summaryrefslogtreecommitdiff
path: root/Master/texmf-dist/tex/latex/l3experimental/l3str/l3regex.sty
diff options
context:
space:
mode:
authorKarl Berry <karl@freefriends.org>2011-12-29 23:03:28 +0000
committerKarl Berry <karl@freefriends.org>2011-12-29 23:03:28 +0000
commit0bc9393ae50667d7ff769a76321a9548a70351fc (patch)
tree6da7cc8ebef423a4653c82a19be6d4e1c6652705 /Master/texmf-dist/tex/latex/l3experimental/l3str/l3regex.sty
parent56ecd0d3d0de54cebdeb29cc8a056001ee93b02e (diff)
l3experimental 3109 (29dec11)
git-svn-id: svn://tug.org/texlive/trunk@24970 c570f23f-e606-0410-a88d-b1316a301751
Diffstat (limited to 'Master/texmf-dist/tex/latex/l3experimental/l3str/l3regex.sty')
-rw-r--r--Master/texmf-dist/tex/latex/l3experimental/l3str/l3regex.sty1443
1 files changed, 885 insertions, 558 deletions
diff --git a/Master/texmf-dist/tex/latex/l3experimental/l3str/l3regex.sty b/Master/texmf-dist/tex/latex/l3experimental/l3str/l3regex.sty
index f3dc0b5a358..facf764c168 100644
--- a/Master/texmf-dist/tex/latex/l3experimental/l3str/l3regex.sty
+++ b/Master/texmf-dist/tex/latex/l3experimental/l3str/l3regex.sty
@@ -46,17 +46,19 @@
%%
%% -----------------------------------------------------------------------
\RequirePackage{expl3}
-\GetIdInfo$Id: l3regex.dtx 3017 2011-11-28 00:10:02Z bruno $
+\GetIdInfo$Id: l3regex.dtx 3073 2011-12-21 01:53:06Z bruno $
{L3 Experimental Regular Expressions}
\ProvidesExplPackage
{\ExplFileName}{\ExplFileDate}{\ExplFileVersion}{\ExplFileDescription}
-\RequirePackage{l3str}
+\RequirePackage{l3str, l3tl-analysis, l3flag}
\cs_new:Npn \regex_tmp:w { }
-\tl_new:N \g_regex_tmpa_tl
\tl_new:N \l_regex_tmpa_tl
\tl_new:N \l_regex_tmpb_tl
+\tl_new:N \l_regex_tmpc_tl
\int_new:N \l_regex_tmpa_int
\int_new:N \l_regex_tmpb_int
+\flag_new:N \l_regex_group_begin_flag
+\flag_new:N \l_regex_group_end_flag
\int_new:N \l_regex_max_state_int
\int_new:N \l_regex_left_state_int
\int_new:N \l_regex_right_state_int
@@ -67,7 +69,10 @@
\int_new:N \l_regex_capturing_group_int
\seq_new:N \l_regex_capturing_group_seq
\int_new:N \l_regex_capturing_group_max_int
+\tl_new:N \l_regex_build_or_class_tl
\tl_new:N \l_regex_one_or_group_tl
+\bool_new:N \l_regex_in_csname_bool
+\tl_new:N \l_regex_tmpa_regex_tl
\cs_new_eq:NN \regex_build_tmp_class:n \use_none:n
\bool_new:N \l_regex_class_bool
\tl_new:N \l_regex_class_tl
@@ -133,12 +138,16 @@
\regex_item_geq:n { \c_eleven } % > lf
\regex_item_range:nn { \c_zero } { \c_nine } % < lf
}
-\tl_new:N \l_regex_query_other_str
+\int_new:N \l_regex_nesting_int
+\int_new:N \l_regex_min_step_int
+\int_new:N \l_regex_max_step_int
\int_new:N \l_regex_current_step_int
\int_new:N \l_regex_start_step_int
\int_new:N \l_regex_success_step_int
-\int_new:N \l_regex_unique_step_int
+\int_new:N \l_regex_unique_id_int
\int_new:N \l_regex_current_char_int
+\int_new:N \l_regex_current_catcode_int
+\tl_new:N \l_regex_current_token_tl
\int_new:N \l_regex_last_char_int
\int_new:N \l_regex_case_changed_char_int
\bool_new:N \l_regex_caseless_bool
@@ -147,23 +156,23 @@
\prop_new:N \l_regex_success_submatches_prop
\int_new:N \l_regex_max_index_int
\tl_new:N \l_regex_every_match_tl
-\bool_new:N \l_regex_success_bool
+\bool_new:N \g_regex_success_bool
+\bool_new:N \l_regex_saved_success_bool
+\bool_new:N \l_regex_success_match_bool
\cs_new_protected:Npn \regex_last_match_empty_no:F #1 {#1}
\cs_new_protected:Npn \regex_last_match_empty_yes:F
{ \int_compare:nNnF \l_regex_start_step_int = \l_regex_current_step_int }
\cs_new_eq:NN \regex_last_match_empty:F \regex_last_match_empty_no:F
\bool_new:N \l_regex_success_empty_bool
\bool_new:N \l_regex_fresh_thread_bool
-\seq_new:N \g_regex_submatches_seq
-\int_new:N \g_regex_match_count_int
-\seq_new:N \g_regex_extract_all_seq
-\seq_new:N \g_regex_split_seq
-\tl_new:N \l_regex_replacement_tl
-\tl_new:N \g_regex_replaced_str
-\cs_new_eq:NN \s_regex_stop \scan_stop:
+\int_new:N \l_regex_replacement_int
+\int_new:N \l_regex_replacement_csnames_int
+\int_new:N \l_regex_submatch_int
+\int_new:N \l_regex_submatch_start_int
+\int_new:N \l_regex_match_count_int
\cs_new_protected:Npn \regex_toks_put_left:Nx #1#2
{
- \cs_set_nopar:Npx \regex_tmp:w \s_regex_stop { \s_regex_stop #2 }
+ \cs_set_nopar:Npx \regex_tmp:w \s_stop { \s_stop #2 }
\tex_toks:D #1 \exp_after:wN \exp_after:wN \exp_after:wN
{ \exp_after:wN \regex_tmp:w \tex_the:D \tex_toks:D #1 }
}
@@ -173,18 +182,24 @@
\tex_toks:D #1 \exp_after:wN
{ \tex_the:D \tex_toks:D \exp_after:wN #1 \regex_tmp:w }
}
-\cs_new_nopar:Npn \regex_query_substr:NN #1#2
+\cs_new:Npn \regex_query_substr:nn #1#2
{
- \exp_after:wN \regex_query_substr_aux:NN
- \exp_after:wN #1
- \exp_after:wN #2
- \l_regex_query_other_str
- \q_stop
+ \if_num:w \int_eval:w #1 < \int_eval:w #2 \int_eval_end:
+ \str_aux_toks_range:nn {#1} {#2}
+ \fi:
+ }
+\cs_new:Npn \regex_query_submatch:nn #1#2
+ {
+ \if_num:w #1 < \l_regex_capturing_group_int
+ \exp_after:wN \regex_query_submatch:w
+ \int_use:N \int_eval:w #1 + #2 ;
+ \fi:
}
-\cs_new:Npn \regex_query_substr_aux:NN #1#2
+\cs_new:Npn \regex_query_submatch:w #1 ;
{
- \str_skip_do:nn {#1}
- { \str_collect_do:nn { #2 - #1 } { \use_i_delimit_by_q_stop:nw } }
+ \regex_query_substr:nn
+ { \tex_skip:D #1 }
+ { \etex_gluestretch:D \tex_skip:D #1 }
}
\cs_new_protected:Npn \regex_seq_pop_int:NN #1#2
{
@@ -208,8 +223,27 @@
\seq_get:NN #1 \l_regex_tmpa_tl
\l_regex_tmpa_tl
}
+\cs_new:Npn \regex_seq_item:Nn #1#2
+ {
+ \exp_after:wN \regex_seq_item_aux:wNn
+ \int_use:N \int_eval:w #2 \exp_after:wN ;
+ #1 { ? \prg_map_break: } { }
+ \prg_break_point:n { }
+ }
+\cs_new:Npn \regex_seq_item_aux:wNn #1; #2#3
+ {
+ \use_none:n #2
+ \if_num:w #1 = \c_zero
+ \exp_after:wN \prg_map_break:n
+ \else:
+ \exp_after:wN \use_none:n
+ \fi:
+ { \exp_not:n {#3} }
+ \exp_after:wN \regex_seq_item_aux:wNn
+ \int_use:N \int_eval:w #1 - \c_one ;
+ }
\cs_new_protected_nopar:Npn \regex_item_dot:T
- { \int_compare:nNnF \l_regex_current_char_int = \c_minus_one }
+ { \int_compare:nNnF \l_regex_current_char_int < \c_minus_one }
\cs_new_protected_nopar:Npn \regex_break_true:w
#1 \regex_break_point:TF #2 #3 {#2}
\cs_new_protected_nopar:Npn \regex_break_false:w
@@ -283,6 +317,37 @@
\cs_set:Npn \regex_item_range:nn { \regex_item_caseful_range:nn }
\cs_set:Npn \regex_item_geq:n { \regex_item_caseful_geq:n }
}
+\cs_new_protected_nopar:Npn \regex_item_catcode:n #1
+ {
+ \if_num:w #1 = \l_regex_current_catcode_int
+ \exp_after:wN \regex_break_true:w
+ \fi:
+ }
+\cs_new_protected_nopar:Npn \regex_item_catcode_neq:n #1
+ {
+ \if_num:w #1 = \l_regex_current_catcode_int
+ \else:
+ \exp_after:wN \regex_break_true:w
+ \fi:
+ }
+\cs_new_protected:Npn \regex_item_cs:n #1
+ {
+ \int_compare:nNnT \l_regex_current_catcode_int = \c_zero
+ {
+ \tl_set:Nn \l_regex_tmpa_regex_tl {#1}
+ \bool_set_eq:NN \l_regex_saved_success_bool \g_regex_success_bool
+ \exp_args:NNx \regex_match:NnTF \l_regex_tmpa_regex_tl
+ {
+ \exp_after:wN \exp_after:wN
+ \exp_after:wN \cs_to_str:N \l_regex_current_token_tl
+ }
+ {
+ \bool_gset_eq:NN \g_regex_success_bool \l_regex_saved_success_bool
+ \regex_break_true:w
+ }
+ { \bool_gset_eq:NN \g_regex_success_bool \l_regex_saved_success_bool }
+ }
+ }
\cs_new_protected:Npn \regex_get_digits:nw #1
{
\tex_afterassignment:D \regex_tmp:w
@@ -362,14 +427,10 @@
}
