Fix issue 15261: Crash in history::TextDatabase::GetTextMatches

The crash in history::TextDatabase::GetTextMatches is caused by unexpected result of tolower() in some locales such as tr_TR.UTF-8. This CL fixes this issue by using following statement to replace tolower():

(ch>='A' && ch<='Z') ? (ch-'A'+'a') : ch

which will always return expected result for ascii characters regardless of current locale.

BUG=15261 Crash in history::TextDatabase::GetTextMatches
TEST=none

Review URL: http://codereview.chromium.org/174387

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@25141 0039d316-1c4b-4281-b951-d872f2087c98

9 files changed, 109 insertions, 8 deletions

diff --git a/third_party/sqlite/README.chromium b/third_party/sqlite/README.chromium
index 7daa083..1050a7d 100644
--- a/third_party/sqlite/README.chromium
+++ b/third_party/sqlite/README.chromium
@@ -68,7 +68,7 @@ Scott Hess <shess@google.com>, December 11, 2007
 
 --------------------------------------------
 
-As of July 20, 2009, these are our changes from sqlite_vendor:
+As of Sep 2, 2009, these are our changes from sqlite_vendor:
 
  - fts2.c disables fts2_tokenizer().
  - sqlite3Poison() in src/btree.c.
@@ -82,6 +82,10 @@ As of July 20, 2009, these are our changes from sqlite_vendor:
  - Compile warning fixed in func.c (check if this is still needed)
  - Fixed a typo bug in fts2_icu.c: "strlen(nInput)" (filed upstream as
    http://www.sqlite.org/cvstrac/tktview?tn=3543)
+ - Avoid using tolower() in fts code which causes problem in some locales, see:
+   safe-tolower.patch
+   http://crbug.com/15261
+   http://www.sqlite.org/src/tktview/991789d9f3136a0460dc83a33e815c1aa9757c26
 
