Mercurial > octave
changeset 22333:2758af148ced
move base_list and regexp classes to octave namespace
* base-list.h (class octave_base_list): Move class to octave
namespace and rename to base_list.
* lo-regexp.cc, lo-regexp.h (class regexp): Move class to octave
namespace. Move global functions to static functions in regexp
class.
* comment-list.h, regexp.cc, symtab.h, txt-eng.h, variables.cc,
pt-arg-list.cc, pt-arg-list.h, pt-array-list.h, pt-classdef.h,
pt-decl.h, pt-mat.cc, pt-misc.h, pt-select.h, pt-stmt.h, child-list.h:
Update for namespace changes.
| author | John W. Eaton <jwe@octave.org> |
|---|---|
| date | Wed, 17 Aug 2016 17:57:49 -0400 |
| parents | 1d85005fc6b7 |
| children | fed9aa2ed01d |
| files | libinterp/corefcn/comment-list.h libinterp/corefcn/regexp.cc libinterp/corefcn/symtab.h libinterp/corefcn/txt-eng.h libinterp/corefcn/variables.cc libinterp/parse-tree/pt-arg-list.cc libinterp/parse-tree/pt-arg-list.h libinterp/parse-tree/pt-array-list.h libinterp/parse-tree/pt-classdef.h libinterp/parse-tree/pt-decl.h libinterp/parse-tree/pt-mat.cc libinterp/parse-tree/pt-misc.h libinterp/parse-tree/pt-select.h libinterp/parse-tree/pt-stmt.h liboctave/system/child-list.h liboctave/util/base-list.h liboctave/util/lo-regexp.cc liboctave/util/lo-regexp.h |
| diffstat | 18 files changed, 850 insertions(+), 796 deletions(-) [+] |
line wrap: on
line diff
--- a/libinterp/corefcn/comment-list.h Wed Aug 17 10:30:50 2016 -0700 +++ b/libinterp/corefcn/comment-list.h Wed Aug 17 17:57:49 2016 -0400 @@ -84,14 +84,14 @@ }; class -octave_comment_list : public octave_base_list<octave_comment_elt> +octave_comment_list : public octave::base_list<octave_comment_elt> { public: octave_comment_list (void) { } void append (const octave_comment_elt& elt) - { octave_base_list<octave_comment_elt>::append (elt); } + { octave::base_list<octave_comment_elt>::append (elt); } void append (const std::string& s, octave_comment_elt::comment_type t = octave_comment_elt::unknown)
--- a/libinterp/corefcn/regexp.cc Wed Aug 17 10:30:50 2016 -0700 +++ b/libinterp/corefcn/regexp.cc Wed Aug 17 17:57:49 2016 -0400 @@ -274,7 +274,7 @@ } static void -parse_options (regexp::opts& options, const octave_value_list& args, +parse_options (octave::regexp::opts& options, const octave_value_list& args, const std::string& who, int skip, bool& extra_args) { extra_args = false; @@ -339,12 +339,13 @@ if (args(1).is_sq_string ()) pattern = do_regexp_ptn_string_escapes (pattern); - regexp::opts options; + octave::regexp::opts options; options.case_insensitive (case_insensitive); bool extra_options = false; parse_options (options, args, who, 2, extra_options); - regexp::match_data rx_lst = regexp_match (pattern, buffer, options, who); + octave::regexp::match_data rx_lst + = octave::regexp::match (pattern, buffer, options, who); string_vector named_pats = rx_lst.named_patterns (); @@ -373,7 +374,7 @@ Cell tmp (dim_vector (1, sz)); i = 0; - for (regexp::match_data::const_iterator p = rx_lst.begin (); + for (octave::regexp::match_data::const_iterator p = rx_lst.begin (); p != rx_lst.end (); p++) { string_vector named_tokens = p->named_tokens (); @@ -389,7 +390,7 @@ if (options.once ()) { - regexp::match_data::const_iterator p = rx_lst.begin (); + octave::regexp::match_data::const_iterator p = rx_lst.begin (); retval(4) = sz ? p->tokens () : Cell (); retval(3) = sz ? p->match_string () : ""; @@ -426,7 +427,7 @@ size_t sp_start = 0; i = 0; - for (regexp::match_data::const_iterator p = rx_lst.begin (); + for (octave::regexp::match_data::const_iterator p = rx_lst.begin (); p != rx_lst.end (); p++) { double s = p->start (); @@ -1322,11 +1323,11 @@ } regexpargs.resize (len); - regexp::opts options; + octave::regexp::opts options; bool extra_args = false; parse_options (options, regexpargs, who, 0, extra_args); - return regexp_replace (pattern, buffer, replacement, options, who); + return octave::regexp::replace (pattern, buffer, replacement, options, who); } DEFUN (regexprep, args, ,
--- a/libinterp/corefcn/symtab.h Wed Aug 17 10:30:50 2016 -0700 +++ b/libinterp/corefcn/symtab.h Wed Aug 17 17:57:49 2016 -0400 @@ -2055,7 +2055,7 @@ { std::list<symbol_record> retval; - ::regexp pat (pattern); + octave::regexp pat (pattern); for (global_table_const_iterator p = global_table.begin (); p != global_table.end (); p++) @@ -2795,7 +2795,7 @@ void do_clear_variable_regexp (const std::string& pat) { - ::regexp pattern (pat); + octave::regexp pattern (pat); for (table_iterator p = table.begin (); p != table.end (); p++) { @@ -2872,7 +2872,7 @@ { std::list<symbol_record> retval; - ::regexp pat (pattern); + octave::regexp pat (pattern); for (table_const_iterator p = table.begin (); p != table.end (); p++) {
--- a/libinterp/corefcn/txt-eng.h Wed Aug 17 10:30:50 2016 -0700 +++ b/libinterp/corefcn/txt-eng.h Wed Aug 17 17:57:49 2016 -0400 @@ -108,14 +108,14 @@ class OCTINTERP_API text_element_list - : public text_element, public octave_base_list<text_element *> + : public text_element, public octave::base_list<text_element *> { public: text_element_list (void) - : text_element (), octave_base_list<text_element*> () { } + : text_element (), octave::base_list<text_element*> () { } text_element_list (text_element* e) - : text_element (), octave_base_list<text_element*> () + : text_element (), octave::base_list<text_element*> () { push_back (e); } ~text_element_list (void)
--- a/libinterp/corefcn/variables.cc Wed Aug 17 10:30:50 2016 -0700 +++ b/libinterp/corefcn/variables.cc Wed Aug 17 17:57:49 2016 -0400 @@ -2119,7 +2119,7 @@ { if (have_regexp) { - if (is_regexp_match (patstr, nm)) + if (octave::regexp::is_match (patstr, nm)) { retval = true; break;
--- a/libinterp/parse-tree/pt-arg-list.cc Wed Aug 17 10:30:50 2016 -0700 +++ b/libinterp/parse-tree/pt-arg-list.cc Wed Aug 17 17:57:49 2016 -0400 @@ -73,7 +73,7 @@ void tree_argument_list::append (const element_type& s) { - octave_base_list<tree_expression *>::append (s); + octave::base_list<tree_expression *>::append (s); if (! list_includes_magic_end && s && s->has_magic_end ()) list_includes_magic_end = true;
--- a/libinterp/parse-tree/pt-arg-list.h Wed Aug 17 10:30:50 2016 -0700 +++ b/libinterp/parse-tree/pt-arg-list.h Wed Aug 17 17:57:49 2016 -0400 @@ -41,7 +41,7 @@ // arguments in a function call or index expression. class -tree_argument_list : public octave_base_list<tree_expression *> +tree_argument_list : public octave::base_list<tree_expression *> { public:
--- a/libinterp/parse-tree/pt-array-list.h Wed Aug 17 10:30:50 2016 -0700 +++ b/libinterp/parse-tree/pt-array-list.h Wed Aug 17 17:57:49 2016 -0400 @@ -34,15 +34,15 @@ class tree_array_list : public tree_expression, - public octave_base_list<tree_argument_list *> + public octave::base_list<tree_argument_list *> { public: - typedef octave_base_list<tree_argument_list *>::iterator iterator; - typedef octave_base_list<tree_argument_list *>::const_iterator const_iterator; + typedef octave::base_list<tree_argument_list *>::iterator iterator; + typedef octave::base_list<tree_argument_list *>::const_iterator const_iterator; tree_array_list (tree_argument_list *row = 0, int l = -1, int c = -1) - : tree_expression (l, c), octave_base_list<tree_argument_list *> () + : tree_expression (l, c), octave::base_list<tree_argument_list *> () { if (row) append (row);