\cs_new_protected_nopar:Npn \regex_build_new_state:
{
- \int_compare:nNnTF \l_regex_max_state_int > { 32766 }
- { \msg_error:nn { regex } { -997 } }
- {
- \int_incr:N \l_regex_max_state_int
- \tex_toks:D \l_regex_max_state_int { \s_regex_stop }
- }
+ \tex_toks:D \l_regex_max_state_int { \s_stop }
\int_set_eq:NN \l_regex_left_state_int \l_regex_right_state_int
\int_set_eq:NN \l_regex_right_state_int \l_regex_max_state_int
+ \int_incr:N \l_regex_max_state_int
}
\cs_new_protected_nopar:Npn \regex_build_transition_aux:NN #1#2
{
@@ -388,16 +449,18 @@
}
\cs_new_protected:Npn \regex_build:n #1
{
- \regex_build_setup:
- \str_aux_escape:NNNn
- \regex_build_i_unescaped:N
- \regex_build_i_escaped:N
- \regex_build_i_raw:N
- { #1 }
+ \bool_set_false:N \l_regex_in_csname_bool
+ \regex_build:w
+ \regex_build_new_state:
+ \regex_toks_put_right:Nx \l_regex_left_state_int
+ { \regex_action_start_wildcard:nn {0} {1} }
\regex_build_open_aux:
- \g_str_result_tl \prg_do_nothing: \prg_do_nothing:
- \regex_seq_push_int:NN \l_regex_capturing_group_seq \c_zero
- \regex_build_close_aux: \regex_build_group_:
+ \str_escape_use:NNNn
+ \regex_build_i_unescaped:N
+ \regex_build_i_escaped:N
+ \regex_build_i_raw:N
+ { #1 }
+ \prg_do_nothing: \prg_do_nothing:
\regex_build_end:
}
\cs_new_nopar:Npn \regex_build_i_unescaped:N #1
@@ -414,38 +477,39 @@
}
\cs_new_nopar:Npn \regex_build_i_raw:N #1
{ \exp_not:N \regex_build_raw:N #1 }
-\cs_new_protected_nopar:Npn \regex_build_default_control:N #1
+\cs_new_protected_nopar:Npn \regex_build_control:N #1
{
- \cs_if_exist_use:cF { regex_build_#1: }
- { \regex_build_default_raw:N #1 }
+ \cs_if_exist_use:cF { regex_ \l_regex_build_or_class_tl _#1: }
+ { \use:c { regex_ \l_regex_build_or_class_tl _default_raw:N } #1 }
}
-\cs_new_protected_nopar:Npn \regex_build_setup:
+\cs_new_protected_nopar:Npn \regex_build:w
{
- \cs_set_eq:NN \regex_build_control:N \regex_build_default_control:N
\cs_set_eq:NN \regex_build_raw:N \regex_build_default_raw:N
+ \tl_set:Nn \l_regex_build_or_class_tl { build }
\int_set_eq:NN \l_regex_capturing_group_int \c_zero
\int_zero:N \l_regex_max_state_int
\regex_build_new_state:
- \regex_build_new_state:
- \regex_toks_put_right:Nx \l_regex_left_state_int
- { \regex_action_start_wildcard:nn {0} {1} }
}
\cs_new_protected_nopar:Npn \regex_build_end:
{
+ \regex_seq_push_int:NN \l_regex_capturing_group_seq \c_zero
+ \regex_build_close_aux: \regex_build_group_:
\seq_if_empty:NF \l_regex_capturing_group_seq
{
- \msg_error:nn { regex } {22}
+ \msg_kernel_error:nnx { regex } { missing-rparen }
+ { \seq_length:N \l_regex_capturing_group_seq }
\prg_replicate:nn
{ \seq_length:N \l_regex_capturing_group_seq }
{ \regex_build_close_aux: \regex_build_group_: }
}
\regex_toks_put_right:Nx \l_regex_right_state_int
{ \regex_action_success: }
+ \int_incr:N \l_regex_capturing_group_int
}
\cs_new_protected_nopar:cpn { regex_build_^: }
- { \regex_build_anchor_start:N \c_zero }
+ { \regex_build_anchor_start:N \l_regex_min_step_int }
\cs_new_protected_nopar:Npn \regex_build_A:
- { \regex_build_anchor_start:N \c_zero }
+ { \regex_build_anchor_start:N \l_regex_min_step_int }
\cs_new_protected_nopar:Npn \regex_build_G:
{ \regex_build_anchor_start:N \l_regex_start_step_int }
\cs_new_protected_nopar:Npn \regex_build_anchor_start:N #1
@@ -453,7 +517,7 @@
\regex_build_new_state:
\regex_toks_put_right:Nx \l_regex_left_state_int
{
- \exp_not:N \int_compare:nNnT {#1} = \l_regex_current_step_int
+ \exp_not:N \int_compare:nNnT #1 = \l_regex_current_step_int
{
\regex_action_free:n
{
@@ -469,7 +533,7 @@
\regex_toks_put_right:Nx \l_regex_left_state_int
{
\exp_not:N \int_compare:nNnT
- \c_minus_one = \l_regex_current_char_int
+ \l_regex_current_char_int < \c_minus_one
{
\regex_action_free:n
{
@@ -519,61 +583,52 @@
\int_set_eq:NN \l_regex_current_char_int \l_regex_last_char_int
\c_regex_w_tl
\regex_break_point:TF
- { \group_end: \c_regex_W_tl \regex_item_equal:n { -1 } }
+ { \group_end: \c_regex_W_tl \regex_item_equal:n { -2 } }
{ \group_end: \c_regex_w_tl }
\regex_break_point:TF
}
-\cs_new_protected_nopar:Npn \regex_build_default_raw:N #1
+\cs_new_protected:Npn \regex_build_one:n #1
+ { \regex_build_one:x { \exp_not:n {#1} } }
+\cs_new_protected:Npn \regex_build_one:x #1
{
- \cs_set:Npx \regex_build_tmp_class:n ##1
+ \cs_set_nopar:Npx \regex_build_tmp_class:n ##1
{
- \regex_item_equal:n { \int_value:w `#1 ~ }
- \regex_break_point:TF { \regex_action_cost:n { ##1 } } { }
+ \exp_not:N \exp_not:n {#1}
+ \regex_break_point:TF { \regex_action_cost:n {##1} } { }
}
\regex_build_one_quantifier:
}
+\cs_new_protected_nopar:Npn \regex_build_default_raw:N #1
+ { \regex_build_one:x { \regex_item_equal:n { \int_value:w `#1 ~ } } }
\cs_new_protected_nopar:cpn { regex_build_.: }
{
\cs_set:Npn \regex_build_tmp_class:n ##1
{ \regex_item_dot:T { \regex_action_cost:n {##1} } }
\regex_build_one_quantifier:
}
-\cs_new_protected_nopar:Npn \regex_build_char_type:N #1
- {
- \cs_set:Npn \regex_build_tmp_class:n ##1
- {
- \exp_not:N #1
- \regex_break_point:TF { \regex_action_cost:n {##1} } { }
- }
- \regex_build_one_quantifier:
- }
\tl_map_inline:nn { dDhHsSvVwWN }
{
\cs_new_protected_nopar:cpx { regex_build_#1: }
- {
- \exp_not:N \regex_build_char_type:N
- \exp_not:c { c_regex_#1_tl }
- }
+ { \regex_build_one:n \exp_not:c { c_regex_#1_tl } }
}
+\cs_new_protected:Npn \regex_class_one:n
+ { \tl_put_right:Nn \l_regex_class_tl }
+\cs_new_protected:Npn \regex_class_one:x
+ { \tl_put_right:Nx \l_regex_class_tl }
\cs_new_protected_nopar:cpn { regex_build_[: }
{
\tl_clear:N \l_regex_class_tl
- \cs_set_eq:NN \regex_build_control:N \regex_class_control:N
- \cs_set_eq:NN \regex_build_raw:N \regex_class_raw:N
+ \cs_set_eq:NN \regex_build_raw:N \regex_class_default_raw:N
+ \tl_set:Nn \l_regex_build_or_class_tl { class }
\regex_class_first:NN
}
-\cs_new_protected_nopar:Npn \regex_class_control:N #1
- {
- \cs_if_exist_use:cF { regex_class_#1: }
- { \regex_class_raw:N #1 }
- }
\cs_new_protected_nopar:cpn { regex_class_]: }
{
\tl_if_empty:NTF \l_regex_class_tl %[
- { \regex_class_raw:N ] }
+ { \regex_class_default_raw:N ] }
{
- \cs_set_eq:NN \regex_build_control:N \regex_build_default_control:N
\cs_set_eq:NN \regex_build_raw:N \regex_build_default_raw:N
+ \tl_set:Nn \l_regex_build_or_class_tl { build }
\cs_set:Npn \regex_build_tmp_class:n ##1
{
\exp_not:o \l_regex_class_tl
@@ -593,7 +648,7 @@
#1 #2
}
}
-\cs_new_protected_nopar:Npn \regex_class_raw:N #1#2#3
+\cs_new_protected_nopar:Npn \regex_class_default_raw:N #1#2#3
{
\str_if_eq:nnTF {#2#3} { \regex_build_control:N - }
{ \regex_class_range:Nw #1 }
@@ -603,10 +658,7 @@
}
}
\cs_new_protected_nopar:Npn \regex_class_single:N #1
- {
- \tl_put_right:Nx \l_regex_class_tl
- { \regex_item_equal:n { \int_value:w `#1 } }
- }
+ { \regex_class_one:x { \regex_item_equal:n { \int_value:w `#1 } } }
\cs_new_protected_nopar:Npn \regex_class_range:Nw #1#2#3
{
\token_if_eq_meaning:NNTF #2 \regex_build_control:N
@@ -624,9 +676,9 @@
\cs_new_protected_nopar:Npn \regex_class_range_put:NN #1#2
{
\if_num:w `#1 > `#2 \exp_stop_f:
- \msg_error:nn { regex } {8}
+ \msg_kernel_error:nnxx { regex } { backwards-range } {#1} {#2}
\else:
- \tl_put_right:Nx \l_regex_class_tl
+ \regex_class_one:x
{
\if_num:w `#1 = `#2 \exp_stop_f:
\regex_item_equal:n