 Changes from Chrome:
  - I marked all changes I made with "evanm", so you can find them with
diff --git a/third_party/sqlite/ext/fts1/fts1.c b/third_party/sqlite/ext/fts1/fts1.c
index f067c55..c471a97 100644
--- a/third_party/sqlite/ext/fts1/fts1.c
+++ b/third_party/sqlite/ext/fts1/fts1.c
@@ -208,7 +208,7 @@ static int safe_isspace(char c){
   return (c&0x80)==0 ? isspace(c) : 0;
 }
 static int safe_tolower(char c){
-  return (c&0x80)==0 ? tolower(c) : c;
+  return (c>='A' && c<='Z') ? (c-'A'+'a') : c;
 }
 static int safe_isalnum(char c){
   return (c&0x80)==0 ? isalnum(c) : 0;
diff --git a/third_party/sqlite/ext/fts1/fts1_tokenizer1.c b/third_party/sqlite/ext/fts1/fts1_tokenizer1.c
index f58fba8..90774da 100644
--- a/third_party/sqlite/ext/fts1/fts1_tokenizer1.c
+++ b/third_party/sqlite/ext/fts1/fts1_tokenizer1.c
@@ -182,7 +182,7 @@ static int simpleNext(
         ** case-insensitivity.
         */
         unsigned char ch = p[iStartOffset+i];
-        c->pToken[i] = ch<0x80 ? tolower(ch) : ch;
+        c->pToken[i] = (ch>='A' && ch<='Z') ? (ch-'A'+'a') : ch;
       }
       *ppToken = c->pToken;
       *pnBytes = n;
diff --git a/third_party/sqlite/ext/fts1/simple_tokenizer.c b/third_party/sqlite/ext/fts1/simple_tokenizer.c
index d00a770..9a52bd1 100644
--- a/third_party/sqlite/ext/fts1/simple_tokenizer.c
+++ b/third_party/sqlite/ext/fts1/simple_tokenizer.c
@@ -138,7 +138,7 @@ static int simpleNext(
         ** case-insensitivity.
         */
         char ch = c->pCurrent[ii];
-        c->zToken[ii] = (unsigned char)ch<0x80 ? tolower(ch) : ch;
+        c->zToken[ii] = ((ch>='A' && ch<='Z') ? (ch-'A'+'a') : ch);
       }
       c->zToken[n] = '\0';
       *ppToken = c->zToken;
diff --git a/third_party/sqlite/ext/fts2/fts2.c b/third_party/sqlite/ext/fts2/fts2.c
index 35ab87c..b261ead 100644
--- a/third_party/sqlite/ext/fts2/fts2.c
+++ b/third_party/sqlite/ext/fts2/fts2.c
@@ -372,7 +372,7 @@ static int safe_isspace(char c){
   return (c&0x80)==0 ? isspace(c) : 0;
 }
 static int safe_tolower(char c){
-  return (c&0x80)==0 ? tolower(c) : c;
+  return (c>='A' && c<='Z') ? (c-'A'+'a') : c;
 }
 static int safe_isalnum(char c){
   return (c&0x80)==0 ? isalnum(c) : 0;
diff --git a/third_party/sqlite/ext/fts2/fts2_tokenizer1.c b/third_party/sqlite/ext/fts2/fts2_tokenizer1.c
index f2ba49e..1f03cc4 100644
--- a/third_party/sqlite/ext/fts2/fts2_tokenizer1.c
+++ b/third_party/sqlite/ext/fts2/fts2_tokenizer1.c
@@ -191,7 +191,7 @@ static int simpleNext(
         ** case-insensitivity.
         */
         unsigned char ch = p[iStartOffset+i];
-        c->pToken[i] = ch<0x80 ? tolower(ch) : ch;
+        c->pToken[i] = (ch>='A' && ch<='Z') ? (ch-'A'+'a') : ch;
       }
       *ppToken = c->pToken;
       *pnBytes = n;
diff --git a/third_party/sqlite/ext/fts3/fts3.c b/third_party/sqlite/ext/fts3/fts3.c
index cc6dcd8..922cf3f 100644
--- a/third_party/sqlite/ext/fts3/fts3.c
+++ b/third_party/sqlite/ext/fts3/fts3.c
@@ -330,7 +330,7 @@ static int safe_isspace(char c){
   return (c&0x80)==0 ? isspace(c) : 0;
 }
 static int safe_tolower(char c){
-  return (c&0x80)==0 ? tolower(c) : c;
+  return (c>='A' && c<='Z') ? (c-'A'+'a') : c;
 }
 static int safe_isalnum(char c){
   return (c&0x80)==0 ? isalnum(c) : 0;
diff --git a/third_party/sqlite/ext/fts3/fts3_tokenizer1.c b/third_party/sqlite/ext/fts3/fts3_tokenizer1.c
index da255d9..1df3910 100644
--- a/third_party/sqlite/ext/fts3/fts3_tokenizer1.c
+++ b/third_party/sqlite/ext/fts3/fts3_tokenizer1.c
@@ -191,7 +191,7 @@ static int simpleNext(
         ** case-insensitivity.
         */
         unsigned char ch = p[iStartOffset+i];
-        c->pToken[i] = ch<0x80 ? tolower(ch) : ch;
+        c->pToken[i] = (ch>='A' && ch<='Z') ? (ch-'A'+'a') : ch;
       }
       *ppToken = c->pToken;
       *pnBytes = n;
