Add methods QPDFTokenizer::Token::isWord

m-holger · Jay Berkenbilt
1 parent dca70f13
Showing 9 changed files with 39 additions and 48 deletions
examples/pdf-filter-tokens.cc
include/qpdf/QPDFTokenizer.hh
libqpdf/Pl_QPDFTokenizer.cc
libqpdf/QPDF.cc
libqpdf/QPDFFormFieldObjectHelper.cc
libqpdf/QPDFPageObjectHelper.cc
libqpdf/QPDFTokenizer.cc
libqpdf/QPDF_linearization.cc
libtests/input_source.cc
@@ -139,8 +139,7 @@ ColorToGray::handleToken(QPDFTokenizer::Token const&amp; token)
     QPDFTokenizer::token_type_e token_type = token.getType();
     if (!isIgnorable(token_type)) {
         this->stack.push_back(token);
-        if ((this->stack.size() == 4) &&
-            (token == QPDFTokenizer::Token(QPDFTokenizer::tt_word, "rg")) &&
+        if ((this->stack.size() == 4) && token.isWord("rg") &&
             (isNumeric(this->stack.at(0).getType())) &&
             (isNumeric(this->stack.at(1).getType())) &&
             (isNumeric(this->stack.at(2).getType()))) {
@@ -112,6 +112,16 @@ class QPDFTokenizer
         {
             return this->type == tt_integer;
         }
+        bool
+        isWord() const
+        {
+            return this->type == tt_word;
+        }
+        bool
+        isWord(std::string const& value) const
+        {
+            return this->type == tt_word && this->value == value;
+        }
       private:
         token_type_e type;
@@ -53,9 +53,7 @@ Pl_QPDFTokenizer::finish()
         this->m->filter->handleToken(token);
         if (token.getType() == QPDFTokenizer::tt_eof) {
             break;
-        } else if (
-            (token.getType() == QPDFTokenizer::tt_word) &&
-            (token.getValue() == "ID")) {
+        } else if (token.isWord("ID")) {
             // Read the space after the ID.
             char ch = ' ';
             input->read(&ch, 1);
@@ -459,13 +459,11 @@ QPDF::findHeader()
 bool
 QPDF::findStartxref()
 {
-    QPDFTokenizer::Token t = readToken(this->m->file);
-    if (t == QPDFTokenizer::Token(QPDFTokenizer::tt_word, "startxref")) {
-        if (readToken(this->m->file).isInteger()) {
-            // Position in front of offset token
-            this->m->file->seek(this->m->file->getLastOffset(), SEEK_SET);
-            return true;
-        }
+    if (readToken(m->file).isWord("startxref") &&
+        readToken(m->file).isInteger()) {
+        // Position in front of offset token
+        this->m->file->seek(this->m->file->getLastOffset(), SEEK_SET);
+        return true;
     }
     return false;
 }
@@ -613,16 +611,13 @@ QPDF::reconstruct_xref(QPDFExc&amp; e)
             // containing this token
         } else if (t1.isInteger()) {
             QPDFTokenizer::Token t2 = readToken(this->m->file, MAX_LEN);
-            QPDFTokenizer::Token t3 = readToken(this->m->file, MAX_LEN);
             if ((t2.isInteger()) &&
-                (t3 == QPDFTokenizer::Token(QPDFTokenizer::tt_word, "obj"))) {
+                (readToken(m->file, MAX_LEN).isWord("obj"))) {
                 int obj = QUtil::string_to_int(t1.getValue().c_str());
                 int gen = QUtil::string_to_int(t2.getValue().c_str());
                 insertXrefEntry(obj, 1, token_start, gen, true);
             }
-        } else if (
-            (!this->m->trailer.isInitialized()) &&
-            (t1 == QPDFTokenizer::Token(QPDFTokenizer::tt_word, "trailer"))) {
+        } else if (!this->m->trailer.isInitialized() && t1.isWord("trailer")) {
             QPDFObjectHandle t =
                 readObject(this->m->file, "trailer", QPDFObjGen(), false);
             if (!t.isDictionary()) {
@@ -922,8 +917,7 @@ QPDF::read_xrefTable(qpdf_offset_t xref_offset)
             }
         }
         qpdf_offset_t pos = this->m->file->tell();
-        QPDFTokenizer::Token t = readToken(this->m->file);
-        if (t == QPDFTokenizer::Token(QPDFTokenizer::tt_word, "trailer")) {
+        if (readToken(m->file).isWord("trailer")) {
             done = true;
         } else {
             this->m->file->seek(pos, SEEK_SET);
@@ -1431,8 +1425,7 @@ QPDF::readObject(
     } else if (object.isDictionary() && (!in_object_stream)) {
         // check for stream
         qpdf_offset_t cur_offset = input->tell();
-        if (readToken(input) ==
-            QPDFTokenizer::Token(QPDFTokenizer::tt_word, "stream")) {
+        if (readToken(input).isWord("stream")) {
             // The PDF specification states that the word "stream"
             // should be followed by either a carriage return and
             // a newline or by a newline alone.  It specifically
@@ -1523,9 +1516,7 @@ QPDF::readObject(
                 // Seek in two steps to avoid potential integer overflow
                 input->seek(stream_offset, SEEK_SET);
                 input->seek(toO(length), SEEK_CUR);
-                if (!(readToken(input) ==
-                      QPDFTokenizer::Token(
-                          QPDFTokenizer::tt_word, "endstream"))) {
+                if (!readToken(input).isWord("endstream")) {
                     QTC::TC("qpdf", "QPDF missing endstream");
                     throw damagedPDF(
                         input, input->getLastOffset(), "expected endstream");
@@ -1556,9 +1547,8 @@ bool
 QPDF::findEndstream()
 {
     // Find endstream or endobj. Position the input at that token.
-    QPDFTokenizer::Token t = readToken(this->m->file, 20);
-    if ((t.getType() == QPDFTokenizer::tt_word) &&
-        ((t.getValue() == "endobj") || (t.getValue() == "endstream"))) {
+    auto t = readToken(m->file, 20);
+    if (t.isWord("endobj") || t.isWord("endstream")) {
         this->m->file->seek(this->m->file->getLastOffset(), SEEK_SET);
         return true;
     }
@@ -1682,8 +1672,8 @@ QPDF::readObjectAtOffset(
     QPDFTokenizer::Token tobj = readToken(this->m->file);
     bool objidok = tobjid.isInteger();
-    int genok = tgen.isInteger();
-    int objok = (tobj == QPDFTokenizer::Token(QPDFTokenizer::tt_word, "obj"));
+    bool genok = tgen.isInteger();
+    bool objok = tobj.isWord("obj");
     QTC::TC("qpdf", "QPDF check objid", objidok ? 1 : 0);
     QTC::TC("qpdf", "QPDF check generation", genok ? 1 : 0);
@@ -1743,8 +1733,7 @@ QPDF::readObjectAtOffset(
     QPDFObjectHandle oh = readObject(this->m->file, description, og, false);
-    if (!(readToken(this->m->file) ==
-          QPDFTokenizer::Token(QPDFTokenizer::tt_word, "endobj"))) {
+    if (!readToken(this->m->file).isWord("endobj")) {
         QTC::TC("qpdf", "QPDF err expected endobj");
         warn(damagedPDF("expected endobj"));
     }
@@ -556,7 +556,7 @@ ValueSetter::handleToken(QPDFTokenizer::Token const&amp; token)
     switch (state) {
     case st_top:
         writeToken(token);
-        if ((ttype == QPDFTokenizer::tt_word) && (value == "BMC")) {
+        if (token.isWord("BMC")) {
             state = st_bmc;
         }
         break;
@@ -571,7 +571,7 @@ ValueSetter::handleToken(QPDFTokenizer::Token const&amp; token)
         // fall through to emc
     case st_emc:
-        if ((ttype == QPDFTokenizer::tt_word) && (value == "EMC")) {
+        if (token.isWord("EMC")) {
             do_replace = true;
             state = st_end;
         }
@@ -751,7 +751,7 @@ TfFinder::handleToken(QPDFTokenizer::Token const&amp; token)
         break;
     case QPDFTokenizer::tt_word:
-        if (value == "Tf") {
+        if (token.isWord("Tf")) {
             if ((last_num > 1.0) && (last_num < 1000.0)) {
                 // These ranges are arbitrary but keep us from doing
                 // insane things or suffering from over/underflow
@@ -210,18 +210,16 @@ InlineImageTracker::handleToken(QPDFTokenizer::Token const&amp; token)
                 writeToken(token);
                 state = st_top;
             }
-        } else if (
-            token == QPDFTokenizer::Token(QPDFTokenizer::tt_word, "ID")) {
+        } else if (token.isWord("ID")) {
             bi_str += token.getValue();
             dict_str += " >>";
-        } else if (
-            token == QPDFTokenizer::Token(QPDFTokenizer::tt_word, "EI")) {
+        } else if (token.isWord("EI")) {
             state = st_top;
         } else {
             bi_str += token.getRawValue();
             dict_str += token.getRawValue();
         }
-    } else if (token == QPDFTokenizer::Token(QPDFTokenizer::tt_word, "BI")) {
+    } else if (token.isWord("BI")) {
         bi_str = token.getValue();
         dict_str = "<< ";
         state = st_bi;
@@ -892,7 +892,7 @@ QPDFTokenizer::findEI(std::shared_ptr&lt;InputSource&gt; input)
                 okay = true;
             } else if (type == tt_bad) {
                 found_bad = true;
-            } else if (type == tt_word) {
+            } else if (t.isWord()) {
                 // The qpdf tokenizer lumps alphabetic and otherwise
                 // uncategorized characters into "words". We recognize
                 // strings of alphabetic characters as potential valid
@@ -121,12 +121,9 @@ QPDF::isLinearized()
         }
         QPDFTokenizer::Token t1 = readToken(this->m->file);
-        QPDFTokenizer::Token t2 = readToken(this->m->file);
-        QPDFTokenizer::Token t3 = readToken(this->m->file);
-        QPDFTokenizer::Token t4 = readToken(this->m->file);
-        if (t1.isInteger() && t2.isInteger() &&
-            (t3 == QPDFTokenizer::Token(QPDFTokenizer::tt_word, "obj")) &&
-            (t4.getType() == QPDFTokenizer::tt_dict_open)) {
+        if (t1.isInteger() && readToken(m->file).isInteger() &&
+            readToken(m->file).isWord("obj") &&
+            (readToken(m->file).getType() == QPDFTokenizer::tt_dict_open)) {
             lindict_obj = toI(QUtil::string_to_ll(t1.getValue().c_str()));
         }
     }
@@ -37,9 +37,9 @@ Finder::check()
 {
     QPDFTokenizer tokenizer;
     QPDFTokenizer::Token t = tokenizer.readToken(is, "finder", true);
-    if (t == QPDFTokenizer::Token(QPDFTokenizer::tt_word, "potato")) {
+    if (t.isWord("potato")) {
         t = tokenizer.readToken(is, "finder", true);
-        return (t == QPDFTokenizer::Token(QPDFTokenizer::tt_word, after));
+        return t.isWord(after);
     }
     return false;
 }