@@ -640,10 +692,93 @@
\tl_map_inline:nn { dDhHsSvVwWN }
{
\cs_new_protected_nopar:cpx { regex_class_#1: }
+ { \regex_class_one:n \exp_not:c { c_regex_#1_tl } }
+ }
+\cs_new_protected_nopar:Npn \regex_build_c:
+ { \regex_aux_c:NNNN \regex_item_catcode:n \use_i:nn }
+\cs_new_protected_nopar:Npn \regex_build_C:
+ { \regex_aux_c:NNNN \regex_item_catcode_neq:n \use_ii:nn }
+\cs_new_eq:NN \regex_class_c: \regex_build_c:
+\cs_new_eq:NN \regex_class_C: \regex_build_C:
+\cs_new_protected:Npn \regex_aux_c:NNNN #1#2#3#4
+ {
+ \token_if_eq_meaning:NNTF #3 \regex_build_raw:N
+ {
+ \regex_aux_c_test:NTF #4
+ {
+ \use:c { regex_ \l_regex_build_or_class_tl _one:x }
+ { #1 { \int_use:N \l_regex_current_catcode_int } }
+ }
+ {
+ \msg_kernel_error:nnx { regex } { build-c } { #4 }
+ #3 #4
+ }
+ }
+ {
+ \if_charcode:w \c_lbrace_str \exp_not:N #4
+ \exp_after:wN #2
+ \else:
+ \exp_after:wN \use_ii:nn
+ \fi:
+ {
+ \group_begin:
+ \bool_set_true:N \l_regex_in_csname_bool
+ \regex_build:w
+ }
+ {
+ \msg_kernel_error:nnx { regex } { build-c } { #4 }
+ #3 #4
+ }
+ }
+ }
+\cs_new_protected:Npn \regex_aux_c_test:NTF #1
+ {
+ \int_set:Nn \l_regex_current_catcode_int
+ {
+ \if_case:w \int_eval:w `#1 - `A \int_eval_end:
+ \c_thirteen % Active
+ \or: \c_one % Begin-group
+ \or: \c_zero % Control sequence
+ \or: \c_eight % Down (subscript)
+ \or: \c_two % End-group
+ \else:
+ \if_case:w \int_eval:w `#1 - `L \int_eval_end:
+ \c_eleven % Letter
+ \or: \c_three % Math toggle
+ \or: \c_minus_one
+ \or: \c_twelve % Other
+ \or: \c_six % Parameter
+ \or: \c_minus_one
+ \or: \c_minus_one
+ \or: \c_ten % Space
+ \or: \c_four % Tab (alignment)
+ \or: \c_seven % Up (superscript)
+ \else: \c_minus_one
+ \fi:
+ \fi:
+ }
+ \if_num:w \l_regex_current_catcode_int = \c_minus_one
+ \exp_after:wN \use_ii:nn
+ \else:
+ \exp_after:wN \use_i:nn
+ \fi:
+ }
+\cs_new_protected:cpn { regex_build_ \c_rbrace_str : }
+ {
+ \bool_if:NTF \l_regex_in_csname_bool
{
- \tl_put_right:Nn \exp_not:N \l_regex_class_tl
- { \exp_not:c { c_regex_#1_tl } }
+ \regex_build_Z:
+ \regex_toks_put_right:Nx \l_regex_right_state_int
+ { \regex_action_success: }
+ \int_incr:N \l_regex_capturing_group_int
+ \use:x
+ {
+ \group_end:
+ \exp_not:n { \use:c { regex_ \l_regex_build_or_class_tl _one:n } }
+ { \regex_item_cs:n { \regex_set_aux:N ? } }
+ }
}
+ { \exp_after:wN \regex_build_raw:N \c_rbrace_str }
}
\cs_new_protected_nopar:Npn \regex_build_quantifier:w #1#2
{
@@ -711,12 +846,8 @@
{
\tl_if_empty:nTF {#2}
{ \regex_build_quantifier_aux:nnNN { {#1} } { n* } }
- {
- \int_compare:nNnT {#1} > {#2}
- { \msg_error:nn { regex } {4} }
- \regex_build_quantifier_aux:nnNN { {#1} {#2} } { nn }
- }
- }
+ { \regex_build_quantifier_aux:nnNN { {#1} {#2} } { nn } }
+ } %^^A todo: catch the case m>n
{
\regex_build_quantifier_end:n { }
\use:x
@@ -833,7 +964,7 @@
\cs_new_protected_nopar:cpn { regex_build_): }
{
\seq_if_empty:NTF \l_regex_capturing_group_seq
- { \msg_error:nn { regex } { 22 } }
+ { \msg_kernel_error:nn { regex } { extra-rparen } }
{
\regex_build_close_aux:
\regex_build_group_quantifier:
@@ -878,7 +1009,8 @@
{
\cs_if_exist_use:cF { regex_build_special_group_\token_to_str:N #2 : }
{
- \msg_error:nn { regex } { -998 }
+ \msg_kernel_error:nnx { regex } { unsupported }
+ { (? \token_to_str:N #2 } %)
\regex_build_control:N ( % )
\regex_build_raw:N ?
#1 #2
@@ -990,7 +1122,7 @@
{
\tex_toks:D \l_regex_tmpa_int
{
- \s_regex_stop
+ \s_stop
#1 { \int_eval:n { \l_regex_right_state_int - \l_regex_tmpa_int } }
}
}
@@ -1052,16 +1184,19 @@
\regex_toks_put_right:Nx \l_regex_right_state_int
{
\regex_action_free:n
- { \int_eval:n { \l_regex_max_state_int - \l_regex_left_state_int } }
+ {
+ \int_eval:n %^^A todo: document why that value.
+ { \l_regex_max_state_int - \c_one - \l_regex_left_state_int }
+ }
}
\int_set_eq:NN \l_regex_tmpa_int \l_regex_left_state_int
- \int_set:Nn \l_regex_tmpb_int { \l_regex_max_state_int + \c_one }
+ \int_set_eq:NN \l_regex_tmpb_int \l_regex_max_state_int
\int_set:Nn \l_regex_max_state_int
{
- \l_regex_left_state_int - \c_one
- + #1 * ( \l_regex_max_state_int - \l_regex_left_state_int + \c_one )
+ \l_regex_left_state_int
+ + #1 * ( \l_regex_max_state_int - \l_regex_left_state_int )
}
- \int_until_do:nNnn \l_regex_tmpb_int > \l_regex_max_state_int
+ \int_while_do:nNnn \l_regex_tmpb_int < \l_regex_max_state_int
{
\tex_toks:D \l_regex_tmpb_int = \tex_toks:D \l_regex_tmpa_int
\int_incr:N \l_regex_tmpa_int
@@ -1114,14 +1249,13 @@
\cs_new_protected_nopar:Npn \regex_state_use_with_submatches:
{ \regex_state_use_aux:n { } }
\cs_new_protected_nopar:Npn \regex_state_use:
- { \regex_state_use_aux:n { \exp_after:wN \regex_state_use_aux_ii:w } }
-\cs_new_nopar:Npn \regex_state_use_aux_ii:w #1 \s_regex_stop { }
+ { \regex_state_use_aux:n { \exp_after:wN \use_none_delimit_by_s_stop:w } }
\cs_new_protected_nopar:Npn \regex_state_use_aux:n #1
{
\if_num:w \tex_dimen:D \l_regex_current_state_int
- < \l_regex_unique_step_int
+ < \l_regex_unique_id_int
\tex_dimen:D \l_regex_current_state_int
- = \l_regex_unique_step_int sp \scan_stop:
+ = \l_regex_unique_id_int sp \scan_stop:
#1 \tex_the:D \tex_toks:D \exp_after:wN \l_regex_current_state_int
\fi:
\scan_stop:
@@ -1144,73 +1278,105 @@
\tex_the:D \tex_toks:D #1
}
}
-\cs_new_protected:Npn \regex_store_submatches_aux:w #1 \s_regex_stop
+\cs_new_protected:Npn \regex_store_submatches_aux:w #1 \s_stop
{
\regex_store_submatches_aux_ii:Nnnw
#1
\regex_state_submatches:nn \c_minus_one \q_prop
- \s_regex_stop
+ \s_stop
}
\cs_new_protected:Npn \regex_store_submatches_aux_ii:Nnnw
- \regex_state_submatches:nn #1 #2 #3 \s_regex_stop
+ \regex_state_submatches:nn #1 #2 #3 \s_stop
{
\exp_after:wN \c_zero
\exp_after:wN \regex_state_submatches:nn \exp_after:wN
{
\int_value:w \int_eval:w
- \l_regex_unique_step_int + \c_one
+ \l_regex_unique_id_int + \c_one
\exp_after:wN
}
\exp_after:wN { \l_regex_current_submatches_prop }
\regex_state_submatches:nn {#1} {#2}
- \s_regex_stop
+ \s_stop
}
\cs_new_protected:Npn \regex_state_submatches:nn #1#2
{
- \if_num:w #1 = \l_regex_unique_step_int
+ \if_num:w #1 = \l_regex_unique_id_int
\tl_set:Nn \l_regex_current_submatches_prop { #2 }
\fi:
}
\cs_new_protected:Npn \regex_match:n #1
{
- \tl_set:Nx \l_regex_query_other_str { \tl_to_other_str:n {#1} }
+ \tl_set_analysis:Nn \l_regex_tmpa_tl {#1}
+
+ \int_zero:N \l_regex_nesting_int
+ \int_set_eq:NN \l_regex_current_step_int \l_regex_max_state_int
+ \regex_query_set:nnn { } { -1 } { -2 }
+ \int_set_eq:NN \l_regex_min_step_int \l_regex_current_step_int
+ \exp_after:wN \regex_query_set_loop:ww
+ \l_regex_tmpa_tl
+ \q_mark { ? \prg_map_break: } ,
+ \prg_break_point:n { }
+ \int_set_eq:NN \l_regex_max_step_int \l_regex_current_step_int
+ \regex_query_set:nnn { } { -1 } { -2 }
+
\regex_match_initial_setup:
\regex_match_once:
}
-\cs_new_protected_nopar:Npn \regex_match_once:
+\cs_new:Npn \regex_query_set_loop:ww #1 \q_mark #2#3 ,
{
- \regex_match_setup:
- \exp_after:wN \regex_match_once_aux: \l_regex_query_other_str
- \q_recursion_tail \q_recursion_stop
- \bool_if:NT \l_regex_success_bool { \l_regex_every_match_tl }
+ \use_none:n #2
+ \regex_query_set:nnn {#1} {"#2} {#3}
+ \if_case:w "#2 \exp_stop_f:
+ \or: \int_incr:N \l_regex_nesting_int
+ \or: \int_decr:N \l_regex_nesting_int
+ \fi:
+ \regex_query_set_loop:ww