--- a/libinterp/parse-tree/pt-classdef.h Wed Aug 17 10:30:50 2016 -0700 +++ b/libinterp/parse-tree/pt-classdef.h Wed Aug 17 17:57:49 2016 -0400 @@ -74,7 +74,7 @@ tree_classdef_attribute& operator = (const tree_classdef_attribute&); }; -class tree_classdef_attribute_list : public octave_base_list<tree_classdef_attribute *> +class tree_classdef_attribute_list : public octave::base_list<tree_classdef_attribute *> { public: @@ -82,8 +82,8 @@ tree_classdef_attribute_list (tree_classdef_attribute *a) { append (a); } - tree_classdef_attribute_list (const octave_base_list<tree_classdef_attribute *>& a) - : octave_base_list<tree_classdef_attribute *> (a) { } + tree_classdef_attribute_list (const octave::base_list<tree_classdef_attribute *>& a) + : octave::base_list<tree_classdef_attribute *> (a) { } ~tree_classdef_attribute_list (void); @@ -122,7 +122,7 @@ tree_classdef_superclass& operator = (const tree_classdef_superclass&); }; -class tree_classdef_superclass_list : public octave_base_list<tree_classdef_superclass *> +class tree_classdef_superclass_list : public octave::base_list<tree_classdef_superclass *> { public: @@ -130,8 +130,8 @@ tree_classdef_superclass_list (tree_classdef_superclass *sc) { append (sc); } - tree_classdef_superclass_list (const octave_base_list<tree_classdef_superclass *>& a) - : octave_base_list<tree_classdef_superclass *> (a) { } + tree_classdef_superclass_list (const octave::base_list<tree_classdef_superclass *>& a) + : octave::base_list<tree_classdef_superclass *> (a) { } ~tree_classdef_superclass_list (void); @@ -152,7 +152,7 @@ public: tree_classdef_element (tree_classdef_attribute_list *a, - octave_base_list<T> *elist, + octave::base_list<T> *elist, octave_comment_list *lc, octave_comment_list *tc, int l = -1, int c = -1) : tree (l, c), attr_list (a), elt_list (elist), @@ -169,7 +169,7 @@ tree_classdef_attribute_list *attribute_list (void) { return attr_list; } - octave_base_list<T> *element_list (void) { return elt_list; } + octave::base_list<T> *element_list (void) { return elt_list; } octave_comment_list *leading_comment (void) { return lead_comm; } @@ -183,7 +183,7 @@ tree_classdef_attribute_list *attr_list; // The list of objects contained in this block. - octave_base_list<T> *elt_list; + octave::base_list<T> *elt_list; // Comment preceding the token marking the beginning of the block. octave_comment_list *lead_comm; @@ -229,7 +229,7 @@ tree_classdef_property& operator = (const tree_classdef_property&); }; -class tree_classdef_property_list : public octave_base_list<tree_classdef_property *> +class tree_classdef_property_list : public octave::base_list<tree_classdef_property *> { public: @@ -237,8 +237,8 @@ tree_classdef_property_list (tree_classdef_property* p) { append (p); } - tree_classdef_property_list (const octave_base_list<tree_classdef_property *>& a) - : octave_base_list<tree_classdef_property *> (a) { } + tree_classdef_property_list (const octave::base_list<tree_classdef_property *>& a) + : octave::base_list<tree_classdef_property *> (a) { } ~tree_classdef_property_list (void); @@ -278,7 +278,7 @@ tree_classdef_properties_block& operator = (const tree_classdef_properties_block&); }; -class tree_classdef_methods_list : public octave_base_list<octave_value> +class tree_classdef_methods_list : public octave::base_list<octave_value> { public: @@ -286,8 +286,8 @@ tree_classdef_methods_list (const octave_value& f) { append (f); } - tree_classdef_methods_list (const octave_base_list<octave_value>& a) - : octave_base_list<octave_value> (a) { } + tree_classdef_methods_list (const octave::base_list<octave_value>& a) + : octave::base_list<octave_value> (a) { } ~tree_classdef_methods_list (void) { } @@ -351,7 +351,7 @@ tree_classdef_event& operator = (const tree_classdef_event&); }; -class tree_classdef_events_list : public octave_base_list<tree_classdef_event *> +class tree_classdef_events_list : public octave::base_list<tree_classdef_event *> { public: @@ -359,8 +359,8 @@ tree_classdef_events_list (tree_classdef_event *e) { append (e); } - tree_classdef_events_list (const octave_base_list<tree_classdef_event *>& a) - : octave_base_list<tree_classdef_event *> (a) { } + tree_classdef_events_list (const octave::base_list<tree_classdef_event *>& a) + : octave::base_list<tree_classdef_event *> (a) { } ~tree_classdef_events_list (void); @@ -432,7 +432,7 @@ tree_classdef_enum& operator = (const tree_classdef_enum&); }; -class tree_classdef_enum_list : public octave_base_list<tree_classdef_enum *> +class tree_classdef_enum_list : public octave::base_list<tree_classdef_enum *> { public: @@ -440,8 +440,8 @@ tree_classdef_enum_list (tree_classdef_enum *e) { append (e); } - tree_classdef_enum_list (const octave_base_list<tree_classdef_enum *>& a) - : octave_base_list<tree_classdef_enum *> (a) { } + tree_classdef_enum_list (const octave::base_list<tree_classdef_enum *>& a) + : octave::base_list<tree_classdef_enum *> (a) { } ~tree_classdef_enum_list (void);
--- a/libinterp/parse-tree/pt-decl.h Wed Aug 17 10:30:50 2016 -0700 +++ b/libinterp/parse-tree/pt-decl.h Wed Aug 17 17:57:49 2016 -0400 @@ -111,7 +111,7 @@ }; class -tree_decl_init_list : public octave_base_list<tree_decl_elt *> +tree_decl_init_list : public octave::base_list<tree_decl_elt *> { public:
--- a/libinterp/parse-tree/pt-mat.cc Wed Aug 17 10:30:50 2016 -0700 +++ b/libinterp/parse-tree/pt-mat.cc Wed Aug 17 17:57:49 2016 -0400 @@ -64,7 +64,7 @@ private: class - tm_row_const_rep : public octave_base_list<octave_value> + tm_row_const_rep : public octave::base_list<octave_value> { public: @@ -462,7 +462,7 @@ } class -tm_const : public octave_base_list<tm_row_const> +tm_const : public octave::base_list<tm_row_const> { public:
--- a/libinterp/parse-tree/pt-misc.h Wed Aug 17 10:30:50 2016 -0700 +++ b/libinterp/parse-tree/pt-misc.h Wed Aug 17 17:57:49 2016 -0400 @@ -45,7 +45,7 @@ // only. class -tree_parameter_list : public octave_base_list<tree_decl_elt *> +tree_parameter_list : public octave::base_list<tree_decl_elt *> { public: @@ -112,7 +112,7 @@ // assignment expressions. class -tree_return_list : public octave_base_list<tree_index_expression *> +tree_return_list : public octave::base_list<tree_index_expression *> { public: @@ -137,7 +137,7 @@ }; class -tree_va_return_list : public octave_base_list<octave_value> +tree_va_return_list : public octave::base_list<octave_value> { public:
--- a/libinterp/parse-tree/pt-select.h Wed Aug 17 10:30:50 2016 -0700 +++ b/libinterp/parse-tree/pt-select.h Wed Aug 17 17:57:49 2016 -0400 @@ -88,7 +88,7 @@ }; class -tree_if_command_list : public octave_base_list<tree_if_clause *> +tree_if_command_list : public octave::base_list<tree_if_clause *> { public: @@ -218,7 +218,7 @@ }; class -tree_switch_case_list : public octave_base_list<tree_switch_case *> +tree_switch_case_list : public octave::base_list<tree_switch_case *> { public:
--- a/libinterp/parse-tree/pt-stmt.h Wed Aug 17 10:30:50 2016 -0700 +++ b/libinterp/parse-tree/pt-stmt.h Wed Aug 17 17:57:49 2016 -0400 @@ -128,7 +128,7 @@ // A list of statements to evaluate. class -tree_statement_list : public octave_base_list<tree_statement *> +tree_statement_list : public octave::base_list<tree_statement *> { public:
--- a/liboctave/system/child-list.h Wed Aug 17 10:30:50 2016 -0700 +++ b/liboctave/system/child-list.h Wed Aug 17 17:57:49 2016 -0400 @@ -90,7 +90,7 @@ child_list (void) { } - class child_list_rep : public octave_base_list<child> + class child_list_rep : public base_list<child> { public:
--- a/liboctave/util/base-list.h Wed Aug 17 10:30:50 2016 -0700 +++ b/liboctave/util/base-list.h Wed Aug 17 17:57:49 2016 -0400 @@ -29,103 +29,106 @@ #include <list> -template <typename elt_type> -class -octave_base_list +namespace octave { -public: - - typedef typename std::list<elt_type>::iterator iterator; - typedef typename std::list<elt_type>::const_iterator const_iterator; - - typedef typename std::list<elt_type>::reverse_iterator reverse_iterator; - typedef typename std::list<elt_type>::const_reverse_iterator const_reverse_iterator; + template <typename elt_type> + class + base_list + { + public: - bool empty (void) const { return lst.empty (); } + typedef typename std::list<elt_type>::iterator iterator; + typedef typename std::list<elt_type>::const_iterator const_iterator; - size_t size (void) const { return lst.size (); } - size_t length (void) const { return size (); } + typedef typename std::list<elt_type>::reverse_iterator reverse_iterator; + typedef typename std::list<elt_type>::const_reverse_iterator const_reverse_iterator; - iterator erase (iterator pos) { return lst.erase (pos); } + bool empty (void) const { return lst.empty (); } - template <typename P> - void remove_if (P pred) - { - lst.remove_if (pred); + size_t size (void) const { return lst.size (); } + size_t length (void) const { return size (); } + + iterator erase (iterator pos) { return lst.erase (pos); } - // FIXME: kluge removed 8/7/13. Eventually this commented - // code should be deleted. - // - // FIXME: this kluge should be removed at some point. - // We would like to simply call - // - // lst.remove_if (pred); - // - // but the Sun Studio compiler chokes on that. - // - // iterator b = lst.begin (); - // iterator e = lst.end (); - // while (b != e) - // { - // iterator n = b; - // n++; - // if (pred (*b)) - // lst.erase (b); - // b = n; - // } - } + template <typename P> + void remove_if (P pred) + { + lst.remove_if (pred); - void clear (void) { lst.clear (); } + // FIXME: kluge removed 8/7/13. Eventually this commented + // code should be deleted. + // + // FIXME: this kluge should be removed at some point. + // We would like to simply call + // + // lst.remove_if (pred); + // + // but the Sun Studio compiler chokes on that. + // + // iterator b = lst.begin (); + // iterator e = lst.end (); + // while (b != e) + // { + // iterator n = b; + // n++; + // if (pred (*b)) + // lst.erase (b); + // b = n; + // } + } + + void clear (void) { lst.clear (); } - iterator begin (void) { return iterator (lst.begin ()); } - const_iterator begin (void) const { return const_iterator (lst.begin ()); } + iterator begin (void) { return iterator (lst.begin ()); } + const_iterator begin (void) const { return const_iterator (lst.begin ()); } - iterator end (void) { return iterator (lst.end ()); } - const_iterator end (void) const { return const_iterator (lst.end ()); } + iterator end (void) { return iterator (lst.end ()); } + const_iterator end (void) const { return const_iterator (lst.end ()); } - reverse_iterator rbegin (void) { return reverse_iterator (lst.rbegin ()); } - const_reverse_iterator rbegin (void) const - { return const_reverse_iterator (lst.rbegin ()); } + reverse_iterator rbegin (void) { return reverse_iterator (lst.rbegin ()); } + const_reverse_iterator rbegin (void) const + { return const_reverse_iterator (lst.rbegin ()); } - reverse_iterator rend (void) { return reverse_iterator (lst.rend ()); } - const_reverse_iterator rend (void) const - { return const_reverse_iterator (lst.rend ()); } + reverse_iterator rend (void) { return reverse_iterator (lst.rend ()); } + const_reverse_iterator rend (void) const + { return const_reverse_iterator (lst.rend ()); } + + elt_type& front (void) { return lst.front (); } + elt_type& back (void) { return lst.back (); } - elt_type& front (void) { return lst.front (); } - elt_type& back (void) { return lst.back (); } + const elt_type& front (void) const { return lst.front (); } + const elt_type& back (void) const { return lst.back (); } - const elt_type& front (void) const { return lst.front (); } - const elt_type& back (void) const { return lst.back (); } + void push_front (const elt_type& s) { lst.push_front (s); } + void push_back (const elt_type& s) { lst.push_back (s); } - void push_front (const elt_type& s) { lst.push_front (s); } - void push_back (const elt_type& s) { lst.push_back (s); } + void pop_front (void) { lst.pop_front (); } + void pop_back (void) { lst.pop_back (); } - void pop_front (void) { lst.pop_front (); } - void pop_back (void) { lst.pop_back (); } - - // For backward compatibility. - void append (const elt_type& s) { lst.push_back (s); } + // For backward compatibility. + void append (const elt_type& s) { lst.push_back (s); } - octave_base_list (void) : lst () { } + base_list (void) : lst () { } - octave_base_list (const std::list<elt_type>& l) : lst (l) { } + base_list (const std::list<elt_type>& l) : lst (l) { } - octave_base_list (const octave_base_list& bl) : lst (bl.lst) { } + base_list (const base_list& bl) : lst (bl.lst) { } - octave_base_list& operator = (const octave_base_list& bl) - { - if (this != &bl) - { - lst = bl.lst; - } - return *this; - } + base_list& operator = (const base_list& bl) + { + if (this != &bl) + { + lst = bl.lst; + } + return *this; + } - virtual ~octave_base_list (void) { } + virtual ~base_list (void) { } + + private: -private: - - std::list<elt_type> lst; -}; + std::list<elt_type> lst; + }; +} #endif
--- a/liboctave/util/lo-regexp.cc Wed Aug 17 10:30:50 2016 -0700 +++ b/liboctave/util/lo-regexp.cc Wed Aug 17 17:57:49 2016 -0400 @@ -57,556 +57,559 @@ // FIXME: don't bother collecting and composing return values // the user doesn't want. -void -regexp::free (void) -{ - if (data) - pcre_free (static_cast<pcre *> (data)); -} - -void -regexp::compile_internal (void) +namespace octave { - // If we had a previously compiled pattern, release it. - free (); - - size_t max_length = MAXLOOKBEHIND; + void + regexp::free (void) + { + if (data) + pcre_free (static_cast<pcre *> (data)); + } - size_t pos = 0; - size_t new_pos; - int inames = 0; - std::ostringstream buf; + void + regexp::compile_internal (void) + { + // If we had a previously compiled pattern, release it. + free (); + + size_t max_length = MAXLOOKBEHIND; + + size_t pos = 0; + size_t new_pos; + int inames = 0; + std::ostringstream buf; - while ((new_pos = pattern.find ("(?", pos)) != std::string::npos) - { - if (pattern.at (new_pos + 2) == '<' - && !(pattern.at (new_pos + 3) == '=' - || pattern.at (new_pos + 3) == '!')) - { - // The syntax of named tokens in pcre is "(?P<name>...)" while - // we need a syntax "(?<name>...)", so fix that here. Also an - // expression like - // "(?<first>\w+)\s+(?<last>\w+)|(?<last>\w+),\s+(?<first>\w+)" - // should be perfectly legal, while pcre does not allow the same - // named token name on both sides of the alternative. Also fix - // that here by replacing name tokens by dummy names, and dealing - // with the dummy names later. + while ((new_pos = pattern.find ("(?", pos)) != std::string::npos) + { + if (pattern.at (new_pos + 2) == '<' + && !(pattern.at (new_pos + 3) == '=' + || pattern.at (new_pos + 3) == '!')) + { + // The syntax of named tokens in pcre is "(?P<name>...)" while + // we need a syntax "(?<name>...)", so fix that here. Also an + // expression like + // "(?<first>\w+)\s+(?<last>\w+)|(?<last>\w+),\s+(?<first>\w+)" + // should be perfectly legal, while pcre does not allow the same + // named token name on both sides of the alternative. Also fix + // that here by replacing name tokens by dummy names, and dealing + // with the dummy names later. - size_t tmp_pos = pattern.find_first_of ('>', new_pos); - - if (tmp_pos == std::string::npos) - (*current_liboctave_error_handler) - ("regexp: syntax error in pattern"); + size_t tmp_pos = pattern.find_first_of ('>', new_pos); - std::string tmp_name = - pattern.substr (new_pos+3, tmp_pos-new_pos-3); + if (tmp_pos == std::string::npos) + (*current_liboctave_error_handler) + ("regexp: syntax error in pattern"); - bool found = false; + std::string tmp_name = + pattern.substr (new_pos+3, tmp_pos-new_pos-3); + + bool found = false; - for (int i = 0; i < nnames; i++) - { - if (named_pats(i) == tmp_name) - { - named_idx.resize (dim_vector (inames+1, 1)); - named_idx(inames) = i; - found = true; - break; - } - } - - if (! found) - { - named_idx.resize (dim_vector (inames+1, 1)); - named_idx(inames) = nnames; - named_pats.append (tmp_name); - nnames++; - } + for (int i = 0; i < nnames; i++) + { + if (named_pats(i) == tmp_name) + { + named_idx.resize (dim_vector (inames+1, 1)); + named_idx(inames) = i; + found = true; + break; + } + } - if (new_pos - pos > 0) - buf << pattern.substr (pos, new_pos-pos); - if (inames < 10) - buf << "(?P<n00" << inames++; - else if (inames < 100) - buf << "(?P<n0" << inames++; - else - buf << "(?P<n" << inames++; + if (! found) + { + named_idx.resize (dim_vector (inames+1, 1)); + named_idx(inames) = nnames; + named_pats.append (tmp_name); + nnames++; + } + + if (new_pos - pos > 0) + buf << pattern.substr (pos, new_pos-pos); + if (inames < 10) + buf << "(?P<n00" << inames++; + else if (inames < 100) + buf << "(?P<n0" << inames++; + else + buf << "(?P<n" << inames++; - pos = tmp_pos; - } - else if (pattern.at (new_pos + 2) == '<') - { - // Find lookbehind operators of arbitrary length (ie like - // "(?<=[a-z]*)") and replace with a maximum length operator - // as PCRE can not yet handle arbitrary length lookahead - // operators. Use the string length as the maximum length to - // avoid issues. + pos = tmp_pos; + } + else if (pattern.at (new_pos + 2) == '<') + { + // Find lookbehind operators of arbitrary length (ie like + // "(?<=[a-z]*)") and replace with a maximum length operator + // as PCRE can not yet handle arbitrary length lookahead + // operators. Use the string length as the maximum length to + // avoid issues. - int brackets = 1; - size_t tmp_pos1 = new_pos + 2; - size_t tmp_pos2 = tmp_pos1; + int brackets = 1; + size_t tmp_pos1 = new_pos + 2; + size_t tmp_pos2 = tmp_pos1; - while (tmp_pos1 < pattern.length () && brackets > 0) - { - char ch = pattern.at (tmp_pos1); - - if (ch == '(') - brackets++; - else if (ch == ')') - { - if (brackets > 1) - tmp_pos2 = tmp_pos1; + while (tmp_pos1 < pattern.length () && brackets > 0) + { + char ch = pattern.at (tmp_pos1); - brackets--; - } - - tmp_pos1++; - } + if (ch == '(') + brackets++; + else if (ch == ')') + { + if (brackets > 1) + tmp_pos2 = tmp_pos1; - if (brackets != 0) - { - buf << pattern.substr (pos, new_pos - pos) << "(?"; - pos = new_pos + 2; - } - else - { - size_t tmp_pos3 = pattern.find_first_of ("*+", tmp_pos2); + brackets--; + } + + tmp_pos1++; + } - if (tmp_pos3 != std::string::npos && tmp_pos3 < tmp_pos1) - { - if (! lookbehind_warned) - { - lookbehind_warned = true; - (*current_liboctave_warning_with_id_handler) - ("Octave:regexp-lookbehind-limit", - "%s: arbitrary length lookbehind patterns are only supported up to length %d", - who.c_str (), MAXLOOKBEHIND); - } + if (brackets != 0) + { + buf << pattern.substr (pos, new_pos - pos) << "(?"; + pos = new_pos + 2; + } + else + { + size_t tmp_pos3 = pattern.find_first_of ("*+", tmp_pos2); - buf << pattern.substr (pos, new_pos - pos) << "("; - - size_t i; + if (tmp_pos3 != std::string::npos && tmp_pos3 < tmp_pos1) + { + if (! lookbehind_warned) + { + lookbehind_warned = true; + (*current_liboctave_warning_with_id_handler) + ("Octave:regexp-lookbehind-limit", + "%s: arbitrary length lookbehind patterns are only supported up to length %d", + who.c_str (), MAXLOOKBEHIND); + } - if (pattern.at (tmp_pos3) == '*') - i = 0; - else - i = 1; + buf << pattern.substr (pos, new_pos - pos) << "("; + + size_t i; - for (; i < max_length + 1; i++) - { - buf << pattern.substr (new_pos, tmp_pos3 - new_pos) - << "{" << i << "}"; - buf << pattern.substr (tmp_pos3 + 1, - tmp_pos1 - tmp_pos3 - 1); - if (i != max_length) - buf << "|"; - } - buf << ")"; - } - else - buf << pattern.substr (pos, tmp_pos1 - pos); + if (pattern.at (tmp_pos3) == '*') + i = 0; + else + i = 1; - pos = tmp_pos1; - } - } - else - { - buf << pattern.substr (pos, new_pos - pos) << "(?"; - pos = new_pos + 2; - } + for (; i < max_length + 1; i++) + { + buf << pattern.substr (new_pos, tmp_pos3 - new_pos) + << "{" << i << "}"; + buf << pattern.substr (tmp_pos3 + 1, + tmp_pos1 - tmp_pos3 - 1); + if (i != max_length) + buf << "|"; + } + buf << ")"; + } + else + buf << pattern.substr (pos, tmp_pos1 - pos); - } - - buf << pattern.substr (pos); + pos = tmp_pos1; + } + } + else + { + buf << pattern.substr (pos, new_pos - pos) << "(?"; + pos = new_pos + 2; + } - const char *err; - int erroffset; - std::string buf_str = buf.str (); + } + + buf << pattern.substr (pos); - int pcre_options - = ((options.case_insensitive () ? PCRE_CASELESS : 0) - | (options.dotexceptnewline () ? 0 : PCRE_DOTALL) - | (options.lineanchors () ? PCRE_MULTILINE : 0) - | (options.freespacing () ? PCRE_EXTENDED : 0)); - - data = pcre_compile (buf_str.c_str (), pcre_options, &err, &erroffset, 0); + const char *err; + int erroffset; + std::string buf_str = buf.str (); - if (! data) - (*current_liboctave_error_handler) - ("%s: %s at position %d of expression", who.c_str (), err, erroffset); -} + int pcre_options + = ((options.case_insensitive () ? PCRE_CASELESS : 0) + | (options.dotexceptnewline () ? 