OpenSystemsDevelopment / qpdf

Commit 611192f3c8e994f4b24c685f91c7c533e5e08b71

Authored by m-holger
1 parent 0518d585

Use nextToken in Tokenizer::findEI and QPDFWordTokenFinder::check

Inline Side-by-side
Showing 1 changed file with 10 additions and 11 deletions
libqpdf/QPDFTokenizer.cc
  View file @611192f
@@ -53,9 +53,9 @@ QPDFWordTokenFinder::check() @@ -53,9 +53,9 @@ QPDFWordTokenFinder::check()
53 // Find a word token matching the given string, preceded by a delimiter, and followed by a 53 // Find a word token matching the given string, preceded by a delimiter, and followed by a
54 // delimiter or EOF. 54 // delimiter or EOF.
55 Tokenizer tokenizer; 55 Tokenizer tokenizer;
56 - auto t = tokenizer.readToken(is, "finder", true, str.size() + 2); 56 + tokenizer.nextToken(is, "finder", str.size() + 2);
57 qpdf_offset_t pos = is.tell(); 57 qpdf_offset_t pos = is.tell();
58 - if (!(t == QPDFTokenizer::Token(QPDFTokenizer::tt_word, str))) { 58 + if (tokenizer.getType() != tt::tt_word || tokenizer.getValue() != str) {
59 QTC::TC("qpdf", "QPDFTokenizer finder found wrong word"); 59 QTC::TC("qpdf", "QPDFTokenizer finder found wrong word");
60 return false; 60 return false;
61 } 61 }
@@ -853,13 +853,13 @@ Tokenizer::findEI(InputSource& input) @@ -853,13 +853,13 @@ Tokenizer::findEI(InputSource& input)
853 // all required as well as a BI and ID. If we get 10 good tokens in a row or hit EOF, we can 853 // all required as well as a BI and ID. If we get 10 good tokens in a row or hit EOF, we can
854 // be pretty sure we've found the actual EI. 854 // be pretty sure we've found the actual EI.
855 for (int i = 0; i < 10; ++i) { 855 for (int i = 0; i < 10; ++i) {
856 - auto t = check.readToken(input, "checker", true);  
857 - auto type = t.getType();  
858 - if (type == tt::tt_eof) { 856 + check.nextToken(input, "checker");
  857 + auto typ = check.getType();
  858 + if (typ == tt::tt_eof) {
859 okay = true; 859 okay = true;
860 - } else if (type == tt::tt_bad) { 860 + } else if (typ == tt::tt_bad) {
861 found_bad = true; 861 found_bad = true;
862 - } else if (t.isWord()) { 862 + } else if (typ == tt::tt_word) {
863 // The qpdf tokenizer lumps alphabetic and otherwise uncategorized characters into 863 // The qpdf tokenizer lumps alphabetic and otherwise uncategorized characters into
864 // "words". We recognize strings of alphabetic characters as potential valid 864 // "words". We recognize strings of alphabetic characters as potential valid
865 // operators for purposes of telling whether we're in valid content or not. It's not 865 // operators for purposes of telling whether we're in valid content or not. It's not
@@ -868,13 +868,12 @@ Tokenizer::findEI(InputSource&amp; input) @@ -868,13 +868,12 @@ Tokenizer::findEI(InputSource&amp; input)
868 bool found_alpha = false; 868 bool found_alpha = false;
869 bool found_non_printable = false; 869 bool found_non_printable = false;
870 bool found_other = false; 870 bool found_other = false;
871 - for (char ch: t.getValue()) {  
872 - if (((ch >= 'a') && (ch <= 'z')) || ((ch >= 'A') && (ch <= 'Z')) ||  
873 - (ch == '*')) { 871 + for (char ch: check.getValue()) {
  872 + if ((ch >= 'a' && ch <= 'z') || (ch >= 'A' && ch <= 'Z') || (ch == '*')) {
874 // Treat '*' as alpha since there are valid PDF operators that contain * 873 // Treat '*' as alpha since there are valid PDF operators that contain *
875 // along with alphabetic characters. 874 // along with alphabetic characters.
876 found_alpha = true; 875 found_alpha = true;
877 - } else if ((static_cast<signed char>(ch) < 32) && (!isSpace(ch))) { 876 + } else if (static_cast<signed char>(ch) < 32 && !isSpace(ch)) {
878 // Compare ch as a signed char so characters outside of 7-bit will be < 0. 877 // Compare ch as a signed char so characters outside of 7-bit will be < 0.
879 found_non_printable = true; 878 found_non_printable = true;
880 break; 879 break;