}
-\cs_new_protected_nopar:Npn \regex_match_once_aux:
+\cs_new_protected:Npn \regex_query_set:nnn #1#2#3
{
- \int_compare:nNnTF \l_regex_start_step_int = \c_zero
- {
- \int_set_eq:NN \l_regex_current_char_int \c_minus_one
- \regex_match_loop:N
- }
- {
- \str_skip_do:nn
- { \l_regex_start_step_int - \c_one }
- { \regex_match_once_aux_ii:N }
- }
+ \tex_muskip:D \l_regex_current_step_int
+ = \etex_gluetomu:D
+ #3 sp
+ plus #2 sp
+ minus \l_regex_nesting_int sp
+ \scan_stop:
+ \tex_toks:D \l_regex_current_step_int {#1}
+ \int_incr:N \l_regex_current_step_int
}
-\cs_new_protected_nopar:Npn \regex_match_once_aux_ii:N #1
+\cs_new_protected_nopar:Npn \regex_query_get:
{
- \int_set:Nn \l_regex_current_char_int { `#1 }
- \regex_match_loop:N
+ \tl_set:Nx \l_regex_current_token_tl
+ { \tex_the:D \tex_toks:D \l_regex_current_step_int }
+ \l_regex_current_char_int
+ = \etex_mutoglue:D \tex_muskip:D \l_regex_current_step_int
+ \l_regex_current_catcode_int = \etex_gluestretch:D
+ \etex_mutoglue:D \tex_muskip:D \l_regex_current_step_int
+ }
+\cs_new_protected_nopar:Npn \regex_match_once:
+ {
+ \regex_match_setup:
+ \regex_query_get:
+ \regex_match_loop:
+ \prg_break_point:n { }
+ \bool_if:NT \l_regex_success_match_bool
+ {
+ \bool_gset_true:N \g_regex_success_bool
+ \l_regex_every_match_tl
+ }
}
\cs_new_protected_nopar:Npn \regex_match_initial_setup:
{
- \prg_stepwise_inline:nnnn {1} {1} { \l_regex_max_state_int }
+ \prg_stepwise_inline:nnnn {0} {1} { \l_regex_max_state_int - \c_one }
{ \tex_dimen:D ##1 \c_minus_one sp \scan_stop: }
- \int_set_eq:NN \l_regex_unique_step_int \c_minus_one
- \int_set_eq:NN \l_regex_start_step_int \c_minus_one
- \int_set_eq:NN \l_regex_current_step_int \c_zero
- \int_set_eq:NN \l_regex_success_step_int \c_zero
+ \int_set_eq:NN \l_regex_unique_id_int \c_minus_one
+ \int_set:Nn \l_regex_start_step_int
+ { \l_regex_min_step_int - \c_one }
+ \int_set_eq:NN \l_regex_current_step_int \l_regex_min_step_int
+ \int_set_eq:NN \l_regex_success_step_int \l_regex_min_step_int
+ \int_set_eq:NN \l_regex_submatch_int \l_regex_max_state_int
\bool_set_false:N \l_regex_success_empty_bool
+ \bool_gset_false:N \g_regex_success_bool
}
\cs_new_protected_nopar:Npn \regex_match_setup:
{
@@ -1219,33 +1385,44 @@
{ \cs_set_eq:NN \regex_last_match_empty:F \regex_last_match_empty_yes:F }
{ \cs_set_eq:NN \regex_last_match_empty:F \regex_last_match_empty_no:F }
\int_set_eq:NN \l_regex_start_step_int \l_regex_success_step_int
- \int_set_eq:NN \l_regex_current_step_int \l_regex_start_step_int
- \int_decr:N \l_regex_current_step_int
- \bool_set_false:N \l_regex_success_bool
+ \int_set:Nn \l_regex_current_step_int
+ { \l_regex_start_step_int - \c_one }
+ \bool_set_false:N \l_regex_success_match_bool
\int_zero:N \l_regex_max_index_int
- \regex_store_state:n {1}
+ \regex_store_state:n {0} %^^A _state_int!
}
-\cs_new_protected_nopar:Npn \regex_match_loop:N #1
+\cs_new_protected_nopar:Npn \regex_match_loop:
{
- \regex_match_loop_setup:N #1
+ \int_incr:N \l_regex_current_step_int
+ \int_incr:N \l_regex_unique_id_int
+ \bool_set_false:N \l_regex_fresh_thread_bool
+ \int_set_eq:NN \l_regex_last_char_int \l_regex_current_char_int
+ \regex_query_get:
+ \regex_match_loop_case_hook:
\cs_set_nopar:Npx \regex_tmp:w
{
\int_zero:N \l_regex_max_index_int
- \prg_stepwise_function:nnnN
- {1} {1} { \l_regex_max_index_int }
- \regex_match_one_index:n
+ \regex_match_one_index:w 1 ; \prg_break_point:n { }
+ \exp_not:N \prg_break_point:n { }
}
- \regex_tmp:w \prg_break_point:n { }
+ \regex_tmp:w
+ \if_num:w \l_regex_current_char_int < \c_minus_one
+ \exp_after:wN \prg_map_break:
+ \fi:
\if_num:w \l_regex_max_index_int = \c_zero
- \exp_after:wN \use_none_delimit_by_q_recursion_stop:w
+ \exp_after:wN \prg_map_break:
\fi:
- \quark_if_recursion_tail_stop:N #1
- \regex_match_loop:N
+ \regex_match_loop:
}
-\cs_new_nopar:Npn \regex_match_one_index:n #1
+\cs_new_nopar:Npn \regex_match_one_index:w #1;
{
+ \if_num:w #1 > \l_regex_max_index_int
+ \exp_after:wN \prg_map_break:
+ \fi:
\regex_match_one_index_aux:n
{ \int_value:w \tex_skip:D #1 }
+ \exp_after:wN \regex_match_one_index:w
+ \int_use:N \int_eval:w #1 + \c_one ;
}
\cs_new_protected_nopar:Npn \regex_match_one_index_aux:n #1
{
@@ -1253,19 +1430,6 @@
\prop_clear:N \l_regex_current_submatches_prop
\regex_state_use_with_submatches:
}
-\cs_new_protected_nopar:Npn \regex_match_loop_setup:N #1
- {
- \int_incr:N \l_regex_current_step_int
- \int_incr:N \l_regex_unique_step_int
- \bool_set_false:N \l_regex_fresh_thread_bool
- \int_set_eq:NN \l_regex_last_char_int \l_regex_current_char_int
- \if_meaning:w #1 \q_recursion_tail
- \int_set_eq:NN \l_regex_current_char_int \c_minus_one
- \else:
- \int_set:Nn \l_regex_current_char_int {`#1}
- \fi:
- \regex_match_loop_case_hook:
- }
\cs_new_protected_nopar:Npn \regex_match_loop_case_hook: { }
\cs_new_protected_nopar:Npn \regex_match_loop_caseless_hook:
{
@@ -1293,18 +1457,17 @@
}
\cs_new_protected_nopar:Npn \regex_action_cost:n #1
{
- \exp_args:Nf \regex_store_state:n %^^A optimize!
+ \exp_args:Nf \regex_store_state:n
{ \int_eval:n { \l_regex_current_state_int + #1 } }
}
\cs_new_protected_nopar:Npn \regex_action_success:
{
\regex_last_match_empty:F
{
- \bool_set_true:N \l_regex_success_bool
+ \bool_set_true:N \l_regex_success_match_bool
\bool_set_eq:NN \l_regex_success_empty_bool
\l_regex_fresh_thread_bool
- \int_set_eq:NN \l_regex_success_step_int
- \l_regex_current_step_int
+ \int_set_eq:NN \l_regex_success_step_int \l_regex_current_step_int
\prop_set_eq:NN \l_regex_success_submatches_prop
\l_regex_current_submatches_prop
\prg_map_break:
@@ -1328,504 +1491,668 @@
\prop_put:Nno \l_regex_current_submatches_prop {#1}
{ \int_use:N \l_regex_current_step_int }
}
-\cs_new_protected_nopar:Npn \regex_extract:
+\cs_new_protected:Npn \regex_submatch_nesting_aux:n #1
{
- \seq_gclear:N \g_regex_submatches_seq
- \prg_stepwise_inline:nnnn
- {0} {1} { \l_regex_capturing_group_int }
- {
- \regex_extract_aux:nTF { ##1 }
- {
- \seq_gput_right:Nx \g_regex_submatches_seq
- { \regex_query_substr:NN \l_regex_tmpa_tl \l_regex_tmpb_tl }
- }
- { \seq_gput_right:Nn \g_regex_submatches_seq { } }
- }
+ + \etex_glueshrink:D \etex_mutoglue:D \etex_muexpr:D
+ \tex_muskip:D \etex_gluestretch:D \tex_skip:D #1
+ - \tex_muskip:D \tex_skip:D #1
+ \scan_stop:
}
-\cs_new_protected_nopar:Npn \regex_extract_aux:nTF #1#2#3
+\cs_new:Npn \regex_nesting:n #1 { } %^^A move. Rename?
+\tl_new:N \l_regex_nesting_tl
+\cs_new_protected:Npn \regex_replacement:n #1
{
- \prop_get:NnNTF \l_regex_success_submatches_prop
- { #1 < } \l_regex_tmpa_tl
+ \int_zero:N \l_regex_replacement_int
+ \int_zero:N \l_regex_nesting_int
+ \tl_clear:N \l_regex_nesting_tl
+ \str_escape_use:NNNn
+ \regex_replacement_unescaped:N
+ \regex_replacement_escaped:N
+ \regex_replacement_raw:N
+ {#1}
+ \prg_do_nothing: \prg_do_nothing:
+ \cs_set_nopar:Npx \regex_nesting:n ##1
+ {
+ + \int_use:N \l_regex_nesting_int
+ \l_regex_nesting_tl
+ - \regex_submatch_nesting_aux:n {##1}
+ }
+ \use:x
{
- \prop_get:NnNTF \l_regex_success_submatches_prop
- { #1 > } \l_regex_tmpb_tl
- {#2}
- {#3}
+ \exp_not:n { \cs_set_nopar:Npn \regex_replacement_tl:n ##1 }
+ { \str_aux_toks_range:nn \c_zero \l_regex_replacement_int }
}
- {#3}
+ % ^^A rename! Careful with \cP\#.