0 : PCRE_DOTALL) + | (options.lineanchors () ? PCRE_MULTILINE : 0) + | (options.freespacing () ? PCRE_EXTENDED : 0)); -regexp::match_data -regexp::match (const std::string& buffer) -{ - regexp::match_data retval; + data = pcre_compile (buf_str.c_str (), pcre_options, &err, &erroffset, 0); + + if (! data) + (*current_liboctave_error_handler) + ("%s: %s at position %d of expression", who.c_str (), err, erroffset); + } - std::list<regexp::match_element> lst; + regexp::match_data + regexp::match (const std::string& buffer) + { + regexp::match_data retval; - int subpatterns; - int namecount; - int nameentrysize; - char *nametable; - size_t idx = 0; + std::list<regexp::match_element> lst; - pcre *re = static_cast<pcre *> (data); + int subpatterns; + int namecount; + int nameentrysize; + char *nametable; + size_t idx = 0; - pcre_fullinfo (re, 0, PCRE_INFO_CAPTURECOUNT, &subpatterns); - pcre_fullinfo (re, 0, PCRE_INFO_NAMECOUNT, &namecount); - pcre_fullinfo (re, 0, PCRE_INFO_NAMEENTRYSIZE, &nameentrysize); - pcre_fullinfo (re, 0, PCRE_INFO_NAMETABLE, &nametable); + pcre *re = static_cast<pcre *> (data); - OCTAVE_LOCAL_BUFFER (int, ovector, (subpatterns+1)*3); - OCTAVE_LOCAL_BUFFER (int, nidx, namecount); + pcre_fullinfo (re, 0, PCRE_INFO_CAPTURECOUNT, &subpatterns); + pcre_fullinfo (re, 0, PCRE_INFO_NAMECOUNT, &namecount); + pcre_fullinfo (re, 0, PCRE_INFO_NAMEENTRYSIZE, &nameentrysize); + pcre_fullinfo (re, 0, PCRE_INFO_NAMETABLE, &nametable); - for (int i = 0; i < namecount; i++) - { - // Index of subpattern in first two bytes MSB first of name. - // Extract index. - nidx[i] = (static_cast<int> (nametable[i*nameentrysize])) << 8 - | static_cast<int> (nametable[i*nameentrysize+1]); - } + OCTAVE_LOCAL_BUFFER (int, ovector, (subpatterns+1)*3); + OCTAVE_LOCAL_BUFFER (int, nidx, namecount); - while (true) - { - OCTAVE_QUIT; + for (int i = 0; i < namecount; i++) + { + // Index of subpattern in first two bytes MSB first of name. + // Extract index. + nidx[i] = (static_cast<int> (nametable[i*nameentrysize])) << 8 + | static_cast<int> (nametable[i*nameentrysize+1]); + } + + while (true) + { + OCTAVE_QUIT; - int matches = pcre_exec (re, 0, buffer.c_str (), - buffer.length (), idx, - (idx ? PCRE_NOTBOL : 0), - ovector, (subpatterns+1)*3); + int matches = pcre_exec (re, 0, buffer.c_str (), + buffer.length (), idx, + (idx ? PCRE_NOTBOL : 0), + ovector, (subpatterns+1)*3); - if (matches == PCRE_ERROR_MATCHLIMIT) - { - // Try harder; start with default value for MATCH_LIMIT - // and increase it. - (*current_liboctave_warning_with_id_handler) - ("Octave:regexp-match-limit", - "your pattern caused PCRE to hit its MATCH_LIMIT; trying harder now, but this will be slow"); + if (matches == PCRE_ERROR_MATCHLIMIT) + { + // Try harder; start with default value for MATCH_LIMIT + // and increase it. + (*current_liboctave_warning_with_id_handler) + ("Octave:regexp-match-limit", + "your pattern caused PCRE to hit its MATCH_LIMIT; trying harder now, but this will be slow"); - pcre_extra pe; + pcre_extra pe; - pcre_config (PCRE_CONFIG_MATCH_LIMIT, - static_cast<void *> (&pe.match_limit)); + pcre_config (PCRE_CONFIG_MATCH_LIMIT, + static_cast<void *> (&pe.match_limit)); - pe.flags = PCRE_EXTRA_MATCH_LIMIT; + pe.flags = PCRE_EXTRA_MATCH_LIMIT; - int i = 0; - while (matches == PCRE_ERROR_MATCHLIMIT - && i++ < PCRE_MATCHLIMIT_MAX) - { - OCTAVE_QUIT; + int i = 0; + while (matches == PCRE_ERROR_MATCHLIMIT + && i++ < PCRE_MATCHLIMIT_MAX) + { + OCTAVE_QUIT; - pe.match_limit *= 10; - matches = pcre_exec (re, &pe, buffer.c_str (), - buffer.length (), idx, - (idx ? PCRE_NOTBOL : 0), - ovector, (subpatterns+1)*3); - } - } + pe.match_limit *= 10; + matches = pcre_exec (re, &pe, buffer.c_str (), + buffer.length (), idx, + (idx ? PCRE_NOTBOL : 0), + ovector, (subpatterns+1)*3); + } + } - if (matches < 0 && matches != PCRE_ERROR_NOMATCH) - (*current_liboctave_error_handler) - ("%s: internal error calling pcre_exec; " - "error code from pcre_exec is %i", who.c_str (), matches); + if (matches < 0 && matches != PCRE_ERROR_NOMATCH) + (*current_liboctave_error_handler) + ("%s: internal error calling pcre_exec; " + "error code from pcre_exec is %i", who.c_str (), matches); - if (matches == PCRE_ERROR_NOMATCH) - break; - else if (ovector[1] <= ovector[0] && ! options.emptymatch ()) - { - // Zero length match. Skip to next char. - idx = ovector[0] + 1; - if (idx < buffer.length ()) - continue; - else - break; - } - else - { - int pos_match = 0; - Matrix token_extents (matches-1, 2); + if (matches == PCRE_ERROR_NOMATCH) + break; + else if (ovector[1] <= ovector[0] && ! options.emptymatch ()) + { + // Zero length match. Skip to next char. + idx = ovector[0] + 1; + if (idx < buffer.length ()) + continue; + else + break; + } + else + { + int pos_match = 0; + Matrix token_extents (matches-1, 2); - for (int i = 1; i < matches; i++) - { - if (ovector[2*i] >= 0 && ovector[2*i+1] > 0 - && (i == 1 || ovector[2*i] != ovector[2*i-2] - || ovector[2*i-1] != ovector[2*i+1])) - { - token_extents(pos_match,0) = double (ovector[2*i]+1); - token_extents(pos_match++,1) = double (ovector[2*i+1]); - } - } + for (int i = 1; i < matches; i++) + { + if (ovector[2*i] >= 0 && ovector[2*i+1] > 0 + && (i == 1 || ovector[2*i] != ovector[2*i-2] + || ovector[2*i-1] != ovector[2*i+1])) + { + token_extents(pos_match,0) = double (ovector[2*i]+1); + token_extents(pos_match++,1) = double (ovector[2*i+1]); + } + } - token_extents.resize (pos_match, 2); + token_extents.resize (pos_match, 2); - double start = double (ovector[0]+1); - double end = double (ovector[1]); + double start = double (ovector[0]+1); + double end = double (ovector[1]); - const char **listptr; - int status = pcre_get_substring_list (buffer.c_str (), ovector, - matches, &listptr); + const char **listptr; + int status = pcre_get_substring_list (buffer.c_str (), ovector, + matches, &listptr); - if (status == PCRE_ERROR_NOMEMORY) - (*current_liboctave_error_handler) - ("%s: cannot allocate memory in pcre_get_substring_list", - who.c_str ()); + if (status == PCRE_ERROR_NOMEMORY) + (*current_liboctave_error_handler) + ("%s: cannot allocate memory in pcre_get_substring_list", + who.c_str ()); - string_vector tokens (pos_match); - string_vector named_tokens (nnames); - int pos_offset = 0; - pos_match = 0; + string_vector tokens (pos_match); + string_vector named_tokens (nnames); + int pos_offset = 0; + pos_match = 0; - for (int i = 1; i < matches; i++) - { - if (ovector[2*i] >= 0 && ovector[2*i+1] > 0) - { - if (i == 1 || ovector[2*i] != ovector[2*i-2] - || ovector[2*i-1] != ovector[2*i+1]) - { - if (namecount > 0) - { - // FIXME: Should probably do this with a map() - // rather than a linear search. However, - // the number of captured, named expressions - // is usually pretty small (< 4) - for (int j = 0; j < namecount; j++) - { - if (nidx[j] == i) - { - named_tokens(named_idx(j)) = - std::string (*(listptr+i-pos_offset)); - break; - } - } - } + for (int i = 1; i < matches; i++) + { + if (ovector[2*i] >= 0 && ovector[2*i+1] > 0) + { + if (i == 1 || ovector[2*i] != ovector[2*i-2] + || ovector[2*i-1] != ovector[2*i+1]) + { + if (namecount > 0) + { + // FIXME: Should probably do this with a map() + // rather than a linear search. However, + // the number of captured, named expressions + // is usually pretty small (< 4) + for (int j = 0; j < namecount; j++) + { + if (nidx[j] == i) + { + named_tokens(named_idx(j)) = + std::string (*(listptr+i-pos_offset)); + break; + } + } + } - tokens(pos_match++) = std::string (*(listptr+i)); - } - else - pos_offset++; - } - } + tokens(pos_match++) = std::string (*(listptr+i)); + } + else + pos_offset++; + } + } - std::string match_string = std::string (*listptr); + std::string match_string = std::string (*listptr); - pcre_free_substring_list (listptr); + pcre_free_substring_list (listptr); - regexp::match_element new_elem (named_tokens, tokens, match_string, - token_extents, start, end); - lst.push_back (new_elem); + regexp::match_element new_elem (named_tokens, tokens, match_string, + token_extents, start, end); + lst.push_back (new_elem); - if (ovector[1] <= ovector[0]) - { - // Zero length