pt::param(3tcl) Parser Tools pt::param(3tcl)
______________________________________________________________________________
NAME
pt::param - PackRat Machine Specification
SYNOPSIS
package require Tcl 8.5
______________________________________________________________________________
DESCRIPTION
Are you lost ? Do you have trouble understanding this document ? In
that case please read the overview provided by the Introduction to
Parser Tools. This document is the entrypoint to the whole system the
current package is a part of.
Welcome to the PackRat Machine (short: PARAM), a virtual machine geared
towards the support of recursive descent parsers, especially packrat
parsers. Towards this end it has features like the caching and reuse of
partial results, the caching of the encountered input, and the ability
to backtrack in both input and AST creation.
This document specifies the machine in terms of its architectural state
and instruction set.
ARCHITECTURAL STATE
Any PARAM implementation has to manage at least the following state:
Input (IN)
This is the channel the characters to process are read from.
This part of the machine's state is used and modified by the in-
structions defined in the section Input Handling.
Current Character (CC)
The character from the input currently tested against its possi-
ble alternatives.
This part of the machine's state is used and modified by the in-
structions defined in the section Character Processing.
Current Location (CL)
The location of the current character in the input, as offset
relative to the beginning of the input. Character offsets are
counted from 0.
This part of the machine's state is used and modified by the in-
structions defined in the sections Character Processing, Loca-
tion Handling, and Nonterminal Execution.
Location Stack (LS)
A stack of locations in the input, saved for possible backtrack-
ing.
This part of the machine's state is used and modified by the in-
structions defined in the sections Character Processing, Loca-
tion Handling, and Nonterminal Execution.
Status (ST)
The status of the last attempt of testing the input, indicating
either success or failure.
This part of the machine's state is used and modified by the in-
structions defined in the sections Status Control, Character
Processing, and Nonterminal Execution.
Semantic Value (SV)
The current semantic value, either empty, or a node for AST con-
structed from the input.
This part of the machine's state is used and modified by the in-
structions defined in the sections Value Construction, and AST
Construction.
AST Reduction Stack (ARS)
The stack of partial ASTs constructed during the processing of
nonterminal symbols.
This part of the machine's state is used and modified by the in-
structions defined in the sections Value Construction, and AST
Construction.
AST Stack (AS)
The stack of reduction stacks, saved for possible backtracking.
This part of the machine's state is used and modified by the in-
structions defined in the sections Value Construction, and AST
Construction.
Error Status (ER)
The machine's current knowledge of errors. This is either empty,
or set to a pair of location in the input and the set of mes-
sages for that location.
Note that this part of the machine's state can be set even if
the last test of the current character was successful. For exam-
ple, the *-operator (matching a sub-expression zero or more
times) in a PEG is always successful, even if it encounters a
problem further in the input and has to backtrack. Such problems
must not be forgotten when continuing the parsing.
This part of the machine's state is used and modified by the in-
structions defined in the sections Error Handling, Character
Processing, and Nonterminal Execution.
Error Stack (ES)
The stack of error stati, saved for backtracking. This enables
the machine to merge current and older error stati when perform-
ing backtracking in choices after an failed match.
This part of the machine's state is used and modified by the in-
structions defined in the sections Error Handling, Character
Processing, and Nonterminal Execution.
Nonterminal Cache (NC)
A cache of machine states keyed by pairs name of nonterminal
symbol and location in the input. Each pair (N, L) is associated
with a 4-tuple holding the values to use for CL, ST, SV, and ER
after the nonterminal N was parsed starting from the location L.
It is a performance aid for backtracking parsers, allowing them
to avoid an expensive reparsing of complex nonterminal symbols
if they have been encountered before at a given location.
The key location is where machine started the attempt to match
the named nonterminal symbol, and the location in the saved
4-tuple is where machine ended up after the attempt completed,
independent of the success of the attempt.
This part of the machine's state is used and modified by the in-
structions defined in the section Nonterminal Execution.
Terminal Cache (TC)
A cache of characters read from IN, with their location in IN as
pair of line and column, keyed by the location in IN, this time
as character offset from the beginning of IN. It is a perfor-
mance aid for backtracking parsers, allowing them to avoid a
possibly expensive rereading of characters from IN, or even en-
abling backtracking at, i.e. in the case of IN not randomly
seekable.