}
-\cs_new_protected_nopar:Npn \regex_set:Nn
- { \regex_set_aux:NNn \tl_set_eq:NN }
-\cs_new_protected_nopar:Npn \regex_gset:Nn
- { \regex_set_aux:NNn \tl_gset_eq:NN }
-\cs_new_protected:Npn \regex_set_aux:NNn #1#2#3
+\cs_new_protected:Npn \regex_replacement_raw:N #1
+ { \regex_replacement_put:n {#1} }
+\cs_new_protected:Npn \regex_replacement_put:n #1
{
- \group_begin:
- \regex_build:n {#3}
- \tl_gset:Nx \g_regex_tmpa_tl
- {
- \exp_not:N \regex_nfa:Nw \exp_not:N #2
- \l_regex_max_state_int
- = \int_use:N \l_regex_max_state_int
- \l_regex_capturing_group_int
- = \int_use:N \l_regex_capturing_group_int
- \token_if_eq_meaning:NNT
- \regex_match_loop_case_hook:
- \regex_match_loop_caseless_hook:
- {
- \cs_set_eq:NN \regex_match_loop_case_hook:
- \regex_match_loop_caseless_hook:
- }
- \prg_stepwise_function:nnnN
- {1} {1} {\l_regex_max_state_int}
- \regex_set_aux:n
- \s_regex_stop
- }
- \group_end:
- #1 #2 \g_regex_tmpa_tl
+ \tex_toks:D \l_regex_replacement_int {#1}
+ \int_incr:N \l_regex_replacement_int
}
-\cs_new_nopar:Npn \regex_set_aux:n #1
- { \tex_toks:D #1 { \tex_the:D \tex_toks:D #1 } }
-\cs_new:Npn \regex_nfa:Nw #1 #2 \s_regex_stop
- { \msg_expandable_kernel_error:nnn { regex } { nfa-misused } {#1} }
-\cs_new_protected_nopar:Npn \regex_const:Nn
- { \regex_set_aux:NNn \cs_new_eq:NN }
-\prg_new_protected_conditional:Npnn \regex_check_nfa:N #1 { T , TF }
- { \exp_after:wN \regex_check_nfa_aux:Nw #1 \q_stop }
-\cs_new:Npn \regex_check_nfa_aux:Nw #1 #2 \q_stop
+\cs_new_protected:Npn \regex_replacement_unescaped:N #1
{
- \if_meaning:w \regex_nfa:Nw #1
- \prg_return_true:
+ \if_charcode:w \c_rbrace_str #1
+ \if_num:w \l_regex_replacement_csnames_int > \c_zero
+ \regex_replacement_put:n \cs_end:
+ \else:
+ \regex_replacement_put:n #1
+ \fi:
\else:
- \msg_error:nnx { regex } { not-nfa } { \token_to_str:N #1 }
- \prg_return_false:
+ \regex_replacement_put:n #1
\fi:
}
-\cs_new_protected_nopar:Npn \regex_use:N #1
- { \exp_after:wN \use_none:nn #1 }
-\cs_new_protected:Npn \regex_user_aux:n #1
+\cs_new_protected:Npn \regex_replacement_escaped:N #1
{
- \group_begin:
- \tl_clear:N \l_regex_every_match_tl
- #1
- \group_end:
- }
-\cs_new_protected_nopar:Npn \regex_return_after_group:
- {
- \if_meaning:w \c_true_bool \l_regex_success_bool
- \group_insert_after:N \prg_return_true:
+ \if_charcode:w c #1
+ \exp_after:wN \regex_replacement_c:w
\else:
- \group_insert_after:N \prg_return_false:
+ \if_charcode:w g #1
+ \exp_after:wN \exp_after:wN \exp_after:wN \regex_replacement_g:w
+ \else:
+ \if_num:w \c_one < 1#1 \exp_stop_f:
+ \regex_replacement_put_submatch:n {#1}
+ \else:
+ \regex_replacement_put:n #1
+ \fi:
+ \fi:
\fi:
}
-\cs_new_protected_nopar:Npn \regex_extract_after_group:N #1
+\cs_new_protected:Npn \regex_replacement_put_submatch:n #1
{
- \if_meaning:w \c_true_bool \l_regex_success_bool
- \regex_extract:
- \group_insert_after:N \seq_set_eq:NN
- \group_insert_after:N #1
- \group_insert_after:N \g_regex_submatches_seq
+ \regex_replacement_put:n
+ { \regex_query_submatch:nn {#1} {##1} }
+ \if_num:w \l_regex_replacement_csnames_int = \c_zero
+ \tl_put_right:Nn \l_regex_nesting_tl
+ {
+ \exp_not:N \if_num:w #1 < \l_regex_capturing_group_int
+ \regex_submatch_nesting_aux:n { \int_eval:w #1+##1 \int_eval_end: }
+ \exp_not:N \fi:
+ }
\fi:
}
-\cs_new_protected_nopar:Npn \regex_count_after_group:N #1
+\cs_new_protected:Npn \regex_replacement_error:NNN #1#2#3
{
- \group_insert_after:N \int_set_eq:NN
- \group_insert_after:N #1
- \group_insert_after:N \g_regex_match_count_int
+ \msg_kernel_error:nn { regex } { replacement-#1 }
+ #2 #3
}
-\prg_new_protected_conditional:Npnn \regex_match:nn #1#2 { T , F , TF }
+\cs_new_protected:Npn \regex_replacement_g:w #1#2
{
- \regex_user_aux:n
+ \str_if_eq:xxTF
+ { \exp_not:n { #1#2 } }
+ { \regex_replacement_unescaped:N \c_lbrace_str }
{
- \regex_disable_submatches:
- \regex_build:n {#1}
- \regex_match:n {#2}
- \regex_return_after_group:
+ \int_zero:N \l_regex_tmpa_int
+ \regex_replacement_g_digits:NN
}
+ { \regex_replacement_error:NNN g #1 #2 }
}
-\prg_new_protected_conditional:Npnn \regex_match:Nn #1#2 { T , F , TF }
+\cs_new_protected:Npn \regex_replacement_g_digits:NN #1#2
{
- \regex_check_nfa:NTF #1
+ \token_if_eq_meaning:NNTF #1 \regex_replacement_unescaped:N
{
- \regex_user_aux:n
- {
- \regex_disable_submatches:
- \regex_use:N #1
- \regex_match:n {#2}
- \regex_return_after_group:
- }
+ \if_num:w \c_one < 1#2 \exp_stop_f:
+ \int_set:Nn \l_regex_tmpa_int { \c_ten * \l_regex_tmpa_int + #2 }
+ \exp_after:wN \use_i:nnn
+ \exp_after:wN \regex_replacement_g_digits:NN
+ \else:
+ \if_charcode:w \c_rbrace_str #2
+ \exp_args:No \regex_replacement_put_submatch:n
+ { \int_use:N \l_regex_tmpa_int }
+ \exp_after:wN \exp_after:wN \exp_after:wN \use_none:nn
+ \else:
+ \exp_after:wN \exp_after:wN
+ \exp_after:wN \regex_replacement_error:NNN
+ \exp_after:wN \exp_after:wN \exp_after:wN g
+ \fi:
+ \fi:
}
- { \prg_return_false: }
+ { \regex_replacement_error:NNN g }
+ #1 #2
}
-\cs_new_protected:Npn \regex_count:nnN #1#2#3
+\cs_new_protected_nopar:Npn \regex_replacement_c:w #1#2
{
- \regex_user_aux:n
+ \token_if_eq_meaning:NNTF #1 \regex_replacement_unescaped:N
{
- \regex_disable_submatches:
- \int_gzero:N \g_regex_match_count_int
- \tl_set:Nn \l_regex_every_match_tl
- {
- \int_gincr:N \g_regex_match_count_int
- \regex_match_once:
- }
- \regex_build:n {#1}
- \regex_match:n {#2}
- \regex_count_after_group:N #3
+ \cs_if_exist_use:cF { regex_replacement_c_#2:w }
+ { \regex_replacement_error:NNN c #1#2 }
}
+ { \regex_replacement_error:NNN c #1#2 }
}
-\cs_new_protected:Npn \regex_count:NnN #1#2#3
+\cs_new_protected_nopar:cpn { regex_replacement_c_ \c_lbrace_str :w }
{
- \regex_check_nfa:NT #1
- {
- \regex_user_aux:n
- {
- \regex_disable_submatches:
- \int_gzero:N \g_regex_match_count_int
- \tl_set:Nn \l_regex_every_match_tl
- {
- \int_gincr:N \g_regex_match_count_int
- \regex_match_once:
- }
- \regex_use:N #1
- \regex_match:n {#2}
- \regex_count_after_group:N #3
- }
- }
+ \int_incr:N \l_regex_replacement_csnames_int
+ \regex_replacement_put:n
+ { \exp_not:n { \exp_after:wN \regex_replacement_exp_not:N \cs:w } }
}
-\cs_new_protected:Npn \regex_extract_once:nnN #1#2#3
- {
- \regex_user_aux:n
- {
- \regex_build:n {#1}
- \regex_match:n {#2}
- \regex_extract_after_group:N #3
- }
+\cs_new_nopar:Npn \regex_replacement_exp_not:N #1 { \exp_not:n {#1} }
+\group_begin:
+
+ \char_set_catcode_math_superscript:N \^^@
+ \cs_new_protected_nopar:Npn \regex_replacement_c_U:w
+ { \regex_replacement_char:nNN { ^^@ } }
+
+ \char_set_catcode_alignment:N \^^@
+ \cs_new_protected_nopar:Npn \regex_replacement_c_T:w
+ { \regex_replacement_char:nNN { ^^@ } }
+
+ \cs_new_protected_nopar:Npn \regex_replacement_c_S:w #1#2
+ {
+ \int_compare:nNnTF { `#2 } = \c_zero
+ { \regex_replacement_error:NNN c #1#2 }
+ {
+ \char_set_lccode:nn {32} { `#2 }
+ \tl_to_lowercase:n { \regex_replacement_put:n {~} }
+ }
+ }
+
+ \char_set_catcode_parameter:N \^^@
+ \cs_new_protected_nopar:Npn \regex_replacement_c_P:w
+ { \regex_replacement_char:nNN { ^^@^^@ } }
+
+ \char_set_catcode_other:N \^^@
+ \cs_new_protected_nopar:Npn \regex_replacement_c_O:w
+ { \regex_replacement_char:nNN { ^^@ } }
+
+ \char_set_catcode_math_toggle:N \^^@
+ \cs_new_protected_nopar:Npn \regex_replacement_c_M:w
+ { \regex_replacement_char:nNN { ^^@ } }
+
+ \char_set_catcode_letter:N \^^@
