Fix buffer overrun for class with very many characters whose codepoints are

above 255.


git-svn-id: svn://vcs.exim.org/pcre/code/trunk@300 2f5784b3-3f2a-0410-8824-cb99058d5e15

5 files changed, 27 insertions, 3 deletions

diff --git a/ChangeLog b/ChangeLog
index a697896..44a9abc 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,6 +1,13 @@
 ChangeLog for PCRE
 ------------------
 
+Version 7.6 14-Jan-08
+---------------------
+
+1.  A class with a very large number of characters with codepoints greater
+    than 255 (in UTF-8 mode, of course), caused a buffer overflow.
+     
+
 Version 7.5 10-Jan-08
 ---------------------
 
diff --git a/configure.ac b/configure.ac
index 19eaf54..f4d1b7e 100644
--- a/configure.ac
+++ b/configure.ac
@@ -7,9 +7,9 @@ dnl be defined as -RC2, for example. For real releases, it should be defined
 dnl empty.
 
 m4_define(pcre_major, [7])
-m4_define(pcre_minor, [5])
+m4_define(pcre_minor, [6])
 m4_define(pcre_prerelease, [])
-m4_define(pcre_date, [2008-01-10])
+m4_define(pcre_date, [2008-01-14])
 
 # Libtool shared library interface versions (current:revision:age)
 m4_define(libpcre_version, [0:1:0])
diff --git a/pcre_compile.c b/pcre_compile.c
index d4840cb..b7b6fc2 100644
--- a/pcre_compile.c
+++ b/pcre_compile.c
@@ -2376,6 +2376,7 @@ uschar classbits[32];
 BOOL class_utf8;
 BOOL utf8 = (options & PCRE_UTF8) != 0;
 uschar *class_utf8data;
+uschar *class_utf8data_base;
 uschar utf8_char[6];
 #else
 BOOL utf8 = FALSE;
@@ -2687,6 +2688,7 @@ for (;; ptr++)
 #ifdef SUPPORT_UTF8
     class_utf8 = FALSE;                       /* No chars >= 256 */
     class_utf8data = code + LINK_SIZE + 2;    /* For UTF-8 items */
+    class_utf8data_base = class_utf8data;     /* For resetting in pass 1 */ 
 #endif
 
     /* Process characters until ] is reached. By writing this as a "do" it
@@ -2702,6 +2704,18 @@ for (;; ptr++)
         {                           /* Braces are required because the */
         GETCHARLEN(c, ptr, ptr);    /* macro generates multiple statements */
         }
+      
+      /* In the pre-compile phase, accumulate the length of any UTF-8 extra
+      data and reset the pointer. This is so that very large classes that 
+      contain a zillion UTF-8 characters no longer overwrite the work space
+      (which is on the stack). */  
+       
+      if (lengthptr != NULL)
+        {
+        *lengthptr += class_utf8data - class_utf8data_base;
+        class_utf8data = class_utf8data_base; 
+        }  
+ 
 #endif
 
       /* Inside \Q...\E everything is literal except \E */
@@ -5807,7 +5821,6 @@ to fill in forward references to subpatterns. */
 
 uschar cworkspace[COMPILE_WORK_SIZE];
 
-
 /* Set this early so that early errors get offset 0. */
 
 ptr = (const uschar *)pattern;
diff --git a/testdata/testinput4 b/testdata/testinput4
index 886dbce..c98c085 100644
--- a/testdata/testinput4
+++ b/testdata/testinput4
@@ -607,4 +607,6 @@
 /[[:^xdigit:]]/8g
     M\x{442}
 