This part of the machine's state is used and modified by the in-
structions defined in the section Input Handling.
INSTRUCTION SET
With the machine's architectural state specified it is now possible to
specify the instruction set operating on that state and to be imple-
mented by any realization of the PARAM. The 37 instructions are grouped
roughly by the state they influence and/or query during their execu-
tion.
INPUT HANDLING
The instructions in this section mainly access IN, pulling the charac-
ters to process into the machine.
input_next msg
This method reads the next character, i.e. the character after
CL, from IN. If successful this character becomes CC, CL is ad-
vanced by one, ES is cleared, and the operation is recorded as a
success in ST.
The operation may read the character from IN if the next charac-
ter is not yet known to TC. If successful the new character is
stored in TC, with its location (line, column), and the opera-
tion otherwise behaves as specified above. Future reads from the
same location, possible due to backtracking, will then be satis-
fied from TC instead of IN.
If, on the other hand, the end of IN was reached, the operation
is recorded as failed in ST, CL is left unchanged, and the pair
of CL and msg becomes the new ES.
CHARACTER PROCESSING
The instructions in this section mainly access CC, testing it against
character classes, ranges, and individual characters.
test_alnum
This instruction implements the special PE operator "alnum",
which checks if CC falls into the character class of the same
name, or not.
Success and failure of the test are both recorded directly in
ST. Success further clears ES, wheras failure sets the pair of
CL and expected input (encoded as a leaf parsing expression) as
the new ES and then rewinds CL by one character, preparing the
machine for another parse attempt by a possible alternative.
test_alpha
This instruction implements the special PE operator "alpha",
which checks if CC falls into the character class of the same
name, or not.
Success and failure of the test are both recorded directly in
ST. Success further clears ES, wheras failure sets the pair of
CL and expected input (encoded as a leaf parsing expression) as
the new ES and then rewinds CL by one character, preparing the
machine for another parse attempt by a possible alternative.
test_ascii
This instruction implements the special PE operator "ascii",
which checks if CC falls into the character class of the same
name, or not.
Success and failure of the test are both recorded directly in
ST. Success further clears ES, wheras failure sets the pair of
CL and expected input (encoded as a leaf parsing expression) as
the new ES and then rewinds CL by one character, preparing the
machine for another parse attempt by a possible alternative.
test_char char
This instruction implements the character matching operator,
i.e. it checks if CC is char.
Success and failure of the test are both recorded directly in
ST. Success further clears ES, wheras failure sets the pair of
CL and expected input (encoded as a leaf parsing expression) as
the new ES and then rewinds CL by one character, preparing the
machine for another parse attempt by a possible alternative.
test_ddigit
This instruction implements the special PE operator "ddigit",
which checks if CC falls into the character class of the same
name, or not.
Success and failure of the test are both recorded directly in
ST. Success further clears ES, wheras failure sets the pair of
CL and expected input (encoded as a leaf parsing expression) as
the new ES and then rewinds CL by one character, preparing the
machine for another parse attempt by a possible alternative.
test_digit
This instruction implements the special PE operator "digit",
which checks if CC falls into the character class of the same
name, or not.
Success and failure of the test are both recorded directly in
ST. Success further clears ES, wheras failure sets the pair of
CL and expected input (encoded as a leaf parsing expression) as
the new ES and then rewinds CL by one character, preparing the
machine for another parse attempt by a possible alternative.
test_graph
This instruction implements the special PE operator "graph",
which checks if CC falls into the character class of the same
name, or not.
Success and failure of the test are both recorded directly in
ST. Success further clears ES, wheras failure sets the pair of
CL and expected input (encoded as a leaf parsing expression) as
the new ES and then rewinds CL by one character, preparing the
machine for another parse attempt by a possible alternative.
test_lower
This instruction implements the special PE operator "lower",
which checks if CC falls into the character class of the same
name, or not.
Success and failure of the test are both recorded directly in
ST. Success further clears ES, wheras failure sets the pair of
CL and expected input (encoded as a leaf parsing expression) as
the new ES and then rewinds CL by one character, preparing the
machine for another parse attempt by a possible alternative.
test_print
This instruction implements the special PE operator "print",
which checks if CC falls into the character class of the same
name, or not.