+ \cs_new_protected_nopar:Npn \regex_replacement_c_L:w
+ { \regex_replacement_char:nNN { ^^@ } }
+
+ \char_set_catcode_group_end:N \^^@
+ \cs_new_protected_nopar:Npn \regex_replacement_c_E:w
+ {
+ \int_decr:N \l_regex_nesting_int
+ \regex_replacement_char:nNN { \if_false: { \fi: ^^@ }
+ }
+
+ \char_set_catcode_math_subscript:N \^^@
+ \cs_new_protected_nopar:Npn \regex_replacement_c_D:w
+ { \regex_replacement_char:nNN { ^^@ } }
+
+ \char_set_catcode_group_begin:N \^^@
+ \cs_new_protected_nopar:Npn \regex_replacement_c_B:w
+ {
+ \int_incr:N \l_regex_nesting_int
+ \regex_replacement_char:nNN { \exp_after:wN ^^@ \if_false: } \fi: }
+ }
+
+ \char_set_catcode_active:N \^^@
+ \cs_new_protected_nopar:Npn \regex_replacement_c_A:w
+ { \regex_replacement_char:nNN { \exp_not:N ^^@ } }
+
+\group_end:
+
+\cs_new_protected_nopar:Npn \regex_replacement_char:nNN #1#2#3
+ {
+ \char_set_lccode:nn \c_zero { `#3 }
+ \tl_to_lowercase:n
+ { \regex_replacement_put:n { \exp_not:n {#1} } }
}
-\prg_new_protected_conditional:Npnn \regex_extract_once:nnN #1#2#3
- { T , F , TF }
+\cs_new_protected_nopar:Npn \regex_set:Nn
+ { \regex_set_aux:NNn \tl_set:Nn }
+\cs_new_protected_nopar:Npn \regex_gset:Nn
+ { \regex_set_aux:NNn \tl_gset:Nn }
+\cs_new_protected_nopar:Npn \regex_const:Nn
+ { \regex_set_aux:NNn \tl_const:Nn }
+\cs_new_protected:Npn \regex_set_aux:NNn #1#2#3
{
- \regex_user_aux:n
- {
- \regex_build:n {#1}
- \regex_match:n {#2}
- \regex_extract_after_group:N #3
- \regex_return_after_group:
- }
+ \group_begin:
+ \regex_build:n {#3}
+ \cs_set_nopar:Npx \regex_tmp:w
+ { #1 \exp_not:N #2 { \regex_set_aux:N #2 } }
+ \exp_after:wN
+ \group_end:
+ \regex_tmp:w
}
-\cs_new_protected:Npn \regex_extract_once:NnN #1#2#3
+\cs_new:Npn \regex_set_aux:N #1
{
- \regex_check_nfa:NT #1
+ \exp_not:n { \regex_nfa:Nw #1 }
+ \l_regex_max_state_int
+ = \int_use:N \l_regex_max_state_int
+ \l_regex_capturing_group_int
+ = \int_use:N \l_regex_capturing_group_int
+ \token_if_eq_meaning:NNT
+ \regex_match_loop_case_hook:
+ \regex_match_loop_caseless_hook:
{
- \regex_user_aux:n
- {
- \regex_use:N #1
- \regex_match:n {#2}
- \regex_extract_after_group:N #3
- }
+ \cs_set_eq:NN \regex_match_loop_case_hook:
+ \regex_match_loop_caseless_hook:
}
+ \prg_stepwise_function:nnnN
+ {0} {1} {\l_regex_max_state_int - \c_one }
+ \regex_set_aux:n
+ \s_stop
}
-\prg_new_protected_conditional:Npnn \regex_extract_once:NnN #1#2#3
- { T , F , TF }
+\cs_new_nopar:Npn \regex_set_aux:n #1
+ { \tex_toks:D #1 { \tex_the:D \tex_toks:D #1 } }
+\cs_new:Npn \regex_nfa:Nw #1
{
- \regex_check_nfa:NTF #1
- {
- \regex_user_aux:n
- {
- \regex_use:N #1
- \regex_match:n {#2}
- \regex_extract_after_group:N #3
- \regex_return_after_group:
- }
- }
- { \prg_return_false: }
+ \msg_expandable_kernel_error:nnn { regex } { nfa-misused } {#1}
+ \use_none_delimit_by_s_stop:w
}
-\cs_new_protected:Npn \regex_extract_all:nnN #1#2#3
+\cs_new_protected_nopar:Npn \regex_use:N
+ { \exp_last_unbraced:No \use_none:nn }
+\cs_new_nopar:Npn \regex_to_str:N #1
{
- \regex_user_aux:n
- {
- \seq_gclear:N \g_regex_extract_all_seq
- \tl_set:Nn \l_regex_every_match_tl { \regex_extract_all_aux: }
- \regex_build:n {#1}
- \regex_match:n {#2}
- \regex_extract_all_after_group:N #3
- }
+ \exp_after:wN \regex_to_str_aux:Nw #1
+ \tex_toks:D \q_stop \prg_map_break: \tex_toks:D
+ \prg_break_point:n { }
}
-\cs_new_protected:Npn \regex_extract_all:NnN #1#2#3
+\cs_new_nopar:Npn \regex_to_str_aux:Nw #1#2\tex_toks:D
{
- \regex_check_nfa:NT #1
- {
- \regex_user_aux:n
- {
- \seq_gclear:N \g_regex_extract_all_seq
- \tl_set:Nn \l_regex_every_match_tl { \regex_extract_all_aux: }
- \regex_use:N #1
- \regex_match:n {#2}
- \regex_extract_all_after_group:N #3
- }
- }
+ \use_none_delimit_by_q_stop:w #2 \q_stop
+ \tl_to_str:n {#1#2} \iow_newline:
+ \regex_to_str_aux:Nw \tex_toks:D
}
-\cs_new_protected_nopar:Npn \regex_extract_all_aux:
+\cs_new_protected_nopar:Npn \regex_aux_return:
{
- \regex_extract:
- \seq_gconcat:NNN \g_regex_extract_all_seq
- \g_regex_extract_all_seq \g_regex_submatches_seq
- \regex_match_once:
+ \if_meaning:w \c_true_bool \g_regex_success_bool
+ \prg_return_true:
+ \else:
+ \prg_return_false:
+ \fi:
}
-\cs_new_protected_nopar:Npn \regex_extract_all_after_group:N #1
+\cs_new_protected:Npn \regex_aux_build_match:nn #1#2
{
- \seq_if_empty:NF \g_regex_extract_all_seq
- {
- \regex_extract:
- \group_insert_after:N \seq_set_eq:NN
- \group_insert_after:N #1
- \group_insert_after:N \g_regex_extract_all_seq
- }
+ \regex_build:n {#1}
+ \regex_match:n {#2}
}
-\cs_new_protected:Npn \regex_split:nnN #1#2#3
+\cs_new_protected:Npn \regex_aux_use_match:Nn #1#2
{
- \regex_user_aux:n
+ \exp_args:No \tl_if_head_eq_meaning:nNTF {#1} \regex_nfa:Nw
{
- \seq_gclear:N \g_regex_split_seq
- \tl_set:Nn \l_regex_every_match_tl { \regex_split_aux: }
- \regex_build:n {#1}
+ \regex_use:N #1
\regex_match:n {#2}
- \regex_split_after_group:N #3
}
- }
-\cs_new_protected:Npn \regex_split:NnN #1#2#3
- {
- \regex_check_nfa:NT #1
- {
- \regex_user_aux:n
- {
- \seq_gclear:N \g_regex_split_seq
- \tl_set:Nn \l_regex_every_match_tl { \regex_split_aux: }
- \regex_use:N #1
- \regex_match:n {#2}
- \regex_split_after_group:N #3
- }
- }
- }
-\cs_new_protected_nopar:Npn \regex_split_aux:
- {
- \int_compare:nNnF \l_regex_start_step_int = \l_regex_success_step_int
{
- \regex_extract:
- \seq_pop:NN \g_regex_submatches_seq \l_regex_tmpa_tl
- \regex_extract_aux:nTF {0}
- {
- \seq_gput_left:Nx \g_regex_submatches_seq
- {
- \regex_query_substr:NN
- \l_regex_start_step_int \l_regex_tmpa_tl
- }
- }
- { \msg_error:nn { regex } { internal } }
- \seq_gconcat:NNN \g_regex_split_seq
- \g_regex_split_seq \g_regex_submatches_seq
+ \msg_kernel_error:nnx { regex } { not-nfa } { \token_to_str:N #1 }
+ \bool_gset_false:N \g_regex_success_bool
}
- \regex_match_once:
}
-\cs_new_protected_nopar:Npn \regex_split_after_group:N #1
+\cs_set_protected:Npn \regex_tmp:w #1#2#3
{
- \int_compare:nNnTF \l_regex_start_step_int = \l_regex_current_step_int
+ \cs_new_protected_nopar:Npn #1
+ { #3 \regex_aux_build_match:nn }
+ \cs_new_protected_nopar:Npn #2
+ { #3 \regex_aux_use_match:Nn }
+ \prg_new_protected_conditional:Npnn #1 ##1##2##3 { T , F , TF }
{
- \bool_if:NF \l_regex_success_empty_bool
- { \seq_gput_right:Nn \g_regex_split_seq { } }
+ #3 \regex_aux_build_match:nn {##1} {##2} ##3
+ \regex_aux_return:
}
+ \prg_new_protected_conditional:Npnn #2 ##1##2##3 { T , F , TF }
{
- \seq_gput_right:Nx \g_regex_split_seq
- {
- \regex_query_substr:NN
- \l_regex_start_step_int \l_regex_current_step_int
- }
+ #3 \regex_aux_use_match:Nn {##1} {##2} ##3
+ \regex_aux_return:
}
- \group_insert_after:N \seq_set_eq:NN
- \group_insert_after:N #1
- \group_insert_after:N \g_regex_split_seq
}
-\cs_new_protected:Npn \regex_replacement:n #1
+\tl_map_inline:nn
{
- \str_aux_escape:NNNn
- \prg_do_nothing:
- \regex_replacement_escaped:N
- \prg_do_nothing:
- {#1}
- \tl_set_eq:NN \l_regex_replacement_tl \g_str_result_tl
- \tl_set:Nx \l_regex_replacement_tl
- { \l_regex_replacement_tl \prg_do_nothing: }
+ { extract_once } { extract_all }
+ { replace_once } { replace_all }
+ { split }
}
-\cs_new_nopar:Npn \regex_replacement_escaped:N #1
{
- \regex_token_if_other_digit:NTF #1
- {
- \exp_not:n
- { \exp_not:n { \seq_item:Nn \g_regex_submatches_seq } } {#1}
- }
- {
- \token_if_eq_charcode:NNTF g #1
- { \exp_not:N \regex_replacement_submatch:w }
- { #1 }
- }
+ \exp_args:Nccc \regex_tmp:w
+ { regex_#1:nnN } { regex_#1:NnN } { regex_#1_aux:NnnN }
}
-\cs_new_nopar:Npn \regex_replacement_submatch:w #1
+\cs_new_protected_nopar:Npn \regex_extract:
{
- \exp_after:wN \token_if_eq_meaning:NNTF \c_lbrace_str #1
- { \regex_replacement_submatch_aux:nN { } }
- {
- \regex_token_if_other_digit:NTF #1
- { \exp_not:n { \seq_item:Nn \g_regex_submatches_seq } }
- { \msg_expandable_kernel_error:nn { regex } { g-misused } }
- #1
- }
+ \int_set_eq:NN \l_regex_submatch_start_int \l_regex_submatch_int
+ \if_meaning:w \c_true_bool \g_regex_success_bool
+ \prg_replicate:nn \l_regex_capturing_group_int
+ {
+ \tex_skip:D \l_regex_submatch_int \c_zero sp \scan_stop:
+ \int_incr:N \l_regex_submatch_int
+ }
+ \prop_map_inline:Nn \l_regex_success_submatches_prop
+ {
+ \if_num:w ##1 \c_max_int
+ \exp_after:wN \regex_extract_aux_b:wn \int_use:N
+ \else:
+ \exp_after:wN \regex_extract_aux_e:wn \int_use:N
+ \fi:
+ \int_eval:w \l_regex_submatch_start_int + ##1 {##2}
+ }
+ \fi:
}
-\cs_new_nopar:Npn \regex_replacement_submatch_aux:nN #1 #2
+\cs_new_protected_nopar:Npn \regex_extract_aux_b:wn #1 < #2
{
- \regex_token_if_other_digit:NTF #2
- { \regex_replacement_submatch_aux:nN {#1#2} }
- {
- \exp_not:n { \seq_item:Nn \g_regex_submatches_seq } {#1}
- \exp_after:wN \token_if_eq_meaning:NNF \c_rbrace_str #2
- {
- \msg_expandable_kernel_error:nn { regex } { g-misused }
- #2
- }
- }
+ \tex_skip:D #1 = #2 sp
+ plus \etex_gluestretch:D \tex_skip:D #1 \scan_stop:
}
-\cs_new_protected_nopar:Npn \regex_replace_after_group:N #1
+\cs_new_protected_nopar:Npn \regex_extract_aux_e:wn #1 > #2
{
- \group_insert_after:N \tl_set:Nx
- \group_insert_after:N #1
- \group_insert_after:N {
- \group_insert_after:N \tl_to_str:N
- \group_insert_after:N \g_regex_replaced_str
- \group_insert_after:N }
+ \tex_skip:D #1
+ = 1 \tex_skip:D #1 plus #2 sp \scan_stop:
}
-\cs_new_protected:Npn \regex_replace_once_aux:Nn #1#2
+\cs_new_protected:Npn \regex_group_end_extract_seq:N #1
{
- \group_begin:
- \regex_replace_after_group:N #1
- \tl_clear:N \l_regex_every_match_tl
- #2
- \exp_args:No \regex_match:n {#1}
- \regex_extract:
- \regex_extract_aux:nTF {0}
+ \cs_set_eq:NN \seq_item:n \scan_stop:
+ \flag_lower:N \l_regex_group_begin_flag
+ \flag_lower:N \l_regex_group_end_flag
+ \tl_set:Nx \l_regex_tmpa_tl
{
- \cs_set_nopar:Npn \regex_tmp:w
- { \str_skip_do:nn { \l_regex_tmpb_tl } { } }
- \tl_gset:Nx \g_regex_replaced_str
- {
- \regex_query_substr:NN \c_zero \l_regex_tmpa_tl
- \l_regex_replacement_tl
- \exp_after:wN \regex_tmp:w \l_regex_query_other_str
- }
+ \prg_stepwise_function:nnnN
+ \l_regex_max_state_int
+ \c_one
+ { \l_regex_submatch_int - \c_one }
+ \regex_extract_seq_aux:n
}
- { \tl_gset_eq:NN \g_regex_replaced_str \l_regex_query_other_str }
- \group_end:
+ \flag_test:NF \l_regex_group_begin_flag
+ { \flag_test:NF \l_regex_group_end_flag { \use_none:nnn } }
+ \msg_kernel_error:nn { regex } { sequence-unbalanced }
+ \tl_set:Nx \l_regex_tmpa_tl { \l_regex_tmpa_tl }
+ \exp_args:NNNo \group_end:
+ \tl_set:Nn #1 \l_regex_tmpa_tl
}
-\cs_new_protected:Npn \regex_replace_once:nnN #1#2#3
+\cs_new_nopar:Npn \regex_extract_seq_aux:n #1
{
- \regex_replace_once_aux:Nn #3
+ \seq_item:n
{
- \regex_build:n {#1}
- \regex_replacement:n {#2}
+ \exp_after:wN \regex_extract_seq_aux:ww
+ \int_value:w \regex_submatch_nesting_aux:n {#1} ; #1;
}
}
-\cs_new_protected:Npn \regex_replace_once:NnN #1#2#3
+\cs_new_nopar:Npn \regex_extract_seq_aux:ww #1; #2;
+ { %^^A todo: use flags
+ \if_num:w #1 < \c_zero
+ \flag_raise:N \l_regex_group_end_flag
+ \prg_replicate:nn {-#1} { \exp_not:n { { \if_false: } \fi: } }
+ \fi:
+ \regex_query_submatch:w #2;
+ \if_num:w #1 > \c_zero
+ \flag_raise:N \l_regex_group_begin_flag
+ \prg_replicate:nn {#1} { \exp_not:n { \if_false: { \fi: } } }
+ \fi:
+ }
+\prg_new_protected_conditional:Npnn \regex_match:nn #1#2 { T , F , TF }
{
- \regex_replace_once_aux:Nn #3
- {
- \regex_use:N #1
- \regex_replacement:n {#2}
- }
+ \regex_match_aux:n
+ { \regex_aux_build_match:nn {#1} {#2} }
}
-\prg_new_protected_conditional:Npnn \regex_replace_once:nnN #1#2#3 {T,F,TF}
+\prg_new_protected_conditional:Npnn \regex_match:Nn #1#2 { T , F , TF }
{
- \regex_replace_once_aux:Nn #3
- {
- \regex_build:n {#1}
- \regex_replacement:n {#2}
- \regex_return_after_group:
- }
+ \regex_match_aux:n
+ { \regex_aux_use_match:Nn #1 {#2} }
}
-\prg_new_protected_conditional:Npnn \regex_replace_once:NnN #1#2#3 {T,F,TF}
+\cs_new_protected:Npn \regex_match_aux:n #1
{
- \regex_replace_once_aux:Nn #3
- {
- \regex_use:N #1
- \regex_replacement:n {#2}
- \regex_return_after_group:
- }
+ \group_begin:
+ \tl_clear:N \l_regex_every_match_tl
+ \regex_disable_submatches:
+ #1
+ \group_end:
+ \regex_aux_return:
}
-\cs_new_protected:Npn \regex_replace_all_aux:Nn #1#2
+\cs_new_protected_nopar:Npn \regex_count:nnN
+ { \regex_count_aux:NnnN \regex_aux_build_match:nn }
+\cs_new_protected_nopar:Npn \regex_count:NnN
+ { \regex_count_aux:NnnN \regex_aux_use_match:Nn }
+\cs_new_protected:Npn \regex_count_aux:NnnN #1#2#3#4
{
\group_begin:
- \regex_replace_after_group:N #1
- \tl_set:Nn \l_regex_every_match_tl { \regex_replace_all_aux: }
- \tl_gclear:N \g_regex_replaced_str
- #2
- \exp_args:No \regex_match:n {#1}
- \cs_set_nopar:Npn \regex_tmp:w
- { \str_skip_do:nn { \l_regex_start_step_int } { } }
- \tl_gput_right:Nx \g_regex_replaced_str
- { \exp_after:wN \regex_tmp:w \l_regex_query_other_str }
+ \regex_disable_submatches:
+ \int_zero:N \l_regex_match_count_int
+ \tl_set:Nn \l_regex_every_match_tl
+ {
+ \int_incr:N \l_regex_match_count_int
+ \regex_match_once:
+ }
+ #1 {#2} {#3}
+ \exp_args:NNNo
\group_end:
+ \int_set:Nn #4 { \int_use:N \l_regex_match_count_int }
}
-\cs_new_protected_nopar:Npn \regex_replace_all_aux:
+\cs_new_protected:Npn \regex_extract_once_aux:NnnN #1#2#3#4
{
- \regex_extract:
- \regex_extract_aux:nTF {0}
- {
- \tl_gput_right:Nx \g_regex_replaced_str
+ \group_begin:
+ \tl_set:Nn \l_regex_every_match_tl { \regex_extract: }
+ #1 {#2} {#3}
+ \regex_group_end_extract_seq:N #4
+ }
+\cs_new_protected:Npn \regex_extract_all_aux:NnnN #1#2#3#4
+ {
+ \group_begin:
+ \tl_set:Nn \l_regex_every_match_tl
+ { \regex_extract: \regex_match_once: }
+ #1 {#2} {#3}
+ \regex_group_end_extract_seq:N #4
+ }
+\cs_new_protected:Npn \regex_split_aux:NnnN #1#2#3#4
+ {
+ \group_begin:
+ \tl_set:Nn \l_regex_every_match_tl
+ {
+ \if_num:w \l_regex_start_step_int < \l_regex_success_step_int
+ \regex_extract:
+ \tex_skip:D \l_regex_submatch_start_int
+ = \l_regex_start_step_int sp
+ plus \tex_skip:D \l_regex_submatch_start_int \scan_stop:
+ \fi:
+ \regex_match_once:
+ }
+ #1 {#2} {#3}
+ \tex_skip:D \l_regex_submatch_int
+ = \l_regex_start_step_int sp
+ plus \l_regex_current_step_int sp \scan_stop:
+ \int_incr:N \l_regex_submatch_int
+ \if_num:w \l_regex_start_step_int = \l_regex_current_step_int
+ \if_meaning:w \c_true_bool \l_regex_success_empty_bool
+ \int_decr:N \l_regex_submatch_int
+ \fi:
+ \fi:
+ \regex_group_end_extract_seq:N #4
+ }
+\cs_new_protected:Npn \regex_replace_once_aux:NnnN #1#2#3#4
+ {
+ \group_begin:
+ \tl_clear:N \l_regex_every_match_tl
+ \regex_replacement:n {#3}
+ \exp_args:Nno #1 {#2} #4
+ \if_meaning:w \c_true_bool \g_regex_success_bool
+ \regex_extract:
+ \int_set:Nn \l_regex_tmpa_int
+ { \regex_nesting:n { \l_regex_submatch_start_int } }
+ \if_num:w \l_regex_tmpa_int = \c_zero
+ \else:
+ \msg_kernel_error:nnx { regex } { replace-unbalanced }
+ { \l_regex_tmpa_int }
+ \fi:
+ \tl_set:Nx \l_regex_tmpa_tl
{
- \regex_query_substr:NN
- \l_regex_start_step_int \l_regex_tmpa_tl
- \l_regex_replacement_tl
+ \if_num:w \l_regex_tmpa_int < \c_zero
+ \prg_replicate:nn { - \l_regex_tmpa_int }
+ { \exp_not:n { { \if_false: } \fi: } }
+ \fi:
+ \regex_query_substr:nn
+ { \l_regex_min_step_int }
+ { \tex_skip:D \l_regex_submatch_start_int }
+ \regex_replacement_tl:n { \l_regex_submatch_start_int }
+ \regex_query_substr:nn
+ { \etex_gluestretch:D \tex_skip:D \l_regex_submatch_start_int }
+ { \l_regex_max_step_int }
+ \if_num:w \l_regex_tmpa_int > \c_zero
+ \prg_replicate:nn { \l_regex_tmpa_int }
+ { \exp_not:n { \if_false: { \fi: } } }
+ \fi:
}
- }
- { \msg_error:nn { regex } { internal } }
- \regex_match_once:
+ \tl_set:Nx \l_regex_tmpa_tl { \l_regex_tmpa_tl }
+ \exp_args:NNNo \group_end:
+ \tl_set:Nn #4 \l_regex_tmpa_tl
+ \else:
+ \group_end:
+ \fi:
}
-\cs_new_protected:Npn \regex_replace_all:nnN #1#2#3
+\cs_new_protected_nopar:Npn \regex_replace_once_aux:w #1;
{
- \regex_replace_all_aux:Nn #3
+ }
+\cs_new_protected:Npn \regex_replace_all_aux:NnnN #1#2#3#4
+ {
+ \group_begin:
+ \tl_set:Nn \l_regex_every_match_tl
+ {
+ \regex_extract:
+ \tex_skip:D \l_regex_submatch_start_int
+ = \tex_the:D \tex_skip:D \l_regex_submatch_start_int
+ minus \l_regex_start_step_int sp \scan_stop:
+ \regex_match_once:
+ }
+ \regex_replacement:n {#3}
+ \exp_args:Nno #1 {#2} #4
+ \int_set:Nn \l_regex_tmpa_int
+ {
+ 0
+ \prg_stepwise_function:nnnN
+ \l_regex_max_state_int
+ \l_regex_capturing_group_int
+ { \l_regex_submatch_int - \c_one }
+ \regex_nesting:n
+ }
+ \if_num:w \l_regex_tmpa_int = \c_zero
+ \else:
+ \msg_kernel_error:nnx { regex } { replace-unbalanced }
+ { \l_regex_tmpa_int }
+ \fi:
+ \tl_set:Nx \l_regex_tmpa_tl
+ {
+ \if_num:w \l_regex_tmpa_int < \c_zero
+ \prg_replicate:nn { - \l_regex_tmpa_int }
+ { \exp_not:n { { \if_false: } \fi: } }
+ \fi:
+ \prg_stepwise_function:nnnN
+ \l_regex_max_state_int
+ \l_regex_capturing_group_int
+ { \l_regex_submatch_int - \c_one }
+ \regex_replace_all_aux:n
+ \regex_query_substr:nn
+ \l_regex_start_step_int \l_regex_max_step_int
+ \if_num:w \l_regex_tmpa_int > \c_zero
+ \prg_replicate:nn { \l_regex_tmpa_int }
+ { \exp_not:n { \if_false: { \fi: } } }
+ \fi:
+ }
+ \tl_set:Nx \l_regex_tmpa_tl { \l_regex_tmpa_tl }
+ \exp_args:NNNo \group_end:
+ \tl_set:Nn #4 \l_regex_tmpa_tl
+ }
+\cs_new_nopar:Npn \regex_replace_all_aux:n #1
+ {
+ \regex_query_substr:nn
+ { \etex_glueshrink:D \tex_skip:D #1 } { \tex_skip:D #1 }
+ \regex_replacement_tl:n {#1}
+ }
+\msg_kernel_new:nnnn { regex } { sequence-unbalanced }
+ {
+ Missing~
+ \flag_test:NTF \l_regex_group_end_flag
{
- \regex_build:n {#1}
- \regex_replacement:n {#2}
+ left~
+ \flag_test:NTF \l_regex_group_begin_flag
+ { and~right~braces } { brace }
}
+ { right~brace }
+ \ inserted~in~extracted~match.
}
-\cs_new_protected:Npn \regex_replace_all:NnN #1#2#3
{
- \regex_replace_all_aux:Nn #3
+ LaTeX~was~asked~to~extract~submatches~or~split~a~token~list~
+ according~to~a~given~regular~expression,~but~some~of~the~resulting~
+ items~were~not~balanced.
+ }
+\msg_kernel_new:nnnn { regex } { replace-unbalanced }
+ { The~result~of~a~replacement~does~not~have~balanced~braces. }
+ {
+ LaTeX~was~asked~to~do~some~regular~expression~replacement,~
+ and~the~resulting~token~list~would~not~have~the~same~number~
+ of~begin-group~and~end-group~tokens. \\ \\
+ \ \ \ \
+ \prg_case_int:nnn {#1}
{
- \regex_use:N #1
- \regex_replacement:n {#2}
+ { -1 } { A~left~brace~was }
+ { 1 } { A~right~brace~was }
}
+ {
+ \int_abs:n {#1} ~
+ \int_compare:nNnTF {#1} < \c_zero { left } { right } ~
+ braces ~ were
+ }
+ \ inserted.
+ }
+\msg_kernel_new:nnnn { regex } { missing-rparen }
+ { Missing~right~parenthesis~added~in~regular~expression. }
+ {
+ LaTeX~was~given~a~regular~expression~with~\int_eval:n{#1}~
+ more~left~parenthes\int_compare:nTF{#1=1}{i}{e}s~than~right~
+ parenthes\int_compare:nTF{#1=1}{i}{e}s.
+ }
+\msg_kernel_new:nnnn { regex } { extra-rparen }
+ { Extra~right~parenthesis~ignored~in~regular~expression. }
+ {
+ LaTeX~came~across~a~closing~parenthesis~when~no~submatch~group~
+ was~open.~The~parenthesis~will~be~ignored.
}
-\msg_new:nnn { regex } { -998 } { Unsupported~construct }
-\msg_new:nnn { regex } { -997 }
- { The~regular~expression~is~too~large~(32768~states). }
-\msg_new:nnn { regex } { 1 } { \iow_char:N\\~at~end~of~pattern }
-\msg_new:nnn { regex } { 4 }
- { Numbers~out~of~order~in~\iow_char:N\{\iow_char\}~quantifier. }
-\msg_new:nnn { regex } { 6 }
- { Missing~terminating~\iow_char:N\]~for~character~class }
-\msg_new:nnn { regex } { 7 }
- { Invalid~escape~sequence~in~character~class }
-\msg_new:nnn { regex } { 8 }
- { Range~out~of~order~in~character~class }
-\msg_new:nnn { regex } { 22 } { Mismatched~parentheses }
-\msg_new:nnn { regex } { 34 }
- { Character~value~in~\iow_char:N\\x{...}~sequence~is~too~large }
-\msg_new:nnn { regex } { not-nfa }
- {
- LaTeX~was~expecting~a~regular~expression~variable.\\
- Instead,~LaTeX~found~'#1'.
+\msg_kernel_new:nnnn { regex } { backwards-range }
+ { Range~[#1-#2]~out~of~order~in~character~class. }
+ {
+ In~ranges~of~characters~[x-y]~appearing~in~character~classes,~
+ the~first~character~code~must~not~be~larger~than~the~second.~
+ Here,~#1~has~character~code~\int_eval:n {`#1},~while~#2~has~
+ character~code~\int_eval:n {`#2}.
+ }
+\msg_kernel_new:nnnn { regex } { unsupported }
+ { Unsupported~construction~`#1'. }
+ {
+ The~construction~`#1'~is~not~supported~by~the~LaTeX~
+ regular~expression~module.~Perhaps~some~character~should~
+ have~been~escaped?
+ }
+\msg_kernel_new:nnnn { regex } { not-nfa }
+ { This~is~not~a~regular~expression~variable. }
+ {
+ LaTeX~was~expecting~`#1'~to~be~a~regular~expression~variable.\\
+ This~control~sequence~is~not~a~regex~variable.~It's~current~meaning~
+ is~\\\\
+ \ \ \ \ \token_to_str:N #1 = \token_to_meaning:N #1 .
+ }
+\msg_kernel_new:nnnn { regex } { build-c }
+ { Misused~\iow_char:N\\c~or\iow_char:N\\C~command~in~regular~expression. }
+ {
+ The~\iow_char:N\\C~escape~sequence~can~be~followed~by~
+ one~of~the~letters~ABCDELMOPSTU.~The~\iow_char:N\\c~
+ escape~sequence~can~also~be~followed~by~a~brace~group.
+ }
+\msg_kernel_new:nnnn { regex } { unknown-option }
+ { Unknown~option~`#1'~for~regular~expressions. }
+ {
+ LaTeX~came~across~something~like~`(?#1)'~in~a~regular~expression,~
+ but~the~option~`#1'~is~not~known.~It~will~be~ignored.
+ }
+\msg_kernel_new:nnnn { regex } { invalid-in-option }
+ { Invalid~character~in~option~of~a~regular~expression. }
+ {
+ The~character~or~escape~sequence~`#1'~is~not~defined~
+ as~an~option~within~regular~expressions.
+ }
+\msg_kernel_new:nnnn { regex } { replacement-g }
+ { Missing~brace~for~the~\iow_char:N\\g~construction~in~a~replacement~text. }
+ {
+ In~the~replacement~text~for~a~regular~expression~search,~
+ submatches~are~represented~either~as~\iow_char:N \\g{dd..d},~
+ or~\\d,~where~`d'~are~single~digits.~Here,~a~brace~is~missing.
+ }
+\msg_kernel_new:nnnn { regex } { replacement-c }
+ { Something~is~wrong~with~the~replacement~text. }
+ {
+ The~\iow_char:N\\c~escape~sequence~can~only~be~followed~by~
+ ABCDELMN~or~by~a~brace.
}
\msg_kernel_new:nnn { regex } { nfa-misused }
{ Automaton~#1 used~incorrectly. }
-\msg_kernel_new:nnn { regex } { g-misused }
- { \g misused~in~replacement~text. }
%%
%%
%% End of file `l3regex.sty'.