From 64193c7de9f53e01dca447a4a902398fe85b9c06 Mon Sep 17 00:00:00 2001
From: Damien George <damien@micropython.org>
Date: Fri, 22 Oct 2021 22:23:47 +1100
Subject: [PATCH] lib/re1.5: Add support for named classes in class sets.

Total code size change of this and previous commit:

   bare-arm:    +0 +0.000%
minimal x86:    +0 +0.000%
   unix x64:   +32 +0.004% standard
      stm32:   +24 +0.006% PYBV10
     cc3200:   +16 +0.009%
    esp8266:   +20 +0.003% GENERIC
      esp32:   +44 +0.003% GENERIC[incl +8(data)]
     mimxrt:   +32 +0.009% TEENSY40
 renesas-ra:   +24 +0.004% RA6M2_EK
        nrf:    +0 +0.000% pca10040
        rp2:   +24 +0.005% PICO
       samd:   +32 +0.012% ADAFRUIT_ITSYBITSY_M4_EXPRESS

Addresses issue #7920.

Signed-off-by: Damien George <damien@micropython.org>
---
 lib/re1.5/charclass.c   | 10 +++++++++-
 lib/re1.5/compilecode.c | 13 ++++++++++---
 lib/re1.5/re1.5.h       |  1 +
 3 files changed, 20 insertions(+), 4 deletions(-)

diff --git a/lib/re1.5/charclass.c b/lib/re1.5/charclass.c
index 7f6388c93d..2553b40530 100644
--- a/lib/re1.5/charclass.c
+++ b/lib/re1.5/charclass.c
@@ -6,7 +6,15 @@ int _re1_5_classmatch(const char *pc, const char *sp)
     int is_positive = (pc[-1] == Class);
     int cnt = *pc++;
     while (cnt--) {
-        if (*sp >= *pc && *sp <= pc[1]) return is_positive;
+        if (*pc == RE15_CLASS_NAMED_CLASS_INDICATOR) {
+            if (_re1_5_namedclassmatch(pc + 1, sp)) {
+                return is_positive;
+            }
+        } else {
+            if (*sp >= *pc && *sp <= pc[1]) {
+                return is_positive;
+            }
+        }
         pc += 2;
     }
     return !is_positive;
diff --git a/lib/re1.5/compilecode.c b/lib/re1.5/compilecode.c
index d4fc2a2895..513a155970 100644
--- a/lib/re1.5/compilecode.c
+++ b/lib/re1.5/compilecode.c
@@ -66,14 +66,21 @@ static const char *_compilecode(const char *re, ByteProg *prog, int sizecode)
             PC++; // Skip # of pair byte
             prog->len++;
             for (cnt = 0; *re != ']'; re++, cnt++) {
-                if (*re == '\\') {
+                char c = *re;
+                if (c == '\\') {
                     ++re;
+                    c = *re;
+                    if (MATCH_NAMED_CLASS_CHAR(c)) {
+                        c = RE15_CLASS_NAMED_CLASS_INDICATOR;
+                        goto emit_char_pair;
+                    }
                 }
-                if (!*re) return NULL;
-                EMIT(PC++, *re);
+                if (!c) return NULL;
                 if (re[1] == '-' && re[2] != ']') {
                     re += 2;
                 }
+            emit_char_pair:
+                EMIT(PC++, c);
                 EMIT(PC++, *re);
             }
             EMIT_CHECKED(term + 1, cnt);
diff --git a/lib/re1.5/re1.5.h b/lib/re1.5/re1.5.h
index 81f43ed7f5..b1ec01cbc5 100644
--- a/lib/re1.5/re1.5.h
+++ b/lib/re1.5/re1.5.h
@@ -138,6 +138,7 @@ struct Subject {
 
 #define NON_ANCHORED_PREFIX 5
 #define HANDLE_ANCHORED(bytecode, is_anchored) ((is_anchored) ? (bytecode) + NON_ANCHORED_PREFIX : (bytecode))
+#define RE15_CLASS_NAMED_CLASS_INDICATOR 0
 
 int re1_5_backtrack(ByteProg*, Subject*, const char**, int, int);
 int re1_5_pikevm(ByteProg*, Subject*, const char**, int, int);