Success and failure of the test are both recorded directly in
ST. Success further clears ES, wheras failure sets the pair of
CL and expected input (encoded as a leaf parsing expression) as
the new ES and then rewinds CL by one character, preparing the
machine for another parse attempt by a possible alternative.
test_punct
This instruction implements the special PE operator "punct",
which checks if CC falls into the character class of the same
name, or not.
Success and failure of the test are both recorded directly in
ST. Success further clears ES, wheras failure sets the pair of
CL and expected input (encoded as a leaf parsing expression) as
the new ES and then rewinds CL by one character, preparing the
machine for another parse attempt by a possible alternative.
test_range chars chare
This instruction implements the range matching operator, i.e. it
checks if CC falls into the interval of characters spanned up by
the two characters from chars to chare, both inclusive.
Success and failure of the test are both recorded directly in
ST. Success further clears ES, wheras failure sets the pair of
CL and expected input (encoded as a leaf parsing expression) as
the new ES and then rewinds CL by one character, preparing the
machine for another parse attempt by a possible alternative.
test_space
This instruction implements the special PE operator "space",
which checks if CC falls into the character class of the same
name, or not.
Success and failure of the test are both recorded directly in
ST. Success further clears ES, wheras failure sets the pair of
CL and expected input (encoded as a leaf parsing expression) as
the new ES and then rewinds CL by one character, preparing the
machine for another parse attempt by a possible alternative.
test_upper
This instruction implements the special PE operator "upper",
which checks if CC falls into the character class of the same
name, or not.
Success and failure of the test are both recorded directly in
ST. Success further clears ES, wheras failure sets the pair of
CL and expected input (encoded as a leaf parsing expression) as
the new ES and then rewinds CL by one character, preparing the
machine for another parse attempt by a possible alternative.
test_wordchar
This instruction implements the special PE operator "wordchar",
which checks if CC falls into the character class of the same
name, or not.
Success and failure of the test are both recorded directly in
ST. Success further clears ES, wheras failure sets the pair of
CL and expected input (encoded as a leaf parsing expression) as
the new ES and then rewinds CL by one character, preparing the
machine for another parse attempt by a possible alternative.
test_xdigit
This instruction implements the special PE operator "xdigit",
which checks if CC falls into the character class of the same
name, or not.
Success and failure of the test are both recorded directly in
ST. Success further clears ES, wheras failure sets the pair of
CL and expected input (encoded as a leaf parsing expression) as
the new ES and then rewinds CL by one character, preparing the
machine for another parse attempt by a possible alternative.
ERROR HANDLING
The instructions in this section mainly access ER and ES.
error_clear
This instruction clears ER.
error_push
This instruction makes a copy of ER and pushes it on ES.
error_pop_merge
This instruction takes the topmost entry of ES and merges the
error status it contains with ES, making the result the new ES.
The merge is governed by four rules, with the merge result
[1] Empty if both states are empty.
[2] The non-empty state if only one of the two states is non-
empty.
[3] The state with the larger location, if the two states
specify different locations.
[4] The pair of the location shared by the two states, and
the set-union of their messages for states at the same
location.
error_nonterminal symbol
This is a guarded instruction. It does nothing if either ES is
empty, or if the location in ES is not just past the last loca-
tion saved in LS. Otherwise it sets the pair of that location
and the nonterminal symbol as the new ES.
Note: In the above "just past" means "that location plus one",
or also "the location of the next character after that loca-
tion".
STATUS CONTROL
The instructions in this section directly manipulate ST.
status_ok
This instruction sets ST to true, recording a success.
status_fail
This instruction sets ST to false, recording a failure.
status_negate
This instruction negates ST, turning a failure into a success
and vice versa.
LOCATION HANDLING
The instructions in this section access CL and LS.
loc_push
This instruction makes a copy of CL and pushes it on LS.
loc_pop_discard
This instructions pops the last saved location from LS.
loc_pop_rewind
This instruction pops the last saved location from LS and re-
stores it as CL.
NONTERMINAL EXECUTION
The instructions in this section access and manipulate NC.
symbol_restore symbol
This instruction checks if NC contains data for the nonterminal
symbol at CL, or not. The result of the instruction is a boolean
flag, with True indicating that data was found in the cache. In
that case the instruction has further updated the architectural
state of the machine with the cached information, namely CL, ST,
ER, and SV.