match. Skip to next char. - idx = ovector[0] + 1; - if (idx <= buffer.length ()) - continue; - } - else - idx = ovector[1]; + if (ovector[1] <= ovector[0]) + { + // Zero length match. Skip to next char. + idx = ovector[0] + 1; + if (idx <= buffer.length ()) + continue; + } + else + idx = ovector[1]; - if (options.once () || idx >= buffer.length ()) - break; - } - } + if (options.once () || idx >= buffer.length ()) + break; + } + } - retval = regexp::match_data (lst, named_pats); + retval = regexp::match_data (lst, named_pats); - return retval; -} + return retval; + } -bool -regexp::is_match (const std::string& buffer) -{ - regexp::match_data rx_lst = match (buffer); + bool + regexp::is_match (const std::string& buffer) + { + regexp::match_data rx_lst = match (buffer); - return rx_lst.size () > 0; -} + return rx_lst.size () > 0; + } -Array<bool> -regexp::is_match (const string_vector& buffer) -{ - octave_idx_type len = buffer.numel (); + Array<bool> + regexp::is_match (const string_vector& buffer) + { + octave_idx_type len = buffer.numel (); - Array<bool> retval (dim_vector (len, 1)); + Array<bool> retval (dim_vector (len, 1)); - for (octave_idx_type i = 0; i < buffer.numel (); i++) - retval(i) = is_match (buffer(i)); + for (octave_idx_type i = 0; i < buffer.numel (); i++) + retval(i) = is_match (buffer(i)); - return retval; -} + return retval; + } -// Declare rep_token_t used in processing replacement string -typedef struct -{ - size_t pos; - int num; -} rep_token_t; + // Declare rep_token_t used in processing replacement string + typedef struct + { + size_t pos; + int num; + } rep_token_t; -std::string -regexp::replace (const std::string& buffer, const std::string& replacement) -{ - std::string retval; + std::string + regexp::replace (const std::string& buffer, const std::string& replacement) + { + std::string retval; - regexp::match_data rx_lst = match (buffer); + regexp::match_data rx_lst = match (buffer); - size_t num_matches = rx_lst.size (); + size_t num_matches = rx_lst.size (); - if (num_matches == 0) - { - retval = buffer; - return retval; - } + if (num_matches == 0) + { + retval = buffer; + return retval; + } - // Identify replacement tokens; build a vector of group numbers in - // the replacement string so that we can quickly calculate the size - // of the replacement. + // Identify replacement tokens; build a vector of group numbers in + // the replacement string so that we can quickly calculate the size + // of the replacement. - // FIXME: All code assumes that only 10 tokens ($0-$9) exist. - // $11 represents $1 followed by the character '1' rather than - // the eleventh capture buffer. + // FIXME: All code assumes that only 10 tokens ($0-$9) exist. + // $11 represents $1 followed by the character '1' rather than + // the eleventh capture buffer. - std::string repstr = replacement; - std::vector<rep_token_t> tokens; - tokens.reserve (5); // Reserve memory for 5 pattern replacements + std::string repstr = replacement; + std::vector<rep_token_t> tokens; + tokens.reserve (5); // Reserve memory for 5 pattern replacements - for (size_t i=0; i < repstr.size (); i++) - { - if (repstr[i] == '\\') - { - if (i < repstr.size () - 1 && repstr[i+1] == '$') - { - repstr.erase (i,1); // erase backslash - i++; // skip over '$' - continue; - } - if (i < repstr.size () - 1 && repstr[i+1] == '\\') - { - repstr.erase (i,1); // erase 1st backslash - continue; - } - } - else if (repstr[i] == '$') - { - if (i < repstr.size () - 1 && isdigit (repstr[i+1])) - { - rep_token_t tmp_token; + for (size_t i=0; i < repstr.size (); i++) + { + if (repstr[i] == '\\') + { + if (i < repstr.size () - 1 && repstr[i+1] == '$') + { + repstr.erase (i,1); // erase backslash + i++; // skip over '$' + continue; + } + if (i < repstr.size () - 1 && repstr[i+1] == '\\') + { + repstr.erase (i,1); // erase 1st backslash + continue; + } + } + else if (repstr[i] == '$') + { + if (i < repstr.size () - 1 && isdigit (repstr[i+1])) + { + rep_token_t tmp_token; - tmp_token.pos = i; - tmp_token.num = repstr[i+1]-'0'; - tokens.push_back (tmp_token); - } - } - } + tmp_token.pos = i; + tmp_token.num = repstr[i+1]-'0'; + tokens.push_back (tmp_token); + } + } + } - std::string rep; - int num_tokens = tokens.size (); + std::string rep; + int num_tokens = tokens.size (); - if (num_tokens > 0) - { - // Determine replacement length - const size_t replen = repstr.size () - 2*num_tokens; - int delta = 0; - regexp::match_data::const_iterator p = rx_lst.begin (); - for (size_t i = 0; i < num_matches; i++) - { - OCTAVE_QUIT; + if (num_tokens > 0) + { + // Determine replacement length + const size_t replen = repstr.size () - 2*num_tokens; + int delta = 0; + regexp::match_data::const_iterator p = rx_lst.begin (); + for (size_t i = 0; i < num_matches; i++) + { + OCTAVE_QUIT; - double start = p->start (); - double end = p->end (); + double start = p->start (); + double end = p->end (); - const Matrix pairs (p->token_extents ()); - size_t pairlen = 0; - for (int j = 0; j < num_tokens; j++) - { - if (tokens[j].num == 0) - pairlen += static_cast<size_t> (end - start) + 1; - else if (tokens[j].num <= pairs.rows ()) - pairlen += static_cast<size_t> (pairs(tokens[j].num-1,1) - - pairs(tokens[j].num-1,0)) + 1; - } - delta += (static_cast<int> (replen + pairlen) - - static_cast<int> (end - start + 1)); - p++; - } + const Matrix pairs (p->token_extents ()); + size_t pairlen = 0; + for (int j = 0; j < num_tokens; j++) + { + if (tokens[j].num == 0) + pairlen += static_cast<size_t> (end - start) + 1; + else if (tokens[j].num <= pairs.rows ()) + pairlen += static_cast<size_t> (pairs(tokens[j].num-1,1) + - pairs(tokens[j].num-1,0)) + 1; + } + delta += (static_cast<int> (replen + pairlen) + - static_cast<int> (end - start + 1)); + p++; + } - // Build replacement string - rep.reserve (buffer.size () + delta); - size_t from = 0; - p = rx_lst.begin (); - for (size_t i = 0; i < num_matches; i++) - { - OCTAVE_QUIT; + // Build replacement string + rep.reserve (buffer.size () + delta); + size_t from = 0; + p = rx_lst.begin (); + for (size_t i = 0; i < num_matches; i++) + { + OCTAVE_QUIT; - double start = p->start (); - double end = p->end (); + double start = p->start (); + double end = p->end (); - const Matrix pairs (p->token_extents ()); - rep.append (&buffer[from], static_cast<size_t> (start - 1) - from); - from = static_cast<size_t> (end); + const Matrix pairs (p->token_extents ()); + rep.append (&buffer[from], static_cast<size_t> (start - 1) - from); + from = static_cast<size_t> (end); - size_t cur_pos = 0; + size_t cur_pos = 0; - for (int j = 0; j < num_tokens; j++) - { - rep.append (&repstr[cur_pos], (tokens[j].pos) - cur_pos); - cur_pos = tokens[j].pos+2; + for (int j = 0; j < num_tokens; j++) + { + rep.append (&repstr[cur_pos], (tokens[j].pos) - cur_pos); + cur_pos = tokens[j].pos+2; - int k = tokens[j].num; - if (k == 0) - { - // replace with entire match - rep.append (&buffer[static_cast<size_t> (end - 1)], - static_cast<size_t> (end - start) + 1); - } - else if (k <= pairs.rows ()) - { - // replace with group capture - rep.append (&buffer[static_cast<size_t> (pairs(k-1,0)-1)], - static_cast<size_t> (pairs(k-1,1) - - pairs(k-1,0)) + 1); - } - else - { - // replace with nothing - } - } - if (cur_pos < repstr.size ()) - rep.append (&repstr[cur_pos], repstr.size () - cur_pos); + int k = tokens[j].num; + if (k == 0) + { + // replace with entire match + rep.append (&buffer[static_cast<size_t> (end - 1)], + static_cast<size_t> (end - start) + 1); + } + else if (k <= pairs.rows ()) + { + // replace with group capture + rep.append (&buffer[static_cast<size_t> (pairs(k-1,0)-1)], + static_cast<size_t> (pairs(k-1,1) + - pairs(k-1,0)) + 1); + } + else + { + // replace with nothing + } + } + if (cur_pos < repstr.size ()) + rep.append (&repstr[cur_pos], repstr.size () - cur_pos); - p++; - } - rep.append (&buffer[from], buffer.size () - from); - } - else - { - // Determine repstr length - const size_t replen = repstr.size (); - int delta = 0; - regexp::match_data::const_iterator p = rx_lst.begin (); - for (size_t i = 0; i < num_matches; i++) - { - OCTAVE_QUIT; - delta += static_cast<int> (replen) - - static_cast<int> (p->end () - p->start () + 1); - p++; - } + p++; + } + rep.append (&buffer[from], buffer.size () - from); + } + else + { + // Determine repstr length + const size_t replen = repstr.size (); + int delta = 0; + regexp::match_data::const_iterator p = rx_lst.begin (); + for (size_t i = 0; i < num_matches; i++) + { + OCTAVE_QUIT; + delta += static_cast<int> (replen) + - static_cast<int> (p->end () - p->start () + 1); + p++; + } - // Build replacement string - rep.reserve (buffer.size () + delta); - size_t from = 0; - p = rx_lst.begin (); - for (size_t i = 0; i < num_matches; i++) - { - OCTAVE_QUIT; - rep.append (&buffer[from], - static_cast<size_t> (p->start () - 1) - from); - from = static_cast<size_t> (p->end ()); - rep.append (repstr); - p++; - } - rep.append (&buffer[from], buffer.size () - from); - } + // Build replacement string + rep.reserve (buffer.size () + delta); + size_t from = 0; + p = rx_lst.begin (); + for (size_t i = 0; i < num_matches; i++) + { + OCTAVE_QUIT; + rep.append (&buffer[from], + static_cast<size_t> (p->start () - 1) - from); + from = static_cast<size_t> (p->end ()); + rep.append (repstr); + p++; + } + rep.append (&buffer[from], buffer.size () - from); + } - retval = rep; - return retval; + retval = rep; + return retval; + } }
--- a/liboctave/util/lo-regexp.h Wed Aug 17 10:30:50 2016 -0700 +++ b/liboctave/util/lo-regexp.h Wed Aug 17 17:57:49 2016 -0400 @@ -35,247 +35,294 @@ #include "base-list.h" #include "str-vec.h" -class -OCTAVE_API -regexp +namespace octave { -public: - - class opts; - class match_data; - - regexp (const std::string& pat = "", - const regexp::opts& opt = regexp::opts (), - const std::string& w = "regexp") - : pattern (pat), options (opt), data (0), named_pats (), - nnames (0), named_idx (), who (w) - { - compile_internal (); - } - - regexp (const regexp& rx) - : pattern (rx.pattern), data (rx.data), named_pats (rx.named_pats), - nnames (rx.nnames), named_idx (rx.named_idx) - { } - - regexp& operator = (const regexp& rx) - { - if (this != &rx) - { - pattern = rx.pattern; - data = rx.data; - named_pats = rx.named_pats; - nnames = rx.nnames; - named_idx = rx.named_idx; - } - - return *this; - } - - ~regexp (void) { free (); } - - void compile (const std::string& pat, - const regexp::opts& opt = regexp::opts ()) - { - pattern = pat; - options = opt; - compile_internal (); - } - - match_data match (const std::string& buffer); - - bool is_match (const std::string& buffer); - - Array<bool> is_match (const string_vector& buffer); - - std::string replace (const std::string& buffer, - const std::string& replacement); - - class opts + class + OCTAVE_API + regexp { public: - opts (void) - : x_case_insensitive (false), x_dotexceptnewline (false), - x_emptymatch (false), x_freespacing (false), x_lineanchors (false), - x_once (false) { } + class opts; + class match_data; + + regexp (const std::string& pat = "", + const regexp::opts& opt = regexp::opts (), + const std::string& w = "regexp") + : pattern (pat), options (opt), data (0), named_pats (), + nnames (0), named_idx (), who (w) + { + compile_internal (); + } + + regexp (const regexp& rx) + : pattern (rx.pattern), data (rx.data), named_pats (rx.named_pats), + nnames (rx.nnames), named_idx (rx.named_idx) + { } - opts (const opts& o) - : x_case_insensitive (o.x_case_insensitive), - x_dotexceptnewline (o.x_dotexceptnewline), - x_emptymatch (o.x_emptymatch), - x_freespacing (o.x_freespacing), - x_lineanchors (o.x_lineanchors), - x_once (o.x_once) - { } + regexp& operator = (const regexp& rx) + { + if (this != &rx) + { + pattern = rx.pattern; + data = rx.data; + named_pats = rx.named_pats; + nnames = rx.nnames; + named_idx = rx.named_idx; + } + + return *this; + } + + ~regexp (void) { free (); } + + void compile (const std::string& pat, + const regexp::opts& opt = regexp::opts ()) + { + pattern = pat; + options = opt; + compile_internal (); + } + + match_data match (const std::string& buffer); + + bool is_match (const std::string& buffer); - opts& operator = (const opts& o) + Array<bool> is_match (const string_vector& buffer); + + std::string replace (const std::string& buffer, + const std::string& replacement); + + static regexp::match_data + match (const std::string& pat, const std::string& buffer, + const regexp::opts& opt = regexp::opts (), + const std::string& who = "regexp") + { + regexp rx (pat, opt, who); + + return rx.match (buffer); + } + + static bool + is_match (const std::string& pat, const std::string& buffer, + const regexp::opts& opt = regexp::opts (), + const std::string& who = "regexp") { - if (this != &o) - { - x_case_insensitive = o.x_case_insensitive; - x_dotexceptnewline = o.x_dotexceptnewline; - x_emptymatch = o.x_emptymatch; - x_freespacing = o.x_freespacing; - x_lineanchors = o.x_lineanchors; - x_once = o.x_once; - } + regexp rx (pat, opt, who); + + return rx.is_match (buffer); + } + + static Array<bool> + is_match (const std::string& pat, const string_vector& buffer, + const regexp::opts& opt = regexp::opts (), + const std::string& who = "regexp") + { + regexp rx (pat, opt, who); - return *this; + return rx.is_match (buffer); + } + + static std::string + replace (const std::string& pat, const std::string& buffer, + const std::string& replacement, + const regexp::opts& opt = regexp::opts (), + const std::string& who = "regexp") + { + regexp rx (pat, opt, who); + + return rx.replace (buffer, replacement); } - ~opts (void) { } + class opts + { + public: + + opts (void) + : x_case_insensitive (false), x_dotexceptnewline (false), + x_emptymatch (false), x_freespacing (false), x_lineanchors (false), + x_once (false) { } + + opts (const opts& o) + : x_case_insensitive (o.x_case_insensitive), + x_dotexceptnewline (o.x_dotexceptnewline), + x_emptymatch (o.x_emptymatch), + x_freespacing (o.x_freespacing), + x_lineanchors (o.x_lineanchors), + x_once (o.x_once) + { } + + opts& operator = (const opts& o) + { + if (this != &o) + { + x_case_insensitive = o.x_case_insensitive; + x_dotexceptnewline = o.x_dotexceptnewline; + x_emptymatch = o.x_emptymatch; + x_freespacing = o.x_freespacing; + x_lineanchors = o.x_lineanchors; + x_once = o.x_once; + } + + return *this; + } + + ~opts (void) { } + + void case_insensitive (bool val) { x_case_insensitive = val; } + void dotexceptnewline (bool val) { x_dotexceptnewline = val; } + void emptymatch (bool