diff --git a/third_party/sqlite/safe-tolower.patch b/third_party/sqlite/safe-tolower.patch
new file mode 100644
index 0000000..a0ec849
--- /dev/null
+++ b/third_party/sqlite/safe-tolower.patch
@@ -0,0 +1,97 @@
+This patch removes the usage of tolower() in fts code, which is not locale
+neutral and causes problem in some locales such as Turkish.
+See http://crbug.com/15261 for details.
+An upstream ticket was also created for this issue:
+http://www.sqlite.org/src/tktview/991789d9f3136a0460dc83a33e815c1aa9757c26
+
+Index: ext/fts3/fts3.c
+===================================================================
+--- ext/fts3/fts3.c	(revision 24387)
++++ ext/fts3/fts3.c	(working copy)
+@@ -330,7 +330,7 @@
+   return (c&0x80)==0 ? isspace(c) : 0;
+ }
+ static int safe_tolower(char c){
+-  return (c&0x80)==0 ? tolower(c) : c;
++  return (c>='A' && c<='Z') ? (c-'A'+'a') : c;
+ }
+ static int safe_isalnum(char c){
+   return (c&0x80)==0 ? isalnum(c) : 0;
+Index: ext/fts3/fts3_tokenizer1.c
+===================================================================
+--- ext/fts3/fts3_tokenizer1.c	(revision 24387)
++++ ext/fts3/fts3_tokenizer1.c	(working copy)
+@@ -191,7 +191,7 @@
+         ** case-insensitivity.
+         */
+         unsigned char ch = p[iStartOffset+i];
+-        c->pToken[i] = ch<0x80 ? tolower(ch) : ch;
++        c->pToken[i] = (ch>='A' && ch<='Z') ? (ch-'A'+'a') : ch;
+       }
+       *ppToken = c->pToken;
+       *pnBytes = n;
+Index: ext/fts1/simple_tokenizer.c
+===================================================================
+--- ext/fts1/simple_tokenizer.c	(revision 24387)
++++ ext/fts1/simple_tokenizer.c	(working copy)
+@@ -138,7 +138,7 @@
+         ** case-insensitivity.
+         */
+         char ch = c->pCurrent[ii];
+-        c->zToken[ii] = (unsigned char)ch<0x80 ? tolower(ch) : ch;
++        c->zToken[ii] = ((ch>='A' && ch<='Z') ? (ch-'A'+'a') : ch);
+       }
+       c->zToken[n] = '\0';
+       *ppToken = c->zToken;
+Index: ext/fts1/fts1_tokenizer1.c
+===================================================================
+--- ext/fts1/fts1_tokenizer1.c	(revision 24387)
++++ ext/fts1/fts1_tokenizer1.c	(working copy)
+@@ -182,7 +182,7 @@
+         ** case-insensitivity.
+         */
+         unsigned char ch = p[iStartOffset+i];
+-        c->pToken[i] = ch<0x80 ? tolower(ch) : ch;
++        c->pToken[i] = (ch>='A' && ch<='Z') ? (ch-'A'+'a') : ch;
+       }
+       *ppToken = c->pToken;
+       *pnBytes = n;
+Index: ext/fts1/fts1.c
+===================================================================
+--- ext/fts1/fts1.c	(revision 24387)
++++ ext/fts1/fts1.c	(working copy)
+@@ -208,7 +208,7 @@
+   return (c&0x80)==0 ? isspace(c) : 0;
+ }
+ static int safe_tolower(char c){
+-  return (c&0x80)==0 ? tolower(c) : c;
++  return (c>='A' && c<='Z') ? (c-'A'+'a') : c;
+ }
+ static int safe_isalnum(char c){
+   return (c&0x80)==0 ? isalnum(c) : 0;
+Index: ext/fts2/fts2.c
+===================================================================
+--- ext/fts2/fts2.c	(revision 24387)
++++ ext/fts2/fts2.c	(working copy)
+@@ -372,7 +372,7 @@
+   return (c&0x80)==0 ? isspace(c) : 0;
+ }
+ static int safe_tolower(char c){
+-  return (c&0x80)==0 ? tolower(c) : c;
++  return (c>='A' && c<='Z') ? (c-'A'+'a') : c;
+ }
+ static int safe_isalnum(char c){
+   return (c&0x80)==0 ? isalnum(c) : 0;
+Index: ext/fts2/fts2_tokenizer1.c
+===================================================================
+--- ext/fts2/fts2_tokenizer1.c	(revision 24387)
++++ ext/fts2/fts2_tokenizer1.c	(working copy)
+@@ -191,7 +191,7 @@
+         ** case-insensitivity.
+         */
+         unsigned char ch = p[iStartOffset+i];
+-        c->pToken[i] = ch<0x80 ? tolower(ch) : ch;
++        c->pToken[i] = (ch>='A' && ch<='Z') ? (ch-'A'+'a') : ch;
+       }
+       *ppToken = c->pToken;
+       *pnBytes = n;