The method with which the instruction's result is transformed
into control flow is left undefined and the responsibility of
the implementation.
symbol_save symbol
This instructions saves the current settings of CL, ST, ER, and
SV in NC, using the pair of nonterminal symbol and the last lo-
cation saved in LS as key.
VALUE CONSTRUCTION
The instructions in this section manipulate SV.
value_clear
This instruction clears SV.
value_leaf symbol
This instruction constructs an AST node for symbol covering the
range of IN from one character after the last location saved on
LS to CL and stores it in SV. ...
value_reduce symbol
This instruction generally behaves like value_nonterminal_leaf,
except that it takes all AST nodes on ARS, if any, and makes
them the children of the new node, with the last node saved on
ARS becoming the right-most / last child. Note that ARS is not
modfied by this operation.
AST CONSTRUCTION
The instructions in this section manipulate ARS and AS.
ast_value_push
This instruction makes a copy of SV and pushes it on ARS.
ast_push
This instruction pushes the current state of ARS on AS and then
clears ARS.
ast_pop_rewind
This instruction pops the last entry saved on AS and restores it
as the new state of ARS.
ast_pop_discard
This instruction pops the last entry saved on AS.
CONTROL FLOW
Normally this section would contain the specifications of the control
flow instructions of the PARAM, i.e. (un)conditional jumps and the
like. However, this part of the PARAM is intentionally left unspeci-
fied. This allows the implementations to freely choose how to implement
control flow.
The implementation of this machine in Parser Tools, i.e the package
pt::rde, is not only coded in Tcl, but also relies on Tcl commands to
provide it with control flow (instructions).
INTERACTION OF THE INSTRUCTIONS WITH THE ARCHITECTURAL STATE
InstructionInputsOutputs
======================= ===========================================
ast_pop_discardAS->AS
ast_pop_rewindAS->AS, ARS
ast_push ARS, AS->AS
ast_value_pushSV, ARS->ARS
======================= ===========================================
error_clear-->ER
error_nonterminal symER, LS->ER
error_pop_merge ES, ER->ER
error_pushES, ER->ES
======================= ===========================================
input_next msgIN->TC, CL, CC, ST, ER
======================= ===========================================
loc_pop_discardLS->LS
loc_pop_rewindLS->LS, CL
loc_push CL, LS->LS
======================= ===========================================
status_fail-->ST
status_negateST->ST
status_ok -->ST
======================= ===========================================
symbol_restore symNC->CL, ST, ER, SV
symbol_save symCL, ST, ER, SV LS->NC
======================= ===========================================
test_alnum CC->ST, ER
test_alphaCC->ST, ER
test_asciiCC->ST, ER
test_char charCC->ST, ER
test_ddigitCC->ST, ER
test_digitCC->ST, ER
test_graphCC->ST, ER
test_lowerCC->ST, ER
test_printCC->ST, ER
test_punctCC->ST, ER
test_range chars chareCC->ST, ER
test_spaceCC->ST, ER
test_upperCC->ST, ER
test_wordcharCC->ST, ER
test_xdigitCC->ST, ER
======================= ===========================================
value_clear-->SV
value_leaf symbolLS, CL->SV
value_reduce symbolARS, LS, CL->SV
======================= ===========================================
BUGS, IDEAS, FEEDBACK
This document, and the package it describes, will undoubtedly contain
bugs and other problems. Please report such in the category pt of the
Tcllib Trackers [http://core.tcl.tk/tcllib/reportlist]. Please also
report any ideas for enhancements you may have for either package
and/or documentation.
When proposing code changes, please provide unified diffs, i.e the out-
put of diff -u.
Note further that attachments are strongly preferred over inlined
patches. Attachments can be made by going to the Edit form of the
ticket immediately after its creation, and then using the left-most
button in the secondary navigation bar.
KEYWORDS
EBNF, LL(k), PEG, TDPL, context-free languages, expression, grammar,
matching, parser, parsing expression, parsing expression grammar, push
down automaton, recursive descent, state, top-down parsing languages,
transducer, virtual machine
CATEGORY
Parsing and Grammars
COPYRIGHT
Copyright (c) 2009 Andreas Kupries <andreas_kupries@users.sourceforge.net>
tcllib 1 pt::param(3tcl)