val) { x_emptymatch = val; } + void freespacing (bool val) { x_freespacing = val; } + void lineanchors (bool val) { x_lineanchors = val; } + void once (bool val) { x_once = val; } + + bool case_insensitive (void) const { return x_case_insensitive; } + bool dotexceptnewline (void) const { return x_dotexceptnewline; } + bool emptymatch (void) const { return x_emptymatch; } + bool freespacing (void) const { return x_freespacing; } + bool lineanchors (void) const { return x_lineanchors; } + bool once (void) const { return x_once; } + + private: + + bool x_case_insensitive; + bool x_dotexceptnewline; + bool x_emptymatch; + bool x_freespacing; + bool x_lineanchors; + bool x_once; + }; + + class match_element + { + public: - void case_insensitive (bool val) { x_case_insensitive = val; } - void dotexceptnewline (bool val) { x_dotexceptnewline = val; } - void emptymatch (bool val) { x_emptymatch = val; } - void freespacing (bool val) { x_freespacing = val; } - void lineanchors (bool val) { x_lineanchors = val; } - void once (bool val) { x_once = val; } + match_element (const string_vector& nt, const string_vector& t, + const std::string& ms, const Matrix& te, + double s, double e) + : x_match_string (ms), x_named_tokens (nt), x_tokens (t), + x_token_extents (te), x_start (s), x_end (e) + { } + + match_element (const match_element &a) + : x_match_string (a.x_match_string), + x_named_tokens (a.x_named_tokens), x_tokens (a.x_tokens), + x_token_extents (a.x_token_extents), + x_start (a.x_start), x_end (a.x_end) + { } + + std::string match_string (void) const { return x_match_string; } + string_vector named_tokens (void) const { return x_named_tokens; } + string_vector tokens (void) const { return x_tokens; } + Matrix token_extents (void) const { return x_token_extents; } + double start (void) const { return x_start; } + double end (void) const { return x_end; } + + private: + + std::string x_match_string; + string_vector x_named_tokens; + string_vector x_tokens; + Matrix x_token_extents; + double x_start; + double x_end; + }; - bool case_insensitive (void) const { return x_case_insensitive; } - bool dotexceptnewline (void) const { return x_dotexceptnewline; } - bool emptymatch (void) const { return x_emptymatch; } - bool freespacing (void) const { return x_freespacing; } - bool lineanchors (void) const { return x_lineanchors; } - bool once (void) const { return x_once; } + class match_data : public base_list<match_element> + { + public: + + match_data (void) + : base_list<match_element> (), named_pats () + { } + + match_data (const std::list<match_element>& l, const string_vector& np) + : base_list<match_element> (l), named_pats (np) + { } + + match_data (const match_data& rx_lst) + : base_list<match_element> (rx_lst), + named_pats (rx_lst.named_pats) + { } + + match_data& operator = (const match_data& rx_lst) + { + if (this != &rx_lst) + { + base_list<match_element>::operator = (rx_lst); + named_pats = rx_lst.named_pats; + } + + return *this; + } + + ~match_data (void) { } + + string_vector named_patterns (void) { return named_pats; } + + private: + + string_vector named_pats; + }; private: - bool x_case_insensitive; - bool x_dotexceptnewline; - bool x_emptymatch; - bool x_freespacing; - bool x_lineanchors; - bool x_once; - }; - - class match_element - { - public: - - match_element (const string_vector& nt, const string_vector& t, - const std::string& ms, const Matrix& te, - double s, double e) - : x_match_string (ms), x_named_tokens (nt), x_tokens (t), - x_token_extents (te), x_start (s), x_end (e) - { } + // The pattern we've been asked to match. + std::string pattern; - match_element (const match_element &a) - : x_match_string (a.x_match_string), - x_named_tokens (a.x_named_tokens), x_tokens (a.x_tokens), - x_token_extents (a.x_token_extents), - x_start (a.x_start), x_end (a.x_end) - { } - - std::string match_string (void) const { return x_match_string; } - string_vector named_tokens (void) const { return x_named_tokens; } - string_vector tokens (void) const { return x_tokens; } - Matrix token_extents (void) const { return x_token_extents; } - double start (void) const { return x_start; } - double end (void) const { return x_end; } - - private: + opts options; - std::string x_match_string; - string_vector x_named_tokens; - string_vector x_tokens; - Matrix x_token_extents; - double x_start; - double x_end; - }; - - class match_data : public octave_base_list<match_element> - { - public: - - match_data (void) - : octave_base_list<match_element> (), named_pats () - { } + // Internal data describing the regular expression. + void *data; - match_data (const std::list<match_element>& l, const string_vector& np) - : octave_base_list<match_element> (l), named_pats (np) - { } - - match_data (const match_data& rx_lst) - : octave_base_list<match_element> (rx_lst), - named_pats (rx_lst.named_pats) - { } + std::string m; + string_vector named_pats; + int nnames; + Array<int> named_idx; + std::string who; - match_data& operator = (const match_data& rx_lst) - { - if (this != &rx_lst) - { - octave_base_list<match_element>::operator = (rx_lst); - named_pats = rx_lst.named_pats; - } - - return *this; - } - - ~match_data (void) { } - - string_vector named_patterns (void) { return named_pats; } + void free (void); - private: - - string_vector named_pats; + void compile_internal (void); }; - -private: - - // The pattern we've been asked to match. - std::string pattern; - - opts options; +} - // Internal data describing the regular expression. - void *data; +#if defined (OCTAVE_USE_DEPRECATED_FUNCTIONS) - std::string m; - string_vector named_pats; - int nnames; - Array<int> named_idx; - std::string who; +OCTAVE_DEPRECATED ("use 'octave::regexp' instead") +typedef octave::regexp regexp; - void free (void); - - void compile_internal (void); -}; - +OCTAVE_DEPRECATED ("use 'octave::regexp::match' instead") inline regexp::match_data regexp_match (const std::string& pat, const std::string& buffer, const regexp::opts& opt = regexp::opts (), const std::string& who = "regexp") { - regexp rx (pat, opt, who); - - return rx.match (buffer); + return octave::regexp::match (pat, buffer, opt, who); } +OCTAVE_DEPRECATED ("use 'octave::regexp::is_match' instead") inline bool is_regexp_match (const std::string& pat, const std::string& buffer, const regexp::opts& opt = regexp::opts (), const std::string& who = "regexp") { - regexp rx (pat, opt, who); - - return rx.is_match (buffer); + return octave::regexp::is_match (pat, buffer, opt, who); } +OCTAVE_DEPRECATED ("use 'octave::regexp::is_match' instead") inline Array<bool> is_regexp_match (const std::string& pat, const string_vector& buffer, const regexp::opts& opt = regexp::opts (), const std::string& who = "regexp") { - regexp rx (pat, opt, who); - - return rx.is_match (buffer); + return octave::regexp::is_match (pat, buffer, opt, who); } +OCTAVE_DEPRECATED ("use 'octave::regexp::replace' instead") inline std::string regexp_replace (const std::string& pat, const std::string& buffer, @@ -283,9 +330,9 @@ const regexp::opts& opt = regexp::opts (), const std::string& who = "regexp") { - regexp rx (pat, opt, who); - - return rx.replace (buffer, replacement); + return octave::regexp::replace (pat, buffer, replacement, opt, who); } #endif + +#endif