+/[^ABCDEFGHIJKLMNOPQRSTUVWXYZÀÁÂÃÄÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖØÙÚÛÜÝÞĀĂĄĆĈĊČĎĐĒĔĖĘĚĜĞĠĢĤĦĨĪĬĮİIJĴĶĹĻĽĿŁŃŅŇŊŌŎŐŒŔŖŘŚŜŞŠŢŤŦŨŪŬŮŰŲŴŶŸŹŻŽƁƂƄƆƇƉƊƋƎƏƐƑƓƔƖƗƘƜƝƟƠƢƤƦƧƩƬƮƯƱƲƳƵƷƸƼDŽLJNJǍǏǑǓǕǗǙǛǞǠǢǤǦǨǪǬǮDZǴǶǷǸǺǼǾȀȂȄȆȈȊȌȎȐȒȔȖȘȚȜȞȠȢȤȦȨȪȬȮȰȲȺȻȽȾɁΆΈΉΊΌΎΏΑΒΓΔΕΖΗΘΙΚΛΜΝΞΟΠΡΣΤΥΦΧΨΩΪΫϒϓϔϘϚϜϞϠϢϤϦϨϪϬϮϴϷϹϺϽϾϿЀЁЂЃЄЅІЇЈЉЊЋЌЍЎЏАБВГДЕЖЗИЙКЛМНОПРСТУФХЦЧШЩЪЫЬЭЮЯѠѢѤѦѨѪѬѮѰѲѴѶѸѺѼѾҀҊҌҎҐҒҔҖҘҚҜҞҠҢҤҦҨҪҬҮҰҲҴҶҸҺҼҾӀӁӃӅӇӉӋӍӐӒӔӖӘӚӜӞӠӢӤӦӨӪӬӮӰӲӴӶӸԀԂԄԆԈԊԌԎԱԲԳԴԵԶԷԸԹԺԻԼԽԾԿՀՁՂՃՄՅՆՇՈՉՊՋՌՍՎՏՐՑՒՓՔՕՖႠႡႢႣႤႥႦႧႨႩႪႫႬႭႮႯႰႱႲႳႴႵႶႷႸႹႺႻႼႽႾႿჀჁჂჃჄჅḀḂḄḆḈḊḌḎḐḒḔḖḘḚḜḞḠḢḤḦḨḪḬḮḰḲḴḶḸḺḼḾṀṂṄṆṈṊṌṎṐṒṔṖṘṚṜṞṠṢṤṦṨṪṬṮṰṲṴṶṸṺṼṾẀẂẄẆẈẊẌẎẐẒẔẠẢẤẦẨẪẬẮẰẲẴẶẸẺẼẾỀỂỄỆỈỊỌỎỐỒỔỖỘỚỜỞỠỢỤỦỨỪỬỮỰỲỴỶỸἈἉἊἋἌἍἎἏἘἙἚἛἜἝἨἩἪἫἬἭἮἯἸἹἺἻἼἽἾἿὈὉὊὋὌὍὙὛὝὟὨὩὪὫὬὭὮὯᾸᾹᾺΆῈΈῊΉῘῙῚΊῨῩῪΎῬῸΌῺΏabcdefghijklmnopqrstuvwxyzªµºßàáâãäåæçèéêëìíîïðñòóôõöøùúûüýþÿāăąćĉċčďđēĕėęěĝğġģĥħĩīĭįıijĵķĸĺļľŀłńņňʼnŋōŏőœŕŗřśŝşšţťŧũūŭůűųŵŷźżžſƀƃƅƈƌƍƒƕƙƚƛƞơƣƥƨƪƫƭưƴƶƹƺƽƾƿdžljnjǎǐǒǔǖǘǚǜǝǟǡǣǥǧǩǫǭǯǰdzǵǹǻǽǿȁȃȅȇȉȋȍȏȑȓȕȗșțȝȟȡȣȥȧȩȫȭȯȱȳȴȵȶȷȸȹȼȿɀɐɑɒɓɔɕɖɗɘəɚɛɜɝɞɟɠɡɢɣɤɥɦɧɨɩɪɫɬɭɮɯɰɱɲɳɴɵɶɷɸɹɺɻɼɽɾɿʀʁʂʃʄʅʆʇʈʉʊʋʌʍʎʏʐʑʒʓʔʕʖʗʘʙʚʛʜʝʞʟʠʡʢʣʤʥʦʧʨʩʪʫʬʭʮʯΐάέήίΰαβγδεζηθικλμνξοπρςστυφχψωϊϋόύώϐϑϕϖϗϙϛϝϟϡϣϥϧϩϫϭϯϰϱϲϳϵϸϻϼабвгдежзийклмнопрстуфхцчшщъыьэюяѐёђѓєѕіїјљњћќѝўџѡѣѥѧѩѫѭѯѱѳѵѷѹѻѽѿҁҋҍҏґғҕҗҙқҝҟҡңҥҧҩҫҭүұҳҵҷҹһҽҿӂӄӆӈӊӌӎӑӓӕӗәӛӝӟӡӣӥӧөӫӭӯӱӳӵӷӹԁԃԅԇԉԋԍԏաբգդեզէըթժիլխծկհձղճմյնշոչպջռսվտրցւփքօֆևᴀᴁᴂᴃᴄᴅᴆᴇᴈᴉᴊᴋᴌᴍᴎᴏᴐᴑᴒᴓᴔᴕᴖᴗᴘᴙᴚᴛᴜᴝᴞᴟᴠᴡᴢᴣᴤᴥᴦᴧᴨᴩᴪᴫᵢᵣᵤᵥᵦᵧᵨᵩᵪᵫᵬᵭᵮᵯᵰᵱᵲᵳᵴᵵᵶᵷᵹᵺᵻᵼᵽᵾᵿᶀᶁᶂᶃᶄᶅᶆᶇᶈᶉᶊᶋᶌᶍᶎᶏᶐᶑᶒᶓᶔᶕᶖᶗᶘᶙᶚḁḃḅḇḉḋḍḏḑḓḕḗḙḛḝḟḡḣḥḧḩḫḭḯḱḳḵḷḹḻḽḿṁṃṅṇṉṋṍṏṑṓṕṗṙṛṝṟṡṣṥṧṩṫṭṯṱṳṵṷṹṻṽṿẁẃẅẇẉẋẍẏẑẓẕẖẗẘẙẚẛạảấầẩẫậắằẳẵặẹẻẽếềểễệỉịọỏốồổỗộớờởỡợụủứừửữựỳỵỷỹἀἁἂἃἄἅἆἇἐἑἒἓἔἕἠἡἢἣἤἥἦἧἰἱἲἳἴἵἶἷὀὁὂὃὄὅὐὑὒὓὔὕὖὗὠὡὢὣὤὥὦὧὰάὲέὴήὶίὸόὺύὼώᾀᾁᾂᾃᾄᾅᾆᾇᾐᾑᾒᾓᾔᾕᾖᾗᾠᾡᾢᾣᾤᾥᾦᾧᾰᾱᾲᾳᾴᾶᾷιῂῃῄῆῇῐῑῒΐῖῗῠῡῢΰῤῥῦῧῲῳῴῶῷⲁⲃⲅⲇⲉⲋⲍⲏⲑⲓⲕⲗⲙⲛⲝⲟⲡⲣⲥⲧⲩⲫⲭⲯⲱⲳⲵⲷⲹⲻⲽⲿⳁⳃⳅⳇⳉⳋⳍⳏⳑⳓⳕⳗⳙⳛⳝⳟⳡⳣⳤⴀⴁⴂⴃⴄⴅⴆⴇⴈⴉⴊⴋⴌⴍⴎⴏⴐⴑⴒⴓⴔⴕⴖⴗⴘⴙⴚⴛⴜⴝⴞⴟⴠⴡⴢⴣⴤⴥfffiflffifflſtstﬓﬔﬕﬖﬗ\d-_^]/8
+
 / End of testinput4 /
diff --git a/testdata/testoutput4 b/testdata/testoutput4
index cb8400f..d87ea4b 100644
--- a/testdata/testoutput4
+++ b/testdata/testoutput4
@@ -1069,4 +1069,6 @@ No match
  0: M
  0: \x{442}
 
