integer pointer case) when PC_Tab_NP => Dout (Img (Node) & "matching Tab ", Node.NP.all); if Cursor <= Node.NP.all then Cursor := Node.NP.all; goto Succeed; else goto Fail; end if; -- Unanchored movement when PC_Unanchored => Dout ("attempting to move anchor point"); -- All done if we tried every position if Cursor > Length then goto Match_Fail; -- Otherwise extend the anchor point, and restack ourself else Cursor := Cursor + 1; Push (Node); goto Succeed; end if; -- Write immediate. This node performs the actual write when PC_Write_Imm => Dout (Img (Node) & "executing immediate write of " & Subject (Stack (Stack_Base - 1).Cursor + 1 .. Cursor)); Put_Line (Node.FP.all, Subject (Stack (Stack_Base - 1).Cursor + 1 .. Cursor)); Pop_Region; goto Succeed; -- Write on match. This node sets up for the eventual write when PC_Write_OnM => Dout (Img (Node) & "registering deferred write"); Stack (Stack_Base - 1).Node := Node; Push (CP_Assign'Access); Pop_Region; Assign_OnM := True; goto Succeed; end case; -- We are NOT allowed to fall though this case statement, since every -- match routine must end by executing a goto to the appropriate point -- in the finite state machine model. pragma Warnings (Off); Logic_Error; pragma Warnings (On); end XMatchD; end GNAT.Spitbol.Patterns;
Go to most recent revision | Compare with Previous | Blame | View Log