Fix compiling a class with UCP and without UTF (port to 10.22)

This commit is contained in:
Petr Písař 2017-01-03 15:34:41 +01:00
parent dc60893013
commit de310a916b

View File

@ -1,4 +1,4 @@
From a83027bb4b195c879d504da051571f22a5ac7ca3 Mon Sep 17 00:00:00 2001
From b3343e2c2c77b85f841a7af5e4121dab11692065 Mon Sep 17 00:00:00 2001
From: ph10 <ph10@6239d852-aaf2-0410-a92c-79f79f948069>
Date: Mon, 26 Dec 2016 17:11:18 +0000
Subject: [PATCH] Fix class bug when UCP but not UTF was set and all wide
@ -7,42 +7,39 @@ MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
git-svn-id: svn://vcs.exim.org/pcre2/code/trunk@628 6239d852-aaf2-0410-a92c-79f79f948069
Ported to 10.22:
commit a83027bb4b195c879d504da051571f22a5ac7ca3
Author: ph10 <ph10@6239d852-aaf2-0410-a92c-79f79f948069>
Date: Mon Dec 26 17:11:18 2016 +0000
Fix class bug when UCP but not UTF was set and all wide characters need to b
e
included.
git-svn-id: svn://vcs.exim.org/pcre2/code/trunk@628 6239d852-aaf2-0410-a92c-79f79f948069
Signed-off-by: Petr Písař <ppisar@redhat.com>
---
ChangeLog | 4 ++++
src/pcre2_compile.c | 8 +++++++-
src/pcre2_compile.c | 10 ++++++++--
testdata/testinput10 | 2 ++
testdata/testinput12 | 2 ++
testdata/testoutput10 | 8 ++++++++
testdata/testoutput12-16 | 8 ++++++++
testdata/testoutput12-32 | 8 ++++++++
7 files changed, 39 insertions(+), 1 deletion(-)
6 files changed, 36 insertions(+), 2 deletions(-)
diff --git a/ChangeLog b/ChangeLog
index 6156b24..d276feb 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -255,6 +255,10 @@ This lifts a restriction on the number of branches in a group (more than about
38. Add the "-ac" command line option to pcre2test as a synonym for "-pattern
auto_callout".
+39. In a library with Unicode support, incorrect data was compiled for a
+pattern with PCRE2_UCP set without PCRE2_UTF if a class required all wide
+characters to match (for example, /[\s[:^ascii:]]/).
+
Version 10.22 29-July-2016
--------------------------
diff --git a/src/pcre2_compile.c b/src/pcre2_compile.c
index 9dc9c98..fce226b 100644
index ae6b5e1..2c9f758 100644
--- a/src/pcre2_compile.c
+++ b/src/pcre2_compile.c
@@ -4927,9 +4927,13 @@ for (;; pptr++)
@@ -4482,10 +4482,14 @@ for (;; ptr++)
In the special case where there are no xclass items, this is
automatically handled by the use of OP_CLASS or OP_NCLASS, but an
explicit range is needed for OP_XCLASS. Setting a flag here
causes the range to be generated later when it is known that
- OP_XCLASS is required. */
explicit range is needed for OP_XCLASS. Setting a flag here causes
- the range to be generated later when it is known that OP_XCLASS is
- required. */
+ the range to be generated later when it is known that
+ OP_XCLASS is required. In the 8-bit library this is relevant only in
+ utf mode, since no wide characters can exist otherwise. */
@ -53,7 +50,7 @@ index 9dc9c98..fce226b 100644
match_all_or_no_wide_chars |= local_negate;
break;
}
@@ -5217,6 +5221,8 @@ for (;; pptr++)
@@ -4993,6 +4997,8 @@ for (;; ptr++)
all wide characters (depending on whether the whole class is or is not
negated). This requirement is indicated by match_all_or_no_wide_chars being
true. We do this by including an explicit range, which works in both cases.
@ -63,21 +60,21 @@ index 9dc9c98..fce226b 100644
When there *are* properties in a positive UTF-8 or any 16-bit or 32_bit
class where \S etc is present without PCRE2_UCP, causing an extended class
diff --git a/testdata/testinput10 b/testdata/testinput10
index a1806ae..85d2005 100644
index 4b80778..1c6134b 100644
--- a/testdata/testinput10
+++ b/testdata/testinput10
@@ -456,4 +456,6 @@
/(*:*++++++++++++''''''''''''''''''''+''+++'+++x+++++++++++++++++++++++++++++++++++(++++++++++++++++++++:++++++%++:''''''''''''''''''''''''+++++++++++++++++++++++++++++++++++++++++++++++++++++-++++++++k+++++++''''+++'+++++++++++++++++++++++''''++++++++++++':ƿ)/utf
@@ -454,4 +454,6 @@
\= Expect no match
123
+/[\s[:^ascii:]]/B,ucp
+
# End of testinput10
diff --git a/testdata/testinput12 b/testdata/testinput12
index 5b29f41..c3b2bfc 100644
index 29934ec..d851ae6 100644
--- a/testdata/testinput12
+++ b/testdata/testinput12
@@ -358,4 +358,6 @@
@@ -354,4 +354,6 @@
\= Expect no match
123
@ -85,12 +82,12 @@ index 5b29f41..c3b2bfc 100644
+
# End of testinput12
diff --git a/testdata/testoutput10 b/testdata/testoutput10
index 3c35f0b..31b7d00 100644
index 0c1e9b2..aef89ca 100644
--- a/testdata/testoutput10
+++ b/testdata/testoutput10
@@ -1567,4 +1567,12 @@ No match
/(*:*++++++++++++''''''''''''''''''''+''+++'+++x+++++++++++++++++++++++++++++++++++(++++++++++++++++++++:++++++%++:''''''''''''''''''''''''+++++++++++++++++++++++++++++++++++++++++++++++++++++-++++++++k+++++++''''+++'+++++++++++++++++++++++''''++++++++++++':ƿ)/utf
Failed: error 176 at offset 259: name is too long in (*MARK), (*PRUNE), (*SKIP), or (*THEN)
@@ -1564,4 +1564,12 @@ Failed: error -22: UTF-8 error: isolated byte with 0x80 bit set at offset 1
123
No match
+/[\s[:^ascii:]]/B,ucp
+------------------------------------------------------------------
@ -102,10 +99,10 @@ index 3c35f0b..31b7d00 100644
+
# End of testinput10
diff --git a/testdata/testoutput12-16 b/testdata/testoutput12-16
index f0f2230..3b5a0cd 100644
index 9cd6640..e2d5b9f 100644
--- a/testdata/testoutput12-16
+++ b/testdata/testoutput12-16
@@ -1407,4 +1407,12 @@ Subject length lower bound = 2
@@ -1396,4 +1396,12 @@ Subject length lower bound = 2
123
No match
@ -119,10 +116,10 @@ index f0f2230..3b5a0cd 100644
+
# End of testinput12
diff --git a/testdata/testoutput12-32 b/testdata/testoutput12-32
index 3718b27..1496159 100644
index 75a5ad7..7479a93 100644
--- a/testdata/testoutput12-32
+++ b/testdata/testoutput12-32
@@ -1401,4 +1401,12 @@ Failed: error -28: UTF-32 error: code points greater than 0x10ffff are not defin
@@ -1390,4 +1390,12 @@ Failed: error -28: UTF-32 error: code points greater than 0x10ffff are not defin
123
No match