+/[^ABCDEFGHIJKLMNOPQRSTUVWXYZÀÁÂÃÄÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖØÙÚÛÜÝÞĀĂĄĆĈĊČĎĐĒĔĖĘĚĜĞĠĢĤĦĨĪĬĮİIJĴĶĹĻĽĿŁŃŅŇŊŌŎŐŒŔŖŘŚŜŞŠŢŤŦŨŪŬŮŰŲŴŶŸŹŻŽƁƂƄƆƇƉƊƋƎƏƐƑƓƔƖƗƘƜƝƟƠƢƤƦƧƩƬƮƯƱƲƳƵƷƸƼDŽLJNJǍǏǑǓǕǗǙǛǞǠǢǤǦǨǪǬǮDZǴǶǷǸǺǼǾȀȂȄȆȈȊȌȎȐȒȔȖȘȚȜȞȠȢȤȦȨȪȬȮȰȲȺȻȽȾɁΆΈΉΊΌΎΏΑΒΓΔΕΖΗΘΙΚΛΜΝΞΟΠΡΣΤΥΦΧΨΩΪΫϒϓϔϘϚϜϞϠϢϤϦϨϪϬϮϴϷϹϺϽϾϿЀЁЂЃЄЅІЇЈЉЊЋЌЍЎЏАБВГДЕЖЗИЙКЛМНОПРСТУФХЦЧШЩЪЫЬЭЮЯѠѢѤѦѨѪѬѮѰѲѴѶѸѺѼѾҀҊҌҎҐҒҔҖҘҚҜҞҠҢҤҦҨҪҬҮҰҲҴҶҸҺҼҾӀӁӃӅӇӉӋӍӐӒӔӖӘӚӜӞӠӢӤӦӨӪӬӮӰӲӴӶӸԀԂԄԆԈԊԌԎԱԲԳԴԵԶԷԸԹԺԻԼԽԾԿՀՁՂՃՄՅՆՇՈՉՊՋՌՍՎՏՐՑՒՓՔՕՖႠႡႢႣႤႥႦႧႨႩႪႫႬႭႮႯႰႱႲႳႴႵႶႷႸႹႺႻႼႽႾႿჀჁჂჃჄჅḀḂḄḆḈḊḌḎḐḒḔḖḘḚḜḞḠḢḤḦḨḪḬḮḰḲḴḶḸḺḼḾṀṂṄṆṈṊṌṎṐṒṔṖṘṚṜṞṠṢṤṦṨṪṬṮṰṲṴṶṸṺṼṾẀẂẄẆẈẊẌẎẐẒẔẠẢẤẦẨẪẬẮẰẲẴẶẸẺẼẾỀỂỄỆỈỊỌỎỐỒỔỖỘỚỜỞỠỢỤỦỨỪỬỮỰỲỴỶỸἈἉἊἋἌἍἎἏἘἙἚἛἜἝἨἩἪἫἬἭἮἯἸἹἺἻἼἽἾἿὈὉὊὋὌὍὙὛὝὟὨὩὪὫὬὭὮὯᾸᾹᾺΆῈΈῊΉῘῙῚΊῨῩῪΎῬῸΌῺΏabcdefghijklmnopqrstuvwxyzªµºßàáâãäåæçèéêëìíîïðñòóôõöøùúûüýþÿāăąćĉċčďđēĕėęěĝğġģĥħĩīĭįıijĵķĸĺļľŀłńņňʼnŋōŏőœŕŗřśŝşšţťŧũūŭůűųŵŷźżžſƀƃƅƈƌƍƒƕƙƚƛƞơƣƥƨƪƫƭưƴƶƹƺƽƾƿdžljnjǎǐǒǔǖǘǚǜǝǟǡǣǥǧǩǫǭǯǰdzǵǹǻǽǿȁȃȅȇȉȋȍȏȑȓȕȗșțȝȟȡȣȥȧȩȫȭȯȱȳȴȵȶȷȸȹȼȿɀɐɑɒɓɔɕɖɗɘəɚɛɜɝɞɟɠɡɢɣɤɥɦɧɨɩɪɫɬɭɮɯɰɱɲɳɴɵɶɷɸɹɺɻɼɽɾɿʀʁʂʃʄʅʆʇʈʉʊʋʌʍʎʏʐʑʒʓʔʕʖʗʘʙʚʛʜʝʞʟʠʡʢʣʤʥʦʧʨʩʪʫʬʭʮʯΐάέήίΰαβγδεζηθικλμνξοπρςστυφχψωϊϋόύώϐϑϕϖϗϙϛϝϟϡϣϥϧϩϫϭϯϰϱϲϳϵϸϻϼабвгдежзийклмнопрстуфхцчшщъыьэюяѐёђѓєѕіїјљњћќѝўџѡѣѥѧѩѫѭѯѱѳѵѷѹѻѽѿҁҋҍҏґғҕҗҙқҝҟҡңҥҧҩҫҭүұҳҵҷҹһҽҿӂӄӆӈӊӌӎӑӓӕӗәӛӝӟӡӣӥӧөӫӭӯӱӳӵӷӹԁԃԅԇԉԋԍԏաբգդեզէըթժիլխծկհձղճմյնշոչպջռսվտրցւփքօֆևᴀᴁᴂᴃᴄᴅᴆᴇᴈᴉᴊᴋᴌᴍᴎᴏᴐᴑᴒᴓᴔᴕᴖᴗᴘᴙᴚᴛᴜᴝᴞᴟᴠᴡᴢᴣᴤᴥᴦᴧᴨᴩᴪᴫᵢᵣᵤᵥᵦᵧᵨᵩᵪᵫᵬᵭᵮᵯᵰᵱᵲᵳᵴᵵᵶᵷᵹᵺᵻᵼᵽᵾᵿᶀᶁᶂᶃᶄᶅᶆᶇᶈᶉᶊᶋᶌᶍᶎᶏᶐᶑᶒᶓᶔᶕᶖᶗᶘᶙᶚḁḃḅḇḉḋḍḏḑḓḕḗḙḛḝḟḡḣḥḧḩḫḭḯḱḳḵḷḹḻḽḿṁṃṅṇṉṋṍṏṑṓṕṗṙṛṝṟṡṣṥṧṩṫṭṯṱṳṵṷṹṻṽṿẁẃẅẇẉẋẍẏẑẓẕẖẗẘẙẚẛạảấầẩẫậắằẳẵặẹẻẽếềểễệỉịọỏốồổỗộớờởỡợụủứừửữựỳỵỷỹἀἁἂἃἄἅἆἇἐἑἒἓἔἕἠἡἢἣἤἥἦἧἰἱἲἳἴἵἶἷὀὁὂὃὄὅὐὑὒὓὔὕὖὗὠὡὢὣὤὥὦὧὰάὲέὴήὶίὸόὺύὼώᾀᾁᾂᾃᾄᾅᾆᾇᾐᾑᾒᾓᾔᾕᾖᾗᾠᾡᾢᾣᾤᾥᾦᾧᾰᾱᾲᾳᾴᾶᾷιῂῃῄῆῇῐῑῒΐῖῗῠῡῢΰῤῥῦῧῲῳῴῶῷⲁⲃⲅⲇⲉⲋⲍⲏⲑⲓⲕⲗⲙⲛⲝⲟⲡⲣⲥⲧⲩⲫⲭⲯⲱⲳⲵⲷⲹⲻⲽⲿⳁⳃⳅⳇⳉⳋⳍⳏⳑⳓⳕⳗⳙⳛⳝⳟⳡⳣⳤⴀⴁⴂⴃⴄⴅⴆⴇⴈⴉⴊⴋⴌⴍⴎⴏⴐⴑⴒⴓⴔⴕⴖⴗⴘⴙⴚⴛⴜⴝⴞⴟⴠⴡⴢⴣⴤⴥfffiflffifflſtstﬓﬔﬕﬖﬗ\d-_^]/8
+
 / End of testinput4 /